Datasets:

BioDEX

/

BioDEX-QA

like 5

Follow

BioDEX 6

Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Hyponatraemia'.

A phase 1/2 trial of an immune-modulatory vaccine against IDO/PD-L1 in combination with nivolumab in metastatic melanoma. Anti-programmed death (PD)-1 (aPD1) therapy is an effective treatment for metastatic melanoma (MM); however, over 50% of patients progress due to resistance. We tested a first-in-class immune-modulatory vaccine (IO102/IO103) against indoleamine 2,3-dioxygenase (IDO) and PD ligand 1 (PD-L1), targeting immunosuppressive cells and tumor cells expressing IDO and/or PD-L1 (IDO/PD-L1), combined with nivolumab. Thirty aPD1 therapy-naive patients with MM were treated in a phase 1/2 study ( https://clinicaltrials.gov/ , NCT03047928). The primary endpoint was feasibility and safety; the systemic toxicity profile was comparable to that of nivolumab monotherapy. Secondary endpoints were efficacy and immunogenicity; an objective response rate (ORR) of 80% (confidence interval (CI), 62.7-90.5%) was reached, with 43% (CI, 27.4-60.8%) complete responses. After a median follow-up of 22.9 months, the median progression-free survival (PFS) was 26 months (CI, 15.4-69 months). Median overall survival (OS) was not reached. Vaccine-specific responses assessed in vitro were detected in the blood of >93% of patients during vaccination. Vaccine-reactive T cells comprised CD4+ and CD8+ T cells with activity against IDO- and PD-L1-expressing cancer and immune cells. T cell influx of peripherally expanded T cells into tumor sites was observed in responding patients, and general enrichment of IDO- and PD-L1-specific clones after treatment was documented. These clinical efficacy and favorable safety data support further validation in a larger randomized trial to confirm the clinical potential of this immunomodulating approach. pmcMain Despite remarkable advances in the treatment of MM with immune checkpoint inhibitors (ICIs) targeting PD-1 and cytotoxic T lymphocyte antigen 4 (CTLA-4), around half of patients are resistant to ICI monotherapy1. The combination of anti-CTLA-4 (aCTLA-4) and aPD1 therapy is to date the most effective therapy resulting in a response rate of around 60%; however, 50% of the patients also develop severe adverse events2,3. Therefore, an equally effective but less toxic treatment is highly needed. Several approaches to enhance ICI efficacy are currently being investigated, such as other ICIs, T cell therapy with tumor-infiltrating T cells or innate immunity stimulators such as Toll-like receptor 9 agonists4–6. Treating cancer patients with vaccines that stimulate a targeted immune response is another attractive approach, with very few side effects observed thus far in combination immunotherapy studies7,8. Immune-modulatory vaccines targeting tumoral immune escape mechanisms offer a new, generalizable strategy compared to patient-specific neoantigen cancer vaccines7,9. The immune-modulatory vaccine strategy in this clinical trial is based on the finding of circulating cytotoxic T cells specific to IDO and PD-L1 in the blood of patients with cancer and, to a lesser extent, in healthy donors. IDO- and PD-L1-specific CD8+ T cells can directly recognize and kill IDO+ and/or PD-L1+ tumor cells and likewise recognize and kill non-malignant cells that express their cognate targets. Furthermore, IDO- and PD-L1-specific CD4+ T cells release pro-inflammatory cytokines in response to IDO- or PD-L1-expressing target cells. IDO and PD-L1 are expressed not only by melanoma cells but also by many other cell types in the tumor microenvironment (TME), which differentiates these antigens from traditional tumor antigens used in other studies10–15. Activation of IDO/PD-L1-specific T cells by vaccination can therefore restrict the range of immunosuppressive signals mediated by immunosuppressive cells and thereby revert the TME from an immune hostile to an immune friendly environment. In animal models of cancer, vaccinations with IDO epitopes resulted in anti-tumor therapeutic effects that were correlated with reductions in IDO expression in myeloid cell populations within the TME16. The IDO/PD-L1 immune-modulating vaccine may lead to a translatable strategy for improving the efficacy of aPD1 therapy through activation of specific T cells. We hypothesize that the IDO/PD-L1 vaccine attracts T cells into the tumor, which induces type 1 helper T (TH1) cell inflammation and reverts the TME into an immune-permissive site, thereby turning the tumor ‘hot’. This would also upregulate PD-L1 expression in cancer and immune cells, generating more susceptible targets to aPD1 therapy (Extended Data Fig. 1a). This theory was confirmed in a mouse model in which aPD1 therapy and IDO vaccination show synergistic effects16. In this phase 1/2 clinical trial, MM1636, patients with MM received a combination of the IDO/PD-L1 (IO102/IO103) peptide vaccine with the adjuvant Montanide and the aPD1 antibody nivolumab. Patients were included in three cohorts: 30 aPD1 therapy-naive patients (cohort A), ten aPD1 therapy-refractory patients (cohort B, de novo resistance) and ten patients who progressed after aPD1 therapy (cohort C, acquired resistance). Here, we report results from cohort A. The vaccine was given biweekly for the first six administrations and thereafter every 4th week. A maximum of 15 vaccines were administered. Nivolumab was given in parallel, biweekly (3 mg per kg) or every 4th week (6 mg per kg) for up to 2 years (Extended Data Fig. 1b). The primary objective was safety and feasibility. Secondary objectives were immunogenicity and clinical efficacy. Results Patients and treatment Thirty patients were enrolled from December 2017 to June 2020. None of the 30 patients dropped out of the study; all received at least three cycles of therapy (Supplementary Fig. 1). At the current database lock (5 October 2020), six patients were still on treatment in the trial. Of the 24 patients who were not on trial treatment at data cutoff, two are still receiving nivolumab monotherapy (6 mg per kg every 4 weeks). Reasons for stopping treatment for the remaining 22 patients included disease progression (37%), toxicity (20%), maximum benefit or complete response (CR) confirmed on two consecutive scans (17%) or completing 2 years of treatment (7%). For the 24 patients who were not on trial treatment at data cutoff, the mean number of vaccinations was 10.5 (range, 3–15 vaccinations). Thirteen of these 24 patients continued nivolumab (6 mg per kg, every 4 weeks) as a standard of care. Nine patients received subsequent therapy after progression (Supplementary Table 1). Baseline characteristics are shown in Table 1. The mean age was 70 years; 37% of patients had elevated lactate dehydrogenase (LDH) levels, 60% were stage M1c, 37% had BRAF mutations, and 43% were negative for PD-L1 (<1%). A total of three patients (10%) had received prior ipilimumab therapy. No patients had brain metastasis (Supplementary Table 2).Table 1 Baseline patient characteristics (n = 30) Characteristic Number (%) Mean age, years (range) 70 (46–85) Sex, male 16 (55%) ECOG PS 0 26 (87%) LDH levels  ≤ULN 19 (63%)  >ULN 11 (37%) M stage (AJCC-8)  M1a 6 (20%)  M1b 6 (20%)  M1c 18 (60%) Number of lesion sites  1 6 (20%)  2–3 17 (57%)  >3 7 (23%) Liver metastases present 10 (33%) BRAF status  Mutant 11 (37%)  Wild type 19 (63%) PD-L1  <1% 13 (43%)  >1% 17 (57%) Previous systemic therapy  Ipilimumab 3 (10%)  No 27 (90%) AJCC-8, eighth edition of the American Joint Committee on Cancer; ECOG PS, Eastern Cooperative Oncology Group performance status; ULN, upper limit of normal. Notable clinical responses to the combination therapy Thirty patients with MM were treated with the IDO/PD-L1 vaccine and nivolumab according to the trial protocol. By investigator review, the ORR reached 80% (CI, 62.7–90.5%), with 43% of patients (CI, 27.4–60.8%) achieving a CR and 37% (CI, 20.9–54.5%) reaching a partial response (PR) as the best overall response, while 20% experienced progressive disease (PD) according to Response Evaluation Criteria in Solid Tumors (RECIST) 1.1 (Fig. 1a). Two of the patients with a PR did not have a confirmatory scan stating PR on two consecutive scans. Early onset of response was frequent, with 22 of 30 patients having an objective response at the first evaluation (after 12 weeks on treatment). Median times to PR and CR were 75 d (range, 54–256 d) and 327 d (range, 73–490 d), respectively (Fig. 2a–c).Fig. 1 Clinical response. a, Pie charts with percent ORR, CR, PR and PD according to RECIST 1.1 by investigator review of all patients (n = 30), PD-L1+ patients (>1%, (n = 17)) and PD-L1− patients (<1%, n = 13)), respectively. Two-sided CIs (95%) were constructed using the Clopper–Pearson method. b, Treatment effect in MM1636 compared with a matched historical control group from the DAMMED database (n = 74). Patients in MM1636 (n = 29) were matched with the exact same combination variable according to age (≤70 years, >70 years), sex, LDH levels (normal, elevated), M stage (M1a, M1b, M1c), BRAF status (wild type, mutated) and PD-L1 status (<1%, ≥1%). Estimates for treatment effects were calculated by weighted logistic regression analyses and weighted Cox proportional hazard model. Bar height indicates the estimated response rate; tops of bars are centers for error bars. Odds ratios (OR), response rates and their corresponding 95% CIs were extracted from the regression model. All P values were two sided, and P values below 0.05 were considered statistically significant. c, Best change in the sum of target lesion size compared with that at baseline (n = 30). The horizontal line at −30 shows the threshold for defining an objective response in the absence of non-target disease progression or new lesions according to RECIST 1.1. Two patients with 100% reduction in target lesion size had non-target lesions present. White stars, six patients had normalization (<10 mm) of fluor-18-deoxyglucose (FDG)-negative lymph nodes (at baseline, lymph nodes were >1.5 cm and FDG+) and 100% reduction of non-lymph node lesions and are considered to have had a CR (green bar). Black star, one patient (MM29) was considered to have had a PR (blue bar), although he had not reached a −30% change in target lesion size. The patient had a single measurable 13-mm lung metastasis at baseline and multiple biopsy-verified cutaneous metastases on the left crus (not detectable by positron-emission tomography–computed tomography (PET–CT) at baseline). The best change in target lesion size was 10 mm, and a post-treatment biopsy from the cutaneous metastases showed no sign of malignancy. Thus, overall, the patient was classified as having a PR. d, Kaplan–Meier curve of the response duration in the 24 patients with an objective response. e, Kaplan–Meier curve of PFS in all 30 treated patients. f, Kaplan–Meier curve of OS in all 30 treated patients. Fig. 2 Clinical response. a, Swimmer plot showing response duration and time to response according to RECIST 1.1 for all treated patients (n = 30). Triangles indicate first evidence of PR, while squares indicate first evidence of CR. Closed circles indicate time of progression. Arrows indicate ongoing responses. Patient MM18 died due to nivolumab-induced side effects. b, Spider plot showing response kinetics in all treated patients (n = 30). Red squares indicate time of progression. c, PET–CT images of patient MM42 before and after treatment (after 12 series of treatment) showing FDG metabolism in target lesions. SC, subcutaneous. The ORR among PD-L1+ (>1% (clone 28.8)) patients (n = 17) was 94.1% (CI, 73–99.7%) and 61.5% (CI, 35.5–82.3%) in PD-L1− patients (n = 13) (Fig. 1a). Objective responses were observed in patients irrespective of human leukocyte antigen (HLA) genotype (Supplementary Fig. 2). Clinical response data were validated by blinded independent external review, in which an ORR of 76.6% (CI, 57.7–90.1%) was reported, with 53.3% of patients achieving a CR, 23.3% achieving a PR and 3.3% experiencing stable disease. Comparisons between investigator review and external review are outlined in Supplementary Table 3. To examine whether the very high response rate should be attributed to nivolumab or to the vaccine, we retrieved clinical response information from a matched historical control group using the Danish Metastatic Melanoma Database (DAMMED) from contemporaneously treated patients with stage 3–4 melanoma who received aPD1 monotherapy17. Patients were matched with the exact same combination variable according to age, sex, PD-L1 status, BRAF status, LDH level and M stage (those at stage M1d were excluded from the control group (no patients with brain metastasis)). Matched controls were identified for 29 patients, and the ORR of 79.3% (CI, 61.0–90.4%) observed in MM1636 was found to be significantly higher (P < 0.0012) than that in the matched control group in which an ORR of 41.7% (CI, 31.0–53.3%) was reached. Furthermore, of the 29 patients in MM1636, a significantly (P < 0.0017) higher percentage (41.4% (CI, 25.2–59.6%)) of patients achieved CR in MM1636 than that (12% (CI, 6.3–21.6%)) in the matched historical control group. ORR and complete response rate (CRR) in the matched historical control group were comparable to those of patients treated in randomized phase 3 pivotal trials with aPD1 monotherapy18 (Fig. 1b). The treatment leads to prolonged PFS At data cutoff, the median duration of response had not been reached, with 87% of all responding patients being free from progression at 12 months (Fig. 1d). Patients were followed for up to 35 months with a median follow-up time of 22.9 months (CI, 14.9–26.2 months). OS and PFS were calculated from the first day of treatment to death or progression or to the date of the last follow-up (5 October 2020). The median PFS (mPFS) for all treated patients was 26 months (CI, 15.4–69 months) and was not reached for responding patients (Fig. 1e). The median OS was not reached at the data cutoff. OS at 12 months was 81.6% (CI, 61.6–92%) (Fig. 1f). One patient (MM18) with a CR died from nivolumab-related severe adverse events; the remaining patients died because of metastatic melanoma. For comparison, the mPFS was 8.3 months (CI, 5.5 months–NR (not reached)) in the matched historical control group (n = 74), while the median OS was 23.2 months (CI, 23.2 months–NR). (Extended Data Fig. 2a,b). The combination of the IDO/PD-L1 vaccine and nivolumab was safe Treatment-related adverse events are listed for all 30 patients in Supplementary Table 4. Common treatment-related grade 1–2 toxicities were fatigue (47%), rash (47%), arthralgia (30%), diarrhea (23%), nausea (23%), dry skin (20%), pruritus (20%), infusion reaction (17%), xerostomia (17%) and myalgia (17%). Four patients (13%) experienced grade 3–4 adverse events: one patient with a grade 3 maculopapular rash (MM01), one patient with grade 3 adrenal insufficiency (MM06) and one patient with grade 3 arthralgia (MM22). Patient MM18 died from urosepsis with multi-organ failure and severe hyponatremia. This patient had experienced multiple immune-related adverse events with grade 3 colitis, grade 2 pneumonitis, grade 3 arthralgia, grade 2 vasculitis and grade 2 nivolumab infusion-related allergic reaction. Additionally, patient MM18 had symptoms of myocarditis at the time of death with highly elevated cardiac troponin I levels. Bedside echocardiography showed an ejection fraction of 15%, which at baseline was 60%, but an autopsy was not conducted, and myocarditis was never confirmed pathologically. Patient MM06 had received first-line treatment with ipilimumab before entering the trial and was on substitution corticosteroids at the time of inclusion. Adrenal insufficiency was aggravated by an erysipelas infection with high fever, reaching grade 3 in Common Terminology Criteria for Adverse Events and resolving quickly after appropriate antibiotic therapy was initiated. As expected, local side effects were common with 77% of the patients who developed injection site reactions. These reactions were classified as granulomas (63%), redness (20%), pain (13%) and pruritus (13%) at the injection site. All local reactions were grade 1–2, most likely related to the Montanide adjuvant and typically transient. However, two patients (MM07 and MM20) decided to discontinue vaccination after eight and 11 injections, respectively, due to granulomas, tenderness and pain that limited instrumental activities of daily living but continued to receive nivolumab (Extended Data Fig. 2c). Vaccine-specific responses in blood were frequently detected First, all 30 patients were assessed for the presence of vaccine-specific responses in peripheral blood mononuclear cells (PBMCs) before, on and after vaccination using a modified interferon (IFN)-γ enzyme-linked immune absorbent spot (ELISPOT) assay. This assay is known to expand antigen-specific memory cells and to improve the detection and the correlative power of ELISPOT during treatment19–21. Prevaccine IDO-specific responses were detectable in ten (33%) patients, while prevaccine PD-L1-specific responses were detectable in eight patients (27%); overlapping (specific to both IDO and PD-L1) prevaccine responses were present in four (13.3%) patients. During vaccination, an increase of IDO-specific T cells or PD-L1-specific T cells in the blood was observed in 28 (93%) and 26 (86%) patients, respectively. In total, 93% of patients had an increase in either PD-L1- or IDO-specific responses on vaccination (Fig. 3a), with a significant (P < 0.0001) median increase from baseline to the post-vaccine response demonstrated for both IDO and PD-L1 (at different time points on treatment), confirming that vaccine-specific immune responses were induced in patients regardless of clinical response (Fig. 3b,c). Immune responses fluctuated in the blood over time (Extended Data Fig. 3a). An increase in IDO- and PD-L1-specific responses in the peripheral blood was also detectable directly ex vivo across the different clinical response groups (Extended Data Fig. 4).Fig. 3 Vaccine-specific responses in blood. a, IDO- and PD-L1-specific T cell responses in PBMCs at baseline and on vaccination as measured by the IFN-γ ELISPOT assay (n = 30). *Responses were calculated as the difference between the average numbers of spots in wells stimulated with IDO or PD-L1 peptide (triplicates) and those from the corresponding control (DMSO), and statistical analyses of ELISPOT responses were performed using a distribution-free resampling method (Moodie et al.50). DR (double response), response was not statistically confirmed due to replicate number, but the number of spots in peptide wells was two times higher than that in control wells (DMSO). NS, no significant response and no DR. For a detailed overview of responses at serial time points on vaccination, see Extended Data Fig. 4a. b, IDO- and PD-L1-specific T cell response in PBMCs in all treated patients measured by the IFN-γ ELISPOT assay at baseline and on vaccination. On-vaccination responses were selected from the ‘best’ ELISPOT response at different time points for each patient (series 3, 6, 12, 18 or 24) during vaccination (n = 30). Wilcoxon matched-pairs signed-rank test was used to compare responses to IDO or PD-L1 peptides in the vaccine between baseline and later time points. c, Representative example of ELISPOT wells with response in patient MM23 in serial PBMCs before and on treatment. d, IDO-specific CD4+ and CD8+ T cells were isolated and expanded from PBMCs stimulated in vitro with the IDO peptide and a low dose of IL-2 for 14–15 d before sorting using the Miltenyi Cytokine Secretion Assay—Cell Enrichment and Detection kit. To assess their cytolytic potential, IDO-specific T cells were stimulated with IDO peptide, and expression of CD107a, IFN-γ and TNF-α was assessed by flow cytometry (the example is from patient MM14). Sustained vaccine-specific responses were observed 3 and 6 months after the last vaccine, indicating induction of memory responses in nine patients with response to clinical treatment who surpassed follow-up at data lock (Extended Data Fig. 3b). Importantly, PD-L1- and IDO-specific responses were observed irrespective of HLA genotype (Supplementary Table 5). To verify the functionality of vaccine-induced T cells, IDO- or PD-L1-specific T cells were isolated and in vitro expanded from PBMCs of five patients. Phenotypic characterization by flow cytometry revealed that the isolated vaccine-specific T cells consisted of both CD4+ and CD8+ T cells. Also, both IDO- and PD-L1-specific CD4+ and CD8+ T cells showed pro-inflammatory properties, as they expressed the cytolytic marker CD107a and secreted the cytokines IFN-γ and tumor necrosis factor (TNF)-α (Fig. 3d and Extended Data Fig. 6a–d). Interestingly, we were also able to detect vaccine-specific CD4+ and CD8+ T cell responses in peripheral blood ex vivo (Extended Data Figs. 4 and 5). We observed a significant increase in the overall percentage of CD107a, CD137 and TNF-α expression in response to peptide stimulation in on- or post-treatment PBMC samples compared to that at baseline, further confirming the expansion and the diverse signature of vaccine-specific T cells (Extended Data Fig. 5). Vaccine-specific responses detected in the skin at the vaccination site To investigate whether vaccine-specific T cells have the potential to migrate to peripheral tissue, delayed-type hypersensitivity (DTH) tests were performed after six cycles of treatment on 15 patients to assess the presence of vaccine-reactive T cells in the skin. Supplementary Table 6 display an overview of skin-infiltrating lymphocyte (SKIL) cultures. We detected IDO-specific T cells in the skin of six of ten patients and PD-L1-specific T cells in nine of 11 patients (Extended Data Fig. 7a). Intracellular cytokine staining was performed on SKILs from five patients after stimulation with either PD-L1 or IDO peptide. Here, we detected mainly CD4+ peptide-reactive T cells that secreted TNF-α and upregulated CD107a, and a minor fraction also secreted IFN-γ. In one patient, we detected CD8+ PD-L1-reactive T cells (Extended Data Fig. 7b–d). Vaccine-induced T cells specifically recognize target cells To confirm the functionality of vaccine-expanded T cells, vaccine-specific T cell clones (clonal purity was confirmed by T cell receptor (TCR) sequencing) were isolated and expanded from patient PBMCs (Extended Data Fig. 8e). We showed that PD-L1-specific T cells were able to recognize PD-L1+ autologous tumor cells in a PD-L1 expression-dependent manner if the cancer cells also expressed HLA-II (Fig. 4a,b). Similarly, an HLA-DR-restricted IDO-specific CD4+ T cell clone was able to recognize an HLA-DR-matched IDO-expressing model cell line, MonoMac1, in an IDO-expression-dependent manner (Fig. 4e–g). As previously described, IDO- and PD-L1-specific T cells’ mode of action is not limited to targeting only cancer cells. We were able to show that vaccine-specific T cell clones also reacted against PD-L1- and IDO-expressing autologous immune cells (Fig. 4c,h). To provide myeloid cells with a tumor-associated phenotype we treated isolated CD14+ myeloid cells with tumor conditioned medium (TCM) derived from an established autologous tumor cell line. We observed that such TCM-treated CD14+ cells had increased expression of PD-L1 and IDO and were effectively recognized by autologous PD-L1- and IDO-specific CD4+ T cell clones (Fig. 4c,d,h,i).Fig. 4 PD-L1- and IDO-specific T cells from vaccinated patients react against PD-L1- and IDO-expressing target cells. a, Left, PD-L1-specific T cell culture (MM1636.05) reactivity against PD-L1 peptide or autologous tumor cells in the IFN-γ ELISPOT assay. Tumor cells were either not treated or pretreated with 200 U ml−1 IFN-γ for 48 h before the assay. Effector:target (E:T) ratio of 10:1 was used. Right, PD-L1 and HLA-II surface expression on melanoma cells with (green) or without (yellow) pretreatment with IFN-γ compared to an isotype control (gray) as assessed by flow cytometry. b, Left, PD-L1-specific T cell (MM1636.05) reactivity in the IFN-γ ELISPOT assay against autologous tumor cells pretreated with IFN-γ (500 U ml−1) and transfected with mock or PD-L1 small interfering (si)RNA 24 h after transfection. E:T ratio, 10:1. Right, PD-L1 surface expression on melanoma tumor cells (MM1636.05) assessed by flow cytometry 24 h after transfection with mock (blue) or PD-L1 (red) siRNA compared to the isotype control (gray). c, Reactivity of the CD4+ PD-L1-specific T cell clone (MM1636.14) against PD-L1 peptide or autologous CD14+ cells; E:T ratio, 10:1. CD14+ cells were isolated using magnetic bead sorting and used as targets in an ELISPOT assay directly or after pretreatment for 2 d with TCM derived from the autologous tumor cell line. d, Quantitative PCR with reverse transcription (RT–qPCR) analysis of PD-L1 (CD274) expression in sorted CD14+ cells before and after treatment with autologous TCM for 48 h. e, Reactivity of the IDO-specific CD4+ T cell clone (MM1636.23) against IDO peptide combined with HLA-DR (L243)-, HLA-DQ (SPV-L3)- or HLA-DP (B7/21)-blocking antibodies (aHLA-DR, aHLA-DQ and aHLA-DP) in an intracellular staining assay (ICS) for IFN-γ and TNF-α production. T cells were incubated with individual blocking antibodies (2 μg ml−1) for 30 min before adding IDO peptide. f, Reactivity of the IDO-specific CD4+ T cell clone (MM1636.23) against the HLA-DR-matched IDO-expressing cell line MonoMac1 transfected with mock or IDO siRNA in an ICS assay, E.T ratio, 4:1. siRNA transfection was performed 48 h before the experiment. g, RT–qPCR analysis of IDO1 expression in MonoMac1 cells 48 h after siRNA transfection. h, Reactivity of the CD4+ IDO-specific T cell clone (MM1636.14) against IDO peptide or autologous CD14+ cells; E:T ratio, 20:1. CD14+ cells were isolated using magnetic bead sorting and used as targets in an ELISPOT assay directly or after pretreatment with TCM derived from the autologous tumor cell line. i, RT–qPCR analysis of IDO1 expression in sorted CD14+ cells before and after treatment with autologous TCM for 48 h. Bars in RT–qPCR data (d,g,i) represent the mean of three (d,i) or six (g) technical replicates ±s.d.; P values were determined by two-tailed parametric t-tests. ELISPOT counts (a,b,c,h) represent the mean value of three technical replicates ±s.e.m.; response P values were determined using the distribution-free resampling (DFR) method. TNTC, too numerous to count. T cell clones in blood and tumors To track the role of treatment-induced T cell responses, TCR sequencing of the complementarity-determining region 3 (CDR3) was performed on five patients in peripheral blood (baseline and cycles 3, 6 and 12) and paired biopsies. These five patients (MM01, MM02, MM08, MM09 and MM13) were selected due to the availability of material and to investigate a balanced patient group with both responders and non-responders. Due to a limited number of available paired biopsies (either as a consequence of patient refusal or unexpectedly rapid and substantial clinical responses), statistical analysis could not be applied. Details on clinical response are shown in Fig. 2a. Additionally, PBMCs (on treatment) or SKILs were stimulated with the IDO/PD-L1 peptides, and then cytokine-producing T cells were sorted to track vaccine-induced T cells both in the periphery and at the tumor site. To identify enriched IDO/PD-L1-specific T cell clones, TCR rearrangements in sorted IDO/PD-L1-specific clones and TCR rearrangements in sorted IDO/PD-L1-specific T cell samples were compared to sequences from baseline PBMC samples for each patient. Clonal expansion of vaccine-specific TCR rearrangements from samples on vaccination were then tracked using a differential abundance framework. Cumulative IDO/PD-L1-specific T cell frequencies were tracked in post-treatment samples. We found no relation between clinical response and the enrichment of vaccine-specific clones, but an increase in IDO/PD-L1-specific T cell clones was observed at different time points in the periphery in all five patients (Extended Data Fig. 3c). We next investigated overall changes in the T cell repertoire in the blood. A modest increase in the peripheral T cell fraction was observed in the three responding patients at cycle 3, while the two non-responding patients had a clear decrease in T cell fraction (Extended Data Fig. 8a). We thereafter investigated TCR clonality and TCR repertoire richness, exploring the proportion of abundant clones and the number of unique rearrangements, respectively. A decreasing peripheral Simpson clonality and increasing TCR repertoire richness was observed in responding patients at cycle 3, which might indicate tumor trafficking upon treatment. The opposite pattern was observed in non-responding patients (Extended Data Fig. 8b,c). Peripherally expanded clones were associated with tumors and persisted until cycle 12 (latest time point analyzed). The largest peripheral expansion was observed at cycle 3, with the most significant increase observed in patient MM01 (CR). Responding patients had a larger fraction of peripherally expanded clones that were also found in tumors compared to that of non-responders. By tracking peripherally expanded clones detected at the tumor site, we observed that patient MM01 had a substantial increase after treatment, indicating tumor trafficking of peripheral expanded clones (Extended Data Fig. 8d). Influx at the tumor site of vaccine-enriched T cell clones Given the observation of increased T cell fraction and enrichment of IDO- and PD-L1-specific clones in the blood after treatment, we investigated whether the same trend was observed at the tumor site. Both TCR sequencing and immunohistochemistry (IHC) of paired biopsies from the five patients described above showed an increase in the T cell fraction with an influx of CD3+ and CD8+ T cells after treatment in the three responding patients (Fig. 5a–c). IHC was not possible for patient MM09 due to tissue loss.Fig. 5 Changes in the TME after treatment. Number of CD3+ and CD8+ T cells, TCR fraction, TCR clonality, TCR repertoire, biopsy expanded TCR clones and enrichment of IDO- and PD-L1-specific T cells at the tumor site. a, Number of CD3+ and CD8+ T cells per mm2 (sum in the validated area) at the tumor site detected by IHC of paired biopsies from four patients. b, Example of IHC of CD3+ and CD8+ T cells at the tumor site before and after treatment (cells per mm2) in one patient (MM01). c, T cell fraction at the tumor site at baseline and cycle 6 by TCR sequencing. The T cell fraction was calculated by taking the total number of T cell templates and dividing by the total number of nucleated cells. d, Tracking of vaccine-associated clones at baseline and cycle 6 in tumor biopsies. Cumulative frequencies of IDO and PD-L1 vaccine-specific TCR rearrangements are represented. e,f, TCR clonality and TCR repertoire richness in five patients at the tumor site at baseline and cycle 6. Simpson clonality measures how evenly TCR sequences are distributed among a set of T cells, where 0 indicates an even distribution of frequencies and 1 indicates an asymmetric distribution in which a few clones dominate. TCR repertoire richness reports the mean number of unique rearrangements. g, Bar chart representing baseline expanded biopsy clones from five patients (colored bars) and the detection of biopsy expanded clones also found in the blood at baseline and series 3, 6 and 12 (white and gray bars) by TCR sequencing. We thereafter investigated whether some of the IDO/PD-L1 vaccine-linked T cells were present at the tumor site. Vaccine-associated clones were tracked as the combined frequency of IDO- and PD-L1-specific T cell rearrangements. In biopsies, the frequencies at cycle 6 were compared to those at baseline and showed an increase in vaccine-specific T cells in four of five patients, irrespective of clinical response (Fig. 5d). TCR sequencing of PD-L1-specific SKILs (a more specific culture than IDO/PD-L1-specific isolations derived from PBMCs on vaccination) and paired biopsies showed that two of the top five PD-L1-specific SKIL clones were present at the tumor site both before and after treatment (Extended Data Fig. 7e,f). With a focus on the more abundant T cell clones, we investigated overall TCR clonality at the tumor site before and after treatment. In addition, we explored the number of unique TCR rearrangements, dissecting the lower-frequency clones. Patient MM01 had a significant increase in TCR clonality and a decrease in repertoire richness at the tumor site after therapy, indicating a focused tumor repertoire response of selected clones. All three responding patients had a decrease in TCR repertoire richness, which again might indicate a focused tumor response (Fig. 5e,f). Deeper analyses showed that the T cell clones that expanded at the tumor site after therapy were also present in the blood at baseline and increased significantly after treatment in four of five patients. The highest proportion was detected early at cycle 3. These data again support trafficking of peripherally expanded clones to the tumor site and could indicate that the T cell response to treatment is derived from pre-existing peripherally tumor-associated T cells (Fig. 5g). Signs of treatment-induced inflammation in the TME To dissect changes in the TME induced by T cell influx upon treatment in responding patients, RNA gene expression analyses using the nCounter PanCancer Immune Profiling Panel from NanoString were performed on paired biopsies from two responding patients (MM01 and MM13). Expression of genes related to adaptive immunity such as T cell activation, effector functions (genes encoding IFN-γ, TNF-α, IL-15 and IL-18) and cytotoxicity was increased in post-treatment biopsies (Extended Data Fig. 9a,b). Also, expression of genes related to checkpoint inhibitors such as those encoding T cell immunoglobulin and mucin-domain containing-3 (TIM-3), IDO, PD-L1, PD-L2, PD-1 and CTLA-4 increased after treatment, indicating activation of immune cells in the TME (Extended Data Fig. 9c). Additionally, IHC of paired biopsies from four patients (MM01, MM02, MM05 and MM13) showed an upregulation of PD-L1, IDO, MHC-I and MHC-II on tumor cells, indicating a treatment-induced pro-inflammatory response in the three responding patients, except for a decrease in MHC-II expression in patient MM13. By contrast, the non-responding patient MM02 had a reduction in T cell numbers present in the tumor after treatment and no expression of PD-L1, IDO or MHC-II and, interestingly, total loss of MHC-I, demonstrating tumor immune escape (Extended Data Fig. 10a). CD8+ T cells and their distance (µm) to PD-L1-expressing cells in baseline biopsies from five patients was investigated by IHC. Except for patient MM13 (PR), distance and clinical responses were associated: the two responders had reduced distance (<20 µm) between cells expressing these markers compared to non-responding patients (>80 µm). This observation indicates that responding patients not only have a higher intratumoral infiltration of CD8+ T cells, but that these cells can surround and attack PD-L1-expressing immune cells and tumor cells (Extended Data Fig. 10b). Discussion In this clinical trial, MM1636, 30 patients with metastatic melanoma were treated with a first-in-class immunomodulatory IDO/PD-L1-targeting peptide vaccine combined with nivolumab. The treatment led to an unprecedented high ORR of 80%, with 43% of patients reaching a CR, and a striking mPFS of 26 months (95% CI, 15.4–69 months) was reached. The vaccine represents a new treatment strategy to activate specific T cells that target cells contributing to immune suppression (including tumor cells), positively modulating the TME by inducing local inflammation. Indeed, we show that vaccine-specific T cells isolated and expanded from vaccinated patients recognize not only tumor cells in a target- and HLA-restricted manner but also myeloid cells polarized toward a tumor-associated phenotype. Hence, myeloid cells become targets for vaccine-activated T cells when they have a tumor-associated phenotype in the TME. These phenomena may further induce checkpoint molecules and rewire the TME toward an increasingly aPD1-permissive state. A drawback of our study is the single-center nonrandomized setup. Comparison between trials or between patients in trials and real-world patients is problematic due to multiple factors, such as period of conduction and other therapies available at the different periods of time. Nevertheless, the rate of investigator-assessed ORR in the phase 3 trial CheckMate 067 was 43.7% in the nivolumab monotherapy group and 57% in the nivolumab and ipilimumab group. CRs occurred in 8.9% and 11.5% of patients, respectively18. The mPFS of 26 months (95% CI, 15.4–69 months) in this trial is more than twice as long as that for patients treated with nivolumab and ipilimumab in CheckMate 067, for which an mPFS of 11.5 months (95% CI, 8.7–19.3 months) was reached. Patient baseline characteristics were in general comparable to those of patients with MM who have been treated in CheckMate 067, although patients in MM1636 were older (mean age, 70 years) and a larger fraction were positive for PD-L1 (57%)3,18,22. Among patients with PD-L1-negative tumors in MM1636, an ORR of 61.5% was still reached, which would be expected to be around 33% for first-line nivolumab monotherapy23. Some studies suggest that older patients might have a tendency to respond better to aPD1 therapy; however, this is still debated24–27. To address potential trial bias and the nonrandomized setup, patients in MM1636 were matched for age, performance status, sex, M stage, LDH level, PD-L1 status and BRAF status with a historical control group from the DAMMED, who were treated contemporarily (2015–2019) with aPD1 monotherapy as standard of care17. We found a significantly higher ORR and CRR in MM1636 compared to those of matched patients, who had an ORR of 43% and a CRR of 13%, comparable to patients treated in CheckMate 067. Restrictions of the synthetic control group are of course that it is partially historic and patient selection outside matching criteria cannot be ruled out28. Numerous contemporary clinical trials are exploring the combination of aPD1 therapy with other immunomodulating agents for advanced melanoma. Talimogene laherparepvec, an oncolytic virus, is approved by the Food and Drug Administration and the European Medicines Agency to treat advanced melanoma. A small phase 1b trial with 21 patients (MASTERKEY-265) combined talimogene laherparepvec and pembrolizumab to treat patients with advanced unresectable melanoma and reached an ORR of 62% and a CR of 33%29,30. Seventy-one percent of the patients in this trial had an M stage below M1c; this number was 40% in our trial. Furthermore, mainly patients with an M stage below M1c responded to treatment, which was not the case in MM1636. Results from a large randomized phase 3 trial are awaited (KEYNOTE-034). Epacadostat, an IDO inhibitor, was tested in combination with pembrolizumab in a nonrandomized phase 2 trial in 40 aPD1 treatment-naive patients with MM with promising results, reaching an ORR of 62%. Unfortunately, the phase 3 trial showed no indication that epacadostat provided improvement in PFS and OS31. Limitations of the phase 3 trial were the sparse information on pharmacodynamics as well as biomarker evaluation to improve the design. The IDO/PD-L1 vaccine is different from epacadostat, as it is not an IDO inhibitor but targets IDO- and PD-L1-expressing cells. Similar vaccines administered as monotherapy induced objective responses in lung cancer and basal cell carcinoma, while epacadostat as monotherapy in 52 patients resulted in no responses32,33. Sahin et al. recently published encouraging data from a first-in-human trial, in which a vaccine containing liposomal RNA targeting four unmutated tumor-associated antigens (NY-ESO-1, MAGE-A3, tyrosinase and TPTE) was administered alone or in combination with nivolumab in patients with advanced melanoma. Responses were observed in both the monotherapy group as well as the combination group in checkpoint inhibitor-experienced patients, suggesting the efficacy of non-mutant shared tumor antigen vaccines34. The overall safety and tolerability findings are comparable to those of aPD1 monotherapy. Injection site reactions were exclusive to the vaccine. However, these side effects were transient and mild in most patients and most likely due to the adjuvant Montanide. IDO- and PD-L1-specific CD8+ and CD4+ T cells exist among peripheral blood lymphocytes in healthy donors35–38 and expand in response to pro-inflammatory stimuli35,39. Furthermore, both IDO and PD-L1 are induced in cells as a counter response to the inflammatory response. This provides a mechanism that ensures immune homeostasis, which keeps IDO/PD-L1-specific T cells in check; therefore, the risk of triggering autoimmune-related adverse events by vaccination appears to be minimal. This was confirmed in the first clinical trials of IDO and PD-L1 vaccination (NCT01543464 (refs. 33,40) and NCT03042793 (ref. 41)). The spontaneous (baseline) immune response to the vaccines observed in the current study is in agreement with our previous observation in various patients with cancer35–38. Numerous vaccine-induced changes in the blood and at the tumor site were observed. Peripheral IDO- and/or PD-L1-specific T cells were detected in vitro in a modified ELISPOT in over 93% of patients on vaccination, unrelated to patient HLA type. Immune responses were persistent in patients who surpassed follow-up at data cutoff and were still detectable up to 6 months after the last vaccine, suggesting induction of memory T cells. We did not observe a correlation between vaccine-induced responses in blood and clinical responses. However, the detection of a highly significant increase in vaccine-specific T cell numbers after vaccination in almost all patients together with the very low number of patients with PD makes it difficult to expect such a correlation, especially because other aspects should be taken into account (for example, the loss of class I expression on tumors cells in patient MM02 with PD after vaccination). Frequencies of peripheral T cells induced against PD-L1 were overall higher than those against IDO. Importantly, however, we also observed that it was not the same patients who responded strongly to both IDO and PD-L1, nor did patients react with a similar response pattern to the two antigens; that is, different time points of response were observed. This suggests that each component of the vaccine plays different roles in the ongoing immune response in patients. The goal of IDO/PD-L1 vaccination was to modulate the TME to increase responsiveness to aPD1 therapy, as we have observed in animal models, in which immune conversion is demonstrated in the TME with an increased influx of T cells. Targeting both IDO and PD-L1 together enables synergy, as the TME is known to use different immune escape mechanisms and IDO and PD-L1 are often overexpressed by different cellular compartments. Ex vivo ELISPOT and flow cytometry assays confirmed the induction of immune response toward both epitopes and the higher immunogenicity of the PD-L1 epitope, although because of lower sensitivity failed to detect some of the responses. TCR sequencing of five patients confirmed enrichment of IDO/PD-L1-specific T cell clones in the blood at different time points after treatment. Furthermore, an increase in enriched IDO/PD-L1-specific clones was observed in four of five patients at the tumor site after treatment, irrespective of clinical response. Phenotypic characterization showed that vaccine-specific T cells, which were expanded in vitro with interleukin (IL)-2 from the blood of vaccinated individuals, were both CD4+ T cells and CD8+ T cells. This was confirmed by ex vivo phenotype description of vaccine-activated T cells. Vaccine-specific T cells expressed CD107a and CD137 and produced IFN-γ and TNF-α upon stimulation with the cognate target, indicating their cytolytic capacity. Overall, the data from immune monitoring supports the importance of both antigens in the generation of the frequent clinical responses in the study. Despite a limited number of paired biopsies (due to either patient refusal or the fact that a large fraction of responding patients had no assessable tumors after six cycles), we observed trends indicating treatment-induced general T cell influx in responding patients. It was shown that proliferation of CD8+ T cells in the tumor after aPD1 treatment is associated with radiographic reduction in tumor size42. Additionally, we showed that a large proportion of expanded peripheral TCR clones were associated with tumors and the most considerable amount of clonal expansion was observed early, at cycle 3. For the patient with a CR included in the TCR sequencing analyses (MM01), the number of peripheral expanded clones present at the tumor site increased after treatment compared to that at baseline, indicating tumor trafficking of peripheral expanded clones. Gene expression analyses (two paired biopsies) and IHC (five paired biopsies) further demonstrated that the combination treatment induced a pro-inflammatory TME in responding patients with signs of T cell activation and cytotoxicity and increased cytokine activity. This may lead to further upregulation of IDO, PD-L1, MHC-I and MHC-II on tumor cells, leading to more treatment targets. It was shown that, following vaccination with a cancer vaccine, PD-L1 expression is increased on tumor cells due to recruitment of tumor-specific T cells and upregulation of adaptive immune resistance pathways in the TME43. Treatment with nivolumab monotherapy enhances PD-L1 expression, and it is therefore problematic to discriminate the effect of the vaccine as compared to that of nivolumab44. IDO- and PD-L1-specific pro-inflammatory effector T cells were hypothesized to counteract the functions of IDO- and PD-L1-expressing immune-suppressive cells as a means to keep the immune balance between immune activation and inhibition45. However, because the expression of these molecules is induced as a counter-regulatory response to inflammatory mediators such as IFNs, they can also be expressed by, for example, activated myeloid cells or T cells. Thus, activation of PD-L1-specific T cells may result in depletion of activated T cells in the tumor and other sites. Importantly, it was recently described that PD-L1+ T cells mainly have tolerogenic effects on tumor immunity and exert tumor-promoting properties, suggesting that targeting this immune population is indeed also beneficial46. We have previously investigated the effect of activating PD-L1-specific T cells in vitro and in vivo and found that, overall, they supported the effector phase of an immune response by removing PD-L1-expressing regulatory immune cells that inhibit PD-1+ effector T cells47,48. The major role of the PD-1 pathway is believed to be the regulation of effector T cell responses. Thus, this protective pathway is more important after activation, rather than at the initial T cell-activation stage49. Accordingly, the presence of PD-L1-specific T cells during the activation phase of an immune response may not increase or support a pro-inflammatory immune response due to the expression of PD-L1 on potent antigen-presenting cells or on the T cells themselves14. Thus, the overall effects of PD-L1-specific T cells may vary depending on the expression of both PD-1 and PD-L1, that is, due to the microenvironment and the state of the immune response. In conclusion, here we report an impressive response rate, CR rate and mPFS for a first-in-class immune-modulating vaccine combined with nivolumab. This may be a first step toward a new treatment strategy for patients with MM. Limitations are the low number of patients treated at a single institution and the lack of a randomized design with aPD1 monotherapy as comparator. Studies in aPD1 therapy-resistant or -refractory melanoma are ongoing as well as biomarker analysis for selecting patients at a higher likelihood to benefit from the combination versus aPD1 monotherapy. A larger randomized trial will be essential to validate these findings and determine the specific contribution of the vaccine to clinical responses and changes in the TME. In December 2020, the Food and Drug Administration granted breakthrough therapy designation for the IO102/IO103 vaccine combined with aPD1 therapy in metastatic melanoma based on data from the MM1636 trial. Methods Trial design and treatment plan MM1636 is an investigator-initiated, nonrandomized, open-label, single-center phase 1/2 study. All patients were treated at the Department of Oncology, Herlev and Gentofte Hospital, University of Copenhagen, Herlev, Denmark. The study was conducted according to the Declaration of Helsinki and Good Clinical Practice (GCP) and monitored by the GCP unit in Copenhagen, Denmark. The protocol was approved by the ethical committee of the Capital region of Denmark (H-17000988), the Danish Medical Agencies (2017011073) and the Capital Region of Denmark Data Unit (P-2019-172). The study was registered at https://clinicaltrials.gov/ under the identifier NCT03047928 and at the EudraCT (no. 2016-0004527-23). This study initially aimed to include 30 aPD1 treatment-naive patients with MM. An amendment with the addition of two other cohorts with ten patients in each cohort was made to evaluate immune responses and clinical efficacy in aPD1 therapy-resistant patients (cohort B, de novo resistance and cohort C, acquired resistance) for a total of 50 patients. The amendment with cohorts B and C was approved at the inclusion of 18 patients in cohort A. The trial is still including patients in cohorts B and C. This article reports results from cohort A. The first six patients were treated as for phase 1, evaluating for safety and tolerability before the remaining 24 patients were included in phase 2. The IDO/PD-L1 vaccine was administered subcutaneously biweekly for the first 10 weeks and thereafter every 4 weeks for approximately 9 months. A maximum of 15 vaccines were administered. Nivolumab was administered according to the approved label (3 mg per kg biweekly) for 24 cycles. The 15th vaccine was administered together with the 24th nivolumab cycle of 3 mg per kg, and responding patients thereafter continued nivolumab monotherapy every 4th week (6 mg per kg) as a standard of care after investigator assessment. Treatment with nivolumab was discontinued at the maximum benefit (assessed by the investigator), after a maximum of 2 years of therapy, at progression or due to severe adverse events (Extended Data Fig. 1b, treatment plan). Vaccine composition Each vaccine was composed of 100 µg IO102, a 21-amino-acid peptide (DTLLKALLEIASCLEKALQVF) from IDO, and 100 µg IO103, a 19-amino-acid peptide (FMTYWHLLNAFTVTVPKDL) from the signal peptide of PD-L1 (PolyPeptide Laboratories). Peptides were dissolved separately in 50 µl DMSO, filtered for sterility and frozen at −20 °C (NUNC CyroTubes CryoLine System Internal Thread, Sigma-Aldrich). At <24 h before administration, peptides were thawed. The PD-L1 peptide was diluted in 400 µl sterile water and, immediately before injection, mixed with the IDO peptide solution and 500 µl Montanide ISA-51 (SEPPIC) to achieve a total volume of 1 ml. Patients Patients above 18 years of age with locally advanced or stage 4 melanoma according to the AJCC (seventh edition), at least one measurable lesion according to RECIST 1.1 and an ECOG PS of 0–1 were eligible. The main exclusion criteria were prior treatment with aPD1 therapy, CNS metastases >1 cm, severe comorbidities and active autoimmune disease. Enrollment was not restricted to PD-L1 status, but it was known before inclusion. Patients were included after informed consent. Patients MM42 and MM20 have confirmed their approval of PET–CT images and clinical images that have been included in the article. Key study assessments Safety and tolerability were evaluated based on changes in clinical laboratory analyses and reported adverse events. Adverse events were assessed according to Common Terminology Criteria for Adverse Events (version 5.0) and were graded from 1 to 5 for all treated patients up to 6 months after the last dose of the IDO/PD-L1 vaccine. Clinical efficacy was assessed using FDG PET–CT scans before treatment and every 3rd month until progression. Objective responses were categorized into CR, PR, stable disease or PD according to RECIST version 1.1. Clinical data were collected at CCIT-DK in the eCRF program OpenClinica version 1.0 and in Microsoft Excel version 2002 on a secure server. Blood samples for immunologic analyses were collected before treatment, before the third cycle, after the sixth, 12th, 18th and 24th cycles (on vaccination) and 3 and 6 months after the last vaccine. Two to three tumor needle biopsies (1.2 mm) were collected at baseline and after six cycles from the same tumor site, when accessible, to evaluate immune responses at the tumor site. DTH skin tests were performed and punch biopsies were taken from the DTH area after cycle 6 for the evaluation of SKILs reactive to PD-L1 and IDO (Supplementary Fig. 1, treatment plan). Statistical analysis of clinical outcome Survival curves were computed in GraphPad Prism software version 9.0.0 according to Kaplan–Meier method. Median follow-up time of enrollment was calculated using the reverse Kaplan–Meier method, also in GraphPad Prism software version 9.0.0. For binary outcomes, 95% two-sided CIs were constructed using the Clopper–Pearson method, also in GraphPad version 9.0.0. An independent board of certified and experienced oncoradiologists performed an external review to evaluate clinical response to address the potential bias of investigator site review. The external review took place at Rigshospitalet, Copenhagen University Hospital. This hospital did not participate in the MM1636 trial, and the external reviewer had no prior knowledge about the clinical trial or the trial therapy. Only PET–CT images were accessed, and arrows indicating target or non-target lesions appeared on baseline images as the only additional information. To address potential trial bias regarding treatment effect, we matched patients in MM1636 with patients from the DAMMED, a population-based database that retrospectively collects data on patients with metastatic melanoma in Denmark. Here, data from 938 patients treated with aPD1 monotherapy contemporaneously (January 2015–October 2019) were extracted. Two hundred and eighteen of these patients were eligible for comparison and matching (all parameters available) (Supplementary Table 1), and 74 patients from the DAMMED were found to match. Patients were matched for age (≤70 years, >70 years), sex, LDH levels (normal, elevated), M stage (M1a, M1b, M1c), BRAF status (wild type, mutated) and PD-L1 status (<1%, ≥1%). An exact matching algorithm was used in which patients in MM1636 were matched with patients from the DAMMED with the same combination of variables. Twenty-nine patients from MM1636 were matched with exact combinations of the six variables. One patient could not be matched. To secure balance of the calculations, control patients were weighted according to the number of patients for each MM1636 patient. Estimates for treatment effects were calculated by weighted logistic regression analyses and weighted Cox proportional hazard model. The R package ‘Matchlt’ was used for matching patients. As the method chosen for matching control patients to protocol patients, a weighted binary logistic regression model was used for comparing response rates in the two matched cohorts. Odds ratios and response rates, including their corresponding 95% CIs, were extracted from the regression models. All P values were two sided, and P values below 0.05 were considered statistically significant. SAS version 9.4M5 was used for the weighted logistic regression models. Processing of peripheral blood mononuclear cells Peripheral blood was collected from all patients in heparinized tubes and was processed within 4 h. In brief, PBMCs were isolated using Lymphoprep (Medinor) separation. PBMCs were counted on a Sysmex XP-300 analyzer and frozen in Human AB Serum (Sigma-Aldrich, H4522-100ML) with 10% DMSO using controlled-rate freezing (CoolCell, BioCision) in a −80 °C freezer and were moved the next day to a freezer at −140 °C until further processing. Needle biopsies at baseline and after the sixth vaccine Two to three needle biopsies (1.2 mm) were taken at baseline and after the sixth cycle of treatment when assessable from the same tumor lesion. One fragment was fixed with formalin and embedded in paraffin (FFPE); one to two fragments were used for expanding tumor-infiltrating T cells and establishing autologous tumor cell lines. Delayed-type hypersensitivity and generation of skin-infiltrating lymphocytes After six cycles of treatment, we performed intradermal injections of vaccine components without adjuvant and one control injection containing DMSO without peptide. Patients were injected with either a mixture of both IDO and PD-L1 peptides at all three injection sites or PD-L1 peptide, IDO peptide or a mixture of the two at the injection sites, respectively (Supplementary Table 6). Eight hours after injection, punch biopsies were resected from the three sites where the peptide was injected and transported immediately to the laboratory and cut into fragments of 1–2 mm3. SKILs were expanded to establish ‘young SKILs’ in CM medium consisting of RPMI 1640 with GlutaMAX, 25 mM HEPES, pH 7.2 (Gibco, 72400-021), IL-2 (100 or 6,000 IU ml−1) (Proleukin Novartis, 004184), 10% heat-inactivated Human AB Serum (Sigma-Aldrich, H4522-100ML), 100 U ml−1 penicillin, 1.25 µg ml−1 Fungizone (Bristol Myers Squibb, 49182) and 100 µg ml−1 streptomycin (Gibco, 15140-122). Half of the medium was replaced three times per week. Young SKILs from samples derived from IDO peptide, PD-L1 peptide and the mixture injection sites were further expanded in a small-scale version of the 14-d rapid expansion protocol as previously described51. Quantification of specific T cells in blood by ELISPOT For enumeration of vaccine-specific T cells in the peripheral blood, PBMCs from patients were stimulated with IDO or PD-L1 peptide in the presence of a low dose of IL-2 (120 U ml−1) for 7–13 d before being used in IFN-γ ELISPOT assays with 2.5–3.2 × 105 cells per well. Briefly, cells were placed in a 96-well PVDF membrane-bottomed ELISPOT plate (MultiScreen MSIPN4W50, Millipore) precoated with IFN-γ-capture antibody (clone 1-D1K, Mabtech). Diluted IDO or PD-L1 peptide stocks in DMSO were added at 5 µM; an equivalent amount of DMSO was added to control wells. PBMCs from each patient were set up in duplicate or triplicate for peptide and control stimulations. Cells were incubated in ELISPOT plates in the presence of the peptide for 16–18 h, after which they were washed off, and biotinylated secondary antibody (anti-human IFN-γ mAB, clone 7-B6-1, Mabtech) was added. After 2 h of incubation, unbound secondary antibody was washed off, and streptavidin-conjugated alkaline phosphatase (Mabtech) was added for 1 h. Next, unbound streptavidin-conjugated enzyme was washed off, and the assay was developed by adding BCIP/NBT substrate (Mabtech). ELISPOT plates were analyzed on the CTL ImmunoSpot S6 Ultimate V analyzer using ImmunoSpot software version 5.1. Responses were calculated as the difference between the average numbers of spots in wells stimulated with IDO or PD-L1 peptide and those from corresponding control wells. For detection of IDO and PD-L1 peptide responses ex vivo, PBMCs were thawed and rested overnight in medium containing DNase I (1 μg ml−1, Sigma-Aldrich, 11284932001) before being used in IFN-γ ELISPOT assays as described above. A total of 6–9 × 105 PBMCs per well were seeded. Statistical analysis of all ELISPOT responses was performed using the DFR method as described by Moodie et al. using RStudio software (RStudio Team, 2016, http://www.rstudio.com/)50. Vaccine-specific ELISPOT responses were defined as true if the difference between the spot count in control and peptide-stimulated wells was statistically significant according to DFR statistical analysis or, for samples performed in duplicate, if the spot count in peptide-stimulated wells was at least 2× the spot count in control wells50. Cancer cell lines and tumor conditioned medium Autologous melanoma cell lines were established from needle biopsies. Briefly, biopsies were chopped into small fragments and seeded in 24-well cultures in RPMI 1640 with GlutaMAX, 25 mM HEPES, pH 7.2 (Gibco, 72400-021), 10% heat-inactivated FBS (Life Technologies, 10500064), 100 U ml−1 penicillin, 1.25 µg ml−1 Fungizone (Bristol Myers Squibb, 49182) and 100 µg ml−1 streptomycin (Gibco, 15140-122). Established adherent melanoma tumor cell lines were cryopreserved at −140 °C in freezing medium containing FBS with 10% DMSO. PD-L1 and HLA-II expression on established tumor cell lines was assessed by flow cytometry staining with anti-PD-L1–PE-Cy7 (1:23 dilution, clone M1h1 (RUO), BD, 558017) and anti-HLA-II–FITC (1:23 dilution, clone Tu39 (RUO), BD, 555558) antibodies. To obtain TCM, established tumor cell lines were cultured in 175-cm2 Nunc cell culture flasks until 80–90% confluency was reached. The culture medium was then replaced with 20 ml fresh X-VIVO 15 with Gentamicin and Phenol Red (Lonza, BE02-060Q), medium with 5% heat-inactivated Human AB Serum (Sigma-Aldrich, H4522-100ML). After 24 h of incubation, TCM was collected and centrifuged to remove any resuspended cells, after which TCM was aliquoted, frozen and stored at −80 °C. The acute monocytic leukemia cell line MonoMac1 was obtained from the DSMZ—German Collection of Microorganisms and Cell Cultures (ACC 252) and cultured in RPMI 1640 with GlutaMAX, 25 mM HEPES, pH 7.2 (Gibco, 72400-021) and 10% heat-inactivated FBS. Isolation of autologous myeloid cells Autologous CD14+ cells were sorted from freshly thawed PBMCs using a magnetic bead-separation kit (Miltenyi Biotec, 130-050-201) according to the manufacturer’s instructions. Isolated CD14+ cells were used as targets in the IFN-γ ELISPOT assay directly after sorting or differentiated in vitro into tumor-associated macrophages by culturing with 1 ml fresh X-VIVO 15 medium with Gentamicin and Phenol Red (Lonza, BE02-060Q) and 5% heat-inactivated Human AB Serum, supplemented with 1 ml autologous TCM in 24-well plates for 2 d. Quantification of vaccine-specific T cells from DTH biopsy sites by ELISPOT SKILs expanded in vitro were rested in medium without IL-2 overnight before being used in the IFN-γ ELISPOT assay as described above to evaluate reactivity of skin-infiltrating T cells. Generation of IDO- and PD-L1-specific T cell cultures from PBMCs or SKILs IDO- or PD-L1-specific T cells were isolated from peptide-stimulated in vitro PBMC cultures on days 14–15 after stimulation or from SKIL cultures expanded in vitro. For specific T cell isolation, PBMCs or SKILs were stimulated with IDO or PD-L1 peptide, and cytokine-producing T cells were sorted using the IFN-γ or TNF-α Secretion Assay—Cell Enrichment and Detection kit (Miltenyi Biotec). Cytokine-production profile of PD-L1- and IDO-specific T cells by intracellular staining To assess the T cell cytokine-production profile, isolated and expanded IDO- and PD-L1-specific T cell cultures were stimulated for 5 h with peptide at 5 μM in a 96-well plate. One hour after the start of the incubation, anti-CD107a–PE (1:133 dilution, clone H4A3, BD Biosciences, 555801) antibody and BD GolgiPlug (1:1,000 dilution, BD Biosciences) were added. After a 5-h incubation, cells were stained using fluorescently labeled surface marker antibodies: anti-CD4–PerCP (1:5 dilution, clone SK3, 345770), anti-CD8–FITC (1:5 dilution, clone SK1, 345772) and anti-CD3–APC-H7 (1:20 dilution, clone SK7, 560275) (all from BD Biosciences). Dead cells were stained using FVS510 (1:250 dilution, BD Biosciences, 564406), followed by overnight fixation and permeabilization using eBioscience Fixation/Permeabilization buffers (00-5123-43, 00-5223-56) according to the manufacturer’s instructions. Cells were then stained intracellularly in eBioscience permeabilization buffer (00-8333-56) with anti-IFN-γ–APC (1:20 dilution, clone 25723.11, BD Biosciences, 341117) and anti-TNF-α–BV421 (1:100 dilution, clone Mab11, BD Biosciences, 562783) antibodies. Samples were analyzed on the FACSCanto II (BD Biosciences) using BD FACSDiva software version 8.0.2. The gating strategy is shown in Supplementary Fig. 3a. For assessing the reactivity against IDO expressing MonoMac1 cells, the appropriate amount of cancer cells was added to the IDO-specific T cells to obtain an effector:target ratio of 4:1 in the 96-well plate and T cell cytokine production was tested after 5 h of co-culture as described above. For HLA-blocking experiments, HLA-blocking antibodies (2 μg ml−1) were added directly into wells 30 min before the addition of peptide. Blocking antibodies used were HLA-DR (1:500 dilution, clone L243, Abcam, ab136320), HLA-DQ (1:500 dilution, clone SPV-L3, Abcam, ab23632) and HLA-DP (1:500 dilution, clone B7/21, Abcam, ab20897). To assess ex vivo T cell reactivity to IDO and PD-L1 peptides in patient PBMCs, cells were thawed and rested for 1–2 d in medium containing DNase I (1 μg ml−1, Sigma-Aldrich, 11284932001). PBMCs were then stimulated with peptide at 5 μM in a 96-well plate for 8 h. An hour after the addition of peptide, anti-CD107a–BV421 (3:50 dilution, clone H4A3, BD Biosciences, 328626) antibody and BD GolgiPlug (1:1,000 dilution, BD Biosciences) were added. Surface and intracellular staining was performed as described above. Antibodies used for surface staining were anti-CD3–PE-CF594 (0.8:30 dilution, clone UCHT1, BD Biosciences, 562280), anti-CD4–BV711 (1:30 dilution, clone SK3, BD Biosciences, 563028) and anti-CD8–Qdot605 (1:150 dilution, clone 3B5, Thermo Fisher, Q10009). Antibodies used for intracellular staining were anti-CD137–PE (1:20 dilution, clone 4B4-1 (RUO), BD Biosciences, 555956), anti-IFN-γ–PE-Cy7 (1.5:20 dilution, clone B27, BD Biosciences, 557643) and anti-TNF-α–APC (1:20 dilution, clone Mab11, BD Biosciences, 554514). Samples were acquired on the NovoCyte Quanteon (ACEA Biosciences) and analyzed using NovoExpress software version 1.4.1. To assess vaccine-specific T cell responses, background values observed in unstimulated PBMC samples were subtracted from values observed in peptide-stimulated conditions. Positive response value threshold was set at a difference of 0.2% from the background values. Based on this response cutoff, only TNF-α, CD107a and CD137 responses were detected in this assay. Statistical analysis comparing baseline with on-treatment or post-treatment cytokine profiles was performed using two-sided Wilcoxon matched-pairs signed-rank test. The gating strategy is shown in Supplementary Fig. 3b. siRNA-mediated PD-L1 and IDO1 silencing A stealth siRNA duplex for targeted silencing of PD-L1 (Invitrogen)52, a custom silencer select siRNA for targeted silencing of IDO1 (Ambion) and the recommended silencer select negative control (Ambion) siRNA for mock transfection were used. The stealth PD-L1 siRNA duplex consisted of the sense sequence 5′-CCUACUGGCAUU-UGCUGAACGCAUU-3′ and the antisense sequence 5′-AAUGCGUUCAGCAAAUGCCAGUAGG-3′. The three silencer IDO siRNA duplexes used were siRNA1 (sense sequence, 5′-ACAUCUGCCUGAUCUCAUATT-3′; antisense sequence, 5′-UAUGAGAUCAGGCAGAUGUTT-3′), siRNA2 (sense sequence, 5′-CCACGAUCAUGUGAACCCATT-3′; antisense sequence, 5′-UGGGUUCACAUGAUCG-UGGAT-3′) and siRNA3 (sense sequence, 5′-CGAUCAUGUGAACCCAAAATT-3′; antisense sequence, 5′-UUUUGGGUUCACAUGAUCGTG-3′). For PD-L1 or IDO1 silencing, cancer cells were electroporated with 0.025 nmol of each siRNA duplex as previously described53. For PD-L1-silencing experiments, cancer cells were treated with IFN-γ (500 U ml−1, PeproTech) 1 h after electroporation. Electroporated cells were used as target cells in ELISPOT and ICS assays 24 h or 48 h after siRNA electroporation. RT–qPCR Total RNA was extracted using the RNeasy Plus Mini kit (Qiagen, 74134) following the manufacturer’s instructions. RNA concentration was quantified using a NanoDrop 2000 (Thermo Fisher Scientific), and a total of 1,000 ng RNA was reverse transcribed with the High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems, 4368814) using 1,000 ng input RNA for transcription. Real-time qPCR analyses were performed using the TaqMan Gene Expression Assay on a Roche LightCycler 480 instrument. RT–qPCR assays were performed with a minimum of three technical replicates and analyzed using the ΔCt method as described in Bookout et al.54 with normalization of IDO1 expression (primer ID, Hs00984148_m1) or PD-L1 expression (primer ID, Hs001125296_m1) to the expression level of the housekeeping gene POLR2A (primer ID, Hs00172187_m1) and the control sample (mock). For low-concentration samples with no amplification, Ct was set to 40. Controls without reverse transcriptase were used as controls for specific amplifications. P values were determined using two-tailed parametric t-tests. BRAF-mutational status and PD-L1 status at baseline in all patients A library of historical FFPE biopsies were assessable for all patients and analyzed locally at Herlev and Gentofte University Hospital by experienced pathologists for BRAF status and PD-L1 expression on tumor cells. The BRAF analysis was carried out with real-time PCR using the EntroGen BRAF Mutation Analysis Kit II (BRAFX-RT64, CE-IVD) to specifically detect mutations corresponding to V600D, V600E and V600K in the BRAF protein. PD-L1 status was assessed using the monoclonal rabbit anti-PD-L1 antibody (clone 28.8, PD-L1 IHC 28-8 pharmDx) in FFPE biopsies following the manufacturer’s instructions. Patients were considered positive for PD-L1 with expression levels ≥1% and negative for PD-L1 with expression levels <1%. Human leukocyte antigen type Blood samples from all 30 patients were genotyped for class I (HLA-A, HLA-B, HLA-C) and three class II types (HLA-DRB1, HLA-DQA1, HLA-DQB1) using LinkSēq HLA Typing Kits (Thermo Fisher, 1580C). These test kits are based on real-time PCR using allele-specific exponential amplification (sequence-specific primers) followed by melting curve analyses. Immunohistochemistry simplex: Immunoscore CR IHC staining was performed using a qualified Ventana Benchmark XT with four different steps: (1) antigen retrieval; (2) staining with the following primary antibodies: anti-CD3 antibody (clone HDX2; provider, HalioDx; HD-FG-000013; 10931065/10636667; concentration, 0.25 µg ml−1), anti-CD8 antibody (clone HDX3; provider, HalioDx; HD-FG-000019; 10931069/10337710/10639301; concentration, 0.5 µg ml−1), anti-IDO monoclonal antibody (clone VINC3IDO; provider, Thermo Fisher Scientific; 14-9750-82; E25003-101; concentration, 0.05 µg ml−1), anti-HLA class 1 ABC antibody (clone, EMR8-5; provider, Abcam; ab70328; GR3248333/GR3186494; concentration, 0.5 µg ml−1), anti-HLA-DR/DP/DQ/DX antibody (clone CR3/43; provider, Santa Cruz; sc-53302; L1714; concentration, 1 µg ml−1) and anti-PD-L1 antibody (clone HDX3; provider, HalioDx; HD-FG-000035; 106312810/106312816; concentration, 3.3 µg ml−1); (3) detection with a secondary antibody using the ultraView Universal DAB Detection kit; and (4) counterstaining using Hematoxylin & Bluing Reagent (staining of cellular nuclei). Control slides were systematically included in each staining run to permit quality control of the obtained measurements. Following coverslipping, slides were scanned with the NanoZoomer-XR to generate digital images (20×). Two consecutive slides were specifically used to perform Immunoscore CR TL staining of CD3+ and CD8+ cells. Digital pathology of T lymphocytes The digital pathology for Immunoscore CR TL allowed the quantification of positive cells (in cells per mm2) into the core tumor and invasive margin if present. Each sample was analyzed using the HalioDx Digital Pathology Platform. Digital pathology and pathologist analysis of IDO Digital pathology for IDO allowed the quantification of stained area (in mm2) into the whole tumor. Each sample was analyzed using the HalioDx Digital Pathology Platform. Analysis of IDO+ cells was performed by a pathologist. Results were expressed as an H score from 0 to 300. The score is obtained by the formula 3 × percentage of cells with strong staining + 2 × percentage of cells with moderate staining + percentage of cells with weak staining. Digital pathologist analysis of MHC-I and MHC-II Analysis of MHC-I+ and MHC-II+ cells was performed by a pathologist. Results are expressed as a percentage of positive tumor cells and a percentage of positive cells in the stroma. Digital pathology Immunoscore immune checkpoint (CD8+ and PD-L1+) The Immunoscore CR IC test allowed the quantification of CD8++ cell density in the whole tumor and a CD8+-centered proximity index (which corresponds to the percentage of CD8++ cells that have at least one PD-L1+ cell in the neighborhood) at different cutoff distances (20 µm, 40 µm, 60 µm and 80 µm). Pathologist analysis of PD-L1 A pathologist performed analysis of PD-L1+ cells. Positivity of a viable tumor cell was considered when partial or complete cell membrane staining was observed (more than 10% of the tumor cell membrane). Results were expressed as a percentage. NanoString RNA profiling and Immunosign RNA was extracted from FFPE tissues using Qiagen RNeasy FFPE extraction kits (Qiagen). Annotations from the pathologist performing H&E staining were used to guide removal of normal tissue from the slides by macrodissection before nucleic acid extraction, which occurred after tissue deparaffinization and lysis. Each extracted RNA sample was independently quantified using a NanoDrop spectrophotometer (NanoDrop Technologies) and qualified (Agilent Bioanalyzer). Degradation was quantified as the percentage of RNA fragments smaller than 300 bp using the RNA 6000 Nano kit (Agilent Bioanalyzer). Good sample quality was defined as less than 50% of RNA fragments being 50–300 bp in size. RNA expression profiling was performed using the nCounter PanCancer Immune Profiling Panel from NanoString (NanoString Technologies). The PanCancer Immune Profiling Panel contains 776 probes and is supplemented with six genes to complete HalioDx Immunosign targets. Hybridization was performed according to the manufacturer’s instructions. Hybridized probes were then purified and immobilized on a streptavidin-coated cartridge using the nCounter Prep Station (NanoString Technologies). Data collection was carried out on the nCounter Digital Analyzer (NanoString Technologies) following the manufacturer’s instructions to count individual fluorescent barcodes and quantify target RNA molecules present in each sample. For each assay, a scan of 490 fields of view was performed. Raw data from the NanoString nCounter were processed using NanoString recommendations. Quality control enables to retain data of good quality with a binding density that ranges between 0.05 and 2.25. The linearity of positive controls was checked using the R2 value of regression between the counts and the concentrations of positive controls. Samples that showed R2 < 0.75 were flagged and removed from the analysis. The background was removed using a thresholding method at the mean + 2 s.d. of negative controls. Raw counts were normalized using a positive normalization factor. Samples showing positive normalization factors outside the range of 0.3–3 were removed from the analysis. A second normalization was performed using the housekeeping gene normalization factor. Only the most stable housekeeping genes were selected for this normalization step using the variance-versus-mean relationship. All samples showing a normalization factor outside the range of 0.1–10 were removed from the analysis. All statistical analyses were performed on normalized counts using R software (version 2.6.2, 2019-12-12). T cell receptor variable β chain sequencing To track longitudinal immune responses to therapy, genomic DNA was extracted from longitudinal pre- and post-treatment PBMCs (five patients), pre- and post-treatment biopsies (five patients) (both FFPE and treated with RNAlater) and IDO- and PD-L1-specific T cell cultures from PBMCs (five patients) or SKILs (one patient). Three clonal (two IDO- and one PD-L1-) specific cultures were also sequenced to confirm clonal purity of cultures. DNA from PBMCs or RNAlater-treated biopsies was extracted with the DNeasy Blood and Tissue kit (Qiagen, 69504), DNA from sorted IDO and PD-L1-specific T cells from either PBMCs or SKILs was extracted using the QIAamp DNA Micro kit (Qiagen, 565304), and DNA from PFFE biopsies was extracted using the Maxwell RSC DNA FFPE kit (Promega, AS1450). Immunosequencing of the CDR3 regions of human TCRβ chains was performed using the immunoSEQÒ Assay (Adaptive Biotechnologies). Extracted genomic DNA was amplified in a bias-controlled multiplex PCR, followed by high-throughput sequencing. Sequences were collapsed and filtered to identify and quantitate the absolute abundance of each unique TCRβ CDR3 region for further analysis as previously described55–57. Statistical analyses of TCRβ sequencing results Two quantitative components of diversity were compared across samples in this study. First, Simpson clonality was calculated on productive rearrangements by ∑i=1Rpi2, where R is the total number of rearrangements and pi is the productive frequency of rearrangement i. Values of Simpson clonality range from 0 to 1 and measure how evenly receptor sequences (rearrangements) are distributed among a set of T cells. Clonality values approaching 0 indicate a very even distribution of frequencies, whereas values approaching 1 indicate an increasingly asymmetric distribution in which a few clones are present at high frequencies. Second, sample richness was calculated as the number of unique productive rearrangements in a sample after computationally downsampling to a common number of T cells to control for variation in sample depth or T cell fraction. Repertoires were randomly sampled without replacement five times, and we report the mean number of unique rearrangements. The T cell fraction was calculated by taking the total number of T cell templates and dividing by the total number of nucleated cells. The total number of nucleated cells was derived from reference genes using the immunoSEQ Analyzer version 3. To identify enriched vaccine clones in each patient, rearrangement frequencies in their baseline PBMCs and each IDO/PD-L1-sorted T cell sample were compared using a binomial distribution framework as previously described58. In brief, for each clone, we performed a two-sided test to determine whether frequencies were the same in the patient’s periphery and a PD-L1- or IDO-specific T cell sample. The Benjamini–Hochberg procedure was used to control the false discovery rate at 0.01 (ref. 59). Clonal expansion in post-treatment samples was similarly assessed using this differential abundance framework, but an IDO/PD-L1-specific T cell sample was replaced with a post-treatment series sample. In biopsies, the six series frequencies were compared to those of baseline tissue. Lastly, vaccine-associated clones were tracked in each PBMC and tissue sample by summing the frequency of each rearrangement enriched in either PD-L1- or IDO-specific T cells. All statistical analyses were performed in R version 3.6.1. Reporting Summary Further information on research design is available in the Nature Research Reporting Summary linked to this article. Online content Any methods, additional references, Nature Research reporting summaries, source data, extended data, supplementary information, acknowledgements, peer review information; details of author contributions and competing interests; and statements of data and code availability are available at 10.1038/s41591-021-01544-x. Supplementary information Supplementary Information Supplementary Figs. 1–3 and Tables 1–6 Reporting Summary Extended data Extended Data Fig. 1 Mode of Action and treatment plan. a) Anticipated mechanism of action of the combination therapy of an IDO/PD-L1 derived peptide vaccine and nivolumab (anti-PD1). 1) The IDO/PD-L1 peptide vaccine is administered subcutaneously (s.c) and nivolumab is administered intravenous (IV). 2) The vaccines peptides are phagocytosed by an antigen presenting cell and presented to IDO and PD-L1 specific T cells, which are activated. 3) The activated T cells migrates to the tumor site where they attack both immune-suppressive cells and tumour cells expressing IDO and/or PD-L1 leading to cytokine production and a pro-inflammatory tumour microenvironment. 4/5) Enhanced tumour killing by both IDO/PD-L1 specific T cells and tumor specific cytotoxic T cells due to PD-1 blockade. Created with BioRender.com b) Treatment plan. After written informed content patients were screened. Before treatment start a baseline PET/CT scan was performed, baseline blood sample for research use and if assessable a baseline needle biopsy. Patients were treated with the IDO/PD-L1 peptide vaccine subcutaneously biweekly for the first 6 injections and thereafter every fourth week for a maximum of 15 vaccines. Nivolumab was administered in parallel biweekly (3mg/kg) up to 24 series. If patients needed subsequent nivolumab after ended vaccination regiment they were treated with 6 mg/kg every fourth week up to two years. Needle biopsy and delayed type hypersensitivity (DTH) was performed after 6 series of treatment if assessable. PET/CT scans was performed every third month. Extended Data Fig. 2 Progression free survival and overall survival in matched historical control group and vaccine injection site reaction. a) Kaplan-Meier curve of progression free survival with corresponding 95% confidence intervals in the matched historical control group (n=74). Patients were matched on BRAF-status, PD-L1-status, age, gender, M-stage and LDH level). b) Kaplan-Meier curve of overall survival with corresponding 95% confidence intervals in the matched historical control group (n=74). c) Images of injection site reaction in patient MM20 (CR) after 11 vaccination show redness, rash and granuloma at injection site. Extended Data Fig. 3 Vaccine specific responses in blood. a) Heatmap of specific (background has been subtracted) IDO and PD-L1 responses in PBMCs at baseline, series 3, 6, 12, 18 and 24 measured by in vitro IFN-γ Elispot assay show fluctuations in the blood during treatment (n=30). b) Heatmap of specific (background has been subtracted) IDO and PD-L1 responses in PBMCs at baseline and 3 and 6 months after last vaccine measured by in vitro IFN- γ Elispot assay (n=30). 2.5–3.2x105 cells per well were used. c) Vaccine associated clones were tracked in the blood by summing the frequency of each rearrangement enriched in either IDO or PD-L1 T cells. Extended Data Fig. 4 Ex vivo vaccine specific responses in blood. Heatmaps of detected specific (background has been subtracted) IDO (a) and PD-L1 (b) responses in PBMCs at baseline and on/after treatment as measured by IFNγ ELISPOT (n=25). C) Example of well images of ex vivo ELISPOT wells for three different patients (n=3). 6-9x105 cells per well were used. Extended Data Fig. 5 CD4 and CD8 vaccine specific T cell responses in blood. Top: heatmaps of IDO specific CD4 (left) and CD8 (right) T cell responses in PBMCs at baseline and on/post treatment. Bottom: heatmaps of PD-L1 specific CD4 (left) and CD8 (right) T cell responses. Responses quantified by flow cytometry by an increased expression of CD107α, CD137 and TNFα after 8h peptide stimulation. Values represent specific responses after background values have been subtracted (n=21). Statistical analysis were performed using two-sided Wilcoxon matched-paired rank test. Extended Data Fig. 6 Pro-inflammatory profiles of sorted CD4 and CD8 T cells from blood. a/b/c/d) Percentage of in vitro stimulated and sorted PD-L1 and IDO specific CD4 and CD8 T cells secreting cytokines. Extended Data Fig. 7 Vaccine specific responses in skin. a) IDO and PD-L1 specific T cell responses in SKILs after 6 series of treatment measured by IFN-γ Elispot assay (n=13). SKILs were grown from DTH injection with either IDO peptide, PD-L1 peptide or a mix as presented by different blue colours. * Responses were calculated as the difference between average numbers of spots in wells stimulated with IDO or PD-L1 peptide (triplicates) and corresponding control (DMSO) and statistical analyses of Elispot responses were performed using distribution-free resampling method (Moodie et al.). DR: Not statistically confirmed response due to replicate number but number of spots in peptide wells are two times higher than control wells (DMSO). NS: No significant response and no DR. b/c/d) Percentage of cytokine secreting/CD107a+ CD4+ and CD8+ IDO and PD-L1 specific T cells in response to in vitro peptide stimulation by flow cytometry. e) Skin infiltrating PD-L1 specific T cell clones also found in biopsy in patient MM01. TCR sequencing was performed on a PD-L1 specific T cell culture generated from DTH area on the lower back injected with PD-L1 peptide on patient MM01. Bars show the frequency of top 25 clones in the culture with which indicates a high Simpson clonality of 0.43. f) Tracking the frequency of the top five skin infiltrating PD-L1 specific clones in tumour before and after treatment. Extended Data Fig. 8 T cell changes in blood after treatment. T cell fraction, TCR clonality and repertoire richness in blood and peripherally expanded TCR clones in both responding and non-responding patients. a) T cell fraction in peripheral blood of five patients at baseline, series 3, 6 and 12 by TCR sequencing. T cell fraction was calculated by taking the total number of T cell templates and dividing by the total number of nucleated cells (n=5.) b) TCR clonality in peripheral blood of five patients at baseline, series 3, 6 and 12 by TCR sequencing (n=5). Simpson clonality measures how evenly TCR sequences are distributed amongst a set of T cells where 0 indicate even distribution of frequencies and 1 indicate an asymmetric distribution where a few clones dominate. c) TCR repertoire richness in peripheral blood of five patients at baseline, series 3, 6 and 12 by TCR sequencing (n=5). TCR repertoire richness report the mean number of unique rearrangements. d) Bar-chart chart representing dynamics of the expanded T cell clones. Coloured bars represent peripherally expanded clones in five patients at series 3, series 6 and series 12 as compared to the baseline PBMC samples (n=5). Light gray bar represents peripherally expanded clones that are present in baseline biopsy samples while dark gray bar represents peripherally expanded clones that are present in the post-treatment biopsy (after 6 series). e) Frequency of the dominant TCR β chain in clonal PD-L1 and IDO specific cultures as determined by CDR3 sequencing. Extended Data Fig. 9 Profiling of genes relevant to T cell activation, cytokines and exhaustion markers on pre- and post-treatment biopsies in two patients. a) RNA expression profiling of genes related to T cell activation was performed using NanoString nCounter (n=2). b) RNA expression profiling of genes related to cytokine activity was performed using NanoString nCounter (n=2). c) RNA expression profiling of genes related to checkpoint inhibitors was performed using NanoString nCounter (n=2). Extended Data Fig. 10 Treatment induced upregulation of PD-L1, IDO, MHC I and MHC II and distance between CD8 T cells and PD-L1 expressing cells. a) IHC on 4 paired biopsies stained for CD3 and CD8 T cells, PD-L1, MHC I and MHC II on tumor cells/mm2 (sum in the validated area) and IDO H-score (from 0 to 300). H-score is the expression of IDO on both immune and tumor cells: The score is obtained by the formula: 3 x percentage of cells with a strong staining + 2 x percentage of cells with a moderate staining + percentage of cells with a weak staining (n=4). b) Distance in µm between CD8+ T cells and PD-L1+ stained cells on five baseline biopsies detected by IHC (n=5). Extended data is available for this paper at 10.1038/s41591-021-01544-x. Supplementary information The online version contains supplementary material available at 10.1038/s41591-021-01544-x. Acknowledgements We thank all patients and their relatives for being a part of the trial. We thank Ö. Met, M. Jonassen, S. Ullitz Færch, B. Saxild, S. Wendt and C. Grønhøj for technical support. We thank M. Cumberbatch for input with translational analysis. We thank the nurses at clinic 5 and the head of the Oncology Department at Herlev and Gentofte Hospital, L. Sengeløv. The study was funded through a research funding agreement between IO Biotech and the CCIT-DK, Herlev Hospital and the Oncology Department at Herlev Hospital. Author contributions I.M.S. and M.H.A. conceived the study. J.W.K. wrote the protocol with input from I.M.S., M.H.A. and M.-B.Z. J.W.K. and C.L.L. were responsible for patient treatment and patient care with help from R.B.H. and consulting from I.M.S., E. Ellebaek and M.D. I.M.S., E. Ellebaek and M.D. were responsible for patient recruitment. J.W.K. and C.L.L. were responsible for coordinating trial procedures and collecting data and samples. J.W.K. and C.L.L. were clinical investigators, and I.M.S. was a sponsor. J.W.K. and C.L.L. analyzed and interpreted clinical data with support from I.M.S. and E. Ehrnrooth. T.W.K. performed statistical analyses. H.W.H. helped with clinical imaging. J.W.K., E.M., C.O.M., S.M.A., S.E.W.-B., M.O.H. and A.W.P. analyzed and interpreted translational data with support from M.H.A. and I.M.S. Funding was acquired by I.M.S., M.H.A. and M.-B.Z. J.W.K. wrote the manuscript with input from I.M.S., M.H.A., E.M. and A.W.P. All authors reviewed the manuscript, interpreted data and approved the final version. Data availability This clinical trial was registered at https://www.clinicaltrials.gov/ before patient enrollment (clinical trial identifier NCT03047928). Each vaccine was composed of 100 µg IO102, a 21-amino-acid peptide (DTLLKALLEIASCLEKALQVF) from the IDO peptide, and 100 µg IO103, a 19-amino-acid peptide (FMTYWHLLNAFTVTVPKDL) from the signal peptide of PD-L1. TCR sequencing data are available from Adaptive Biotechnologies. Upon request, the CCIT-DK office will provide a username and a password to access the designated data within approximately 2–4 weeks (https://www.herlevhospital.dk/ccit-denmark/find-us/Sider/Contact-information.aspx). All requests for the remaining data including raw data and analyzed data and materials will, within a reasonable time frame, be reviewed by the CCIT-DK office (https://www.herlevhospital.dk/ccit-denmark/find-us/Sider/Contact-information.aspx) to verify whether the request is subject to any intellectual property or obligations. Patient-related data not included in the paper were generated as part of clinical trials and may be subject to patient confidentiality. Any data and materials that can be shared will be released via a material-transfer agreement. The following database was used in the study: https://research.regionh.dk/da/publications/the-danish-metastatic-melanoma-database-dammed(32749d99-095f-4cae-b5de-769bae27f01e).html. Competing interests M.H.A. is named as an inventor on various patent applications relating to therapeutic uses of IDO and PD-L1 peptides. These patent applications are assigned to the company IO Biotech, which is developing immune-modulating cancer treatments. M.H.A. is a founder, shareholder and advisor for IO Biotech. I.M.S. is a cofounder, shareholder and advisor for IO Biotech. I.M.S. has an advisory board relationship with or lectured for Roche, Novartis, MSD, Celgene, Incyte, TILT Bio, Pfizer and BMS AstraZeneca and has received limited grants for translational research from BMS, Roche and Novartis. M.-B.Z. is the CEO, founder and shareholder at IO Biotech. E.M., A.W.P. and E. Ehrnrooth are employees at IO Biotech. E. Ellebaek has received honoraria from BMS, Pierre Fabre, Roche and Kyowa Kirin and travel support from MSD. M.D. has received honoraria from Genzyme, MSD, BMS, Roche and Novartis and travel support from Novartis, MSD, BMS, Roche and Pfizer. The remaining authors declare that they have no conflict of interest. Peer review information Nature Medicine thanks George Coukos, Michael Schell and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. Javier Carmona was the primary editor on this article and managed its editorial process and peer review in collaboration with the rest of the editorial team. Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. These authors contributed equally: Julie Westerlin Kjeldsen, Cathrine Lund Lorentzen. Change history 3/8/2022 A Correction to this paper has been published: 10.1038/s41591-022-01771-w

NIVOLUMAB

DrugsGivenReaction

CC BY

34887574

2021-12-09

Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Immune-mediated enterocolitis'.

A phase 1/2 trial of an immune-modulatory vaccine against IDO/PD-L1 in combination with nivolumab in metastatic melanoma. Anti-programmed death (PD)-1 (aPD1) therapy is an effective treatment for metastatic melanoma (MM); however, over 50% of patients progress due to resistance. We tested a first-in-class immune-modulatory vaccine (IO102/IO103) against indoleamine 2,3-dioxygenase (IDO) and PD ligand 1 (PD-L1), targeting immunosuppressive cells and tumor cells expressing IDO and/or PD-L1 (IDO/PD-L1), combined with nivolumab. Thirty aPD1 therapy-naive patients with MM were treated in a phase 1/2 study ( https://clinicaltrials.gov/ , NCT03047928). The primary endpoint was feasibility and safety; the systemic toxicity profile was comparable to that of nivolumab monotherapy. Secondary endpoints were efficacy and immunogenicity; an objective response rate (ORR) of 80% (confidence interval (CI), 62.7-90.5%) was reached, with 43% (CI, 27.4-60.8%) complete responses. After a median follow-up of 22.9 months, the median progression-free survival (PFS) was 26 months (CI, 15.4-69 months). Median overall survival (OS) was not reached. Vaccine-specific responses assessed in vitro were detected in the blood of >93% of patients during vaccination. Vaccine-reactive T cells comprised CD4+ and CD8+ T cells with activity against IDO- and PD-L1-expressing cancer and immune cells. T cell influx of peripherally expanded T cells into tumor sites was observed in responding patients, and general enrichment of IDO- and PD-L1-specific clones after treatment was documented. These clinical efficacy and favorable safety data support further validation in a larger randomized trial to confirm the clinical potential of this immunomodulating approach. pmcMain Despite remarkable advances in the treatment of MM with immune checkpoint inhibitors (ICIs) targeting PD-1 and cytotoxic T lymphocyte antigen 4 (CTLA-4), around half of patients are resistant to ICI monotherapy1. The combination of anti-CTLA-4 (aCTLA-4) and aPD1 therapy is to date the most effective therapy resulting in a response rate of around 60%; however, 50% of the patients also develop severe adverse events2,3. Therefore, an equally effective but less toxic treatment is highly needed. Several approaches to enhance ICI efficacy are currently being investigated, such as other ICIs, T cell therapy with tumor-infiltrating T cells or innate immunity stimulators such as Toll-like receptor 9 agonists4–6. Treating cancer patients with vaccines that stimulate a targeted immune response is another attractive approach, with very few side effects observed thus far in combination immunotherapy studies7,8. Immune-modulatory vaccines targeting tumoral immune escape mechanisms offer a new, generalizable strategy compared to patient-specific neoantigen cancer vaccines7,9. The immune-modulatory vaccine strategy in this clinical trial is based on the finding of circulating cytotoxic T cells specific to IDO and PD-L1 in the blood of patients with cancer and, to a lesser extent, in healthy donors. IDO- and PD-L1-specific CD8+ T cells can directly recognize and kill IDO+ and/or PD-L1+ tumor cells and likewise recognize and kill non-malignant cells that express their cognate targets. Furthermore, IDO- and PD-L1-specific CD4+ T cells release pro-inflammatory cytokines in response to IDO- or PD-L1-expressing target cells. IDO and PD-L1 are expressed not only by melanoma cells but also by many other cell types in the tumor microenvironment (TME), which differentiates these antigens from traditional tumor antigens used in other studies10–15. Activation of IDO/PD-L1-specific T cells by vaccination can therefore restrict the range of immunosuppressive signals mediated by immunosuppressive cells and thereby revert the TME from an immune hostile to an immune friendly environment. In animal models of cancer, vaccinations with IDO epitopes resulted in anti-tumor therapeutic effects that were correlated with reductions in IDO expression in myeloid cell populations within the TME16. The IDO/PD-L1 immune-modulating vaccine may lead to a translatable strategy for improving the efficacy of aPD1 therapy through activation of specific T cells. We hypothesize that the IDO/PD-L1 vaccine attracts T cells into the tumor, which induces type 1 helper T (TH1) cell inflammation and reverts the TME into an immune-permissive site, thereby turning the tumor ‘hot’. This would also upregulate PD-L1 expression in cancer and immune cells, generating more susceptible targets to aPD1 therapy (Extended Data Fig. 1a). This theory was confirmed in a mouse model in which aPD1 therapy and IDO vaccination show synergistic effects16. In this phase 1/2 clinical trial, MM1636, patients with MM received a combination of the IDO/PD-L1 (IO102/IO103) peptide vaccine with the adjuvant Montanide and the aPD1 antibody nivolumab. Patients were included in three cohorts: 30 aPD1 therapy-naive patients (cohort A), ten aPD1 therapy-refractory patients (cohort B, de novo resistance) and ten patients who progressed after aPD1 therapy (cohort C, acquired resistance). Here, we report results from cohort A. The vaccine was given biweekly for the first six administrations and thereafter every 4th week. A maximum of 15 vaccines were administered. Nivolumab was given in parallel, biweekly (3 mg per kg) or every 4th week (6 mg per kg) for up to 2 years (Extended Data Fig. 1b). The primary objective was safety and feasibility. Secondary objectives were immunogenicity and clinical efficacy. Results Patients and treatment Thirty patients were enrolled from December 2017 to June 2020. None of the 30 patients dropped out of the study; all received at least three cycles of therapy (Supplementary Fig. 1). At the current database lock (5 October 2020), six patients were still on treatment in the trial. Of the 24 patients who were not on trial treatment at data cutoff, two are still receiving nivolumab monotherapy (6 mg per kg every 4 weeks). Reasons for stopping treatment for the remaining 22 patients included disease progression (37%), toxicity (20%), maximum benefit or complete response (CR) confirmed on two consecutive scans (17%) or completing 2 years of treatment (7%). For the 24 patients who were not on trial treatment at data cutoff, the mean number of vaccinations was 10.5 (range, 3–15 vaccinations). Thirteen of these 24 patients continued nivolumab (6 mg per kg, every 4 weeks) as a standard of care. Nine patients received subsequent therapy after progression (Supplementary Table 1). Baseline characteristics are shown in Table 1. The mean age was 70 years; 37% of patients had elevated lactate dehydrogenase (LDH) levels, 60% were stage M1c, 37% had BRAF mutations, and 43% were negative for PD-L1 (<1%). A total of three patients (10%) had received prior ipilimumab therapy. No patients had brain metastasis (Supplementary Table 2).Table 1 Baseline patient characteristics (n = 30) Characteristic Number (%) Mean age, years (range) 70 (46–85) Sex, male 16 (55%) ECOG PS 0 26 (87%) LDH levels  ≤ULN 19 (63%)  >ULN 11 (37%) M stage (AJCC-8)  M1a 6 (20%)  M1b 6 (20%)  M1c 18 (60%) Number of lesion sites  1 6 (20%)  2–3 17 (57%)  >3 7 (23%) Liver metastases present 10 (33%) BRAF status  Mutant 11 (37%)  Wild type 19 (63%) PD-L1  <1% 13 (43%)  >1% 17 (57%) Previous systemic therapy  Ipilimumab 3 (10%)  No 27 (90%) AJCC-8, eighth edition of the American Joint Committee on Cancer; ECOG PS, Eastern Cooperative Oncology Group performance status; ULN, upper limit of normal. Notable clinical responses to the combination therapy Thirty patients with MM were treated with the IDO/PD-L1 vaccine and nivolumab according to the trial protocol. By investigator review, the ORR reached 80% (CI, 62.7–90.5%), with 43% of patients (CI, 27.4–60.8%) achieving a CR and 37% (CI, 20.9–54.5%) reaching a partial response (PR) as the best overall response, while 20% experienced progressive disease (PD) according to Response Evaluation Criteria in Solid Tumors (RECIST) 1.1 (Fig. 1a). Two of the patients with a PR did not have a confirmatory scan stating PR on two consecutive scans. Early onset of response was frequent, with 22 of 30 patients having an objective response at the first evaluation (after 12 weeks on treatment). Median times to PR and CR were 75 d (range, 54–256 d) and 327 d (range, 73–490 d), respectively (Fig. 2a–c).Fig. 1 Clinical response. a, Pie charts with percent ORR, CR, PR and PD according to RECIST 1.1 by investigator review of all patients (n = 30), PD-L1+ patients (>1%, (n = 17)) and PD-L1− patients (<1%, n = 13)), respectively. Two-sided CIs (95%) were constructed using the Clopper–Pearson method. b, Treatment effect in MM1636 compared with a matched historical control group from the DAMMED database (n = 74). Patients in MM1636 (n = 29) were matched with the exact same combination variable according to age (≤70 years, >70 years), sex, LDH levels (normal, elevated), M stage (M1a, M1b, M1c), BRAF status (wild type, mutated) and PD-L1 status (<1%, ≥1%). Estimates for treatment effects were calculated by weighted logistic regression analyses and weighted Cox proportional hazard model. Bar height indicates the estimated response rate; tops of bars are centers for error bars. Odds ratios (OR), response rates and their corresponding 95% CIs were extracted from the regression model. All P values were two sided, and P values below 0.05 were considered statistically significant. c, Best change in the sum of target lesion size compared with that at baseline (n = 30). The horizontal line at −30 shows the threshold for defining an objective response in the absence of non-target disease progression or new lesions according to RECIST 1.1. Two patients with 100% reduction in target lesion size had non-target lesions present. White stars, six patients had normalization (<10 mm) of fluor-18-deoxyglucose (FDG)-negative lymph nodes (at baseline, lymph nodes were >1.5 cm and FDG+) and 100% reduction of non-lymph node lesions and are considered to have had a CR (green bar). Black star, one patient (MM29) was considered to have had a PR (blue bar), although he had not reached a −30% change in target lesion size. The patient had a single measurable 13-mm lung metastasis at baseline and multiple biopsy-verified cutaneous metastases on the left crus (not detectable by positron-emission tomography–computed tomography (PET–CT) at baseline). The best change in target lesion size was 10 mm, and a post-treatment biopsy from the cutaneous metastases showed no sign of malignancy. Thus, overall, the patient was classified as having a PR. d, Kaplan–Meier curve of the response duration in the 24 patients with an objective response. e, Kaplan–Meier curve of PFS in all 30 treated patients. f, Kaplan–Meier curve of OS in all 30 treated patients. Fig. 2 Clinical response. a, Swimmer plot showing response duration and time to response according to RECIST 1.1 for all treated patients (n = 30). Triangles indicate first evidence of PR, while squares indicate first evidence of CR. Closed circles indicate time of progression. Arrows indicate ongoing responses. Patient MM18 died due to nivolumab-induced side effects. b, Spider plot showing response kinetics in all treated patients (n = 30). Red squares indicate time of progression. c, PET–CT images of patient MM42 before and after treatment (after 12 series of treatment) showing FDG metabolism in target lesions. SC, subcutaneous. The ORR among PD-L1+ (>1% (clone 28.8)) patients (n = 17) was 94.1% (CI, 73–99.7%) and 61.5% (CI, 35.5–82.3%) in PD-L1− patients (n = 13) (Fig. 1a). Objective responses were observed in patients irrespective of human leukocyte antigen (HLA) genotype (Supplementary Fig. 2). Clinical response data were validated by blinded independent external review, in which an ORR of 76.6% (CI, 57.7–90.1%) was reported, with 53.3% of patients achieving a CR, 23.3% achieving a PR and 3.3% experiencing stable disease. Comparisons between investigator review and external review are outlined in Supplementary Table 3. To examine whether the very high response rate should be attributed to nivolumab or to the vaccine, we retrieved clinical response information from a matched historical control group using the Danish Metastatic Melanoma Database (DAMMED) from contemporaneously treated patients with stage 3–4 melanoma who received aPD1 monotherapy17. Patients were matched with the exact same combination variable according to age, sex, PD-L1 status, BRAF status, LDH level and M stage (those at stage M1d were excluded from the control group (no patients with brain metastasis)). Matched controls were identified for 29 patients, and the ORR of 79.3% (CI, 61.0–90.4%) observed in MM1636 was found to be significantly higher (P < 0.0012) than that in the matched control group in which an ORR of 41.7% (CI, 31.0–53.3%) was reached. Furthermore, of the 29 patients in MM1636, a significantly (P < 0.0017) higher percentage (41.4% (CI, 25.2–59.6%)) of patients achieved CR in MM1636 than that (12% (CI, 6.3–21.6%)) in the matched historical control group. ORR and complete response rate (CRR) in the matched historical control group were comparable to those of patients treated in randomized phase 3 pivotal trials with aPD1 monotherapy18 (Fig. 1b). The treatment leads to prolonged PFS At data cutoff, the median duration of response had not been reached, with 87% of all responding patients being free from progression at 12 months (Fig. 1d). Patients were followed for up to 35 months with a median follow-up time of 22.9 months (CI, 14.9–26.2 months). OS and PFS were calculated from the first day of treatment to death or progression or to the date of the last follow-up (5 October 2020). The median PFS (mPFS) for all treated patients was 26 months (CI, 15.4–69 months) and was not reached for responding patients (Fig. 1e). The median OS was not reached at the data cutoff. OS at 12 months was 81.6% (CI, 61.6–92%) (Fig. 1f). One patient (MM18) with a CR died from nivolumab-related severe adverse events; the remaining patients died because of metastatic melanoma. For comparison, the mPFS was 8.3 months (CI, 5.5 months–NR (not reached)) in the matched historical control group (n = 74), while the median OS was 23.2 months (CI, 23.2 months–NR). (Extended Data Fig. 2a,b). The combination of the IDO/PD-L1 vaccine and nivolumab was safe Treatment-related adverse events are listed for all 30 patients in Supplementary Table 4. Common treatment-related grade 1–2 toxicities were fatigue (47%), rash (47%), arthralgia (30%), diarrhea (23%), nausea (23%), dry skin (20%), pruritus (20%), infusion reaction (17%), xerostomia (17%) and myalgia (17%). Four patients (13%) experienced grade 3–4 adverse events: one patient with a grade 3 maculopapular rash (MM01), one patient with grade 3 adrenal insufficiency (MM06) and one patient with grade 3 arthralgia (MM22). Patient MM18 died from urosepsis with multi-organ failure and severe hyponatremia. This patient had experienced multiple immune-related adverse events with grade 3 colitis, grade 2 pneumonitis, grade 3 arthralgia, grade 2 vasculitis and grade 2 nivolumab infusion-related allergic reaction. Additionally, patient MM18 had symptoms of myocarditis at the time of death with highly elevated cardiac troponin I levels. Bedside echocardiography showed an ejection fraction of 15%, which at baseline was 60%, but an autopsy was not conducted, and myocarditis was never confirmed pathologically. Patient MM06 had received first-line treatment with ipilimumab before entering the trial and was on substitution corticosteroids at the time of inclusion. Adrenal insufficiency was aggravated by an erysipelas infection with high fever, reaching grade 3 in Common Terminology Criteria for Adverse Events and resolving quickly after appropriate antibiotic therapy was initiated. As expected, local side effects were common with 77% of the patients who developed injection site reactions. These reactions were classified as granulomas (63%), redness (20%), pain (13%) and pruritus (13%) at the injection site. All local reactions were grade 1–2, most likely related to the Montanide adjuvant and typically transient. However, two patients (MM07 and MM20) decided to discontinue vaccination after eight and 11 injections, respectively, due to granulomas, tenderness and pain that limited instrumental activities of daily living but continued to receive nivolumab (Extended Data Fig. 2c). Vaccine-specific responses in blood were frequently detected First, all 30 patients were assessed for the presence of vaccine-specific responses in peripheral blood mononuclear cells (PBMCs) before, on and after vaccination using a modified interferon (IFN)-γ enzyme-linked immune absorbent spot (ELISPOT) assay. This assay is known to expand antigen-specific memory cells and to improve the detection and the correlative power of ELISPOT during treatment19–21. Prevaccine IDO-specific responses were detectable in ten (33%) patients, while prevaccine PD-L1-specific responses were detectable in eight patients (27%); overlapping (specific to both IDO and PD-L1) prevaccine responses were present in four (13.3%) patients. During vaccination, an increase of IDO-specific T cells or PD-L1-specific T cells in the blood was observed in 28 (93%) and 26 (86%) patients, respectively. In total, 93% of patients had an increase in either PD-L1- or IDO-specific responses on vaccination (Fig. 3a), with a significant (P < 0.0001) median increase from baseline to the post-vaccine response demonstrated for both IDO and PD-L1 (at different time points on treatment), confirming that vaccine-specific immune responses were induced in patients regardless of clinical response (Fig. 3b,c). Immune responses fluctuated in the blood over time (Extended Data Fig. 3a). An increase in IDO- and PD-L1-specific responses in the peripheral blood was also detectable directly ex vivo across the different clinical response groups (Extended Data Fig. 4).Fig. 3 Vaccine-specific responses in blood. a, IDO- and PD-L1-specific T cell responses in PBMCs at baseline and on vaccination as measured by the IFN-γ ELISPOT assay (n = 30). *Responses were calculated as the difference between the average numbers of spots in wells stimulated with IDO or PD-L1 peptide (triplicates) and those from the corresponding control (DMSO), and statistical analyses of ELISPOT responses were performed using a distribution-free resampling method (Moodie et al.50). DR (double response), response was not statistically confirmed due to replicate number, but the number of spots in peptide wells was two times higher than that in control wells (DMSO). NS, no significant response and no DR. For a detailed overview of responses at serial time points on vaccination, see Extended Data Fig. 4a. b, IDO- and PD-L1-specific T cell response in PBMCs in all treated patients measured by the IFN-γ ELISPOT assay at baseline and on vaccination. On-vaccination responses were selected from the ‘best’ ELISPOT response at different time points for each patient (series 3, 6, 12, 18 or 24) during vaccination (n = 30). Wilcoxon matched-pairs signed-rank test was used to compare responses to IDO or PD-L1 peptides in the vaccine between baseline and later time points. c, Representative example of ELISPOT wells with response in patient MM23 in serial PBMCs before and on treatment. d, IDO-specific CD4+ and CD8+ T cells were isolated and expanded from PBMCs stimulated in vitro with the IDO peptide and a low dose of IL-2 for 14–15 d before sorting using the Miltenyi Cytokine Secretion Assay—Cell Enrichment and Detection kit. To assess their cytolytic potential, IDO-specific T cells were stimulated with IDO peptide, and expression of CD107a, IFN-γ and TNF-α was assessed by flow cytometry (the example is from patient MM14). Sustained vaccine-specific responses were observed 3 and 6 months after the last vaccine, indicating induction of memory responses in nine patients with response to clinical treatment who surpassed follow-up at data lock (Extended Data Fig. 3b). Importantly, PD-L1- and IDO-specific responses were observed irrespective of HLA genotype (Supplementary Table 5). To verify the functionality of vaccine-induced T cells, IDO- or PD-L1-specific T cells were isolated and in vitro expanded from PBMCs of five patients. Phenotypic characterization by flow cytometry revealed that the isolated vaccine-specific T cells consisted of both CD4+ and CD8+ T cells. Also, both IDO- and PD-L1-specific CD4+ and CD8+ T cells showed pro-inflammatory properties, as they expressed the cytolytic marker CD107a and secreted the cytokines IFN-γ and tumor necrosis factor (TNF)-α (Fig. 3d and Extended Data Fig. 6a–d). Interestingly, we were also able to detect vaccine-specific CD4+ and CD8+ T cell responses in peripheral blood ex vivo (Extended Data Figs. 4 and 5). We observed a significant increase in the overall percentage of CD107a, CD137 and TNF-α expression in response to peptide stimulation in on- or post-treatment PBMC samples compared to that at baseline, further confirming the expansion and the diverse signature of vaccine-specific T cells (Extended Data Fig. 5). Vaccine-specific responses detected in the skin at the vaccination site To investigate whether vaccine-specific T cells have the potential to migrate to peripheral tissue, delayed-type hypersensitivity (DTH) tests were performed after six cycles of treatment on 15 patients to assess the presence of vaccine-reactive T cells in the skin. Supplementary Table 6 display an overview of skin-infiltrating lymphocyte (SKIL) cultures. We detected IDO-specific T cells in the skin of six of ten patients and PD-L1-specific T cells in nine of 11 patients (Extended Data Fig. 7a). Intracellular cytokine staining was performed on SKILs from five patients after stimulation with either PD-L1 or IDO peptide. Here, we detected mainly CD4+ peptide-reactive T cells that secreted TNF-α and upregulated CD107a, and a minor fraction also secreted IFN-γ. In one patient, we detected CD8+ PD-L1-reactive T cells (Extended Data Fig. 7b–d). Vaccine-induced T cells specifically recognize target cells To confirm the functionality of vaccine-expanded T cells, vaccine-specific T cell clones (clonal purity was confirmed by T cell receptor (TCR) sequencing) were isolated and expanded from patient PBMCs (Extended Data Fig. 8e). We showed that PD-L1-specific T cells were able to recognize PD-L1+ autologous tumor cells in a PD-L1 expression-dependent manner if the cancer cells also expressed HLA-II (Fig. 4a,b). Similarly, an HLA-DR-restricted IDO-specific CD4+ T cell clone was able to recognize an HLA-DR-matched IDO-expressing model cell line, MonoMac1, in an IDO-expression-dependent manner (Fig. 4e–g). As previously described, IDO- and PD-L1-specific T cells’ mode of action is not limited to targeting only cancer cells. We were able to show that vaccine-specific T cell clones also reacted against PD-L1- and IDO-expressing autologous immune cells (Fig. 4c,h). To provide myeloid cells with a tumor-associated phenotype we treated isolated CD14+ myeloid cells with tumor conditioned medium (TCM) derived from an established autologous tumor cell line. We observed that such TCM-treated CD14+ cells had increased expression of PD-L1 and IDO and were effectively recognized by autologous PD-L1- and IDO-specific CD4+ T cell clones (Fig. 4c,d,h,i).Fig. 4 PD-L1- and IDO-specific T cells from vaccinated patients react against PD-L1- and IDO-expressing target cells. a, Left, PD-L1-specific T cell culture (MM1636.05) reactivity against PD-L1 peptide or autologous tumor cells in the IFN-γ ELISPOT assay. Tumor cells were either not treated or pretreated with 200 U ml−1 IFN-γ for 48 h before the assay. Effector:target (E:T) ratio of 10:1 was used. Right, PD-L1 and HLA-II surface expression on melanoma cells with (green) or without (yellow) pretreatment with IFN-γ compared to an isotype control (gray) as assessed by flow cytometry. b, Left, PD-L1-specific T cell (MM1636.05) reactivity in the IFN-γ ELISPOT assay against autologous tumor cells pretreated with IFN-γ (500 U ml−1) and transfected with mock or PD-L1 small interfering (si)RNA 24 h after transfection. E:T ratio, 10:1. Right, PD-L1 surface expression on melanoma tumor cells (MM1636.05) assessed by flow cytometry 24 h after transfection with mock (blue) or PD-L1 (red) siRNA compared to the isotype control (gray). c, Reactivity of the CD4+ PD-L1-specific T cell clone (MM1636.14) against PD-L1 peptide or autologous CD14+ cells; E:T ratio, 10:1. CD14+ cells were isolated using magnetic bead sorting and used as targets in an ELISPOT assay directly or after pretreatment for 2 d with TCM derived from the autologous tumor cell line. d, Quantitative PCR with reverse transcription (RT–qPCR) analysis of PD-L1 (CD274) expression in sorted CD14+ cells before and after treatment with autologous TCM for 48 h. e, Reactivity of the IDO-specific CD4+ T cell clone (MM1636.23) against IDO peptide combined with HLA-DR (L243)-, HLA-DQ (SPV-L3)- or HLA-DP (B7/21)-blocking antibodies (aHLA-DR, aHLA-DQ and aHLA-DP) in an intracellular staining assay (ICS) for IFN-γ and TNF-α production. T cells were incubated with individual blocking antibodies (2 μg ml−1) for 30 min before adding IDO peptide. f, Reactivity of the IDO-specific CD4+ T cell clone (MM1636.23) against the HLA-DR-matched IDO-expressing cell line MonoMac1 transfected with mock or IDO siRNA in an ICS assay, E.T ratio, 4:1. siRNA transfection was performed 48 h before the experiment. g, RT–qPCR analysis of IDO1 expression in MonoMac1 cells 48 h after siRNA transfection. h, Reactivity of the CD4+ IDO-specific T cell clone (MM1636.14) against IDO peptide or autologous CD14+ cells; E:T ratio, 20:1. CD14+ cells were isolated using magnetic bead sorting and used as targets in an ELISPOT assay directly or after pretreatment with TCM derived from the autologous tumor cell line. i, RT–qPCR analysis of IDO1 expression in sorted CD14+ cells before and after treatment with autologous TCM for 48 h. Bars in RT–qPCR data (d,g,i) represent the mean of three (d,i) or six (g) technical replicates ±s.d.; P values were determined by two-tailed parametric t-tests. ELISPOT counts (a,b,c,h) represent the mean value of three technical replicates ±s.e.m.; response P values were determined using the distribution-free resampling (DFR) method. TNTC, too numerous to count. T cell clones in blood and tumors To track the role of treatment-induced T cell responses, TCR sequencing of the complementarity-determining region 3 (CDR3) was performed on five patients in peripheral blood (baseline and cycles 3, 6 and 12) and paired biopsies. These five patients (MM01, MM02, MM08, MM09 and MM13) were selected due to the availability of material and to investigate a balanced patient group with both responders and non-responders. Due to a limited number of available paired biopsies (either as a consequence of patient refusal or unexpectedly rapid and substantial clinical responses), statistical analysis could not be applied. Details on clinical response are shown in Fig. 2a. Additionally, PBMCs (on treatment) or SKILs were stimulated with the IDO/PD-L1 peptides, and then cytokine-producing T cells were sorted to track vaccine-induced T cells both in the periphery and at the tumor site. To identify enriched IDO/PD-L1-specific T cell clones, TCR rearrangements in sorted IDO/PD-L1-specific clones and TCR rearrangements in sorted IDO/PD-L1-specific T cell samples were compared to sequences from baseline PBMC samples for each patient. Clonal expansion of vaccine-specific TCR rearrangements from samples on vaccination were then tracked using a differential abundance framework. Cumulative IDO/PD-L1-specific T cell frequencies were tracked in post-treatment samples. We found no relation between clinical response and the enrichment of vaccine-specific clones, but an increase in IDO/PD-L1-specific T cell clones was observed at different time points in the periphery in all five patients (Extended Data Fig. 3c). We next investigated overall changes in the T cell repertoire in the blood. A modest increase in the peripheral T cell fraction was observed in the three responding patients at cycle 3, while the two non-responding patients had a clear decrease in T cell fraction (Extended Data Fig. 8a). We thereafter investigated TCR clonality and TCR repertoire richness, exploring the proportion of abundant clones and the number of unique rearrangements, respectively. A decreasing peripheral Simpson clonality and increasing TCR repertoire richness was observed in responding patients at cycle 3, which might indicate tumor trafficking upon treatment. The opposite pattern was observed in non-responding patients (Extended Data Fig. 8b,c). Peripherally expanded clones were associated with tumors and persisted until cycle 12 (latest time point analyzed). The largest peripheral expansion was observed at cycle 3, with the most significant increase observed in patient MM01 (CR). Responding patients had a larger fraction of peripherally expanded clones that were also found in tumors compared to that of non-responders. By tracking peripherally expanded clones detected at the tumor site, we observed that patient MM01 had a substantial increase after treatment, indicating tumor trafficking of peripheral expanded clones (Extended Data Fig. 8d). Influx at the tumor site of vaccine-enriched T cell clones Given the observation of increased T cell fraction and enrichment of IDO- and PD-L1-specific clones in the blood after treatment, we investigated whether the same trend was observed at the tumor site. Both TCR sequencing and immunohistochemistry (IHC) of paired biopsies from the five patients described above showed an increase in the T cell fraction with an influx of CD3+ and CD8+ T cells after treatment in the three responding patients (Fig. 5a–c). IHC was not possible for patient MM09 due to tissue loss.Fig. 5 Changes in the TME after treatment. Number of CD3+ and CD8+ T cells, TCR fraction, TCR clonality, TCR repertoire, biopsy expanded TCR clones and enrichment of IDO- and PD-L1-specific T cells at the tumor site. a, Number of CD3+ and CD8+ T cells per mm2 (sum in the validated area) at the tumor site detected by IHC of paired biopsies from four patients. b, Example of IHC of CD3+ and CD8+ T cells at the tumor site before and after treatment (cells per mm2) in one patient (MM01). c, T cell fraction at the tumor site at baseline and cycle 6 by TCR sequencing. The T cell fraction was calculated by taking the total number of T cell templates and dividing by the total number of nucleated cells. d, Tracking of vaccine-associated clones at baseline and cycle 6 in tumor biopsies. Cumulative frequencies of IDO and PD-L1 vaccine-specific TCR rearrangements are represented. e,f, TCR clonality and TCR repertoire richness in five patients at the tumor site at baseline and cycle 6. Simpson clonality measures how evenly TCR sequences are distributed among a set of T cells, where 0 indicates an even distribution of frequencies and 1 indicates an asymmetric distribution in which a few clones dominate. TCR repertoire richness reports the mean number of unique rearrangements. g, Bar chart representing baseline expanded biopsy clones from five patients (colored bars) and the detection of biopsy expanded clones also found in the blood at baseline and series 3, 6 and 12 (white and gray bars) by TCR sequencing. We thereafter investigated whether some of the IDO/PD-L1 vaccine-linked T cells were present at the tumor site. Vaccine-associated clones were tracked as the combined frequency of IDO- and PD-L1-specific T cell rearrangements. In biopsies, the frequencies at cycle 6 were compared to those at baseline and showed an increase in vaccine-specific T cells in four of five patients, irrespective of clinical response (Fig. 5d). TCR sequencing of PD-L1-specific SKILs (a more specific culture than IDO/PD-L1-specific isolations derived from PBMCs on vaccination) and paired biopsies showed that two of the top five PD-L1-specific SKIL clones were present at the tumor site both before and after treatment (Extended Data Fig. 7e,f). With a focus on the more abundant T cell clones, we investigated overall TCR clonality at the tumor site before and after treatment. In addition, we explored the number of unique TCR rearrangements, dissecting the lower-frequency clones. Patient MM01 had a significant increase in TCR clonality and a decrease in repertoire richness at the tumor site after therapy, indicating a focused tumor repertoire response of selected clones. All three responding patients had a decrease in TCR repertoire richness, which again might indicate a focused tumor response (Fig. 5e,f). Deeper analyses showed that the T cell clones that expanded at the tumor site after therapy were also present in the blood at baseline and increased significantly after treatment in four of five patients. The highest proportion was detected early at cycle 3. These data again support trafficking of peripherally expanded clones to the tumor site and could indicate that the T cell response to treatment is derived from pre-existing peripherally tumor-associated T cells (Fig. 5g). Signs of treatment-induced inflammation in the TME To dissect changes in the TME induced by T cell influx upon treatment in responding patients, RNA gene expression analyses using the nCounter PanCancer Immune Profiling Panel from NanoString were performed on paired biopsies from two responding patients (MM01 and MM13). Expression of genes related to adaptive immunity such as T cell activation, effector functions (genes encoding IFN-γ, TNF-α, IL-15 and IL-18) and cytotoxicity was increased in post-treatment biopsies (Extended Data Fig. 9a,b). Also, expression of genes related to checkpoint inhibitors such as those encoding T cell immunoglobulin and mucin-domain containing-3 (TIM-3), IDO, PD-L1, PD-L2, PD-1 and CTLA-4 increased after treatment, indicating activation of immune cells in the TME (Extended Data Fig. 9c). Additionally, IHC of paired biopsies from four patients (MM01, MM02, MM05 and MM13) showed an upregulation of PD-L1, IDO, MHC-I and MHC-II on tumor cells, indicating a treatment-induced pro-inflammatory response in the three responding patients, except for a decrease in MHC-II expression in patient MM13. By contrast, the non-responding patient MM02 had a reduction in T cell numbers present in the tumor after treatment and no expression of PD-L1, IDO or MHC-II and, interestingly, total loss of MHC-I, demonstrating tumor immune escape (Extended Data Fig. 10a). CD8+ T cells and their distance (µm) to PD-L1-expressing cells in baseline biopsies from five patients was investigated by IHC. Except for patient MM13 (PR), distance and clinical responses were associated: the two responders had reduced distance (<20 µm) between cells expressing these markers compared to non-responding patients (>80 µm). This observation indicates that responding patients not only have a higher intratumoral infiltration of CD8+ T cells, but that these cells can surround and attack PD-L1-expressing immune cells and tumor cells (Extended Data Fig. 10b). Discussion In this clinical trial, MM1636, 30 patients with metastatic melanoma were treated with a first-in-class immunomodulatory IDO/PD-L1-targeting peptide vaccine combined with nivolumab. The treatment led to an unprecedented high ORR of 80%, with 43% of patients reaching a CR, and a striking mPFS of 26 months (95% CI, 15.4–69 months) was reached. The vaccine represents a new treatment strategy to activate specific T cells that target cells contributing to immune suppression (including tumor cells), positively modulating the TME by inducing local inflammation. Indeed, we show that vaccine-specific T cells isolated and expanded from vaccinated patients recognize not only tumor cells in a target- and HLA-restricted manner but also myeloid cells polarized toward a tumor-associated phenotype. Hence, myeloid cells become targets for vaccine-activated T cells when they have a tumor-associated phenotype in the TME. These phenomena may further induce checkpoint molecules and rewire the TME toward an increasingly aPD1-permissive state. A drawback of our study is the single-center nonrandomized setup. Comparison between trials or between patients in trials and real-world patients is problematic due to multiple factors, such as period of conduction and other therapies available at the different periods of time. Nevertheless, the rate of investigator-assessed ORR in the phase 3 trial CheckMate 067 was 43.7% in the nivolumab monotherapy group and 57% in the nivolumab and ipilimumab group. CRs occurred in 8.9% and 11.5% of patients, respectively18. The mPFS of 26 months (95% CI, 15.4–69 months) in this trial is more than twice as long as that for patients treated with nivolumab and ipilimumab in CheckMate 067, for which an mPFS of 11.5 months (95% CI, 8.7–19.3 months) was reached. Patient baseline characteristics were in general comparable to those of patients with MM who have been treated in CheckMate 067, although patients in MM1636 were older (mean age, 70 years) and a larger fraction were positive for PD-L1 (57%)3,18,22. Among patients with PD-L1-negative tumors in MM1636, an ORR of 61.5% was still reached, which would be expected to be around 33% for first-line nivolumab monotherapy23. Some studies suggest that older patients might have a tendency to respond better to aPD1 therapy; however, this is still debated24–27. To address potential trial bias and the nonrandomized setup, patients in MM1636 were matched for age, performance status, sex, M stage, LDH level, PD-L1 status and BRAF status with a historical control group from the DAMMED, who were treated contemporarily (2015–2019) with aPD1 monotherapy as standard of care17. We found a significantly higher ORR and CRR in MM1636 compared to those of matched patients, who had an ORR of 43% and a CRR of 13%, comparable to patients treated in CheckMate 067. Restrictions of the synthetic control group are of course that it is partially historic and patient selection outside matching criteria cannot be ruled out28. Numerous contemporary clinical trials are exploring the combination of aPD1 therapy with other immunomodulating agents for advanced melanoma. Talimogene laherparepvec, an oncolytic virus, is approved by the Food and Drug Administration and the European Medicines Agency to treat advanced melanoma. A small phase 1b trial with 21 patients (MASTERKEY-265) combined talimogene laherparepvec and pembrolizumab to treat patients with advanced unresectable melanoma and reached an ORR of 62% and a CR of 33%29,30. Seventy-one percent of the patients in this trial had an M stage below M1c; this number was 40% in our trial. Furthermore, mainly patients with an M stage below M1c responded to treatment, which was not the case in MM1636. Results from a large randomized phase 3 trial are awaited (KEYNOTE-034). Epacadostat, an IDO inhibitor, was tested in combination with pembrolizumab in a nonrandomized phase 2 trial in 40 aPD1 treatment-naive patients with MM with promising results, reaching an ORR of 62%. Unfortunately, the phase 3 trial showed no indication that epacadostat provided improvement in PFS and OS31. Limitations of the phase 3 trial were the sparse information on pharmacodynamics as well as biomarker evaluation to improve the design. The IDO/PD-L1 vaccine is different from epacadostat, as it is not an IDO inhibitor but targets IDO- and PD-L1-expressing cells. Similar vaccines administered as monotherapy induced objective responses in lung cancer and basal cell carcinoma, while epacadostat as monotherapy in 52 patients resulted in no responses32,33. Sahin et al. recently published encouraging data from a first-in-human trial, in which a vaccine containing liposomal RNA targeting four unmutated tumor-associated antigens (NY-ESO-1, MAGE-A3, tyrosinase and TPTE) was administered alone or in combination with nivolumab in patients with advanced melanoma. Responses were observed in both the monotherapy group as well as the combination group in checkpoint inhibitor-experienced patients, suggesting the efficacy of non-mutant shared tumor antigen vaccines34. The overall safety and tolerability findings are comparable to those of aPD1 monotherapy. Injection site reactions were exclusive to the vaccine. However, these side effects were transient and mild in most patients and most likely due to the adjuvant Montanide. IDO- and PD-L1-specific CD8+ and CD4+ T cells exist among peripheral blood lymphocytes in healthy donors35–38 and expand in response to pro-inflammatory stimuli35,39. Furthermore, both IDO and PD-L1 are induced in cells as a counter response to the inflammatory response. This provides a mechanism that ensures immune homeostasis, which keeps IDO/PD-L1-specific T cells in check; therefore, the risk of triggering autoimmune-related adverse events by vaccination appears to be minimal. This was confirmed in the first clinical trials of IDO and PD-L1 vaccination (NCT01543464 (refs. 33,40) and NCT03042793 (ref. 41)). The spontaneous (baseline) immune response to the vaccines observed in the current study is in agreement with our previous observation in various patients with cancer35–38. Numerous vaccine-induced changes in the blood and at the tumor site were observed. Peripheral IDO- and/or PD-L1-specific T cells were detected in vitro in a modified ELISPOT in over 93% of patients on vaccination, unrelated to patient HLA type. Immune responses were persistent in patients who surpassed follow-up at data cutoff and were still detectable up to 6 months after the last vaccine, suggesting induction of memory T cells. We did not observe a correlation between vaccine-induced responses in blood and clinical responses. However, the detection of a highly significant increase in vaccine-specific T cell numbers after vaccination in almost all patients together with the very low number of patients with PD makes it difficult to expect such a correlation, especially because other aspects should be taken into account (for example, the loss of class I expression on tumors cells in patient MM02 with PD after vaccination). Frequencies of peripheral T cells induced against PD-L1 were overall higher than those against IDO. Importantly, however, we also observed that it was not the same patients who responded strongly to both IDO and PD-L1, nor did patients react with a similar response pattern to the two antigens; that is, different time points of response were observed. This suggests that each component of the vaccine plays different roles in the ongoing immune response in patients. The goal of IDO/PD-L1 vaccination was to modulate the TME to increase responsiveness to aPD1 therapy, as we have observed in animal models, in which immune conversion is demonstrated in the TME with an increased influx of T cells. Targeting both IDO and PD-L1 together enables synergy, as the TME is known to use different immune escape mechanisms and IDO and PD-L1 are often overexpressed by different cellular compartments. Ex vivo ELISPOT and flow cytometry assays confirmed the induction of immune response toward both epitopes and the higher immunogenicity of the PD-L1 epitope, although because of lower sensitivity failed to detect some of the responses. TCR sequencing of five patients confirmed enrichment of IDO/PD-L1-specific T cell clones in the blood at different time points after treatment. Furthermore, an increase in enriched IDO/PD-L1-specific clones was observed in four of five patients at the tumor site after treatment, irrespective of clinical response. Phenotypic characterization showed that vaccine-specific T cells, which were expanded in vitro with interleukin (IL)-2 from the blood of vaccinated individuals, were both CD4+ T cells and CD8+ T cells. This was confirmed by ex vivo phenotype description of vaccine-activated T cells. Vaccine-specific T cells expressed CD107a and CD137 and produced IFN-γ and TNF-α upon stimulation with the cognate target, indicating their cytolytic capacity. Overall, the data from immune monitoring supports the importance of both antigens in the generation of the frequent clinical responses in the study. Despite a limited number of paired biopsies (due to either patient refusal or the fact that a large fraction of responding patients had no assessable tumors after six cycles), we observed trends indicating treatment-induced general T cell influx in responding patients. It was shown that proliferation of CD8+ T cells in the tumor after aPD1 treatment is associated with radiographic reduction in tumor size42. Additionally, we showed that a large proportion of expanded peripheral TCR clones were associated with tumors and the most considerable amount of clonal expansion was observed early, at cycle 3. For the patient with a CR included in the TCR sequencing analyses (MM01), the number of peripheral expanded clones present at the tumor site increased after treatment compared to that at baseline, indicating tumor trafficking of peripheral expanded clones. Gene expression analyses (two paired biopsies) and IHC (five paired biopsies) further demonstrated that the combination treatment induced a pro-inflammatory TME in responding patients with signs of T cell activation and cytotoxicity and increased cytokine activity. This may lead to further upregulation of IDO, PD-L1, MHC-I and MHC-II on tumor cells, leading to more treatment targets. It was shown that, following vaccination with a cancer vaccine, PD-L1 expression is increased on tumor cells due to recruitment of tumor-specific T cells and upregulation of adaptive immune resistance pathways in the TME43. Treatment with nivolumab monotherapy enhances PD-L1 expression, and it is therefore problematic to discriminate the effect of the vaccine as compared to that of nivolumab44. IDO- and PD-L1-specific pro-inflammatory effector T cells were hypothesized to counteract the functions of IDO- and PD-L1-expressing immune-suppressive cells as a means to keep the immune balance between immune activation and inhibition45. However, because the expression of these molecules is induced as a counter-regulatory response to inflammatory mediators such as IFNs, they can also be expressed by, for example, activated myeloid cells or T cells. Thus, activation of PD-L1-specific T cells may result in depletion of activated T cells in the tumor and other sites. Importantly, it was recently described that PD-L1+ T cells mainly have tolerogenic effects on tumor immunity and exert tumor-promoting properties, suggesting that targeting this immune population is indeed also beneficial46. We have previously investigated the effect of activating PD-L1-specific T cells in vitro and in vivo and found that, overall, they supported the effector phase of an immune response by removing PD-L1-expressing regulatory immune cells that inhibit PD-1+ effector T cells47,48. The major role of the PD-1 pathway is believed to be the regulation of effector T cell responses. Thus, this protective pathway is more important after activation, rather than at the initial T cell-activation stage49. Accordingly, the presence of PD-L1-specific T cells during the activation phase of an immune response may not increase or support a pro-inflammatory immune response due to the expression of PD-L1 on potent antigen-presenting cells or on the T cells themselves14. Thus, the overall effects of PD-L1-specific T cells may vary depending on the expression of both PD-1 and PD-L1, that is, due to the microenvironment and the state of the immune response. In conclusion, here we report an impressive response rate, CR rate and mPFS for a first-in-class immune-modulating vaccine combined with nivolumab. This may be a first step toward a new treatment strategy for patients with MM. Limitations are the low number of patients treated at a single institution and the lack of a randomized design with aPD1 monotherapy as comparator. Studies in aPD1 therapy-resistant or -refractory melanoma are ongoing as well as biomarker analysis for selecting patients at a higher likelihood to benefit from the combination versus aPD1 monotherapy. A larger randomized trial will be essential to validate these findings and determine the specific contribution of the vaccine to clinical responses and changes in the TME. In December 2020, the Food and Drug Administration granted breakthrough therapy designation for the IO102/IO103 vaccine combined with aPD1 therapy in metastatic melanoma based on data from the MM1636 trial. Methods Trial design and treatment plan MM1636 is an investigator-initiated, nonrandomized, open-label, single-center phase 1/2 study. All patients were treated at the Department of Oncology, Herlev and Gentofte Hospital, University of Copenhagen, Herlev, Denmark. The study was conducted according to the Declaration of Helsinki and Good Clinical Practice (GCP) and monitored by the GCP unit in Copenhagen, Denmark. The protocol was approved by the ethical committee of the Capital region of Denmark (H-17000988), the Danish Medical Agencies (2017011073) and the Capital Region of Denmark Data Unit (P-2019-172). The study was registered at https://clinicaltrials.gov/ under the identifier NCT03047928 and at the EudraCT (no. 2016-0004527-23). This study initially aimed to include 30 aPD1 treatment-naive patients with MM. An amendment with the addition of two other cohorts with ten patients in each cohort was made to evaluate immune responses and clinical efficacy in aPD1 therapy-resistant patients (cohort B, de novo resistance and cohort C, acquired resistance) for a total of 50 patients. The amendment with cohorts B and C was approved at the inclusion of 18 patients in cohort A. The trial is still including patients in cohorts B and C. This article reports results from cohort A. The first six patients were treated as for phase 1, evaluating for safety and tolerability before the remaining 24 patients were included in phase 2. The IDO/PD-L1 vaccine was administered subcutaneously biweekly for the first 10 weeks and thereafter every 4 weeks for approximately 9 months. A maximum of 15 vaccines were administered. Nivolumab was administered according to the approved label (3 mg per kg biweekly) for 24 cycles. The 15th vaccine was administered together with the 24th nivolumab cycle of 3 mg per kg, and responding patients thereafter continued nivolumab monotherapy every 4th week (6 mg per kg) as a standard of care after investigator assessment. Treatment with nivolumab was discontinued at the maximum benefit (assessed by the investigator), after a maximum of 2 years of therapy, at progression or due to severe adverse events (Extended Data Fig. 1b, treatment plan). Vaccine composition Each vaccine was composed of 100 µg IO102, a 21-amino-acid peptide (DTLLKALLEIASCLEKALQVF) from IDO, and 100 µg IO103, a 19-amino-acid peptide (FMTYWHLLNAFTVTVPKDL) from the signal peptide of PD-L1 (PolyPeptide Laboratories). Peptides were dissolved separately in 50 µl DMSO, filtered for sterility and frozen at −20 °C (NUNC CyroTubes CryoLine System Internal Thread, Sigma-Aldrich). At <24 h before administration, peptides were thawed. The PD-L1 peptide was diluted in 400 µl sterile water and, immediately before injection, mixed with the IDO peptide solution and 500 µl Montanide ISA-51 (SEPPIC) to achieve a total volume of 1 ml. Patients Patients above 18 years of age with locally advanced or stage 4 melanoma according to the AJCC (seventh edition), at least one measurable lesion according to RECIST 1.1 and an ECOG PS of 0–1 were eligible. The main exclusion criteria were prior treatment with aPD1 therapy, CNS metastases >1 cm, severe comorbidities and active autoimmune disease. Enrollment was not restricted to PD-L1 status, but it was known before inclusion. Patients were included after informed consent. Patients MM42 and MM20 have confirmed their approval of PET–CT images and clinical images that have been included in the article. Key study assessments Safety and tolerability were evaluated based on changes in clinical laboratory analyses and reported adverse events. Adverse events were assessed according to Common Terminology Criteria for Adverse Events (version 5.0) and were graded from 1 to 5 for all treated patients up to 6 months after the last dose of the IDO/PD-L1 vaccine. Clinical efficacy was assessed using FDG PET–CT scans before treatment and every 3rd month until progression. Objective responses were categorized into CR, PR, stable disease or PD according to RECIST version 1.1. Clinical data were collected at CCIT-DK in the eCRF program OpenClinica version 1.0 and in Microsoft Excel version 2002 on a secure server. Blood samples for immunologic analyses were collected before treatment, before the third cycle, after the sixth, 12th, 18th and 24th cycles (on vaccination) and 3 and 6 months after the last vaccine. Two to three tumor needle biopsies (1.2 mm) were collected at baseline and after six cycles from the same tumor site, when accessible, to evaluate immune responses at the tumor site. DTH skin tests were performed and punch biopsies were taken from the DTH area after cycle 6 for the evaluation of SKILs reactive to PD-L1 and IDO (Supplementary Fig. 1, treatment plan). Statistical analysis of clinical outcome Survival curves were computed in GraphPad Prism software version 9.0.0 according to Kaplan–Meier method. Median follow-up time of enrollment was calculated using the reverse Kaplan–Meier method, also in GraphPad Prism software version 9.0.0. For binary outcomes, 95% two-sided CIs were constructed using the Clopper–Pearson method, also in GraphPad version 9.0.0. An independent board of certified and experienced oncoradiologists performed an external review to evaluate clinical response to address the potential bias of investigator site review. The external review took place at Rigshospitalet, Copenhagen University Hospital. This hospital did not participate in the MM1636 trial, and the external reviewer had no prior knowledge about the clinical trial or the trial therapy. Only PET–CT images were accessed, and arrows indicating target or non-target lesions appeared on baseline images as the only additional information. To address potential trial bias regarding treatment effect, we matched patients in MM1636 with patients from the DAMMED, a population-based database that retrospectively collects data on patients with metastatic melanoma in Denmark. Here, data from 938 patients treated with aPD1 monotherapy contemporaneously (January 2015–October 2019) were extracted. Two hundred and eighteen of these patients were eligible for comparison and matching (all parameters available) (Supplementary Table 1), and 74 patients from the DAMMED were found to match. Patients were matched for age (≤70 years, >70 years), sex, LDH levels (normal, elevated), M stage (M1a, M1b, M1c), BRAF status (wild type, mutated) and PD-L1 status (<1%, ≥1%). An exact matching algorithm was used in which patients in MM1636 were matched with patients from the DAMMED with the same combination of variables. Twenty-nine patients from MM1636 were matched with exact combinations of the six variables. One patient could not be matched. To secure balance of the calculations, control patients were weighted according to the number of patients for each MM1636 patient. Estimates for treatment effects were calculated by weighted logistic regression analyses and weighted Cox proportional hazard model. The R package ‘Matchlt’ was used for matching patients. As the method chosen for matching control patients to protocol patients, a weighted binary logistic regression model was used for comparing response rates in the two matched cohorts. Odds ratios and response rates, including their corresponding 95% CIs, were extracted from the regression models. All P values were two sided, and P values below 0.05 were considered statistically significant. SAS version 9.4M5 was used for the weighted logistic regression models. Processing of peripheral blood mononuclear cells Peripheral blood was collected from all patients in heparinized tubes and was processed within 4 h. In brief, PBMCs were isolated using Lymphoprep (Medinor) separation. PBMCs were counted on a Sysmex XP-300 analyzer and frozen in Human AB Serum (Sigma-Aldrich, H4522-100ML) with 10% DMSO using controlled-rate freezing (CoolCell, BioCision) in a −80 °C freezer and were moved the next day to a freezer at −140 °C until further processing. Needle biopsies at baseline and after the sixth vaccine Two to three needle biopsies (1.2 mm) were taken at baseline and after the sixth cycle of treatment when assessable from the same tumor lesion. One fragment was fixed with formalin and embedded in paraffin (FFPE); one to two fragments were used for expanding tumor-infiltrating T cells and establishing autologous tumor cell lines. Delayed-type hypersensitivity and generation of skin-infiltrating lymphocytes After six cycles of treatment, we performed intradermal injections of vaccine components without adjuvant and one control injection containing DMSO without peptide. Patients were injected with either a mixture of both IDO and PD-L1 peptides at all three injection sites or PD-L1 peptide, IDO peptide or a mixture of the two at the injection sites, respectively (Supplementary Table 6). Eight hours after injection, punch biopsies were resected from the three sites where the peptide was injected and transported immediately to the laboratory and cut into fragments of 1–2 mm3. SKILs were expanded to establish ‘young SKILs’ in CM medium consisting of RPMI 1640 with GlutaMAX, 25 mM HEPES, pH 7.2 (Gibco, 72400-021), IL-2 (100 or 6,000 IU ml−1) (Proleukin Novartis, 004184), 10% heat-inactivated Human AB Serum (Sigma-Aldrich, H4522-100ML), 100 U ml−1 penicillin, 1.25 µg ml−1 Fungizone (Bristol Myers Squibb, 49182) and 100 µg ml−1 streptomycin (Gibco, 15140-122). Half of the medium was replaced three times per week. Young SKILs from samples derived from IDO peptide, PD-L1 peptide and the mixture injection sites were further expanded in a small-scale version of the 14-d rapid expansion protocol as previously described51. Quantification of specific T cells in blood by ELISPOT For enumeration of vaccine-specific T cells in the peripheral blood, PBMCs from patients were stimulated with IDO or PD-L1 peptide in the presence of a low dose of IL-2 (120 U ml−1) for 7–13 d before being used in IFN-γ ELISPOT assays with 2.5–3.2 × 105 cells per well. Briefly, cells were placed in a 96-well PVDF membrane-bottomed ELISPOT plate (MultiScreen MSIPN4W50, Millipore) precoated with IFN-γ-capture antibody (clone 1-D1K, Mabtech). Diluted IDO or PD-L1 peptide stocks in DMSO were added at 5 µM; an equivalent amount of DMSO was added to control wells. PBMCs from each patient were set up in duplicate or triplicate for peptide and control stimulations. Cells were incubated in ELISPOT plates in the presence of the peptide for 16–18 h, after which they were washed off, and biotinylated secondary antibody (anti-human IFN-γ mAB, clone 7-B6-1, Mabtech) was added. After 2 h of incubation, unbound secondary antibody was washed off, and streptavidin-conjugated alkaline phosphatase (Mabtech) was added for 1 h. Next, unbound streptavidin-conjugated enzyme was washed off, and the assay was developed by adding BCIP/NBT substrate (Mabtech). ELISPOT plates were analyzed on the CTL ImmunoSpot S6 Ultimate V analyzer using ImmunoSpot software version 5.1. Responses were calculated as the difference between the average numbers of spots in wells stimulated with IDO or PD-L1 peptide and those from corresponding control wells. For detection of IDO and PD-L1 peptide responses ex vivo, PBMCs were thawed and rested overnight in medium containing DNase I (1 μg ml−1, Sigma-Aldrich, 11284932001) before being used in IFN-γ ELISPOT assays as described above. A total of 6–9 × 105 PBMCs per well were seeded. Statistical analysis of all ELISPOT responses was performed using the DFR method as described by Moodie et al. using RStudio software (RStudio Team, 2016, http://www.rstudio.com/)50. Vaccine-specific ELISPOT responses were defined as true if the difference between the spot count in control and peptide-stimulated wells was statistically significant according to DFR statistical analysis or, for samples performed in duplicate, if the spot count in peptide-stimulated wells was at least 2× the spot count in control wells50. Cancer cell lines and tumor conditioned medium Autologous melanoma cell lines were established from needle biopsies. Briefly, biopsies were chopped into small fragments and seeded in 24-well cultures in RPMI 1640 with GlutaMAX, 25 mM HEPES, pH 7.2 (Gibco, 72400-021), 10% heat-inactivated FBS (Life Technologies, 10500064), 100 U ml−1 penicillin, 1.25 µg ml−1 Fungizone (Bristol Myers Squibb, 49182) and 100 µg ml−1 streptomycin (Gibco, 15140-122). Established adherent melanoma tumor cell lines were cryopreserved at −140 °C in freezing medium containing FBS with 10% DMSO. PD-L1 and HLA-II expression on established tumor cell lines was assessed by flow cytometry staining with anti-PD-L1–PE-Cy7 (1:23 dilution, clone M1h1 (RUO), BD, 558017) and anti-HLA-II–FITC (1:23 dilution, clone Tu39 (RUO), BD, 555558) antibodies. To obtain TCM, established tumor cell lines were cultured in 175-cm2 Nunc cell culture flasks until 80–90% confluency was reached. The culture medium was then replaced with 20 ml fresh X-VIVO 15 with Gentamicin and Phenol Red (Lonza, BE02-060Q), medium with 5% heat-inactivated Human AB Serum (Sigma-Aldrich, H4522-100ML). After 24 h of incubation, TCM was collected and centrifuged to remove any resuspended cells, after which TCM was aliquoted, frozen and stored at −80 °C. The acute monocytic leukemia cell line MonoMac1 was obtained from the DSMZ—German Collection of Microorganisms and Cell Cultures (ACC 252) and cultured in RPMI 1640 with GlutaMAX, 25 mM HEPES, pH 7.2 (Gibco, 72400-021) and 10% heat-inactivated FBS. Isolation of autologous myeloid cells Autologous CD14+ cells were sorted from freshly thawed PBMCs using a magnetic bead-separation kit (Miltenyi Biotec, 130-050-201) according to the manufacturer’s instructions. Isolated CD14+ cells were used as targets in the IFN-γ ELISPOT assay directly after sorting or differentiated in vitro into tumor-associated macrophages by culturing with 1 ml fresh X-VIVO 15 medium with Gentamicin and Phenol Red (Lonza, BE02-060Q) and 5% heat-inactivated Human AB Serum, supplemented with 1 ml autologous TCM in 24-well plates for 2 d. Quantification of vaccine-specific T cells from DTH biopsy sites by ELISPOT SKILs expanded in vitro were rested in medium without IL-2 overnight before being used in the IFN-γ ELISPOT assay as described above to evaluate reactivity of skin-infiltrating T cells. Generation of IDO- and PD-L1-specific T cell cultures from PBMCs or SKILs IDO- or PD-L1-specific T cells were isolated from peptide-stimulated in vitro PBMC cultures on days 14–15 after stimulation or from SKIL cultures expanded in vitro. For specific T cell isolation, PBMCs or SKILs were stimulated with IDO or PD-L1 peptide, and cytokine-producing T cells were sorted using the IFN-γ or TNF-α Secretion Assay—Cell Enrichment and Detection kit (Miltenyi Biotec). Cytokine-production profile of PD-L1- and IDO-specific T cells by intracellular staining To assess the T cell cytokine-production profile, isolated and expanded IDO- and PD-L1-specific T cell cultures were stimulated for 5 h with peptide at 5 μM in a 96-well plate. One hour after the start of the incubation, anti-CD107a–PE (1:133 dilution, clone H4A3, BD Biosciences, 555801) antibody and BD GolgiPlug (1:1,000 dilution, BD Biosciences) were added. After a 5-h incubation, cells were stained using fluorescently labeled surface marker antibodies: anti-CD4–PerCP (1:5 dilution, clone SK3, 345770), anti-CD8–FITC (1:5 dilution, clone SK1, 345772) and anti-CD3–APC-H7 (1:20 dilution, clone SK7, 560275) (all from BD Biosciences). Dead cells were stained using FVS510 (1:250 dilution, BD Biosciences, 564406), followed by overnight fixation and permeabilization using eBioscience Fixation/Permeabilization buffers (00-5123-43, 00-5223-56) according to the manufacturer’s instructions. Cells were then stained intracellularly in eBioscience permeabilization buffer (00-8333-56) with anti-IFN-γ–APC (1:20 dilution, clone 25723.11, BD Biosciences, 341117) and anti-TNF-α–BV421 (1:100 dilution, clone Mab11, BD Biosciences, 562783) antibodies. Samples were analyzed on the FACSCanto II (BD Biosciences) using BD FACSDiva software version 8.0.2. The gating strategy is shown in Supplementary Fig. 3a. For assessing the reactivity against IDO expressing MonoMac1 cells, the appropriate amount of cancer cells was added to the IDO-specific T cells to obtain an effector:target ratio of 4:1 in the 96-well plate and T cell cytokine production was tested after 5 h of co-culture as described above. For HLA-blocking experiments, HLA-blocking antibodies (2 μg ml−1) were added directly into wells 30 min before the addition of peptide. Blocking antibodies used were HLA-DR (1:500 dilution, clone L243, Abcam, ab136320), HLA-DQ (1:500 dilution, clone SPV-L3, Abcam, ab23632) and HLA-DP (1:500 dilution, clone B7/21, Abcam, ab20897). To assess ex vivo T cell reactivity to IDO and PD-L1 peptides in patient PBMCs, cells were thawed and rested for 1–2 d in medium containing DNase I (1 μg ml−1, Sigma-Aldrich, 11284932001). PBMCs were then stimulated with peptide at 5 μM in a 96-well plate for 8 h. An hour after the addition of peptide, anti-CD107a–BV421 (3:50 dilution, clone H4A3, BD Biosciences, 328626) antibody and BD GolgiPlug (1:1,000 dilution, BD Biosciences) were added. Surface and intracellular staining was performed as described above. Antibodies used for surface staining were anti-CD3–PE-CF594 (0.8:30 dilution, clone UCHT1, BD Biosciences, 562280), anti-CD4–BV711 (1:30 dilution, clone SK3, BD Biosciences, 563028) and anti-CD8–Qdot605 (1:150 dilution, clone 3B5, Thermo Fisher, Q10009). Antibodies used for intracellular staining were anti-CD137–PE (1:20 dilution, clone 4B4-1 (RUO), BD Biosciences, 555956), anti-IFN-γ–PE-Cy7 (1.5:20 dilution, clone B27, BD Biosciences, 557643) and anti-TNF-α–APC (1:20 dilution, clone Mab11, BD Biosciences, 554514). Samples were acquired on the NovoCyte Quanteon (ACEA Biosciences) and analyzed using NovoExpress software version 1.4.1. To assess vaccine-specific T cell responses, background values observed in unstimulated PBMC samples were subtracted from values observed in peptide-stimulated conditions. Positive response value threshold was set at a difference of 0.2% from the background values. Based on this response cutoff, only TNF-α, CD107a and CD137 responses were detected in this assay. Statistical analysis comparing baseline with on-treatment or post-treatment cytokine profiles was performed using two-sided Wilcoxon matched-pairs signed-rank test. The gating strategy is shown in Supplementary Fig. 3b. siRNA-mediated PD-L1 and IDO1 silencing A stealth siRNA duplex for targeted silencing of PD-L1 (Invitrogen)52, a custom silencer select siRNA for targeted silencing of IDO1 (Ambion) and the recommended silencer select negative control (Ambion) siRNA for mock transfection were used. The stealth PD-L1 siRNA duplex consisted of the sense sequence 5′-CCUACUGGCAUU-UGCUGAACGCAUU-3′ and the antisense sequence 5′-AAUGCGUUCAGCAAAUGCCAGUAGG-3′. The three silencer IDO siRNA duplexes used were siRNA1 (sense sequence, 5′-ACAUCUGCCUGAUCUCAUATT-3′; antisense sequence, 5′-UAUGAGAUCAGGCAGAUGUTT-3′), siRNA2 (sense sequence, 5′-CCACGAUCAUGUGAACCCATT-3′; antisense sequence, 5′-UGGGUUCACAUGAUCG-UGGAT-3′) and siRNA3 (sense sequence, 5′-CGAUCAUGUGAACCCAAAATT-3′; antisense sequence, 5′-UUUUGGGUUCACAUGAUCGTG-3′). For PD-L1 or IDO1 silencing, cancer cells were electroporated with 0.025 nmol of each siRNA duplex as previously described53. For PD-L1-silencing experiments, cancer cells were treated with IFN-γ (500 U ml−1, PeproTech) 1 h after electroporation. Electroporated cells were used as target cells in ELISPOT and ICS assays 24 h or 48 h after siRNA electroporation. RT–qPCR Total RNA was extracted using the RNeasy Plus Mini kit (Qiagen, 74134) following the manufacturer’s instructions. RNA concentration was quantified using a NanoDrop 2000 (Thermo Fisher Scientific), and a total of 1,000 ng RNA was reverse transcribed with the High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems, 4368814) using 1,000 ng input RNA for transcription. Real-time qPCR analyses were performed using the TaqMan Gene Expression Assay on a Roche LightCycler 480 instrument. RT–qPCR assays were performed with a minimum of three technical replicates and analyzed using the ΔCt method as described in Bookout et al.54 with normalization of IDO1 expression (primer ID, Hs00984148_m1) or PD-L1 expression (primer ID, Hs001125296_m1) to the expression level of the housekeeping gene POLR2A (primer ID, Hs00172187_m1) and the control sample (mock). For low-concentration samples with no amplification, Ct was set to 40. Controls without reverse transcriptase were used as controls for specific amplifications. P values were determined using two-tailed parametric t-tests. BRAF-mutational status and PD-L1 status at baseline in all patients A library of historical FFPE biopsies were assessable for all patients and analyzed locally at Herlev and Gentofte University Hospital by experienced pathologists for BRAF status and PD-L1 expression on tumor cells. The BRAF analysis was carried out with real-time PCR using the EntroGen BRAF Mutation Analysis Kit II (BRAFX-RT64, CE-IVD) to specifically detect mutations corresponding to V600D, V600E and V600K in the BRAF protein. PD-L1 status was assessed using the monoclonal rabbit anti-PD-L1 antibody (clone 28.8, PD-L1 IHC 28-8 pharmDx) in FFPE biopsies following the manufacturer’s instructions. Patients were considered positive for PD-L1 with expression levels ≥1% and negative for PD-L1 with expression levels <1%. Human leukocyte antigen type Blood samples from all 30 patients were genotyped for class I (HLA-A, HLA-B, HLA-C) and three class II types (HLA-DRB1, HLA-DQA1, HLA-DQB1) using LinkSēq HLA Typing Kits (Thermo Fisher, 1580C). These test kits are based on real-time PCR using allele-specific exponential amplification (sequence-specific primers) followed by melting curve analyses. Immunohistochemistry simplex: Immunoscore CR IHC staining was performed using a qualified Ventana Benchmark XT with four different steps: (1) antigen retrieval; (2) staining with the following primary antibodies: anti-CD3 antibody (clone HDX2; provider, HalioDx; HD-FG-000013; 10931065/10636667; concentration, 0.25 µg ml−1), anti-CD8 antibody (clone HDX3; provider, HalioDx; HD-FG-000019; 10931069/10337710/10639301; concentration, 0.5 µg ml−1), anti-IDO monoclonal antibody (clone VINC3IDO; provider, Thermo Fisher Scientific; 14-9750-82; E25003-101; concentration, 0.05 µg ml−1), anti-HLA class 1 ABC antibody (clone, EMR8-5; provider, Abcam; ab70328; GR3248333/GR3186494; concentration, 0.5 µg ml−1), anti-HLA-DR/DP/DQ/DX antibody (clone CR3/43; provider, Santa Cruz; sc-53302; L1714; concentration, 1 µg ml−1) and anti-PD-L1 antibody (clone HDX3; provider, HalioDx; HD-FG-000035; 106312810/106312816; concentration, 3.3 µg ml−1); (3) detection with a secondary antibody using the ultraView Universal DAB Detection kit; and (4) counterstaining using Hematoxylin & Bluing Reagent (staining of cellular nuclei). Control slides were systematically included in each staining run to permit quality control of the obtained measurements. Following coverslipping, slides were scanned with the NanoZoomer-XR to generate digital images (20×). Two consecutive slides were specifically used to perform Immunoscore CR TL staining of CD3+ and CD8+ cells. Digital pathology of T lymphocytes The digital pathology for Immunoscore CR TL allowed the quantification of positive cells (in cells per mm2) into the core tumor and invasive margin if present. Each sample was analyzed using the HalioDx Digital Pathology Platform. Digital pathology and pathologist analysis of IDO Digital pathology for IDO allowed the quantification of stained area (in mm2) into the whole tumor. Each sample was analyzed using the HalioDx Digital Pathology Platform. Analysis of IDO+ cells was performed by a pathologist. Results were expressed as an H score from 0 to 300. The score is obtained by the formula 3 × percentage of cells with strong staining + 2 × percentage of cells with moderate staining + percentage of cells with weak staining. Digital pathologist analysis of MHC-I and MHC-II Analysis of MHC-I+ and MHC-II+ cells was performed by a pathologist. Results are expressed as a percentage of positive tumor cells and a percentage of positive cells in the stroma. Digital pathology Immunoscore immune checkpoint (CD8+ and PD-L1+) The Immunoscore CR IC test allowed the quantification of CD8++ cell density in the whole tumor and a CD8+-centered proximity index (which corresponds to the percentage of CD8++ cells that have at least one PD-L1+ cell in the neighborhood) at different cutoff distances (20 µm, 40 µm, 60 µm and 80 µm). Pathologist analysis of PD-L1 A pathologist performed analysis of PD-L1+ cells. Positivity of a viable tumor cell was considered when partial or complete cell membrane staining was observed (more than 10% of the tumor cell membrane). Results were expressed as a percentage. NanoString RNA profiling and Immunosign RNA was extracted from FFPE tissues using Qiagen RNeasy FFPE extraction kits (Qiagen). Annotations from the pathologist performing H&E staining were used to guide removal of normal tissue from the slides by macrodissection before nucleic acid extraction, which occurred after tissue deparaffinization and lysis. Each extracted RNA sample was independently quantified using a NanoDrop spectrophotometer (NanoDrop Technologies) and qualified (Agilent Bioanalyzer). Degradation was quantified as the percentage of RNA fragments smaller than 300 bp using the RNA 6000 Nano kit (Agilent Bioanalyzer). Good sample quality was defined as less than 50% of RNA fragments being 50–300 bp in size. RNA expression profiling was performed using the nCounter PanCancer Immune Profiling Panel from NanoString (NanoString Technologies). The PanCancer Immune Profiling Panel contains 776 probes and is supplemented with six genes to complete HalioDx Immunosign targets. Hybridization was performed according to the manufacturer’s instructions. Hybridized probes were then purified and immobilized on a streptavidin-coated cartridge using the nCounter Prep Station (NanoString Technologies). Data collection was carried out on the nCounter Digital Analyzer (NanoString Technologies) following the manufacturer’s instructions to count individual fluorescent barcodes and quantify target RNA molecules present in each sample. For each assay, a scan of 490 fields of view was performed. Raw data from the NanoString nCounter were processed using NanoString recommendations. Quality control enables to retain data of good quality with a binding density that ranges between 0.05 and 2.25. The linearity of positive controls was checked using the R2 value of regression between the counts and the concentrations of positive controls. Samples that showed R2 < 0.75 were flagged and removed from the analysis. The background was removed using a thresholding method at the mean + 2 s.d. of negative controls. Raw counts were normalized using a positive normalization factor. Samples showing positive normalization factors outside the range of 0.3–3 were removed from the analysis. A second normalization was performed using the housekeeping gene normalization factor. Only the most stable housekeeping genes were selected for this normalization step using the variance-versus-mean relationship. All samples showing a normalization factor outside the range of 0.1–10 were removed from the analysis. All statistical analyses were performed on normalized counts using R software (version 2.6.2, 2019-12-12). T cell receptor variable β chain sequencing To track longitudinal immune responses to therapy, genomic DNA was extracted from longitudinal pre- and post-treatment PBMCs (five patients), pre- and post-treatment biopsies (five patients) (both FFPE and treated with RNAlater) and IDO- and PD-L1-specific T cell cultures from PBMCs (five patients) or SKILs (one patient). Three clonal (two IDO- and one PD-L1-) specific cultures were also sequenced to confirm clonal purity of cultures. DNA from PBMCs or RNAlater-treated biopsies was extracted with the DNeasy Blood and Tissue kit (Qiagen, 69504), DNA from sorted IDO and PD-L1-specific T cells from either PBMCs or SKILs was extracted using the QIAamp DNA Micro kit (Qiagen, 565304), and DNA from PFFE biopsies was extracted using the Maxwell RSC DNA FFPE kit (Promega, AS1450). Immunosequencing of the CDR3 regions of human TCRβ chains was performed using the immunoSEQÒ Assay (Adaptive Biotechnologies). Extracted genomic DNA was amplified in a bias-controlled multiplex PCR, followed by high-throughput sequencing. Sequences were collapsed and filtered to identify and quantitate the absolute abundance of each unique TCRβ CDR3 region for further analysis as previously described55–57. Statistical analyses of TCRβ sequencing results Two quantitative components of diversity were compared across samples in this study. First, Simpson clonality was calculated on productive rearrangements by ∑i=1Rpi2, where R is the total number of rearrangements and pi is the productive frequency of rearrangement i. Values of Simpson clonality range from 0 to 1 and measure how evenly receptor sequences (rearrangements) are distributed among a set of T cells. Clonality values approaching 0 indicate a very even distribution of frequencies, whereas values approaching 1 indicate an increasingly asymmetric distribution in which a few clones are present at high frequencies. Second, sample richness was calculated as the number of unique productive rearrangements in a sample after computationally downsampling to a common number of T cells to control for variation in sample depth or T cell fraction. Repertoires were randomly sampled without replacement five times, and we report the mean number of unique rearrangements. The T cell fraction was calculated by taking the total number of T cell templates and dividing by the total number of nucleated cells. The total number of nucleated cells was derived from reference genes using the immunoSEQ Analyzer version 3. To identify enriched vaccine clones in each patient, rearrangement frequencies in their baseline PBMCs and each IDO/PD-L1-sorted T cell sample were compared using a binomial distribution framework as previously described58. In brief, for each clone, we performed a two-sided test to determine whether frequencies were the same in the patient’s periphery and a PD-L1- or IDO-specific T cell sample. The Benjamini–Hochberg procedure was used to control the false discovery rate at 0.01 (ref. 59). Clonal expansion in post-treatment samples was similarly assessed using this differential abundance framework, but an IDO/PD-L1-specific T cell sample was replaced with a post-treatment series sample. In biopsies, the six series frequencies were compared to those of baseline tissue. Lastly, vaccine-associated clones were tracked in each PBMC and tissue sample by summing the frequency of each rearrangement enriched in either PD-L1- or IDO-specific T cells. All statistical analyses were performed in R version 3.6.1. Reporting Summary Further information on research design is available in the Nature Research Reporting Summary linked to this article. Online content Any methods, additional references, Nature Research reporting summaries, source data, extended data, supplementary information, acknowledgements, peer review information; details of author contributions and competing interests; and statements of data and code availability are available at 10.1038/s41591-021-01544-x. Supplementary information Supplementary Information Supplementary Figs. 1–3 and Tables 1–6 Reporting Summary Extended data Extended Data Fig. 1 Mode of Action and treatment plan. a) Anticipated mechanism of action of the combination therapy of an IDO/PD-L1 derived peptide vaccine and nivolumab (anti-PD1). 1) The IDO/PD-L1 peptide vaccine is administered subcutaneously (s.c) and nivolumab is administered intravenous (IV). 2) The vaccines peptides are phagocytosed by an antigen presenting cell and presented to IDO and PD-L1 specific T cells, which are activated. 3) The activated T cells migrates to the tumor site where they attack both immune-suppressive cells and tumour cells expressing IDO and/or PD-L1 leading to cytokine production and a pro-inflammatory tumour microenvironment. 4/5) Enhanced tumour killing by both IDO/PD-L1 specific T cells and tumor specific cytotoxic T cells due to PD-1 blockade. Created with BioRender.com b) Treatment plan. After written informed content patients were screened. Before treatment start a baseline PET/CT scan was performed, baseline blood sample for research use and if assessable a baseline needle biopsy. Patients were treated with the IDO/PD-L1 peptide vaccine subcutaneously biweekly for the first 6 injections and thereafter every fourth week for a maximum of 15 vaccines. Nivolumab was administered in parallel biweekly (3mg/kg) up to 24 series. If patients needed subsequent nivolumab after ended vaccination regiment they were treated with 6 mg/kg every fourth week up to two years. Needle biopsy and delayed type hypersensitivity (DTH) was performed after 6 series of treatment if assessable. PET/CT scans was performed every third month. Extended Data Fig. 2 Progression free survival and overall survival in matched historical control group and vaccine injection site reaction. a) Kaplan-Meier curve of progression free survival with corresponding 95% confidence intervals in the matched historical control group (n=74). Patients were matched on BRAF-status, PD-L1-status, age, gender, M-stage and LDH level). b) Kaplan-Meier curve of overall survival with corresponding 95% confidence intervals in the matched historical control group (n=74). c) Images of injection site reaction in patient MM20 (CR) after 11 vaccination show redness, rash and granuloma at injection site. Extended Data Fig. 3 Vaccine specific responses in blood. a) Heatmap of specific (background has been subtracted) IDO and PD-L1 responses in PBMCs at baseline, series 3, 6, 12, 18 and 24 measured by in vitro IFN-γ Elispot assay show fluctuations in the blood during treatment (n=30). b) Heatmap of specific (background has been subtracted) IDO and PD-L1 responses in PBMCs at baseline and 3 and 6 months after last vaccine measured by in vitro IFN- γ Elispot assay (n=30). 2.5–3.2x105 cells per well were used. c) Vaccine associated clones were tracked in the blood by summing the frequency of each rearrangement enriched in either IDO or PD-L1 T cells. Extended Data Fig. 4 Ex vivo vaccine specific responses in blood. Heatmaps of detected specific (background has been subtracted) IDO (a) and PD-L1 (b) responses in PBMCs at baseline and on/after treatment as measured by IFNγ ELISPOT (n=25). C) Example of well images of ex vivo ELISPOT wells for three different patients (n=3). 6-9x105 cells per well were used. Extended Data Fig. 5 CD4 and CD8 vaccine specific T cell responses in blood. Top: heatmaps of IDO specific CD4 (left) and CD8 (right) T cell responses in PBMCs at baseline and on/post treatment. Bottom: heatmaps of PD-L1 specific CD4 (left) and CD8 (right) T cell responses. Responses quantified by flow cytometry by an increased expression of CD107α, CD137 and TNFα after 8h peptide stimulation. Values represent specific responses after background values have been subtracted (n=21). Statistical analysis were performed using two-sided Wilcoxon matched-paired rank test. Extended Data Fig. 6 Pro-inflammatory profiles of sorted CD4 and CD8 T cells from blood. a/b/c/d) Percentage of in vitro stimulated and sorted PD-L1 and IDO specific CD4 and CD8 T cells secreting cytokines. Extended Data Fig. 7 Vaccine specific responses in skin. a) IDO and PD-L1 specific T cell responses in SKILs after 6 series of treatment measured by IFN-γ Elispot assay (n=13). SKILs were grown from DTH injection with either IDO peptide, PD-L1 peptide or a mix as presented by different blue colours. * Responses were calculated as the difference between average numbers of spots in wells stimulated with IDO or PD-L1 peptide (triplicates) and corresponding control (DMSO) and statistical analyses of Elispot responses were performed using distribution-free resampling method (Moodie et al.). DR: Not statistically confirmed response due to replicate number but number of spots in peptide wells are two times higher than control wells (DMSO). NS: No significant response and no DR. b/c/d) Percentage of cytokine secreting/CD107a+ CD4+ and CD8+ IDO and PD-L1 specific T cells in response to in vitro peptide stimulation by flow cytometry. e) Skin infiltrating PD-L1 specific T cell clones also found in biopsy in patient MM01. TCR sequencing was performed on a PD-L1 specific T cell culture generated from DTH area on the lower back injected with PD-L1 peptide on patient MM01. Bars show the frequency of top 25 clones in the culture with which indicates a high Simpson clonality of 0.43. f) Tracking the frequency of the top five skin infiltrating PD-L1 specific clones in tumour before and after treatment. Extended Data Fig. 8 T cell changes in blood after treatment. T cell fraction, TCR clonality and repertoire richness in blood and peripherally expanded TCR clones in both responding and non-responding patients. a) T cell fraction in peripheral blood of five patients at baseline, series 3, 6 and 12 by TCR sequencing. T cell fraction was calculated by taking the total number of T cell templates and dividing by the total number of nucleated cells (n=5.) b) TCR clonality in peripheral blood of five patients at baseline, series 3, 6 and 12 by TCR sequencing (n=5). Simpson clonality measures how evenly TCR sequences are distributed amongst a set of T cells where 0 indicate even distribution of frequencies and 1 indicate an asymmetric distribution where a few clones dominate. c) TCR repertoire richness in peripheral blood of five patients at baseline, series 3, 6 and 12 by TCR sequencing (n=5). TCR repertoire richness report the mean number of unique rearrangements. d) Bar-chart chart representing dynamics of the expanded T cell clones. Coloured bars represent peripherally expanded clones in five patients at series 3, series 6 and series 12 as compared to the baseline PBMC samples (n=5). Light gray bar represents peripherally expanded clones that are present in baseline biopsy samples while dark gray bar represents peripherally expanded clones that are present in the post-treatment biopsy (after 6 series). e) Frequency of the dominant TCR β chain in clonal PD-L1 and IDO specific cultures as determined by CDR3 sequencing. Extended Data Fig. 9 Profiling of genes relevant to T cell activation, cytokines and exhaustion markers on pre- and post-treatment biopsies in two patients. a) RNA expression profiling of genes related to T cell activation was performed using NanoString nCounter (n=2). b) RNA expression profiling of genes related to cytokine activity was performed using NanoString nCounter (n=2). c) RNA expression profiling of genes related to checkpoint inhibitors was performed using NanoString nCounter (n=2). Extended Data Fig. 10 Treatment induced upregulation of PD-L1, IDO, MHC I and MHC II and distance between CD8 T cells and PD-L1 expressing cells. a) IHC on 4 paired biopsies stained for CD3 and CD8 T cells, PD-L1, MHC I and MHC II on tumor cells/mm2 (sum in the validated area) and IDO H-score (from 0 to 300). H-score is the expression of IDO on both immune and tumor cells: The score is obtained by the formula: 3 x percentage of cells with a strong staining + 2 x percentage of cells with a moderate staining + percentage of cells with a weak staining (n=4). b) Distance in µm between CD8+ T cells and PD-L1+ stained cells on five baseline biopsies detected by IHC (n=5). Extended data is available for this paper at 10.1038/s41591-021-01544-x. Supplementary information The online version contains supplementary material available at 10.1038/s41591-021-01544-x. Acknowledgements We thank all patients and their relatives for being a part of the trial. We thank Ö. Met, M. Jonassen, S. Ullitz Færch, B. Saxild, S. Wendt and C. Grønhøj for technical support. We thank M. Cumberbatch for input with translational analysis. We thank the nurses at clinic 5 and the head of the Oncology Department at Herlev and Gentofte Hospital, L. Sengeløv. The study was funded through a research funding agreement between IO Biotech and the CCIT-DK, Herlev Hospital and the Oncology Department at Herlev Hospital. Author contributions I.M.S. and M.H.A. conceived the study. J.W.K. wrote the protocol with input from I.M.S., M.H.A. and M.-B.Z. J.W.K. and C.L.L. were responsible for patient treatment and patient care with help from R.B.H. and consulting from I.M.S., E. Ellebaek and M.D. I.M.S., E. Ellebaek and M.D. were responsible for patient recruitment. J.W.K. and C.L.L. were responsible for coordinating trial procedures and collecting data and samples. J.W.K. and C.L.L. were clinical investigators, and I.M.S. was a sponsor. J.W.K. and C.L.L. analyzed and interpreted clinical data with support from I.M.S. and E. Ehrnrooth. T.W.K. performed statistical analyses. H.W.H. helped with clinical imaging. J.W.K., E.M., C.O.M., S.M.A., S.E.W.-B., M.O.H. and A.W.P. analyzed and interpreted translational data with support from M.H.A. and I.M.S. Funding was acquired by I.M.S., M.H.A. and M.-B.Z. J.W.K. wrote the manuscript with input from I.M.S., M.H.A., E.M. and A.W.P. All authors reviewed the manuscript, interpreted data and approved the final version. Data availability This clinical trial was registered at https://www.clinicaltrials.gov/ before patient enrollment (clinical trial identifier NCT03047928). Each vaccine was composed of 100 µg IO102, a 21-amino-acid peptide (DTLLKALLEIASCLEKALQVF) from the IDO peptide, and 100 µg IO103, a 19-amino-acid peptide (FMTYWHLLNAFTVTVPKDL) from the signal peptide of PD-L1. TCR sequencing data are available from Adaptive Biotechnologies. Upon request, the CCIT-DK office will provide a username and a password to access the designated data within approximately 2–4 weeks (https://www.herlevhospital.dk/ccit-denmark/find-us/Sider/Contact-information.aspx). All requests for the remaining data including raw data and analyzed data and materials will, within a reasonable time frame, be reviewed by the CCIT-DK office (https://www.herlevhospital.dk/ccit-denmark/find-us/Sider/Contact-information.aspx) to verify whether the request is subject to any intellectual property or obligations. Patient-related data not included in the paper were generated as part of clinical trials and may be subject to patient confidentiality. Any data and materials that can be shared will be released via a material-transfer agreement. The following database was used in the study: https://research.regionh.dk/da/publications/the-danish-metastatic-melanoma-database-dammed(32749d99-095f-4cae-b5de-769bae27f01e).html. Competing interests M.H.A. is named as an inventor on various patent applications relating to therapeutic uses of IDO and PD-L1 peptides. These patent applications are assigned to the company IO Biotech, which is developing immune-modulating cancer treatments. M.H.A. is a founder, shareholder and advisor for IO Biotech. I.M.S. is a cofounder, shareholder and advisor for IO Biotech. I.M.S. has an advisory board relationship with or lectured for Roche, Novartis, MSD, Celgene, Incyte, TILT Bio, Pfizer and BMS AstraZeneca and has received limited grants for translational research from BMS, Roche and Novartis. M.-B.Z. is the CEO, founder and shareholder at IO Biotech. E.M., A.W.P. and E. Ehrnrooth are employees at IO Biotech. E. Ellebaek has received honoraria from BMS, Pierre Fabre, Roche and Kyowa Kirin and travel support from MSD. M.D. has received honoraria from Genzyme, MSD, BMS, Roche and Novartis and travel support from Novartis, MSD, BMS, Roche and Pfizer. The remaining authors declare that they have no conflict of interest. Peer review information Nature Medicine thanks George Coukos, Michael Schell and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. Javier Carmona was the primary editor on this article and managed its editorial process and peer review in collaboration with the rest of the editorial team. Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. These authors contributed equally: Julie Westerlin Kjeldsen, Cathrine Lund Lorentzen. Change history 3/8/2022 A Correction to this paper has been published: 10.1038/s41591-022-01771-w

NIVOLUMAB

DrugsGivenReaction

CC BY

34887574

2021-12-09

Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Immune-mediated lung disease'.

A phase 1/2 trial of an immune-modulatory vaccine against IDO/PD-L1 in combination with nivolumab in metastatic melanoma. Anti-programmed death (PD)-1 (aPD1) therapy is an effective treatment for metastatic melanoma (MM); however, over 50% of patients progress due to resistance. We tested a first-in-class immune-modulatory vaccine (IO102/IO103) against indoleamine 2,3-dioxygenase (IDO) and PD ligand 1 (PD-L1), targeting immunosuppressive cells and tumor cells expressing IDO and/or PD-L1 (IDO/PD-L1), combined with nivolumab. Thirty aPD1 therapy-naive patients with MM were treated in a phase 1/2 study ( https://clinicaltrials.gov/ , NCT03047928). The primary endpoint was feasibility and safety; the systemic toxicity profile was comparable to that of nivolumab monotherapy. Secondary endpoints were efficacy and immunogenicity; an objective response rate (ORR) of 80% (confidence interval (CI), 62.7-90.5%) was reached, with 43% (CI, 27.4-60.8%) complete responses. After a median follow-up of 22.9 months, the median progression-free survival (PFS) was 26 months (CI, 15.4-69 months). Median overall survival (OS) was not reached. Vaccine-specific responses assessed in vitro were detected in the blood of >93% of patients during vaccination. Vaccine-reactive T cells comprised CD4+ and CD8+ T cells with activity against IDO- and PD-L1-expressing cancer and immune cells. T cell influx of peripherally expanded T cells into tumor sites was observed in responding patients, and general enrichment of IDO- and PD-L1-specific clones after treatment was documented. These clinical efficacy and favorable safety data support further validation in a larger randomized trial to confirm the clinical potential of this immunomodulating approach. pmcMain Despite remarkable advances in the treatment of MM with immune checkpoint inhibitors (ICIs) targeting PD-1 and cytotoxic T lymphocyte antigen 4 (CTLA-4), around half of patients are resistant to ICI monotherapy1. The combination of anti-CTLA-4 (aCTLA-4) and aPD1 therapy is to date the most effective therapy resulting in a response rate of around 60%; however, 50% of the patients also develop severe adverse events2,3. Therefore, an equally effective but less toxic treatment is highly needed. Several approaches to enhance ICI efficacy are currently being investigated, such as other ICIs, T cell therapy with tumor-infiltrating T cells or innate immunity stimulators such as Toll-like receptor 9 agonists4–6. Treating cancer patients with vaccines that stimulate a targeted immune response is another attractive approach, with very few side effects observed thus far in combination immunotherapy studies7,8. Immune-modulatory vaccines targeting tumoral immune escape mechanisms offer a new, generalizable strategy compared to patient-specific neoantigen cancer vaccines7,9. The immune-modulatory vaccine strategy in this clinical trial is based on the finding of circulating cytotoxic T cells specific to IDO and PD-L1 in the blood of patients with cancer and, to a lesser extent, in healthy donors. IDO- and PD-L1-specific CD8+ T cells can directly recognize and kill IDO+ and/or PD-L1+ tumor cells and likewise recognize and kill non-malignant cells that express their cognate targets. Furthermore, IDO- and PD-L1-specific CD4+ T cells release pro-inflammatory cytokines in response to IDO- or PD-L1-expressing target cells. IDO and PD-L1 are expressed not only by melanoma cells but also by many other cell types in the tumor microenvironment (TME), which differentiates these antigens from traditional tumor antigens used in other studies10–15. Activation of IDO/PD-L1-specific T cells by vaccination can therefore restrict the range of immunosuppressive signals mediated by immunosuppressive cells and thereby revert the TME from an immune hostile to an immune friendly environment. In animal models of cancer, vaccinations with IDO epitopes resulted in anti-tumor therapeutic effects that were correlated with reductions in IDO expression in myeloid cell populations within the TME16. The IDO/PD-L1 immune-modulating vaccine may lead to a translatable strategy for improving the efficacy of aPD1 therapy through activation of specific T cells. We hypothesize that the IDO/PD-L1 vaccine attracts T cells into the tumor, which induces type 1 helper T (TH1) cell inflammation and reverts the TME into an immune-permissive site, thereby turning the tumor ‘hot’. This would also upregulate PD-L1 expression in cancer and immune cells, generating more susceptible targets to aPD1 therapy (Extended Data Fig. 1a). This theory was confirmed in a mouse model in which aPD1 therapy and IDO vaccination show synergistic effects16. In this phase 1/2 clinical trial, MM1636, patients with MM received a combination of the IDO/PD-L1 (IO102/IO103) peptide vaccine with the adjuvant Montanide and the aPD1 antibody nivolumab. Patients were included in three cohorts: 30 aPD1 therapy-naive patients (cohort A), ten aPD1 therapy-refractory patients (cohort B, de novo resistance) and ten patients who progressed after aPD1 therapy (cohort C, acquired resistance). Here, we report results from cohort A. The vaccine was given biweekly for the first six administrations and thereafter every 4th week. A maximum of 15 vaccines were administered. Nivolumab was given in parallel, biweekly (3 mg per kg) or every 4th week (6 mg per kg) for up to 2 years (Extended Data Fig. 1b). The primary objective was safety and feasibility. Secondary objectives were immunogenicity and clinical efficacy. Results Patients and treatment Thirty patients were enrolled from December 2017 to June 2020. None of the 30 patients dropped out of the study; all received at least three cycles of therapy (Supplementary Fig. 1). At the current database lock (5 October 2020), six patients were still on treatment in the trial. Of the 24 patients who were not on trial treatment at data cutoff, two are still receiving nivolumab monotherapy (6 mg per kg every 4 weeks). Reasons for stopping treatment for the remaining 22 patients included disease progression (37%), toxicity (20%), maximum benefit or complete response (CR) confirmed on two consecutive scans (17%) or completing 2 years of treatment (7%). For the 24 patients who were not on trial treatment at data cutoff, the mean number of vaccinations was 10.5 (range, 3–15 vaccinations). Thirteen of these 24 patients continued nivolumab (6 mg per kg, every 4 weeks) as a standard of care. Nine patients received subsequent therapy after progression (Supplementary Table 1). Baseline characteristics are shown in Table 1. The mean age was 70 years; 37% of patients had elevated lactate dehydrogenase (LDH) levels, 60% were stage M1c, 37% had BRAF mutations, and 43% were negative for PD-L1 (<1%). A total of three patients (10%) had received prior ipilimumab therapy. No patients had brain metastasis (Supplementary Table 2).Table 1 Baseline patient characteristics (n = 30) Characteristic Number (%) Mean age, years (range) 70 (46–85) Sex, male 16 (55%) ECOG PS 0 26 (87%) LDH levels  ≤ULN 19 (63%)  >ULN 11 (37%) M stage (AJCC-8)  M1a 6 (20%)  M1b 6 (20%)  M1c 18 (60%) Number of lesion sites  1 6 (20%)  2–3 17 (57%)  >3 7 (23%) Liver metastases present 10 (33%) BRAF status  Mutant 11 (37%)  Wild type 19 (63%) PD-L1  <1% 13 (43%)  >1% 17 (57%) Previous systemic therapy  Ipilimumab 3 (10%)  No 27 (90%) AJCC-8, eighth edition of the American Joint Committee on Cancer; ECOG PS, Eastern Cooperative Oncology Group performance status; ULN, upper limit of normal. Notable clinical responses to the combination therapy Thirty patients with MM were treated with the IDO/PD-L1 vaccine and nivolumab according to the trial protocol. By investigator review, the ORR reached 80% (CI, 62.7–90.5%), with 43% of patients (CI, 27.4–60.8%) achieving a CR and 37% (CI, 20.9–54.5%) reaching a partial response (PR) as the best overall response, while 20% experienced progressive disease (PD) according to Response Evaluation Criteria in Solid Tumors (RECIST) 1.1 (Fig. 1a). Two of the patients with a PR did not have a confirmatory scan stating PR on two consecutive scans. Early onset of response was frequent, with 22 of 30 patients having an objective response at the first evaluation (after 12 weeks on treatment). Median times to PR and CR were 75 d (range, 54–256 d) and 327 d (range, 73–490 d), respectively (Fig. 2a–c).Fig. 1 Clinical response. a, Pie charts with percent ORR, CR, PR and PD according to RECIST 1.1 by investigator review of all patients (n = 30), PD-L1+ patients (>1%, (n = 17)) and PD-L1− patients (<1%, n = 13)), respectively. Two-sided CIs (95%) were constructed using the Clopper–Pearson method. b, Treatment effect in MM1636 compared with a matched historical control group from the DAMMED database (n = 74). Patients in MM1636 (n = 29) were matched with the exact same combination variable according to age (≤70 years, >70 years), sex, LDH levels (normal, elevated), M stage (M1a, M1b, M1c), BRAF status (wild type, mutated) and PD-L1 status (<1%, ≥1%). Estimates for treatment effects were calculated by weighted logistic regression analyses and weighted Cox proportional hazard model. Bar height indicates the estimated response rate; tops of bars are centers for error bars. Odds ratios (OR), response rates and their corresponding 95% CIs were extracted from the regression model. All P values were two sided, and P values below 0.05 were considered statistically significant. c, Best change in the sum of target lesion size compared with that at baseline (n = 30). The horizontal line at −30 shows the threshold for defining an objective response in the absence of non-target disease progression or new lesions according to RECIST 1.1. Two patients with 100% reduction in target lesion size had non-target lesions present. White stars, six patients had normalization (<10 mm) of fluor-18-deoxyglucose (FDG)-negative lymph nodes (at baseline, lymph nodes were >1.5 cm and FDG+) and 100% reduction of non-lymph node lesions and are considered to have had a CR (green bar). Black star, one patient (MM29) was considered to have had a PR (blue bar), although he had not reached a −30% change in target lesion size. The patient had a single measurable 13-mm lung metastasis at baseline and multiple biopsy-verified cutaneous metastases on the left crus (not detectable by positron-emission tomography–computed tomography (PET–CT) at baseline). The best change in target lesion size was 10 mm, and a post-treatment biopsy from the cutaneous metastases showed no sign of malignancy. Thus, overall, the patient was classified as having a PR. d, Kaplan–Meier curve of the response duration in the 24 patients with an objective response. e, Kaplan–Meier curve of PFS in all 30 treated patients. f, Kaplan–Meier curve of OS in all 30 treated patients. Fig. 2 Clinical response. a, Swimmer plot showing response duration and time to response according to RECIST 1.1 for all treated patients (n = 30). Triangles indicate first evidence of PR, while squares indicate first evidence of CR. Closed circles indicate time of progression. Arrows indicate ongoing responses. Patient MM18 died due to nivolumab-induced side effects. b, Spider plot showing response kinetics in all treated patients (n = 30). Red squares indicate time of progression. c, PET–CT images of patient MM42 before and after treatment (after 12 series of treatment) showing FDG metabolism in target lesions. SC, subcutaneous. The ORR among PD-L1+ (>1% (clone 28.8)) patients (n = 17) was 94.1% (CI, 73–99.7%) and 61.5% (CI, 35.5–82.3%) in PD-L1− patients (n = 13) (Fig. 1a). Objective responses were observed in patients irrespective of human leukocyte antigen (HLA) genotype (Supplementary Fig. 2). Clinical response data were validated by blinded independent external review, in which an ORR of 76.6% (CI, 57.7–90.1%) was reported, with 53.3% of patients achieving a CR, 23.3% achieving a PR and 3.3% experiencing stable disease. Comparisons between investigator review and external review are outlined in Supplementary Table 3. To examine whether the very high response rate should be attributed to nivolumab or to the vaccine, we retrieved clinical response information from a matched historical control group using the Danish Metastatic Melanoma Database (DAMMED) from contemporaneously treated patients with stage 3–4 melanoma who received aPD1 monotherapy17. Patients were matched with the exact same combination variable according to age, sex, PD-L1 status, BRAF status, LDH level and M stage (those at stage M1d were excluded from the control group (no patients with brain metastasis)). Matched controls were identified for 29 patients, and the ORR of 79.3% (CI, 61.0–90.4%) observed in MM1636 was found to be significantly higher (P < 0.0012) than that in the matched control group in which an ORR of 41.7% (CI, 31.0–53.3%) was reached. Furthermore, of the 29 patients in MM1636, a significantly (P < 0.0017) higher percentage (41.4% (CI, 25.2–59.6%)) of patients achieved CR in MM1636 than that (12% (CI, 6.3–21.6%)) in the matched historical control group. ORR and complete response rate (CRR) in the matched historical control group were comparable to those of patients treated in randomized phase 3 pivotal trials with aPD1 monotherapy18 (Fig. 1b). The treatment leads to prolonged PFS At data cutoff, the median duration of response had not been reached, with 87% of all responding patients being free from progression at 12 months (Fig. 1d). Patients were followed for up to 35 months with a median follow-up time of 22.9 months (CI, 14.9–26.2 months). OS and PFS were calculated from the first day of treatment to death or progression or to the date of the last follow-up (5 October 2020). The median PFS (mPFS) for all treated patients was 26 months (CI, 15.4–69 months) and was not reached for responding patients (Fig. 1e). The median OS was not reached at the data cutoff. OS at 12 months was 81.6% (CI, 61.6–92%) (Fig. 1f). One patient (MM18) with a CR died from nivolumab-related severe adverse events; the remaining patients died because of metastatic melanoma. For comparison, the mPFS was 8.3 months (CI, 5.5 months–NR (not reached)) in the matched historical control group (n = 74), while the median OS was 23.2 months (CI, 23.2 months–NR). (Extended Data Fig. 2a,b). The combination of the IDO/PD-L1 vaccine and nivolumab was safe Treatment-related adverse events are listed for all 30 patients in Supplementary Table 4. Common treatment-related grade 1–2 toxicities were fatigue (47%), rash (47%), arthralgia (30%), diarrhea (23%), nausea (23%), dry skin (20%), pruritus (20%), infusion reaction (17%), xerostomia (17%) and myalgia (17%). Four patients (13%) experienced grade 3–4 adverse events: one patient with a grade 3 maculopapular rash (MM01), one patient with grade 3 adrenal insufficiency (MM06) and one patient with grade 3 arthralgia (MM22). Patient MM18 died from urosepsis with multi-organ failure and severe hyponatremia. This patient had experienced multiple immune-related adverse events with grade 3 colitis, grade 2 pneumonitis, grade 3 arthralgia, grade 2 vasculitis and grade 2 nivolumab infusion-related allergic reaction. Additionally, patient MM18 had symptoms of myocarditis at the time of death with highly elevated cardiac troponin I levels. Bedside echocardiography showed an ejection fraction of 15%, which at baseline was 60%, but an autopsy was not conducted, and myocarditis was never confirmed pathologically. Patient MM06 had received first-line treatment with ipilimumab before entering the trial and was on substitution corticosteroids at the time of inclusion. Adrenal insufficiency was aggravated by an erysipelas infection with high fever, reaching grade 3 in Common Terminology Criteria for Adverse Events and resolving quickly after appropriate antibiotic therapy was initiated. As expected, local side effects were common with 77% of the patients who developed injection site reactions. These reactions were classified as granulomas (63%), redness (20%), pain (13%) and pruritus (13%) at the injection site. All local reactions were grade 1–2, most likely related to the Montanide adjuvant and typically transient. However, two patients (MM07 and MM20) decided to discontinue vaccination after eight and 11 injections, respectively, due to granulomas, tenderness and pain that limited instrumental activities of daily living but continued to receive nivolumab (Extended Data Fig. 2c). Vaccine-specific responses in blood were frequently detected First, all 30 patients were assessed for the presence of vaccine-specific responses in peripheral blood mononuclear cells (PBMCs) before, on and after vaccination using a modified interferon (IFN)-γ enzyme-linked immune absorbent spot (ELISPOT) assay. This assay is known to expand antigen-specific memory cells and to improve the detection and the correlative power of ELISPOT during treatment19–21. Prevaccine IDO-specific responses were detectable in ten (33%) patients, while prevaccine PD-L1-specific responses were detectable in eight patients (27%); overlapping (specific to both IDO and PD-L1) prevaccine responses were present in four (13.3%) patients. During vaccination, an increase of IDO-specific T cells or PD-L1-specific T cells in the blood was observed in 28 (93%) and 26 (86%) patients, respectively. In total, 93% of patients had an increase in either PD-L1- or IDO-specific responses on vaccination (Fig. 3a), with a significant (P < 0.0001) median increase from baseline to the post-vaccine response demonstrated for both IDO and PD-L1 (at different time points on treatment), confirming that vaccine-specific immune responses were induced in patients regardless of clinical response (Fig. 3b,c). Immune responses fluctuated in the blood over time (Extended Data Fig. 3a). An increase in IDO- and PD-L1-specific responses in the peripheral blood was also detectable directly ex vivo across the different clinical response groups (Extended Data Fig. 4).Fig. 3 Vaccine-specific responses in blood. a, IDO- and PD-L1-specific T cell responses in PBMCs at baseline and on vaccination as measured by the IFN-γ ELISPOT assay (n = 30). *Responses were calculated as the difference between the average numbers of spots in wells stimulated with IDO or PD-L1 peptide (triplicates) and those from the corresponding control (DMSO), and statistical analyses of ELISPOT responses were performed using a distribution-free resampling method (Moodie et al.50). DR (double response), response was not statistically confirmed due to replicate number, but the number of spots in peptide wells was two times higher than that in control wells (DMSO). NS, no significant response and no DR. For a detailed overview of responses at serial time points on vaccination, see Extended Data Fig. 4a. b, IDO- and PD-L1-specific T cell response in PBMCs in all treated patients measured by the IFN-γ ELISPOT assay at baseline and on vaccination. On-vaccination responses were selected from the ‘best’ ELISPOT response at different time points for each patient (series 3, 6, 12, 18 or 24) during vaccination (n = 30). Wilcoxon matched-pairs signed-rank test was used to compare responses to IDO or PD-L1 peptides in the vaccine between baseline and later time points. c, Representative example of ELISPOT wells with response in patient MM23 in serial PBMCs before and on treatment. d, IDO-specific CD4+ and CD8+ T cells were isolated and expanded from PBMCs stimulated in vitro with the IDO peptide and a low dose of IL-2 for 14–15 d before sorting using the Miltenyi Cytokine Secretion Assay—Cell Enrichment and Detection kit. To assess their cytolytic potential, IDO-specific T cells were stimulated with IDO peptide, and expression of CD107a, IFN-γ and TNF-α was assessed by flow cytometry (the example is from patient MM14). Sustained vaccine-specific responses were observed 3 and 6 months after the last vaccine, indicating induction of memory responses in nine patients with response to clinical treatment who surpassed follow-up at data lock (Extended Data Fig. 3b). Importantly, PD-L1- and IDO-specific responses were observed irrespective of HLA genotype (Supplementary Table 5). To verify the functionality of vaccine-induced T cells, IDO- or PD-L1-specific T cells were isolated and in vitro expanded from PBMCs of five patients. Phenotypic characterization by flow cytometry revealed that the isolated vaccine-specific T cells consisted of both CD4+ and CD8+ T cells. Also, both IDO- and PD-L1-specific CD4+ and CD8+ T cells showed pro-inflammatory properties, as they expressed the cytolytic marker CD107a and secreted the cytokines IFN-γ and tumor necrosis factor (TNF)-α (Fig. 3d and Extended Data Fig. 6a–d). Interestingly, we were also able to detect vaccine-specific CD4+ and CD8+ T cell responses in peripheral blood ex vivo (Extended Data Figs. 4 and 5). We observed a significant increase in the overall percentage of CD107a, CD137 and TNF-α expression in response to peptide stimulation in on- or post-treatment PBMC samples compared to that at baseline, further confirming the expansion and the diverse signature of vaccine-specific T cells (Extended Data Fig. 5). Vaccine-specific responses detected in the skin at the vaccination site To investigate whether vaccine-specific T cells have the potential to migrate to peripheral tissue, delayed-type hypersensitivity (DTH) tests were performed after six cycles of treatment on 15 patients to assess the presence of vaccine-reactive T cells in the skin. Supplementary Table 6 display an overview of skin-infiltrating lymphocyte (SKIL) cultures. We detected IDO-specific T cells in the skin of six of ten patients and PD-L1-specific T cells in nine of 11 patients (Extended Data Fig. 7a). Intracellular cytokine staining was performed on SKILs from five patients after stimulation with either PD-L1 or IDO peptide. Here, we detected mainly CD4+ peptide-reactive T cells that secreted TNF-α and upregulated CD107a, and a minor fraction also secreted IFN-γ. In one patient, we detected CD8+ PD-L1-reactive T cells (Extended Data Fig. 7b–d). Vaccine-induced T cells specifically recognize target cells To confirm the functionality of vaccine-expanded T cells, vaccine-specific T cell clones (clonal purity was confirmed by T cell receptor (TCR) sequencing) were isolated and expanded from patient PBMCs (Extended Data Fig. 8e). We showed that PD-L1-specific T cells were able to recognize PD-L1+ autologous tumor cells in a PD-L1 expression-dependent manner if the cancer cells also expressed HLA-II (Fig. 4a,b). Similarly, an HLA-DR-restricted IDO-specific CD4+ T cell clone was able to recognize an HLA-DR-matched IDO-expressing model cell line, MonoMac1, in an IDO-expression-dependent manner (Fig. 4e–g). As previously described, IDO- and PD-L1-specific T cells’ mode of action is not limited to targeting only cancer cells. We were able to show that vaccine-specific T cell clones also reacted against PD-L1- and IDO-expressing autologous immune cells (Fig. 4c,h). To provide myeloid cells with a tumor-associated phenotype we treated isolated CD14+ myeloid cells with tumor conditioned medium (TCM) derived from an established autologous tumor cell line. We observed that such TCM-treated CD14+ cells had increased expression of PD-L1 and IDO and were effectively recognized by autologous PD-L1- and IDO-specific CD4+ T cell clones (Fig. 4c,d,h,i).Fig. 4 PD-L1- and IDO-specific T cells from vaccinated patients react against PD-L1- and IDO-expressing target cells. a, Left, PD-L1-specific T cell culture (MM1636.05) reactivity against PD-L1 peptide or autologous tumor cells in the IFN-γ ELISPOT assay. Tumor cells were either not treated or pretreated with 200 U ml−1 IFN-γ for 48 h before the assay. Effector:target (E:T) ratio of 10:1 was used. Right, PD-L1 and HLA-II surface expression on melanoma cells with (green) or without (yellow) pretreatment with IFN-γ compared to an isotype control (gray) as assessed by flow cytometry. b, Left, PD-L1-specific T cell (MM1636.05) reactivity in the IFN-γ ELISPOT assay against autologous tumor cells pretreated with IFN-γ (500 U ml−1) and transfected with mock or PD-L1 small interfering (si)RNA 24 h after transfection. E:T ratio, 10:1. Right, PD-L1 surface expression on melanoma tumor cells (MM1636.05) assessed by flow cytometry 24 h after transfection with mock (blue) or PD-L1 (red) siRNA compared to the isotype control (gray). c, Reactivity of the CD4+ PD-L1-specific T cell clone (MM1636.14) against PD-L1 peptide or autologous CD14+ cells; E:T ratio, 10:1. CD14+ cells were isolated using magnetic bead sorting and used as targets in an ELISPOT assay directly or after pretreatment for 2 d with TCM derived from the autologous tumor cell line. d, Quantitative PCR with reverse transcription (RT–qPCR) analysis of PD-L1 (CD274) expression in sorted CD14+ cells before and after treatment with autologous TCM for 48 h. e, Reactivity of the IDO-specific CD4+ T cell clone (MM1636.23) against IDO peptide combined with HLA-DR (L243)-, HLA-DQ (SPV-L3)- or HLA-DP (B7/21)-blocking antibodies (aHLA-DR, aHLA-DQ and aHLA-DP) in an intracellular staining assay (ICS) for IFN-γ and TNF-α production. T cells were incubated with individual blocking antibodies (2 μg ml−1) for 30 min before adding IDO peptide. f, Reactivity of the IDO-specific CD4+ T cell clone (MM1636.23) against the HLA-DR-matched IDO-expressing cell line MonoMac1 transfected with mock or IDO siRNA in an ICS assay, E.T ratio, 4:1. siRNA transfection was performed 48 h before the experiment. g, RT–qPCR analysis of IDO1 expression in MonoMac1 cells 48 h after siRNA transfection. h, Reactivity of the CD4+ IDO-specific T cell clone (MM1636.14) against IDO peptide or autologous CD14+ cells; E:T ratio, 20:1. CD14+ cells were isolated using magnetic bead sorting and used as targets in an ELISPOT assay directly or after pretreatment with TCM derived from the autologous tumor cell line. i, RT–qPCR analysis of IDO1 expression in sorted CD14+ cells before and after treatment with autologous TCM for 48 h. Bars in RT–qPCR data (d,g,i) represent the mean of three (d,i) or six (g) technical replicates ±s.d.; P values were determined by two-tailed parametric t-tests. ELISPOT counts (a,b,c,h) represent the mean value of three technical replicates ±s.e.m.; response P values were determined using the distribution-free resampling (DFR) method. TNTC, too numerous to count. T cell clones in blood and tumors To track the role of treatment-induced T cell responses, TCR sequencing of the complementarity-determining region 3 (CDR3) was performed on five patients in peripheral blood (baseline and cycles 3, 6 and 12) and paired biopsies. These five patients (MM01, MM02, MM08, MM09 and MM13) were selected due to the availability of material and to investigate a balanced patient group with both responders and non-responders. Due to a limited number of available paired biopsies (either as a consequence of patient refusal or unexpectedly rapid and substantial clinical responses), statistical analysis could not be applied. Details on clinical response are shown in Fig. 2a. Additionally, PBMCs (on treatment) or SKILs were stimulated with the IDO/PD-L1 peptides, and then cytokine-producing T cells were sorted to track vaccine-induced T cells both in the periphery and at the tumor site. To identify enriched IDO/PD-L1-specific T cell clones, TCR rearrangements in sorted IDO/PD-L1-specific clones and TCR rearrangements in sorted IDO/PD-L1-specific T cell samples were compared to sequences from baseline PBMC samples for each patient. Clonal expansion of vaccine-specific TCR rearrangements from samples on vaccination were then tracked using a differential abundance framework. Cumulative IDO/PD-L1-specific T cell frequencies were tracked in post-treatment samples. We found no relation between clinical response and the enrichment of vaccine-specific clones, but an increase in IDO/PD-L1-specific T cell clones was observed at different time points in the periphery in all five patients (Extended Data Fig. 3c). We next investigated overall changes in the T cell repertoire in the blood. A modest increase in the peripheral T cell fraction was observed in the three responding patients at cycle 3, while the two non-responding patients had a clear decrease in T cell fraction (Extended Data Fig. 8a). We thereafter investigated TCR clonality and TCR repertoire richness, exploring the proportion of abundant clones and the number of unique rearrangements, respectively. A decreasing peripheral Simpson clonality and increasing TCR repertoire richness was observed in responding patients at cycle 3, which might indicate tumor trafficking upon treatment. The opposite pattern was observed in non-responding patients (Extended Data Fig. 8b,c). Peripherally expanded clones were associated with tumors and persisted until cycle 12 (latest time point analyzed). The largest peripheral expansion was observed at cycle 3, with the most significant increase observed in patient MM01 (CR). Responding patients had a larger fraction of peripherally expanded clones that were also found in tumors compared to that of non-responders. By tracking peripherally expanded clones detected at the tumor site, we observed that patient MM01 had a substantial increase after treatment, indicating tumor trafficking of peripheral expanded clones (Extended Data Fig. 8d). Influx at the tumor site of vaccine-enriched T cell clones Given the observation of increased T cell fraction and enrichment of IDO- and PD-L1-specific clones in the blood after treatment, we investigated whether the same trend was observed at the tumor site. Both TCR sequencing and immunohistochemistry (IHC) of paired biopsies from the five patients described above showed an increase in the T cell fraction with an influx of CD3+ and CD8+ T cells after treatment in the three responding patients (Fig. 5a–c). IHC was not possible for patient MM09 due to tissue loss.Fig. 5 Changes in the TME after treatment. Number of CD3+ and CD8+ T cells, TCR fraction, TCR clonality, TCR repertoire, biopsy expanded TCR clones and enrichment of IDO- and PD-L1-specific T cells at the tumor site. a, Number of CD3+ and CD8+ T cells per mm2 (sum in the validated area) at the tumor site detected by IHC of paired biopsies from four patients. b, Example of IHC of CD3+ and CD8+ T cells at the tumor site before and after treatment (cells per mm2) in one patient (MM01). c, T cell fraction at the tumor site at baseline and cycle 6 by TCR sequencing. The T cell fraction was calculated by taking the total number of T cell templates and dividing by the total number of nucleated cells. d, Tracking of vaccine-associated clones at baseline and cycle 6 in tumor biopsies. Cumulative frequencies of IDO and PD-L1 vaccine-specific TCR rearrangements are represented. e,f, TCR clonality and TCR repertoire richness in five patients at the tumor site at baseline and cycle 6. Simpson clonality measures how evenly TCR sequences are distributed among a set of T cells, where 0 indicates an even distribution of frequencies and 1 indicates an asymmetric distribution in which a few clones dominate. TCR repertoire richness reports the mean number of unique rearrangements. g, Bar chart representing baseline expanded biopsy clones from five patients (colored bars) and the detection of biopsy expanded clones also found in the blood at baseline and series 3, 6 and 12 (white and gray bars) by TCR sequencing. We thereafter investigated whether some of the IDO/PD-L1 vaccine-linked T cells were present at the tumor site. Vaccine-associated clones were tracked as the combined frequency of IDO- and PD-L1-specific T cell rearrangements. In biopsies, the frequencies at cycle 6 were compared to those at baseline and showed an increase in vaccine-specific T cells in four of five patients, irrespective of clinical response (Fig. 5d). TCR sequencing of PD-L1-specific SKILs (a more specific culture than IDO/PD-L1-specific isolations derived from PBMCs on vaccination) and paired biopsies showed that two of the top five PD-L1-specific SKIL clones were present at the tumor site both before and after treatment (Extended Data Fig. 7e,f). With a focus on the more abundant T cell clones, we investigated overall TCR clonality at the tumor site before and after treatment. In addition, we explored the number of unique TCR rearrangements, dissecting the lower-frequency clones. Patient MM01 had a significant increase in TCR clonality and a decrease in repertoire richness at the tumor site after therapy, indicating a focused tumor repertoire response of selected clones. All three responding patients had a decrease in TCR repertoire richness, which again might indicate a focused tumor response (Fig. 5e,f). Deeper analyses showed that the T cell clones that expanded at the tumor site after therapy were also present in the blood at baseline and increased significantly after treatment in four of five patients. The highest proportion was detected early at cycle 3. These data again support trafficking of peripherally expanded clones to the tumor site and could indicate that the T cell response to treatment is derived from pre-existing peripherally tumor-associated T cells (Fig. 5g). Signs of treatment-induced inflammation in the TME To dissect changes in the TME induced by T cell influx upon treatment in responding patients, RNA gene expression analyses using the nCounter PanCancer Immune Profiling Panel from NanoString were performed on paired biopsies from two responding patients (MM01 and MM13). Expression of genes related to adaptive immunity such as T cell activation, effector functions (genes encoding IFN-γ, TNF-α, IL-15 and IL-18) and cytotoxicity was increased in post-treatment biopsies (Extended Data Fig. 9a,b). Also, expression of genes related to checkpoint inhibitors such as those encoding T cell immunoglobulin and mucin-domain containing-3 (TIM-3), IDO, PD-L1, PD-L2, PD-1 and CTLA-4 increased after treatment, indicating activation of immune cells in the TME (Extended Data Fig. 9c). Additionally, IHC of paired biopsies from four patients (MM01, MM02, MM05 and MM13) showed an upregulation of PD-L1, IDO, MHC-I and MHC-II on tumor cells, indicating a treatment-induced pro-inflammatory response in the three responding patients, except for a decrease in MHC-II expression in patient MM13. By contrast, the non-responding patient MM02 had a reduction in T cell numbers present in the tumor after treatment and no expression of PD-L1, IDO or MHC-II and, interestingly, total loss of MHC-I, demonstrating tumor immune escape (Extended Data Fig. 10a). CD8+ T cells and their distance (µm) to PD-L1-expressing cells in baseline biopsies from five patients was investigated by IHC. Except for patient MM13 (PR), distance and clinical responses were associated: the two responders had reduced distance (<20 µm) between cells expressing these markers compared to non-responding patients (>80 µm). This observation indicates that responding patients not only have a higher intratumoral infiltration of CD8+ T cells, but that these cells can surround and attack PD-L1-expressing immune cells and tumor cells (Extended Data Fig. 10b). Discussion In this clinical trial, MM1636, 30 patients with metastatic melanoma were treated with a first-in-class immunomodulatory IDO/PD-L1-targeting peptide vaccine combined with nivolumab. The treatment led to an unprecedented high ORR of 80%, with 43% of patients reaching a CR, and a striking mPFS of 26 months (95% CI, 15.4–69 months) was reached. The vaccine represents a new treatment strategy to activate specific T cells that target cells contributing to immune suppression (including tumor cells), positively modulating the TME by inducing local inflammation. Indeed, we show that vaccine-specific T cells isolated and expanded from vaccinated patients recognize not only tumor cells in a target- and HLA-restricted manner but also myeloid cells polarized toward a tumor-associated phenotype. Hence, myeloid cells become targets for vaccine-activated T cells when they have a tumor-associated phenotype in the TME. These phenomena may further induce checkpoint molecules and rewire the TME toward an increasingly aPD1-permissive state. A drawback of our study is the single-center nonrandomized setup. Comparison between trials or between patients in trials and real-world patients is problematic due to multiple factors, such as period of conduction and other therapies available at the different periods of time. Nevertheless, the rate of investigator-assessed ORR in the phase 3 trial CheckMate 067 was 43.7% in the nivolumab monotherapy group and 57% in the nivolumab and ipilimumab group. CRs occurred in 8.9% and 11.5% of patients, respectively18. The mPFS of 26 months (95% CI, 15.4–69 months) in this trial is more than twice as long as that for patients treated with nivolumab and ipilimumab in CheckMate 067, for which an mPFS of 11.5 months (95% CI, 8.7–19.3 months) was reached. Patient baseline characteristics were in general comparable to those of patients with MM who have been treated in CheckMate 067, although patients in MM1636 were older (mean age, 70 years) and a larger fraction were positive for PD-L1 (57%)3,18,22. Among patients with PD-L1-negative tumors in MM1636, an ORR of 61.5% was still reached, which would be expected to be around 33% for first-line nivolumab monotherapy23. Some studies suggest that older patients might have a tendency to respond better to aPD1 therapy; however, this is still debated24–27. To address potential trial bias and the nonrandomized setup, patients in MM1636 were matched for age, performance status, sex, M stage, LDH level, PD-L1 status and BRAF status with a historical control group from the DAMMED, who were treated contemporarily (2015–2019) with aPD1 monotherapy as standard of care17. We found a significantly higher ORR and CRR in MM1636 compared to those of matched patients, who had an ORR of 43% and a CRR of 13%, comparable to patients treated in CheckMate 067. Restrictions of the synthetic control group are of course that it is partially historic and patient selection outside matching criteria cannot be ruled out28. Numerous contemporary clinical trials are exploring the combination of aPD1 therapy with other immunomodulating agents for advanced melanoma. Talimogene laherparepvec, an oncolytic virus, is approved by the Food and Drug Administration and the European Medicines Agency to treat advanced melanoma. A small phase 1b trial with 21 patients (MASTERKEY-265) combined talimogene laherparepvec and pembrolizumab to treat patients with advanced unresectable melanoma and reached an ORR of 62% and a CR of 33%29,30. Seventy-one percent of the patients in this trial had an M stage below M1c; this number was 40% in our trial. Furthermore, mainly patients with an M stage below M1c responded to treatment, which was not the case in MM1636. Results from a large randomized phase 3 trial are awaited (KEYNOTE-034). Epacadostat, an IDO inhibitor, was tested in combination with pembrolizumab in a nonrandomized phase 2 trial in 40 aPD1 treatment-naive patients with MM with promising results, reaching an ORR of 62%. Unfortunately, the phase 3 trial showed no indication that epacadostat provided improvement in PFS and OS31. Limitations of the phase 3 trial were the sparse information on pharmacodynamics as well as biomarker evaluation to improve the design. The IDO/PD-L1 vaccine is different from epacadostat, as it is not an IDO inhibitor but targets IDO- and PD-L1-expressing cells. Similar vaccines administered as monotherapy induced objective responses in lung cancer and basal cell carcinoma, while epacadostat as monotherapy in 52 patients resulted in no responses32,33. Sahin et al. recently published encouraging data from a first-in-human trial, in which a vaccine containing liposomal RNA targeting four unmutated tumor-associated antigens (NY-ESO-1, MAGE-A3, tyrosinase and TPTE) was administered alone or in combination with nivolumab in patients with advanced melanoma. Responses were observed in both the monotherapy group as well as the combination group in checkpoint inhibitor-experienced patients, suggesting the efficacy of non-mutant shared tumor antigen vaccines34. The overall safety and tolerability findings are comparable to those of aPD1 monotherapy. Injection site reactions were exclusive to the vaccine. However, these side effects were transient and mild in most patients and most likely due to the adjuvant Montanide. IDO- and PD-L1-specific CD8+ and CD4+ T cells exist among peripheral blood lymphocytes in healthy donors35–38 and expand in response to pro-inflammatory stimuli35,39. Furthermore, both IDO and PD-L1 are induced in cells as a counter response to the inflammatory response. This provides a mechanism that ensures immune homeostasis, which keeps IDO/PD-L1-specific T cells in check; therefore, the risk of triggering autoimmune-related adverse events by vaccination appears to be minimal. This was confirmed in the first clinical trials of IDO and PD-L1 vaccination (NCT01543464 (refs. 33,40) and NCT03042793 (ref. 41)). The spontaneous (baseline) immune response to the vaccines observed in the current study is in agreement with our previous observation in various patients with cancer35–38. Numerous vaccine-induced changes in the blood and at the tumor site were observed. Peripheral IDO- and/or PD-L1-specific T cells were detected in vitro in a modified ELISPOT in over 93% of patients on vaccination, unrelated to patient HLA type. Immune responses were persistent in patients who surpassed follow-up at data cutoff and were still detectable up to 6 months after the last vaccine, suggesting induction of memory T cells. We did not observe a correlation between vaccine-induced responses in blood and clinical responses. However, the detection of a highly significant increase in vaccine-specific T cell numbers after vaccination in almost all patients together with the very low number of patients with PD makes it difficult to expect such a correlation, especially because other aspects should be taken into account (for example, the loss of class I expression on tumors cells in patient MM02 with PD after vaccination). Frequencies of peripheral T cells induced against PD-L1 were overall higher than those against IDO. Importantly, however, we also observed that it was not the same patients who responded strongly to both IDO and PD-L1, nor did patients react with a similar response pattern to the two antigens; that is, different time points of response were observed. This suggests that each component of the vaccine plays different roles in the ongoing immune response in patients. The goal of IDO/PD-L1 vaccination was to modulate the TME to increase responsiveness to aPD1 therapy, as we have observed in animal models, in which immune conversion is demonstrated in the TME with an increased influx of T cells. Targeting both IDO and PD-L1 together enables synergy, as the TME is known to use different immune escape mechanisms and IDO and PD-L1 are often overexpressed by different cellular compartments. Ex vivo ELISPOT and flow cytometry assays confirmed the induction of immune response toward both epitopes and the higher immunogenicity of the PD-L1 epitope, although because of lower sensitivity failed to detect some of the responses. TCR sequencing of five patients confirmed enrichment of IDO/PD-L1-specific T cell clones in the blood at different time points after treatment. Furthermore, an increase in enriched IDO/PD-L1-specific clones was observed in four of five patients at the tumor site after treatment, irrespective of clinical response. Phenotypic characterization showed that vaccine-specific T cells, which were expanded in vitro with interleukin (IL)-2 from the blood of vaccinated individuals, were both CD4+ T cells and CD8+ T cells. This was confirmed by ex vivo phenotype description of vaccine-activated T cells. Vaccine-specific T cells expressed CD107a and CD137 and produced IFN-γ and TNF-α upon stimulation with the cognate target, indicating their cytolytic capacity. Overall, the data from immune monitoring supports the importance of both antigens in the generation of the frequent clinical responses in the study. Despite a limited number of paired biopsies (due to either patient refusal or the fact that a large fraction of responding patients had no assessable tumors after six cycles), we observed trends indicating treatment-induced general T cell influx in responding patients. It was shown that proliferation of CD8+ T cells in the tumor after aPD1 treatment is associated with radiographic reduction in tumor size42. Additionally, we showed that a large proportion of expanded peripheral TCR clones were associated with tumors and the most considerable amount of clonal expansion was observed early, at cycle 3. For the patient with a CR included in the TCR sequencing analyses (MM01), the number of peripheral expanded clones present at the tumor site increased after treatment compared to that at baseline, indicating tumor trafficking of peripheral expanded clones. Gene expression analyses (two paired biopsies) and IHC (five paired biopsies) further demonstrated that the combination treatment induced a pro-inflammatory TME in responding patients with signs of T cell activation and cytotoxicity and increased cytokine activity. This may lead to further upregulation of IDO, PD-L1, MHC-I and MHC-II on tumor cells, leading to more treatment targets. It was shown that, following vaccination with a cancer vaccine, PD-L1 expression is increased on tumor cells due to recruitment of tumor-specific T cells and upregulation of adaptive immune resistance pathways in the TME43. Treatment with nivolumab monotherapy enhances PD-L1 expression, and it is therefore problematic to discriminate the effect of the vaccine as compared to that of nivolumab44. IDO- and PD-L1-specific pro-inflammatory effector T cells were hypothesized to counteract the functions of IDO- and PD-L1-expressing immune-suppressive cells as a means to keep the immune balance between immune activation and inhibition45. However, because the expression of these molecules is induced as a counter-regulatory response to inflammatory mediators such as IFNs, they can also be expressed by, for example, activated myeloid cells or T cells. Thus, activation of PD-L1-specific T cells may result in depletion of activated T cells in the tumor and other sites. Importantly, it was recently described that PD-L1+ T cells mainly have tolerogenic effects on tumor immunity and exert tumor-promoting properties, suggesting that targeting this immune population is indeed also beneficial46. We have previously investigated the effect of activating PD-L1-specific T cells in vitro and in vivo and found that, overall, they supported the effector phase of an immune response by removing PD-L1-expressing regulatory immune cells that inhibit PD-1+ effector T cells47,48. The major role of the PD-1 pathway is believed to be the regulation of effector T cell responses. Thus, this protective pathway is more important after activation, rather than at the initial T cell-activation stage49. Accordingly, the presence of PD-L1-specific T cells during the activation phase of an immune response may not increase or support a pro-inflammatory immune response due to the expression of PD-L1 on potent antigen-presenting cells or on the T cells themselves14. Thus, the overall effects of PD-L1-specific T cells may vary depending on the expression of both PD-1 and PD-L1, that is, due to the microenvironment and the state of the immune response. In conclusion, here we report an impressive response rate, CR rate and mPFS for a first-in-class immune-modulating vaccine combined with nivolumab. This may be a first step toward a new treatment strategy for patients with MM. Limitations are the low number of patients treated at a single institution and the lack of a randomized design with aPD1 monotherapy as comparator. Studies in aPD1 therapy-resistant or -refractory melanoma are ongoing as well as biomarker analysis for selecting patients at a higher likelihood to benefit from the combination versus aPD1 monotherapy. A larger randomized trial will be essential to validate these findings and determine the specific contribution of the vaccine to clinical responses and changes in the TME. In December 2020, the Food and Drug Administration granted breakthrough therapy designation for the IO102/IO103 vaccine combined with aPD1 therapy in metastatic melanoma based on data from the MM1636 trial. Methods Trial design and treatment plan MM1636 is an investigator-initiated, nonrandomized, open-label, single-center phase 1/2 study. All patients were treated at the Department of Oncology, Herlev and Gentofte Hospital, University of Copenhagen, Herlev, Denmark. The study was conducted according to the Declaration of Helsinki and Good Clinical Practice (GCP) and monitored by the GCP unit in Copenhagen, Denmark. The protocol was approved by the ethical committee of the Capital region of Denmark (H-17000988), the Danish Medical Agencies (2017011073) and the Capital Region of Denmark Data Unit (P-2019-172). The study was registered at https://clinicaltrials.gov/ under the identifier NCT03047928 and at the EudraCT (no. 2016-0004527-23). This study initially aimed to include 30 aPD1 treatment-naive patients with MM. An amendment with the addition of two other cohorts with ten patients in each cohort was made to evaluate immune responses and clinical efficacy in aPD1 therapy-resistant patients (cohort B, de novo resistance and cohort C, acquired resistance) for a total of 50 patients. The amendment with cohorts B and C was approved at the inclusion of 18 patients in cohort A. The trial is still including patients in cohorts B and C. This article reports results from cohort A. The first six patients were treated as for phase 1, evaluating for safety and tolerability before the remaining 24 patients were included in phase 2. The IDO/PD-L1 vaccine was administered subcutaneously biweekly for the first 10 weeks and thereafter every 4 weeks for approximately 9 months. A maximum of 15 vaccines were administered. Nivolumab was administered according to the approved label (3 mg per kg biweekly) for 24 cycles. The 15th vaccine was administered together with the 24th nivolumab cycle of 3 mg per kg, and responding patients thereafter continued nivolumab monotherapy every 4th week (6 mg per kg) as a standard of care after investigator assessment. Treatment with nivolumab was discontinued at the maximum benefit (assessed by the investigator), after a maximum of 2 years of therapy, at progression or due to severe adverse events (Extended Data Fig. 1b, treatment plan). Vaccine composition Each vaccine was composed of 100 µg IO102, a 21-amino-acid peptide (DTLLKALLEIASCLEKALQVF) from IDO, and 100 µg IO103, a 19-amino-acid peptide (FMTYWHLLNAFTVTVPKDL) from the signal peptide of PD-L1 (PolyPeptide Laboratories). Peptides were dissolved separately in 50 µl DMSO, filtered for sterility and frozen at −20 °C (NUNC CyroTubes CryoLine System Internal Thread, Sigma-Aldrich). At <24 h before administration, peptides were thawed. The PD-L1 peptide was diluted in 400 µl sterile water and, immediately before injection, mixed with the IDO peptide solution and 500 µl Montanide ISA-51 (SEPPIC) to achieve a total volume of 1 ml. Patients Patients above 18 years of age with locally advanced or stage 4 melanoma according to the AJCC (seventh edition), at least one measurable lesion according to RECIST 1.1 and an ECOG PS of 0–1 were eligible. The main exclusion criteria were prior treatment with aPD1 therapy, CNS metastases >1 cm, severe comorbidities and active autoimmune disease. Enrollment was not restricted to PD-L1 status, but it was known before inclusion. Patients were included after informed consent. Patients MM42 and MM20 have confirmed their approval of PET–CT images and clinical images that have been included in the article. Key study assessments Safety and tolerability were evaluated based on changes in clinical laboratory analyses and reported adverse events. Adverse events were assessed according to Common Terminology Criteria for Adverse Events (version 5.0) and were graded from 1 to 5 for all treated patients up to 6 months after the last dose of the IDO/PD-L1 vaccine. Clinical efficacy was assessed using FDG PET–CT scans before treatment and every 3rd month until progression. Objective responses were categorized into CR, PR, stable disease or PD according to RECIST version 1.1. Clinical data were collected at CCIT-DK in the eCRF program OpenClinica version 1.0 and in Microsoft Excel version 2002 on a secure server. Blood samples for immunologic analyses were collected before treatment, before the third cycle, after the sixth, 12th, 18th and 24th cycles (on vaccination) and 3 and 6 months after the last vaccine. Two to three tumor needle biopsies (1.2 mm) were collected at baseline and after six cycles from the same tumor site, when accessible, to evaluate immune responses at the tumor site. DTH skin tests were performed and punch biopsies were taken from the DTH area after cycle 6 for the evaluation of SKILs reactive to PD-L1 and IDO (Supplementary Fig. 1, treatment plan). Statistical analysis of clinical outcome Survival curves were computed in GraphPad Prism software version 9.0.0 according to Kaplan–Meier method. Median follow-up time of enrollment was calculated using the reverse Kaplan–Meier method, also in GraphPad Prism software version 9.0.0. For binary outcomes, 95% two-sided CIs were constructed using the Clopper–Pearson method, also in GraphPad version 9.0.0. An independent board of certified and experienced oncoradiologists performed an external review to evaluate clinical response to address the potential bias of investigator site review. The external review took place at Rigshospitalet, Copenhagen University Hospital. This hospital did not participate in the MM1636 trial, and the external reviewer had no prior knowledge about the clinical trial or the trial therapy. Only PET–CT images were accessed, and arrows indicating target or non-target lesions appeared on baseline images as the only additional information. To address potential trial bias regarding treatment effect, we matched patients in MM1636 with patients from the DAMMED, a population-based database that retrospectively collects data on patients with metastatic melanoma in Denmark. Here, data from 938 patients treated with aPD1 monotherapy contemporaneously (January 2015–October 2019) were extracted. Two hundred and eighteen of these patients were eligible for comparison and matching (all parameters available) (Supplementary Table 1), and 74 patients from the DAMMED were found to match. Patients were matched for age (≤70 years, >70 years), sex, LDH levels (normal, elevated), M stage (M1a, M1b, M1c), BRAF status (wild type, mutated) and PD-L1 status (<1%, ≥1%). An exact matching algorithm was used in which patients in MM1636 were matched with patients from the DAMMED with the same combination of variables. Twenty-nine patients from MM1636 were matched with exact combinations of the six variables. One patient could not be matched. To secure balance of the calculations, control patients were weighted according to the number of patients for each MM1636 patient. Estimates for treatment effects were calculated by weighted logistic regression analyses and weighted Cox proportional hazard model. The R package ‘Matchlt’ was used for matching patients. As the method chosen for matching control patients to protocol patients, a weighted binary logistic regression model was used for comparing response rates in the two matched cohorts. Odds ratios and response rates, including their corresponding 95% CIs, were extracted from the regression models. All P values were two sided, and P values below 0.05 were considered statistically significant. SAS version 9.4M5 was used for the weighted logistic regression models. Processing of peripheral blood mononuclear cells Peripheral blood was collected from all patients in heparinized tubes and was processed within 4 h. In brief, PBMCs were isolated using Lymphoprep (Medinor) separation. PBMCs were counted on a Sysmex XP-300 analyzer and frozen in Human AB Serum (Sigma-Aldrich, H4522-100ML) with 10% DMSO using controlled-rate freezing (CoolCell, BioCision) in a −80 °C freezer and were moved the next day to a freezer at −140 °C until further processing. Needle biopsies at baseline and after the sixth vaccine Two to three needle biopsies (1.2 mm) were taken at baseline and after the sixth cycle of treatment when assessable from the same tumor lesion. One fragment was fixed with formalin and embedded in paraffin (FFPE); one to two fragments were used for expanding tumor-infiltrating T cells and establishing autologous tumor cell lines. Delayed-type hypersensitivity and generation of skin-infiltrating lymphocytes After six cycles of treatment, we performed intradermal injections of vaccine components without adjuvant and one control injection containing DMSO without peptide. Patients were injected with either a mixture of both IDO and PD-L1 peptides at all three injection sites or PD-L1 peptide, IDO peptide or a mixture of the two at the injection sites, respectively (Supplementary Table 6). Eight hours after injection, punch biopsies were resected from the three sites where the peptide was injected and transported immediately to the laboratory and cut into fragments of 1–2 mm3. SKILs were expanded to establish ‘young SKILs’ in CM medium consisting of RPMI 1640 with GlutaMAX, 25 mM HEPES, pH 7.2 (Gibco, 72400-021), IL-2 (100 or 6,000 IU ml−1) (Proleukin Novartis, 004184), 10% heat-inactivated Human AB Serum (Sigma-Aldrich, H4522-100ML), 100 U ml−1 penicillin, 1.25 µg ml−1 Fungizone (Bristol Myers Squibb, 49182) and 100 µg ml−1 streptomycin (Gibco, 15140-122). Half of the medium was replaced three times per week. Young SKILs from samples derived from IDO peptide, PD-L1 peptide and the mixture injection sites were further expanded in a small-scale version of the 14-d rapid expansion protocol as previously described51. Quantification of specific T cells in blood by ELISPOT For enumeration of vaccine-specific T cells in the peripheral blood, PBMCs from patients were stimulated with IDO or PD-L1 peptide in the presence of a low dose of IL-2 (120 U ml−1) for 7–13 d before being used in IFN-γ ELISPOT assays with 2.5–3.2 × 105 cells per well. Briefly, cells were placed in a 96-well PVDF membrane-bottomed ELISPOT plate (MultiScreen MSIPN4W50, Millipore) precoated with IFN-γ-capture antibody (clone 1-D1K, Mabtech). Diluted IDO or PD-L1 peptide stocks in DMSO were added at 5 µM; an equivalent amount of DMSO was added to control wells. PBMCs from each patient were set up in duplicate or triplicate for peptide and control stimulations. Cells were incubated in ELISPOT plates in the presence of the peptide for 16–18 h, after which they were washed off, and biotinylated secondary antibody (anti-human IFN-γ mAB, clone 7-B6-1, Mabtech) was added. After 2 h of incubation, unbound secondary antibody was washed off, and streptavidin-conjugated alkaline phosphatase (Mabtech) was added for 1 h. Next, unbound streptavidin-conjugated enzyme was washed off, and the assay was developed by adding BCIP/NBT substrate (Mabtech). ELISPOT plates were analyzed on the CTL ImmunoSpot S6 Ultimate V analyzer using ImmunoSpot software version 5.1. Responses were calculated as the difference between the average numbers of spots in wells stimulated with IDO or PD-L1 peptide and those from corresponding control wells. For detection of IDO and PD-L1 peptide responses ex vivo, PBMCs were thawed and rested overnight in medium containing DNase I (1 μg ml−1, Sigma-Aldrich, 11284932001) before being used in IFN-γ ELISPOT assays as described above. A total of 6–9 × 105 PBMCs per well were seeded. Statistical analysis of all ELISPOT responses was performed using the DFR method as described by Moodie et al. using RStudio software (RStudio Team, 2016, http://www.rstudio.com/)50. Vaccine-specific ELISPOT responses were defined as true if the difference between the spot count in control and peptide-stimulated wells was statistically significant according to DFR statistical analysis or, for samples performed in duplicate, if the spot count in peptide-stimulated wells was at least 2× the spot count in control wells50. Cancer cell lines and tumor conditioned medium Autologous melanoma cell lines were established from needle biopsies. Briefly, biopsies were chopped into small fragments and seeded in 24-well cultures in RPMI 1640 with GlutaMAX, 25 mM HEPES, pH 7.2 (Gibco, 72400-021), 10% heat-inactivated FBS (Life Technologies, 10500064), 100 U ml−1 penicillin, 1.25 µg ml−1 Fungizone (Bristol Myers Squibb, 49182) and 100 µg ml−1 streptomycin (Gibco, 15140-122). Established adherent melanoma tumor cell lines were cryopreserved at −140 °C in freezing medium containing FBS with 10% DMSO. PD-L1 and HLA-II expression on established tumor cell lines was assessed by flow cytometry staining with anti-PD-L1–PE-Cy7 (1:23 dilution, clone M1h1 (RUO), BD, 558017) and anti-HLA-II–FITC (1:23 dilution, clone Tu39 (RUO), BD, 555558) antibodies. To obtain TCM, established tumor cell lines were cultured in 175-cm2 Nunc cell culture flasks until 80–90% confluency was reached. The culture medium was then replaced with 20 ml fresh X-VIVO 15 with Gentamicin and Phenol Red (Lonza, BE02-060Q), medium with 5% heat-inactivated Human AB Serum (Sigma-Aldrich, H4522-100ML). After 24 h of incubation, TCM was collected and centrifuged to remove any resuspended cells, after which TCM was aliquoted, frozen and stored at −80 °C. The acute monocytic leukemia cell line MonoMac1 was obtained from the DSMZ—German Collection of Microorganisms and Cell Cultures (ACC 252) and cultured in RPMI 1640 with GlutaMAX, 25 mM HEPES, pH 7.2 (Gibco, 72400-021) and 10% heat-inactivated FBS. Isolation of autologous myeloid cells Autologous CD14+ cells were sorted from freshly thawed PBMCs using a magnetic bead-separation kit (Miltenyi Biotec, 130-050-201) according to the manufacturer’s instructions. Isolated CD14+ cells were used as targets in the IFN-γ ELISPOT assay directly after sorting or differentiated in vitro into tumor-associated macrophages by culturing with 1 ml fresh X-VIVO 15 medium with Gentamicin and Phenol Red (Lonza, BE02-060Q) and 5% heat-inactivated Human AB Serum, supplemented with 1 ml autologous TCM in 24-well plates for 2 d. Quantification of vaccine-specific T cells from DTH biopsy sites by ELISPOT SKILs expanded in vitro were rested in medium without IL-2 overnight before being used in the IFN-γ ELISPOT assay as described above to evaluate reactivity of skin-infiltrating T cells. Generation of IDO- and PD-L1-specific T cell cultures from PBMCs or SKILs IDO- or PD-L1-specific T cells were isolated from peptide-stimulated in vitro PBMC cultures on days 14–15 after stimulation or from SKIL cultures expanded in vitro. For specific T cell isolation, PBMCs or SKILs were stimulated with IDO or PD-L1 peptide, and cytokine-producing T cells were sorted using the IFN-γ or TNF-α Secretion Assay—Cell Enrichment and Detection kit (Miltenyi Biotec). Cytokine-production profile of PD-L1- and IDO-specific T cells by intracellular staining To assess the T cell cytokine-production profile, isolated and expanded IDO- and PD-L1-specific T cell cultures were stimulated for 5 h with peptide at 5 μM in a 96-well plate. One hour after the start of the incubation, anti-CD107a–PE (1:133 dilution, clone H4A3, BD Biosciences, 555801) antibody and BD GolgiPlug (1:1,000 dilution, BD Biosciences) were added. After a 5-h incubation, cells were stained using fluorescently labeled surface marker antibodies: anti-CD4–PerCP (1:5 dilution, clone SK3, 345770), anti-CD8–FITC (1:5 dilution, clone SK1, 345772) and anti-CD3–APC-H7 (1:20 dilution, clone SK7, 560275) (all from BD Biosciences). Dead cells were stained using FVS510 (1:250 dilution, BD Biosciences, 564406), followed by overnight fixation and permeabilization using eBioscience Fixation/Permeabilization buffers (00-5123-43, 00-5223-56) according to the manufacturer’s instructions. Cells were then stained intracellularly in eBioscience permeabilization buffer (00-8333-56) with anti-IFN-γ–APC (1:20 dilution, clone 25723.11, BD Biosciences, 341117) and anti-TNF-α–BV421 (1:100 dilution, clone Mab11, BD Biosciences, 562783) antibodies. Samples were analyzed on the FACSCanto II (BD Biosciences) using BD FACSDiva software version 8.0.2. The gating strategy is shown in Supplementary Fig. 3a. For assessing the reactivity against IDO expressing MonoMac1 cells, the appropriate amount of cancer cells was added to the IDO-specific T cells to obtain an effector:target ratio of 4:1 in the 96-well plate and T cell cytokine production was tested after 5 h of co-culture as described above. For HLA-blocking experiments, HLA-blocking antibodies (2 μg ml−1) were added directly into wells 30 min before the addition of peptide. Blocking antibodies used were HLA-DR (1:500 dilution, clone L243, Abcam, ab136320), HLA-DQ (1:500 dilution, clone SPV-L3, Abcam, ab23632) and HLA-DP (1:500 dilution, clone B7/21, Abcam, ab20897). To assess ex vivo T cell reactivity to IDO and PD-L1 peptides in patient PBMCs, cells were thawed and rested for 1–2 d in medium containing DNase I (1 μg ml−1, Sigma-Aldrich, 11284932001). PBMCs were then stimulated with peptide at 5 μM in a 96-well plate for 8 h. An hour after the addition of peptide, anti-CD107a–BV421 (3:50 dilution, clone H4A3, BD Biosciences, 328626) antibody and BD GolgiPlug (1:1,000 dilution, BD Biosciences) were added. Surface and intracellular staining was performed as described above. Antibodies used for surface staining were anti-CD3–PE-CF594 (0.8:30 dilution, clone UCHT1, BD Biosciences, 562280), anti-CD4–BV711 (1:30 dilution, clone SK3, BD Biosciences, 563028) and anti-CD8–Qdot605 (1:150 dilution, clone 3B5, Thermo Fisher, Q10009). Antibodies used for intracellular staining were anti-CD137–PE (1:20 dilution, clone 4B4-1 (RUO), BD Biosciences, 555956), anti-IFN-γ–PE-Cy7 (1.5:20 dilution, clone B27, BD Biosciences, 557643) and anti-TNF-α–APC (1:20 dilution, clone Mab11, BD Biosciences, 554514). Samples were acquired on the NovoCyte Quanteon (ACEA Biosciences) and analyzed using NovoExpress software version 1.4.1. To assess vaccine-specific T cell responses, background values observed in unstimulated PBMC samples were subtracted from values observed in peptide-stimulated conditions. Positive response value threshold was set at a difference of 0.2% from the background values. Based on this response cutoff, only TNF-α, CD107a and CD137 responses were detected in this assay. Statistical analysis comparing baseline with on-treatment or post-treatment cytokine profiles was performed using two-sided Wilcoxon matched-pairs signed-rank test. The gating strategy is shown in Supplementary Fig. 3b. siRNA-mediated PD-L1 and IDO1 silencing A stealth siRNA duplex for targeted silencing of PD-L1 (Invitrogen)52, a custom silencer select siRNA for targeted silencing of IDO1 (Ambion) and the recommended silencer select negative control (Ambion) siRNA for mock transfection were used. The stealth PD-L1 siRNA duplex consisted of the sense sequence 5′-CCUACUGGCAUU-UGCUGAACGCAUU-3′ and the antisense sequence 5′-AAUGCGUUCAGCAAAUGCCAGUAGG-3′. The three silencer IDO siRNA duplexes used were siRNA1 (sense sequence, 5′-ACAUCUGCCUGAUCUCAUATT-3′; antisense sequence, 5′-UAUGAGAUCAGGCAGAUGUTT-3′), siRNA2 (sense sequence, 5′-CCACGAUCAUGUGAACCCATT-3′; antisense sequence, 5′-UGGGUUCACAUGAUCG-UGGAT-3′) and siRNA3 (sense sequence, 5′-CGAUCAUGUGAACCCAAAATT-3′; antisense sequence, 5′-UUUUGGGUUCACAUGAUCGTG-3′). For PD-L1 or IDO1 silencing, cancer cells were electroporated with 0.025 nmol of each siRNA duplex as previously described53. For PD-L1-silencing experiments, cancer cells were treated with IFN-γ (500 U ml−1, PeproTech) 1 h after electroporation. Electroporated cells were used as target cells in ELISPOT and ICS assays 24 h or 48 h after siRNA electroporation. RT–qPCR Total RNA was extracted using the RNeasy Plus Mini kit (Qiagen, 74134) following the manufacturer’s instructions. RNA concentration was quantified using a NanoDrop 2000 (Thermo Fisher Scientific), and a total of 1,000 ng RNA was reverse transcribed with the High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems, 4368814) using 1,000 ng input RNA for transcription. Real-time qPCR analyses were performed using the TaqMan Gene Expression Assay on a Roche LightCycler 480 instrument. RT–qPCR assays were performed with a minimum of three technical replicates and analyzed using the ΔCt method as described in Bookout et al.54 with normalization of IDO1 expression (primer ID, Hs00984148_m1) or PD-L1 expression (primer ID, Hs001125296_m1) to the expression level of the housekeeping gene POLR2A (primer ID, Hs00172187_m1) and the control sample (mock). For low-concentration samples with no amplification, Ct was set to 40. Controls without reverse transcriptase were used as controls for specific amplifications. P values were determined using two-tailed parametric t-tests. BRAF-mutational status and PD-L1 status at baseline in all patients A library of historical FFPE biopsies were assessable for all patients and analyzed locally at Herlev and Gentofte University Hospital by experienced pathologists for BRAF status and PD-L1 expression on tumor cells. The BRAF analysis was carried out with real-time PCR using the EntroGen BRAF Mutation Analysis Kit II (BRAFX-RT64, CE-IVD) to specifically detect mutations corresponding to V600D, V600E and V600K in the BRAF protein. PD-L1 status was assessed using the monoclonal rabbit anti-PD-L1 antibody (clone 28.8, PD-L1 IHC 28-8 pharmDx) in FFPE biopsies following the manufacturer’s instructions. Patients were considered positive for PD-L1 with expression levels ≥1% and negative for PD-L1 with expression levels <1%. Human leukocyte antigen type Blood samples from all 30 patients were genotyped for class I (HLA-A, HLA-B, HLA-C) and three class II types (HLA-DRB1, HLA-DQA1, HLA-DQB1) using LinkSēq HLA Typing Kits (Thermo Fisher, 1580C). These test kits are based on real-time PCR using allele-specific exponential amplification (sequence-specific primers) followed by melting curve analyses. Immunohistochemistry simplex: Immunoscore CR IHC staining was performed using a qualified Ventana Benchmark XT with four different steps: (1) antigen retrieval; (2) staining with the following primary antibodies: anti-CD3 antibody (clone HDX2; provider, HalioDx; HD-FG-000013; 10931065/10636667; concentration, 0.25 µg ml−1), anti-CD8 antibody (clone HDX3; provider, HalioDx; HD-FG-000019; 10931069/10337710/10639301; concentration, 0.5 µg ml−1), anti-IDO monoclonal antibody (clone VINC3IDO; provider, Thermo Fisher Scientific; 14-9750-82; E25003-101; concentration, 0.05 µg ml−1), anti-HLA class 1 ABC antibody (clone, EMR8-5; provider, Abcam; ab70328; GR3248333/GR3186494; concentration, 0.5 µg ml−1), anti-HLA-DR/DP/DQ/DX antibody (clone CR3/43; provider, Santa Cruz; sc-53302; L1714; concentration, 1 µg ml−1) and anti-PD-L1 antibody (clone HDX3; provider, HalioDx; HD-FG-000035; 106312810/106312816; concentration, 3.3 µg ml−1); (3) detection with a secondary antibody using the ultraView Universal DAB Detection kit; and (4) counterstaining using Hematoxylin & Bluing Reagent (staining of cellular nuclei). Control slides were systematically included in each staining run to permit quality control of the obtained measurements. Following coverslipping, slides were scanned with the NanoZoomer-XR to generate digital images (20×). Two consecutive slides were specifically used to perform Immunoscore CR TL staining of CD3+ and CD8+ cells. Digital pathology of T lymphocytes The digital pathology for Immunoscore CR TL allowed the quantification of positive cells (in cells per mm2) into the core tumor and invasive margin if present. Each sample was analyzed using the HalioDx Digital Pathology Platform. Digital pathology and pathologist analysis of IDO Digital pathology for IDO allowed the quantification of stained area (in mm2) into the whole tumor. Each sample was analyzed using the HalioDx Digital Pathology Platform. Analysis of IDO+ cells was performed by a pathologist. Results were expressed as an H score from 0 to 300. The score is obtained by the formula 3 × percentage of cells with strong staining + 2 × percentage of cells with moderate staining + percentage of cells with weak staining. Digital pathologist analysis of MHC-I and MHC-II Analysis of MHC-I+ and MHC-II+ cells was performed by a pathologist. Results are expressed as a percentage of positive tumor cells and a percentage of positive cells in the stroma. Digital pathology Immunoscore immune checkpoint (CD8+ and PD-L1+) The Immunoscore CR IC test allowed the quantification of CD8++ cell density in the whole tumor and a CD8+-centered proximity index (which corresponds to the percentage of CD8++ cells that have at least one PD-L1+ cell in the neighborhood) at different cutoff distances (20 µm, 40 µm, 60 µm and 80 µm). Pathologist analysis of PD-L1 A pathologist performed analysis of PD-L1+ cells. Positivity of a viable tumor cell was considered when partial or complete cell membrane staining was observed (more than 10% of the tumor cell membrane). Results were expressed as a percentage. NanoString RNA profiling and Immunosign RNA was extracted from FFPE tissues using Qiagen RNeasy FFPE extraction kits (Qiagen). Annotations from the pathologist performing H&E staining were used to guide removal of normal tissue from the slides by macrodissection before nucleic acid extraction, which occurred after tissue deparaffinization and lysis. Each extracted RNA sample was independently quantified using a NanoDrop spectrophotometer (NanoDrop Technologies) and qualified (Agilent Bioanalyzer). Degradation was quantified as the percentage of RNA fragments smaller than 300 bp using the RNA 6000 Nano kit (Agilent Bioanalyzer). Good sample quality was defined as less than 50% of RNA fragments being 50–300 bp in size. RNA expression profiling was performed using the nCounter PanCancer Immune Profiling Panel from NanoString (NanoString Technologies). The PanCancer Immune Profiling Panel contains 776 probes and is supplemented with six genes to complete HalioDx Immunosign targets. Hybridization was performed according to the manufacturer’s instructions. Hybridized probes were then purified and immobilized on a streptavidin-coated cartridge using the nCounter Prep Station (NanoString Technologies). Data collection was carried out on the nCounter Digital Analyzer (NanoString Technologies) following the manufacturer’s instructions to count individual fluorescent barcodes and quantify target RNA molecules present in each sample. For each assay, a scan of 490 fields of view was performed. Raw data from the NanoString nCounter were processed using NanoString recommendations. Quality control enables to retain data of good quality with a binding density that ranges between 0.05 and 2.25. The linearity of positive controls was checked using the R2 value of regression between the counts and the concentrations of positive controls. Samples that showed R2 < 0.75 were flagged and removed from the analysis. The background was removed using a thresholding method at the mean + 2 s.d. of negative controls. Raw counts were normalized using a positive normalization factor. Samples showing positive normalization factors outside the range of 0.3–3 were removed from the analysis. A second normalization was performed using the housekeeping gene normalization factor. Only the most stable housekeeping genes were selected for this normalization step using the variance-versus-mean relationship. All samples showing a normalization factor outside the range of 0.1–10 were removed from the analysis. All statistical analyses were performed on normalized counts using R software (version 2.6.2, 2019-12-12). T cell receptor variable β chain sequencing To track longitudinal immune responses to therapy, genomic DNA was extracted from longitudinal pre- and post-treatment PBMCs (five patients), pre- and post-treatment biopsies (five patients) (both FFPE and treated with RNAlater) and IDO- and PD-L1-specific T cell cultures from PBMCs (five patients) or SKILs (one patient). Three clonal (two IDO- and one PD-L1-) specific cultures were also sequenced to confirm clonal purity of cultures. DNA from PBMCs or RNAlater-treated biopsies was extracted with the DNeasy Blood and Tissue kit (Qiagen, 69504), DNA from sorted IDO and PD-L1-specific T cells from either PBMCs or SKILs was extracted using the QIAamp DNA Micro kit (Qiagen, 565304), and DNA from PFFE biopsies was extracted using the Maxwell RSC DNA FFPE kit (Promega, AS1450). Immunosequencing of the CDR3 regions of human TCRβ chains was performed using the immunoSEQÒ Assay (Adaptive Biotechnologies). Extracted genomic DNA was amplified in a bias-controlled multiplex PCR, followed by high-throughput sequencing. Sequences were collapsed and filtered to identify and quantitate the absolute abundance of each unique TCRβ CDR3 region for further analysis as previously described55–57. Statistical analyses of TCRβ sequencing results Two quantitative components of diversity were compared across samples in this study. First, Simpson clonality was calculated on productive rearrangements by ∑i=1Rpi2, where R is the total number of rearrangements and pi is the productive frequency of rearrangement i. Values of Simpson clonality range from 0 to 1 and measure how evenly receptor sequences (rearrangements) are distributed among a set of T cells. Clonality values approaching 0 indicate a very even distribution of frequencies, whereas values approaching 1 indicate an increasingly asymmetric distribution in which a few clones are present at high frequencies. Second, sample richness was calculated as the number of unique productive rearrangements in a sample after computationally downsampling to a common number of T cells to control for variation in sample depth or T cell fraction. Repertoires were randomly sampled without replacement five times, and we report the mean number of unique rearrangements. The T cell fraction was calculated by taking the total number of T cell templates and dividing by the total number of nucleated cells. The total number of nucleated cells was derived from reference genes using the immunoSEQ Analyzer version 3. To identify enriched vaccine clones in each patient, rearrangement frequencies in their baseline PBMCs and each IDO/PD-L1-sorted T cell sample were compared using a binomial distribution framework as previously described58. In brief, for each clone, we performed a two-sided test to determine whether frequencies were the same in the patient’s periphery and a PD-L1- or IDO-specific T cell sample. The Benjamini–Hochberg procedure was used to control the false discovery rate at 0.01 (ref. 59). Clonal expansion in post-treatment samples was similarly assessed using this differential abundance framework, but an IDO/PD-L1-specific T cell sample was replaced with a post-treatment series sample. In biopsies, the six series frequencies were compared to those of baseline tissue. Lastly, vaccine-associated clones were tracked in each PBMC and tissue sample by summing the frequency of each rearrangement enriched in either PD-L1- or IDO-specific T cells. All statistical analyses were performed in R version 3.6.1. Reporting Summary Further information on research design is available in the Nature Research Reporting Summary linked to this article. Online content Any methods, additional references, Nature Research reporting summaries, source data, extended data, supplementary information, acknowledgements, peer review information; details of author contributions and competing interests; and statements of data and code availability are available at 10.1038/s41591-021-01544-x. Supplementary information Supplementary Information Supplementary Figs. 1–3 and Tables 1–6 Reporting Summary Extended data Extended Data Fig. 1 Mode of Action and treatment plan. a) Anticipated mechanism of action of the combination therapy of an IDO/PD-L1 derived peptide vaccine and nivolumab (anti-PD1). 1) The IDO/PD-L1 peptide vaccine is administered subcutaneously (s.c) and nivolumab is administered intravenous (IV). 2) The vaccines peptides are phagocytosed by an antigen presenting cell and presented to IDO and PD-L1 specific T cells, which are activated. 3) The activated T cells migrates to the tumor site where they attack both immune-suppressive cells and tumour cells expressing IDO and/or PD-L1 leading to cytokine production and a pro-inflammatory tumour microenvironment. 4/5) Enhanced tumour killing by both IDO/PD-L1 specific T cells and tumor specific cytotoxic T cells due to PD-1 blockade. Created with BioRender.com b) Treatment plan. After written informed content patients were screened. Before treatment start a baseline PET/CT scan was performed, baseline blood sample for research use and if assessable a baseline needle biopsy. Patients were treated with the IDO/PD-L1 peptide vaccine subcutaneously biweekly for the first 6 injections and thereafter every fourth week for a maximum of 15 vaccines. Nivolumab was administered in parallel biweekly (3mg/kg) up to 24 series. If patients needed subsequent nivolumab after ended vaccination regiment they were treated with 6 mg/kg every fourth week up to two years. Needle biopsy and delayed type hypersensitivity (DTH) was performed after 6 series of treatment if assessable. PET/CT scans was performed every third month. Extended Data Fig. 2 Progression free survival and overall survival in matched historical control group and vaccine injection site reaction. a) Kaplan-Meier curve of progression free survival with corresponding 95% confidence intervals in the matched historical control group (n=74). Patients were matched on BRAF-status, PD-L1-status, age, gender, M-stage and LDH level). b) Kaplan-Meier curve of overall survival with corresponding 95% confidence intervals in the matched historical control group (n=74). c) Images of injection site reaction in patient MM20 (CR) after 11 vaccination show redness, rash and granuloma at injection site. Extended Data Fig. 3 Vaccine specific responses in blood. a) Heatmap of specific (background has been subtracted) IDO and PD-L1 responses in PBMCs at baseline, series 3, 6, 12, 18 and 24 measured by in vitro IFN-γ Elispot assay show fluctuations in the blood during treatment (n=30). b) Heatmap of specific (background has been subtracted) IDO and PD-L1 responses in PBMCs at baseline and 3 and 6 months after last vaccine measured by in vitro IFN- γ Elispot assay (n=30). 2.5–3.2x105 cells per well were used. c) Vaccine associated clones were tracked in the blood by summing the frequency of each rearrangement enriched in either IDO or PD-L1 T cells. Extended Data Fig. 4 Ex vivo vaccine specific responses in blood. Heatmaps of detected specific (background has been subtracted) IDO (a) and PD-L1 (b) responses in PBMCs at baseline and on/after treatment as measured by IFNγ ELISPOT (n=25). C) Example of well images of ex vivo ELISPOT wells for three different patients (n=3). 6-9x105 cells per well were used. Extended Data Fig. 5 CD4 and CD8 vaccine specific T cell responses in blood. Top: heatmaps of IDO specific CD4 (left) and CD8 (right) T cell responses in PBMCs at baseline and on/post treatment. Bottom: heatmaps of PD-L1 specific CD4 (left) and CD8 (right) T cell responses. Responses quantified by flow cytometry by an increased expression of CD107α, CD137 and TNFα after 8h peptide stimulation. Values represent specific responses after background values have been subtracted (n=21). Statistical analysis were performed using two-sided Wilcoxon matched-paired rank test. Extended Data Fig. 6 Pro-inflammatory profiles of sorted CD4 and CD8 T cells from blood. a/b/c/d) Percentage of in vitro stimulated and sorted PD-L1 and IDO specific CD4 and CD8 T cells secreting cytokines. Extended Data Fig. 7 Vaccine specific responses in skin. a) IDO and PD-L1 specific T cell responses in SKILs after 6 series of treatment measured by IFN-γ Elispot assay (n=13). SKILs were grown from DTH injection with either IDO peptide, PD-L1 peptide or a mix as presented by different blue colours. * Responses were calculated as the difference between average numbers of spots in wells stimulated with IDO or PD-L1 peptide (triplicates) and corresponding control (DMSO) and statistical analyses of Elispot responses were performed using distribution-free resampling method (Moodie et al.). DR: Not statistically confirmed response due to replicate number but number of spots in peptide wells are two times higher than control wells (DMSO). NS: No significant response and no DR. b/c/d) Percentage of cytokine secreting/CD107a+ CD4+ and CD8+ IDO and PD-L1 specific T cells in response to in vitro peptide stimulation by flow cytometry. e) Skin infiltrating PD-L1 specific T cell clones also found in biopsy in patient MM01. TCR sequencing was performed on a PD-L1 specific T cell culture generated from DTH area on the lower back injected with PD-L1 peptide on patient MM01. Bars show the frequency of top 25 clones in the culture with which indicates a high Simpson clonality of 0.43. f) Tracking the frequency of the top five skin infiltrating PD-L1 specific clones in tumour before and after treatment. Extended Data Fig. 8 T cell changes in blood after treatment. T cell fraction, TCR clonality and repertoire richness in blood and peripherally expanded TCR clones in both responding and non-responding patients. a) T cell fraction in peripheral blood of five patients at baseline, series 3, 6 and 12 by TCR sequencing. T cell fraction was calculated by taking the total number of T cell templates and dividing by the total number of nucleated cells (n=5.) b) TCR clonality in peripheral blood of five patients at baseline, series 3, 6 and 12 by TCR sequencing (n=5). Simpson clonality measures how evenly TCR sequences are distributed amongst a set of T cells where 0 indicate even distribution of frequencies and 1 indicate an asymmetric distribution where a few clones dominate. c) TCR repertoire richness in peripheral blood of five patients at baseline, series 3, 6 and 12 by TCR sequencing (n=5). TCR repertoire richness report the mean number of unique rearrangements. d) Bar-chart chart representing dynamics of the expanded T cell clones. Coloured bars represent peripherally expanded clones in five patients at series 3, series 6 and series 12 as compared to the baseline PBMC samples (n=5). Light gray bar represents peripherally expanded clones that are present in baseline biopsy samples while dark gray bar represents peripherally expanded clones that are present in the post-treatment biopsy (after 6 series). e) Frequency of the dominant TCR β chain in clonal PD-L1 and IDO specific cultures as determined by CDR3 sequencing. Extended Data Fig. 9 Profiling of genes relevant to T cell activation, cytokines and exhaustion markers on pre- and post-treatment biopsies in two patients. a) RNA expression profiling of genes related to T cell activation was performed using NanoString nCounter (n=2). b) RNA expression profiling of genes related to cytokine activity was performed using NanoString nCounter (n=2). c) RNA expression profiling of genes related to checkpoint inhibitors was performed using NanoString nCounter (n=2). Extended Data Fig. 10 Treatment induced upregulation of PD-L1, IDO, MHC I and MHC II and distance between CD8 T cells and PD-L1 expressing cells. a) IHC on 4 paired biopsies stained for CD3 and CD8 T cells, PD-L1, MHC I and MHC II on tumor cells/mm2 (sum in the validated area) and IDO H-score (from 0 to 300). H-score is the expression of IDO on both immune and tumor cells: The score is obtained by the formula: 3 x percentage of cells with a strong staining + 2 x percentage of cells with a moderate staining + percentage of cells with a weak staining (n=4). b) Distance in µm between CD8+ T cells and PD-L1+ stained cells on five baseline biopsies detected by IHC (n=5). Extended data is available for this paper at 10.1038/s41591-021-01544-x. Supplementary information The online version contains supplementary material available at 10.1038/s41591-021-01544-x. Acknowledgements We thank all patients and their relatives for being a part of the trial. We thank Ö. Met, M. Jonassen, S. Ullitz Færch, B. Saxild, S. Wendt and C. Grønhøj for technical support. We thank M. Cumberbatch for input with translational analysis. We thank the nurses at clinic 5 and the head of the Oncology Department at Herlev and Gentofte Hospital, L. Sengeløv. The study was funded through a research funding agreement between IO Biotech and the CCIT-DK, Herlev Hospital and the Oncology Department at Herlev Hospital. Author contributions I.M.S. and M.H.A. conceived the study. J.W.K. wrote the protocol with input from I.M.S., M.H.A. and M.-B.Z. J.W.K. and C.L.L. were responsible for patient treatment and patient care with help from R.B.H. and consulting from I.M.S., E. Ellebaek and M.D. I.M.S., E. Ellebaek and M.D. were responsible for patient recruitment. J.W.K. and C.L.L. were responsible for coordinating trial procedures and collecting data and samples. J.W.K. and C.L.L. were clinical investigators, and I.M.S. was a sponsor. J.W.K. and C.L.L. analyzed and interpreted clinical data with support from I.M.S. and E. Ehrnrooth. T.W.K. performed statistical analyses. H.W.H. helped with clinical imaging. J.W.K., E.M., C.O.M., S.M.A., S.E.W.-B., M.O.H. and A.W.P. analyzed and interpreted translational data with support from M.H.A. and I.M.S. Funding was acquired by I.M.S., M.H.A. and M.-B.Z. J.W.K. wrote the manuscript with input from I.M.S., M.H.A., E.M. and A.W.P. All authors reviewed the manuscript, interpreted data and approved the final version. Data availability This clinical trial was registered at https://www.clinicaltrials.gov/ before patient enrollment (clinical trial identifier NCT03047928). Each vaccine was composed of 100 µg IO102, a 21-amino-acid peptide (DTLLKALLEIASCLEKALQVF) from the IDO peptide, and 100 µg IO103, a 19-amino-acid peptide (FMTYWHLLNAFTVTVPKDL) from the signal peptide of PD-L1. TCR sequencing data are available from Adaptive Biotechnologies. Upon request, the CCIT-DK office will provide a username and a password to access the designated data within approximately 2–4 weeks (https://www.herlevhospital.dk/ccit-denmark/find-us/Sider/Contact-information.aspx). All requests for the remaining data including raw data and analyzed data and materials will, within a reasonable time frame, be reviewed by the CCIT-DK office (https://www.herlevhospital.dk/ccit-denmark/find-us/Sider/Contact-information.aspx) to verify whether the request is subject to any intellectual property or obligations. Patient-related data not included in the paper were generated as part of clinical trials and may be subject to patient confidentiality. Any data and materials that can be shared will be released via a material-transfer agreement. The following database was used in the study: https://research.regionh.dk/da/publications/the-danish-metastatic-melanoma-database-dammed(32749d99-095f-4cae-b5de-769bae27f01e).html. Competing interests M.H.A. is named as an inventor on various patent applications relating to therapeutic uses of IDO and PD-L1 peptides. These patent applications are assigned to the company IO Biotech, which is developing immune-modulating cancer treatments. M.H.A. is a founder, shareholder and advisor for IO Biotech. I.M.S. is a cofounder, shareholder and advisor for IO Biotech. I.M.S. has an advisory board relationship with or lectured for Roche, Novartis, MSD, Celgene, Incyte, TILT Bio, Pfizer and BMS AstraZeneca and has received limited grants for translational research from BMS, Roche and Novartis. M.-B.Z. is the CEO, founder and shareholder at IO Biotech. E.M., A.W.P. and E. Ehrnrooth are employees at IO Biotech. E. Ellebaek has received honoraria from BMS, Pierre Fabre, Roche and Kyowa Kirin and travel support from MSD. M.D. has received honoraria from Genzyme, MSD, BMS, Roche and Novartis and travel support from Novartis, MSD, BMS, Roche and Pfizer. The remaining authors declare that they have no conflict of interest. Peer review information Nature Medicine thanks George Coukos, Michael Schell and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. Javier Carmona was the primary editor on this article and managed its editorial process and peer review in collaboration with the rest of the editorial team. Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. These authors contributed equally: Julie Westerlin Kjeldsen, Cathrine Lund Lorentzen. Change history 3/8/2022 A Correction to this paper has been published: 10.1038/s41591-022-01771-w

NIVOLUMAB

DrugsGivenReaction

CC BY

34887574

2021-12-09

Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Infusion related hypersensitivity reaction'.

A phase 1/2 trial of an immune-modulatory vaccine against IDO/PD-L1 in combination with nivolumab in metastatic melanoma. Anti-programmed death (PD)-1 (aPD1) therapy is an effective treatment for metastatic melanoma (MM); however, over 50% of patients progress due to resistance. We tested a first-in-class immune-modulatory vaccine (IO102/IO103) against indoleamine 2,3-dioxygenase (IDO) and PD ligand 1 (PD-L1), targeting immunosuppressive cells and tumor cells expressing IDO and/or PD-L1 (IDO/PD-L1), combined with nivolumab. Thirty aPD1 therapy-naive patients with MM were treated in a phase 1/2 study ( https://clinicaltrials.gov/ , NCT03047928). The primary endpoint was feasibility and safety; the systemic toxicity profile was comparable to that of nivolumab monotherapy. Secondary endpoints were efficacy and immunogenicity; an objective response rate (ORR) of 80% (confidence interval (CI), 62.7-90.5%) was reached, with 43% (CI, 27.4-60.8%) complete responses. After a median follow-up of 22.9 months, the median progression-free survival (PFS) was 26 months (CI, 15.4-69 months). Median overall survival (OS) was not reached. Vaccine-specific responses assessed in vitro were detected in the blood of >93% of patients during vaccination. Vaccine-reactive T cells comprised CD4+ and CD8+ T cells with activity against IDO- and PD-L1-expressing cancer and immune cells. T cell influx of peripherally expanded T cells into tumor sites was observed in responding patients, and general enrichment of IDO- and PD-L1-specific clones after treatment was documented. These clinical efficacy and favorable safety data support further validation in a larger randomized trial to confirm the clinical potential of this immunomodulating approach. pmcMain Despite remarkable advances in the treatment of MM with immune checkpoint inhibitors (ICIs) targeting PD-1 and cytotoxic T lymphocyte antigen 4 (CTLA-4), around half of patients are resistant to ICI monotherapy1. The combination of anti-CTLA-4 (aCTLA-4) and aPD1 therapy is to date the most effective therapy resulting in a response rate of around 60%; however, 50% of the patients also develop severe adverse events2,3. Therefore, an equally effective but less toxic treatment is highly needed. Several approaches to enhance ICI efficacy are currently being investigated, such as other ICIs, T cell therapy with tumor-infiltrating T cells or innate immunity stimulators such as Toll-like receptor 9 agonists4–6. Treating cancer patients with vaccines that stimulate a targeted immune response is another attractive approach, with very few side effects observed thus far in combination immunotherapy studies7,8. Immune-modulatory vaccines targeting tumoral immune escape mechanisms offer a new, generalizable strategy compared to patient-specific neoantigen cancer vaccines7,9. The immune-modulatory vaccine strategy in this clinical trial is based on the finding of circulating cytotoxic T cells specific to IDO and PD-L1 in the blood of patients with cancer and, to a lesser extent, in healthy donors. IDO- and PD-L1-specific CD8+ T cells can directly recognize and kill IDO+ and/or PD-L1+ tumor cells and likewise recognize and kill non-malignant cells that express their cognate targets. Furthermore, IDO- and PD-L1-specific CD4+ T cells release pro-inflammatory cytokines in response to IDO- or PD-L1-expressing target cells. IDO and PD-L1 are expressed not only by melanoma cells but also by many other cell types in the tumor microenvironment (TME), which differentiates these antigens from traditional tumor antigens used in other studies10–15. Activation of IDO/PD-L1-specific T cells by vaccination can therefore restrict the range of immunosuppressive signals mediated by immunosuppressive cells and thereby revert the TME from an immune hostile to an immune friendly environment. In animal models of cancer, vaccinations with IDO epitopes resulted in anti-tumor therapeutic effects that were correlated with reductions in IDO expression in myeloid cell populations within the TME16. The IDO/PD-L1 immune-modulating vaccine may lead to a translatable strategy for improving the efficacy of aPD1 therapy through activation of specific T cells. We hypothesize that the IDO/PD-L1 vaccine attracts T cells into the tumor, which induces type 1 helper T (TH1) cell inflammation and reverts the TME into an immune-permissive site, thereby turning the tumor ‘hot’. This would also upregulate PD-L1 expression in cancer and immune cells, generating more susceptible targets to aPD1 therapy (Extended Data Fig. 1a). This theory was confirmed in a mouse model in which aPD1 therapy and IDO vaccination show synergistic effects16. In this phase 1/2 clinical trial, MM1636, patients with MM received a combination of the IDO/PD-L1 (IO102/IO103) peptide vaccine with the adjuvant Montanide and the aPD1 antibody nivolumab. Patients were included in three cohorts: 30 aPD1 therapy-naive patients (cohort A), ten aPD1 therapy-refractory patients (cohort B, de novo resistance) and ten patients who progressed after aPD1 therapy (cohort C, acquired resistance). Here, we report results from cohort A. The vaccine was given biweekly for the first six administrations and thereafter every 4th week. A maximum of 15 vaccines were administered. Nivolumab was given in parallel, biweekly (3 mg per kg) or every 4th week (6 mg per kg) for up to 2 years (Extended Data Fig. 1b). The primary objective was safety and feasibility. Secondary objectives were immunogenicity and clinical efficacy. Results Patients and treatment Thirty patients were enrolled from December 2017 to June 2020. None of the 30 patients dropped out of the study; all received at least three cycles of therapy (Supplementary Fig. 1). At the current database lock (5 October 2020), six patients were still on treatment in the trial. Of the 24 patients who were not on trial treatment at data cutoff, two are still receiving nivolumab monotherapy (6 mg per kg every 4 weeks). Reasons for stopping treatment for the remaining 22 patients included disease progression (37%), toxicity (20%), maximum benefit or complete response (CR) confirmed on two consecutive scans (17%) or completing 2 years of treatment (7%). For the 24 patients who were not on trial treatment at data cutoff, the mean number of vaccinations was 10.5 (range, 3–15 vaccinations). Thirteen of these 24 patients continued nivolumab (6 mg per kg, every 4 weeks) as a standard of care. Nine patients received subsequent therapy after progression (Supplementary Table 1). Baseline characteristics are shown in Table 1. The mean age was 70 years; 37% of patients had elevated lactate dehydrogenase (LDH) levels, 60% were stage M1c, 37% had BRAF mutations, and 43% were negative for PD-L1 (<1%). A total of three patients (10%) had received prior ipilimumab therapy. No patients had brain metastasis (Supplementary Table 2).Table 1 Baseline patient characteristics (n = 30) Characteristic Number (%) Mean age, years (range) 70 (46–85) Sex, male 16 (55%) ECOG PS 0 26 (87%) LDH levels  ≤ULN 19 (63%)  >ULN 11 (37%) M stage (AJCC-8)  M1a 6 (20%)  M1b 6 (20%)  M1c 18 (60%) Number of lesion sites  1 6 (20%)  2–3 17 (57%)  >3 7 (23%) Liver metastases present 10 (33%) BRAF status  Mutant 11 (37%)  Wild type 19 (63%) PD-L1  <1% 13 (43%)  >1% 17 (57%) Previous systemic therapy  Ipilimumab 3 (10%)  No 27 (90%) AJCC-8, eighth edition of the American Joint Committee on Cancer; ECOG PS, Eastern Cooperative Oncology Group performance status; ULN, upper limit of normal. Notable clinical responses to the combination therapy Thirty patients with MM were treated with the IDO/PD-L1 vaccine and nivolumab according to the trial protocol. By investigator review, the ORR reached 80% (CI, 62.7–90.5%), with 43% of patients (CI, 27.4–60.8%) achieving a CR and 37% (CI, 20.9–54.5%) reaching a partial response (PR) as the best overall response, while 20% experienced progressive disease (PD) according to Response Evaluation Criteria in Solid Tumors (RECIST) 1.1 (Fig. 1a). Two of the patients with a PR did not have a confirmatory scan stating PR on two consecutive scans. Early onset of response was frequent, with 22 of 30 patients having an objective response at the first evaluation (after 12 weeks on treatment). Median times to PR and CR were 75 d (range, 54–256 d) and 327 d (range, 73–490 d), respectively (Fig. 2a–c).Fig. 1 Clinical response. a, Pie charts with percent ORR, CR, PR and PD according to RECIST 1.1 by investigator review of all patients (n = 30), PD-L1+ patients (>1%, (n = 17)) and PD-L1− patients (<1%, n = 13)), respectively. Two-sided CIs (95%) were constructed using the Clopper–Pearson method. b, Treatment effect in MM1636 compared with a matched historical control group from the DAMMED database (n = 74). Patients in MM1636 (n = 29) were matched with the exact same combination variable according to age (≤70 years, >70 years), sex, LDH levels (normal, elevated), M stage (M1a, M1b, M1c), BRAF status (wild type, mutated) and PD-L1 status (<1%, ≥1%). Estimates for treatment effects were calculated by weighted logistic regression analyses and weighted Cox proportional hazard model. Bar height indicates the estimated response rate; tops of bars are centers for error bars. Odds ratios (OR), response rates and their corresponding 95% CIs were extracted from the regression model. All P values were two sided, and P values below 0.05 were considered statistically significant. c, Best change in the sum of target lesion size compared with that at baseline (n = 30). The horizontal line at −30 shows the threshold for defining an objective response in the absence of non-target disease progression or new lesions according to RECIST 1.1. Two patients with 100% reduction in target lesion size had non-target lesions present. White stars, six patients had normalization (<10 mm) of fluor-18-deoxyglucose (FDG)-negative lymph nodes (at baseline, lymph nodes were >1.5 cm and FDG+) and 100% reduction of non-lymph node lesions and are considered to have had a CR (green bar). Black star, one patient (MM29) was considered to have had a PR (blue bar), although he had not reached a −30% change in target lesion size. The patient had a single measurable 13-mm lung metastasis at baseline and multiple biopsy-verified cutaneous metastases on the left crus (not detectable by positron-emission tomography–computed tomography (PET–CT) at baseline). The best change in target lesion size was 10 mm, and a post-treatment biopsy from the cutaneous metastases showed no sign of malignancy. Thus, overall, the patient was classified as having a PR. d, Kaplan–Meier curve of the response duration in the 24 patients with an objective response. e, Kaplan–Meier curve of PFS in all 30 treated patients. f, Kaplan–Meier curve of OS in all 30 treated patients. Fig. 2 Clinical response. a, Swimmer plot showing response duration and time to response according to RECIST 1.1 for all treated patients (n = 30). Triangles indicate first evidence of PR, while squares indicate first evidence of CR. Closed circles indicate time of progression. Arrows indicate ongoing responses. Patient MM18 died due to nivolumab-induced side effects. b, Spider plot showing response kinetics in all treated patients (n = 30). Red squares indicate time of progression. c, PET–CT images of patient MM42 before and after treatment (after 12 series of treatment) showing FDG metabolism in target lesions. SC, subcutaneous. The ORR among PD-L1+ (>1% (clone 28.8)) patients (n = 17) was 94.1% (CI, 73–99.7%) and 61.5% (CI, 35.5–82.3%) in PD-L1− patients (n = 13) (Fig. 1a). Objective responses were observed in patients irrespective of human leukocyte antigen (HLA) genotype (Supplementary Fig. 2). Clinical response data were validated by blinded independent external review, in which an ORR of 76.6% (CI, 57.7–90.1%) was reported, with 53.3% of patients achieving a CR, 23.3% achieving a PR and 3.3% experiencing stable disease. Comparisons between investigator review and external review are outlined in Supplementary Table 3. To examine whether the very high response rate should be attributed to nivolumab or to the vaccine, we retrieved clinical response information from a matched historical control group using the Danish Metastatic Melanoma Database (DAMMED) from contemporaneously treated patients with stage 3–4 melanoma who received aPD1 monotherapy17. Patients were matched with the exact same combination variable according to age, sex, PD-L1 status, BRAF status, LDH level and M stage (those at stage M1d were excluded from the control group (no patients with brain metastasis)). Matched controls were identified for 29 patients, and the ORR of 79.3% (CI, 61.0–90.4%) observed in MM1636 was found to be significantly higher (P < 0.0012) than that in the matched control group in which an ORR of 41.7% (CI, 31.0–53.3%) was reached. Furthermore, of the 29 patients in MM1636, a significantly (P < 0.0017) higher percentage (41.4% (CI, 25.2–59.6%)) of patients achieved CR in MM1636 than that (12% (CI, 6.3–21.6%)) in the matched historical control group. ORR and complete response rate (CRR) in the matched historical control group were comparable to those of patients treated in randomized phase 3 pivotal trials with aPD1 monotherapy18 (Fig. 1b). The treatment leads to prolonged PFS At data cutoff, the median duration of response had not been reached, with 87% of all responding patients being free from progression at 12 months (Fig. 1d). Patients were followed for up to 35 months with a median follow-up time of 22.9 months (CI, 14.9–26.2 months). OS and PFS were calculated from the first day of treatment to death or progression or to the date of the last follow-up (5 October 2020). The median PFS (mPFS) for all treated patients was 26 months (CI, 15.4–69 months) and was not reached for responding patients (Fig. 1e). The median OS was not reached at the data cutoff. OS at 12 months was 81.6% (CI, 61.6–92%) (Fig. 1f). One patient (MM18) with a CR died from nivolumab-related severe adverse events; the remaining patients died because of metastatic melanoma. For comparison, the mPFS was 8.3 months (CI, 5.5 months–NR (not reached)) in the matched historical control group (n = 74), while the median OS was 23.2 months (CI, 23.2 months–NR). (Extended Data Fig. 2a,b). The combination of the IDO/PD-L1 vaccine and nivolumab was safe Treatment-related adverse events are listed for all 30 patients in Supplementary Table 4. Common treatment-related grade 1–2 toxicities were fatigue (47%), rash (47%), arthralgia (30%), diarrhea (23%), nausea (23%), dry skin (20%), pruritus (20%), infusion reaction (17%), xerostomia (17%) and myalgia (17%). Four patients (13%) experienced grade 3–4 adverse events: one patient with a grade 3 maculopapular rash (MM01), one patient with grade 3 adrenal insufficiency (MM06) and one patient with grade 3 arthralgia (MM22). Patient MM18 died from urosepsis with multi-organ failure and severe hyponatremia. This patient had experienced multiple immune-related adverse events with grade 3 colitis, grade 2 pneumonitis, grade 3 arthralgia, grade 2 vasculitis and grade 2 nivolumab infusion-related allergic reaction. Additionally, patient MM18 had symptoms of myocarditis at the time of death with highly elevated cardiac troponin I levels. Bedside echocardiography showed an ejection fraction of 15%, which at baseline was 60%, but an autopsy was not conducted, and myocarditis was never confirmed pathologically. Patient MM06 had received first-line treatment with ipilimumab before entering the trial and was on substitution corticosteroids at the time of inclusion. Adrenal insufficiency was aggravated by an erysipelas infection with high fever, reaching grade 3 in Common Terminology Criteria for Adverse Events and resolving quickly after appropriate antibiotic therapy was initiated. As expected, local side effects were common with 77% of the patients who developed injection site reactions. These reactions were classified as granulomas (63%), redness (20%), pain (13%) and pruritus (13%) at the injection site. All local reactions were grade 1–2, most likely related to the Montanide adjuvant and typically transient. However, two patients (MM07 and MM20) decided to discontinue vaccination after eight and 11 injections, respectively, due to granulomas, tenderness and pain that limited instrumental activities of daily living but continued to receive nivolumab (Extended Data Fig. 2c). Vaccine-specific responses in blood were frequently detected First, all 30 patients were assessed for the presence of vaccine-specific responses in peripheral blood mononuclear cells (PBMCs) before, on and after vaccination using a modified interferon (IFN)-γ enzyme-linked immune absorbent spot (ELISPOT) assay. This assay is known to expand antigen-specific memory cells and to improve the detection and the correlative power of ELISPOT during treatment19–21. Prevaccine IDO-specific responses were detectable in ten (33%) patients, while prevaccine PD-L1-specific responses were detectable in eight patients (27%); overlapping (specific to both IDO and PD-L1) prevaccine responses were present in four (13.3%) patients. During vaccination, an increase of IDO-specific T cells or PD-L1-specific T cells in the blood was observed in 28 (93%) and 26 (86%) patients, respectively. In total, 93% of patients had an increase in either PD-L1- or IDO-specific responses on vaccination (Fig. 3a), with a significant (P < 0.0001) median increase from baseline to the post-vaccine response demonstrated for both IDO and PD-L1 (at different time points on treatment), confirming that vaccine-specific immune responses were induced in patients regardless of clinical response (Fig. 3b,c). Immune responses fluctuated in the blood over time (Extended Data Fig. 3a). An increase in IDO- and PD-L1-specific responses in the peripheral blood was also detectable directly ex vivo across the different clinical response groups (Extended Data Fig. 4).Fig. 3 Vaccine-specific responses in blood. a, IDO- and PD-L1-specific T cell responses in PBMCs at baseline and on vaccination as measured by the IFN-γ ELISPOT assay (n = 30). *Responses were calculated as the difference between the average numbers of spots in wells stimulated with IDO or PD-L1 peptide (triplicates) and those from the corresponding control (DMSO), and statistical analyses of ELISPOT responses were performed using a distribution-free resampling method (Moodie et al.50). DR (double response), response was not statistically confirmed due to replicate number, but the number of spots in peptide wells was two times higher than that in control wells (DMSO). NS, no significant response and no DR. For a detailed overview of responses at serial time points on vaccination, see Extended Data Fig. 4a. b, IDO- and PD-L1-specific T cell response in PBMCs in all treated patients measured by the IFN-γ ELISPOT assay at baseline and on vaccination. On-vaccination responses were selected from the ‘best’ ELISPOT response at different time points for each patient (series 3, 6, 12, 18 or 24) during vaccination (n = 30). Wilcoxon matched-pairs signed-rank test was used to compare responses to IDO or PD-L1 peptides in the vaccine between baseline and later time points. c, Representative example of ELISPOT wells with response in patient MM23 in serial PBMCs before and on treatment. d, IDO-specific CD4+ and CD8+ T cells were isolated and expanded from PBMCs stimulated in vitro with the IDO peptide and a low dose of IL-2 for 14–15 d before sorting using the Miltenyi Cytokine Secretion Assay—Cell Enrichment and Detection kit. To assess their cytolytic potential, IDO-specific T cells were stimulated with IDO peptide, and expression of CD107a, IFN-γ and TNF-α was assessed by flow cytometry (the example is from patient MM14). Sustained vaccine-specific responses were observed 3 and 6 months after the last vaccine, indicating induction of memory responses in nine patients with response to clinical treatment who surpassed follow-up at data lock (Extended Data Fig. 3b). Importantly, PD-L1- and IDO-specific responses were observed irrespective of HLA genotype (Supplementary Table 5). To verify the functionality of vaccine-induced T cells, IDO- or PD-L1-specific T cells were isolated and in vitro expanded from PBMCs of five patients. Phenotypic characterization by flow cytometry revealed that the isolated vaccine-specific T cells consisted of both CD4+ and CD8+ T cells. Also, both IDO- and PD-L1-specific CD4+ and CD8+ T cells showed pro-inflammatory properties, as they expressed the cytolytic marker CD107a and secreted the cytokines IFN-γ and tumor necrosis factor (TNF)-α (Fig. 3d and Extended Data Fig. 6a–d). Interestingly, we were also able to detect vaccine-specific CD4+ and CD8+ T cell responses in peripheral blood ex vivo (Extended Data Figs. 4 and 5). We observed a significant increase in the overall percentage of CD107a, CD137 and TNF-α expression in response to peptide stimulation in on- or post-treatment PBMC samples compared to that at baseline, further confirming the expansion and the diverse signature of vaccine-specific T cells (Extended Data Fig. 5). Vaccine-specific responses detected in the skin at the vaccination site To investigate whether vaccine-specific T cells have the potential to migrate to peripheral tissue, delayed-type hypersensitivity (DTH) tests were performed after six cycles of treatment on 15 patients to assess the presence of vaccine-reactive T cells in the skin. Supplementary Table 6 display an overview of skin-infiltrating lymphocyte (SKIL) cultures. We detected IDO-specific T cells in the skin of six of ten patients and PD-L1-specific T cells in nine of 11 patients (Extended Data Fig. 7a). Intracellular cytokine staining was performed on SKILs from five patients after stimulation with either PD-L1 or IDO peptide. Here, we detected mainly CD4+ peptide-reactive T cells that secreted TNF-α and upregulated CD107a, and a minor fraction also secreted IFN-γ. In one patient, we detected CD8+ PD-L1-reactive T cells (Extended Data Fig. 7b–d). Vaccine-induced T cells specifically recognize target cells To confirm the functionality of vaccine-expanded T cells, vaccine-specific T cell clones (clonal purity was confirmed by T cell receptor (TCR) sequencing) were isolated and expanded from patient PBMCs (Extended Data Fig. 8e). We showed that PD-L1-specific T cells were able to recognize PD-L1+ autologous tumor cells in a PD-L1 expression-dependent manner if the cancer cells also expressed HLA-II (Fig. 4a,b). Similarly, an HLA-DR-restricted IDO-specific CD4+ T cell clone was able to recognize an HLA-DR-matched IDO-expressing model cell line, MonoMac1, in an IDO-expression-dependent manner (Fig. 4e–g). As previously described, IDO- and PD-L1-specific T cells’ mode of action is not limited to targeting only cancer cells. We were able to show that vaccine-specific T cell clones also reacted against PD-L1- and IDO-expressing autologous immune cells (Fig. 4c,h). To provide myeloid cells with a tumor-associated phenotype we treated isolated CD14+ myeloid cells with tumor conditioned medium (TCM) derived from an established autologous tumor cell line. We observed that such TCM-treated CD14+ cells had increased expression of PD-L1 and IDO and were effectively recognized by autologous PD-L1- and IDO-specific CD4+ T cell clones (Fig. 4c,d,h,i).Fig. 4 PD-L1- and IDO-specific T cells from vaccinated patients react against PD-L1- and IDO-expressing target cells. a, Left, PD-L1-specific T cell culture (MM1636.05) reactivity against PD-L1 peptide or autologous tumor cells in the IFN-γ ELISPOT assay. Tumor cells were either not treated or pretreated with 200 U ml−1 IFN-γ for 48 h before the assay. Effector:target (E:T) ratio of 10:1 was used. Right, PD-L1 and HLA-II surface expression on melanoma cells with (green) or without (yellow) pretreatment with IFN-γ compared to an isotype control (gray) as assessed by flow cytometry. b, Left, PD-L1-specific T cell (MM1636.05) reactivity in the IFN-γ ELISPOT assay against autologous tumor cells pretreated with IFN-γ (500 U ml−1) and transfected with mock or PD-L1 small interfering (si)RNA 24 h after transfection. E:T ratio, 10:1. Right, PD-L1 surface expression on melanoma tumor cells (MM1636.05) assessed by flow cytometry 24 h after transfection with mock (blue) or PD-L1 (red) siRNA compared to the isotype control (gray). c, Reactivity of the CD4+ PD-L1-specific T cell clone (MM1636.14) against PD-L1 peptide or autologous CD14+ cells; E:T ratio, 10:1. CD14+ cells were isolated using magnetic bead sorting and used as targets in an ELISPOT assay directly or after pretreatment for 2 d with TCM derived from the autologous tumor cell line. d, Quantitative PCR with reverse transcription (RT–qPCR) analysis of PD-L1 (CD274) expression in sorted CD14+ cells before and after treatment with autologous TCM for 48 h. e, Reactivity of the IDO-specific CD4+ T cell clone (MM1636.23) against IDO peptide combined with HLA-DR (L243)-, HLA-DQ (SPV-L3)- or HLA-DP (B7/21)-blocking antibodies (aHLA-DR, aHLA-DQ and aHLA-DP) in an intracellular staining assay (ICS) for IFN-γ and TNF-α production. T cells were incubated with individual blocking antibodies (2 μg ml−1) for 30 min before adding IDO peptide. f, Reactivity of the IDO-specific CD4+ T cell clone (MM1636.23) against the HLA-DR-matched IDO-expressing cell line MonoMac1 transfected with mock or IDO siRNA in an ICS assay, E.T ratio, 4:1. siRNA transfection was performed 48 h before the experiment. g, RT–qPCR analysis of IDO1 expression in MonoMac1 cells 48 h after siRNA transfection. h, Reactivity of the CD4+ IDO-specific T cell clone (MM1636.14) against IDO peptide or autologous CD14+ cells; E:T ratio, 20:1. CD14+ cells were isolated using magnetic bead sorting and used as targets in an ELISPOT assay directly or after pretreatment with TCM derived from the autologous tumor cell line. i, RT–qPCR analysis of IDO1 expression in sorted CD14+ cells before and after treatment with autologous TCM for 48 h. Bars in RT–qPCR data (d,g,i) represent the mean of three (d,i) or six (g) technical replicates ±s.d.; P values were determined by two-tailed parametric t-tests. ELISPOT counts (a,b,c,h) represent the mean value of three technical replicates ±s.e.m.; response P values were determined using the distribution-free resampling (DFR) method. TNTC, too numerous to count. T cell clones in blood and tumors To track the role of treatment-induced T cell responses, TCR sequencing of the complementarity-determining region 3 (CDR3) was performed on five patients in peripheral blood (baseline and cycles 3, 6 and 12) and paired biopsies. These five patients (MM01, MM02, MM08, MM09 and MM13) were selected due to the availability of material and to investigate a balanced patient group with both responders and non-responders. Due to a limited number of available paired biopsies (either as a consequence of patient refusal or unexpectedly rapid and substantial clinical responses), statistical analysis could not be applied. Details on clinical response are shown in Fig. 2a. Additionally, PBMCs (on treatment) or SKILs were stimulated with the IDO/PD-L1 peptides, and then cytokine-producing T cells were sorted to track vaccine-induced T cells both in the periphery and at the tumor site. To identify enriched IDO/PD-L1-specific T cell clones, TCR rearrangements in sorted IDO/PD-L1-specific clones and TCR rearrangements in sorted IDO/PD-L1-specific T cell samples were compared to sequences from baseline PBMC samples for each patient. Clonal expansion of vaccine-specific TCR rearrangements from samples on vaccination were then tracked using a differential abundance framework. Cumulative IDO/PD-L1-specific T cell frequencies were tracked in post-treatment samples. We found no relation between clinical response and the enrichment of vaccine-specific clones, but an increase in IDO/PD-L1-specific T cell clones was observed at different time points in the periphery in all five patients (Extended Data Fig. 3c). We next investigated overall changes in the T cell repertoire in the blood. A modest increase in the peripheral T cell fraction was observed in the three responding patients at cycle 3, while the two non-responding patients had a clear decrease in T cell fraction (Extended Data Fig. 8a). We thereafter investigated TCR clonality and TCR repertoire richness, exploring the proportion of abundant clones and the number of unique rearrangements, respectively. A decreasing peripheral Simpson clonality and increasing TCR repertoire richness was observed in responding patients at cycle 3, which might indicate tumor trafficking upon treatment. The opposite pattern was observed in non-responding patients (Extended Data Fig. 8b,c). Peripherally expanded clones were associated with tumors and persisted until cycle 12 (latest time point analyzed). The largest peripheral expansion was observed at cycle 3, with the most significant increase observed in patient MM01 (CR). Responding patients had a larger fraction of peripherally expanded clones that were also found in tumors compared to that of non-responders. By tracking peripherally expanded clones detected at the tumor site, we observed that patient MM01 had a substantial increase after treatment, indicating tumor trafficking of peripheral expanded clones (Extended Data Fig. 8d). Influx at the tumor site of vaccine-enriched T cell clones Given the observation of increased T cell fraction and enrichment of IDO- and PD-L1-specific clones in the blood after treatment, we investigated whether the same trend was observed at the tumor site. Both TCR sequencing and immunohistochemistry (IHC) of paired biopsies from the five patients described above showed an increase in the T cell fraction with an influx of CD3+ and CD8+ T cells after treatment in the three responding patients (Fig. 5a–c). IHC was not possible for patient MM09 due to tissue loss.Fig. 5 Changes in the TME after treatment. Number of CD3+ and CD8+ T cells, TCR fraction, TCR clonality, TCR repertoire, biopsy expanded TCR clones and enrichment of IDO- and PD-L1-specific T cells at the tumor site. a, Number of CD3+ and CD8+ T cells per mm2 (sum in the validated area) at the tumor site detected by IHC of paired biopsies from four patients. b, Example of IHC of CD3+ and CD8+ T cells at the tumor site before and after treatment (cells per mm2) in one patient (MM01). c, T cell fraction at the tumor site at baseline and cycle 6 by TCR sequencing. The T cell fraction was calculated by taking the total number of T cell templates and dividing by the total number of nucleated cells. d, Tracking of vaccine-associated clones at baseline and cycle 6 in tumor biopsies. Cumulative frequencies of IDO and PD-L1 vaccine-specific TCR rearrangements are represented. e,f, TCR clonality and TCR repertoire richness in five patients at the tumor site at baseline and cycle 6. Simpson clonality measures how evenly TCR sequences are distributed among a set of T cells, where 0 indicates an even distribution of frequencies and 1 indicates an asymmetric distribution in which a few clones dominate. TCR repertoire richness reports the mean number of unique rearrangements. g, Bar chart representing baseline expanded biopsy clones from five patients (colored bars) and the detection of biopsy expanded clones also found in the blood at baseline and series 3, 6 and 12 (white and gray bars) by TCR sequencing. We thereafter investigated whether some of the IDO/PD-L1 vaccine-linked T cells were present at the tumor site. Vaccine-associated clones were tracked as the combined frequency of IDO- and PD-L1-specific T cell rearrangements. In biopsies, the frequencies at cycle 6 were compared to those at baseline and showed an increase in vaccine-specific T cells in four of five patients, irrespective of clinical response (Fig. 5d). TCR sequencing of PD-L1-specific SKILs (a more specific culture than IDO/PD-L1-specific isolations derived from PBMCs on vaccination) and paired biopsies showed that two of the top five PD-L1-specific SKIL clones were present at the tumor site both before and after treatment (Extended Data Fig. 7e,f). With a focus on the more abundant T cell clones, we investigated overall TCR clonality at the tumor site before and after treatment. In addition, we explored the number of unique TCR rearrangements, dissecting the lower-frequency clones. Patient MM01 had a significant increase in TCR clonality and a decrease in repertoire richness at the tumor site after therapy, indicating a focused tumor repertoire response of selected clones. All three responding patients had a decrease in TCR repertoire richness, which again might indicate a focused tumor response (Fig. 5e,f). Deeper analyses showed that the T cell clones that expanded at the tumor site after therapy were also present in the blood at baseline and increased significantly after treatment in four of five patients. The highest proportion was detected early at cycle 3. These data again support trafficking of peripherally expanded clones to the tumor site and could indicate that the T cell response to treatment is derived from pre-existing peripherally tumor-associated T cells (Fig. 5g). Signs of treatment-induced inflammation in the TME To dissect changes in the TME induced by T cell influx upon treatment in responding patients, RNA gene expression analyses using the nCounter PanCancer Immune Profiling Panel from NanoString were performed on paired biopsies from two responding patients (MM01 and MM13). Expression of genes related to adaptive immunity such as T cell activation, effector functions (genes encoding IFN-γ, TNF-α, IL-15 and IL-18) and cytotoxicity was increased in post-treatment biopsies (Extended Data Fig. 9a,b). Also, expression of genes related to checkpoint inhibitors such as those encoding T cell immunoglobulin and mucin-domain containing-3 (TIM-3), IDO, PD-L1, PD-L2, PD-1 and CTLA-4 increased after treatment, indicating activation of immune cells in the TME (Extended Data Fig. 9c). Additionally, IHC of paired biopsies from four patients (MM01, MM02, MM05 and MM13) showed an upregulation of PD-L1, IDO, MHC-I and MHC-II on tumor cells, indicating a treatment-induced pro-inflammatory response in the three responding patients, except for a decrease in MHC-II expression in patient MM13. By contrast, the non-responding patient MM02 had a reduction in T cell numbers present in the tumor after treatment and no expression of PD-L1, IDO or MHC-II and, interestingly, total loss of MHC-I, demonstrating tumor immune escape (Extended Data Fig. 10a). CD8+ T cells and their distance (µm) to PD-L1-expressing cells in baseline biopsies from five patients was investigated by IHC. Except for patient MM13 (PR), distance and clinical responses were associated: the two responders had reduced distance (<20 µm) between cells expressing these markers compared to non-responding patients (>80 µm). This observation indicates that responding patients not only have a higher intratumoral infiltration of CD8+ T cells, but that these cells can surround and attack PD-L1-expressing immune cells and tumor cells (Extended Data Fig. 10b). Discussion In this clinical trial, MM1636, 30 patients with metastatic melanoma were treated with a first-in-class immunomodulatory IDO/PD-L1-targeting peptide vaccine combined with nivolumab. The treatment led to an unprecedented high ORR of 80%, with 43% of patients reaching a CR, and a striking mPFS of 26 months (95% CI, 15.4–69 months) was reached. The vaccine represents a new treatment strategy to activate specific T cells that target cells contributing to immune suppression (including tumor cells), positively modulating the TME by inducing local inflammation. Indeed, we show that vaccine-specific T cells isolated and expanded from vaccinated patients recognize not only tumor cells in a target- and HLA-restricted manner but also myeloid cells polarized toward a tumor-associated phenotype. Hence, myeloid cells become targets for vaccine-activated T cells when they have a tumor-associated phenotype in the TME. These phenomena may further induce checkpoint molecules and rewire the TME toward an increasingly aPD1-permissive state. A drawback of our study is the single-center nonrandomized setup. Comparison between trials or between patients in trials and real-world patients is problematic due to multiple factors, such as period of conduction and other therapies available at the different periods of time. Nevertheless, the rate of investigator-assessed ORR in the phase 3 trial CheckMate 067 was 43.7% in the nivolumab monotherapy group and 57% in the nivolumab and ipilimumab group. CRs occurred in 8.9% and 11.5% of patients, respectively18. The mPFS of 26 months (95% CI, 15.4–69 months) in this trial is more than twice as long as that for patients treated with nivolumab and ipilimumab in CheckMate 067, for which an mPFS of 11.5 months (95% CI, 8.7–19.3 months) was reached. Patient baseline characteristics were in general comparable to those of patients with MM who have been treated in CheckMate 067, although patients in MM1636 were older (mean age, 70 years) and a larger fraction were positive for PD-L1 (57%)3,18,22. Among patients with PD-L1-negative tumors in MM1636, an ORR of 61.5% was still reached, which would be expected to be around 33% for first-line nivolumab monotherapy23. Some studies suggest that older patients might have a tendency to respond better to aPD1 therapy; however, this is still debated24–27. To address potential trial bias and the nonrandomized setup, patients in MM1636 were matched for age, performance status, sex, M stage, LDH level, PD-L1 status and BRAF status with a historical control group from the DAMMED, who were treated contemporarily (2015–2019) with aPD1 monotherapy as standard of care17. We found a significantly higher ORR and CRR in MM1636 compared to those of matched patients, who had an ORR of 43% and a CRR of 13%, comparable to patients treated in CheckMate 067. Restrictions of the synthetic control group are of course that it is partially historic and patient selection outside matching criteria cannot be ruled out28. Numerous contemporary clinical trials are exploring the combination of aPD1 therapy with other immunomodulating agents for advanced melanoma. Talimogene laherparepvec, an oncolytic virus, is approved by the Food and Drug Administration and the European Medicines Agency to treat advanced melanoma. A small phase 1b trial with 21 patients (MASTERKEY-265) combined talimogene laherparepvec and pembrolizumab to treat patients with advanced unresectable melanoma and reached an ORR of 62% and a CR of 33%29,30. Seventy-one percent of the patients in this trial had an M stage below M1c; this number was 40% in our trial. Furthermore, mainly patients with an M stage below M1c responded to treatment, which was not the case in MM1636. Results from a large randomized phase 3 trial are awaited (KEYNOTE-034). Epacadostat, an IDO inhibitor, was tested in combination with pembrolizumab in a nonrandomized phase 2 trial in 40 aPD1 treatment-naive patients with MM with promising results, reaching an ORR of 62%. Unfortunately, the phase 3 trial showed no indication that epacadostat provided improvement in PFS and OS31. Limitations of the phase 3 trial were the sparse information on pharmacodynamics as well as biomarker evaluation to improve the design. The IDO/PD-L1 vaccine is different from epacadostat, as it is not an IDO inhibitor but targets IDO- and PD-L1-expressing cells. Similar vaccines administered as monotherapy induced objective responses in lung cancer and basal cell carcinoma, while epacadostat as monotherapy in 52 patients resulted in no responses32,33. Sahin et al. recently published encouraging data from a first-in-human trial, in which a vaccine containing liposomal RNA targeting four unmutated tumor-associated antigens (NY-ESO-1, MAGE-A3, tyrosinase and TPTE) was administered alone or in combination with nivolumab in patients with advanced melanoma. Responses were observed in both the monotherapy group as well as the combination group in checkpoint inhibitor-experienced patients, suggesting the efficacy of non-mutant shared tumor antigen vaccines34. The overall safety and tolerability findings are comparable to those of aPD1 monotherapy. Injection site reactions were exclusive to the vaccine. However, these side effects were transient and mild in most patients and most likely due to the adjuvant Montanide. IDO- and PD-L1-specific CD8+ and CD4+ T cells exist among peripheral blood lymphocytes in healthy donors35–38 and expand in response to pro-inflammatory stimuli35,39. Furthermore, both IDO and PD-L1 are induced in cells as a counter response to the inflammatory response. This provides a mechanism that ensures immune homeostasis, which keeps IDO/PD-L1-specific T cells in check; therefore, the risk of triggering autoimmune-related adverse events by vaccination appears to be minimal. This was confirmed in the first clinical trials of IDO and PD-L1 vaccination (NCT01543464 (refs. 33,40) and NCT03042793 (ref. 41)). The spontaneous (baseline) immune response to the vaccines observed in the current study is in agreement with our previous observation in various patients with cancer35–38. Numerous vaccine-induced changes in the blood and at the tumor site were observed. Peripheral IDO- and/or PD-L1-specific T cells were detected in vitro in a modified ELISPOT in over 93% of patients on vaccination, unrelated to patient HLA type. Immune responses were persistent in patients who surpassed follow-up at data cutoff and were still detectable up to 6 months after the last vaccine, suggesting induction of memory T cells. We did not observe a correlation between vaccine-induced responses in blood and clinical responses. However, the detection of a highly significant increase in vaccine-specific T cell numbers after vaccination in almost all patients together with the very low number of patients with PD makes it difficult to expect such a correlation, especially because other aspects should be taken into account (for example, the loss of class I expression on tumors cells in patient MM02 with PD after vaccination). Frequencies of peripheral T cells induced against PD-L1 were overall higher than those against IDO. Importantly, however, we also observed that it was not the same patients who responded strongly to both IDO and PD-L1, nor did patients react with a similar response pattern to the two antigens; that is, different time points of response were observed. This suggests that each component of the vaccine plays different roles in the ongoing immune response in patients. The goal of IDO/PD-L1 vaccination was to modulate the TME to increase responsiveness to aPD1 therapy, as we have observed in animal models, in which immune conversion is demonstrated in the TME with an increased influx of T cells. Targeting both IDO and PD-L1 together enables synergy, as the TME is known to use different immune escape mechanisms and IDO and PD-L1 are often overexpressed by different cellular compartments. Ex vivo ELISPOT and flow cytometry assays confirmed the induction of immune response toward both epitopes and the higher immunogenicity of the PD-L1 epitope, although because of lower sensitivity failed to detect some of the responses. TCR sequencing of five patients confirmed enrichment of IDO/PD-L1-specific T cell clones in the blood at different time points after treatment. Furthermore, an increase in enriched IDO/PD-L1-specific clones was observed in four of five patients at the tumor site after treatment, irrespective of clinical response. Phenotypic characterization showed that vaccine-specific T cells, which were expanded in vitro with interleukin (IL)-2 from the blood of vaccinated individuals, were both CD4+ T cells and CD8+ T cells. This was confirmed by ex vivo phenotype description of vaccine-activated T cells. Vaccine-specific T cells expressed CD107a and CD137 and produced IFN-γ and TNF-α upon stimulation with the cognate target, indicating their cytolytic capacity. Overall, the data from immune monitoring supports the importance of both antigens in the generation of the frequent clinical responses in the study. Despite a limited number of paired biopsies (due to either patient refusal or the fact that a large fraction of responding patients had no assessable tumors after six cycles), we observed trends indicating treatment-induced general T cell influx in responding patients. It was shown that proliferation of CD8+ T cells in the tumor after aPD1 treatment is associated with radiographic reduction in tumor size42. Additionally, we showed that a large proportion of expanded peripheral TCR clones were associated with tumors and the most considerable amount of clonal expansion was observed early, at cycle 3. For the patient with a CR included in the TCR sequencing analyses (MM01), the number of peripheral expanded clones present at the tumor site increased after treatment compared to that at baseline, indicating tumor trafficking of peripheral expanded clones. Gene expression analyses (two paired biopsies) and IHC (five paired biopsies) further demonstrated that the combination treatment induced a pro-inflammatory TME in responding patients with signs of T cell activation and cytotoxicity and increased cytokine activity. This may lead to further upregulation of IDO, PD-L1, MHC-I and MHC-II on tumor cells, leading to more treatment targets. It was shown that, following vaccination with a cancer vaccine, PD-L1 expression is increased on tumor cells due to recruitment of tumor-specific T cells and upregulation of adaptive immune resistance pathways in the TME43. Treatment with nivolumab monotherapy enhances PD-L1 expression, and it is therefore problematic to discriminate the effect of the vaccine as compared to that of nivolumab44. IDO- and PD-L1-specific pro-inflammatory effector T cells were hypothesized to counteract the functions of IDO- and PD-L1-expressing immune-suppressive cells as a means to keep the immune balance between immune activation and inhibition45. However, because the expression of these molecules is induced as a counter-regulatory response to inflammatory mediators such as IFNs, they can also be expressed by, for example, activated myeloid cells or T cells. Thus, activation of PD-L1-specific T cells may result in depletion of activated T cells in the tumor and other sites. Importantly, it was recently described that PD-L1+ T cells mainly have tolerogenic effects on tumor immunity and exert tumor-promoting properties, suggesting that targeting this immune population is indeed also beneficial46. We have previously investigated the effect of activating PD-L1-specific T cells in vitro and in vivo and found that, overall, they supported the effector phase of an immune response by removing PD-L1-expressing regulatory immune cells that inhibit PD-1+ effector T cells47,48. The major role of the PD-1 pathway is believed to be the regulation of effector T cell responses. Thus, this protective pathway is more important after activation, rather than at the initial T cell-activation stage49. Accordingly, the presence of PD-L1-specific T cells during the activation phase of an immune response may not increase or support a pro-inflammatory immune response due to the expression of PD-L1 on potent antigen-presenting cells or on the T cells themselves14. Thus, the overall effects of PD-L1-specific T cells may vary depending on the expression of both PD-1 and PD-L1, that is, due to the microenvironment and the state of the immune response. In conclusion, here we report an impressive response rate, CR rate and mPFS for a first-in-class immune-modulating vaccine combined with nivolumab. This may be a first step toward a new treatment strategy for patients with MM. Limitations are the low number of patients treated at a single institution and the lack of a randomized design with aPD1 monotherapy as comparator. Studies in aPD1 therapy-resistant or -refractory melanoma are ongoing as well as biomarker analysis for selecting patients at a higher likelihood to benefit from the combination versus aPD1 monotherapy. A larger randomized trial will be essential to validate these findings and determine the specific contribution of the vaccine to clinical responses and changes in the TME. In December 2020, the Food and Drug Administration granted breakthrough therapy designation for the IO102/IO103 vaccine combined with aPD1 therapy in metastatic melanoma based on data from the MM1636 trial. Methods Trial design and treatment plan MM1636 is an investigator-initiated, nonrandomized, open-label, single-center phase 1/2 study. All patients were treated at the Department of Oncology, Herlev and Gentofte Hospital, University of Copenhagen, Herlev, Denmark. The study was conducted according to the Declaration of Helsinki and Good Clinical Practice (GCP) and monitored by the GCP unit in Copenhagen, Denmark. The protocol was approved by the ethical committee of the Capital region of Denmark (H-17000988), the Danish Medical Agencies (2017011073) and the Capital Region of Denmark Data Unit (P-2019-172). The study was registered at https://clinicaltrials.gov/ under the identifier NCT03047928 and at the EudraCT (no. 2016-0004527-23). This study initially aimed to include 30 aPD1 treatment-naive patients with MM. An amendment with the addition of two other cohorts with ten patients in each cohort was made to evaluate immune responses and clinical efficacy in aPD1 therapy-resistant patients (cohort B, de novo resistance and cohort C, acquired resistance) for a total of 50 patients. The amendment with cohorts B and C was approved at the inclusion of 18 patients in cohort A. The trial is still including patients in cohorts B and C. This article reports results from cohort A. The first six patients were treated as for phase 1, evaluating for safety and tolerability before the remaining 24 patients were included in phase 2. The IDO/PD-L1 vaccine was administered subcutaneously biweekly for the first 10 weeks and thereafter every 4 weeks for approximately 9 months. A maximum of 15 vaccines were administered. Nivolumab was administered according to the approved label (3 mg per kg biweekly) for 24 cycles. The 15th vaccine was administered together with the 24th nivolumab cycle of 3 mg per kg, and responding patients thereafter continued nivolumab monotherapy every 4th week (6 mg per kg) as a standard of care after investigator assessment. Treatment with nivolumab was discontinued at the maximum benefit (assessed by the investigator), after a maximum of 2 years of therapy, at progression or due to severe adverse events (Extended Data Fig. 1b, treatment plan). Vaccine composition Each vaccine was composed of 100 µg IO102, a 21-amino-acid peptide (DTLLKALLEIASCLEKALQVF) from IDO, and 100 µg IO103, a 19-amino-acid peptide (FMTYWHLLNAFTVTVPKDL) from the signal peptide of PD-L1 (PolyPeptide Laboratories). Peptides were dissolved separately in 50 µl DMSO, filtered for sterility and frozen at −20 °C (NUNC CyroTubes CryoLine System Internal Thread, Sigma-Aldrich). At <24 h before administration, peptides were thawed. The PD-L1 peptide was diluted in 400 µl sterile water and, immediately before injection, mixed with the IDO peptide solution and 500 µl Montanide ISA-51 (SEPPIC) to achieve a total volume of 1 ml. Patients Patients above 18 years of age with locally advanced or stage 4 melanoma according to the AJCC (seventh edition), at least one measurable lesion according to RECIST 1.1 and an ECOG PS of 0–1 were eligible. The main exclusion criteria were prior treatment with aPD1 therapy, CNS metastases >1 cm, severe comorbidities and active autoimmune disease. Enrollment was not restricted to PD-L1 status, but it was known before inclusion. Patients were included after informed consent. Patients MM42 and MM20 have confirmed their approval of PET–CT images and clinical images that have been included in the article. Key study assessments Safety and tolerability were evaluated based on changes in clinical laboratory analyses and reported adverse events. Adverse events were assessed according to Common Terminology Criteria for Adverse Events (version 5.0) and were graded from 1 to 5 for all treated patients up to 6 months after the last dose of the IDO/PD-L1 vaccine. Clinical efficacy was assessed using FDG PET–CT scans before treatment and every 3rd month until progression. Objective responses were categorized into CR, PR, stable disease or PD according to RECIST version 1.1. Clinical data were collected at CCIT-DK in the eCRF program OpenClinica version 1.0 and in Microsoft Excel version 2002 on a secure server. Blood samples for immunologic analyses were collected before treatment, before the third cycle, after the sixth, 12th, 18th and 24th cycles (on vaccination) and 3 and 6 months after the last vaccine. Two to three tumor needle biopsies (1.2 mm) were collected at baseline and after six cycles from the same tumor site, when accessible, to evaluate immune responses at the tumor site. DTH skin tests were performed and punch biopsies were taken from the DTH area after cycle 6 for the evaluation of SKILs reactive to PD-L1 and IDO (Supplementary Fig. 1, treatment plan). Statistical analysis of clinical outcome Survival curves were computed in GraphPad Prism software version 9.0.0 according to Kaplan–Meier method. Median follow-up time of enrollment was calculated using the reverse Kaplan–Meier method, also in GraphPad Prism software version 9.0.0. For binary outcomes, 95% two-sided CIs were constructed using the Clopper–Pearson method, also in GraphPad version 9.0.0. An independent board of certified and experienced oncoradiologists performed an external review to evaluate clinical response to address the potential bias of investigator site review. The external review took place at Rigshospitalet, Copenhagen University Hospital. This hospital did not participate in the MM1636 trial, and the external reviewer had no prior knowledge about the clinical trial or the trial therapy. Only PET–CT images were accessed, and arrows indicating target or non-target lesions appeared on baseline images as the only additional information. To address potential trial bias regarding treatment effect, we matched patients in MM1636 with patients from the DAMMED, a population-based database that retrospectively collects data on patients with metastatic melanoma in Denmark. Here, data from 938 patients treated with aPD1 monotherapy contemporaneously (January 2015–October 2019) were extracted. Two hundred and eighteen of these patients were eligible for comparison and matching (all parameters available) (Supplementary Table 1), and 74 patients from the DAMMED were found to match. Patients were matched for age (≤70 years, >70 years), sex, LDH levels (normal, elevated), M stage (M1a, M1b, M1c), BRAF status (wild type, mutated) and PD-L1 status (<1%, ≥1%). An exact matching algorithm was used in which patients in MM1636 were matched with patients from the DAMMED with the same combination of variables. Twenty-nine patients from MM1636 were matched with exact combinations of the six variables. One patient could not be matched. To secure balance of the calculations, control patients were weighted according to the number of patients for each MM1636 patient. Estimates for treatment effects were calculated by weighted logistic regression analyses and weighted Cox proportional hazard model. The R package ‘Matchlt’ was used for matching patients. As the method chosen for matching control patients to protocol patients, a weighted binary logistic regression model was used for comparing response rates in the two matched cohorts. Odds ratios and response rates, including their corresponding 95% CIs, were extracted from the regression models. All P values were two sided, and P values below 0.05 were considered statistically significant. SAS version 9.4M5 was used for the weighted logistic regression models. Processing of peripheral blood mononuclear cells Peripheral blood was collected from all patients in heparinized tubes and was processed within 4 h. In brief, PBMCs were isolated using Lymphoprep (Medinor) separation. PBMCs were counted on a Sysmex XP-300 analyzer and frozen in Human AB Serum (Sigma-Aldrich, H4522-100ML) with 10% DMSO using controlled-rate freezing (CoolCell, BioCision) in a −80 °C freezer and were moved the next day to a freezer at −140 °C until further processing. Needle biopsies at baseline and after the sixth vaccine Two to three needle biopsies (1.2 mm) were taken at baseline and after the sixth cycle of treatment when assessable from the same tumor lesion. One fragment was fixed with formalin and embedded in paraffin (FFPE); one to two fragments were used for expanding tumor-infiltrating T cells and establishing autologous tumor cell lines. Delayed-type hypersensitivity and generation of skin-infiltrating lymphocytes After six cycles of treatment, we performed intradermal injections of vaccine components without adjuvant and one control injection containing DMSO without peptide. Patients were injected with either a mixture of both IDO and PD-L1 peptides at all three injection sites or PD-L1 peptide, IDO peptide or a mixture of the two at the injection sites, respectively (Supplementary Table 6). Eight hours after injection, punch biopsies were resected from the three sites where the peptide was injected and transported immediately to the laboratory and cut into fragments of 1–2 mm3. SKILs were expanded to establish ‘young SKILs’ in CM medium consisting of RPMI 1640 with GlutaMAX, 25 mM HEPES, pH 7.2 (Gibco, 72400-021), IL-2 (100 or 6,000 IU ml−1) (Proleukin Novartis, 004184), 10% heat-inactivated Human AB Serum (Sigma-Aldrich, H4522-100ML), 100 U ml−1 penicillin, 1.25 µg ml−1 Fungizone (Bristol Myers Squibb, 49182) and 100 µg ml−1 streptomycin (Gibco, 15140-122). Half of the medium was replaced three times per week. Young SKILs from samples derived from IDO peptide, PD-L1 peptide and the mixture injection sites were further expanded in a small-scale version of the 14-d rapid expansion protocol as previously described51. Quantification of specific T cells in blood by ELISPOT For enumeration of vaccine-specific T cells in the peripheral blood, PBMCs from patients were stimulated with IDO or PD-L1 peptide in the presence of a low dose of IL-2 (120 U ml−1) for 7–13 d before being used in IFN-γ ELISPOT assays with 2.5–3.2 × 105 cells per well. Briefly, cells were placed in a 96-well PVDF membrane-bottomed ELISPOT plate (MultiScreen MSIPN4W50, Millipore) precoated with IFN-γ-capture antibody (clone 1-D1K, Mabtech). Diluted IDO or PD-L1 peptide stocks in DMSO were added at 5 µM; an equivalent amount of DMSO was added to control wells. PBMCs from each patient were set up in duplicate or triplicate for peptide and control stimulations. Cells were incubated in ELISPOT plates in the presence of the peptide for 16–18 h, after which they were washed off, and biotinylated secondary antibody (anti-human IFN-γ mAB, clone 7-B6-1, Mabtech) was added. After 2 h of incubation, unbound secondary antibody was washed off, and streptavidin-conjugated alkaline phosphatase (Mabtech) was added for 1 h. Next, unbound streptavidin-conjugated enzyme was washed off, and the assay was developed by adding BCIP/NBT substrate (Mabtech). ELISPOT plates were analyzed on the CTL ImmunoSpot S6 Ultimate V analyzer using ImmunoSpot software version 5.1. Responses were calculated as the difference between the average numbers of spots in wells stimulated with IDO or PD-L1 peptide and those from corresponding control wells. For detection of IDO and PD-L1 peptide responses ex vivo, PBMCs were thawed and rested overnight in medium containing DNase I (1 μg ml−1, Sigma-Aldrich, 11284932001) before being used in IFN-γ ELISPOT assays as described above. A total of 6–9 × 105 PBMCs per well were seeded. Statistical analysis of all ELISPOT responses was performed using the DFR method as described by Moodie et al. using RStudio software (RStudio Team, 2016, http://www.rstudio.com/)50. Vaccine-specific ELISPOT responses were defined as true if the difference between the spot count in control and peptide-stimulated wells was statistically significant according to DFR statistical analysis or, for samples performed in duplicate, if the spot count in peptide-stimulated wells was at least 2× the spot count in control wells50. Cancer cell lines and tumor conditioned medium Autologous melanoma cell lines were established from needle biopsies. Briefly, biopsies were chopped into small fragments and seeded in 24-well cultures in RPMI 1640 with GlutaMAX, 25 mM HEPES, pH 7.2 (Gibco, 72400-021), 10% heat-inactivated FBS (Life Technologies, 10500064), 100 U ml−1 penicillin, 1.25 µg ml−1 Fungizone (Bristol Myers Squibb, 49182) and 100 µg ml−1 streptomycin (Gibco, 15140-122). Established adherent melanoma tumor cell lines were cryopreserved at −140 °C in freezing medium containing FBS with 10% DMSO. PD-L1 and HLA-II expression on established tumor cell lines was assessed by flow cytometry staining with anti-PD-L1–PE-Cy7 (1:23 dilution, clone M1h1 (RUO), BD, 558017) and anti-HLA-II–FITC (1:23 dilution, clone Tu39 (RUO), BD, 555558) antibodies. To obtain TCM, established tumor cell lines were cultured in 175-cm2 Nunc cell culture flasks until 80–90% confluency was reached. The culture medium was then replaced with 20 ml fresh X-VIVO 15 with Gentamicin and Phenol Red (Lonza, BE02-060Q), medium with 5% heat-inactivated Human AB Serum (Sigma-Aldrich, H4522-100ML). After 24 h of incubation, TCM was collected and centrifuged to remove any resuspended cells, after which TCM was aliquoted, frozen and stored at −80 °C. The acute monocytic leukemia cell line MonoMac1 was obtained from the DSMZ—German Collection of Microorganisms and Cell Cultures (ACC 252) and cultured in RPMI 1640 with GlutaMAX, 25 mM HEPES, pH 7.2 (Gibco, 72400-021) and 10% heat-inactivated FBS. Isolation of autologous myeloid cells Autologous CD14+ cells were sorted from freshly thawed PBMCs using a magnetic bead-separation kit (Miltenyi Biotec, 130-050-201) according to the manufacturer’s instructions. Isolated CD14+ cells were used as targets in the IFN-γ ELISPOT assay directly after sorting or differentiated in vitro into tumor-associated macrophages by culturing with 1 ml fresh X-VIVO 15 medium with Gentamicin and Phenol Red (Lonza, BE02-060Q) and 5% heat-inactivated Human AB Serum, supplemented with 1 ml autologous TCM in 24-well plates for 2 d. Quantification of vaccine-specific T cells from DTH biopsy sites by ELISPOT SKILs expanded in vitro were rested in medium without IL-2 overnight before being used in the IFN-γ ELISPOT assay as described above to evaluate reactivity of skin-infiltrating T cells. Generation of IDO- and PD-L1-specific T cell cultures from PBMCs or SKILs IDO- or PD-L1-specific T cells were isolated from peptide-stimulated in vitro PBMC cultures on days 14–15 after stimulation or from SKIL cultures expanded in vitro. For specific T cell isolation, PBMCs or SKILs were stimulated with IDO or PD-L1 peptide, and cytokine-producing T cells were sorted using the IFN-γ or TNF-α Secretion Assay—Cell Enrichment and Detection kit (Miltenyi Biotec). Cytokine-production profile of PD-L1- and IDO-specific T cells by intracellular staining To assess the T cell cytokine-production profile, isolated and expanded IDO- and PD-L1-specific T cell cultures were stimulated for 5 h with peptide at 5 μM in a 96-well plate. One hour after the start of the incubation, anti-CD107a–PE (1:133 dilution, clone H4A3, BD Biosciences, 555801) antibody and BD GolgiPlug (1:1,000 dilution, BD Biosciences) were added. After a 5-h incubation, cells were stained using fluorescently labeled surface marker antibodies: anti-CD4–PerCP (1:5 dilution, clone SK3, 345770), anti-CD8–FITC (1:5 dilution, clone SK1, 345772) and anti-CD3–APC-H7 (1:20 dilution, clone SK7, 560275) (all from BD Biosciences). Dead cells were stained using FVS510 (1:250 dilution, BD Biosciences, 564406), followed by overnight fixation and permeabilization using eBioscience Fixation/Permeabilization buffers (00-5123-43, 00-5223-56) according to the manufacturer’s instructions. Cells were then stained intracellularly in eBioscience permeabilization buffer (00-8333-56) with anti-IFN-γ–APC (1:20 dilution, clone 25723.11, BD Biosciences, 341117) and anti-TNF-α–BV421 (1:100 dilution, clone Mab11, BD Biosciences, 562783) antibodies. Samples were analyzed on the FACSCanto II (BD Biosciences) using BD FACSDiva software version 8.0.2. The gating strategy is shown in Supplementary Fig. 3a. For assessing the reactivity against IDO expressing MonoMac1 cells, the appropriate amount of cancer cells was added to the IDO-specific T cells to obtain an effector:target ratio of 4:1 in the 96-well plate and T cell cytokine production was tested after 5 h of co-culture as described above. For HLA-blocking experiments, HLA-blocking antibodies (2 μg ml−1) were added directly into wells 30 min before the addition of peptide. Blocking antibodies used were HLA-DR (1:500 dilution, clone L243, Abcam, ab136320), HLA-DQ (1:500 dilution, clone SPV-L3, Abcam, ab23632) and HLA-DP (1:500 dilution, clone B7/21, Abcam, ab20897). To assess ex vivo T cell reactivity to IDO and PD-L1 peptides in patient PBMCs, cells were thawed and rested for 1–2 d in medium containing DNase I (1 μg ml−1, Sigma-Aldrich, 11284932001). PBMCs were then stimulated with peptide at 5 μM in a 96-well plate for 8 h. An hour after the addition of peptide, anti-CD107a–BV421 (3:50 dilution, clone H4A3, BD Biosciences, 328626) antibody and BD GolgiPlug (1:1,000 dilution, BD Biosciences) were added. Surface and intracellular staining was performed as described above. Antibodies used for surface staining were anti-CD3–PE-CF594 (0.8:30 dilution, clone UCHT1, BD Biosciences, 562280), anti-CD4–BV711 (1:30 dilution, clone SK3, BD Biosciences, 563028) and anti-CD8–Qdot605 (1:150 dilution, clone 3B5, Thermo Fisher, Q10009). Antibodies used for intracellular staining were anti-CD137–PE (1:20 dilution, clone 4B4-1 (RUO), BD Biosciences, 555956), anti-IFN-γ–PE-Cy7 (1.5:20 dilution, clone B27, BD Biosciences, 557643) and anti-TNF-α–APC (1:20 dilution, clone Mab11, BD Biosciences, 554514). Samples were acquired on the NovoCyte Quanteon (ACEA Biosciences) and analyzed using NovoExpress software version 1.4.1. To assess vaccine-specific T cell responses, background values observed in unstimulated PBMC samples were subtracted from values observed in peptide-stimulated conditions. Positive response value threshold was set at a difference of 0.2% from the background values. Based on this response cutoff, only TNF-α, CD107a and CD137 responses were detected in this assay. Statistical analysis comparing baseline with on-treatment or post-treatment cytokine profiles was performed using two-sided Wilcoxon matched-pairs signed-rank test. The gating strategy is shown in Supplementary Fig. 3b. siRNA-mediated PD-L1 and IDO1 silencing A stealth siRNA duplex for targeted silencing of PD-L1 (Invitrogen)52, a custom silencer select siRNA for targeted silencing of IDO1 (Ambion) and the recommended silencer select negative control (Ambion) siRNA for mock transfection were used. The stealth PD-L1 siRNA duplex consisted of the sense sequence 5′-CCUACUGGCAUU-UGCUGAACGCAUU-3′ and the antisense sequence 5′-AAUGCGUUCAGCAAAUGCCAGUAGG-3′. The three silencer IDO siRNA duplexes used were siRNA1 (sense sequence, 5′-ACAUCUGCCUGAUCUCAUATT-3′; antisense sequence, 5′-UAUGAGAUCAGGCAGAUGUTT-3′), siRNA2 (sense sequence, 5′-CCACGAUCAUGUGAACCCATT-3′; antisense sequence, 5′-UGGGUUCACAUGAUCG-UGGAT-3′) and siRNA3 (sense sequence, 5′-CGAUCAUGUGAACCCAAAATT-3′; antisense sequence, 5′-UUUUGGGUUCACAUGAUCGTG-3′). For PD-L1 or IDO1 silencing, cancer cells were electroporated with 0.025 nmol of each siRNA duplex as previously described53. For PD-L1-silencing experiments, cancer cells were treated with IFN-γ (500 U ml−1, PeproTech) 1 h after electroporation. Electroporated cells were used as target cells in ELISPOT and ICS assays 24 h or 48 h after siRNA electroporation. RT–qPCR Total RNA was extracted using the RNeasy Plus Mini kit (Qiagen, 74134) following the manufacturer’s instructions. RNA concentration was quantified using a NanoDrop 2000 (Thermo Fisher Scientific), and a total of 1,000 ng RNA was reverse transcribed with the High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems, 4368814) using 1,000 ng input RNA for transcription. Real-time qPCR analyses were performed using the TaqMan Gene Expression Assay on a Roche LightCycler 480 instrument. RT–qPCR assays were performed with a minimum of three technical replicates and analyzed using the ΔCt method as described in Bookout et al.54 with normalization of IDO1 expression (primer ID, Hs00984148_m1) or PD-L1 expression (primer ID, Hs001125296_m1) to the expression level of the housekeeping gene POLR2A (primer ID, Hs00172187_m1) and the control sample (mock). For low-concentration samples with no amplification, Ct was set to 40. Controls without reverse transcriptase were used as controls for specific amplifications. P values were determined using two-tailed parametric t-tests. BRAF-mutational status and PD-L1 status at baseline in all patients A library of historical FFPE biopsies were assessable for all patients and analyzed locally at Herlev and Gentofte University Hospital by experienced pathologists for BRAF status and PD-L1 expression on tumor cells. The BRAF analysis was carried out with real-time PCR using the EntroGen BRAF Mutation Analysis Kit II (BRAFX-RT64, CE-IVD) to specifically detect mutations corresponding to V600D, V600E and V600K in the BRAF protein. PD-L1 status was assessed using the monoclonal rabbit anti-PD-L1 antibody (clone 28.8, PD-L1 IHC 28-8 pharmDx) in FFPE biopsies following the manufacturer’s instructions. Patients were considered positive for PD-L1 with expression levels ≥1% and negative for PD-L1 with expression levels <1%. Human leukocyte antigen type Blood samples from all 30 patients were genotyped for class I (HLA-A, HLA-B, HLA-C) and three class II types (HLA-DRB1, HLA-DQA1, HLA-DQB1) using LinkSēq HLA Typing Kits (Thermo Fisher, 1580C). These test kits are based on real-time PCR using allele-specific exponential amplification (sequence-specific primers) followed by melting curve analyses. Immunohistochemistry simplex: Immunoscore CR IHC staining was performed using a qualified Ventana Benchmark XT with four different steps: (1) antigen retrieval; (2) staining with the following primary antibodies: anti-CD3 antibody (clone HDX2; provider, HalioDx; HD-FG-000013; 10931065/10636667; concentration, 0.25 µg ml−1), anti-CD8 antibody (clone HDX3; provider, HalioDx; HD-FG-000019; 10931069/10337710/10639301; concentration, 0.5 µg ml−1), anti-IDO monoclonal antibody (clone VINC3IDO; provider, Thermo Fisher Scientific; 14-9750-82; E25003-101; concentration, 0.05 µg ml−1), anti-HLA class 1 ABC antibody (clone, EMR8-5; provider, Abcam; ab70328; GR3248333/GR3186494; concentration, 0.5 µg ml−1), anti-HLA-DR/DP/DQ/DX antibody (clone CR3/43; provider, Santa Cruz; sc-53302; L1714; concentration, 1 µg ml−1) and anti-PD-L1 antibody (clone HDX3; provider, HalioDx; HD-FG-000035; 106312810/106312816; concentration, 3.3 µg ml−1); (3) detection with a secondary antibody using the ultraView Universal DAB Detection kit; and (4) counterstaining using Hematoxylin & Bluing Reagent (staining of cellular nuclei). Control slides were systematically included in each staining run to permit quality control of the obtained measurements. Following coverslipping, slides were scanned with the NanoZoomer-XR to generate digital images (20×). Two consecutive slides were specifically used to perform Immunoscore CR TL staining of CD3+ and CD8+ cells. Digital pathology of T lymphocytes The digital pathology for Immunoscore CR TL allowed the quantification of positive cells (in cells per mm2) into the core tumor and invasive margin if present. Each sample was analyzed using the HalioDx Digital Pathology Platform. Digital pathology and pathologist analysis of IDO Digital pathology for IDO allowed the quantification of stained area (in mm2) into the whole tumor. Each sample was analyzed using the HalioDx Digital Pathology Platform. Analysis of IDO+ cells was performed by a pathologist. Results were expressed as an H score from 0 to 300. The score is obtained by the formula 3 × percentage of cells with strong staining + 2 × percentage of cells with moderate staining + percentage of cells with weak staining. Digital pathologist analysis of MHC-I and MHC-II Analysis of MHC-I+ and MHC-II+ cells was performed by a pathologist. Results are expressed as a percentage of positive tumor cells and a percentage of positive cells in the stroma. Digital pathology Immunoscore immune checkpoint (CD8+ and PD-L1+) The Immunoscore CR IC test allowed the quantification of CD8++ cell density in the whole tumor and a CD8+-centered proximity index (which corresponds to the percentage of CD8++ cells that have at least one PD-L1+ cell in the neighborhood) at different cutoff distances (20 µm, 40 µm, 60 µm and 80 µm). Pathologist analysis of PD-L1 A pathologist performed analysis of PD-L1+ cells. Positivity of a viable tumor cell was considered when partial or complete cell membrane staining was observed (more than 10% of the tumor cell membrane). Results were expressed as a percentage. NanoString RNA profiling and Immunosign RNA was extracted from FFPE tissues using Qiagen RNeasy FFPE extraction kits (Qiagen). Annotations from the pathologist performing H&E staining were used to guide removal of normal tissue from the slides by macrodissection before nucleic acid extraction, which occurred after tissue deparaffinization and lysis. Each extracted RNA sample was independently quantified using a NanoDrop spectrophotometer (NanoDrop Technologies) and qualified (Agilent Bioanalyzer). Degradation was quantified as the percentage of RNA fragments smaller than 300 bp using the RNA 6000 Nano kit (Agilent Bioanalyzer). Good sample quality was defined as less than 50% of RNA fragments being 50–300 bp in size. RNA expression profiling was performed using the nCounter PanCancer Immune Profiling Panel from NanoString (NanoString Technologies). The PanCancer Immune Profiling Panel contains 776 probes and is supplemented with six genes to complete HalioDx Immunosign targets. Hybridization was performed according to the manufacturer’s instructions. Hybridized probes were then purified and immobilized on a streptavidin-coated cartridge using the nCounter Prep Station (NanoString Technologies). Data collection was carried out on the nCounter Digital Analyzer (NanoString Technologies) following the manufacturer’s instructions to count individual fluorescent barcodes and quantify target RNA molecules present in each sample. For each assay, a scan of 490 fields of view was performed. Raw data from the NanoString nCounter were processed using NanoString recommendations. Quality control enables to retain data of good quality with a binding density that ranges between 0.05 and 2.25. The linearity of positive controls was checked using the R2 value of regression between the counts and the concentrations of positive controls. Samples that showed R2 < 0.75 were flagged and removed from the analysis. The background was removed using a thresholding method at the mean + 2 s.d. of negative controls. Raw counts were normalized using a positive normalization factor. Samples showing positive normalization factors outside the range of 0.3–3 were removed from the analysis. A second normalization was performed using the housekeeping gene normalization factor. Only the most stable housekeeping genes were selected for this normalization step using the variance-versus-mean relationship. All samples showing a normalization factor outside the range of 0.1–10 were removed from the analysis. All statistical analyses were performed on normalized counts using R software (version 2.6.2, 2019-12-12). T cell receptor variable β chain sequencing To track longitudinal immune responses to therapy, genomic DNA was extracted from longitudinal pre- and post-treatment PBMCs (five patients), pre- and post-treatment biopsies (five patients) (both FFPE and treated with RNAlater) and IDO- and PD-L1-specific T cell cultures from PBMCs (five patients) or SKILs (one patient). Three clonal (two IDO- and one PD-L1-) specific cultures were also sequenced to confirm clonal purity of cultures. DNA from PBMCs or RNAlater-treated biopsies was extracted with the DNeasy Blood and Tissue kit (Qiagen, 69504), DNA from sorted IDO and PD-L1-specific T cells from either PBMCs or SKILs was extracted using the QIAamp DNA Micro kit (Qiagen, 565304), and DNA from PFFE biopsies was extracted using the Maxwell RSC DNA FFPE kit (Promega, AS1450). Immunosequencing of the CDR3 regions of human TCRβ chains was performed using the immunoSEQÒ Assay (Adaptive Biotechnologies). Extracted genomic DNA was amplified in a bias-controlled multiplex PCR, followed by high-throughput sequencing. Sequences were collapsed and filtered to identify and quantitate the absolute abundance of each unique TCRβ CDR3 region for further analysis as previously described55–57. Statistical analyses of TCRβ sequencing results Two quantitative components of diversity were compared across samples in this study. First, Simpson clonality was calculated on productive rearrangements by ∑i=1Rpi2, where R is the total number of rearrangements and pi is the productive frequency of rearrangement i. Values of Simpson clonality range from 0 to 1 and measure how evenly receptor sequences (rearrangements) are distributed among a set of T cells. Clonality values approaching 0 indicate a very even distribution of frequencies, whereas values approaching 1 indicate an increasingly asymmetric distribution in which a few clones are present at high frequencies. Second, sample richness was calculated as the number of unique productive rearrangements in a sample after computationally downsampling to a common number of T cells to control for variation in sample depth or T cell fraction. Repertoires were randomly sampled without replacement five times, and we report the mean number of unique rearrangements. The T cell fraction was calculated by taking the total number of T cell templates and dividing by the total number of nucleated cells. The total number of nucleated cells was derived from reference genes using the immunoSEQ Analyzer version 3. To identify enriched vaccine clones in each patient, rearrangement frequencies in their baseline PBMCs and each IDO/PD-L1-sorted T cell sample were compared using a binomial distribution framework as previously described58. In brief, for each clone, we performed a two-sided test to determine whether frequencies were the same in the patient’s periphery and a PD-L1- or IDO-specific T cell sample. The Benjamini–Hochberg procedure was used to control the false discovery rate at 0.01 (ref. 59). Clonal expansion in post-treatment samples was similarly assessed using this differential abundance framework, but an IDO/PD-L1-specific T cell sample was replaced with a post-treatment series sample. In biopsies, the six series frequencies were compared to those of baseline tissue. Lastly, vaccine-associated clones were tracked in each PBMC and tissue sample by summing the frequency of each rearrangement enriched in either PD-L1- or IDO-specific T cells. All statistical analyses were performed in R version 3.6.1. Reporting Summary Further information on research design is available in the Nature Research Reporting Summary linked to this article. Online content Any methods, additional references, Nature Research reporting summaries, source data, extended data, supplementary information, acknowledgements, peer review information; details of author contributions and competing interests; and statements of data and code availability are available at 10.1038/s41591-021-01544-x. Supplementary information Supplementary Information Supplementary Figs. 1–3 and Tables 1–6 Reporting Summary Extended data Extended Data Fig. 1 Mode of Action and treatment plan. a) Anticipated mechanism of action of the combination therapy of an IDO/PD-L1 derived peptide vaccine and nivolumab (anti-PD1). 1) The IDO/PD-L1 peptide vaccine is administered subcutaneously (s.c) and nivolumab is administered intravenous (IV). 2) The vaccines peptides are phagocytosed by an antigen presenting cell and presented to IDO and PD-L1 specific T cells, which are activated. 3) The activated T cells migrates to the tumor site where they attack both immune-suppressive cells and tumour cells expressing IDO and/or PD-L1 leading to cytokine production and a pro-inflammatory tumour microenvironment. 4/5) Enhanced tumour killing by both IDO/PD-L1 specific T cells and tumor specific cytotoxic T cells due to PD-1 blockade. Created with BioRender.com b) Treatment plan. After written informed content patients were screened. Before treatment start a baseline PET/CT scan was performed, baseline blood sample for research use and if assessable a baseline needle biopsy. Patients were treated with the IDO/PD-L1 peptide vaccine subcutaneously biweekly for the first 6 injections and thereafter every fourth week for a maximum of 15 vaccines. Nivolumab was administered in parallel biweekly (3mg/kg) up to 24 series. If patients needed subsequent nivolumab after ended vaccination regiment they were treated with 6 mg/kg every fourth week up to two years. Needle biopsy and delayed type hypersensitivity (DTH) was performed after 6 series of treatment if assessable. PET/CT scans was performed every third month. Extended Data Fig. 2 Progression free survival and overall survival in matched historical control group and vaccine injection site reaction. a) Kaplan-Meier curve of progression free survival with corresponding 95% confidence intervals in the matched historical control group (n=74). Patients were matched on BRAF-status, PD-L1-status, age, gender, M-stage and LDH level). b) Kaplan-Meier curve of overall survival with corresponding 95% confidence intervals in the matched historical control group (n=74). c) Images of injection site reaction in patient MM20 (CR) after 11 vaccination show redness, rash and granuloma at injection site. Extended Data Fig. 3 Vaccine specific responses in blood. a) Heatmap of specific (background has been subtracted) IDO and PD-L1 responses in PBMCs at baseline, series 3, 6, 12, 18 and 24 measured by in vitro IFN-γ Elispot assay show fluctuations in the blood during treatment (n=30). b) Heatmap of specific (background has been subtracted) IDO and PD-L1 responses in PBMCs at baseline and 3 and 6 months after last vaccine measured by in vitro IFN- γ Elispot assay (n=30). 2.5–3.2x105 cells per well were used. c) Vaccine associated clones were tracked in the blood by summing the frequency of each rearrangement enriched in either IDO or PD-L1 T cells. Extended Data Fig. 4 Ex vivo vaccine specific responses in blood. Heatmaps of detected specific (background has been subtracted) IDO (a) and PD-L1 (b) responses in PBMCs at baseline and on/after treatment as measured by IFNγ ELISPOT (n=25). C) Example of well images of ex vivo ELISPOT wells for three different patients (n=3). 6-9x105 cells per well were used. Extended Data Fig. 5 CD4 and CD8 vaccine specific T cell responses in blood. Top: heatmaps of IDO specific CD4 (left) and CD8 (right) T cell responses in PBMCs at baseline and on/post treatment. Bottom: heatmaps of PD-L1 specific CD4 (left) and CD8 (right) T cell responses. Responses quantified by flow cytometry by an increased expression of CD107α, CD137 and TNFα after 8h peptide stimulation. Values represent specific responses after background values have been subtracted (n=21). Statistical analysis were performed using two-sided Wilcoxon matched-paired rank test. Extended Data Fig. 6 Pro-inflammatory profiles of sorted CD4 and CD8 T cells from blood. a/b/c/d) Percentage of in vitro stimulated and sorted PD-L1 and IDO specific CD4 and CD8 T cells secreting cytokines. Extended Data Fig. 7 Vaccine specific responses in skin. a) IDO and PD-L1 specific T cell responses in SKILs after 6 series of treatment measured by IFN-γ Elispot assay (n=13). SKILs were grown from DTH injection with either IDO peptide, PD-L1 peptide or a mix as presented by different blue colours. * Responses were calculated as the difference between average numbers of spots in wells stimulated with IDO or PD-L1 peptide (triplicates) and corresponding control (DMSO) and statistical analyses of Elispot responses were performed using distribution-free resampling method (Moodie et al.). DR: Not statistically confirmed response due to replicate number but number of spots in peptide wells are two times higher than control wells (DMSO). NS: No significant response and no DR. b/c/d) Percentage of cytokine secreting/CD107a+ CD4+ and CD8+ IDO and PD-L1 specific T cells in response to in vitro peptide stimulation by flow cytometry. e) Skin infiltrating PD-L1 specific T cell clones also found in biopsy in patient MM01. TCR sequencing was performed on a PD-L1 specific T cell culture generated from DTH area on the lower back injected with PD-L1 peptide on patient MM01. Bars show the frequency of top 25 clones in the culture with which indicates a high Simpson clonality of 0.43. f) Tracking the frequency of the top five skin infiltrating PD-L1 specific clones in tumour before and after treatment. Extended Data Fig. 8 T cell changes in blood after treatment. T cell fraction, TCR clonality and repertoire richness in blood and peripherally expanded TCR clones in both responding and non-responding patients. a) T cell fraction in peripheral blood of five patients at baseline, series 3, 6 and 12 by TCR sequencing. T cell fraction was calculated by taking the total number of T cell templates and dividing by the total number of nucleated cells (n=5.) b) TCR clonality in peripheral blood of five patients at baseline, series 3, 6 and 12 by TCR sequencing (n=5). Simpson clonality measures how evenly TCR sequences are distributed amongst a set of T cells where 0 indicate even distribution of frequencies and 1 indicate an asymmetric distribution where a few clones dominate. c) TCR repertoire richness in peripheral blood of five patients at baseline, series 3, 6 and 12 by TCR sequencing (n=5). TCR repertoire richness report the mean number of unique rearrangements. d) Bar-chart chart representing dynamics of the expanded T cell clones. Coloured bars represent peripherally expanded clones in five patients at series 3, series 6 and series 12 as compared to the baseline PBMC samples (n=5). Light gray bar represents peripherally expanded clones that are present in baseline biopsy samples while dark gray bar represents peripherally expanded clones that are present in the post-treatment biopsy (after 6 series). e) Frequency of the dominant TCR β chain in clonal PD-L1 and IDO specific cultures as determined by CDR3 sequencing. Extended Data Fig. 9 Profiling of genes relevant to T cell activation, cytokines and exhaustion markers on pre- and post-treatment biopsies in two patients. a) RNA expression profiling of genes related to T cell activation was performed using NanoString nCounter (n=2). b) RNA expression profiling of genes related to cytokine activity was performed using NanoString nCounter (n=2). c) RNA expression profiling of genes related to checkpoint inhibitors was performed using NanoString nCounter (n=2). Extended Data Fig. 10 Treatment induced upregulation of PD-L1, IDO, MHC I and MHC II and distance between CD8 T cells and PD-L1 expressing cells. a) IHC on 4 paired biopsies stained for CD3 and CD8 T cells, PD-L1, MHC I and MHC II on tumor cells/mm2 (sum in the validated area) and IDO H-score (from 0 to 300). H-score is the expression of IDO on both immune and tumor cells: The score is obtained by the formula: 3 x percentage of cells with a strong staining + 2 x percentage of cells with a moderate staining + percentage of cells with a weak staining (n=4). b) Distance in µm between CD8+ T cells and PD-L1+ stained cells on five baseline biopsies detected by IHC (n=5). Extended data is available for this paper at 10.1038/s41591-021-01544-x. Supplementary information The online version contains supplementary material available at 10.1038/s41591-021-01544-x. Acknowledgements We thank all patients and their relatives for being a part of the trial. We thank Ö. Met, M. Jonassen, S. Ullitz Færch, B. Saxild, S. Wendt and C. Grønhøj for technical support. We thank M. Cumberbatch for input with translational analysis. We thank the nurses at clinic 5 and the head of the Oncology Department at Herlev and Gentofte Hospital, L. Sengeløv. The study was funded through a research funding agreement between IO Biotech and the CCIT-DK, Herlev Hospital and the Oncology Department at Herlev Hospital. Author contributions I.M.S. and M.H.A. conceived the study. J.W.K. wrote the protocol with input from I.M.S., M.H.A. and M.-B.Z. J.W.K. and C.L.L. were responsible for patient treatment and patient care with help from R.B.H. and consulting from I.M.S., E. Ellebaek and M.D. I.M.S., E. Ellebaek and M.D. were responsible for patient recruitment. J.W.K. and C.L.L. were responsible for coordinating trial procedures and collecting data and samples. J.W.K. and C.L.L. were clinical investigators, and I.M.S. was a sponsor. J.W.K. and C.L.L. analyzed and interpreted clinical data with support from I.M.S. and E. Ehrnrooth. T.W.K. performed statistical analyses. H.W.H. helped with clinical imaging. J.W.K., E.M., C.O.M., S.M.A., S.E.W.-B., M.O.H. and A.W.P. analyzed and interpreted translational data with support from M.H.A. and I.M.S. Funding was acquired by I.M.S., M.H.A. and M.-B.Z. J.W.K. wrote the manuscript with input from I.M.S., M.H.A., E.M. and A.W.P. All authors reviewed the manuscript, interpreted data and approved the final version. Data availability This clinical trial was registered at https://www.clinicaltrials.gov/ before patient enrollment (clinical trial identifier NCT03047928). Each vaccine was composed of 100 µg IO102, a 21-amino-acid peptide (DTLLKALLEIASCLEKALQVF) from the IDO peptide, and 100 µg IO103, a 19-amino-acid peptide (FMTYWHLLNAFTVTVPKDL) from the signal peptide of PD-L1. TCR sequencing data are available from Adaptive Biotechnologies. Upon request, the CCIT-DK office will provide a username and a password to access the designated data within approximately 2–4 weeks (https://www.herlevhospital.dk/ccit-denmark/find-us/Sider/Contact-information.aspx). All requests for the remaining data including raw data and analyzed data and materials will, within a reasonable time frame, be reviewed by the CCIT-DK office (https://www.herlevhospital.dk/ccit-denmark/find-us/Sider/Contact-information.aspx) to verify whether the request is subject to any intellectual property or obligations. Patient-related data not included in the paper were generated as part of clinical trials and may be subject to patient confidentiality. Any data and materials that can be shared will be released via a material-transfer agreement. The following database was used in the study: https://research.regionh.dk/da/publications/the-danish-metastatic-melanoma-database-dammed(32749d99-095f-4cae-b5de-769bae27f01e).html. Competing interests M.H.A. is named as an inventor on various patent applications relating to therapeutic uses of IDO and PD-L1 peptides. These patent applications are assigned to the company IO Biotech, which is developing immune-modulating cancer treatments. M.H.A. is a founder, shareholder and advisor for IO Biotech. I.M.S. is a cofounder, shareholder and advisor for IO Biotech. I.M.S. has an advisory board relationship with or lectured for Roche, Novartis, MSD, Celgene, Incyte, TILT Bio, Pfizer and BMS AstraZeneca and has received limited grants for translational research from BMS, Roche and Novartis. M.-B.Z. is the CEO, founder and shareholder at IO Biotech. E.M., A.W.P. and E. Ehrnrooth are employees at IO Biotech. E. Ellebaek has received honoraria from BMS, Pierre Fabre, Roche and Kyowa Kirin and travel support from MSD. M.D. has received honoraria from Genzyme, MSD, BMS, Roche and Novartis and travel support from Novartis, MSD, BMS, Roche and Pfizer. The remaining authors declare that they have no conflict of interest. Peer review information Nature Medicine thanks George Coukos, Michael Schell and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. Javier Carmona was the primary editor on this article and managed its editorial process and peer review in collaboration with the rest of the editorial team. Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. These authors contributed equally: Julie Westerlin Kjeldsen, Cathrine Lund Lorentzen. Change history 3/8/2022 A Correction to this paper has been published: 10.1038/s41591-022-01771-w

NIVOLUMAB

DrugsGivenReaction

CC BY

34887574

2021-12-09

Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Multiple organ dysfunction syndrome'.

A phase 1/2 trial of an immune-modulatory vaccine against IDO/PD-L1 in combination with nivolumab in metastatic melanoma. Anti-programmed death (PD)-1 (aPD1) therapy is an effective treatment for metastatic melanoma (MM); however, over 50% of patients progress due to resistance. We tested a first-in-class immune-modulatory vaccine (IO102/IO103) against indoleamine 2,3-dioxygenase (IDO) and PD ligand 1 (PD-L1), targeting immunosuppressive cells and tumor cells expressing IDO and/or PD-L1 (IDO/PD-L1), combined with nivolumab. Thirty aPD1 therapy-naive patients with MM were treated in a phase 1/2 study ( https://clinicaltrials.gov/ , NCT03047928). The primary endpoint was feasibility and safety; the systemic toxicity profile was comparable to that of nivolumab monotherapy. Secondary endpoints were efficacy and immunogenicity; an objective response rate (ORR) of 80% (confidence interval (CI), 62.7-90.5%) was reached, with 43% (CI, 27.4-60.8%) complete responses. After a median follow-up of 22.9 months, the median progression-free survival (PFS) was 26 months (CI, 15.4-69 months). Median overall survival (OS) was not reached. Vaccine-specific responses assessed in vitro were detected in the blood of >93% of patients during vaccination. Vaccine-reactive T cells comprised CD4+ and CD8+ T cells with activity against IDO- and PD-L1-expressing cancer and immune cells. T cell influx of peripherally expanded T cells into tumor sites was observed in responding patients, and general enrichment of IDO- and PD-L1-specific clones after treatment was documented. These clinical efficacy and favorable safety data support further validation in a larger randomized trial to confirm the clinical potential of this immunomodulating approach. pmcMain Despite remarkable advances in the treatment of MM with immune checkpoint inhibitors (ICIs) targeting PD-1 and cytotoxic T lymphocyte antigen 4 (CTLA-4), around half of patients are resistant to ICI monotherapy1. The combination of anti-CTLA-4 (aCTLA-4) and aPD1 therapy is to date the most effective therapy resulting in a response rate of around 60%; however, 50% of the patients also develop severe adverse events2,3. Therefore, an equally effective but less toxic treatment is highly needed. Several approaches to enhance ICI efficacy are currently being investigated, such as other ICIs, T cell therapy with tumor-infiltrating T cells or innate immunity stimulators such as Toll-like receptor 9 agonists4–6. Treating cancer patients with vaccines that stimulate a targeted immune response is another attractive approach, with very few side effects observed thus far in combination immunotherapy studies7,8. Immune-modulatory vaccines targeting tumoral immune escape mechanisms offer a new, generalizable strategy compared to patient-specific neoantigen cancer vaccines7,9. The immune-modulatory vaccine strategy in this clinical trial is based on the finding of circulating cytotoxic T cells specific to IDO and PD-L1 in the blood of patients with cancer and, to a lesser extent, in healthy donors. IDO- and PD-L1-specific CD8+ T cells can directly recognize and kill IDO+ and/or PD-L1+ tumor cells and likewise recognize and kill non-malignant cells that express their cognate targets. Furthermore, IDO- and PD-L1-specific CD4+ T cells release pro-inflammatory cytokines in response to IDO- or PD-L1-expressing target cells. IDO and PD-L1 are expressed not only by melanoma cells but also by many other cell types in the tumor microenvironment (TME), which differentiates these antigens from traditional tumor antigens used in other studies10–15. Activation of IDO/PD-L1-specific T cells by vaccination can therefore restrict the range of immunosuppressive signals mediated by immunosuppressive cells and thereby revert the TME from an immune hostile to an immune friendly environment. In animal models of cancer, vaccinations with IDO epitopes resulted in anti-tumor therapeutic effects that were correlated with reductions in IDO expression in myeloid cell populations within the TME16. The IDO/PD-L1 immune-modulating vaccine may lead to a translatable strategy for improving the efficacy of aPD1 therapy through activation of specific T cells. We hypothesize that the IDO/PD-L1 vaccine attracts T cells into the tumor, which induces type 1 helper T (TH1) cell inflammation and reverts the TME into an immune-permissive site, thereby turning the tumor ‘hot’. This would also upregulate PD-L1 expression in cancer and immune cells, generating more susceptible targets to aPD1 therapy (Extended Data Fig. 1a). This theory was confirmed in a mouse model in which aPD1 therapy and IDO vaccination show synergistic effects16. In this phase 1/2 clinical trial, MM1636, patients with MM received a combination of the IDO/PD-L1 (IO102/IO103) peptide vaccine with the adjuvant Montanide and the aPD1 antibody nivolumab. Patients were included in three cohorts: 30 aPD1 therapy-naive patients (cohort A), ten aPD1 therapy-refractory patients (cohort B, de novo resistance) and ten patients who progressed after aPD1 therapy (cohort C, acquired resistance). Here, we report results from cohort A. The vaccine was given biweekly for the first six administrations and thereafter every 4th week. A maximum of 15 vaccines were administered. Nivolumab was given in parallel, biweekly (3 mg per kg) or every 4th week (6 mg per kg) for up to 2 years (Extended Data Fig. 1b). The primary objective was safety and feasibility. Secondary objectives were immunogenicity and clinical efficacy. Results Patients and treatment Thirty patients were enrolled from December 2017 to June 2020. None of the 30 patients dropped out of the study; all received at least three cycles of therapy (Supplementary Fig. 1). At the current database lock (5 October 2020), six patients were still on treatment in the trial. Of the 24 patients who were not on trial treatment at data cutoff, two are still receiving nivolumab monotherapy (6 mg per kg every 4 weeks). Reasons for stopping treatment for the remaining 22 patients included disease progression (37%), toxicity (20%), maximum benefit or complete response (CR) confirmed on two consecutive scans (17%) or completing 2 years of treatment (7%). For the 24 patients who were not on trial treatment at data cutoff, the mean number of vaccinations was 10.5 (range, 3–15 vaccinations). Thirteen of these 24 patients continued nivolumab (6 mg per kg, every 4 weeks) as a standard of care. Nine patients received subsequent therapy after progression (Supplementary Table 1). Baseline characteristics are shown in Table 1. The mean age was 70 years; 37% of patients had elevated lactate dehydrogenase (LDH) levels, 60% were stage M1c, 37% had BRAF mutations, and 43% were negative for PD-L1 (<1%). A total of three patients (10%) had received prior ipilimumab therapy. No patients had brain metastasis (Supplementary Table 2).Table 1 Baseline patient characteristics (n = 30) Characteristic Number (%) Mean age, years (range) 70 (46–85) Sex, male 16 (55%) ECOG PS 0 26 (87%) LDH levels  ≤ULN 19 (63%)  >ULN 11 (37%) M stage (AJCC-8)  M1a 6 (20%)  M1b 6 (20%)  M1c 18 (60%) Number of lesion sites  1 6 (20%)  2–3 17 (57%)  >3 7 (23%) Liver metastases present 10 (33%) BRAF status  Mutant 11 (37%)  Wild type 19 (63%) PD-L1  <1% 13 (43%)  >1% 17 (57%) Previous systemic therapy  Ipilimumab 3 (10%)  No 27 (90%) AJCC-8, eighth edition of the American Joint Committee on Cancer; ECOG PS, Eastern Cooperative Oncology Group performance status; ULN, upper limit of normal. Notable clinical responses to the combination therapy Thirty patients with MM were treated with the IDO/PD-L1 vaccine and nivolumab according to the trial protocol. By investigator review, the ORR reached 80% (CI, 62.7–90.5%), with 43% of patients (CI, 27.4–60.8%) achieving a CR and 37% (CI, 20.9–54.5%) reaching a partial response (PR) as the best overall response, while 20% experienced progressive disease (PD) according to Response Evaluation Criteria in Solid Tumors (RECIST) 1.1 (Fig. 1a). Two of the patients with a PR did not have a confirmatory scan stating PR on two consecutive scans. Early onset of response was frequent, with 22 of 30 patients having an objective response at the first evaluation (after 12 weeks on treatment). Median times to PR and CR were 75 d (range, 54–256 d) and 327 d (range, 73–490 d), respectively (Fig. 2a–c).Fig. 1 Clinical response. a, Pie charts with percent ORR, CR, PR and PD according to RECIST 1.1 by investigator review of all patients (n = 30), PD-L1+ patients (>1%, (n = 17)) and PD-L1− patients (<1%, n = 13)), respectively. Two-sided CIs (95%) were constructed using the Clopper–Pearson method. b, Treatment effect in MM1636 compared with a matched historical control group from the DAMMED database (n = 74). Patients in MM1636 (n = 29) were matched with the exact same combination variable according to age (≤70 years, >70 years), sex, LDH levels (normal, elevated), M stage (M1a, M1b, M1c), BRAF status (wild type, mutated) and PD-L1 status (<1%, ≥1%). Estimates for treatment effects were calculated by weighted logistic regression analyses and weighted Cox proportional hazard model. Bar height indicates the estimated response rate; tops of bars are centers for error bars. Odds ratios (OR), response rates and their corresponding 95% CIs were extracted from the regression model. All P values were two sided, and P values below 0.05 were considered statistically significant. c, Best change in the sum of target lesion size compared with that at baseline (n = 30). The horizontal line at −30 shows the threshold for defining an objective response in the absence of non-target disease progression or new lesions according to RECIST 1.1. Two patients with 100% reduction in target lesion size had non-target lesions present. White stars, six patients had normalization (<10 mm) of fluor-18-deoxyglucose (FDG)-negative lymph nodes (at baseline, lymph nodes were >1.5 cm and FDG+) and 100% reduction of non-lymph node lesions and are considered to have had a CR (green bar). Black star, one patient (MM29) was considered to have had a PR (blue bar), although he had not reached a −30% change in target lesion size. The patient had a single measurable 13-mm lung metastasis at baseline and multiple biopsy-verified cutaneous metastases on the left crus (not detectable by positron-emission tomography–computed tomography (PET–CT) at baseline). The best change in target lesion size was 10 mm, and a post-treatment biopsy from the cutaneous metastases showed no sign of malignancy. Thus, overall, the patient was classified as having a PR. d, Kaplan–Meier curve of the response duration in the 24 patients with an objective response. e, Kaplan–Meier curve of PFS in all 30 treated patients. f, Kaplan–Meier curve of OS in all 30 treated patients. Fig. 2 Clinical response. a, Swimmer plot showing response duration and time to response according to RECIST 1.1 for all treated patients (n = 30). Triangles indicate first evidence of PR, while squares indicate first evidence of CR. Closed circles indicate time of progression. Arrows indicate ongoing responses. Patient MM18 died due to nivolumab-induced side effects. b, Spider plot showing response kinetics in all treated patients (n = 30). Red squares indicate time of progression. c, PET–CT images of patient MM42 before and after treatment (after 12 series of treatment) showing FDG metabolism in target lesions. SC, subcutaneous. The ORR among PD-L1+ (>1% (clone 28.8)) patients (n = 17) was 94.1% (CI, 73–99.7%) and 61.5% (CI, 35.5–82.3%) in PD-L1− patients (n = 13) (Fig. 1a). Objective responses were observed in patients irrespective of human leukocyte antigen (HLA) genotype (Supplementary Fig. 2). Clinical response data were validated by blinded independent external review, in which an ORR of 76.6% (CI, 57.7–90.1%) was reported, with 53.3% of patients achieving a CR, 23.3% achieving a PR and 3.3% experiencing stable disease. Comparisons between investigator review and external review are outlined in Supplementary Table 3. To examine whether the very high response rate should be attributed to nivolumab or to the vaccine, we retrieved clinical response information from a matched historical control group using the Danish Metastatic Melanoma Database (DAMMED) from contemporaneously treated patients with stage 3–4 melanoma who received aPD1 monotherapy17. Patients were matched with the exact same combination variable according to age, sex, PD-L1 status, BRAF status, LDH level and M stage (those at stage M1d were excluded from the control group (no patients with brain metastasis)). Matched controls were identified for 29 patients, and the ORR of 79.3% (CI, 61.0–90.4%) observed in MM1636 was found to be significantly higher (P < 0.0012) than that in the matched control group in which an ORR of 41.7% (CI, 31.0–53.3%) was reached. Furthermore, of the 29 patients in MM1636, a significantly (P < 0.0017) higher percentage (41.4% (CI, 25.2–59.6%)) of patients achieved CR in MM1636 than that (12% (CI, 6.3–21.6%)) in the matched historical control group. ORR and complete response rate (CRR) in the matched historical control group were comparable to those of patients treated in randomized phase 3 pivotal trials with aPD1 monotherapy18 (Fig. 1b). The treatment leads to prolonged PFS At data cutoff, the median duration of response had not been reached, with 87% of all responding patients being free from progression at 12 months (Fig. 1d). Patients were followed for up to 35 months with a median follow-up time of 22.9 months (CI, 14.9–26.2 months). OS and PFS were calculated from the first day of treatment to death or progression or to the date of the last follow-up (5 October 2020). The median PFS (mPFS) for all treated patients was 26 months (CI, 15.4–69 months) and was not reached for responding patients (Fig. 1e). The median OS was not reached at the data cutoff. OS at 12 months was 81.6% (CI, 61.6–92%) (Fig. 1f). One patient (MM18) with a CR died from nivolumab-related severe adverse events; the remaining patients died because of metastatic melanoma. For comparison, the mPFS was 8.3 months (CI, 5.5 months–NR (not reached)) in the matched historical control group (n = 74), while the median OS was 23.2 months (CI, 23.2 months–NR). (Extended Data Fig. 2a,b). The combination of the IDO/PD-L1 vaccine and nivolumab was safe Treatment-related adverse events are listed for all 30 patients in Supplementary Table 4. Common treatment-related grade 1–2 toxicities were fatigue (47%), rash (47%), arthralgia (30%), diarrhea (23%), nausea (23%), dry skin (20%), pruritus (20%), infusion reaction (17%), xerostomia (17%) and myalgia (17%). Four patients (13%) experienced grade 3–4 adverse events: one patient with a grade 3 maculopapular rash (MM01), one patient with grade 3 adrenal insufficiency (MM06) and one patient with grade 3 arthralgia (MM22). Patient MM18 died from urosepsis with multi-organ failure and severe hyponatremia. This patient had experienced multiple immune-related adverse events with grade 3 colitis, grade 2 pneumonitis, grade 3 arthralgia, grade 2 vasculitis and grade 2 nivolumab infusion-related allergic reaction. Additionally, patient MM18 had symptoms of myocarditis at the time of death with highly elevated cardiac troponin I levels. Bedside echocardiography showed an ejection fraction of 15%, which at baseline was 60%, but an autopsy was not conducted, and myocarditis was never confirmed pathologically. Patient MM06 had received first-line treatment with ipilimumab before entering the trial and was on substitution corticosteroids at the time of inclusion. Adrenal insufficiency was aggravated by an erysipelas infection with high fever, reaching grade 3 in Common Terminology Criteria for Adverse Events and resolving quickly after appropriate antibiotic therapy was initiated. As expected, local side effects were common with 77% of the patients who developed injection site reactions. These reactions were classified as granulomas (63%), redness (20%), pain (13%) and pruritus (13%) at the injection site. All local reactions were grade 1–2, most likely related to the Montanide adjuvant and typically transient. However, two patients (MM07 and MM20) decided to discontinue vaccination after eight and 11 injections, respectively, due to granulomas, tenderness and pain that limited instrumental activities of daily living but continued to receive nivolumab (Extended Data Fig. 2c). Vaccine-specific responses in blood were frequently detected First, all 30 patients were assessed for the presence of vaccine-specific responses in peripheral blood mononuclear cells (PBMCs) before, on and after vaccination using a modified interferon (IFN)-γ enzyme-linked immune absorbent spot (ELISPOT) assay. This assay is known to expand antigen-specific memory cells and to improve the detection and the correlative power of ELISPOT during treatment19–21. Prevaccine IDO-specific responses were detectable in ten (33%) patients, while prevaccine PD-L1-specific responses were detectable in eight patients (27%); overlapping (specific to both IDO and PD-L1) prevaccine responses were present in four (13.3%) patients. During vaccination, an increase of IDO-specific T cells or PD-L1-specific T cells in the blood was observed in 28 (93%) and 26 (86%) patients, respectively. In total, 93% of patients had an increase in either PD-L1- or IDO-specific responses on vaccination (Fig. 3a), with a significant (P < 0.0001) median increase from baseline to the post-vaccine response demonstrated for both IDO and PD-L1 (at different time points on treatment), confirming that vaccine-specific immune responses were induced in patients regardless of clinical response (Fig. 3b,c). Immune responses fluctuated in the blood over time (Extended Data Fig. 3a). An increase in IDO- and PD-L1-specific responses in the peripheral blood was also detectable directly ex vivo across the different clinical response groups (Extended Data Fig. 4).Fig. 3 Vaccine-specific responses in blood. a, IDO- and PD-L1-specific T cell responses in PBMCs at baseline and on vaccination as measured by the IFN-γ ELISPOT assay (n = 30). *Responses were calculated as the difference between the average numbers of spots in wells stimulated with IDO or PD-L1 peptide (triplicates) and those from the corresponding control (DMSO), and statistical analyses of ELISPOT responses were performed using a distribution-free resampling method (Moodie et al.50). DR (double response), response was not statistically confirmed due to replicate number, but the number of spots in peptide wells was two times higher than that in control wells (DMSO). NS, no significant response and no DR. For a detailed overview of responses at serial time points on vaccination, see Extended Data Fig. 4a. b, IDO- and PD-L1-specific T cell response in PBMCs in all treated patients measured by the IFN-γ ELISPOT assay at baseline and on vaccination. On-vaccination responses were selected from the ‘best’ ELISPOT response at different time points for each patient (series 3, 6, 12, 18 or 24) during vaccination (n = 30). Wilcoxon matched-pairs signed-rank test was used to compare responses to IDO or PD-L1 peptides in the vaccine between baseline and later time points. c, Representative example of ELISPOT wells with response in patient MM23 in serial PBMCs before and on treatment. d, IDO-specific CD4+ and CD8+ T cells were isolated and expanded from PBMCs stimulated in vitro with the IDO peptide and a low dose of IL-2 for 14–15 d before sorting using the Miltenyi Cytokine Secretion Assay—Cell Enrichment and Detection kit. To assess their cytolytic potential, IDO-specific T cells were stimulated with IDO peptide, and expression of CD107a, IFN-γ and TNF-α was assessed by flow cytometry (the example is from patient MM14). Sustained vaccine-specific responses were observed 3 and 6 months after the last vaccine, indicating induction of memory responses in nine patients with response to clinical treatment who surpassed follow-up at data lock (Extended Data Fig. 3b). Importantly, PD-L1- and IDO-specific responses were observed irrespective of HLA genotype (Supplementary Table 5). To verify the functionality of vaccine-induced T cells, IDO- or PD-L1-specific T cells were isolated and in vitro expanded from PBMCs of five patients. Phenotypic characterization by flow cytometry revealed that the isolated vaccine-specific T cells consisted of both CD4+ and CD8+ T cells. Also, both IDO- and PD-L1-specific CD4+ and CD8+ T cells showed pro-inflammatory properties, as they expressed the cytolytic marker CD107a and secreted the cytokines IFN-γ and tumor necrosis factor (TNF)-α (Fig. 3d and Extended Data Fig. 6a–d). Interestingly, we were also able to detect vaccine-specific CD4+ and CD8+ T cell responses in peripheral blood ex vivo (Extended Data Figs. 4 and 5). We observed a significant increase in the overall percentage of CD107a, CD137 and TNF-α expression in response to peptide stimulation in on- or post-treatment PBMC samples compared to that at baseline, further confirming the expansion and the diverse signature of vaccine-specific T cells (Extended Data Fig. 5). Vaccine-specific responses detected in the skin at the vaccination site To investigate whether vaccine-specific T cells have the potential to migrate to peripheral tissue, delayed-type hypersensitivity (DTH) tests were performed after six cycles of treatment on 15 patients to assess the presence of vaccine-reactive T cells in the skin. Supplementary Table 6 display an overview of skin-infiltrating lymphocyte (SKIL) cultures. We detected IDO-specific T cells in the skin of six of ten patients and PD-L1-specific T cells in nine of 11 patients (Extended Data Fig. 7a). Intracellular cytokine staining was performed on SKILs from five patients after stimulation with either PD-L1 or IDO peptide. Here, we detected mainly CD4+ peptide-reactive T cells that secreted TNF-α and upregulated CD107a, and a minor fraction also secreted IFN-γ. In one patient, we detected CD8+ PD-L1-reactive T cells (Extended Data Fig. 7b–d). Vaccine-induced T cells specifically recognize target cells To confirm the functionality of vaccine-expanded T cells, vaccine-specific T cell clones (clonal purity was confirmed by T cell receptor (TCR) sequencing) were isolated and expanded from patient PBMCs (Extended Data Fig. 8e). We showed that PD-L1-specific T cells were able to recognize PD-L1+ autologous tumor cells in a PD-L1 expression-dependent manner if the cancer cells also expressed HLA-II (Fig. 4a,b). Similarly, an HLA-DR-restricted IDO-specific CD4+ T cell clone was able to recognize an HLA-DR-matched IDO-expressing model cell line, MonoMac1, in an IDO-expression-dependent manner (Fig. 4e–g). As previously described, IDO- and PD-L1-specific T cells’ mode of action is not limited to targeting only cancer cells. We were able to show that vaccine-specific T cell clones also reacted against PD-L1- and IDO-expressing autologous immune cells (Fig. 4c,h). To provide myeloid cells with a tumor-associated phenotype we treated isolated CD14+ myeloid cells with tumor conditioned medium (TCM) derived from an established autologous tumor cell line. We observed that such TCM-treated CD14+ cells had increased expression of PD-L1 and IDO and were effectively recognized by autologous PD-L1- and IDO-specific CD4+ T cell clones (Fig. 4c,d,h,i).Fig. 4 PD-L1- and IDO-specific T cells from vaccinated patients react against PD-L1- and IDO-expressing target cells. a, Left, PD-L1-specific T cell culture (MM1636.05) reactivity against PD-L1 peptide or autologous tumor cells in the IFN-γ ELISPOT assay. Tumor cells were either not treated or pretreated with 200 U ml−1 IFN-γ for 48 h before the assay. Effector:target (E:T) ratio of 10:1 was used. Right, PD-L1 and HLA-II surface expression on melanoma cells with (green) or without (yellow) pretreatment with IFN-γ compared to an isotype control (gray) as assessed by flow cytometry. b, Left, PD-L1-specific T cell (MM1636.05) reactivity in the IFN-γ ELISPOT assay against autologous tumor cells pretreated with IFN-γ (500 U ml−1) and transfected with mock or PD-L1 small interfering (si)RNA 24 h after transfection. E:T ratio, 10:1. Right, PD-L1 surface expression on melanoma tumor cells (MM1636.05) assessed by flow cytometry 24 h after transfection with mock (blue) or PD-L1 (red) siRNA compared to the isotype control (gray). c, Reactivity of the CD4+ PD-L1-specific T cell clone (MM1636.14) against PD-L1 peptide or autologous CD14+ cells; E:T ratio, 10:1. CD14+ cells were isolated using magnetic bead sorting and used as targets in an ELISPOT assay directly or after pretreatment for 2 d with TCM derived from the autologous tumor cell line. d, Quantitative PCR with reverse transcription (RT–qPCR) analysis of PD-L1 (CD274) expression in sorted CD14+ cells before and after treatment with autologous TCM for 48 h. e, Reactivity of the IDO-specific CD4+ T cell clone (MM1636.23) against IDO peptide combined with HLA-DR (L243)-, HLA-DQ (SPV-L3)- or HLA-DP (B7/21)-blocking antibodies (aHLA-DR, aHLA-DQ and aHLA-DP) in an intracellular staining assay (ICS) for IFN-γ and TNF-α production. T cells were incubated with individual blocking antibodies (2 μg ml−1) for 30 min before adding IDO peptide. f, Reactivity of the IDO-specific CD4+ T cell clone (MM1636.23) against the HLA-DR-matched IDO-expressing cell line MonoMac1 transfected with mock or IDO siRNA in an ICS assay, E.T ratio, 4:1. siRNA transfection was performed 48 h before the experiment. g, RT–qPCR analysis of IDO1 expression in MonoMac1 cells 48 h after siRNA transfection. h, Reactivity of the CD4+ IDO-specific T cell clone (MM1636.14) against IDO peptide or autologous CD14+ cells; E:T ratio, 20:1. CD14+ cells were isolated using magnetic bead sorting and used as targets in an ELISPOT assay directly or after pretreatment with TCM derived from the autologous tumor cell line. i, RT–qPCR analysis of IDO1 expression in sorted CD14+ cells before and after treatment with autologous TCM for 48 h. Bars in RT–qPCR data (d,g,i) represent the mean of three (d,i) or six (g) technical replicates ±s.d.; P values were determined by two-tailed parametric t-tests. ELISPOT counts (a,b,c,h) represent the mean value of three technical replicates ±s.e.m.; response P values were determined using the distribution-free resampling (DFR) method. TNTC, too numerous to count. T cell clones in blood and tumors To track the role of treatment-induced T cell responses, TCR sequencing of the complementarity-determining region 3 (CDR3) was performed on five patients in peripheral blood (baseline and cycles 3, 6 and 12) and paired biopsies. These five patients (MM01, MM02, MM08, MM09 and MM13) were selected due to the availability of material and to investigate a balanced patient group with both responders and non-responders. Due to a limited number of available paired biopsies (either as a consequence of patient refusal or unexpectedly rapid and substantial clinical responses), statistical analysis could not be applied. Details on clinical response are shown in Fig. 2a. Additionally, PBMCs (on treatment) or SKILs were stimulated with the IDO/PD-L1 peptides, and then cytokine-producing T cells were sorted to track vaccine-induced T cells both in the periphery and at the tumor site. To identify enriched IDO/PD-L1-specific T cell clones, TCR rearrangements in sorted IDO/PD-L1-specific clones and TCR rearrangements in sorted IDO/PD-L1-specific T cell samples were compared to sequences from baseline PBMC samples for each patient. Clonal expansion of vaccine-specific TCR rearrangements from samples on vaccination were then tracked using a differential abundance framework. Cumulative IDO/PD-L1-specific T cell frequencies were tracked in post-treatment samples. We found no relation between clinical response and the enrichment of vaccine-specific clones, but an increase in IDO/PD-L1-specific T cell clones was observed at different time points in the periphery in all five patients (Extended Data Fig. 3c). We next investigated overall changes in the T cell repertoire in the blood. A modest increase in the peripheral T cell fraction was observed in the three responding patients at cycle 3, while the two non-responding patients had a clear decrease in T cell fraction (Extended Data Fig. 8a). We thereafter investigated TCR clonality and TCR repertoire richness, exploring the proportion of abundant clones and the number of unique rearrangements, respectively. A decreasing peripheral Simpson clonality and increasing TCR repertoire richness was observed in responding patients at cycle 3, which might indicate tumor trafficking upon treatment. The opposite pattern was observed in non-responding patients (Extended Data Fig. 8b,c). Peripherally expanded clones were associated with tumors and persisted until cycle 12 (latest time point analyzed). The largest peripheral expansion was observed at cycle 3, with the most significant increase observed in patient MM01 (CR). Responding patients had a larger fraction of peripherally expanded clones that were also found in tumors compared to that of non-responders. By tracking peripherally expanded clones detected at the tumor site, we observed that patient MM01 had a substantial increase after treatment, indicating tumor trafficking of peripheral expanded clones (Extended Data Fig. 8d). Influx at the tumor site of vaccine-enriched T cell clones Given the observation of increased T cell fraction and enrichment of IDO- and PD-L1-specific clones in the blood after treatment, we investigated whether the same trend was observed at the tumor site. Both TCR sequencing and immunohistochemistry (IHC) of paired biopsies from the five patients described above showed an increase in the T cell fraction with an influx of CD3+ and CD8+ T cells after treatment in the three responding patients (Fig. 5a–c). IHC was not possible for patient MM09 due to tissue loss.Fig. 5 Changes in the TME after treatment. Number of CD3+ and CD8+ T cells, TCR fraction, TCR clonality, TCR repertoire, biopsy expanded TCR clones and enrichment of IDO- and PD-L1-specific T cells at the tumor site. a, Number of CD3+ and CD8+ T cells per mm2 (sum in the validated area) at the tumor site detected by IHC of paired biopsies from four patients. b, Example of IHC of CD3+ and CD8+ T cells at the tumor site before and after treatment (cells per mm2) in one patient (MM01). c, T cell fraction at the tumor site at baseline and cycle 6 by TCR sequencing. The T cell fraction was calculated by taking the total number of T cell templates and dividing by the total number of nucleated cells. d, Tracking of vaccine-associated clones at baseline and cycle 6 in tumor biopsies. Cumulative frequencies of IDO and PD-L1 vaccine-specific TCR rearrangements are represented. e,f, TCR clonality and TCR repertoire richness in five patients at the tumor site at baseline and cycle 6. Simpson clonality measures how evenly TCR sequences are distributed among a set of T cells, where 0 indicates an even distribution of frequencies and 1 indicates an asymmetric distribution in which a few clones dominate. TCR repertoire richness reports the mean number of unique rearrangements. g, Bar chart representing baseline expanded biopsy clones from five patients (colored bars) and the detection of biopsy expanded clones also found in the blood at baseline and series 3, 6 and 12 (white and gray bars) by TCR sequencing. We thereafter investigated whether some of the IDO/PD-L1 vaccine-linked T cells were present at the tumor site. Vaccine-associated clones were tracked as the combined frequency of IDO- and PD-L1-specific T cell rearrangements. In biopsies, the frequencies at cycle 6 were compared to those at baseline and showed an increase in vaccine-specific T cells in four of five patients, irrespective of clinical response (Fig. 5d). TCR sequencing of PD-L1-specific SKILs (a more specific culture than IDO/PD-L1-specific isolations derived from PBMCs on vaccination) and paired biopsies showed that two of the top five PD-L1-specific SKIL clones were present at the tumor site both before and after treatment (Extended Data Fig. 7e,f). With a focus on the more abundant T cell clones, we investigated overall TCR clonality at the tumor site before and after treatment. In addition, we explored the number of unique TCR rearrangements, dissecting the lower-frequency clones. Patient MM01 had a significant increase in TCR clonality and a decrease in repertoire richness at the tumor site after therapy, indicating a focused tumor repertoire response of selected clones. All three responding patients had a decrease in TCR repertoire richness, which again might indicate a focused tumor response (Fig. 5e,f). Deeper analyses showed that the T cell clones that expanded at the tumor site after therapy were also present in the blood at baseline and increased significantly after treatment in four of five patients. The highest proportion was detected early at cycle 3. These data again support trafficking of peripherally expanded clones to the tumor site and could indicate that the T cell response to treatment is derived from pre-existing peripherally tumor-associated T cells (Fig. 5g). Signs of treatment-induced inflammation in the TME To dissect changes in the TME induced by T cell influx upon treatment in responding patients, RNA gene expression analyses using the nCounter PanCancer Immune Profiling Panel from NanoString were performed on paired biopsies from two responding patients (MM01 and MM13). Expression of genes related to adaptive immunity such as T cell activation, effector functions (genes encoding IFN-γ, TNF-α, IL-15 and IL-18) and cytotoxicity was increased in post-treatment biopsies (Extended Data Fig. 9a,b). Also, expression of genes related to checkpoint inhibitors such as those encoding T cell immunoglobulin and mucin-domain containing-3 (TIM-3), IDO, PD-L1, PD-L2, PD-1 and CTLA-4 increased after treatment, indicating activation of immune cells in the TME (Extended Data Fig. 9c). Additionally, IHC of paired biopsies from four patients (MM01, MM02, MM05 and MM13) showed an upregulation of PD-L1, IDO, MHC-I and MHC-II on tumor cells, indicating a treatment-induced pro-inflammatory response in the three responding patients, except for a decrease in MHC-II expression in patient MM13. By contrast, the non-responding patient MM02 had a reduction in T cell numbers present in the tumor after treatment and no expression of PD-L1, IDO or MHC-II and, interestingly, total loss of MHC-I, demonstrating tumor immune escape (Extended Data Fig. 10a). CD8+ T cells and their distance (µm) to PD-L1-expressing cells in baseline biopsies from five patients was investigated by IHC. Except for patient MM13 (PR), distance and clinical responses were associated: the two responders had reduced distance (<20 µm) between cells expressing these markers compared to non-responding patients (>80 µm). This observation indicates that responding patients not only have a higher intratumoral infiltration of CD8+ T cells, but that these cells can surround and attack PD-L1-expressing immune cells and tumor cells (Extended Data Fig. 10b). Discussion In this clinical trial, MM1636, 30 patients with metastatic melanoma were treated with a first-in-class immunomodulatory IDO/PD-L1-targeting peptide vaccine combined with nivolumab. The treatment led to an unprecedented high ORR of 80%, with 43% of patients reaching a CR, and a striking mPFS of 26 months (95% CI, 15.4–69 months) was reached. The vaccine represents a new treatment strategy to activate specific T cells that target cells contributing to immune suppression (including tumor cells), positively modulating the TME by inducing local inflammation. Indeed, we show that vaccine-specific T cells isolated and expanded from vaccinated patients recognize not only tumor cells in a target- and HLA-restricted manner but also myeloid cells polarized toward a tumor-associated phenotype. Hence, myeloid cells become targets for vaccine-activated T cells when they have a tumor-associated phenotype in the TME. These phenomena may further induce checkpoint molecules and rewire the TME toward an increasingly aPD1-permissive state. A drawback of our study is the single-center nonrandomized setup. Comparison between trials or between patients in trials and real-world patients is problematic due to multiple factors, such as period of conduction and other therapies available at the different periods of time. Nevertheless, the rate of investigator-assessed ORR in the phase 3 trial CheckMate 067 was 43.7% in the nivolumab monotherapy group and 57% in the nivolumab and ipilimumab group. CRs occurred in 8.9% and 11.5% of patients, respectively18. The mPFS of 26 months (95% CI, 15.4–69 months) in this trial is more than twice as long as that for patients treated with nivolumab and ipilimumab in CheckMate 067, for which an mPFS of 11.5 months (95% CI, 8.7–19.3 months) was reached. Patient baseline characteristics were in general comparable to those of patients with MM who have been treated in CheckMate 067, although patients in MM1636 were older (mean age, 70 years) and a larger fraction were positive for PD-L1 (57%)3,18,22. Among patients with PD-L1-negative tumors in MM1636, an ORR of 61.5% was still reached, which would be expected to be around 33% for first-line nivolumab monotherapy23. Some studies suggest that older patients might have a tendency to respond better to aPD1 therapy; however, this is still debated24–27. To address potential trial bias and the nonrandomized setup, patients in MM1636 were matched for age, performance status, sex, M stage, LDH level, PD-L1 status and BRAF status with a historical control group from the DAMMED, who were treated contemporarily (2015–2019) with aPD1 monotherapy as standard of care17. We found a significantly higher ORR and CRR in MM1636 compared to those of matched patients, who had an ORR of 43% and a CRR of 13%, comparable to patients treated in CheckMate 067. Restrictions of the synthetic control group are of course that it is partially historic and patient selection outside matching criteria cannot be ruled out28. Numerous contemporary clinical trials are exploring the combination of aPD1 therapy with other immunomodulating agents for advanced melanoma. Talimogene laherparepvec, an oncolytic virus, is approved by the Food and Drug Administration and the European Medicines Agency to treat advanced melanoma. A small phase 1b trial with 21 patients (MASTERKEY-265) combined talimogene laherparepvec and pembrolizumab to treat patients with advanced unresectable melanoma and reached an ORR of 62% and a CR of 33%29,30. Seventy-one percent of the patients in this trial had an M stage below M1c; this number was 40% in our trial. Furthermore, mainly patients with an M stage below M1c responded to treatment, which was not the case in MM1636. Results from a large randomized phase 3 trial are awaited (KEYNOTE-034). Epacadostat, an IDO inhibitor, was tested in combination with pembrolizumab in a nonrandomized phase 2 trial in 40 aPD1 treatment-naive patients with MM with promising results, reaching an ORR of 62%. Unfortunately, the phase 3 trial showed no indication that epacadostat provided improvement in PFS and OS31. Limitations of the phase 3 trial were the sparse information on pharmacodynamics as well as biomarker evaluation to improve the design. The IDO/PD-L1 vaccine is different from epacadostat, as it is not an IDO inhibitor but targets IDO- and PD-L1-expressing cells. Similar vaccines administered as monotherapy induced objective responses in lung cancer and basal cell carcinoma, while epacadostat as monotherapy in 52 patients resulted in no responses32,33. Sahin et al. recently published encouraging data from a first-in-human trial, in which a vaccine containing liposomal RNA targeting four unmutated tumor-associated antigens (NY-ESO-1, MAGE-A3, tyrosinase and TPTE) was administered alone or in combination with nivolumab in patients with advanced melanoma. Responses were observed in both the monotherapy group as well as the combination group in checkpoint inhibitor-experienced patients, suggesting the efficacy of non-mutant shared tumor antigen vaccines34. The overall safety and tolerability findings are comparable to those of aPD1 monotherapy. Injection site reactions were exclusive to the vaccine. However, these side effects were transient and mild in most patients and most likely due to the adjuvant Montanide. IDO- and PD-L1-specific CD8+ and CD4+ T cells exist among peripheral blood lymphocytes in healthy donors35–38 and expand in response to pro-inflammatory stimuli35,39. Furthermore, both IDO and PD-L1 are induced in cells as a counter response to the inflammatory response. This provides a mechanism that ensures immune homeostasis, which keeps IDO/PD-L1-specific T cells in check; therefore, the risk of triggering autoimmune-related adverse events by vaccination appears to be minimal. This was confirmed in the first clinical trials of IDO and PD-L1 vaccination (NCT01543464 (refs. 33,40) and NCT03042793 (ref. 41)). The spontaneous (baseline) immune response to the vaccines observed in the current study is in agreement with our previous observation in various patients with cancer35–38. Numerous vaccine-induced changes in the blood and at the tumor site were observed. Peripheral IDO- and/or PD-L1-specific T cells were detected in vitro in a modified ELISPOT in over 93% of patients on vaccination, unrelated to patient HLA type. Immune responses were persistent in patients who surpassed follow-up at data cutoff and were still detectable up to 6 months after the last vaccine, suggesting induction of memory T cells. We did not observe a correlation between vaccine-induced responses in blood and clinical responses. However, the detection of a highly significant increase in vaccine-specific T cell numbers after vaccination in almost all patients together with the very low number of patients with PD makes it difficult to expect such a correlation, especially because other aspects should be taken into account (for example, the loss of class I expression on tumors cells in patient MM02 with PD after vaccination). Frequencies of peripheral T cells induced against PD-L1 were overall higher than those against IDO. Importantly, however, we also observed that it was not the same patients who responded strongly to both IDO and PD-L1, nor did patients react with a similar response pattern to the two antigens; that is, different time points of response were observed. This suggests that each component of the vaccine plays different roles in the ongoing immune response in patients. The goal of IDO/PD-L1 vaccination was to modulate the TME to increase responsiveness to aPD1 therapy, as we have observed in animal models, in which immune conversion is demonstrated in the TME with an increased influx of T cells. Targeting both IDO and PD-L1 together enables synergy, as the TME is known to use different immune escape mechanisms and IDO and PD-L1 are often overexpressed by different cellular compartments. Ex vivo ELISPOT and flow cytometry assays confirmed the induction of immune response toward both epitopes and the higher immunogenicity of the PD-L1 epitope, although because of lower sensitivity failed to detect some of the responses. TCR sequencing of five patients confirmed enrichment of IDO/PD-L1-specific T cell clones in the blood at different time points after treatment. Furthermore, an increase in enriched IDO/PD-L1-specific clones was observed in four of five patients at the tumor site after treatment, irrespective of clinical response. Phenotypic characterization showed that vaccine-specific T cells, which were expanded in vitro with interleukin (IL)-2 from the blood of vaccinated individuals, were both CD4+ T cells and CD8+ T cells. This was confirmed by ex vivo phenotype description of vaccine-activated T cells. Vaccine-specific T cells expressed CD107a and CD137 and produced IFN-γ and TNF-α upon stimulation with the cognate target, indicating their cytolytic capacity. Overall, the data from immune monitoring supports the importance of both antigens in the generation of the frequent clinical responses in the study. Despite a limited number of paired biopsies (due to either patient refusal or the fact that a large fraction of responding patients had no assessable tumors after six cycles), we observed trends indicating treatment-induced general T cell influx in responding patients. It was shown that proliferation of CD8+ T cells in the tumor after aPD1 treatment is associated with radiographic reduction in tumor size42. Additionally, we showed that a large proportion of expanded peripheral TCR clones were associated with tumors and the most considerable amount of clonal expansion was observed early, at cycle 3. For the patient with a CR included in the TCR sequencing analyses (MM01), the number of peripheral expanded clones present at the tumor site increased after treatment compared to that at baseline, indicating tumor trafficking of peripheral expanded clones. Gene expression analyses (two paired biopsies) and IHC (five paired biopsies) further demonstrated that the combination treatment induced a pro-inflammatory TME in responding patients with signs of T cell activation and cytotoxicity and increased cytokine activity. This may lead to further upregulation of IDO, PD-L1, MHC-I and MHC-II on tumor cells, leading to more treatment targets. It was shown that, following vaccination with a cancer vaccine, PD-L1 expression is increased on tumor cells due to recruitment of tumor-specific T cells and upregulation of adaptive immune resistance pathways in the TME43. Treatment with nivolumab monotherapy enhances PD-L1 expression, and it is therefore problematic to discriminate the effect of the vaccine as compared to that of nivolumab44. IDO- and PD-L1-specific pro-inflammatory effector T cells were hypothesized to counteract the functions of IDO- and PD-L1-expressing immune-suppressive cells as a means to keep the immune balance between immune activation and inhibition45. However, because the expression of these molecules is induced as a counter-regulatory response to inflammatory mediators such as IFNs, they can also be expressed by, for example, activated myeloid cells or T cells. Thus, activation of PD-L1-specific T cells may result in depletion of activated T cells in the tumor and other sites. Importantly, it was recently described that PD-L1+ T cells mainly have tolerogenic effects on tumor immunity and exert tumor-promoting properties, suggesting that targeting this immune population is indeed also beneficial46. We have previously investigated the effect of activating PD-L1-specific T cells in vitro and in vivo and found that, overall, they supported the effector phase of an immune response by removing PD-L1-expressing regulatory immune cells that inhibit PD-1+ effector T cells47,48. The major role of the PD-1 pathway is believed to be the regulation of effector T cell responses. Thus, this protective pathway is more important after activation, rather than at the initial T cell-activation stage49. Accordingly, the presence of PD-L1-specific T cells during the activation phase of an immune response may not increase or support a pro-inflammatory immune response due to the expression of PD-L1 on potent antigen-presenting cells or on the T cells themselves14. Thus, the overall effects of PD-L1-specific T cells may vary depending on the expression of both PD-1 and PD-L1, that is, due to the microenvironment and the state of the immune response. In conclusion, here we report an impressive response rate, CR rate and mPFS for a first-in-class immune-modulating vaccine combined with nivolumab. This may be a first step toward a new treatment strategy for patients with MM. Limitations are the low number of patients treated at a single institution and the lack of a randomized design with aPD1 monotherapy as comparator. Studies in aPD1 therapy-resistant or -refractory melanoma are ongoing as well as biomarker analysis for selecting patients at a higher likelihood to benefit from the combination versus aPD1 monotherapy. A larger randomized trial will be essential to validate these findings and determine the specific contribution of the vaccine to clinical responses and changes in the TME. In December 2020, the Food and Drug Administration granted breakthrough therapy designation for the IO102/IO103 vaccine combined with aPD1 therapy in metastatic melanoma based on data from the MM1636 trial. Methods Trial design and treatment plan MM1636 is an investigator-initiated, nonrandomized, open-label, single-center phase 1/2 study. All patients were treated at the Department of Oncology, Herlev and Gentofte Hospital, University of Copenhagen, Herlev, Denmark. The study was conducted according to the Declaration of Helsinki and Good Clinical Practice (GCP) and monitored by the GCP unit in Copenhagen, Denmark. The protocol was approved by the ethical committee of the Capital region of Denmark (H-17000988), the Danish Medical Agencies (2017011073) and the Capital Region of Denmark Data Unit (P-2019-172). The study was registered at https://clinicaltrials.gov/ under the identifier NCT03047928 and at the EudraCT (no. 2016-0004527-23). This study initially aimed to include 30 aPD1 treatment-naive patients with MM. An amendment with the addition of two other cohorts with ten patients in each cohort was made to evaluate immune responses and clinical efficacy in aPD1 therapy-resistant patients (cohort B, de novo resistance and cohort C, acquired resistance) for a total of 50 patients. The amendment with cohorts B and C was approved at the inclusion of 18 patients in cohort A. The trial is still including patients in cohorts B and C. This article reports results from cohort A. The first six patients were treated as for phase 1, evaluating for safety and tolerability before the remaining 24 patients were included in phase 2. The IDO/PD-L1 vaccine was administered subcutaneously biweekly for the first 10 weeks and thereafter every 4 weeks for approximately 9 months. A maximum of 15 vaccines were administered. Nivolumab was administered according to the approved label (3 mg per kg biweekly) for 24 cycles. The 15th vaccine was administered together with the 24th nivolumab cycle of 3 mg per kg, and responding patients thereafter continued nivolumab monotherapy every 4th week (6 mg per kg) as a standard of care after investigator assessment. Treatment with nivolumab was discontinued at the maximum benefit (assessed by the investigator), after a maximum of 2 years of therapy, at progression or due to severe adverse events (Extended Data Fig. 1b, treatment plan). Vaccine composition Each vaccine was composed of 100 µg IO102, a 21-amino-acid peptide (DTLLKALLEIASCLEKALQVF) from IDO, and 100 µg IO103, a 19-amino-acid peptide (FMTYWHLLNAFTVTVPKDL) from the signal peptide of PD-L1 (PolyPeptide Laboratories). Peptides were dissolved separately in 50 µl DMSO, filtered for sterility and frozen at −20 °C (NUNC CyroTubes CryoLine System Internal Thread, Sigma-Aldrich). At <24 h before administration, peptides were thawed. The PD-L1 peptide was diluted in 400 µl sterile water and, immediately before injection, mixed with the IDO peptide solution and 500 µl Montanide ISA-51 (SEPPIC) to achieve a total volume of 1 ml. Patients Patients above 18 years of age with locally advanced or stage 4 melanoma according to the AJCC (seventh edition), at least one measurable lesion according to RECIST 1.1 and an ECOG PS of 0–1 were eligible. The main exclusion criteria were prior treatment with aPD1 therapy, CNS metastases >1 cm, severe comorbidities and active autoimmune disease. Enrollment was not restricted to PD-L1 status, but it was known before inclusion. Patients were included after informed consent. Patients MM42 and MM20 have confirmed their approval of PET–CT images and clinical images that have been included in the article. Key study assessments Safety and tolerability were evaluated based on changes in clinical laboratory analyses and reported adverse events. Adverse events were assessed according to Common Terminology Criteria for Adverse Events (version 5.0) and were graded from 1 to 5 for all treated patients up to 6 months after the last dose of the IDO/PD-L1 vaccine. Clinical efficacy was assessed using FDG PET–CT scans before treatment and every 3rd month until progression. Objective responses were categorized into CR, PR, stable disease or PD according to RECIST version 1.1. Clinical data were collected at CCIT-DK in the eCRF program OpenClinica version 1.0 and in Microsoft Excel version 2002 on a secure server. Blood samples for immunologic analyses were collected before treatment, before the third cycle, after the sixth, 12th, 18th and 24th cycles (on vaccination) and 3 and 6 months after the last vaccine. Two to three tumor needle biopsies (1.2 mm) were collected at baseline and after six cycles from the same tumor site, when accessible, to evaluate immune responses at the tumor site. DTH skin tests were performed and punch biopsies were taken from the DTH area after cycle 6 for the evaluation of SKILs reactive to PD-L1 and IDO (Supplementary Fig. 1, treatment plan). Statistical analysis of clinical outcome Survival curves were computed in GraphPad Prism software version 9.0.0 according to Kaplan–Meier method. Median follow-up time of enrollment was calculated using the reverse Kaplan–Meier method, also in GraphPad Prism software version 9.0.0. For binary outcomes, 95% two-sided CIs were constructed using the Clopper–Pearson method, also in GraphPad version 9.0.0. An independent board of certified and experienced oncoradiologists performed an external review to evaluate clinical response to address the potential bias of investigator site review. The external review took place at Rigshospitalet, Copenhagen University Hospital. This hospital did not participate in the MM1636 trial, and the external reviewer had no prior knowledge about the clinical trial or the trial therapy. Only PET–CT images were accessed, and arrows indicating target or non-target lesions appeared on baseline images as the only additional information. To address potential trial bias regarding treatment effect, we matched patients in MM1636 with patients from the DAMMED, a population-based database that retrospectively collects data on patients with metastatic melanoma in Denmark. Here, data from 938 patients treated with aPD1 monotherapy contemporaneously (January 2015–October 2019) were extracted. Two hundred and eighteen of these patients were eligible for comparison and matching (all parameters available) (Supplementary Table 1), and 74 patients from the DAMMED were found to match. Patients were matched for age (≤70 years, >70 years), sex, LDH levels (normal, elevated), M stage (M1a, M1b, M1c), BRAF status (wild type, mutated) and PD-L1 status (<1%, ≥1%). An exact matching algorithm was used in which patients in MM1636 were matched with patients from the DAMMED with the same combination of variables. Twenty-nine patients from MM1636 were matched with exact combinations of the six variables. One patient could not be matched. To secure balance of the calculations, control patients were weighted according to the number of patients for each MM1636 patient. Estimates for treatment effects were calculated by weighted logistic regression analyses and weighted Cox proportional hazard model. The R package ‘Matchlt’ was used for matching patients. As the method chosen for matching control patients to protocol patients, a weighted binary logistic regression model was used for comparing response rates in the two matched cohorts. Odds ratios and response rates, including their corresponding 95% CIs, were extracted from the regression models. All P values were two sided, and P values below 0.05 were considered statistically significant. SAS version 9.4M5 was used for the weighted logistic regression models. Processing of peripheral blood mononuclear cells Peripheral blood was collected from all patients in heparinized tubes and was processed within 4 h. In brief, PBMCs were isolated using Lymphoprep (Medinor) separation. PBMCs were counted on a Sysmex XP-300 analyzer and frozen in Human AB Serum (Sigma-Aldrich, H4522-100ML) with 10% DMSO using controlled-rate freezing (CoolCell, BioCision) in a −80 °C freezer and were moved the next day to a freezer at −140 °C until further processing. Needle biopsies at baseline and after the sixth vaccine Two to three needle biopsies (1.2 mm) were taken at baseline and after the sixth cycle of treatment when assessable from the same tumor lesion. One fragment was fixed with formalin and embedded in paraffin (FFPE); one to two fragments were used for expanding tumor-infiltrating T cells and establishing autologous tumor cell lines. Delayed-type hypersensitivity and generation of skin-infiltrating lymphocytes After six cycles of treatment, we performed intradermal injections of vaccine components without adjuvant and one control injection containing DMSO without peptide. Patients were injected with either a mixture of both IDO and PD-L1 peptides at all three injection sites or PD-L1 peptide, IDO peptide or a mixture of the two at the injection sites, respectively (Supplementary Table 6). Eight hours after injection, punch biopsies were resected from the three sites where the peptide was injected and transported immediately to the laboratory and cut into fragments of 1–2 mm3. SKILs were expanded to establish ‘young SKILs’ in CM medium consisting of RPMI 1640 with GlutaMAX, 25 mM HEPES, pH 7.2 (Gibco, 72400-021), IL-2 (100 or 6,000 IU ml−1) (Proleukin Novartis, 004184), 10% heat-inactivated Human AB Serum (Sigma-Aldrich, H4522-100ML), 100 U ml−1 penicillin, 1.25 µg ml−1 Fungizone (Bristol Myers Squibb, 49182) and 100 µg ml−1 streptomycin (Gibco, 15140-122). Half of the medium was replaced three times per week. Young SKILs from samples derived from IDO peptide, PD-L1 peptide and the mixture injection sites were further expanded in a small-scale version of the 14-d rapid expansion protocol as previously described51. Quantification of specific T cells in blood by ELISPOT For enumeration of vaccine-specific T cells in the peripheral blood, PBMCs from patients were stimulated with IDO or PD-L1 peptide in the presence of a low dose of IL-2 (120 U ml−1) for 7–13 d before being used in IFN-γ ELISPOT assays with 2.5–3.2 × 105 cells per well. Briefly, cells were placed in a 96-well PVDF membrane-bottomed ELISPOT plate (MultiScreen MSIPN4W50, Millipore) precoated with IFN-γ-capture antibody (clone 1-D1K, Mabtech). Diluted IDO or PD-L1 peptide stocks in DMSO were added at 5 µM; an equivalent amount of DMSO was added to control wells. PBMCs from each patient were set up in duplicate or triplicate for peptide and control stimulations. Cells were incubated in ELISPOT plates in the presence of the peptide for 16–18 h, after which they were washed off, and biotinylated secondary antibody (anti-human IFN-γ mAB, clone 7-B6-1, Mabtech) was added. After 2 h of incubation, unbound secondary antibody was washed off, and streptavidin-conjugated alkaline phosphatase (Mabtech) was added for 1 h. Next, unbound streptavidin-conjugated enzyme was washed off, and the assay was developed by adding BCIP/NBT substrate (Mabtech). ELISPOT plates were analyzed on the CTL ImmunoSpot S6 Ultimate V analyzer using ImmunoSpot software version 5.1. Responses were calculated as the difference between the average numbers of spots in wells stimulated with IDO or PD-L1 peptide and those from corresponding control wells. For detection of IDO and PD-L1 peptide responses ex vivo, PBMCs were thawed and rested overnight in medium containing DNase I (1 μg ml−1, Sigma-Aldrich, 11284932001) before being used in IFN-γ ELISPOT assays as described above. A total of 6–9 × 105 PBMCs per well were seeded. Statistical analysis of all ELISPOT responses was performed using the DFR method as described by Moodie et al. using RStudio software (RStudio Team, 2016, http://www.rstudio.com/)50. Vaccine-specific ELISPOT responses were defined as true if the difference between the spot count in control and peptide-stimulated wells was statistically significant according to DFR statistical analysis or, for samples performed in duplicate, if the spot count in peptide-stimulated wells was at least 2× the spot count in control wells50. Cancer cell lines and tumor conditioned medium Autologous melanoma cell lines were established from needle biopsies. Briefly, biopsies were chopped into small fragments and seeded in 24-well cultures in RPMI 1640 with GlutaMAX, 25 mM HEPES, pH 7.2 (Gibco, 72400-021), 10% heat-inactivated FBS (Life Technologies, 10500064), 100 U ml−1 penicillin, 1.25 µg ml−1 Fungizone (Bristol Myers Squibb, 49182) and 100 µg ml−1 streptomycin (Gibco, 15140-122). Established adherent melanoma tumor cell lines were cryopreserved at −140 °C in freezing medium containing FBS with 10% DMSO. PD-L1 and HLA-II expression on established tumor cell lines was assessed by flow cytometry staining with anti-PD-L1–PE-Cy7 (1:23 dilution, clone M1h1 (RUO), BD, 558017) and anti-HLA-II–FITC (1:23 dilution, clone Tu39 (RUO), BD, 555558) antibodies. To obtain TCM, established tumor cell lines were cultured in 175-cm2 Nunc cell culture flasks until 80–90% confluency was reached. The culture medium was then replaced with 20 ml fresh X-VIVO 15 with Gentamicin and Phenol Red (Lonza, BE02-060Q), medium with 5% heat-inactivated Human AB Serum (Sigma-Aldrich, H4522-100ML). After 24 h of incubation, TCM was collected and centrifuged to remove any resuspended cells, after which TCM was aliquoted, frozen and stored at −80 °C. The acute monocytic leukemia cell line MonoMac1 was obtained from the DSMZ—German Collection of Microorganisms and Cell Cultures (ACC 252) and cultured in RPMI 1640 with GlutaMAX, 25 mM HEPES, pH 7.2 (Gibco, 72400-021) and 10% heat-inactivated FBS. Isolation of autologous myeloid cells Autologous CD14+ cells were sorted from freshly thawed PBMCs using a magnetic bead-separation kit (Miltenyi Biotec, 130-050-201) according to the manufacturer’s instructions. Isolated CD14+ cells were used as targets in the IFN-γ ELISPOT assay directly after sorting or differentiated in vitro into tumor-associated macrophages by culturing with 1 ml fresh X-VIVO 15 medium with Gentamicin and Phenol Red (Lonza, BE02-060Q) and 5% heat-inactivated Human AB Serum, supplemented with 1 ml autologous TCM in 24-well plates for 2 d. Quantification of vaccine-specific T cells from DTH biopsy sites by ELISPOT SKILs expanded in vitro were rested in medium without IL-2 overnight before being used in the IFN-γ ELISPOT assay as described above to evaluate reactivity of skin-infiltrating T cells. Generation of IDO- and PD-L1-specific T cell cultures from PBMCs or SKILs IDO- or PD-L1-specific T cells were isolated from peptide-stimulated in vitro PBMC cultures on days 14–15 after stimulation or from SKIL cultures expanded in vitro. For specific T cell isolation, PBMCs or SKILs were stimulated with IDO or PD-L1 peptide, and cytokine-producing T cells were sorted using the IFN-γ or TNF-α Secretion Assay—Cell Enrichment and Detection kit (Miltenyi Biotec). Cytokine-production profile of PD-L1- and IDO-specific T cells by intracellular staining To assess the T cell cytokine-production profile, isolated and expanded IDO- and PD-L1-specific T cell cultures were stimulated for 5 h with peptide at 5 μM in a 96-well plate. One hour after the start of the incubation, anti-CD107a–PE (1:133 dilution, clone H4A3, BD Biosciences, 555801) antibody and BD GolgiPlug (1:1,000 dilution, BD Biosciences) were added. After a 5-h incubation, cells were stained using fluorescently labeled surface marker antibodies: anti-CD4–PerCP (1:5 dilution, clone SK3, 345770), anti-CD8–FITC (1:5 dilution, clone SK1, 345772) and anti-CD3–APC-H7 (1:20 dilution, clone SK7, 560275) (all from BD Biosciences). Dead cells were stained using FVS510 (1:250 dilution, BD Biosciences, 564406), followed by overnight fixation and permeabilization using eBioscience Fixation/Permeabilization buffers (00-5123-43, 00-5223-56) according to the manufacturer’s instructions. Cells were then stained intracellularly in eBioscience permeabilization buffer (00-8333-56) with anti-IFN-γ–APC (1:20 dilution, clone 25723.11, BD Biosciences, 341117) and anti-TNF-α–BV421 (1:100 dilution, clone Mab11, BD Biosciences, 562783) antibodies. Samples were analyzed on the FACSCanto II (BD Biosciences) using BD FACSDiva software version 8.0.2. The gating strategy is shown in Supplementary Fig. 3a. For assessing the reactivity against IDO expressing MonoMac1 cells, the appropriate amount of cancer cells was added to the IDO-specific T cells to obtain an effector:target ratio of 4:1 in the 96-well plate and T cell cytokine production was tested after 5 h of co-culture as described above. For HLA-blocking experiments, HLA-blocking antibodies (2 μg ml−1) were added directly into wells 30 min before the addition of peptide. Blocking antibodies used were HLA-DR (1:500 dilution, clone L243, Abcam, ab136320), HLA-DQ (1:500 dilution, clone SPV-L3, Abcam, ab23632) and HLA-DP (1:500 dilution, clone B7/21, Abcam, ab20897). To assess ex vivo T cell reactivity to IDO and PD-L1 peptides in patient PBMCs, cells were thawed and rested for 1–2 d in medium containing DNase I (1 μg ml−1, Sigma-Aldrich, 11284932001). PBMCs were then stimulated with peptide at 5 μM in a 96-well plate for 8 h. An hour after the addition of peptide, anti-CD107a–BV421 (3:50 dilution, clone H4A3, BD Biosciences, 328626) antibody and BD GolgiPlug (1:1,000 dilution, BD Biosciences) were added. Surface and intracellular staining was performed as described above. Antibodies used for surface staining were anti-CD3–PE-CF594 (0.8:30 dilution, clone UCHT1, BD Biosciences, 562280), anti-CD4–BV711 (1:30 dilution, clone SK3, BD Biosciences, 563028) and anti-CD8–Qdot605 (1:150 dilution, clone 3B5, Thermo Fisher, Q10009). Antibodies used for intracellular staining were anti-CD137–PE (1:20 dilution, clone 4B4-1 (RUO), BD Biosciences, 555956), anti-IFN-γ–PE-Cy7 (1.5:20 dilution, clone B27, BD Biosciences, 557643) and anti-TNF-α–APC (1:20 dilution, clone Mab11, BD Biosciences, 554514). Samples were acquired on the NovoCyte Quanteon (ACEA Biosciences) and analyzed using NovoExpress software version 1.4.1. To assess vaccine-specific T cell responses, background values observed in unstimulated PBMC samples were subtracted from values observed in peptide-stimulated conditions. Positive response value threshold was set at a difference of 0.2% from the background values. Based on this response cutoff, only TNF-α, CD107a and CD137 responses were detected in this assay. Statistical analysis comparing baseline with on-treatment or post-treatment cytokine profiles was performed using two-sided Wilcoxon matched-pairs signed-rank test. The gating strategy is shown in Supplementary Fig. 3b. siRNA-mediated PD-L1 and IDO1 silencing A stealth siRNA duplex for targeted silencing of PD-L1 (Invitrogen)52, a custom silencer select siRNA for targeted silencing of IDO1 (Ambion) and the recommended silencer select negative control (Ambion) siRNA for mock transfection were used. The stealth PD-L1 siRNA duplex consisted of the sense sequence 5′-CCUACUGGCAUU-UGCUGAACGCAUU-3′ and the antisense sequence 5′-AAUGCGUUCAGCAAAUGCCAGUAGG-3′. The three silencer IDO siRNA duplexes used were siRNA1 (sense sequence, 5′-ACAUCUGCCUGAUCUCAUATT-3′; antisense sequence, 5′-UAUGAGAUCAGGCAGAUGUTT-3′), siRNA2 (sense sequence, 5′-CCACGAUCAUGUGAACCCATT-3′; antisense sequence, 5′-UGGGUUCACAUGAUCG-UGGAT-3′) and siRNA3 (sense sequence, 5′-CGAUCAUGUGAACCCAAAATT-3′; antisense sequence, 5′-UUUUGGGUUCACAUGAUCGTG-3′). For PD-L1 or IDO1 silencing, cancer cells were electroporated with 0.025 nmol of each siRNA duplex as previously described53. For PD-L1-silencing experiments, cancer cells were treated with IFN-γ (500 U ml−1, PeproTech) 1 h after electroporation. Electroporated cells were used as target cells in ELISPOT and ICS assays 24 h or 48 h after siRNA electroporation. RT–qPCR Total RNA was extracted using the RNeasy Plus Mini kit (Qiagen, 74134) following the manufacturer’s instructions. RNA concentration was quantified using a NanoDrop 2000 (Thermo Fisher Scientific), and a total of 1,000 ng RNA was reverse transcribed with the High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems, 4368814) using 1,000 ng input RNA for transcription. Real-time qPCR analyses were performed using the TaqMan Gene Expression Assay on a Roche LightCycler 480 instrument. RT–qPCR assays were performed with a minimum of three technical replicates and analyzed using the ΔCt method as described in Bookout et al.54 with normalization of IDO1 expression (primer ID, Hs00984148_m1) or PD-L1 expression (primer ID, Hs001125296_m1) to the expression level of the housekeeping gene POLR2A (primer ID, Hs00172187_m1) and the control sample (mock). For low-concentration samples with no amplification, Ct was set to 40. Controls without reverse transcriptase were used as controls for specific amplifications. P values were determined using two-tailed parametric t-tests. BRAF-mutational status and PD-L1 status at baseline in all patients A library of historical FFPE biopsies were assessable for all patients and analyzed locally at Herlev and Gentofte University Hospital by experienced pathologists for BRAF status and PD-L1 expression on tumor cells. The BRAF analysis was carried out with real-time PCR using the EntroGen BRAF Mutation Analysis Kit II (BRAFX-RT64, CE-IVD) to specifically detect mutations corresponding to V600D, V600E and V600K in the BRAF protein. PD-L1 status was assessed using the monoclonal rabbit anti-PD-L1 antibody (clone 28.8, PD-L1 IHC 28-8 pharmDx) in FFPE biopsies following the manufacturer’s instructions. Patients were considered positive for PD-L1 with expression levels ≥1% and negative for PD-L1 with expression levels <1%. Human leukocyte antigen type Blood samples from all 30 patients were genotyped for class I (HLA-A, HLA-B, HLA-C) and three class II types (HLA-DRB1, HLA-DQA1, HLA-DQB1) using LinkSēq HLA Typing Kits (Thermo Fisher, 1580C). These test kits are based on real-time PCR using allele-specific exponential amplification (sequence-specific primers) followed by melting curve analyses. Immunohistochemistry simplex: Immunoscore CR IHC staining was performed using a qualified Ventana Benchmark XT with four different steps: (1) antigen retrieval; (2) staining with the following primary antibodies: anti-CD3 antibody (clone HDX2; provider, HalioDx; HD-FG-000013; 10931065/10636667; concentration, 0.25 µg ml−1), anti-CD8 antibody (clone HDX3; provider, HalioDx; HD-FG-000019; 10931069/10337710/10639301; concentration, 0.5 µg ml−1), anti-IDO monoclonal antibody (clone VINC3IDO; provider, Thermo Fisher Scientific; 14-9750-82; E25003-101; concentration, 0.05 µg ml−1), anti-HLA class 1 ABC antibody (clone, EMR8-5; provider, Abcam; ab70328; GR3248333/GR3186494; concentration, 0.5 µg ml−1), anti-HLA-DR/DP/DQ/DX antibody (clone CR3/43; provider, Santa Cruz; sc-53302; L1714; concentration, 1 µg ml−1) and anti-PD-L1 antibody (clone HDX3; provider, HalioDx; HD-FG-000035; 106312810/106312816; concentration, 3.3 µg ml−1); (3) detection with a secondary antibody using the ultraView Universal DAB Detection kit; and (4) counterstaining using Hematoxylin & Bluing Reagent (staining of cellular nuclei). Control slides were systematically included in each staining run to permit quality control of the obtained measurements. Following coverslipping, slides were scanned with the NanoZoomer-XR to generate digital images (20×). Two consecutive slides were specifically used to perform Immunoscore CR TL staining of CD3+ and CD8+ cells. Digital pathology of T lymphocytes The digital pathology for Immunoscore CR TL allowed the quantification of positive cells (in cells per mm2) into the core tumor and invasive margin if present. Each sample was analyzed using the HalioDx Digital Pathology Platform. Digital pathology and pathologist analysis of IDO Digital pathology for IDO allowed the quantification of stained area (in mm2) into the whole tumor. Each sample was analyzed using the HalioDx Digital Pathology Platform. Analysis of IDO+ cells was performed by a pathologist. Results were expressed as an H score from 0 to 300. The score is obtained by the formula 3 × percentage of cells with strong staining + 2 × percentage of cells with moderate staining + percentage of cells with weak staining. Digital pathologist analysis of MHC-I and MHC-II Analysis of MHC-I+ and MHC-II+ cells was performed by a pathologist. Results are expressed as a percentage of positive tumor cells and a percentage of positive cells in the stroma. Digital pathology Immunoscore immune checkpoint (CD8+ and PD-L1+) The Immunoscore CR IC test allowed the quantification of CD8++ cell density in the whole tumor and a CD8+-centered proximity index (which corresponds to the percentage of CD8++ cells that have at least one PD-L1+ cell in the neighborhood) at different cutoff distances (20 µm, 40 µm, 60 µm and 80 µm). Pathologist analysis of PD-L1 A pathologist performed analysis of PD-L1+ cells. Positivity of a viable tumor cell was considered when partial or complete cell membrane staining was observed (more than 10% of the tumor cell membrane). Results were expressed as a percentage. NanoString RNA profiling and Immunosign RNA was extracted from FFPE tissues using Qiagen RNeasy FFPE extraction kits (Qiagen). Annotations from the pathologist performing H&E staining were used to guide removal of normal tissue from the slides by macrodissection before nucleic acid extraction, which occurred after tissue deparaffinization and lysis. Each extracted RNA sample was independently quantified using a NanoDrop spectrophotometer (NanoDrop Technologies) and qualified (Agilent Bioanalyzer). Degradation was quantified as the percentage of RNA fragments smaller than 300 bp using the RNA 6000 Nano kit (Agilent Bioanalyzer). Good sample quality was defined as less than 50% of RNA fragments being 50–300 bp in size. RNA expression profiling was performed using the nCounter PanCancer Immune Profiling Panel from NanoString (NanoString Technologies). The PanCancer Immune Profiling Panel contains 776 probes and is supplemented with six genes to complete HalioDx Immunosign targets. Hybridization was performed according to the manufacturer’s instructions. Hybridized probes were then purified and immobilized on a streptavidin-coated cartridge using the nCounter Prep Station (NanoString Technologies). Data collection was carried out on the nCounter Digital Analyzer (NanoString Technologies) following the manufacturer’s instructions to count individual fluorescent barcodes and quantify target RNA molecules present in each sample. For each assay, a scan of 490 fields of view was performed. Raw data from the NanoString nCounter were processed using NanoString recommendations. Quality control enables to retain data of good quality with a binding density that ranges between 0.05 and 2.25. The linearity of positive controls was checked using the R2 value of regression between the counts and the concentrations of positive controls. Samples that showed R2 < 0.75 were flagged and removed from the analysis. The background was removed using a thresholding method at the mean + 2 s.d. of negative controls. Raw counts were normalized using a positive normalization factor. Samples showing positive normalization factors outside the range of 0.3–3 were removed from the analysis. A second normalization was performed using the housekeeping gene normalization factor. Only the most stable housekeeping genes were selected for this normalization step using the variance-versus-mean relationship. All samples showing a normalization factor outside the range of 0.1–10 were removed from the analysis. All statistical analyses were performed on normalized counts using R software (version 2.6.2, 2019-12-12). T cell receptor variable β chain sequencing To track longitudinal immune responses to therapy, genomic DNA was extracted from longitudinal pre- and post-treatment PBMCs (five patients), pre- and post-treatment biopsies (five patients) (both FFPE and treated with RNAlater) and IDO- and PD-L1-specific T cell cultures from PBMCs (five patients) or SKILs (one patient). Three clonal (two IDO- and one PD-L1-) specific cultures were also sequenced to confirm clonal purity of cultures. DNA from PBMCs or RNAlater-treated biopsies was extracted with the DNeasy Blood and Tissue kit (Qiagen, 69504), DNA from sorted IDO and PD-L1-specific T cells from either PBMCs or SKILs was extracted using the QIAamp DNA Micro kit (Qiagen, 565304), and DNA from PFFE biopsies was extracted using the Maxwell RSC DNA FFPE kit (Promega, AS1450). Immunosequencing of the CDR3 regions of human TCRβ chains was performed using the immunoSEQÒ Assay (Adaptive Biotechnologies). Extracted genomic DNA was amplified in a bias-controlled multiplex PCR, followed by high-throughput sequencing. Sequences were collapsed and filtered to identify and quantitate the absolute abundance of each unique TCRβ CDR3 region for further analysis as previously described55–57. Statistical analyses of TCRβ sequencing results Two quantitative components of diversity were compared across samples in this study. First, Simpson clonality was calculated on productive rearrangements by ∑i=1Rpi2, where R is the total number of rearrangements and pi is the productive frequency of rearrangement i. Values of Simpson clonality range from 0 to 1 and measure how evenly receptor sequences (rearrangements) are distributed among a set of T cells. Clonality values approaching 0 indicate a very even distribution of frequencies, whereas values approaching 1 indicate an increasingly asymmetric distribution in which a few clones are present at high frequencies. Second, sample richness was calculated as the number of unique productive rearrangements in a sample after computationally downsampling to a common number of T cells to control for variation in sample depth or T cell fraction. Repertoires were randomly sampled without replacement five times, and we report the mean number of unique rearrangements. The T cell fraction was calculated by taking the total number of T cell templates and dividing by the total number of nucleated cells. The total number of nucleated cells was derived from reference genes using the immunoSEQ Analyzer version 3. To identify enriched vaccine clones in each patient, rearrangement frequencies in their baseline PBMCs and each IDO/PD-L1-sorted T cell sample were compared using a binomial distribution framework as previously described58. In brief, for each clone, we performed a two-sided test to determine whether frequencies were the same in the patient’s periphery and a PD-L1- or IDO-specific T cell sample. The Benjamini–Hochberg procedure was used to control the false discovery rate at 0.01 (ref. 59). Clonal expansion in post-treatment samples was similarly assessed using this differential abundance framework, but an IDO/PD-L1-specific T cell sample was replaced with a post-treatment series sample. In biopsies, the six series frequencies were compared to those of baseline tissue. Lastly, vaccine-associated clones were tracked in each PBMC and tissue sample by summing the frequency of each rearrangement enriched in either PD-L1- or IDO-specific T cells. All statistical analyses were performed in R version 3.6.1. Reporting Summary Further information on research design is available in the Nature Research Reporting Summary linked to this article. Online content Any methods, additional references, Nature Research reporting summaries, source data, extended data, supplementary information, acknowledgements, peer review information; details of author contributions and competing interests; and statements of data and code availability are available at 10.1038/s41591-021-01544-x. Supplementary information Supplementary Information Supplementary Figs. 1–3 and Tables 1–6 Reporting Summary Extended data Extended Data Fig. 1 Mode of Action and treatment plan. a) Anticipated mechanism of action of the combination therapy of an IDO/PD-L1 derived peptide vaccine and nivolumab (anti-PD1). 1) The IDO/PD-L1 peptide vaccine is administered subcutaneously (s.c) and nivolumab is administered intravenous (IV). 2) The vaccines peptides are phagocytosed by an antigen presenting cell and presented to IDO and PD-L1 specific T cells, which are activated. 3) The activated T cells migrates to the tumor site where they attack both immune-suppressive cells and tumour cells expressing IDO and/or PD-L1 leading to cytokine production and a pro-inflammatory tumour microenvironment. 4/5) Enhanced tumour killing by both IDO/PD-L1 specific T cells and tumor specific cytotoxic T cells due to PD-1 blockade. Created with BioRender.com b) Treatment plan. After written informed content patients were screened. Before treatment start a baseline PET/CT scan was performed, baseline blood sample for research use and if assessable a baseline needle biopsy. Patients were treated with the IDO/PD-L1 peptide vaccine subcutaneously biweekly for the first 6 injections and thereafter every fourth week for a maximum of 15 vaccines. Nivolumab was administered in parallel biweekly (3mg/kg) up to 24 series. If patients needed subsequent nivolumab after ended vaccination regiment they were treated with 6 mg/kg every fourth week up to two years. Needle biopsy and delayed type hypersensitivity (DTH) was performed after 6 series of treatment if assessable. PET/CT scans was performed every third month. Extended Data Fig. 2 Progression free survival and overall survival in matched historical control group and vaccine injection site reaction. a) Kaplan-Meier curve of progression free survival with corresponding 95% confidence intervals in the matched historical control group (n=74). Patients were matched on BRAF-status, PD-L1-status, age, gender, M-stage and LDH level). b) Kaplan-Meier curve of overall survival with corresponding 95% confidence intervals in the matched historical control group (n=74). c) Images of injection site reaction in patient MM20 (CR) after 11 vaccination show redness, rash and granuloma at injection site. Extended Data Fig. 3 Vaccine specific responses in blood. a) Heatmap of specific (background has been subtracted) IDO and PD-L1 responses in PBMCs at baseline, series 3, 6, 12, 18 and 24 measured by in vitro IFN-γ Elispot assay show fluctuations in the blood during treatment (n=30). b) Heatmap of specific (background has been subtracted) IDO and PD-L1 responses in PBMCs at baseline and 3 and 6 months after last vaccine measured by in vitro IFN- γ Elispot assay (n=30). 2.5–3.2x105 cells per well were used. c) Vaccine associated clones were tracked in the blood by summing the frequency of each rearrangement enriched in either IDO or PD-L1 T cells. Extended Data Fig. 4 Ex vivo vaccine specific responses in blood. Heatmaps of detected specific (background has been subtracted) IDO (a) and PD-L1 (b) responses in PBMCs at baseline and on/after treatment as measured by IFNγ ELISPOT (n=25). C) Example of well images of ex vivo ELISPOT wells for three different patients (n=3). 6-9x105 cells per well were used. Extended Data Fig. 5 CD4 and CD8 vaccine specific T cell responses in blood. Top: heatmaps of IDO specific CD4 (left) and CD8 (right) T cell responses in PBMCs at baseline and on/post treatment. Bottom: heatmaps of PD-L1 specific CD4 (left) and CD8 (right) T cell responses. Responses quantified by flow cytometry by an increased expression of CD107α, CD137 and TNFα after 8h peptide stimulation. Values represent specific responses after background values have been subtracted (n=21). Statistical analysis were performed using two-sided Wilcoxon matched-paired rank test. Extended Data Fig. 6 Pro-inflammatory profiles of sorted CD4 and CD8 T cells from blood. a/b/c/d) Percentage of in vitro stimulated and sorted PD-L1 and IDO specific CD4 and CD8 T cells secreting cytokines. Extended Data Fig. 7 Vaccine specific responses in skin. a) IDO and PD-L1 specific T cell responses in SKILs after 6 series of treatment measured by IFN-γ Elispot assay (n=13). SKILs were grown from DTH injection with either IDO peptide, PD-L1 peptide or a mix as presented by different blue colours. * Responses were calculated as the difference between average numbers of spots in wells stimulated with IDO or PD-L1 peptide (triplicates) and corresponding control (DMSO) and statistical analyses of Elispot responses were performed using distribution-free resampling method (Moodie et al.). DR: Not statistically confirmed response due to replicate number but number of spots in peptide wells are two times higher than control wells (DMSO). NS: No significant response and no DR. b/c/d) Percentage of cytokine secreting/CD107a+ CD4+ and CD8+ IDO and PD-L1 specific T cells in response to in vitro peptide stimulation by flow cytometry. e) Skin infiltrating PD-L1 specific T cell clones also found in biopsy in patient MM01. TCR sequencing was performed on a PD-L1 specific T cell culture generated from DTH area on the lower back injected with PD-L1 peptide on patient MM01. Bars show the frequency of top 25 clones in the culture with which indicates a high Simpson clonality of 0.43. f) Tracking the frequency of the top five skin infiltrating PD-L1 specific clones in tumour before and after treatment. Extended Data Fig. 8 T cell changes in blood after treatment. T cell fraction, TCR clonality and repertoire richness in blood and peripherally expanded TCR clones in both responding and non-responding patients. a) T cell fraction in peripheral blood of five patients at baseline, series 3, 6 and 12 by TCR sequencing. T cell fraction was calculated by taking the total number of T cell templates and dividing by the total number of nucleated cells (n=5.) b) TCR clonality in peripheral blood of five patients at baseline, series 3, 6 and 12 by TCR sequencing (n=5). Simpson clonality measures how evenly TCR sequences are distributed amongst a set of T cells where 0 indicate even distribution of frequencies and 1 indicate an asymmetric distribution where a few clones dominate. c) TCR repertoire richness in peripheral blood of five patients at baseline, series 3, 6 and 12 by TCR sequencing (n=5). TCR repertoire richness report the mean number of unique rearrangements. d) Bar-chart chart representing dynamics of the expanded T cell clones. Coloured bars represent peripherally expanded clones in five patients at series 3, series 6 and series 12 as compared to the baseline PBMC samples (n=5). Light gray bar represents peripherally expanded clones that are present in baseline biopsy samples while dark gray bar represents peripherally expanded clones that are present in the post-treatment biopsy (after 6 series). e) Frequency of the dominant TCR β chain in clonal PD-L1 and IDO specific cultures as determined by CDR3 sequencing. Extended Data Fig. 9 Profiling of genes relevant to T cell activation, cytokines and exhaustion markers on pre- and post-treatment biopsies in two patients. a) RNA expression profiling of genes related to T cell activation was performed using NanoString nCounter (n=2). b) RNA expression profiling of genes related to cytokine activity was performed using NanoString nCounter (n=2). c) RNA expression profiling of genes related to checkpoint inhibitors was performed using NanoString nCounter (n=2). Extended Data Fig. 10 Treatment induced upregulation of PD-L1, IDO, MHC I and MHC II and distance between CD8 T cells and PD-L1 expressing cells. a) IHC on 4 paired biopsies stained for CD3 and CD8 T cells, PD-L1, MHC I and MHC II on tumor cells/mm2 (sum in the validated area) and IDO H-score (from 0 to 300). H-score is the expression of IDO on both immune and tumor cells: The score is obtained by the formula: 3 x percentage of cells with a strong staining + 2 x percentage of cells with a moderate staining + percentage of cells with a weak staining (n=4). b) Distance in µm between CD8+ T cells and PD-L1+ stained cells on five baseline biopsies detected by IHC (n=5). Extended data is available for this paper at 10.1038/s41591-021-01544-x. Supplementary information The online version contains supplementary material available at 10.1038/s41591-021-01544-x. Acknowledgements We thank all patients and their relatives for being a part of the trial. We thank Ö. Met, M. Jonassen, S. Ullitz Færch, B. Saxild, S. Wendt and C. Grønhøj for technical support. We thank M. Cumberbatch for input with translational analysis. We thank the nurses at clinic 5 and the head of the Oncology Department at Herlev and Gentofte Hospital, L. Sengeløv. The study was funded through a research funding agreement between IO Biotech and the CCIT-DK, Herlev Hospital and the Oncology Department at Herlev Hospital. Author contributions I.M.S. and M.H.A. conceived the study. J.W.K. wrote the protocol with input from I.M.S., M.H.A. and M.-B.Z. J.W.K. and C.L.L. were responsible for patient treatment and patient care with help from R.B.H. and consulting from I.M.S., E. Ellebaek and M.D. I.M.S., E. Ellebaek and M.D. were responsible for patient recruitment. J.W.K. and C.L.L. were responsible for coordinating trial procedures and collecting data and samples. J.W.K. and C.L.L. were clinical investigators, and I.M.S. was a sponsor. J.W.K. and C.L.L. analyzed and interpreted clinical data with support from I.M.S. and E. Ehrnrooth. T.W.K. performed statistical analyses. H.W.H. helped with clinical imaging. J.W.K., E.M., C.O.M., S.M.A., S.E.W.-B., M.O.H. and A.W.P. analyzed and interpreted translational data with support from M.H.A. and I.M.S. Funding was acquired by I.M.S., M.H.A. and M.-B.Z. J.W.K. wrote the manuscript with input from I.M.S., M.H.A., E.M. and A.W.P. All authors reviewed the manuscript, interpreted data and approved the final version. Data availability This clinical trial was registered at https://www.clinicaltrials.gov/ before patient enrollment (clinical trial identifier NCT03047928). Each vaccine was composed of 100 µg IO102, a 21-amino-acid peptide (DTLLKALLEIASCLEKALQVF) from the IDO peptide, and 100 µg IO103, a 19-amino-acid peptide (FMTYWHLLNAFTVTVPKDL) from the signal peptide of PD-L1. TCR sequencing data are available from Adaptive Biotechnologies. Upon request, the CCIT-DK office will provide a username and a password to access the designated data within approximately 2–4 weeks (https://www.herlevhospital.dk/ccit-denmark/find-us/Sider/Contact-information.aspx). All requests for the remaining data including raw data and analyzed data and materials will, within a reasonable time frame, be reviewed by the CCIT-DK office (https://www.herlevhospital.dk/ccit-denmark/find-us/Sider/Contact-information.aspx) to verify whether the request is subject to any intellectual property or obligations. Patient-related data not included in the paper were generated as part of clinical trials and may be subject to patient confidentiality. Any data and materials that can be shared will be released via a material-transfer agreement. The following database was used in the study: https://research.regionh.dk/da/publications/the-danish-metastatic-melanoma-database-dammed(32749d99-095f-4cae-b5de-769bae27f01e).html. Competing interests M.H.A. is named as an inventor on various patent applications relating to therapeutic uses of IDO and PD-L1 peptides. These patent applications are assigned to the company IO Biotech, which is developing immune-modulating cancer treatments. M.H.A. is a founder, shareholder and advisor for IO Biotech. I.M.S. is a cofounder, shareholder and advisor for IO Biotech. I.M.S. has an advisory board relationship with or lectured for Roche, Novartis, MSD, Celgene, Incyte, TILT Bio, Pfizer and BMS AstraZeneca and has received limited grants for translational research from BMS, Roche and Novartis. M.-B.Z. is the CEO, founder and shareholder at IO Biotech. E.M., A.W.P. and E. Ehrnrooth are employees at IO Biotech. E. Ellebaek has received honoraria from BMS, Pierre Fabre, Roche and Kyowa Kirin and travel support from MSD. M.D. has received honoraria from Genzyme, MSD, BMS, Roche and Novartis and travel support from Novartis, MSD, BMS, Roche and Pfizer. The remaining authors declare that they have no conflict of interest. Peer review information Nature Medicine thanks George Coukos, Michael Schell and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. Javier Carmona was the primary editor on this article and managed its editorial process and peer review in collaboration with the rest of the editorial team. Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. These authors contributed equally: Julie Westerlin Kjeldsen, Cathrine Lund Lorentzen. Change history 3/8/2022 A Correction to this paper has been published: 10.1038/s41591-022-01771-w

NIVOLUMAB

DrugsGivenReaction

CC BY

34887574

2021-12-09

Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Off label use'.

A phase 1/2 trial of an immune-modulatory vaccine against IDO/PD-L1 in combination with nivolumab in metastatic melanoma. Anti-programmed death (PD)-1 (aPD1) therapy is an effective treatment for metastatic melanoma (MM); however, over 50% of patients progress due to resistance. We tested a first-in-class immune-modulatory vaccine (IO102/IO103) against indoleamine 2,3-dioxygenase (IDO) and PD ligand 1 (PD-L1), targeting immunosuppressive cells and tumor cells expressing IDO and/or PD-L1 (IDO/PD-L1), combined with nivolumab. Thirty aPD1 therapy-naive patients with MM were treated in a phase 1/2 study ( https://clinicaltrials.gov/ , NCT03047928). The primary endpoint was feasibility and safety; the systemic toxicity profile was comparable to that of nivolumab monotherapy. Secondary endpoints were efficacy and immunogenicity; an objective response rate (ORR) of 80% (confidence interval (CI), 62.7-90.5%) was reached, with 43% (CI, 27.4-60.8%) complete responses. After a median follow-up of 22.9 months, the median progression-free survival (PFS) was 26 months (CI, 15.4-69 months). Median overall survival (OS) was not reached. Vaccine-specific responses assessed in vitro were detected in the blood of >93% of patients during vaccination. Vaccine-reactive T cells comprised CD4+ and CD8+ T cells with activity against IDO- and PD-L1-expressing cancer and immune cells. T cell influx of peripherally expanded T cells into tumor sites was observed in responding patients, and general enrichment of IDO- and PD-L1-specific clones after treatment was documented. These clinical efficacy and favorable safety data support further validation in a larger randomized trial to confirm the clinical potential of this immunomodulating approach. pmcMain Despite remarkable advances in the treatment of MM with immune checkpoint inhibitors (ICIs) targeting PD-1 and cytotoxic T lymphocyte antigen 4 (CTLA-4), around half of patients are resistant to ICI monotherapy1. The combination of anti-CTLA-4 (aCTLA-4) and aPD1 therapy is to date the most effective therapy resulting in a response rate of around 60%; however, 50% of the patients also develop severe adverse events2,3. Therefore, an equally effective but less toxic treatment is highly needed. Several approaches to enhance ICI efficacy are currently being investigated, such as other ICIs, T cell therapy with tumor-infiltrating T cells or innate immunity stimulators such as Toll-like receptor 9 agonists4–6. Treating cancer patients with vaccines that stimulate a targeted immune response is another attractive approach, with very few side effects observed thus far in combination immunotherapy studies7,8. Immune-modulatory vaccines targeting tumoral immune escape mechanisms offer a new, generalizable strategy compared to patient-specific neoantigen cancer vaccines7,9. The immune-modulatory vaccine strategy in this clinical trial is based on the finding of circulating cytotoxic T cells specific to IDO and PD-L1 in the blood of patients with cancer and, to a lesser extent, in healthy donors. IDO- and PD-L1-specific CD8+ T cells can directly recognize and kill IDO+ and/or PD-L1+ tumor cells and likewise recognize and kill non-malignant cells that express their cognate targets. Furthermore, IDO- and PD-L1-specific CD4+ T cells release pro-inflammatory cytokines in response to IDO- or PD-L1-expressing target cells. IDO and PD-L1 are expressed not only by melanoma cells but also by many other cell types in the tumor microenvironment (TME), which differentiates these antigens from traditional tumor antigens used in other studies10–15. Activation of IDO/PD-L1-specific T cells by vaccination can therefore restrict the range of immunosuppressive signals mediated by immunosuppressive cells and thereby revert the TME from an immune hostile to an immune friendly environment. In animal models of cancer, vaccinations with IDO epitopes resulted in anti-tumor therapeutic effects that were correlated with reductions in IDO expression in myeloid cell populations within the TME16. The IDO/PD-L1 immune-modulating vaccine may lead to a translatable strategy for improving the efficacy of aPD1 therapy through activation of specific T cells. We hypothesize that the IDO/PD-L1 vaccine attracts T cells into the tumor, which induces type 1 helper T (TH1) cell inflammation and reverts the TME into an immune-permissive site, thereby turning the tumor ‘hot’. This would also upregulate PD-L1 expression in cancer and immune cells, generating more susceptible targets to aPD1 therapy (Extended Data Fig. 1a). This theory was confirmed in a mouse model in which aPD1 therapy and IDO vaccination show synergistic effects16. In this phase 1/2 clinical trial, MM1636, patients with MM received a combination of the IDO/PD-L1 (IO102/IO103) peptide vaccine with the adjuvant Montanide and the aPD1 antibody nivolumab. Patients were included in three cohorts: 30 aPD1 therapy-naive patients (cohort A), ten aPD1 therapy-refractory patients (cohort B, de novo resistance) and ten patients who progressed after aPD1 therapy (cohort C, acquired resistance). Here, we report results from cohort A. The vaccine was given biweekly for the first six administrations and thereafter every 4th week. A maximum of 15 vaccines were administered. Nivolumab was given in parallel, biweekly (3 mg per kg) or every 4th week (6 mg per kg) for up to 2 years (Extended Data Fig. 1b). The primary objective was safety and feasibility. Secondary objectives were immunogenicity and clinical efficacy. Results Patients and treatment Thirty patients were enrolled from December 2017 to June 2020. None of the 30 patients dropped out of the study; all received at least three cycles of therapy (Supplementary Fig. 1). At the current database lock (5 October 2020), six patients were still on treatment in the trial. Of the 24 patients who were not on trial treatment at data cutoff, two are still receiving nivolumab monotherapy (6 mg per kg every 4 weeks). Reasons for stopping treatment for the remaining 22 patients included disease progression (37%), toxicity (20%), maximum benefit or complete response (CR) confirmed on two consecutive scans (17%) or completing 2 years of treatment (7%). For the 24 patients who were not on trial treatment at data cutoff, the mean number of vaccinations was 10.5 (range, 3–15 vaccinations). Thirteen of these 24 patients continued nivolumab (6 mg per kg, every 4 weeks) as a standard of care. Nine patients received subsequent therapy after progression (Supplementary Table 1). Baseline characteristics are shown in Table 1. The mean age was 70 years; 37% of patients had elevated lactate dehydrogenase (LDH) levels, 60% were stage M1c, 37% had BRAF mutations, and 43% were negative for PD-L1 (<1%). A total of three patients (10%) had received prior ipilimumab therapy. No patients had brain metastasis (Supplementary Table 2).Table 1 Baseline patient characteristics (n = 30) Characteristic Number (%) Mean age, years (range) 70 (46–85) Sex, male 16 (55%) ECOG PS 0 26 (87%) LDH levels  ≤ULN 19 (63%)  >ULN 11 (37%) M stage (AJCC-8)  M1a 6 (20%)  M1b 6 (20%)  M1c 18 (60%) Number of lesion sites  1 6 (20%)  2–3 17 (57%)  >3 7 (23%) Liver metastases present 10 (33%) BRAF status  Mutant 11 (37%)  Wild type 19 (63%) PD-L1  <1% 13 (43%)  >1% 17 (57%) Previous systemic therapy  Ipilimumab 3 (10%)  No 27 (90%) AJCC-8, eighth edition of the American Joint Committee on Cancer; ECOG PS, Eastern Cooperative Oncology Group performance status; ULN, upper limit of normal. Notable clinical responses to the combination therapy Thirty patients with MM were treated with the IDO/PD-L1 vaccine and nivolumab according to the trial protocol. By investigator review, the ORR reached 80% (CI, 62.7–90.5%), with 43% of patients (CI, 27.4–60.8%) achieving a CR and 37% (CI, 20.9–54.5%) reaching a partial response (PR) as the best overall response, while 20% experienced progressive disease (PD) according to Response Evaluation Criteria in Solid Tumors (RECIST) 1.1 (Fig. 1a). Two of the patients with a PR did not have a confirmatory scan stating PR on two consecutive scans. Early onset of response was frequent, with 22 of 30 patients having an objective response at the first evaluation (after 12 weeks on treatment). Median times to PR and CR were 75 d (range, 54–256 d) and 327 d (range, 73–490 d), respectively (Fig. 2a–c).Fig. 1 Clinical response. a, Pie charts with percent ORR, CR, PR and PD according to RECIST 1.1 by investigator review of all patients (n = 30), PD-L1+ patients (>1%, (n = 17)) and PD-L1− patients (<1%, n = 13)), respectively. Two-sided CIs (95%) were constructed using the Clopper–Pearson method. b, Treatment effect in MM1636 compared with a matched historical control group from the DAMMED database (n = 74). Patients in MM1636 (n = 29) were matched with the exact same combination variable according to age (≤70 years, >70 years), sex, LDH levels (normal, elevated), M stage (M1a, M1b, M1c), BRAF status (wild type, mutated) and PD-L1 status (<1%, ≥1%). Estimates for treatment effects were calculated by weighted logistic regression analyses and weighted Cox proportional hazard model. Bar height indicates the estimated response rate; tops of bars are centers for error bars. Odds ratios (OR), response rates and their corresponding 95% CIs were extracted from the regression model. All P values were two sided, and P values below 0.05 were considered statistically significant. c, Best change in the sum of target lesion size compared with that at baseline (n = 30). The horizontal line at −30 shows the threshold for defining an objective response in the absence of non-target disease progression or new lesions according to RECIST 1.1. Two patients with 100% reduction in target lesion size had non-target lesions present. White stars, six patients had normalization (<10 mm) of fluor-18-deoxyglucose (FDG)-negative lymph nodes (at baseline, lymph nodes were >1.5 cm and FDG+) and 100% reduction of non-lymph node lesions and are considered to have had a CR (green bar). Black star, one patient (MM29) was considered to have had a PR (blue bar), although he had not reached a −30% change in target lesion size. The patient had a single measurable 13-mm lung metastasis at baseline and multiple biopsy-verified cutaneous metastases on the left crus (not detectable by positron-emission tomography–computed tomography (PET–CT) at baseline). The best change in target lesion size was 10 mm, and a post-treatment biopsy from the cutaneous metastases showed no sign of malignancy. Thus, overall, the patient was classified as having a PR. d, Kaplan–Meier curve of the response duration in the 24 patients with an objective response. e, Kaplan–Meier curve of PFS in all 30 treated patients. f, Kaplan–Meier curve of OS in all 30 treated patients. Fig. 2 Clinical response. a, Swimmer plot showing response duration and time to response according to RECIST 1.1 for all treated patients (n = 30). Triangles indicate first evidence of PR, while squares indicate first evidence of CR. Closed circles indicate time of progression. Arrows indicate ongoing responses. Patient MM18 died due to nivolumab-induced side effects. b, Spider plot showing response kinetics in all treated patients (n = 30). Red squares indicate time of progression. c, PET–CT images of patient MM42 before and after treatment (after 12 series of treatment) showing FDG metabolism in target lesions. SC, subcutaneous. The ORR among PD-L1+ (>1% (clone 28.8)) patients (n = 17) was 94.1% (CI, 73–99.7%) and 61.5% (CI, 35.5–82.3%) in PD-L1− patients (n = 13) (Fig. 1a). Objective responses were observed in patients irrespective of human leukocyte antigen (HLA) genotype (Supplementary Fig. 2). Clinical response data were validated by blinded independent external review, in which an ORR of 76.6% (CI, 57.7–90.1%) was reported, with 53.3% of patients achieving a CR, 23.3% achieving a PR and 3.3% experiencing stable disease. Comparisons between investigator review and external review are outlined in Supplementary Table 3. To examine whether the very high response rate should be attributed to nivolumab or to the vaccine, we retrieved clinical response information from a matched historical control group using the Danish Metastatic Melanoma Database (DAMMED) from contemporaneously treated patients with stage 3–4 melanoma who received aPD1 monotherapy17. Patients were matched with the exact same combination variable according to age, sex, PD-L1 status, BRAF status, LDH level and M stage (those at stage M1d were excluded from the control group (no patients with brain metastasis)). Matched controls were identified for 29 patients, and the ORR of 79.3% (CI, 61.0–90.4%) observed in MM1636 was found to be significantly higher (P < 0.0012) than that in the matched control group in which an ORR of 41.7% (CI, 31.0–53.3%) was reached. Furthermore, of the 29 patients in MM1636, a significantly (P < 0.0017) higher percentage (41.4% (CI, 25.2–59.6%)) of patients achieved CR in MM1636 than that (12% (CI, 6.3–21.6%)) in the matched historical control group. ORR and complete response rate (CRR) in the matched historical control group were comparable to those of patients treated in randomized phase 3 pivotal trials with aPD1 monotherapy18 (Fig. 1b). The treatment leads to prolonged PFS At data cutoff, the median duration of response had not been reached, with 87% of all responding patients being free from progression at 12 months (Fig. 1d). Patients were followed for up to 35 months with a median follow-up time of 22.9 months (CI, 14.9–26.2 months). OS and PFS were calculated from the first day of treatment to death or progression or to the date of the last follow-up (5 October 2020). The median PFS (mPFS) for all treated patients was 26 months (CI, 15.4–69 months) and was not reached for responding patients (Fig. 1e). The median OS was not reached at the data cutoff. OS at 12 months was 81.6% (CI, 61.6–92%) (Fig. 1f). One patient (MM18) with a CR died from nivolumab-related severe adverse events; the remaining patients died because of metastatic melanoma. For comparison, the mPFS was 8.3 months (CI, 5.5 months–NR (not reached)) in the matched historical control group (n = 74), while the median OS was 23.2 months (CI, 23.2 months–NR). (Extended Data Fig. 2a,b). The combination of the IDO/PD-L1 vaccine and nivolumab was safe Treatment-related adverse events are listed for all 30 patients in Supplementary Table 4. Common treatment-related grade 1–2 toxicities were fatigue (47%), rash (47%), arthralgia (30%), diarrhea (23%), nausea (23%), dry skin (20%), pruritus (20%), infusion reaction (17%), xerostomia (17%) and myalgia (17%). Four patients (13%) experienced grade 3–4 adverse events: one patient with a grade 3 maculopapular rash (MM01), one patient with grade 3 adrenal insufficiency (MM06) and one patient with grade 3 arthralgia (MM22). Patient MM18 died from urosepsis with multi-organ failure and severe hyponatremia. This patient had experienced multiple immune-related adverse events with grade 3 colitis, grade 2 pneumonitis, grade 3 arthralgia, grade 2 vasculitis and grade 2 nivolumab infusion-related allergic reaction. Additionally, patient MM18 had symptoms of myocarditis at the time of death with highly elevated cardiac troponin I levels. Bedside echocardiography showed an ejection fraction of 15%, which at baseline was 60%, but an autopsy was not conducted, and myocarditis was never confirmed pathologically. Patient MM06 had received first-line treatment with ipilimumab before entering the trial and was on substitution corticosteroids at the time of inclusion. Adrenal insufficiency was aggravated by an erysipelas infection with high fever, reaching grade 3 in Common Terminology Criteria for Adverse Events and resolving quickly after appropriate antibiotic therapy was initiated. As expected, local side effects were common with 77% of the patients who developed injection site reactions. These reactions were classified as granulomas (63%), redness (20%), pain (13%) and pruritus (13%) at the injection site. All local reactions were grade 1–2, most likely related to the Montanide adjuvant and typically transient. However, two patients (MM07 and MM20) decided to discontinue vaccination after eight and 11 injections, respectively, due to granulomas, tenderness and pain that limited instrumental activities of daily living but continued to receive nivolumab (Extended Data Fig. 2c). Vaccine-specific responses in blood were frequently detected First, all 30 patients were assessed for the presence of vaccine-specific responses in peripheral blood mononuclear cells (PBMCs) before, on and after vaccination using a modified interferon (IFN)-γ enzyme-linked immune absorbent spot (ELISPOT) assay. This assay is known to expand antigen-specific memory cells and to improve the detection and the correlative power of ELISPOT during treatment19–21. Prevaccine IDO-specific responses were detectable in ten (33%) patients, while prevaccine PD-L1-specific responses were detectable in eight patients (27%); overlapping (specific to both IDO and PD-L1) prevaccine responses were present in four (13.3%) patients. During vaccination, an increase of IDO-specific T cells or PD-L1-specific T cells in the blood was observed in 28 (93%) and 26 (86%) patients, respectively. In total, 93% of patients had an increase in either PD-L1- or IDO-specific responses on vaccination (Fig. 3a), with a significant (P < 0.0001) median increase from baseline to the post-vaccine response demonstrated for both IDO and PD-L1 (at different time points on treatment), confirming that vaccine-specific immune responses were induced in patients regardless of clinical response (Fig. 3b,c). Immune responses fluctuated in the blood over time (Extended Data Fig. 3a). An increase in IDO- and PD-L1-specific responses in the peripheral blood was also detectable directly ex vivo across the different clinical response groups (Extended Data Fig. 4).Fig. 3 Vaccine-specific responses in blood. a, IDO- and PD-L1-specific T cell responses in PBMCs at baseline and on vaccination as measured by the IFN-γ ELISPOT assay (n = 30). *Responses were calculated as the difference between the average numbers of spots in wells stimulated with IDO or PD-L1 peptide (triplicates) and those from the corresponding control (DMSO), and statistical analyses of ELISPOT responses were performed using a distribution-free resampling method (Moodie et al.50). DR (double response), response was not statistically confirmed due to replicate number, but the number of spots in peptide wells was two times higher than that in control wells (DMSO). NS, no significant response and no DR. For a detailed overview of responses at serial time points on vaccination, see Extended Data Fig. 4a. b, IDO- and PD-L1-specific T cell response in PBMCs in all treated patients measured by the IFN-γ ELISPOT assay at baseline and on vaccination. On-vaccination responses were selected from the ‘best’ ELISPOT response at different time points for each patient (series 3, 6, 12, 18 or 24) during vaccination (n = 30). Wilcoxon matched-pairs signed-rank test was used to compare responses to IDO or PD-L1 peptides in the vaccine between baseline and later time points. c, Representative example of ELISPOT wells with response in patient MM23 in serial PBMCs before and on treatment. d, IDO-specific CD4+ and CD8+ T cells were isolated and expanded from PBMCs stimulated in vitro with the IDO peptide and a low dose of IL-2 for 14–15 d before sorting using the Miltenyi Cytokine Secretion Assay—Cell Enrichment and Detection kit. To assess their cytolytic potential, IDO-specific T cells were stimulated with IDO peptide, and expression of CD107a, IFN-γ and TNF-α was assessed by flow cytometry (the example is from patient MM14). Sustained vaccine-specific responses were observed 3 and 6 months after the last vaccine, indicating induction of memory responses in nine patients with response to clinical treatment who surpassed follow-up at data lock (Extended Data Fig. 3b). Importantly, PD-L1- and IDO-specific responses were observed irrespective of HLA genotype (Supplementary Table 5). To verify the functionality of vaccine-induced T cells, IDO- or PD-L1-specific T cells were isolated and in vitro expanded from PBMCs of five patients. Phenotypic characterization by flow cytometry revealed that the isolated vaccine-specific T cells consisted of both CD4+ and CD8+ T cells. Also, both IDO- and PD-L1-specific CD4+ and CD8+ T cells showed pro-inflammatory properties, as they expressed the cytolytic marker CD107a and secreted the cytokines IFN-γ and tumor necrosis factor (TNF)-α (Fig. 3d and Extended Data Fig. 6a–d). Interestingly, we were also able to detect vaccine-specific CD4+ and CD8+ T cell responses in peripheral blood ex vivo (Extended Data Figs. 4 and 5). We observed a significant increase in the overall percentage of CD107a, CD137 and TNF-α expression in response to peptide stimulation in on- or post-treatment PBMC samples compared to that at baseline, further confirming the expansion and the diverse signature of vaccine-specific T cells (Extended Data Fig. 5). Vaccine-specific responses detected in the skin at the vaccination site To investigate whether vaccine-specific T cells have the potential to migrate to peripheral tissue, delayed-type hypersensitivity (DTH) tests were performed after six cycles of treatment on 15 patients to assess the presence of vaccine-reactive T cells in the skin. Supplementary Table 6 display an overview of skin-infiltrating lymphocyte (SKIL) cultures. We detected IDO-specific T cells in the skin of six of ten patients and PD-L1-specific T cells in nine of 11 patients (Extended Data Fig. 7a). Intracellular cytokine staining was performed on SKILs from five patients after stimulation with either PD-L1 or IDO peptide. Here, we detected mainly CD4+ peptide-reactive T cells that secreted TNF-α and upregulated CD107a, and a minor fraction also secreted IFN-γ. In one patient, we detected CD8+ PD-L1-reactive T cells (Extended Data Fig. 7b–d). Vaccine-induced T cells specifically recognize target cells To confirm the functionality of vaccine-expanded T cells, vaccine-specific T cell clones (clonal purity was confirmed by T cell receptor (TCR) sequencing) were isolated and expanded from patient PBMCs (Extended Data Fig. 8e). We showed that PD-L1-specific T cells were able to recognize PD-L1+ autologous tumor cells in a PD-L1 expression-dependent manner if the cancer cells also expressed HLA-II (Fig. 4a,b). Similarly, an HLA-DR-restricted IDO-specific CD4+ T cell clone was able to recognize an HLA-DR-matched IDO-expressing model cell line, MonoMac1, in an IDO-expression-dependent manner (Fig. 4e–g). As previously described, IDO- and PD-L1-specific T cells’ mode of action is not limited to targeting only cancer cells. We were able to show that vaccine-specific T cell clones also reacted against PD-L1- and IDO-expressing autologous immune cells (Fig. 4c,h). To provide myeloid cells with a tumor-associated phenotype we treated isolated CD14+ myeloid cells with tumor conditioned medium (TCM) derived from an established autologous tumor cell line. We observed that such TCM-treated CD14+ cells had increased expression of PD-L1 and IDO and were effectively recognized by autologous PD-L1- and IDO-specific CD4+ T cell clones (Fig. 4c,d,h,i).Fig. 4 PD-L1- and IDO-specific T cells from vaccinated patients react against PD-L1- and IDO-expressing target cells. a, Left, PD-L1-specific T cell culture (MM1636.05) reactivity against PD-L1 peptide or autologous tumor cells in the IFN-γ ELISPOT assay. Tumor cells were either not treated or pretreated with 200 U ml−1 IFN-γ for 48 h before the assay. Effector:target (E:T) ratio of 10:1 was used. Right, PD-L1 and HLA-II surface expression on melanoma cells with (green) or without (yellow) pretreatment with IFN-γ compared to an isotype control (gray) as assessed by flow cytometry. b, Left, PD-L1-specific T cell (MM1636.05) reactivity in the IFN-γ ELISPOT assay against autologous tumor cells pretreated with IFN-γ (500 U ml−1) and transfected with mock or PD-L1 small interfering (si)RNA 24 h after transfection. E:T ratio, 10:1. Right, PD-L1 surface expression on melanoma tumor cells (MM1636.05) assessed by flow cytometry 24 h after transfection with mock (blue) or PD-L1 (red) siRNA compared to the isotype control (gray). c, Reactivity of the CD4+ PD-L1-specific T cell clone (MM1636.14) against PD-L1 peptide or autologous CD14+ cells; E:T ratio, 10:1. CD14+ cells were isolated using magnetic bead sorting and used as targets in an ELISPOT assay directly or after pretreatment for 2 d with TCM derived from the autologous tumor cell line. d, Quantitative PCR with reverse transcription (RT–qPCR) analysis of PD-L1 (CD274) expression in sorted CD14+ cells before and after treatment with autologous TCM for 48 h. e, Reactivity of the IDO-specific CD4+ T cell clone (MM1636.23) against IDO peptide combined with HLA-DR (L243)-, HLA-DQ (SPV-L3)- or HLA-DP (B7/21)-blocking antibodies (aHLA-DR, aHLA-DQ and aHLA-DP) in an intracellular staining assay (ICS) for IFN-γ and TNF-α production. T cells were incubated with individual blocking antibodies (2 μg ml−1) for 30 min before adding IDO peptide. f, Reactivity of the IDO-specific CD4+ T cell clone (MM1636.23) against the HLA-DR-matched IDO-expressing cell line MonoMac1 transfected with mock or IDO siRNA in an ICS assay, E.T ratio, 4:1. siRNA transfection was performed 48 h before the experiment. g, RT–qPCR analysis of IDO1 expression in MonoMac1 cells 48 h after siRNA transfection. h, Reactivity of the CD4+ IDO-specific T cell clone (MM1636.14) against IDO peptide or autologous CD14+ cells; E:T ratio, 20:1. CD14+ cells were isolated using magnetic bead sorting and used as targets in an ELISPOT assay directly or after pretreatment with TCM derived from the autologous tumor cell line. i, RT–qPCR analysis of IDO1 expression in sorted CD14+ cells before and after treatment with autologous TCM for 48 h. Bars in RT–qPCR data (d,g,i) represent the mean of three (d,i) or six (g) technical replicates ±s.d.; P values were determined by two-tailed parametric t-tests. ELISPOT counts (a,b,c,h) represent the mean value of three technical replicates ±s.e.m.; response P values were determined using the distribution-free resampling (DFR) method. TNTC, too numerous to count. T cell clones in blood and tumors To track the role of treatment-induced T cell responses, TCR sequencing of the complementarity-determining region 3 (CDR3) was performed on five patients in peripheral blood (baseline and cycles 3, 6 and 12) and paired biopsies. These five patients (MM01, MM02, MM08, MM09 and MM13) were selected due to the availability of material and to investigate a balanced patient group with both responders and non-responders. Due to a limited number of available paired biopsies (either as a consequence of patient refusal or unexpectedly rapid and substantial clinical responses), statistical analysis could not be applied. Details on clinical response are shown in Fig. 2a. Additionally, PBMCs (on treatment) or SKILs were stimulated with the IDO/PD-L1 peptides, and then cytokine-producing T cells were sorted to track vaccine-induced T cells both in the periphery and at the tumor site. To identify enriched IDO/PD-L1-specific T cell clones, TCR rearrangements in sorted IDO/PD-L1-specific clones and TCR rearrangements in sorted IDO/PD-L1-specific T cell samples were compared to sequences from baseline PBMC samples for each patient. Clonal expansion of vaccine-specific TCR rearrangements from samples on vaccination were then tracked using a differential abundance framework. Cumulative IDO/PD-L1-specific T cell frequencies were tracked in post-treatment samples. We found no relation between clinical response and the enrichment of vaccine-specific clones, but an increase in IDO/PD-L1-specific T cell clones was observed at different time points in the periphery in all five patients (Extended Data Fig. 3c). We next investigated overall changes in the T cell repertoire in the blood. A modest increase in the peripheral T cell fraction was observed in the three responding patients at cycle 3, while the two non-responding patients had a clear decrease in T cell fraction (Extended Data Fig. 8a). We thereafter investigated TCR clonality and TCR repertoire richness, exploring the proportion of abundant clones and the number of unique rearrangements, respectively. A decreasing peripheral Simpson clonality and increasing TCR repertoire richness was observed in responding patients at cycle 3, which might indicate tumor trafficking upon treatment. The opposite pattern was observed in non-responding patients (Extended Data Fig. 8b,c). Peripherally expanded clones were associated with tumors and persisted until cycle 12 (latest time point analyzed). The largest peripheral expansion was observed at cycle 3, with the most significant increase observed in patient MM01 (CR). Responding patients had a larger fraction of peripherally expanded clones that were also found in tumors compared to that of non-responders. By tracking peripherally expanded clones detected at the tumor site, we observed that patient MM01 had a substantial increase after treatment, indicating tumor trafficking of peripheral expanded clones (Extended Data Fig. 8d). Influx at the tumor site of vaccine-enriched T cell clones Given the observation of increased T cell fraction and enrichment of IDO- and PD-L1-specific clones in the blood after treatment, we investigated whether the same trend was observed at the tumor site. Both TCR sequencing and immunohistochemistry (IHC) of paired biopsies from the five patients described above showed an increase in the T cell fraction with an influx of CD3+ and CD8+ T cells after treatment in the three responding patients (Fig. 5a–c). IHC was not possible for patient MM09 due to tissue loss.Fig. 5 Changes in the TME after treatment. Number of CD3+ and CD8+ T cells, TCR fraction, TCR clonality, TCR repertoire, biopsy expanded TCR clones and enrichment of IDO- and PD-L1-specific T cells at the tumor site. a, Number of CD3+ and CD8+ T cells per mm2 (sum in the validated area) at the tumor site detected by IHC of paired biopsies from four patients. b, Example of IHC of CD3+ and CD8+ T cells at the tumor site before and after treatment (cells per mm2) in one patient (MM01). c, T cell fraction at the tumor site at baseline and cycle 6 by TCR sequencing. The T cell fraction was calculated by taking the total number of T cell templates and dividing by the total number of nucleated cells. d, Tracking of vaccine-associated clones at baseline and cycle 6 in tumor biopsies. Cumulative frequencies of IDO and PD-L1 vaccine-specific TCR rearrangements are represented. e,f, TCR clonality and TCR repertoire richness in five patients at the tumor site at baseline and cycle 6. Simpson clonality measures how evenly TCR sequences are distributed among a set of T cells, where 0 indicates an even distribution of frequencies and 1 indicates an asymmetric distribution in which a few clones dominate. TCR repertoire richness reports the mean number of unique rearrangements. g, Bar chart representing baseline expanded biopsy clones from five patients (colored bars) and the detection of biopsy expanded clones also found in the blood at baseline and series 3, 6 and 12 (white and gray bars) by TCR sequencing. We thereafter investigated whether some of the IDO/PD-L1 vaccine-linked T cells were present at the tumor site. Vaccine-associated clones were tracked as the combined frequency of IDO- and PD-L1-specific T cell rearrangements. In biopsies, the frequencies at cycle 6 were compared to those at baseline and showed an increase in vaccine-specific T cells in four of five patients, irrespective of clinical response (Fig. 5d). TCR sequencing of PD-L1-specific SKILs (a more specific culture than IDO/PD-L1-specific isolations derived from PBMCs on vaccination) and paired biopsies showed that two of the top five PD-L1-specific SKIL clones were present at the tumor site both before and after treatment (Extended Data Fig. 7e,f). With a focus on the more abundant T cell clones, we investigated overall TCR clonality at the tumor site before and after treatment. In addition, we explored the number of unique TCR rearrangements, dissecting the lower-frequency clones. Patient MM01 had a significant increase in TCR clonality and a decrease in repertoire richness at the tumor site after therapy, indicating a focused tumor repertoire response of selected clones. All three responding patients had a decrease in TCR repertoire richness, which again might indicate a focused tumor response (Fig. 5e,f). Deeper analyses showed that the T cell clones that expanded at the tumor site after therapy were also present in the blood at baseline and increased significantly after treatment in four of five patients. The highest proportion was detected early at cycle 3. These data again support trafficking of peripherally expanded clones to the tumor site and could indicate that the T cell response to treatment is derived from pre-existing peripherally tumor-associated T cells (Fig. 5g). Signs of treatment-induced inflammation in the TME To dissect changes in the TME induced by T cell influx upon treatment in responding patients, RNA gene expression analyses using the nCounter PanCancer Immune Profiling Panel from NanoString were performed on paired biopsies from two responding patients (MM01 and MM13). Expression of genes related to adaptive immunity such as T cell activation, effector functions (genes encoding IFN-γ, TNF-α, IL-15 and IL-18) and cytotoxicity was increased in post-treatment biopsies (Extended Data Fig. 9a,b). Also, expression of genes related to checkpoint inhibitors such as those encoding T cell immunoglobulin and mucin-domain containing-3 (TIM-3), IDO, PD-L1, PD-L2, PD-1 and CTLA-4 increased after treatment, indicating activation of immune cells in the TME (Extended Data Fig. 9c). Additionally, IHC of paired biopsies from four patients (MM01, MM02, MM05 and MM13) showed an upregulation of PD-L1, IDO, MHC-I and MHC-II on tumor cells, indicating a treatment-induced pro-inflammatory response in the three responding patients, except for a decrease in MHC-II expression in patient MM13. By contrast, the non-responding patient MM02 had a reduction in T cell numbers present in the tumor after treatment and no expression of PD-L1, IDO or MHC-II and, interestingly, total loss of MHC-I, demonstrating tumor immune escape (Extended Data Fig. 10a). CD8+ T cells and their distance (µm) to PD-L1-expressing cells in baseline biopsies from five patients was investigated by IHC. Except for patient MM13 (PR), distance and clinical responses were associated: the two responders had reduced distance (<20 µm) between cells expressing these markers compared to non-responding patients (>80 µm). This observation indicates that responding patients not only have a higher intratumoral infiltration of CD8+ T cells, but that these cells can surround and attack PD-L1-expressing immune cells and tumor cells (Extended Data Fig. 10b). Discussion In this clinical trial, MM1636, 30 patients with metastatic melanoma were treated with a first-in-class immunomodulatory IDO/PD-L1-targeting peptide vaccine combined with nivolumab. The treatment led to an unprecedented high ORR of 80%, with 43% of patients reaching a CR, and a striking mPFS of 26 months (95% CI, 15.4–69 months) was reached. The vaccine represents a new treatment strategy to activate specific T cells that target cells contributing to immune suppression (including tumor cells), positively modulating the TME by inducing local inflammation. Indeed, we show that vaccine-specific T cells isolated and expanded from vaccinated patients recognize not only tumor cells in a target- and HLA-restricted manner but also myeloid cells polarized toward a tumor-associated phenotype. Hence, myeloid cells become targets for vaccine-activated T cells when they have a tumor-associated phenotype in the TME. These phenomena may further induce checkpoint molecules and rewire the TME toward an increasingly aPD1-permissive state. A drawback of our study is the single-center nonrandomized setup. Comparison between trials or between patients in trials and real-world patients is problematic due to multiple factors, such as period of conduction and other therapies available at the different periods of time. Nevertheless, the rate of investigator-assessed ORR in the phase 3 trial CheckMate 067 was 43.7% in the nivolumab monotherapy group and 57% in the nivolumab and ipilimumab group. CRs occurred in 8.9% and 11.5% of patients, respectively18. The mPFS of 26 months (95% CI, 15.4–69 months) in this trial is more than twice as long as that for patients treated with nivolumab and ipilimumab in CheckMate 067, for which an mPFS of 11.5 months (95% CI, 8.7–19.3 months) was reached. Patient baseline characteristics were in general comparable to those of patients with MM who have been treated in CheckMate 067, although patients in MM1636 were older (mean age, 70 years) and a larger fraction were positive for PD-L1 (57%)3,18,22. Among patients with PD-L1-negative tumors in MM1636, an ORR of 61.5% was still reached, which would be expected to be around 33% for first-line nivolumab monotherapy23. Some studies suggest that older patients might have a tendency to respond better to aPD1 therapy; however, this is still debated24–27. To address potential trial bias and the nonrandomized setup, patients in MM1636 were matched for age, performance status, sex, M stage, LDH level, PD-L1 status and BRAF status with a historical control group from the DAMMED, who were treated contemporarily (2015–2019) with aPD1 monotherapy as standard of care17. We found a significantly higher ORR and CRR in MM1636 compared to those of matched patients, who had an ORR of 43% and a CRR of 13%, comparable to patients treated in CheckMate 067. Restrictions of the synthetic control group are of course that it is partially historic and patient selection outside matching criteria cannot be ruled out28. Numerous contemporary clinical trials are exploring the combination of aPD1 therapy with other immunomodulating agents for advanced melanoma. Talimogene laherparepvec, an oncolytic virus, is approved by the Food and Drug Administration and the European Medicines Agency to treat advanced melanoma. A small phase 1b trial with 21 patients (MASTERKEY-265) combined talimogene laherparepvec and pembrolizumab to treat patients with advanced unresectable melanoma and reached an ORR of 62% and a CR of 33%29,30. Seventy-one percent of the patients in this trial had an M stage below M1c; this number was 40% in our trial. Furthermore, mainly patients with an M stage below M1c responded to treatment, which was not the case in MM1636. Results from a large randomized phase 3 trial are awaited (KEYNOTE-034). Epacadostat, an IDO inhibitor, was tested in combination with pembrolizumab in a nonrandomized phase 2 trial in 40 aPD1 treatment-naive patients with MM with promising results, reaching an ORR of 62%. Unfortunately, the phase 3 trial showed no indication that epacadostat provided improvement in PFS and OS31. Limitations of the phase 3 trial were the sparse information on pharmacodynamics as well as biomarker evaluation to improve the design. The IDO/PD-L1 vaccine is different from epacadostat, as it is not an IDO inhibitor but targets IDO- and PD-L1-expressing cells. Similar vaccines administered as monotherapy induced objective responses in lung cancer and basal cell carcinoma, while epacadostat as monotherapy in 52 patients resulted in no responses32,33. Sahin et al. recently published encouraging data from a first-in-human trial, in which a vaccine containing liposomal RNA targeting four unmutated tumor-associated antigens (NY-ESO-1, MAGE-A3, tyrosinase and TPTE) was administered alone or in combination with nivolumab in patients with advanced melanoma. Responses were observed in both the monotherapy group as well as the combination group in checkpoint inhibitor-experienced patients, suggesting the efficacy of non-mutant shared tumor antigen vaccines34. The overall safety and tolerability findings are comparable to those of aPD1 monotherapy. Injection site reactions were exclusive to the vaccine. However, these side effects were transient and mild in most patients and most likely due to the adjuvant Montanide. IDO- and PD-L1-specific CD8+ and CD4+ T cells exist among peripheral blood lymphocytes in healthy donors35–38 and expand in response to pro-inflammatory stimuli35,39. Furthermore, both IDO and PD-L1 are induced in cells as a counter response to the inflammatory response. This provides a mechanism that ensures immune homeostasis, which keeps IDO/PD-L1-specific T cells in check; therefore, the risk of triggering autoimmune-related adverse events by vaccination appears to be minimal. This was confirmed in the first clinical trials of IDO and PD-L1 vaccination (NCT01543464 (refs. 33,40) and NCT03042793 (ref. 41)). The spontaneous (baseline) immune response to the vaccines observed in the current study is in agreement with our previous observation in various patients with cancer35–38. Numerous vaccine-induced changes in the blood and at the tumor site were observed. Peripheral IDO- and/or PD-L1-specific T cells were detected in vitro in a modified ELISPOT in over 93% of patients on vaccination, unrelated to patient HLA type. Immune responses were persistent in patients who surpassed follow-up at data cutoff and were still detectable up to 6 months after the last vaccine, suggesting induction of memory T cells. We did not observe a correlation between vaccine-induced responses in blood and clinical responses. However, the detection of a highly significant increase in vaccine-specific T cell numbers after vaccination in almost all patients together with the very low number of patients with PD makes it difficult to expect such a correlation, especially because other aspects should be taken into account (for example, the loss of class I expression on tumors cells in patient MM02 with PD after vaccination). Frequencies of peripheral T cells induced against PD-L1 were overall higher than those against IDO. Importantly, however, we also observed that it was not the same patients who responded strongly to both IDO and PD-L1, nor did patients react with a similar response pattern to the two antigens; that is, different time points of response were observed. This suggests that each component of the vaccine plays different roles in the ongoing immune response in patients. The goal of IDO/PD-L1 vaccination was to modulate the TME to increase responsiveness to aPD1 therapy, as we have observed in animal models, in which immune conversion is demonstrated in the TME with an increased influx of T cells. Targeting both IDO and PD-L1 together enables synergy, as the TME is known to use different immune escape mechanisms and IDO and PD-L1 are often overexpressed by different cellular compartments. Ex vivo ELISPOT and flow cytometry assays confirmed the induction of immune response toward both epitopes and the higher immunogenicity of the PD-L1 epitope, although because of lower sensitivity failed to detect some of the responses. TCR sequencing of five patients confirmed enrichment of IDO/PD-L1-specific T cell clones in the blood at different time points after treatment. Furthermore, an increase in enriched IDO/PD-L1-specific clones was observed in four of five patients at the tumor site after treatment, irrespective of clinical response. Phenotypic characterization showed that vaccine-specific T cells, which were expanded in vitro with interleukin (IL)-2 from the blood of vaccinated individuals, were both CD4+ T cells and CD8+ T cells. This was confirmed by ex vivo phenotype description of vaccine-activated T cells. Vaccine-specific T cells expressed CD107a and CD137 and produced IFN-γ and TNF-α upon stimulation with the cognate target, indicating their cytolytic capacity. Overall, the data from immune monitoring supports the importance of both antigens in the generation of the frequent clinical responses in the study. Despite a limited number of paired biopsies (due to either patient refusal or the fact that a large fraction of responding patients had no assessable tumors after six cycles), we observed trends indicating treatment-induced general T cell influx in responding patients. It was shown that proliferation of CD8+ T cells in the tumor after aPD1 treatment is associated with radiographic reduction in tumor size42. Additionally, we showed that a large proportion of expanded peripheral TCR clones were associated with tumors and the most considerable amount of clonal expansion was observed early, at cycle 3. For the patient with a CR included in the TCR sequencing analyses (MM01), the number of peripheral expanded clones present at the tumor site increased after treatment compared to that at baseline, indicating tumor trafficking of peripheral expanded clones. Gene expression analyses (two paired biopsies) and IHC (five paired biopsies) further demonstrated that the combination treatment induced a pro-inflammatory TME in responding patients with signs of T cell activation and cytotoxicity and increased cytokine activity. This may lead to further upregulation of IDO, PD-L1, MHC-I and MHC-II on tumor cells, leading to more treatment targets. It was shown that, following vaccination with a cancer vaccine, PD-L1 expression is increased on tumor cells due to recruitment of tumor-specific T cells and upregulation of adaptive immune resistance pathways in the TME43. Treatment with nivolumab monotherapy enhances PD-L1 expression, and it is therefore problematic to discriminate the effect of the vaccine as compared to that of nivolumab44. IDO- and PD-L1-specific pro-inflammatory effector T cells were hypothesized to counteract the functions of IDO- and PD-L1-expressing immune-suppressive cells as a means to keep the immune balance between immune activation and inhibition45. However, because the expression of these molecules is induced as a counter-regulatory response to inflammatory mediators such as IFNs, they can also be expressed by, for example, activated myeloid cells or T cells. Thus, activation of PD-L1-specific T cells may result in depletion of activated T cells in the tumor and other sites. Importantly, it was recently described that PD-L1+ T cells mainly have tolerogenic effects on tumor immunity and exert tumor-promoting properties, suggesting that targeting this immune population is indeed also beneficial46. We have previously investigated the effect of activating PD-L1-specific T cells in vitro and in vivo and found that, overall, they supported the effector phase of an immune response by removing PD-L1-expressing regulatory immune cells that inhibit PD-1+ effector T cells47,48. The major role of the PD-1 pathway is believed to be the regulation of effector T cell responses. Thus, this protective pathway is more important after activation, rather than at the initial T cell-activation stage49. Accordingly, the presence of PD-L1-specific T cells during the activation phase of an immune response may not increase or support a pro-inflammatory immune response due to the expression of PD-L1 on potent antigen-presenting cells or on the T cells themselves14. Thus, the overall effects of PD-L1-specific T cells may vary depending on the expression of both PD-1 and PD-L1, that is, due to the microenvironment and the state of the immune response. In conclusion, here we report an impressive response rate, CR rate and mPFS for a first-in-class immune-modulating vaccine combined with nivolumab. This may be a first step toward a new treatment strategy for patients with MM. Limitations are the low number of patients treated at a single institution and the lack of a randomized design with aPD1 monotherapy as comparator. Studies in aPD1 therapy-resistant or -refractory melanoma are ongoing as well as biomarker analysis for selecting patients at a higher likelihood to benefit from the combination versus aPD1 monotherapy. A larger randomized trial will be essential to validate these findings and determine the specific contribution of the vaccine to clinical responses and changes in the TME. In December 2020, the Food and Drug Administration granted breakthrough therapy designation for the IO102/IO103 vaccine combined with aPD1 therapy in metastatic melanoma based on data from the MM1636 trial. Methods Trial design and treatment plan MM1636 is an investigator-initiated, nonrandomized, open-label, single-center phase 1/2 study. All patients were treated at the Department of Oncology, Herlev and Gentofte Hospital, University of Copenhagen, Herlev, Denmark. The study was conducted according to the Declaration of Helsinki and Good Clinical Practice (GCP) and monitored by the GCP unit in Copenhagen, Denmark. The protocol was approved by the ethical committee of the Capital region of Denmark (H-17000988), the Danish Medical Agencies (2017011073) and the Capital Region of Denmark Data Unit (P-2019-172). The study was registered at https://clinicaltrials.gov/ under the identifier NCT03047928 and at the EudraCT (no. 2016-0004527-23). This study initially aimed to include 30 aPD1 treatment-naive patients with MM. An amendment with the addition of two other cohorts with ten patients in each cohort was made to evaluate immune responses and clinical efficacy in aPD1 therapy-resistant patients (cohort B, de novo resistance and cohort C, acquired resistance) for a total of 50 patients. The amendment with cohorts B and C was approved at the inclusion of 18 patients in cohort A. The trial is still including patients in cohorts B and C. This article reports results from cohort A. The first six patients were treated as for phase 1, evaluating for safety and tolerability before the remaining 24 patients were included in phase 2. The IDO/PD-L1 vaccine was administered subcutaneously biweekly for the first 10 weeks and thereafter every 4 weeks for approximately 9 months. A maximum of 15 vaccines were administered. Nivolumab was administered according to the approved label (3 mg per kg biweekly) for 24 cycles. The 15th vaccine was administered together with the 24th nivolumab cycle of 3 mg per kg, and responding patients thereafter continued nivolumab monotherapy every 4th week (6 mg per kg) as a standard of care after investigator assessment. Treatment with nivolumab was discontinued at the maximum benefit (assessed by the investigator), after a maximum of 2 years of therapy, at progression or due to severe adverse events (Extended Data Fig. 1b, treatment plan). Vaccine composition Each vaccine was composed of 100 µg IO102, a 21-amino-acid peptide (DTLLKALLEIASCLEKALQVF) from IDO, and 100 µg IO103, a 19-amino-acid peptide (FMTYWHLLNAFTVTVPKDL) from the signal peptide of PD-L1 (PolyPeptide Laboratories). Peptides were dissolved separately in 50 µl DMSO, filtered for sterility and frozen at −20 °C (NUNC CyroTubes CryoLine System Internal Thread, Sigma-Aldrich). At <24 h before administration, peptides were thawed. The PD-L1 peptide was diluted in 400 µl sterile water and, immediately before injection, mixed with the IDO peptide solution and 500 µl Montanide ISA-51 (SEPPIC) to achieve a total volume of 1 ml. Patients Patients above 18 years of age with locally advanced or stage 4 melanoma according to the AJCC (seventh edition), at least one measurable lesion according to RECIST 1.1 and an ECOG PS of 0–1 were eligible. The main exclusion criteria were prior treatment with aPD1 therapy, CNS metastases >1 cm, severe comorbidities and active autoimmune disease. Enrollment was not restricted to PD-L1 status, but it was known before inclusion. Patients were included after informed consent. Patients MM42 and MM20 have confirmed their approval of PET–CT images and clinical images that have been included in the article. Key study assessments Safety and tolerability were evaluated based on changes in clinical laboratory analyses and reported adverse events. Adverse events were assessed according to Common Terminology Criteria for Adverse Events (version 5.0) and were graded from 1 to 5 for all treated patients up to 6 months after the last dose of the IDO/PD-L1 vaccine. Clinical efficacy was assessed using FDG PET–CT scans before treatment and every 3rd month until progression. Objective responses were categorized into CR, PR, stable disease or PD according to RECIST version 1.1. Clinical data were collected at CCIT-DK in the eCRF program OpenClinica version 1.0 and in Microsoft Excel version 2002 on a secure server. Blood samples for immunologic analyses were collected before treatment, before the third cycle, after the sixth, 12th, 18th and 24th cycles (on vaccination) and 3 and 6 months after the last vaccine. Two to three tumor needle biopsies (1.2 mm) were collected at baseline and after six cycles from the same tumor site, when accessible, to evaluate immune responses at the tumor site. DTH skin tests were performed and punch biopsies were taken from the DTH area after cycle 6 for the evaluation of SKILs reactive to PD-L1 and IDO (Supplementary Fig. 1, treatment plan). Statistical analysis of clinical outcome Survival curves were computed in GraphPad Prism software version 9.0.0 according to Kaplan–Meier method. Median follow-up time of enrollment was calculated using the reverse Kaplan–Meier method, also in GraphPad Prism software version 9.0.0. For binary outcomes, 95% two-sided CIs were constructed using the Clopper–Pearson method, also in GraphPad version 9.0.0. An independent board of certified and experienced oncoradiologists performed an external review to evaluate clinical response to address the potential bias of investigator site review. The external review took place at Rigshospitalet, Copenhagen University Hospital. This hospital did not participate in the MM1636 trial, and the external reviewer had no prior knowledge about the clinical trial or the trial therapy. Only PET–CT images were accessed, and arrows indicating target or non-target lesions appeared on baseline images as the only additional information. To address potential trial bias regarding treatment effect, we matched patients in MM1636 with patients from the DAMMED, a population-based database that retrospectively collects data on patients with metastatic melanoma in Denmark. Here, data from 938 patients treated with aPD1 monotherapy contemporaneously (January 2015–October 2019) were extracted. Two hundred and eighteen of these patients were eligible for comparison and matching (all parameters available) (Supplementary Table 1), and 74 patients from the DAMMED were found to match. Patients were matched for age (≤70 years, >70 years), sex, LDH levels (normal, elevated), M stage (M1a, M1b, M1c), BRAF status (wild type, mutated) and PD-L1 status (<1%, ≥1%). An exact matching algorithm was used in which patients in MM1636 were matched with patients from the DAMMED with the same combination of variables. Twenty-nine patients from MM1636 were matched with exact combinations of the six variables. One patient could not be matched. To secure balance of the calculations, control patients were weighted according to the number of patients for each MM1636 patient. Estimates for treatment effects were calculated by weighted logistic regression analyses and weighted Cox proportional hazard model. The R package ‘Matchlt’ was used for matching patients. As the method chosen for matching control patients to protocol patients, a weighted binary logistic regression model was used for comparing response rates in the two matched cohorts. Odds ratios and response rates, including their corresponding 95% CIs, were extracted from the regression models. All P values were two sided, and P values below 0.05 were considered statistically significant. SAS version 9.4M5 was used for the weighted logistic regression models. Processing of peripheral blood mononuclear cells Peripheral blood was collected from all patients in heparinized tubes and was processed within 4 h. In brief, PBMCs were isolated using Lymphoprep (Medinor) separation. PBMCs were counted on a Sysmex XP-300 analyzer and frozen in Human AB Serum (Sigma-Aldrich, H4522-100ML) with 10% DMSO using controlled-rate freezing (CoolCell, BioCision) in a −80 °C freezer and were moved the next day to a freezer at −140 °C until further processing. Needle biopsies at baseline and after the sixth vaccine Two to three needle biopsies (1.2 mm) were taken at baseline and after the sixth cycle of treatment when assessable from the same tumor lesion. One fragment was fixed with formalin and embedded in paraffin (FFPE); one to two fragments were used for expanding tumor-infiltrating T cells and establishing autologous tumor cell lines. Delayed-type hypersensitivity and generation of skin-infiltrating lymphocytes After six cycles of treatment, we performed intradermal injections of vaccine components without adjuvant and one control injection containing DMSO without peptide. Patients were injected with either a mixture of both IDO and PD-L1 peptides at all three injection sites or PD-L1 peptide, IDO peptide or a mixture of the two at the injection sites, respectively (Supplementary Table 6). Eight hours after injection, punch biopsies were resected from the three sites where the peptide was injected and transported immediately to the laboratory and cut into fragments of 1–2 mm3. SKILs were expanded to establish ‘young SKILs’ in CM medium consisting of RPMI 1640 with GlutaMAX, 25 mM HEPES, pH 7.2 (Gibco, 72400-021), IL-2 (100 or 6,000 IU ml−1) (Proleukin Novartis, 004184), 10% heat-inactivated Human AB Serum (Sigma-Aldrich, H4522-100ML), 100 U ml−1 penicillin, 1.25 µg ml−1 Fungizone (Bristol Myers Squibb, 49182) and 100 µg ml−1 streptomycin (Gibco, 15140-122). Half of the medium was replaced three times per week. Young SKILs from samples derived from IDO peptide, PD-L1 peptide and the mixture injection sites were further expanded in a small-scale version of the 14-d rapid expansion protocol as previously described51. Quantification of specific T cells in blood by ELISPOT For enumeration of vaccine-specific T cells in the peripheral blood, PBMCs from patients were stimulated with IDO or PD-L1 peptide in the presence of a low dose of IL-2 (120 U ml−1) for 7–13 d before being used in IFN-γ ELISPOT assays with 2.5–3.2 × 105 cells per well. Briefly, cells were placed in a 96-well PVDF membrane-bottomed ELISPOT plate (MultiScreen MSIPN4W50, Millipore) precoated with IFN-γ-capture antibody (clone 1-D1K, Mabtech). Diluted IDO or PD-L1 peptide stocks in DMSO were added at 5 µM; an equivalent amount of DMSO was added to control wells. PBMCs from each patient were set up in duplicate or triplicate for peptide and control stimulations. Cells were incubated in ELISPOT plates in the presence of the peptide for 16–18 h, after which they were washed off, and biotinylated secondary antibody (anti-human IFN-γ mAB, clone 7-B6-1, Mabtech) was added. After 2 h of incubation, unbound secondary antibody was washed off, and streptavidin-conjugated alkaline phosphatase (Mabtech) was added for 1 h. Next, unbound streptavidin-conjugated enzyme was washed off, and the assay was developed by adding BCIP/NBT substrate (Mabtech). ELISPOT plates were analyzed on the CTL ImmunoSpot S6 Ultimate V analyzer using ImmunoSpot software version 5.1. Responses were calculated as the difference between the average numbers of spots in wells stimulated with IDO or PD-L1 peptide and those from corresponding control wells. For detection of IDO and PD-L1 peptide responses ex vivo, PBMCs were thawed and rested overnight in medium containing DNase I (1 μg ml−1, Sigma-Aldrich, 11284932001) before being used in IFN-γ ELISPOT assays as described above. A total of 6–9 × 105 PBMCs per well were seeded. Statistical analysis of all ELISPOT responses was performed using the DFR method as described by Moodie et al. using RStudio software (RStudio Team, 2016, http://www.rstudio.com/)50. Vaccine-specific ELISPOT responses were defined as true if the difference between the spot count in control and peptide-stimulated wells was statistically significant according to DFR statistical analysis or, for samples performed in duplicate, if the spot count in peptide-stimulated wells was at least 2× the spot count in control wells50. Cancer cell lines and tumor conditioned medium Autologous melanoma cell lines were established from needle biopsies. Briefly, biopsies were chopped into small fragments and seeded in 24-well cultures in RPMI 1640 with GlutaMAX, 25 mM HEPES, pH 7.2 (Gibco, 72400-021), 10% heat-inactivated FBS (Life Technologies, 10500064), 100 U ml−1 penicillin, 1.25 µg ml−1 Fungizone (Bristol Myers Squibb, 49182) and 100 µg ml−1 streptomycin (Gibco, 15140-122). Established adherent melanoma tumor cell lines were cryopreserved at −140 °C in freezing medium containing FBS with 10% DMSO. PD-L1 and HLA-II expression on established tumor cell lines was assessed by flow cytometry staining with anti-PD-L1–PE-Cy7 (1:23 dilution, clone M1h1 (RUO), BD, 558017) and anti-HLA-II–FITC (1:23 dilution, clone Tu39 (RUO), BD, 555558) antibodies. To obtain TCM, established tumor cell lines were cultured in 175-cm2 Nunc cell culture flasks until 80–90% confluency was reached. The culture medium was then replaced with 20 ml fresh X-VIVO 15 with Gentamicin and Phenol Red (Lonza, BE02-060Q), medium with 5% heat-inactivated Human AB Serum (Sigma-Aldrich, H4522-100ML). After 24 h of incubation, TCM was collected and centrifuged to remove any resuspended cells, after which TCM was aliquoted, frozen and stored at −80 °C. The acute monocytic leukemia cell line MonoMac1 was obtained from the DSMZ—German Collection of Microorganisms and Cell Cultures (ACC 252) and cultured in RPMI 1640 with GlutaMAX, 25 mM HEPES, pH 7.2 (Gibco, 72400-021) and 10% heat-inactivated FBS. Isolation of autologous myeloid cells Autologous CD14+ cells were sorted from freshly thawed PBMCs using a magnetic bead-separation kit (Miltenyi Biotec, 130-050-201) according to the manufacturer’s instructions. Isolated CD14+ cells were used as targets in the IFN-γ ELISPOT assay directly after sorting or differentiated in vitro into tumor-associated macrophages by culturing with 1 ml fresh X-VIVO 15 medium with Gentamicin and Phenol Red (Lonza, BE02-060Q) and 5% heat-inactivated Human AB Serum, supplemented with 1 ml autologous TCM in 24-well plates for 2 d. Quantification of vaccine-specific T cells from DTH biopsy sites by ELISPOT SKILs expanded in vitro were rested in medium without IL-2 overnight before being used in the IFN-γ ELISPOT assay as described above to evaluate reactivity of skin-infiltrating T cells. Generation of IDO- and PD-L1-specific T cell cultures from PBMCs or SKILs IDO- or PD-L1-specific T cells were isolated from peptide-stimulated in vitro PBMC cultures on days 14–15 after stimulation or from SKIL cultures expanded in vitro. For specific T cell isolation, PBMCs or SKILs were stimulated with IDO or PD-L1 peptide, and cytokine-producing T cells were sorted using the IFN-γ or TNF-α Secretion Assay—Cell Enrichment and Detection kit (Miltenyi Biotec). Cytokine-production profile of PD-L1- and IDO-specific T cells by intracellular staining To assess the T cell cytokine-production profile, isolated and expanded IDO- and PD-L1-specific T cell cultures were stimulated for 5 h with peptide at 5 μM in a 96-well plate. One hour after the start of the incubation, anti-CD107a–PE (1:133 dilution, clone H4A3, BD Biosciences, 555801) antibody and BD GolgiPlug (1:1,000 dilution, BD Biosciences) were added. After a 5-h incubation, cells were stained using fluorescently labeled surface marker antibodies: anti-CD4–PerCP (1:5 dilution, clone SK3, 345770), anti-CD8–FITC (1:5 dilution, clone SK1, 345772) and anti-CD3–APC-H7 (1:20 dilution, clone SK7, 560275) (all from BD Biosciences). Dead cells were stained using FVS510 (1:250 dilution, BD Biosciences, 564406), followed by overnight fixation and permeabilization using eBioscience Fixation/Permeabilization buffers (00-5123-43, 00-5223-56) according to the manufacturer’s instructions. Cells were then stained intracellularly in eBioscience permeabilization buffer (00-8333-56) with anti-IFN-γ–APC (1:20 dilution, clone 25723.11, BD Biosciences, 341117) and anti-TNF-α–BV421 (1:100 dilution, clone Mab11, BD Biosciences, 562783) antibodies. Samples were analyzed on the FACSCanto II (BD Biosciences) using BD FACSDiva software version 8.0.2. The gating strategy is shown in Supplementary Fig. 3a. For assessing the reactivity against IDO expressing MonoMac1 cells, the appropriate amount of cancer cells was added to the IDO-specific T cells to obtain an effector:target ratio of 4:1 in the 96-well plate and T cell cytokine production was tested after 5 h of co-culture as described above. For HLA-blocking experiments, HLA-blocking antibodies (2 μg ml−1) were added directly into wells 30 min before the addition of peptide. Blocking antibodies used were HLA-DR (1:500 dilution, clone L243, Abcam, ab136320), HLA-DQ (1:500 dilution, clone SPV-L3, Abcam, ab23632) and HLA-DP (1:500 dilution, clone B7/21, Abcam, ab20897). To assess ex vivo T cell reactivity to IDO and PD-L1 peptides in patient PBMCs, cells were thawed and rested for 1–2 d in medium containing DNase I (1 μg ml−1, Sigma-Aldrich, 11284932001). PBMCs were then stimulated with peptide at 5 μM in a 96-well plate for 8 h. An hour after the addition of peptide, anti-CD107a–BV421 (3:50 dilution, clone H4A3, BD Biosciences, 328626) antibody and BD GolgiPlug (1:1,000 dilution, BD Biosciences) were added. Surface and intracellular staining was performed as described above. Antibodies used for surface staining were anti-CD3–PE-CF594 (0.8:30 dilution, clone UCHT1, BD Biosciences, 562280), anti-CD4–BV711 (1:30 dilution, clone SK3, BD Biosciences, 563028) and anti-CD8–Qdot605 (1:150 dilution, clone 3B5, Thermo Fisher, Q10009). Antibodies used for intracellular staining were anti-CD137–PE (1:20 dilution, clone 4B4-1 (RUO), BD Biosciences, 555956), anti-IFN-γ–PE-Cy7 (1.5:20 dilution, clone B27, BD Biosciences, 557643) and anti-TNF-α–APC (1:20 dilution, clone Mab11, BD Biosciences, 554514). Samples were acquired on the NovoCyte Quanteon (ACEA Biosciences) and analyzed using NovoExpress software version 1.4.1. To assess vaccine-specific T cell responses, background values observed in unstimulated PBMC samples were subtracted from values observed in peptide-stimulated conditions. Positive response value threshold was set at a difference of 0.2% from the background values. Based on this response cutoff, only TNF-α, CD107a and CD137 responses were detected in this assay. Statistical analysis comparing baseline with on-treatment or post-treatment cytokine profiles was performed using two-sided Wilcoxon matched-pairs signed-rank test. The gating strategy is shown in Supplementary Fig. 3b. siRNA-mediated PD-L1 and IDO1 silencing A stealth siRNA duplex for targeted silencing of PD-L1 (Invitrogen)52, a custom silencer select siRNA for targeted silencing of IDO1 (Ambion) and the recommended silencer select negative control (Ambion) siRNA for mock transfection were used. The stealth PD-L1 siRNA duplex consisted of the sense sequence 5′-CCUACUGGCAUU-UGCUGAACGCAUU-3′ and the antisense sequence 5′-AAUGCGUUCAGCAAAUGCCAGUAGG-3′. The three silencer IDO siRNA duplexes used were siRNA1 (sense sequence, 5′-ACAUCUGCCUGAUCUCAUATT-3′; antisense sequence, 5′-UAUGAGAUCAGGCAGAUGUTT-3′), siRNA2 (sense sequence, 5′-CCACGAUCAUGUGAACCCATT-3′; antisense sequence, 5′-UGGGUUCACAUGAUCG-UGGAT-3′) and siRNA3 (sense sequence, 5′-CGAUCAUGUGAACCCAAAATT-3′; antisense sequence, 5′-UUUUGGGUUCACAUGAUCGTG-3′). For PD-L1 or IDO1 silencing, cancer cells were electroporated with 0.025 nmol of each siRNA duplex as previously described53. For PD-L1-silencing experiments, cancer cells were treated with IFN-γ (500 U ml−1, PeproTech) 1 h after electroporation. Electroporated cells were used as target cells in ELISPOT and ICS assays 24 h or 48 h after siRNA electroporation. RT–qPCR Total RNA was extracted using the RNeasy Plus Mini kit (Qiagen, 74134) following the manufacturer’s instructions. RNA concentration was quantified using a NanoDrop 2000 (Thermo Fisher Scientific), and a total of 1,000 ng RNA was reverse transcribed with the High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems, 4368814) using 1,000 ng input RNA for transcription. Real-time qPCR analyses were performed using the TaqMan Gene Expression Assay on a Roche LightCycler 480 instrument. RT–qPCR assays were performed with a minimum of three technical replicates and analyzed using the ΔCt method as described in Bookout et al.54 with normalization of IDO1 expression (primer ID, Hs00984148_m1) or PD-L1 expression (primer ID, Hs001125296_m1) to the expression level of the housekeeping gene POLR2A (primer ID, Hs00172187_m1) and the control sample (mock). For low-concentration samples with no amplification, Ct was set to 40. Controls without reverse transcriptase were used as controls for specific amplifications. P values were determined using two-tailed parametric t-tests. BRAF-mutational status and PD-L1 status at baseline in all patients A library of historical FFPE biopsies were assessable for all patients and analyzed locally at Herlev and Gentofte University Hospital by experienced pathologists for BRAF status and PD-L1 expression on tumor cells. The BRAF analysis was carried out with real-time PCR using the EntroGen BRAF Mutation Analysis Kit II (BRAFX-RT64, CE-IVD) to specifically detect mutations corresponding to V600D, V600E and V600K in the BRAF protein. PD-L1 status was assessed using the monoclonal rabbit anti-PD-L1 antibody (clone 28.8, PD-L1 IHC 28-8 pharmDx) in FFPE biopsies following the manufacturer’s instructions. Patients were considered positive for PD-L1 with expression levels ≥1% and negative for PD-L1 with expression levels <1%. Human leukocyte antigen type Blood samples from all 30 patients were genotyped for class I (HLA-A, HLA-B, HLA-C) and three class II types (HLA-DRB1, HLA-DQA1, HLA-DQB1) using LinkSēq HLA Typing Kits (Thermo Fisher, 1580C). These test kits are based on real-time PCR using allele-specific exponential amplification (sequence-specific primers) followed by melting curve analyses. Immunohistochemistry simplex: Immunoscore CR IHC staining was performed using a qualified Ventana Benchmark XT with four different steps: (1) antigen retrieval; (2) staining with the following primary antibodies: anti-CD3 antibody (clone HDX2; provider, HalioDx; HD-FG-000013; 10931065/10636667; concentration, 0.25 µg ml−1), anti-CD8 antibody (clone HDX3; provider, HalioDx; HD-FG-000019; 10931069/10337710/10639301; concentration, 0.5 µg ml−1), anti-IDO monoclonal antibody (clone VINC3IDO; provider, Thermo Fisher Scientific; 14-9750-82; E25003-101; concentration, 0.05 µg ml−1), anti-HLA class 1 ABC antibody (clone, EMR8-5; provider, Abcam; ab70328; GR3248333/GR3186494; concentration, 0.5 µg ml−1), anti-HLA-DR/DP/DQ/DX antibody (clone CR3/43; provider, Santa Cruz; sc-53302; L1714; concentration, 1 µg ml−1) and anti-PD-L1 antibody (clone HDX3; provider, HalioDx; HD-FG-000035; 106312810/106312816; concentration, 3.3 µg ml−1); (3) detection with a secondary antibody using the ultraView Universal DAB Detection kit; and (4) counterstaining using Hematoxylin & Bluing Reagent (staining of cellular nuclei). Control slides were systematically included in each staining run to permit quality control of the obtained measurements. Following coverslipping, slides were scanned with the NanoZoomer-XR to generate digital images (20×). Two consecutive slides were specifically used to perform Immunoscore CR TL staining of CD3+ and CD8+ cells. Digital pathology of T lymphocytes The digital pathology for Immunoscore CR TL allowed the quantification of positive cells (in cells per mm2) into the core tumor and invasive margin if present. Each sample was analyzed using the HalioDx Digital Pathology Platform. Digital pathology and pathologist analysis of IDO Digital pathology for IDO allowed the quantification of stained area (in mm2) into the whole tumor. Each sample was analyzed using the HalioDx Digital Pathology Platform. Analysis of IDO+ cells was performed by a pathologist. Results were expressed as an H score from 0 to 300. The score is obtained by the formula 3 × percentage of cells with strong staining + 2 × percentage of cells with moderate staining + percentage of cells with weak staining. Digital pathologist analysis of MHC-I and MHC-II Analysis of MHC-I+ and MHC-II+ cells was performed by a pathologist. Results are expressed as a percentage of positive tumor cells and a percentage of positive cells in the stroma. Digital pathology Immunoscore immune checkpoint (CD8+ and PD-L1+) The Immunoscore CR IC test allowed the quantification of CD8++ cell density in the whole tumor and a CD8+-centered proximity index (which corresponds to the percentage of CD8++ cells that have at least one PD-L1+ cell in the neighborhood) at different cutoff distances (20 µm, 40 µm, 60 µm and 80 µm). Pathologist analysis of PD-L1 A pathologist performed analysis of PD-L1+ cells. Positivity of a viable tumor cell was considered when partial or complete cell membrane staining was observed (more than 10% of the tumor cell membrane). Results were expressed as a percentage. NanoString RNA profiling and Immunosign RNA was extracted from FFPE tissues using Qiagen RNeasy FFPE extraction kits (Qiagen). Annotations from the pathologist performing H&E staining were used to guide removal of normal tissue from the slides by macrodissection before nucleic acid extraction, which occurred after tissue deparaffinization and lysis. Each extracted RNA sample was independently quantified using a NanoDrop spectrophotometer (NanoDrop Technologies) and qualified (Agilent Bioanalyzer). Degradation was quantified as the percentage of RNA fragments smaller than 300 bp using the RNA 6000 Nano kit (Agilent Bioanalyzer). Good sample quality was defined as less than 50% of RNA fragments being 50–300 bp in size. RNA expression profiling was performed using the nCounter PanCancer Immune Profiling Panel from NanoString (NanoString Technologies). The PanCancer Immune Profiling Panel contains 776 probes and is supplemented with six genes to complete HalioDx Immunosign targets. Hybridization was performed according to the manufacturer’s instructions. Hybridized probes were then purified and immobilized on a streptavidin-coated cartridge using the nCounter Prep Station (NanoString Technologies). Data collection was carried out on the nCounter Digital Analyzer (NanoString Technologies) following the manufacturer’s instructions to count individual fluorescent barcodes and quantify target RNA molecules present in each sample. For each assay, a scan of 490 fields of view was performed. Raw data from the NanoString nCounter were processed using NanoString recommendations. Quality control enables to retain data of good quality with a binding density that ranges between 0.05 and 2.25. The linearity of positive controls was checked using the R2 value of regression between the counts and the concentrations of positive controls. Samples that showed R2 < 0.75 were flagged and removed from the analysis. The background was removed using a thresholding method at the mean + 2 s.d. of negative controls. Raw counts were normalized using a positive normalization factor. Samples showing positive normalization factors outside the range of 0.3–3 were removed from the analysis. A second normalization was performed using the housekeeping gene normalization factor. Only the most stable housekeeping genes were selected for this normalization step using the variance-versus-mean relationship. All samples showing a normalization factor outside the range of 0.1–10 were removed from the analysis. All statistical analyses were performed on normalized counts using R software (version 2.6.2, 2019-12-12). T cell receptor variable β chain sequencing To track longitudinal immune responses to therapy, genomic DNA was extracted from longitudinal pre- and post-treatment PBMCs (five patients), pre- and post-treatment biopsies (five patients) (both FFPE and treated with RNAlater) and IDO- and PD-L1-specific T cell cultures from PBMCs (five patients) or SKILs (one patient). Three clonal (two IDO- and one PD-L1-) specific cultures were also sequenced to confirm clonal purity of cultures. DNA from PBMCs or RNAlater-treated biopsies was extracted with the DNeasy Blood and Tissue kit (Qiagen, 69504), DNA from sorted IDO and PD-L1-specific T cells from either PBMCs or SKILs was extracted using the QIAamp DNA Micro kit (Qiagen, 565304), and DNA from PFFE biopsies was extracted using the Maxwell RSC DNA FFPE kit (Promega, AS1450). Immunosequencing of the CDR3 regions of human TCRβ chains was performed using the immunoSEQÒ Assay (Adaptive Biotechnologies). Extracted genomic DNA was amplified in a bias-controlled multiplex PCR, followed by high-throughput sequencing. Sequences were collapsed and filtered to identify and quantitate the absolute abundance of each unique TCRβ CDR3 region for further analysis as previously described55–57. Statistical analyses of TCRβ sequencing results Two quantitative components of diversity were compared across samples in this study. First, Simpson clonality was calculated on productive rearrangements by ∑i=1Rpi2, where R is the total number of rearrangements and pi is the productive frequency of rearrangement i. Values of Simpson clonality range from 0 to 1 and measure how evenly receptor sequences (rearrangements) are distributed among a set of T cells. Clonality values approaching 0 indicate a very even distribution of frequencies, whereas values approaching 1 indicate an increasingly asymmetric distribution in which a few clones are present at high frequencies. Second, sample richness was calculated as the number of unique productive rearrangements in a sample after computationally downsampling to a common number of T cells to control for variation in sample depth or T cell fraction. Repertoires were randomly sampled without replacement five times, and we report the mean number of unique rearrangements. The T cell fraction was calculated by taking the total number of T cell templates and dividing by the total number of nucleated cells. The total number of nucleated cells was derived from reference genes using the immunoSEQ Analyzer version 3. To identify enriched vaccine clones in each patient, rearrangement frequencies in their baseline PBMCs and each IDO/PD-L1-sorted T cell sample were compared using a binomial distribution framework as previously described58. In brief, for each clone, we performed a two-sided test to determine whether frequencies were the same in the patient’s periphery and a PD-L1- or IDO-specific T cell sample. The Benjamini–Hochberg procedure was used to control the false discovery rate at 0.01 (ref. 59). Clonal expansion in post-treatment samples was similarly assessed using this differential abundance framework, but an IDO/PD-L1-specific T cell sample was replaced with a post-treatment series sample. In biopsies, the six series frequencies were compared to those of baseline tissue. Lastly, vaccine-associated clones were tracked in each PBMC and tissue sample by summing the frequency of each rearrangement enriched in either PD-L1- or IDO-specific T cells. All statistical analyses were performed in R version 3.6.1. Reporting Summary Further information on research design is available in the Nature Research Reporting Summary linked to this article. Online content Any methods, additional references, Nature Research reporting summaries, source data, extended data, supplementary information, acknowledgements, peer review information; details of author contributions and competing interests; and statements of data and code availability are available at 10.1038/s41591-021-01544-x. Supplementary information Supplementary Information Supplementary Figs. 1–3 and Tables 1–6 Reporting Summary Extended data Extended Data Fig. 1 Mode of Action and treatment plan. a) Anticipated mechanism of action of the combination therapy of an IDO/PD-L1 derived peptide vaccine and nivolumab (anti-PD1). 1) The IDO/PD-L1 peptide vaccine is administered subcutaneously (s.c) and nivolumab is administered intravenous (IV). 2) The vaccines peptides are phagocytosed by an antigen presenting cell and presented to IDO and PD-L1 specific T cells, which are activated. 3) The activated T cells migrates to the tumor site where they attack both immune-suppressive cells and tumour cells expressing IDO and/or PD-L1 leading to cytokine production and a pro-inflammatory tumour microenvironment. 4/5) Enhanced tumour killing by both IDO/PD-L1 specific T cells and tumor specific cytotoxic T cells due to PD-1 blockade. Created with BioRender.com b) Treatment plan. After written informed content patients were screened. Before treatment start a baseline PET/CT scan was performed, baseline blood sample for research use and if assessable a baseline needle biopsy. Patients were treated with the IDO/PD-L1 peptide vaccine subcutaneously biweekly for the first 6 injections and thereafter every fourth week for a maximum of 15 vaccines. Nivolumab was administered in parallel biweekly (3mg/kg) up to 24 series. If patients needed subsequent nivolumab after ended vaccination regiment they were treated with 6 mg/kg every fourth week up to two years. Needle biopsy and delayed type hypersensitivity (DTH) was performed after 6 series of treatment if assessable. PET/CT scans was performed every third month. Extended Data Fig. 2 Progression free survival and overall survival in matched historical control group and vaccine injection site reaction. a) Kaplan-Meier curve of progression free survival with corresponding 95% confidence intervals in the matched historical control group (n=74). Patients were matched on BRAF-status, PD-L1-status, age, gender, M-stage and LDH level). b) Kaplan-Meier curve of overall survival with corresponding 95% confidence intervals in the matched historical control group (n=74). c) Images of injection site reaction in patient MM20 (CR) after 11 vaccination show redness, rash and granuloma at injection site. Extended Data Fig. 3 Vaccine specific responses in blood. a) Heatmap of specific (background has been subtracted) IDO and PD-L1 responses in PBMCs at baseline, series 3, 6, 12, 18 and 24 measured by in vitro IFN-γ Elispot assay show fluctuations in the blood during treatment (n=30). b) Heatmap of specific (background has been subtracted) IDO and PD-L1 responses in PBMCs at baseline and 3 and 6 months after last vaccine measured by in vitro IFN- γ Elispot assay (n=30). 2.5–3.2x105 cells per well were used. c) Vaccine associated clones were tracked in the blood by summing the frequency of each rearrangement enriched in either IDO or PD-L1 T cells. Extended Data Fig. 4 Ex vivo vaccine specific responses in blood. Heatmaps of detected specific (background has been subtracted) IDO (a) and PD-L1 (b) responses in PBMCs at baseline and on/after treatment as measured by IFNγ ELISPOT (n=25). C) Example of well images of ex vivo ELISPOT wells for three different patients (n=3). 6-9x105 cells per well were used. Extended Data Fig. 5 CD4 and CD8 vaccine specific T cell responses in blood. Top: heatmaps of IDO specific CD4 (left) and CD8 (right) T cell responses in PBMCs at baseline and on/post treatment. Bottom: heatmaps of PD-L1 specific CD4 (left) and CD8 (right) T cell responses. Responses quantified by flow cytometry by an increased expression of CD107α, CD137 and TNFα after 8h peptide stimulation. Values represent specific responses after background values have been subtracted (n=21). Statistical analysis were performed using two-sided Wilcoxon matched-paired rank test. Extended Data Fig. 6 Pro-inflammatory profiles of sorted CD4 and CD8 T cells from blood. a/b/c/d) Percentage of in vitro stimulated and sorted PD-L1 and IDO specific CD4 and CD8 T cells secreting cytokines. Extended Data Fig. 7 Vaccine specific responses in skin. a) IDO and PD-L1 specific T cell responses in SKILs after 6 series of treatment measured by IFN-γ Elispot assay (n=13). SKILs were grown from DTH injection with either IDO peptide, PD-L1 peptide or a mix as presented by different blue colours. * Responses were calculated as the difference between average numbers of spots in wells stimulated with IDO or PD-L1 peptide (triplicates) and corresponding control (DMSO) and statistical analyses of Elispot responses were performed using distribution-free resampling method (Moodie et al.). DR: Not statistically confirmed response due to replicate number but number of spots in peptide wells are two times higher than control wells (DMSO). NS: No significant response and no DR. b/c/d) Percentage of cytokine secreting/CD107a+ CD4+ and CD8+ IDO and PD-L1 specific T cells in response to in vitro peptide stimulation by flow cytometry. e) Skin infiltrating PD-L1 specific T cell clones also found in biopsy in patient MM01. TCR sequencing was performed on a PD-L1 specific T cell culture generated from DTH area on the lower back injected with PD-L1 peptide on patient MM01. Bars show the frequency of top 25 clones in the culture with which indicates a high Simpson clonality of 0.43. f) Tracking the frequency of the top five skin infiltrating PD-L1 specific clones in tumour before and after treatment. Extended Data Fig. 8 T cell changes in blood after treatment. T cell fraction, TCR clonality and repertoire richness in blood and peripherally expanded TCR clones in both responding and non-responding patients. a) T cell fraction in peripheral blood of five patients at baseline, series 3, 6 and 12 by TCR sequencing. T cell fraction was calculated by taking the total number of T cell templates and dividing by the total number of nucleated cells (n=5.) b) TCR clonality in peripheral blood of five patients at baseline, series 3, 6 and 12 by TCR sequencing (n=5). Simpson clonality measures how evenly TCR sequences are distributed amongst a set of T cells where 0 indicate even distribution of frequencies and 1 indicate an asymmetric distribution where a few clones dominate. c) TCR repertoire richness in peripheral blood of five patients at baseline, series 3, 6 and 12 by TCR sequencing (n=5). TCR repertoire richness report the mean number of unique rearrangements. d) Bar-chart chart representing dynamics of the expanded T cell clones. Coloured bars represent peripherally expanded clones in five patients at series 3, series 6 and series 12 as compared to the baseline PBMC samples (n=5). Light gray bar represents peripherally expanded clones that are present in baseline biopsy samples while dark gray bar represents peripherally expanded clones that are present in the post-treatment biopsy (after 6 series). e) Frequency of the dominant TCR β chain in clonal PD-L1 and IDO specific cultures as determined by CDR3 sequencing. Extended Data Fig. 9 Profiling of genes relevant to T cell activation, cytokines and exhaustion markers on pre- and post-treatment biopsies in two patients. a) RNA expression profiling of genes related to T cell activation was performed using NanoString nCounter (n=2). b) RNA expression profiling of genes related to cytokine activity was performed using NanoString nCounter (n=2). c) RNA expression profiling of genes related to checkpoint inhibitors was performed using NanoString nCounter (n=2). Extended Data Fig. 10 Treatment induced upregulation of PD-L1, IDO, MHC I and MHC II and distance between CD8 T cells and PD-L1 expressing cells. a) IHC on 4 paired biopsies stained for CD3 and CD8 T cells, PD-L1, MHC I and MHC II on tumor cells/mm2 (sum in the validated area) and IDO H-score (from 0 to 300). H-score is the expression of IDO on both immune and tumor cells: The score is obtained by the formula: 3 x percentage of cells with a strong staining + 2 x percentage of cells with a moderate staining + percentage of cells with a weak staining (n=4). b) Distance in µm between CD8+ T cells and PD-L1+ stained cells on five baseline biopsies detected by IHC (n=5). Extended data is available for this paper at 10.1038/s41591-021-01544-x. Supplementary information The online version contains supplementary material available at 10.1038/s41591-021-01544-x. Acknowledgements We thank all patients and their relatives for being a part of the trial. We thank Ö. Met, M. Jonassen, S. Ullitz Færch, B. Saxild, S. Wendt and C. Grønhøj for technical support. We thank M. Cumberbatch for input with translational analysis. We thank the nurses at clinic 5 and the head of the Oncology Department at Herlev and Gentofte Hospital, L. Sengeløv. The study was funded through a research funding agreement between IO Biotech and the CCIT-DK, Herlev Hospital and the Oncology Department at Herlev Hospital. Author contributions I.M.S. and M.H.A. conceived the study. J.W.K. wrote the protocol with input from I.M.S., M.H.A. and M.-B.Z. J.W.K. and C.L.L. were responsible for patient treatment and patient care with help from R.B.H. and consulting from I.M.S., E. Ellebaek and M.D. I.M.S., E. Ellebaek and M.D. were responsible for patient recruitment. J.W.K. and C.L.L. were responsible for coordinating trial procedures and collecting data and samples. J.W.K. and C.L.L. were clinical investigators, and I.M.S. was a sponsor. J.W.K. and C.L.L. analyzed and interpreted clinical data with support from I.M.S. and E. Ehrnrooth. T.W.K. performed statistical analyses. H.W.H. helped with clinical imaging. J.W.K., E.M., C.O.M., S.M.A., S.E.W.-B., M.O.H. and A.W.P. analyzed and interpreted translational data with support from M.H.A. and I.M.S. Funding was acquired by I.M.S., M.H.A. and M.-B.Z. J.W.K. wrote the manuscript with input from I.M.S., M.H.A., E.M. and A.W.P. All authors reviewed the manuscript, interpreted data and approved the final version. Data availability This clinical trial was registered at https://www.clinicaltrials.gov/ before patient enrollment (clinical trial identifier NCT03047928). Each vaccine was composed of 100 µg IO102, a 21-amino-acid peptide (DTLLKALLEIASCLEKALQVF) from the IDO peptide, and 100 µg IO103, a 19-amino-acid peptide (FMTYWHLLNAFTVTVPKDL) from the signal peptide of PD-L1. TCR sequencing data are available from Adaptive Biotechnologies. Upon request, the CCIT-DK office will provide a username and a password to access the designated data within approximately 2–4 weeks (https://www.herlevhospital.dk/ccit-denmark/find-us/Sider/Contact-information.aspx). All requests for the remaining data including raw data and analyzed data and materials will, within a reasonable time frame, be reviewed by the CCIT-DK office (https://www.herlevhospital.dk/ccit-denmark/find-us/Sider/Contact-information.aspx) to verify whether the request is subject to any intellectual property or obligations. Patient-related data not included in the paper were generated as part of clinical trials and may be subject to patient confidentiality. Any data and materials that can be shared will be released via a material-transfer agreement. The following database was used in the study: https://research.regionh.dk/da/publications/the-danish-metastatic-melanoma-database-dammed(32749d99-095f-4cae-b5de-769bae27f01e).html. Competing interests M.H.A. is named as an inventor on various patent applications relating to therapeutic uses of IDO and PD-L1 peptides. These patent applications are assigned to the company IO Biotech, which is developing immune-modulating cancer treatments. M.H.A. is a founder, shareholder and advisor for IO Biotech. I.M.S. is a cofounder, shareholder and advisor for IO Biotech. I.M.S. has an advisory board relationship with or lectured for Roche, Novartis, MSD, Celgene, Incyte, TILT Bio, Pfizer and BMS AstraZeneca and has received limited grants for translational research from BMS, Roche and Novartis. M.-B.Z. is the CEO, founder and shareholder at IO Biotech. E.M., A.W.P. and E. Ehrnrooth are employees at IO Biotech. E. Ellebaek has received honoraria from BMS, Pierre Fabre, Roche and Kyowa Kirin and travel support from MSD. M.D. has received honoraria from Genzyme, MSD, BMS, Roche and Novartis and travel support from Novartis, MSD, BMS, Roche and Pfizer. The remaining authors declare that they have no conflict of interest. Peer review information Nature Medicine thanks George Coukos, Michael Schell and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. Javier Carmona was the primary editor on this article and managed its editorial process and peer review in collaboration with the rest of the editorial team. Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. These authors contributed equally: Julie Westerlin Kjeldsen, Cathrine Lund Lorentzen. Change history 3/8/2022 A Correction to this paper has been published: 10.1038/s41591-022-01771-w

NIVOLUMAB

DrugsGivenReaction

CC BY

34887574

2021-12-09

Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Urosepsis'.

A phase 1/2 trial of an immune-modulatory vaccine against IDO/PD-L1 in combination with nivolumab in metastatic melanoma. Anti-programmed death (PD)-1 (aPD1) therapy is an effective treatment for metastatic melanoma (MM); however, over 50% of patients progress due to resistance. We tested a first-in-class immune-modulatory vaccine (IO102/IO103) against indoleamine 2,3-dioxygenase (IDO) and PD ligand 1 (PD-L1), targeting immunosuppressive cells and tumor cells expressing IDO and/or PD-L1 (IDO/PD-L1), combined with nivolumab. Thirty aPD1 therapy-naive patients with MM were treated in a phase 1/2 study ( https://clinicaltrials.gov/ , NCT03047928). The primary endpoint was feasibility and safety; the systemic toxicity profile was comparable to that of nivolumab monotherapy. Secondary endpoints were efficacy and immunogenicity; an objective response rate (ORR) of 80% (confidence interval (CI), 62.7-90.5%) was reached, with 43% (CI, 27.4-60.8%) complete responses. After a median follow-up of 22.9 months, the median progression-free survival (PFS) was 26 months (CI, 15.4-69 months). Median overall survival (OS) was not reached. Vaccine-specific responses assessed in vitro were detected in the blood of >93% of patients during vaccination. Vaccine-reactive T cells comprised CD4+ and CD8+ T cells with activity against IDO- and PD-L1-expressing cancer and immune cells. T cell influx of peripherally expanded T cells into tumor sites was observed in responding patients, and general enrichment of IDO- and PD-L1-specific clones after treatment was documented. These clinical efficacy and favorable safety data support further validation in a larger randomized trial to confirm the clinical potential of this immunomodulating approach. pmcMain Despite remarkable advances in the treatment of MM with immune checkpoint inhibitors (ICIs) targeting PD-1 and cytotoxic T lymphocyte antigen 4 (CTLA-4), around half of patients are resistant to ICI monotherapy1. The combination of anti-CTLA-4 (aCTLA-4) and aPD1 therapy is to date the most effective therapy resulting in a response rate of around 60%; however, 50% of the patients also develop severe adverse events2,3. Therefore, an equally effective but less toxic treatment is highly needed. Several approaches to enhance ICI efficacy are currently being investigated, such as other ICIs, T cell therapy with tumor-infiltrating T cells or innate immunity stimulators such as Toll-like receptor 9 agonists4–6. Treating cancer patients with vaccines that stimulate a targeted immune response is another attractive approach, with very few side effects observed thus far in combination immunotherapy studies7,8. Immune-modulatory vaccines targeting tumoral immune escape mechanisms offer a new, generalizable strategy compared to patient-specific neoantigen cancer vaccines7,9. The immune-modulatory vaccine strategy in this clinical trial is based on the finding of circulating cytotoxic T cells specific to IDO and PD-L1 in the blood of patients with cancer and, to a lesser extent, in healthy donors. IDO- and PD-L1-specific CD8+ T cells can directly recognize and kill IDO+ and/or PD-L1+ tumor cells and likewise recognize and kill non-malignant cells that express their cognate targets. Furthermore, IDO- and PD-L1-specific CD4+ T cells release pro-inflammatory cytokines in response to IDO- or PD-L1-expressing target cells. IDO and PD-L1 are expressed not only by melanoma cells but also by many other cell types in the tumor microenvironment (TME), which differentiates these antigens from traditional tumor antigens used in other studies10–15. Activation of IDO/PD-L1-specific T cells by vaccination can therefore restrict the range of immunosuppressive signals mediated by immunosuppressive cells and thereby revert the TME from an immune hostile to an immune friendly environment. In animal models of cancer, vaccinations with IDO epitopes resulted in anti-tumor therapeutic effects that were correlated with reductions in IDO expression in myeloid cell populations within the TME16. The IDO/PD-L1 immune-modulating vaccine may lead to a translatable strategy for improving the efficacy of aPD1 therapy through activation of specific T cells. We hypothesize that the IDO/PD-L1 vaccine attracts T cells into the tumor, which induces type 1 helper T (TH1) cell inflammation and reverts the TME into an immune-permissive site, thereby turning the tumor ‘hot’. This would also upregulate PD-L1 expression in cancer and immune cells, generating more susceptible targets to aPD1 therapy (Extended Data Fig. 1a). This theory was confirmed in a mouse model in which aPD1 therapy and IDO vaccination show synergistic effects16. In this phase 1/2 clinical trial, MM1636, patients with MM received a combination of the IDO/PD-L1 (IO102/IO103) peptide vaccine with the adjuvant Montanide and the aPD1 antibody nivolumab. Patients were included in three cohorts: 30 aPD1 therapy-naive patients (cohort A), ten aPD1 therapy-refractory patients (cohort B, de novo resistance) and ten patients who progressed after aPD1 therapy (cohort C, acquired resistance). Here, we report results from cohort A. The vaccine was given biweekly for the first six administrations and thereafter every 4th week. A maximum of 15 vaccines were administered. Nivolumab was given in parallel, biweekly (3 mg per kg) or every 4th week (6 mg per kg) for up to 2 years (Extended Data Fig. 1b). The primary objective was safety and feasibility. Secondary objectives were immunogenicity and clinical efficacy. Results Patients and treatment Thirty patients were enrolled from December 2017 to June 2020. None of the 30 patients dropped out of the study; all received at least three cycles of therapy (Supplementary Fig. 1). At the current database lock (5 October 2020), six patients were still on treatment in the trial. Of the 24 patients who were not on trial treatment at data cutoff, two are still receiving nivolumab monotherapy (6 mg per kg every 4 weeks). Reasons for stopping treatment for the remaining 22 patients included disease progression (37%), toxicity (20%), maximum benefit or complete response (CR) confirmed on two consecutive scans (17%) or completing 2 years of treatment (7%). For the 24 patients who were not on trial treatment at data cutoff, the mean number of vaccinations was 10.5 (range, 3–15 vaccinations). Thirteen of these 24 patients continued nivolumab (6 mg per kg, every 4 weeks) as a standard of care. Nine patients received subsequent therapy after progression (Supplementary Table 1). Baseline characteristics are shown in Table 1. The mean age was 70 years; 37% of patients had elevated lactate dehydrogenase (LDH) levels, 60% were stage M1c, 37% had BRAF mutations, and 43% were negative for PD-L1 (<1%). A total of three patients (10%) had received prior ipilimumab therapy. No patients had brain metastasis (Supplementary Table 2).Table 1 Baseline patient characteristics (n = 30) Characteristic Number (%) Mean age, years (range) 70 (46–85) Sex, male 16 (55%) ECOG PS 0 26 (87%) LDH levels  ≤ULN 19 (63%)  >ULN 11 (37%) M stage (AJCC-8)  M1a 6 (20%)  M1b 6 (20%)  M1c 18 (60%) Number of lesion sites  1 6 (20%)  2–3 17 (57%)  >3 7 (23%) Liver metastases present 10 (33%) BRAF status  Mutant 11 (37%)  Wild type 19 (63%) PD-L1  <1% 13 (43%)  >1% 17 (57%) Previous systemic therapy  Ipilimumab 3 (10%)  No 27 (90%) AJCC-8, eighth edition of the American Joint Committee on Cancer; ECOG PS, Eastern Cooperative Oncology Group performance status; ULN, upper limit of normal. Notable clinical responses to the combination therapy Thirty patients with MM were treated with the IDO/PD-L1 vaccine and nivolumab according to the trial protocol. By investigator review, the ORR reached 80% (CI, 62.7–90.5%), with 43% of patients (CI, 27.4–60.8%) achieving a CR and 37% (CI, 20.9–54.5%) reaching a partial response (PR) as the best overall response, while 20% experienced progressive disease (PD) according to Response Evaluation Criteria in Solid Tumors (RECIST) 1.1 (Fig. 1a). Two of the patients with a PR did not have a confirmatory scan stating PR on two consecutive scans. Early onset of response was frequent, with 22 of 30 patients having an objective response at the first evaluation (after 12 weeks on treatment). Median times to PR and CR were 75 d (range, 54–256 d) and 327 d (range, 73–490 d), respectively (Fig. 2a–c).Fig. 1 Clinical response. a, Pie charts with percent ORR, CR, PR and PD according to RECIST 1.1 by investigator review of all patients (n = 30), PD-L1+ patients (>1%, (n = 17)) and PD-L1− patients (<1%, n = 13)), respectively. Two-sided CIs (95%) were constructed using the Clopper–Pearson method. b, Treatment effect in MM1636 compared with a matched historical control group from the DAMMED database (n = 74). Patients in MM1636 (n = 29) were matched with the exact same combination variable according to age (≤70 years, >70 years), sex, LDH levels (normal, elevated), M stage (M1a, M1b, M1c), BRAF status (wild type, mutated) and PD-L1 status (<1%, ≥1%). Estimates for treatment effects were calculated by weighted logistic regression analyses and weighted Cox proportional hazard model. Bar height indicates the estimated response rate; tops of bars are centers for error bars. Odds ratios (OR), response rates and their corresponding 95% CIs were extracted from the regression model. All P values were two sided, and P values below 0.05 were considered statistically significant. c, Best change in the sum of target lesion size compared with that at baseline (n = 30). The horizontal line at −30 shows the threshold for defining an objective response in the absence of non-target disease progression or new lesions according to RECIST 1.1. Two patients with 100% reduction in target lesion size had non-target lesions present. White stars, six patients had normalization (<10 mm) of fluor-18-deoxyglucose (FDG)-negative lymph nodes (at baseline, lymph nodes were >1.5 cm and FDG+) and 100% reduction of non-lymph node lesions and are considered to have had a CR (green bar). Black star, one patient (MM29) was considered to have had a PR (blue bar), although he had not reached a −30% change in target lesion size. The patient had a single measurable 13-mm lung metastasis at baseline and multiple biopsy-verified cutaneous metastases on the left crus (not detectable by positron-emission tomography–computed tomography (PET–CT) at baseline). The best change in target lesion size was 10 mm, and a post-treatment biopsy from the cutaneous metastases showed no sign of malignancy. Thus, overall, the patient was classified as having a PR. d, Kaplan–Meier curve of the response duration in the 24 patients with an objective response. e, Kaplan–Meier curve of PFS in all 30 treated patients. f, Kaplan–Meier curve of OS in all 30 treated patients. Fig. 2 Clinical response. a, Swimmer plot showing response duration and time to response according to RECIST 1.1 for all treated patients (n = 30). Triangles indicate first evidence of PR, while squares indicate first evidence of CR. Closed circles indicate time of progression. Arrows indicate ongoing responses. Patient MM18 died due to nivolumab-induced side effects. b, Spider plot showing response kinetics in all treated patients (n = 30). Red squares indicate time of progression. c, PET–CT images of patient MM42 before and after treatment (after 12 series of treatment) showing FDG metabolism in target lesions. SC, subcutaneous. The ORR among PD-L1+ (>1% (clone 28.8)) patients (n = 17) was 94.1% (CI, 73–99.7%) and 61.5% (CI, 35.5–82.3%) in PD-L1− patients (n = 13) (Fig. 1a). Objective responses were observed in patients irrespective of human leukocyte antigen (HLA) genotype (Supplementary Fig. 2). Clinical response data were validated by blinded independent external review, in which an ORR of 76.6% (CI, 57.7–90.1%) was reported, with 53.3% of patients achieving a CR, 23.3% achieving a PR and 3.3% experiencing stable disease. Comparisons between investigator review and external review are outlined in Supplementary Table 3. To examine whether the very high response rate should be attributed to nivolumab or to the vaccine, we retrieved clinical response information from a matched historical control group using the Danish Metastatic Melanoma Database (DAMMED) from contemporaneously treated patients with stage 3–4 melanoma who received aPD1 monotherapy17. Patients were matched with the exact same combination variable according to age, sex, PD-L1 status, BRAF status, LDH level and M stage (those at stage M1d were excluded from the control group (no patients with brain metastasis)). Matched controls were identified for 29 patients, and the ORR of 79.3% (CI, 61.0–90.4%) observed in MM1636 was found to be significantly higher (P < 0.0012) than that in the matched control group in which an ORR of 41.7% (CI, 31.0–53.3%) was reached. Furthermore, of the 29 patients in MM1636, a significantly (P < 0.0017) higher percentage (41.4% (CI, 25.2–59.6%)) of patients achieved CR in MM1636 than that (12% (CI, 6.3–21.6%)) in the matched historical control group. ORR and complete response rate (CRR) in the matched historical control group were comparable to those of patients treated in randomized phase 3 pivotal trials with aPD1 monotherapy18 (Fig. 1b). The treatment leads to prolonged PFS At data cutoff, the median duration of response had not been reached, with 87% of all responding patients being free from progression at 12 months (Fig. 1d). Patients were followed for up to 35 months with a median follow-up time of 22.9 months (CI, 14.9–26.2 months). OS and PFS were calculated from the first day of treatment to death or progression or to the date of the last follow-up (5 October 2020). The median PFS (mPFS) for all treated patients was 26 months (CI, 15.4–69 months) and was not reached for responding patients (Fig. 1e). The median OS was not reached at the data cutoff. OS at 12 months was 81.6% (CI, 61.6–92%) (Fig. 1f). One patient (MM18) with a CR died from nivolumab-related severe adverse events; the remaining patients died because of metastatic melanoma. For comparison, the mPFS was 8.3 months (CI, 5.5 months–NR (not reached)) in the matched historical control group (n = 74), while the median OS was 23.2 months (CI, 23.2 months–NR). (Extended Data Fig. 2a,b). The combination of the IDO/PD-L1 vaccine and nivolumab was safe Treatment-related adverse events are listed for all 30 patients in Supplementary Table 4. Common treatment-related grade 1–2 toxicities were fatigue (47%), rash (47%), arthralgia (30%), diarrhea (23%), nausea (23%), dry skin (20%), pruritus (20%), infusion reaction (17%), xerostomia (17%) and myalgia (17%). Four patients (13%) experienced grade 3–4 adverse events: one patient with a grade 3 maculopapular rash (MM01), one patient with grade 3 adrenal insufficiency (MM06) and one patient with grade 3 arthralgia (MM22). Patient MM18 died from urosepsis with multi-organ failure and severe hyponatremia. This patient had experienced multiple immune-related adverse events with grade 3 colitis, grade 2 pneumonitis, grade 3 arthralgia, grade 2 vasculitis and grade 2 nivolumab infusion-related allergic reaction. Additionally, patient MM18 had symptoms of myocarditis at the time of death with highly elevated cardiac troponin I levels. Bedside echocardiography showed an ejection fraction of 15%, which at baseline was 60%, but an autopsy was not conducted, and myocarditis was never confirmed pathologically. Patient MM06 had received first-line treatment with ipilimumab before entering the trial and was on substitution corticosteroids at the time of inclusion. Adrenal insufficiency was aggravated by an erysipelas infection with high fever, reaching grade 3 in Common Terminology Criteria for Adverse Events and resolving quickly after appropriate antibiotic therapy was initiated. As expected, local side effects were common with 77% of the patients who developed injection site reactions. These reactions were classified as granulomas (63%), redness (20%), pain (13%) and pruritus (13%) at the injection site. All local reactions were grade 1–2, most likely related to the Montanide adjuvant and typically transient. However, two patients (MM07 and MM20) decided to discontinue vaccination after eight and 11 injections, respectively, due to granulomas, tenderness and pain that limited instrumental activities of daily living but continued to receive nivolumab (Extended Data Fig. 2c). Vaccine-specific responses in blood were frequently detected First, all 30 patients were assessed for the presence of vaccine-specific responses in peripheral blood mononuclear cells (PBMCs) before, on and after vaccination using a modified interferon (IFN)-γ enzyme-linked immune absorbent spot (ELISPOT) assay. This assay is known to expand antigen-specific memory cells and to improve the detection and the correlative power of ELISPOT during treatment19–21. Prevaccine IDO-specific responses were detectable in ten (33%) patients, while prevaccine PD-L1-specific responses were detectable in eight patients (27%); overlapping (specific to both IDO and PD-L1) prevaccine responses were present in four (13.3%) patients. During vaccination, an increase of IDO-specific T cells or PD-L1-specific T cells in the blood was observed in 28 (93%) and 26 (86%) patients, respectively. In total, 93% of patients had an increase in either PD-L1- or IDO-specific responses on vaccination (Fig. 3a), with a significant (P < 0.0001) median increase from baseline to the post-vaccine response demonstrated for both IDO and PD-L1 (at different time points on treatment), confirming that vaccine-specific immune responses were induced in patients regardless of clinical response (Fig. 3b,c). Immune responses fluctuated in the blood over time (Extended Data Fig. 3a). An increase in IDO- and PD-L1-specific responses in the peripheral blood was also detectable directly ex vivo across the different clinical response groups (Extended Data Fig. 4).Fig. 3 Vaccine-specific responses in blood. a, IDO- and PD-L1-specific T cell responses in PBMCs at baseline and on vaccination as measured by the IFN-γ ELISPOT assay (n = 30). *Responses were calculated as the difference between the average numbers of spots in wells stimulated with IDO or PD-L1 peptide (triplicates) and those from the corresponding control (DMSO), and statistical analyses of ELISPOT responses were performed using a distribution-free resampling method (Moodie et al.50). DR (double response), response was not statistically confirmed due to replicate number, but the number of spots in peptide wells was two times higher than that in control wells (DMSO). NS, no significant response and no DR. For a detailed overview of responses at serial time points on vaccination, see Extended Data Fig. 4a. b, IDO- and PD-L1-specific T cell response in PBMCs in all treated patients measured by the IFN-γ ELISPOT assay at baseline and on vaccination. On-vaccination responses were selected from the ‘best’ ELISPOT response at different time points for each patient (series 3, 6, 12, 18 or 24) during vaccination (n = 30). Wilcoxon matched-pairs signed-rank test was used to compare responses to IDO or PD-L1 peptides in the vaccine between baseline and later time points. c, Representative example of ELISPOT wells with response in patient MM23 in serial PBMCs before and on treatment. d, IDO-specific CD4+ and CD8+ T cells were isolated and expanded from PBMCs stimulated in vitro with the IDO peptide and a low dose of IL-2 for 14–15 d before sorting using the Miltenyi Cytokine Secretion Assay—Cell Enrichment and Detection kit. To assess their cytolytic potential, IDO-specific T cells were stimulated with IDO peptide, and expression of CD107a, IFN-γ and TNF-α was assessed by flow cytometry (the example is from patient MM14). Sustained vaccine-specific responses were observed 3 and 6 months after the last vaccine, indicating induction of memory responses in nine patients with response to clinical treatment who surpassed follow-up at data lock (Extended Data Fig. 3b). Importantly, PD-L1- and IDO-specific responses were observed irrespective of HLA genotype (Supplementary Table 5). To verify the functionality of vaccine-induced T cells, IDO- or PD-L1-specific T cells were isolated and in vitro expanded from PBMCs of five patients. Phenotypic characterization by flow cytometry revealed that the isolated vaccine-specific T cells consisted of both CD4+ and CD8+ T cells. Also, both IDO- and PD-L1-specific CD4+ and CD8+ T cells showed pro-inflammatory properties, as they expressed the cytolytic marker CD107a and secreted the cytokines IFN-γ and tumor necrosis factor (TNF)-α (Fig. 3d and Extended Data Fig. 6a–d). Interestingly, we were also able to detect vaccine-specific CD4+ and CD8+ T cell responses in peripheral blood ex vivo (Extended Data Figs. 4 and 5). We observed a significant increase in the overall percentage of CD107a, CD137 and TNF-α expression in response to peptide stimulation in on- or post-treatment PBMC samples compared to that at baseline, further confirming the expansion and the diverse signature of vaccine-specific T cells (Extended Data Fig. 5). Vaccine-specific responses detected in the skin at the vaccination site To investigate whether vaccine-specific T cells have the potential to migrate to peripheral tissue, delayed-type hypersensitivity (DTH) tests were performed after six cycles of treatment on 15 patients to assess the presence of vaccine-reactive T cells in the skin. Supplementary Table 6 display an overview of skin-infiltrating lymphocyte (SKIL) cultures. We detected IDO-specific T cells in the skin of six of ten patients and PD-L1-specific T cells in nine of 11 patients (Extended Data Fig. 7a). Intracellular cytokine staining was performed on SKILs from five patients after stimulation with either PD-L1 or IDO peptide. Here, we detected mainly CD4+ peptide-reactive T cells that secreted TNF-α and upregulated CD107a, and a minor fraction also secreted IFN-γ. In one patient, we detected CD8+ PD-L1-reactive T cells (Extended Data Fig. 7b–d). Vaccine-induced T cells specifically recognize target cells To confirm the functionality of vaccine-expanded T cells, vaccine-specific T cell clones (clonal purity was confirmed by T cell receptor (TCR) sequencing) were isolated and expanded from patient PBMCs (Extended Data Fig. 8e). We showed that PD-L1-specific T cells were able to recognize PD-L1+ autologous tumor cells in a PD-L1 expression-dependent manner if the cancer cells also expressed HLA-II (Fig. 4a,b). Similarly, an HLA-DR-restricted IDO-specific CD4+ T cell clone was able to recognize an HLA-DR-matched IDO-expressing model cell line, MonoMac1, in an IDO-expression-dependent manner (Fig. 4e–g). As previously described, IDO- and PD-L1-specific T cells’ mode of action is not limited to targeting only cancer cells. We were able to show that vaccine-specific T cell clones also reacted against PD-L1- and IDO-expressing autologous immune cells (Fig. 4c,h). To provide myeloid cells with a tumor-associated phenotype we treated isolated CD14+ myeloid cells with tumor conditioned medium (TCM) derived from an established autologous tumor cell line. We observed that such TCM-treated CD14+ cells had increased expression of PD-L1 and IDO and were effectively recognized by autologous PD-L1- and IDO-specific CD4+ T cell clones (Fig. 4c,d,h,i).Fig. 4 PD-L1- and IDO-specific T cells from vaccinated patients react against PD-L1- and IDO-expressing target cells. a, Left, PD-L1-specific T cell culture (MM1636.05) reactivity against PD-L1 peptide or autologous tumor cells in the IFN-γ ELISPOT assay. Tumor cells were either not treated or pretreated with 200 U ml−1 IFN-γ for 48 h before the assay. Effector:target (E:T) ratio of 10:1 was used. Right, PD-L1 and HLA-II surface expression on melanoma cells with (green) or without (yellow) pretreatment with IFN-γ compared to an isotype control (gray) as assessed by flow cytometry. b, Left, PD-L1-specific T cell (MM1636.05) reactivity in the IFN-γ ELISPOT assay against autologous tumor cells pretreated with IFN-γ (500 U ml−1) and transfected with mock or PD-L1 small interfering (si)RNA 24 h after transfection. E:T ratio, 10:1. Right, PD-L1 surface expression on melanoma tumor cells (MM1636.05) assessed by flow cytometry 24 h after transfection with mock (blue) or PD-L1 (red) siRNA compared to the isotype control (gray). c, Reactivity of the CD4+ PD-L1-specific T cell clone (MM1636.14) against PD-L1 peptide or autologous CD14+ cells; E:T ratio, 10:1. CD14+ cells were isolated using magnetic bead sorting and used as targets in an ELISPOT assay directly or after pretreatment for 2 d with TCM derived from the autologous tumor cell line. d, Quantitative PCR with reverse transcription (RT–qPCR) analysis of PD-L1 (CD274) expression in sorted CD14+ cells before and after treatment with autologous TCM for 48 h. e, Reactivity of the IDO-specific CD4+ T cell clone (MM1636.23) against IDO peptide combined with HLA-DR (L243)-, HLA-DQ (SPV-L3)- or HLA-DP (B7/21)-blocking antibodies (aHLA-DR, aHLA-DQ and aHLA-DP) in an intracellular staining assay (ICS) for IFN-γ and TNF-α production. T cells were incubated with individual blocking antibodies (2 μg ml−1) for 30 min before adding IDO peptide. f, Reactivity of the IDO-specific CD4+ T cell clone (MM1636.23) against the HLA-DR-matched IDO-expressing cell line MonoMac1 transfected with mock or IDO siRNA in an ICS assay, E.T ratio, 4:1. siRNA transfection was performed 48 h before the experiment. g, RT–qPCR analysis of IDO1 expression in MonoMac1 cells 48 h after siRNA transfection. h, Reactivity of the CD4+ IDO-specific T cell clone (MM1636.14) against IDO peptide or autologous CD14+ cells; E:T ratio, 20:1. CD14+ cells were isolated using magnetic bead sorting and used as targets in an ELISPOT assay directly or after pretreatment with TCM derived from the autologous tumor cell line. i, RT–qPCR analysis of IDO1 expression in sorted CD14+ cells before and after treatment with autologous TCM for 48 h. Bars in RT–qPCR data (d,g,i) represent the mean of three (d,i) or six (g) technical replicates ±s.d.; P values were determined by two-tailed parametric t-tests. ELISPOT counts (a,b,c,h) represent the mean value of three technical replicates ±s.e.m.; response P values were determined using the distribution-free resampling (DFR) method. TNTC, too numerous to count. T cell clones in blood and tumors To track the role of treatment-induced T cell responses, TCR sequencing of the complementarity-determining region 3 (CDR3) was performed on five patients in peripheral blood (baseline and cycles 3, 6 and 12) and paired biopsies. These five patients (MM01, MM02, MM08, MM09 and MM13) were selected due to the availability of material and to investigate a balanced patient group with both responders and non-responders. Due to a limited number of available paired biopsies (either as a consequence of patient refusal or unexpectedly rapid and substantial clinical responses), statistical analysis could not be applied. Details on clinical response are shown in Fig. 2a. Additionally, PBMCs (on treatment) or SKILs were stimulated with the IDO/PD-L1 peptides, and then cytokine-producing T cells were sorted to track vaccine-induced T cells both in the periphery and at the tumor site. To identify enriched IDO/PD-L1-specific T cell clones, TCR rearrangements in sorted IDO/PD-L1-specific clones and TCR rearrangements in sorted IDO/PD-L1-specific T cell samples were compared to sequences from baseline PBMC samples for each patient. Clonal expansion of vaccine-specific TCR rearrangements from samples on vaccination were then tracked using a differential abundance framework. Cumulative IDO/PD-L1-specific T cell frequencies were tracked in post-treatment samples. We found no relation between clinical response and the enrichment of vaccine-specific clones, but an increase in IDO/PD-L1-specific T cell clones was observed at different time points in the periphery in all five patients (Extended Data Fig. 3c). We next investigated overall changes in the T cell repertoire in the blood. A modest increase in the peripheral T cell fraction was observed in the three responding patients at cycle 3, while the two non-responding patients had a clear decrease in T cell fraction (Extended Data Fig. 8a). We thereafter investigated TCR clonality and TCR repertoire richness, exploring the proportion of abundant clones and the number of unique rearrangements, respectively. A decreasing peripheral Simpson clonality and increasing TCR repertoire richness was observed in responding patients at cycle 3, which might indicate tumor trafficking upon treatment. The opposite pattern was observed in non-responding patients (Extended Data Fig. 8b,c). Peripherally expanded clones were associated with tumors and persisted until cycle 12 (latest time point analyzed). The largest peripheral expansion was observed at cycle 3, with the most significant increase observed in patient MM01 (CR). Responding patients had a larger fraction of peripherally expanded clones that were also found in tumors compared to that of non-responders. By tracking peripherally expanded clones detected at the tumor site, we observed that patient MM01 had a substantial increase after treatment, indicating tumor trafficking of peripheral expanded clones (Extended Data Fig. 8d). Influx at the tumor site of vaccine-enriched T cell clones Given the observation of increased T cell fraction and enrichment of IDO- and PD-L1-specific clones in the blood after treatment, we investigated whether the same trend was observed at the tumor site. Both TCR sequencing and immunohistochemistry (IHC) of paired biopsies from the five patients described above showed an increase in the T cell fraction with an influx of CD3+ and CD8+ T cells after treatment in the three responding patients (Fig. 5a–c). IHC was not possible for patient MM09 due to tissue loss.Fig. 5 Changes in the TME after treatment. Number of CD3+ and CD8+ T cells, TCR fraction, TCR clonality, TCR repertoire, biopsy expanded TCR clones and enrichment of IDO- and PD-L1-specific T cells at the tumor site. a, Number of CD3+ and CD8+ T cells per mm2 (sum in the validated area) at the tumor site detected by IHC of paired biopsies from four patients. b, Example of IHC of CD3+ and CD8+ T cells at the tumor site before and after treatment (cells per mm2) in one patient (MM01). c, T cell fraction at the tumor site at baseline and cycle 6 by TCR sequencing. The T cell fraction was calculated by taking the total number of T cell templates and dividing by the total number of nucleated cells. d, Tracking of vaccine-associated clones at baseline and cycle 6 in tumor biopsies. Cumulative frequencies of IDO and PD-L1 vaccine-specific TCR rearrangements are represented. e,f, TCR clonality and TCR repertoire richness in five patients at the tumor site at baseline and cycle 6. Simpson clonality measures how evenly TCR sequences are distributed among a set of T cells, where 0 indicates an even distribution of frequencies and 1 indicates an asymmetric distribution in which a few clones dominate. TCR repertoire richness reports the mean number of unique rearrangements. g, Bar chart representing baseline expanded biopsy clones from five patients (colored bars) and the detection of biopsy expanded clones also found in the blood at baseline and series 3, 6 and 12 (white and gray bars) by TCR sequencing. We thereafter investigated whether some of the IDO/PD-L1 vaccine-linked T cells were present at the tumor site. Vaccine-associated clones were tracked as the combined frequency of IDO- and PD-L1-specific T cell rearrangements. In biopsies, the frequencies at cycle 6 were compared to those at baseline and showed an increase in vaccine-specific T cells in four of five patients, irrespective of clinical response (Fig. 5d). TCR sequencing of PD-L1-specific SKILs (a more specific culture than IDO/PD-L1-specific isolations derived from PBMCs on vaccination) and paired biopsies showed that two of the top five PD-L1-specific SKIL clones were present at the tumor site both before and after treatment (Extended Data Fig. 7e,f). With a focus on the more abundant T cell clones, we investigated overall TCR clonality at the tumor site before and after treatment. In addition, we explored the number of unique TCR rearrangements, dissecting the lower-frequency clones. Patient MM01 had a significant increase in TCR clonality and a decrease in repertoire richness at the tumor site after therapy, indicating a focused tumor repertoire response of selected clones. All three responding patients had a decrease in TCR repertoire richness, which again might indicate a focused tumor response (Fig. 5e,f). Deeper analyses showed that the T cell clones that expanded at the tumor site after therapy were also present in the blood at baseline and increased significantly after treatment in four of five patients. The highest proportion was detected early at cycle 3. These data again support trafficking of peripherally expanded clones to the tumor site and could indicate that the T cell response to treatment is derived from pre-existing peripherally tumor-associated T cells (Fig. 5g). Signs of treatment-induced inflammation in the TME To dissect changes in the TME induced by T cell influx upon treatment in responding patients, RNA gene expression analyses using the nCounter PanCancer Immune Profiling Panel from NanoString were performed on paired biopsies from two responding patients (MM01 and MM13). Expression of genes related to adaptive immunity such as T cell activation, effector functions (genes encoding IFN-γ, TNF-α, IL-15 and IL-18) and cytotoxicity was increased in post-treatment biopsies (Extended Data Fig. 9a,b). Also, expression of genes related to checkpoint inhibitors such as those encoding T cell immunoglobulin and mucin-domain containing-3 (TIM-3), IDO, PD-L1, PD-L2, PD-1 and CTLA-4 increased after treatment, indicating activation of immune cells in the TME (Extended Data Fig. 9c). Additionally, IHC of paired biopsies from four patients (MM01, MM02, MM05 and MM13) showed an upregulation of PD-L1, IDO, MHC-I and MHC-II on tumor cells, indicating a treatment-induced pro-inflammatory response in the three responding patients, except for a decrease in MHC-II expression in patient MM13. By contrast, the non-responding patient MM02 had a reduction in T cell numbers present in the tumor after treatment and no expression of PD-L1, IDO or MHC-II and, interestingly, total loss of MHC-I, demonstrating tumor immune escape (Extended Data Fig. 10a). CD8+ T cells and their distance (µm) to PD-L1-expressing cells in baseline biopsies from five patients was investigated by IHC. Except for patient MM13 (PR), distance and clinical responses were associated: the two responders had reduced distance (<20 µm) between cells expressing these markers compared to non-responding patients (>80 µm). This observation indicates that responding patients not only have a higher intratumoral infiltration of CD8+ T cells, but that these cells can surround and attack PD-L1-expressing immune cells and tumor cells (Extended Data Fig. 10b). Discussion In this clinical trial, MM1636, 30 patients with metastatic melanoma were treated with a first-in-class immunomodulatory IDO/PD-L1-targeting peptide vaccine combined with nivolumab. The treatment led to an unprecedented high ORR of 80%, with 43% of patients reaching a CR, and a striking mPFS of 26 months (95% CI, 15.4–69 months) was reached. The vaccine represents a new treatment strategy to activate specific T cells that target cells contributing to immune suppression (including tumor cells), positively modulating the TME by inducing local inflammation. Indeed, we show that vaccine-specific T cells isolated and expanded from vaccinated patients recognize not only tumor cells in a target- and HLA-restricted manner but also myeloid cells polarized toward a tumor-associated phenotype. Hence, myeloid cells become targets for vaccine-activated T cells when they have a tumor-associated phenotype in the TME. These phenomena may further induce checkpoint molecules and rewire the TME toward an increasingly aPD1-permissive state. A drawback of our study is the single-center nonrandomized setup. Comparison between trials or between patients in trials and real-world patients is problematic due to multiple factors, such as period of conduction and other therapies available at the different periods of time. Nevertheless, the rate of investigator-assessed ORR in the phase 3 trial CheckMate 067 was 43.7% in the nivolumab monotherapy group and 57% in the nivolumab and ipilimumab group. CRs occurred in 8.9% and 11.5% of patients, respectively18. The mPFS of 26 months (95% CI, 15.4–69 months) in this trial is more than twice as long as that for patients treated with nivolumab and ipilimumab in CheckMate 067, for which an mPFS of 11.5 months (95% CI, 8.7–19.3 months) was reached. Patient baseline characteristics were in general comparable to those of patients with MM who have been treated in CheckMate 067, although patients in MM1636 were older (mean age, 70 years) and a larger fraction were positive for PD-L1 (57%)3,18,22. Among patients with PD-L1-negative tumors in MM1636, an ORR of 61.5% was still reached, which would be expected to be around 33% for first-line nivolumab monotherapy23. Some studies suggest that older patients might have a tendency to respond better to aPD1 therapy; however, this is still debated24–27. To address potential trial bias and the nonrandomized setup, patients in MM1636 were matched for age, performance status, sex, M stage, LDH level, PD-L1 status and BRAF status with a historical control group from the DAMMED, who were treated contemporarily (2015–2019) with aPD1 monotherapy as standard of care17. We found a significantly higher ORR and CRR in MM1636 compared to those of matched patients, who had an ORR of 43% and a CRR of 13%, comparable to patients treated in CheckMate 067. Restrictions of the synthetic control group are of course that it is partially historic and patient selection outside matching criteria cannot be ruled out28. Numerous contemporary clinical trials are exploring the combination of aPD1 therapy with other immunomodulating agents for advanced melanoma. Talimogene laherparepvec, an oncolytic virus, is approved by the Food and Drug Administration and the European Medicines Agency to treat advanced melanoma. A small phase 1b trial with 21 patients (MASTERKEY-265) combined talimogene laherparepvec and pembrolizumab to treat patients with advanced unresectable melanoma and reached an ORR of 62% and a CR of 33%29,30. Seventy-one percent of the patients in this trial had an M stage below M1c; this number was 40% in our trial. Furthermore, mainly patients with an M stage below M1c responded to treatment, which was not the case in MM1636. Results from a large randomized phase 3 trial are awaited (KEYNOTE-034). Epacadostat, an IDO inhibitor, was tested in combination with pembrolizumab in a nonrandomized phase 2 trial in 40 aPD1 treatment-naive patients with MM with promising results, reaching an ORR of 62%. Unfortunately, the phase 3 trial showed no indication that epacadostat provided improvement in PFS and OS31. Limitations of the phase 3 trial were the sparse information on pharmacodynamics as well as biomarker evaluation to improve the design. The IDO/PD-L1 vaccine is different from epacadostat, as it is not an IDO inhibitor but targets IDO- and PD-L1-expressing cells. Similar vaccines administered as monotherapy induced objective responses in lung cancer and basal cell carcinoma, while epacadostat as monotherapy in 52 patients resulted in no responses32,33. Sahin et al. recently published encouraging data from a first-in-human trial, in which a vaccine containing liposomal RNA targeting four unmutated tumor-associated antigens (NY-ESO-1, MAGE-A3, tyrosinase and TPTE) was administered alone or in combination with nivolumab in patients with advanced melanoma. Responses were observed in both the monotherapy group as well as the combination group in checkpoint inhibitor-experienced patients, suggesting the efficacy of non-mutant shared tumor antigen vaccines34. The overall safety and tolerability findings are comparable to those of aPD1 monotherapy. Injection site reactions were exclusive to the vaccine. However, these side effects were transient and mild in most patients and most likely due to the adjuvant Montanide. IDO- and PD-L1-specific CD8+ and CD4+ T cells exist among peripheral blood lymphocytes in healthy donors35–38 and expand in response to pro-inflammatory stimuli35,39. Furthermore, both IDO and PD-L1 are induced in cells as a counter response to the inflammatory response. This provides a mechanism that ensures immune homeostasis, which keeps IDO/PD-L1-specific T cells in check; therefore, the risk of triggering autoimmune-related adverse events by vaccination appears to be minimal. This was confirmed in the first clinical trials of IDO and PD-L1 vaccination (NCT01543464 (refs. 33,40) and NCT03042793 (ref. 41)). The spontaneous (baseline) immune response to the vaccines observed in the current study is in agreement with our previous observation in various patients with cancer35–38. Numerous vaccine-induced changes in the blood and at the tumor site were observed. Peripheral IDO- and/or PD-L1-specific T cells were detected in vitro in a modified ELISPOT in over 93% of patients on vaccination, unrelated to patient HLA type. Immune responses were persistent in patients who surpassed follow-up at data cutoff and were still detectable up to 6 months after the last vaccine, suggesting induction of memory T cells. We did not observe a correlation between vaccine-induced responses in blood and clinical responses. However, the detection of a highly significant increase in vaccine-specific T cell numbers after vaccination in almost all patients together with the very low number of patients with PD makes it difficult to expect such a correlation, especially because other aspects should be taken into account (for example, the loss of class I expression on tumors cells in patient MM02 with PD after vaccination). Frequencies of peripheral T cells induced against PD-L1 were overall higher than those against IDO. Importantly, however, we also observed that it was not the same patients who responded strongly to both IDO and PD-L1, nor did patients react with a similar response pattern to the two antigens; that is, different time points of response were observed. This suggests that each component of the vaccine plays different roles in the ongoing immune response in patients. The goal of IDO/PD-L1 vaccination was to modulate the TME to increase responsiveness to aPD1 therapy, as we have observed in animal models, in which immune conversion is demonstrated in the TME with an increased influx of T cells. Targeting both IDO and PD-L1 together enables synergy, as the TME is known to use different immune escape mechanisms and IDO and PD-L1 are often overexpressed by different cellular compartments. Ex vivo ELISPOT and flow cytometry assays confirmed the induction of immune response toward both epitopes and the higher immunogenicity of the PD-L1 epitope, although because of lower sensitivity failed to detect some of the responses. TCR sequencing of five patients confirmed enrichment of IDO/PD-L1-specific T cell clones in the blood at different time points after treatment. Furthermore, an increase in enriched IDO/PD-L1-specific clones was observed in four of five patients at the tumor site after treatment, irrespective of clinical response. Phenotypic characterization showed that vaccine-specific T cells, which were expanded in vitro with interleukin (IL)-2 from the blood of vaccinated individuals, were both CD4+ T cells and CD8+ T cells. This was confirmed by ex vivo phenotype description of vaccine-activated T cells. Vaccine-specific T cells expressed CD107a and CD137 and produced IFN-γ and TNF-α upon stimulation with the cognate target, indicating their cytolytic capacity. Overall, the data from immune monitoring supports the importance of both antigens in the generation of the frequent clinical responses in the study. Despite a limited number of paired biopsies (due to either patient refusal or the fact that a large fraction of responding patients had no assessable tumors after six cycles), we observed trends indicating treatment-induced general T cell influx in responding patients. It was shown that proliferation of CD8+ T cells in the tumor after aPD1 treatment is associated with radiographic reduction in tumor size42. Additionally, we showed that a large proportion of expanded peripheral TCR clones were associated with tumors and the most considerable amount of clonal expansion was observed early, at cycle 3. For the patient with a CR included in the TCR sequencing analyses (MM01), the number of peripheral expanded clones present at the tumor site increased after treatment compared to that at baseline, indicating tumor trafficking of peripheral expanded clones. Gene expression analyses (two paired biopsies) and IHC (five paired biopsies) further demonstrated that the combination treatment induced a pro-inflammatory TME in responding patients with signs of T cell activation and cytotoxicity and increased cytokine activity. This may lead to further upregulation of IDO, PD-L1, MHC-I and MHC-II on tumor cells, leading to more treatment targets. It was shown that, following vaccination with a cancer vaccine, PD-L1 expression is increased on tumor cells due to recruitment of tumor-specific T cells and upregulation of adaptive immune resistance pathways in the TME43. Treatment with nivolumab monotherapy enhances PD-L1 expression, and it is therefore problematic to discriminate the effect of the vaccine as compared to that of nivolumab44. IDO- and PD-L1-specific pro-inflammatory effector T cells were hypothesized to counteract the functions of IDO- and PD-L1-expressing immune-suppressive cells as a means to keep the immune balance between immune activation and inhibition45. However, because the expression of these molecules is induced as a counter-regulatory response to inflammatory mediators such as IFNs, they can also be expressed by, for example, activated myeloid cells or T cells. Thus, activation of PD-L1-specific T cells may result in depletion of activated T cells in the tumor and other sites. Importantly, it was recently described that PD-L1+ T cells mainly have tolerogenic effects on tumor immunity and exert tumor-promoting properties, suggesting that targeting this immune population is indeed also beneficial46. We have previously investigated the effect of activating PD-L1-specific T cells in vitro and in vivo and found that, overall, they supported the effector phase of an immune response by removing PD-L1-expressing regulatory immune cells that inhibit PD-1+ effector T cells47,48. The major role of the PD-1 pathway is believed to be the regulation of effector T cell responses. Thus, this protective pathway is more important after activation, rather than at the initial T cell-activation stage49. Accordingly, the presence of PD-L1-specific T cells during the activation phase of an immune response may not increase or support a pro-inflammatory immune response due to the expression of PD-L1 on potent antigen-presenting cells or on the T cells themselves14. Thus, the overall effects of PD-L1-specific T cells may vary depending on the expression of both PD-1 and PD-L1, that is, due to the microenvironment and the state of the immune response. In conclusion, here we report an impressive response rate, CR rate and mPFS for a first-in-class immune-modulating vaccine combined with nivolumab. This may be a first step toward a new treatment strategy for patients with MM. Limitations are the low number of patients treated at a single institution and the lack of a randomized design with aPD1 monotherapy as comparator. Studies in aPD1 therapy-resistant or -refractory melanoma are ongoing as well as biomarker analysis for selecting patients at a higher likelihood to benefit from the combination versus aPD1 monotherapy. A larger randomized trial will be essential to validate these findings and determine the specific contribution of the vaccine to clinical responses and changes in the TME. In December 2020, the Food and Drug Administration granted breakthrough therapy designation for the IO102/IO103 vaccine combined with aPD1 therapy in metastatic melanoma based on data from the MM1636 trial. Methods Trial design and treatment plan MM1636 is an investigator-initiated, nonrandomized, open-label, single-center phase 1/2 study. All patients were treated at the Department of Oncology, Herlev and Gentofte Hospital, University of Copenhagen, Herlev, Denmark. The study was conducted according to the Declaration of Helsinki and Good Clinical Practice (GCP) and monitored by the GCP unit in Copenhagen, Denmark. The protocol was approved by the ethical committee of the Capital region of Denmark (H-17000988), the Danish Medical Agencies (2017011073) and the Capital Region of Denmark Data Unit (P-2019-172). The study was registered at https://clinicaltrials.gov/ under the identifier NCT03047928 and at the EudraCT (no. 2016-0004527-23). This study initially aimed to include 30 aPD1 treatment-naive patients with MM. An amendment with the addition of two other cohorts with ten patients in each cohort was made to evaluate immune responses and clinical efficacy in aPD1 therapy-resistant patients (cohort B, de novo resistance and cohort C, acquired resistance) for a total of 50 patients. The amendment with cohorts B and C was approved at the inclusion of 18 patients in cohort A. The trial is still including patients in cohorts B and C. This article reports results from cohort A. The first six patients were treated as for phase 1, evaluating for safety and tolerability before the remaining 24 patients were included in phase 2. The IDO/PD-L1 vaccine was administered subcutaneously biweekly for the first 10 weeks and thereafter every 4 weeks for approximately 9 months. A maximum of 15 vaccines were administered. Nivolumab was administered according to the approved label (3 mg per kg biweekly) for 24 cycles. The 15th vaccine was administered together with the 24th nivolumab cycle of 3 mg per kg, and responding patients thereafter continued nivolumab monotherapy every 4th week (6 mg per kg) as a standard of care after investigator assessment. Treatment with nivolumab was discontinued at the maximum benefit (assessed by the investigator), after a maximum of 2 years of therapy, at progression or due to severe adverse events (Extended Data Fig. 1b, treatment plan). Vaccine composition Each vaccine was composed of 100 µg IO102, a 21-amino-acid peptide (DTLLKALLEIASCLEKALQVF) from IDO, and 100 µg IO103, a 19-amino-acid peptide (FMTYWHLLNAFTVTVPKDL) from the signal peptide of PD-L1 (PolyPeptide Laboratories). Peptides were dissolved separately in 50 µl DMSO, filtered for sterility and frozen at −20 °C (NUNC CyroTubes CryoLine System Internal Thread, Sigma-Aldrich). At <24 h before administration, peptides were thawed. The PD-L1 peptide was diluted in 400 µl sterile water and, immediately before injection, mixed with the IDO peptide solution and 500 µl Montanide ISA-51 (SEPPIC) to achieve a total volume of 1 ml. Patients Patients above 18 years of age with locally advanced or stage 4 melanoma according to the AJCC (seventh edition), at least one measurable lesion according to RECIST 1.1 and an ECOG PS of 0–1 were eligible. The main exclusion criteria were prior treatment with aPD1 therapy, CNS metastases >1 cm, severe comorbidities and active autoimmune disease. Enrollment was not restricted to PD-L1 status, but it was known before inclusion. Patients were included after informed consent. Patients MM42 and MM20 have confirmed their approval of PET–CT images and clinical images that have been included in the article. Key study assessments Safety and tolerability were evaluated based on changes in clinical laboratory analyses and reported adverse events. Adverse events were assessed according to Common Terminology Criteria for Adverse Events (version 5.0) and were graded from 1 to 5 for all treated patients up to 6 months after the last dose of the IDO/PD-L1 vaccine. Clinical efficacy was assessed using FDG PET–CT scans before treatment and every 3rd month until progression. Objective responses were categorized into CR, PR, stable disease or PD according to RECIST version 1.1. Clinical data were collected at CCIT-DK in the eCRF program OpenClinica version 1.0 and in Microsoft Excel version 2002 on a secure server. Blood samples for immunologic analyses were collected before treatment, before the third cycle, after the sixth, 12th, 18th and 24th cycles (on vaccination) and 3 and 6 months after the last vaccine. Two to three tumor needle biopsies (1.2 mm) were collected at baseline and after six cycles from the same tumor site, when accessible, to evaluate immune responses at the tumor site. DTH skin tests were performed and punch biopsies were taken from the DTH area after cycle 6 for the evaluation of SKILs reactive to PD-L1 and IDO (Supplementary Fig. 1, treatment plan). Statistical analysis of clinical outcome Survival curves were computed in GraphPad Prism software version 9.0.0 according to Kaplan–Meier method. Median follow-up time of enrollment was calculated using the reverse Kaplan–Meier method, also in GraphPad Prism software version 9.0.0. For binary outcomes, 95% two-sided CIs were constructed using the Clopper–Pearson method, also in GraphPad version 9.0.0. An independent board of certified and experienced oncoradiologists performed an external review to evaluate clinical response to address the potential bias of investigator site review. The external review took place at Rigshospitalet, Copenhagen University Hospital. This hospital did not participate in the MM1636 trial, and the external reviewer had no prior knowledge about the clinical trial or the trial therapy. Only PET–CT images were accessed, and arrows indicating target or non-target lesions appeared on baseline images as the only additional information. To address potential trial bias regarding treatment effect, we matched patients in MM1636 with patients from the DAMMED, a population-based database that retrospectively collects data on patients with metastatic melanoma in Denmark. Here, data from 938 patients treated with aPD1 monotherapy contemporaneously (January 2015–October 2019) were extracted. Two hundred and eighteen of these patients were eligible for comparison and matching (all parameters available) (Supplementary Table 1), and 74 patients from the DAMMED were found to match. Patients were matched for age (≤70 years, >70 years), sex, LDH levels (normal, elevated), M stage (M1a, M1b, M1c), BRAF status (wild type, mutated) and PD-L1 status (<1%, ≥1%). An exact matching algorithm was used in which patients in MM1636 were matched with patients from the DAMMED with the same combination of variables. Twenty-nine patients from MM1636 were matched with exact combinations of the six variables. One patient could not be matched. To secure balance of the calculations, control patients were weighted according to the number of patients for each MM1636 patient. Estimates for treatment effects were calculated by weighted logistic regression analyses and weighted Cox proportional hazard model. The R package ‘Matchlt’ was used for matching patients. As the method chosen for matching control patients to protocol patients, a weighted binary logistic regression model was used for comparing response rates in the two matched cohorts. Odds ratios and response rates, including their corresponding 95% CIs, were extracted from the regression models. All P values were two sided, and P values below 0.05 were considered statistically significant. SAS version 9.4M5 was used for the weighted logistic regression models. Processing of peripheral blood mononuclear cells Peripheral blood was collected from all patients in heparinized tubes and was processed within 4 h. In brief, PBMCs were isolated using Lymphoprep (Medinor) separation. PBMCs were counted on a Sysmex XP-300 analyzer and frozen in Human AB Serum (Sigma-Aldrich, H4522-100ML) with 10% DMSO using controlled-rate freezing (CoolCell, BioCision) in a −80 °C freezer and were moved the next day to a freezer at −140 °C until further processing. Needle biopsies at baseline and after the sixth vaccine Two to three needle biopsies (1.2 mm) were taken at baseline and after the sixth cycle of treatment when assessable from the same tumor lesion. One fragment was fixed with formalin and embedded in paraffin (FFPE); one to two fragments were used for expanding tumor-infiltrating T cells and establishing autologous tumor cell lines. Delayed-type hypersensitivity and generation of skin-infiltrating lymphocytes After six cycles of treatment, we performed intradermal injections of vaccine components without adjuvant and one control injection containing DMSO without peptide. Patients were injected with either a mixture of both IDO and PD-L1 peptides at all three injection sites or PD-L1 peptide, IDO peptide or a mixture of the two at the injection sites, respectively (Supplementary Table 6). Eight hours after injection, punch biopsies were resected from the three sites where the peptide was injected and transported immediately to the laboratory and cut into fragments of 1–2 mm3. SKILs were expanded to establish ‘young SKILs’ in CM medium consisting of RPMI 1640 with GlutaMAX, 25 mM HEPES, pH 7.2 (Gibco, 72400-021), IL-2 (100 or 6,000 IU ml−1) (Proleukin Novartis, 004184), 10% heat-inactivated Human AB Serum (Sigma-Aldrich, H4522-100ML), 100 U ml−1 penicillin, 1.25 µg ml−1 Fungizone (Bristol Myers Squibb, 49182) and 100 µg ml−1 streptomycin (Gibco, 15140-122). Half of the medium was replaced three times per week. Young SKILs from samples derived from IDO peptide, PD-L1 peptide and the mixture injection sites were further expanded in a small-scale version of the 14-d rapid expansion protocol as previously described51. Quantification of specific T cells in blood by ELISPOT For enumeration of vaccine-specific T cells in the peripheral blood, PBMCs from patients were stimulated with IDO or PD-L1 peptide in the presence of a low dose of IL-2 (120 U ml−1) for 7–13 d before being used in IFN-γ ELISPOT assays with 2.5–3.2 × 105 cells per well. Briefly, cells were placed in a 96-well PVDF membrane-bottomed ELISPOT plate (MultiScreen MSIPN4W50, Millipore) precoated with IFN-γ-capture antibody (clone 1-D1K, Mabtech). Diluted IDO or PD-L1 peptide stocks in DMSO were added at 5 µM; an equivalent amount of DMSO was added to control wells. PBMCs from each patient were set up in duplicate or triplicate for peptide and control stimulations. Cells were incubated in ELISPOT plates in the presence of the peptide for 16–18 h, after which they were washed off, and biotinylated secondary antibody (anti-human IFN-γ mAB, clone 7-B6-1, Mabtech) was added. After 2 h of incubation, unbound secondary antibody was washed off, and streptavidin-conjugated alkaline phosphatase (Mabtech) was added for 1 h. Next, unbound streptavidin-conjugated enzyme was washed off, and the assay was developed by adding BCIP/NBT substrate (Mabtech). ELISPOT plates were analyzed on the CTL ImmunoSpot S6 Ultimate V analyzer using ImmunoSpot software version 5.1. Responses were calculated as the difference between the average numbers of spots in wells stimulated with IDO or PD-L1 peptide and those from corresponding control wells. For detection of IDO and PD-L1 peptide responses ex vivo, PBMCs were thawed and rested overnight in medium containing DNase I (1 μg ml−1, Sigma-Aldrich, 11284932001) before being used in IFN-γ ELISPOT assays as described above. A total of 6–9 × 105 PBMCs per well were seeded. Statistical analysis of all ELISPOT responses was performed using the DFR method as described by Moodie et al. using RStudio software (RStudio Team, 2016, http://www.rstudio.com/)50. Vaccine-specific ELISPOT responses were defined as true if the difference between the spot count in control and peptide-stimulated wells was statistically significant according to DFR statistical analysis or, for samples performed in duplicate, if the spot count in peptide-stimulated wells was at least 2× the spot count in control wells50. Cancer cell lines and tumor conditioned medium Autologous melanoma cell lines were established from needle biopsies. Briefly, biopsies were chopped into small fragments and seeded in 24-well cultures in RPMI 1640 with GlutaMAX, 25 mM HEPES, pH 7.2 (Gibco, 72400-021), 10% heat-inactivated FBS (Life Technologies, 10500064), 100 U ml−1 penicillin, 1.25 µg ml−1 Fungizone (Bristol Myers Squibb, 49182) and 100 µg ml−1 streptomycin (Gibco, 15140-122). Established adherent melanoma tumor cell lines were cryopreserved at −140 °C in freezing medium containing FBS with 10% DMSO. PD-L1 and HLA-II expression on established tumor cell lines was assessed by flow cytometry staining with anti-PD-L1–PE-Cy7 (1:23 dilution, clone M1h1 (RUO), BD, 558017) and anti-HLA-II–FITC (1:23 dilution, clone Tu39 (RUO), BD, 555558) antibodies. To obtain TCM, established tumor cell lines were cultured in 175-cm2 Nunc cell culture flasks until 80–90% confluency was reached. The culture medium was then replaced with 20 ml fresh X-VIVO 15 with Gentamicin and Phenol Red (Lonza, BE02-060Q), medium with 5% heat-inactivated Human AB Serum (Sigma-Aldrich, H4522-100ML). After 24 h of incubation, TCM was collected and centrifuged to remove any resuspended cells, after which TCM was aliquoted, frozen and stored at −80 °C. The acute monocytic leukemia cell line MonoMac1 was obtained from the DSMZ—German Collection of Microorganisms and Cell Cultures (ACC 252) and cultured in RPMI 1640 with GlutaMAX, 25 mM HEPES, pH 7.2 (Gibco, 72400-021) and 10% heat-inactivated FBS. Isolation of autologous myeloid cells Autologous CD14+ cells were sorted from freshly thawed PBMCs using a magnetic bead-separation kit (Miltenyi Biotec, 130-050-201) according to the manufacturer’s instructions. Isolated CD14+ cells were used as targets in the IFN-γ ELISPOT assay directly after sorting or differentiated in vitro into tumor-associated macrophages by culturing with 1 ml fresh X-VIVO 15 medium with Gentamicin and Phenol Red (Lonza, BE02-060Q) and 5% heat-inactivated Human AB Serum, supplemented with 1 ml autologous TCM in 24-well plates for 2 d. Quantification of vaccine-specific T cells from DTH biopsy sites by ELISPOT SKILs expanded in vitro were rested in medium without IL-2 overnight before being used in the IFN-γ ELISPOT assay as described above to evaluate reactivity of skin-infiltrating T cells. Generation of IDO- and PD-L1-specific T cell cultures from PBMCs or SKILs IDO- or PD-L1-specific T cells were isolated from peptide-stimulated in vitro PBMC cultures on days 14–15 after stimulation or from SKIL cultures expanded in vitro. For specific T cell isolation, PBMCs or SKILs were stimulated with IDO or PD-L1 peptide, and cytokine-producing T cells were sorted using the IFN-γ or TNF-α Secretion Assay—Cell Enrichment and Detection kit (Miltenyi Biotec). Cytokine-production profile of PD-L1- and IDO-specific T cells by intracellular staining To assess the T cell cytokine-production profile, isolated and expanded IDO- and PD-L1-specific T cell cultures were stimulated for 5 h with peptide at 5 μM in a 96-well plate. One hour after the start of the incubation, anti-CD107a–PE (1:133 dilution, clone H4A3, BD Biosciences, 555801) antibody and BD GolgiPlug (1:1,000 dilution, BD Biosciences) were added. After a 5-h incubation, cells were stained using fluorescently labeled surface marker antibodies: anti-CD4–PerCP (1:5 dilution, clone SK3, 345770), anti-CD8–FITC (1:5 dilution, clone SK1, 345772) and anti-CD3–APC-H7 (1:20 dilution, clone SK7, 560275) (all from BD Biosciences). Dead cells were stained using FVS510 (1:250 dilution, BD Biosciences, 564406), followed by overnight fixation and permeabilization using eBioscience Fixation/Permeabilization buffers (00-5123-43, 00-5223-56) according to the manufacturer’s instructions. Cells were then stained intracellularly in eBioscience permeabilization buffer (00-8333-56) with anti-IFN-γ–APC (1:20 dilution, clone 25723.11, BD Biosciences, 341117) and anti-TNF-α–BV421 (1:100 dilution, clone Mab11, BD Biosciences, 562783) antibodies. Samples were analyzed on the FACSCanto II (BD Biosciences) using BD FACSDiva software version 8.0.2. The gating strategy is shown in Supplementary Fig. 3a. For assessing the reactivity against IDO expressing MonoMac1 cells, the appropriate amount of cancer cells was added to the IDO-specific T cells to obtain an effector:target ratio of 4:1 in the 96-well plate and T cell cytokine production was tested after 5 h of co-culture as described above. For HLA-blocking experiments, HLA-blocking antibodies (2 μg ml−1) were added directly into wells 30 min before the addition of peptide. Blocking antibodies used were HLA-DR (1:500 dilution, clone L243, Abcam, ab136320), HLA-DQ (1:500 dilution, clone SPV-L3, Abcam, ab23632) and HLA-DP (1:500 dilution, clone B7/21, Abcam, ab20897). To assess ex vivo T cell reactivity to IDO and PD-L1 peptides in patient PBMCs, cells were thawed and rested for 1–2 d in medium containing DNase I (1 μg ml−1, Sigma-Aldrich, 11284932001). PBMCs were then stimulated with peptide at 5 μM in a 96-well plate for 8 h. An hour after the addition of peptide, anti-CD107a–BV421 (3:50 dilution, clone H4A3, BD Biosciences, 328626) antibody and BD GolgiPlug (1:1,000 dilution, BD Biosciences) were added. Surface and intracellular staining was performed as described above. Antibodies used for surface staining were anti-CD3–PE-CF594 (0.8:30 dilution, clone UCHT1, BD Biosciences, 562280), anti-CD4–BV711 (1:30 dilution, clone SK3, BD Biosciences, 563028) and anti-CD8–Qdot605 (1:150 dilution, clone 3B5, Thermo Fisher, Q10009). Antibodies used for intracellular staining were anti-CD137–PE (1:20 dilution, clone 4B4-1 (RUO), BD Biosciences, 555956), anti-IFN-γ–PE-Cy7 (1.5:20 dilution, clone B27, BD Biosciences, 557643) and anti-TNF-α–APC (1:20 dilution, clone Mab11, BD Biosciences, 554514). Samples were acquired on the NovoCyte Quanteon (ACEA Biosciences) and analyzed using NovoExpress software version 1.4.1. To assess vaccine-specific T cell responses, background values observed in unstimulated PBMC samples were subtracted from values observed in peptide-stimulated conditions. Positive response value threshold was set at a difference of 0.2% from the background values. Based on this response cutoff, only TNF-α, CD107a and CD137 responses were detected in this assay. Statistical analysis comparing baseline with on-treatment or post-treatment cytokine profiles was performed using two-sided Wilcoxon matched-pairs signed-rank test. The gating strategy is shown in Supplementary Fig. 3b. siRNA-mediated PD-L1 and IDO1 silencing A stealth siRNA duplex for targeted silencing of PD-L1 (Invitrogen)52, a custom silencer select siRNA for targeted silencing of IDO1 (Ambion) and the recommended silencer select negative control (Ambion) siRNA for mock transfection were used. The stealth PD-L1 siRNA duplex consisted of the sense sequence 5′-CCUACUGGCAUU-UGCUGAACGCAUU-3′ and the antisense sequence 5′-AAUGCGUUCAGCAAAUGCCAGUAGG-3′. The three silencer IDO siRNA duplexes used were siRNA1 (sense sequence, 5′-ACAUCUGCCUGAUCUCAUATT-3′; antisense sequence, 5′-UAUGAGAUCAGGCAGAUGUTT-3′), siRNA2 (sense sequence, 5′-CCACGAUCAUGUGAACCCATT-3′; antisense sequence, 5′-UGGGUUCACAUGAUCG-UGGAT-3′) and siRNA3 (sense sequence, 5′-CGAUCAUGUGAACCCAAAATT-3′; antisense sequence, 5′-UUUUGGGUUCACAUGAUCGTG-3′). For PD-L1 or IDO1 silencing, cancer cells were electroporated with 0.025 nmol of each siRNA duplex as previously described53. For PD-L1-silencing experiments, cancer cells were treated with IFN-γ (500 U ml−1, PeproTech) 1 h after electroporation. Electroporated cells were used as target cells in ELISPOT and ICS assays 24 h or 48 h after siRNA electroporation. RT–qPCR Total RNA was extracted using the RNeasy Plus Mini kit (Qiagen, 74134) following the manufacturer’s instructions. RNA concentration was quantified using a NanoDrop 2000 (Thermo Fisher Scientific), and a total of 1,000 ng RNA was reverse transcribed with the High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems, 4368814) using 1,000 ng input RNA for transcription. Real-time qPCR analyses were performed using the TaqMan Gene Expression Assay on a Roche LightCycler 480 instrument. RT–qPCR assays were performed with a minimum of three technical replicates and analyzed using the ΔCt method as described in Bookout et al.54 with normalization of IDO1 expression (primer ID, Hs00984148_m1) or PD-L1 expression (primer ID, Hs001125296_m1) to the expression level of the housekeeping gene POLR2A (primer ID, Hs00172187_m1) and the control sample (mock). For low-concentration samples with no amplification, Ct was set to 40. Controls without reverse transcriptase were used as controls for specific amplifications. P values were determined using two-tailed parametric t-tests. BRAF-mutational status and PD-L1 status at baseline in all patients A library of historical FFPE biopsies were assessable for all patients and analyzed locally at Herlev and Gentofte University Hospital by experienced pathologists for BRAF status and PD-L1 expression on tumor cells. The BRAF analysis was carried out with real-time PCR using the EntroGen BRAF Mutation Analysis Kit II (BRAFX-RT64, CE-IVD) to specifically detect mutations corresponding to V600D, V600E and V600K in the BRAF protein. PD-L1 status was assessed using the monoclonal rabbit anti-PD-L1 antibody (clone 28.8, PD-L1 IHC 28-8 pharmDx) in FFPE biopsies following the manufacturer’s instructions. Patients were considered positive for PD-L1 with expression levels ≥1% and negative for PD-L1 with expression levels <1%. Human leukocyte antigen type Blood samples from all 30 patients were genotyped for class I (HLA-A, HLA-B, HLA-C) and three class II types (HLA-DRB1, HLA-DQA1, HLA-DQB1) using LinkSēq HLA Typing Kits (Thermo Fisher, 1580C). These test kits are based on real-time PCR using allele-specific exponential amplification (sequence-specific primers) followed by melting curve analyses. Immunohistochemistry simplex: Immunoscore CR IHC staining was performed using a qualified Ventana Benchmark XT with four different steps: (1) antigen retrieval; (2) staining with the following primary antibodies: anti-CD3 antibody (clone HDX2; provider, HalioDx; HD-FG-000013; 10931065/10636667; concentration, 0.25 µg ml−1), anti-CD8 antibody (clone HDX3; provider, HalioDx; HD-FG-000019; 10931069/10337710/10639301; concentration, 0.5 µg ml−1), anti-IDO monoclonal antibody (clone VINC3IDO; provider, Thermo Fisher Scientific; 14-9750-82; E25003-101; concentration, 0.05 µg ml−1), anti-HLA class 1 ABC antibody (clone, EMR8-5; provider, Abcam; ab70328; GR3248333/GR3186494; concentration, 0.5 µg ml−1), anti-HLA-DR/DP/DQ/DX antibody (clone CR3/43; provider, Santa Cruz; sc-53302; L1714; concentration, 1 µg ml−1) and anti-PD-L1 antibody (clone HDX3; provider, HalioDx; HD-FG-000035; 106312810/106312816; concentration, 3.3 µg ml−1); (3) detection with a secondary antibody using the ultraView Universal DAB Detection kit; and (4) counterstaining using Hematoxylin & Bluing Reagent (staining of cellular nuclei). Control slides were systematically included in each staining run to permit quality control of the obtained measurements. Following coverslipping, slides were scanned with the NanoZoomer-XR to generate digital images (20×). Two consecutive slides were specifically used to perform Immunoscore CR TL staining of CD3+ and CD8+ cells. Digital pathology of T lymphocytes The digital pathology for Immunoscore CR TL allowed the quantification of positive cells (in cells per mm2) into the core tumor and invasive margin if present. Each sample was analyzed using the HalioDx Digital Pathology Platform. Digital pathology and pathologist analysis of IDO Digital pathology for IDO allowed the quantification of stained area (in mm2) into the whole tumor. Each sample was analyzed using the HalioDx Digital Pathology Platform. Analysis of IDO+ cells was performed by a pathologist. Results were expressed as an H score from 0 to 300. The score is obtained by the formula 3 × percentage of cells with strong staining + 2 × percentage of cells with moderate staining + percentage of cells with weak staining. Digital pathologist analysis of MHC-I and MHC-II Analysis of MHC-I+ and MHC-II+ cells was performed by a pathologist. Results are expressed as a percentage of positive tumor cells and a percentage of positive cells in the stroma. Digital pathology Immunoscore immune checkpoint (CD8+ and PD-L1+) The Immunoscore CR IC test allowed the quantification of CD8++ cell density in the whole tumor and a CD8+-centered proximity index (which corresponds to the percentage of CD8++ cells that have at least one PD-L1+ cell in the neighborhood) at different cutoff distances (20 µm, 40 µm, 60 µm and 80 µm). Pathologist analysis of PD-L1 A pathologist performed analysis of PD-L1+ cells. Positivity of a viable tumor cell was considered when partial or complete cell membrane staining was observed (more than 10% of the tumor cell membrane). Results were expressed as a percentage. NanoString RNA profiling and Immunosign RNA was extracted from FFPE tissues using Qiagen RNeasy FFPE extraction kits (Qiagen). Annotations from the pathologist performing H&E staining were used to guide removal of normal tissue from the slides by macrodissection before nucleic acid extraction, which occurred after tissue deparaffinization and lysis. Each extracted RNA sample was independently quantified using a NanoDrop spectrophotometer (NanoDrop Technologies) and qualified (Agilent Bioanalyzer). Degradation was quantified as the percentage of RNA fragments smaller than 300 bp using the RNA 6000 Nano kit (Agilent Bioanalyzer). Good sample quality was defined as less than 50% of RNA fragments being 50–300 bp in size. RNA expression profiling was performed using the nCounter PanCancer Immune Profiling Panel from NanoString (NanoString Technologies). The PanCancer Immune Profiling Panel contains 776 probes and is supplemented with six genes to complete HalioDx Immunosign targets. Hybridization was performed according to the manufacturer’s instructions. Hybridized probes were then purified and immobilized on a streptavidin-coated cartridge using the nCounter Prep Station (NanoString Technologies). Data collection was carried out on the nCounter Digital Analyzer (NanoString Technologies) following the manufacturer’s instructions to count individual fluorescent barcodes and quantify target RNA molecules present in each sample. For each assay, a scan of 490 fields of view was performed. Raw data from the NanoString nCounter were processed using NanoString recommendations. Quality control enables to retain data of good quality with a binding density that ranges between 0.05 and 2.25. The linearity of positive controls was checked using the R2 value of regression between the counts and the concentrations of positive controls. Samples that showed R2 < 0.75 were flagged and removed from the analysis. The background was removed using a thresholding method at the mean + 2 s.d. of negative controls. Raw counts were normalized using a positive normalization factor. Samples showing positive normalization factors outside the range of 0.3–3 were removed from the analysis. A second normalization was performed using the housekeeping gene normalization factor. Only the most stable housekeeping genes were selected for this normalization step using the variance-versus-mean relationship. All samples showing a normalization factor outside the range of 0.1–10 were removed from the analysis. All statistical analyses were performed on normalized counts using R software (version 2.6.2, 2019-12-12). T cell receptor variable β chain sequencing To track longitudinal immune responses to therapy, genomic DNA was extracted from longitudinal pre- and post-treatment PBMCs (five patients), pre- and post-treatment biopsies (five patients) (both FFPE and treated with RNAlater) and IDO- and PD-L1-specific T cell cultures from PBMCs (five patients) or SKILs (one patient). Three clonal (two IDO- and one PD-L1-) specific cultures were also sequenced to confirm clonal purity of cultures. DNA from PBMCs or RNAlater-treated biopsies was extracted with the DNeasy Blood and Tissue kit (Qiagen, 69504), DNA from sorted IDO and PD-L1-specific T cells from either PBMCs or SKILs was extracted using the QIAamp DNA Micro kit (Qiagen, 565304), and DNA from PFFE biopsies was extracted using the Maxwell RSC DNA FFPE kit (Promega, AS1450). Immunosequencing of the CDR3 regions of human TCRβ chains was performed using the immunoSEQÒ Assay (Adaptive Biotechnologies). Extracted genomic DNA was amplified in a bias-controlled multiplex PCR, followed by high-throughput sequencing. Sequences were collapsed and filtered to identify and quantitate the absolute abundance of each unique TCRβ CDR3 region for further analysis as previously described55–57. Statistical analyses of TCRβ sequencing results Two quantitative components of diversity were compared across samples in this study. First, Simpson clonality was calculated on productive rearrangements by ∑i=1Rpi2, where R is the total number of rearrangements and pi is the productive frequency of rearrangement i. Values of Simpson clonality range from 0 to 1 and measure how evenly receptor sequences (rearrangements) are distributed among a set of T cells. Clonality values approaching 0 indicate a very even distribution of frequencies, whereas values approaching 1 indicate an increasingly asymmetric distribution in which a few clones are present at high frequencies. Second, sample richness was calculated as the number of unique productive rearrangements in a sample after computationally downsampling to a common number of T cells to control for variation in sample depth or T cell fraction. Repertoires were randomly sampled without replacement five times, and we report the mean number of unique rearrangements. The T cell fraction was calculated by taking the total number of T cell templates and dividing by the total number of nucleated cells. The total number of nucleated cells was derived from reference genes using the immunoSEQ Analyzer version 3. To identify enriched vaccine clones in each patient, rearrangement frequencies in their baseline PBMCs and each IDO/PD-L1-sorted T cell sample were compared using a binomial distribution framework as previously described58. In brief, for each clone, we performed a two-sided test to determine whether frequencies were the same in the patient’s periphery and a PD-L1- or IDO-specific T cell sample. The Benjamini–Hochberg procedure was used to control the false discovery rate at 0.01 (ref. 59). Clonal expansion in post-treatment samples was similarly assessed using this differential abundance framework, but an IDO/PD-L1-specific T cell sample was replaced with a post-treatment series sample. In biopsies, the six series frequencies were compared to those of baseline tissue. Lastly, vaccine-associated clones were tracked in each PBMC and tissue sample by summing the frequency of each rearrangement enriched in either PD-L1- or IDO-specific T cells. All statistical analyses were performed in R version 3.6.1. Reporting Summary Further information on research design is available in the Nature Research Reporting Summary linked to this article. Online content Any methods, additional references, Nature Research reporting summaries, source data, extended data, supplementary information, acknowledgements, peer review information; details of author contributions and competing interests; and statements of data and code availability are available at 10.1038/s41591-021-01544-x. Supplementary information Supplementary Information Supplementary Figs. 1–3 and Tables 1–6 Reporting Summary Extended data Extended Data Fig. 1 Mode of Action and treatment plan. a) Anticipated mechanism of action of the combination therapy of an IDO/PD-L1 derived peptide vaccine and nivolumab (anti-PD1). 1) The IDO/PD-L1 peptide vaccine is administered subcutaneously (s.c) and nivolumab is administered intravenous (IV). 2) The vaccines peptides are phagocytosed by an antigen presenting cell and presented to IDO and PD-L1 specific T cells, which are activated. 3) The activated T cells migrates to the tumor site where they attack both immune-suppressive cells and tumour cells expressing IDO and/or PD-L1 leading to cytokine production and a pro-inflammatory tumour microenvironment. 4/5) Enhanced tumour killing by both IDO/PD-L1 specific T cells and tumor specific cytotoxic T cells due to PD-1 blockade. Created with BioRender.com b) Treatment plan. After written informed content patients were screened. Before treatment start a baseline PET/CT scan was performed, baseline blood sample for research use and if assessable a baseline needle biopsy. Patients were treated with the IDO/PD-L1 peptide vaccine subcutaneously biweekly for the first 6 injections and thereafter every fourth week for a maximum of 15 vaccines. Nivolumab was administered in parallel biweekly (3mg/kg) up to 24 series. If patients needed subsequent nivolumab after ended vaccination regiment they were treated with 6 mg/kg every fourth week up to two years. Needle biopsy and delayed type hypersensitivity (DTH) was performed after 6 series of treatment if assessable. PET/CT scans was performed every third month. Extended Data Fig. 2 Progression free survival and overall survival in matched historical control group and vaccine injection site reaction. a) Kaplan-Meier curve of progression free survival with corresponding 95% confidence intervals in the matched historical control group (n=74). Patients were matched on BRAF-status, PD-L1-status, age, gender, M-stage and LDH level). b) Kaplan-Meier curve of overall survival with corresponding 95% confidence intervals in the matched historical control group (n=74). c) Images of injection site reaction in patient MM20 (CR) after 11 vaccination show redness, rash and granuloma at injection site. Extended Data Fig. 3 Vaccine specific responses in blood. a) Heatmap of specific (background has been subtracted) IDO and PD-L1 responses in PBMCs at baseline, series 3, 6, 12, 18 and 24 measured by in vitro IFN-γ Elispot assay show fluctuations in the blood during treatment (n=30). b) Heatmap of specific (background has been subtracted) IDO and PD-L1 responses in PBMCs at baseline and 3 and 6 months after last vaccine measured by in vitro IFN- γ Elispot assay (n=30). 2.5–3.2x105 cells per well were used. c) Vaccine associated clones were tracked in the blood by summing the frequency of each rearrangement enriched in either IDO or PD-L1 T cells. Extended Data Fig. 4 Ex vivo vaccine specific responses in blood. Heatmaps of detected specific (background has been subtracted) IDO (a) and PD-L1 (b) responses in PBMCs at baseline and on/after treatment as measured by IFNγ ELISPOT (n=25). C) Example of well images of ex vivo ELISPOT wells for three different patients (n=3). 6-9x105 cells per well were used. Extended Data Fig. 5 CD4 and CD8 vaccine specific T cell responses in blood. Top: heatmaps of IDO specific CD4 (left) and CD8 (right) T cell responses in PBMCs at baseline and on/post treatment. Bottom: heatmaps of PD-L1 specific CD4 (left) and CD8 (right) T cell responses. Responses quantified by flow cytometry by an increased expression of CD107α, CD137 and TNFα after 8h peptide stimulation. Values represent specific responses after background values have been subtracted (n=21). Statistical analysis were performed using two-sided Wilcoxon matched-paired rank test. Extended Data Fig. 6 Pro-inflammatory profiles of sorted CD4 and CD8 T cells from blood. a/b/c/d) Percentage of in vitro stimulated and sorted PD-L1 and IDO specific CD4 and CD8 T cells secreting cytokines. Extended Data Fig. 7 Vaccine specific responses in skin. a) IDO and PD-L1 specific T cell responses in SKILs after 6 series of treatment measured by IFN-γ Elispot assay (n=13). SKILs were grown from DTH injection with either IDO peptide, PD-L1 peptide or a mix as presented by different blue colours. * Responses were calculated as the difference between average numbers of spots in wells stimulated with IDO or PD-L1 peptide (triplicates) and corresponding control (DMSO) and statistical analyses of Elispot responses were performed using distribution-free resampling method (Moodie et al.). DR: Not statistically confirmed response due to replicate number but number of spots in peptide wells are two times higher than control wells (DMSO). NS: No significant response and no DR. b/c/d) Percentage of cytokine secreting/CD107a+ CD4+ and CD8+ IDO and PD-L1 specific T cells in response to in vitro peptide stimulation by flow cytometry. e) Skin infiltrating PD-L1 specific T cell clones also found in biopsy in patient MM01. TCR sequencing was performed on a PD-L1 specific T cell culture generated from DTH area on the lower back injected with PD-L1 peptide on patient MM01. Bars show the frequency of top 25 clones in the culture with which indicates a high Simpson clonality of 0.43. f) Tracking the frequency of the top five skin infiltrating PD-L1 specific clones in tumour before and after treatment. Extended Data Fig. 8 T cell changes in blood after treatment. T cell fraction, TCR clonality and repertoire richness in blood and peripherally expanded TCR clones in both responding and non-responding patients. a) T cell fraction in peripheral blood of five patients at baseline, series 3, 6 and 12 by TCR sequencing. T cell fraction was calculated by taking the total number of T cell templates and dividing by the total number of nucleated cells (n=5.) b) TCR clonality in peripheral blood of five patients at baseline, series 3, 6 and 12 by TCR sequencing (n=5). Simpson clonality measures how evenly TCR sequences are distributed amongst a set of T cells where 0 indicate even distribution of frequencies and 1 indicate an asymmetric distribution where a few clones dominate. c) TCR repertoire richness in peripheral blood of five patients at baseline, series 3, 6 and 12 by TCR sequencing (n=5). TCR repertoire richness report the mean number of unique rearrangements. d) Bar-chart chart representing dynamics of the expanded T cell clones. Coloured bars represent peripherally expanded clones in five patients at series 3, series 6 and series 12 as compared to the baseline PBMC samples (n=5). Light gray bar represents peripherally expanded clones that are present in baseline biopsy samples while dark gray bar represents peripherally expanded clones that are present in the post-treatment biopsy (after 6 series). e) Frequency of the dominant TCR β chain in clonal PD-L1 and IDO specific cultures as determined by CDR3 sequencing. Extended Data Fig. 9 Profiling of genes relevant to T cell activation, cytokines and exhaustion markers on pre- and post-treatment biopsies in two patients. a) RNA expression profiling of genes related to T cell activation was performed using NanoString nCounter (n=2). b) RNA expression profiling of genes related to cytokine activity was performed using NanoString nCounter (n=2). c) RNA expression profiling of genes related to checkpoint inhibitors was performed using NanoString nCounter (n=2). Extended Data Fig. 10 Treatment induced upregulation of PD-L1, IDO, MHC I and MHC II and distance between CD8 T cells and PD-L1 expressing cells. a) IHC on 4 paired biopsies stained for CD3 and CD8 T cells, PD-L1, MHC I and MHC II on tumor cells/mm2 (sum in the validated area) and IDO H-score (from 0 to 300). H-score is the expression of IDO on both immune and tumor cells: The score is obtained by the formula: 3 x percentage of cells with a strong staining + 2 x percentage of cells with a moderate staining + percentage of cells with a weak staining (n=4). b) Distance in µm between CD8+ T cells and PD-L1+ stained cells on five baseline biopsies detected by IHC (n=5). Extended data is available for this paper at 10.1038/s41591-021-01544-x. Supplementary information The online version contains supplementary material available at 10.1038/s41591-021-01544-x. Acknowledgements We thank all patients and their relatives for being a part of the trial. We thank Ö. Met, M. Jonassen, S. Ullitz Færch, B. Saxild, S. Wendt and C. Grønhøj for technical support. We thank M. Cumberbatch for input with translational analysis. We thank the nurses at clinic 5 and the head of the Oncology Department at Herlev and Gentofte Hospital, L. Sengeløv. The study was funded through a research funding agreement between IO Biotech and the CCIT-DK, Herlev Hospital and the Oncology Department at Herlev Hospital. Author contributions I.M.S. and M.H.A. conceived the study. J.W.K. wrote the protocol with input from I.M.S., M.H.A. and M.-B.Z. J.W.K. and C.L.L. were responsible for patient treatment and patient care with help from R.B.H. and consulting from I.M.S., E. Ellebaek and M.D. I.M.S., E. Ellebaek and M.D. were responsible for patient recruitment. J.W.K. and C.L.L. were responsible for coordinating trial procedures and collecting data and samples. J.W.K. and C.L.L. were clinical investigators, and I.M.S. was a sponsor. J.W.K. and C.L.L. analyzed and interpreted clinical data with support from I.M.S. and E. Ehrnrooth. T.W.K. performed statistical analyses. H.W.H. helped with clinical imaging. J.W.K., E.M., C.O.M., S.M.A., S.E.W.-B., M.O.H. and A.W.P. analyzed and interpreted translational data with support from M.H.A. and I.M.S. Funding was acquired by I.M.S., M.H.A. and M.-B.Z. J.W.K. wrote the manuscript with input from I.M.S., M.H.A., E.M. and A.W.P. All authors reviewed the manuscript, interpreted data and approved the final version. Data availability This clinical trial was registered at https://www.clinicaltrials.gov/ before patient enrollment (clinical trial identifier NCT03047928). Each vaccine was composed of 100 µg IO102, a 21-amino-acid peptide (DTLLKALLEIASCLEKALQVF) from the IDO peptide, and 100 µg IO103, a 19-amino-acid peptide (FMTYWHLLNAFTVTVPKDL) from the signal peptide of PD-L1. TCR sequencing data are available from Adaptive Biotechnologies. Upon request, the CCIT-DK office will provide a username and a password to access the designated data within approximately 2–4 weeks (https://www.herlevhospital.dk/ccit-denmark/find-us/Sider/Contact-information.aspx). All requests for the remaining data including raw data and analyzed data and materials will, within a reasonable time frame, be reviewed by the CCIT-DK office (https://www.herlevhospital.dk/ccit-denmark/find-us/Sider/Contact-information.aspx) to verify whether the request is subject to any intellectual property or obligations. Patient-related data not included in the paper were generated as part of clinical trials and may be subject to patient confidentiality. Any data and materials that can be shared will be released via a material-transfer agreement. The following database was used in the study: https://research.regionh.dk/da/publications/the-danish-metastatic-melanoma-database-dammed(32749d99-095f-4cae-b5de-769bae27f01e).html. Competing interests M.H.A. is named as an inventor on various patent applications relating to therapeutic uses of IDO and PD-L1 peptides. These patent applications are assigned to the company IO Biotech, which is developing immune-modulating cancer treatments. M.H.A. is a founder, shareholder and advisor for IO Biotech. I.M.S. is a cofounder, shareholder and advisor for IO Biotech. I.M.S. has an advisory board relationship with or lectured for Roche, Novartis, MSD, Celgene, Incyte, TILT Bio, Pfizer and BMS AstraZeneca and has received limited grants for translational research from BMS, Roche and Novartis. M.-B.Z. is the CEO, founder and shareholder at IO Biotech. E.M., A.W.P. and E. Ehrnrooth are employees at IO Biotech. E. Ellebaek has received honoraria from BMS, Pierre Fabre, Roche and Kyowa Kirin and travel support from MSD. M.D. has received honoraria from Genzyme, MSD, BMS, Roche and Novartis and travel support from Novartis, MSD, BMS, Roche and Pfizer. The remaining authors declare that they have no conflict of interest. Peer review information Nature Medicine thanks George Coukos, Michael Schell and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. Javier Carmona was the primary editor on this article and managed its editorial process and peer review in collaboration with the rest of the editorial team. Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. These authors contributed equally: Julie Westerlin Kjeldsen, Cathrine Lund Lorentzen. Change history 3/8/2022 A Correction to this paper has been published: 10.1038/s41591-022-01771-w

NIVOLUMAB

DrugsGivenReaction

CC BY

34887574

2021-12-09

Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Vasculitis'.

A phase 1/2 trial of an immune-modulatory vaccine against IDO/PD-L1 in combination with nivolumab in metastatic melanoma. Anti-programmed death (PD)-1 (aPD1) therapy is an effective treatment for metastatic melanoma (MM); however, over 50% of patients progress due to resistance. We tested a first-in-class immune-modulatory vaccine (IO102/IO103) against indoleamine 2,3-dioxygenase (IDO) and PD ligand 1 (PD-L1), targeting immunosuppressive cells and tumor cells expressing IDO and/or PD-L1 (IDO/PD-L1), combined with nivolumab. Thirty aPD1 therapy-naive patients with MM were treated in a phase 1/2 study ( https://clinicaltrials.gov/ , NCT03047928). The primary endpoint was feasibility and safety; the systemic toxicity profile was comparable to that of nivolumab monotherapy. Secondary endpoints were efficacy and immunogenicity; an objective response rate (ORR) of 80% (confidence interval (CI), 62.7-90.5%) was reached, with 43% (CI, 27.4-60.8%) complete responses. After a median follow-up of 22.9 months, the median progression-free survival (PFS) was 26 months (CI, 15.4-69 months). Median overall survival (OS) was not reached. Vaccine-specific responses assessed in vitro were detected in the blood of >93% of patients during vaccination. Vaccine-reactive T cells comprised CD4+ and CD8+ T cells with activity against IDO- and PD-L1-expressing cancer and immune cells. T cell influx of peripherally expanded T cells into tumor sites was observed in responding patients, and general enrichment of IDO- and PD-L1-specific clones after treatment was documented. These clinical efficacy and favorable safety data support further validation in a larger randomized trial to confirm the clinical potential of this immunomodulating approach. pmcMain Despite remarkable advances in the treatment of MM with immune checkpoint inhibitors (ICIs) targeting PD-1 and cytotoxic T lymphocyte antigen 4 (CTLA-4), around half of patients are resistant to ICI monotherapy1. The combination of anti-CTLA-4 (aCTLA-4) and aPD1 therapy is to date the most effective therapy resulting in a response rate of around 60%; however, 50% of the patients also develop severe adverse events2,3. Therefore, an equally effective but less toxic treatment is highly needed. Several approaches to enhance ICI efficacy are currently being investigated, such as other ICIs, T cell therapy with tumor-infiltrating T cells or innate immunity stimulators such as Toll-like receptor 9 agonists4–6. Treating cancer patients with vaccines that stimulate a targeted immune response is another attractive approach, with very few side effects observed thus far in combination immunotherapy studies7,8. Immune-modulatory vaccines targeting tumoral immune escape mechanisms offer a new, generalizable strategy compared to patient-specific neoantigen cancer vaccines7,9. The immune-modulatory vaccine strategy in this clinical trial is based on the finding of circulating cytotoxic T cells specific to IDO and PD-L1 in the blood of patients with cancer and, to a lesser extent, in healthy donors. IDO- and PD-L1-specific CD8+ T cells can directly recognize and kill IDO+ and/or PD-L1+ tumor cells and likewise recognize and kill non-malignant cells that express their cognate targets. Furthermore, IDO- and PD-L1-specific CD4+ T cells release pro-inflammatory cytokines in response to IDO- or PD-L1-expressing target cells. IDO and PD-L1 are expressed not only by melanoma cells but also by many other cell types in the tumor microenvironment (TME), which differentiates these antigens from traditional tumor antigens used in other studies10–15. Activation of IDO/PD-L1-specific T cells by vaccination can therefore restrict the range of immunosuppressive signals mediated by immunosuppressive cells and thereby revert the TME from an immune hostile to an immune friendly environment. In animal models of cancer, vaccinations with IDO epitopes resulted in anti-tumor therapeutic effects that were correlated with reductions in IDO expression in myeloid cell populations within the TME16. The IDO/PD-L1 immune-modulating vaccine may lead to a translatable strategy for improving the efficacy of aPD1 therapy through activation of specific T cells. We hypothesize that the IDO/PD-L1 vaccine attracts T cells into the tumor, which induces type 1 helper T (TH1) cell inflammation and reverts the TME into an immune-permissive site, thereby turning the tumor ‘hot’. This would also upregulate PD-L1 expression in cancer and immune cells, generating more susceptible targets to aPD1 therapy (Extended Data Fig. 1a). This theory was confirmed in a mouse model in which aPD1 therapy and IDO vaccination show synergistic effects16. In this phase 1/2 clinical trial, MM1636, patients with MM received a combination of the IDO/PD-L1 (IO102/IO103) peptide vaccine with the adjuvant Montanide and the aPD1 antibody nivolumab. Patients were included in three cohorts: 30 aPD1 therapy-naive patients (cohort A), ten aPD1 therapy-refractory patients (cohort B, de novo resistance) and ten patients who progressed after aPD1 therapy (cohort C, acquired resistance). Here, we report results from cohort A. The vaccine was given biweekly for the first six administrations and thereafter every 4th week. A maximum of 15 vaccines were administered. Nivolumab was given in parallel, biweekly (3 mg per kg) or every 4th week (6 mg per kg) for up to 2 years (Extended Data Fig. 1b). The primary objective was safety and feasibility. Secondary objectives were immunogenicity and clinical efficacy. Results Patients and treatment Thirty patients were enrolled from December 2017 to June 2020. None of the 30 patients dropped out of the study; all received at least three cycles of therapy (Supplementary Fig. 1). At the current database lock (5 October 2020), six patients were still on treatment in the trial. Of the 24 patients who were not on trial treatment at data cutoff, two are still receiving nivolumab monotherapy (6 mg per kg every 4 weeks). Reasons for stopping treatment for the remaining 22 patients included disease progression (37%), toxicity (20%), maximum benefit or complete response (CR) confirmed on two consecutive scans (17%) or completing 2 years of treatment (7%). For the 24 patients who were not on trial treatment at data cutoff, the mean number of vaccinations was 10.5 (range, 3–15 vaccinations). Thirteen of these 24 patients continued nivolumab (6 mg per kg, every 4 weeks) as a standard of care. Nine patients received subsequent therapy after progression (Supplementary Table 1). Baseline characteristics are shown in Table 1. The mean age was 70 years; 37% of patients had elevated lactate dehydrogenase (LDH) levels, 60% were stage M1c, 37% had BRAF mutations, and 43% were negative for PD-L1 (<1%). A total of three patients (10%) had received prior ipilimumab therapy. No patients had brain metastasis (Supplementary Table 2).Table 1 Baseline patient characteristics (n = 30) Characteristic Number (%) Mean age, years (range) 70 (46–85) Sex, male 16 (55%) ECOG PS 0 26 (87%) LDH levels  ≤ULN 19 (63%)  >ULN 11 (37%) M stage (AJCC-8)  M1a 6 (20%)  M1b 6 (20%)  M1c 18 (60%) Number of lesion sites  1 6 (20%)  2–3 17 (57%)  >3 7 (23%) Liver metastases present 10 (33%) BRAF status  Mutant 11 (37%)  Wild type 19 (63%) PD-L1  <1% 13 (43%)  >1% 17 (57%) Previous systemic therapy  Ipilimumab 3 (10%)  No 27 (90%) AJCC-8, eighth edition of the American Joint Committee on Cancer; ECOG PS, Eastern Cooperative Oncology Group performance status; ULN, upper limit of normal. Notable clinical responses to the combination therapy Thirty patients with MM were treated with the IDO/PD-L1 vaccine and nivolumab according to the trial protocol. By investigator review, the ORR reached 80% (CI, 62.7–90.5%), with 43% of patients (CI, 27.4–60.8%) achieving a CR and 37% (CI, 20.9–54.5%) reaching a partial response (PR) as the best overall response, while 20% experienced progressive disease (PD) according to Response Evaluation Criteria in Solid Tumors (RECIST) 1.1 (Fig. 1a). Two of the patients with a PR did not have a confirmatory scan stating PR on two consecutive scans. Early onset of response was frequent, with 22 of 30 patients having an objective response at the first evaluation (after 12 weeks on treatment). Median times to PR and CR were 75 d (range, 54–256 d) and 327 d (range, 73–490 d), respectively (Fig. 2a–c).Fig. 1 Clinical response. a, Pie charts with percent ORR, CR, PR and PD according to RECIST 1.1 by investigator review of all patients (n = 30), PD-L1+ patients (>1%, (n = 17)) and PD-L1− patients (<1%, n = 13)), respectively. Two-sided CIs (95%) were constructed using the Clopper–Pearson method. b, Treatment effect in MM1636 compared with a matched historical control group from the DAMMED database (n = 74). Patients in MM1636 (n = 29) were matched with the exact same combination variable according to age (≤70 years, >70 years), sex, LDH levels (normal, elevated), M stage (M1a, M1b, M1c), BRAF status (wild type, mutated) and PD-L1 status (<1%, ≥1%). Estimates for treatment effects were calculated by weighted logistic regression analyses and weighted Cox proportional hazard model. Bar height indicates the estimated response rate; tops of bars are centers for error bars. Odds ratios (OR), response rates and their corresponding 95% CIs were extracted from the regression model. All P values were two sided, and P values below 0.05 were considered statistically significant. c, Best change in the sum of target lesion size compared with that at baseline (n = 30). The horizontal line at −30 shows the threshold for defining an objective response in the absence of non-target disease progression or new lesions according to RECIST 1.1. Two patients with 100% reduction in target lesion size had non-target lesions present. White stars, six patients had normalization (<10 mm) of fluor-18-deoxyglucose (FDG)-negative lymph nodes (at baseline, lymph nodes were >1.5 cm and FDG+) and 100% reduction of non-lymph node lesions and are considered to have had a CR (green bar). Black star, one patient (MM29) was considered to have had a PR (blue bar), although he had not reached a −30% change in target lesion size. The patient had a single measurable 13-mm lung metastasis at baseline and multiple biopsy-verified cutaneous metastases on the left crus (not detectable by positron-emission tomography–computed tomography (PET–CT) at baseline). The best change in target lesion size was 10 mm, and a post-treatment biopsy from the cutaneous metastases showed no sign of malignancy. Thus, overall, the patient was classified as having a PR. d, Kaplan–Meier curve of the response duration in the 24 patients with an objective response. e, Kaplan–Meier curve of PFS in all 30 treated patients. f, Kaplan–Meier curve of OS in all 30 treated patients. Fig. 2 Clinical response. a, Swimmer plot showing response duration and time to response according to RECIST 1.1 for all treated patients (n = 30). Triangles indicate first evidence of PR, while squares indicate first evidence of CR. Closed circles indicate time of progression. Arrows indicate ongoing responses. Patient MM18 died due to nivolumab-induced side effects. b, Spider plot showing response kinetics in all treated patients (n = 30). Red squares indicate time of progression. c, PET–CT images of patient MM42 before and after treatment (after 12 series of treatment) showing FDG metabolism in target lesions. SC, subcutaneous. The ORR among PD-L1+ (>1% (clone 28.8)) patients (n = 17) was 94.1% (CI, 73–99.7%) and 61.5% (CI, 35.5–82.3%) in PD-L1− patients (n = 13) (Fig. 1a). Objective responses were observed in patients irrespective of human leukocyte antigen (HLA) genotype (Supplementary Fig. 2). Clinical response data were validated by blinded independent external review, in which an ORR of 76.6% (CI, 57.7–90.1%) was reported, with 53.3% of patients achieving a CR, 23.3% achieving a PR and 3.3% experiencing stable disease. Comparisons between investigator review and external review are outlined in Supplementary Table 3. To examine whether the very high response rate should be attributed to nivolumab or to the vaccine, we retrieved clinical response information from a matched historical control group using the Danish Metastatic Melanoma Database (DAMMED) from contemporaneously treated patients with stage 3–4 melanoma who received aPD1 monotherapy17. Patients were matched with the exact same combination variable according to age, sex, PD-L1 status, BRAF status, LDH level and M stage (those at stage M1d were excluded from the control group (no patients with brain metastasis)). Matched controls were identified for 29 patients, and the ORR of 79.3% (CI, 61.0–90.4%) observed in MM1636 was found to be significantly higher (P < 0.0012) than that in the matched control group in which an ORR of 41.7% (CI, 31.0–53.3%) was reached. Furthermore, of the 29 patients in MM1636, a significantly (P < 0.0017) higher percentage (41.4% (CI, 25.2–59.6%)) of patients achieved CR in MM1636 than that (12% (CI, 6.3–21.6%)) in the matched historical control group. ORR and complete response rate (CRR) in the matched historical control group were comparable to those of patients treated in randomized phase 3 pivotal trials with aPD1 monotherapy18 (Fig. 1b). The treatment leads to prolonged PFS At data cutoff, the median duration of response had not been reached, with 87% of all responding patients being free from progression at 12 months (Fig. 1d). Patients were followed for up to 35 months with a median follow-up time of 22.9 months (CI, 14.9–26.2 months). OS and PFS were calculated from the first day of treatment to death or progression or to the date of the last follow-up (5 October 2020). The median PFS (mPFS) for all treated patients was 26 months (CI, 15.4–69 months) and was not reached for responding patients (Fig. 1e). The median OS was not reached at the data cutoff. OS at 12 months was 81.6% (CI, 61.6–92%) (Fig. 1f). One patient (MM18) with a CR died from nivolumab-related severe adverse events; the remaining patients died because of metastatic melanoma. For comparison, the mPFS was 8.3 months (CI, 5.5 months–NR (not reached)) in the matched historical control group (n = 74), while the median OS was 23.2 months (CI, 23.2 months–NR). (Extended Data Fig. 2a,b). The combination of the IDO/PD-L1 vaccine and nivolumab was safe Treatment-related adverse events are listed for all 30 patients in Supplementary Table 4. Common treatment-related grade 1–2 toxicities were fatigue (47%), rash (47%), arthralgia (30%), diarrhea (23%), nausea (23%), dry skin (20%), pruritus (20%), infusion reaction (17%), xerostomia (17%) and myalgia (17%). Four patients (13%) experienced grade 3–4 adverse events: one patient with a grade 3 maculopapular rash (MM01), one patient with grade 3 adrenal insufficiency (MM06) and one patient with grade 3 arthralgia (MM22). Patient MM18 died from urosepsis with multi-organ failure and severe hyponatremia. This patient had experienced multiple immune-related adverse events with grade 3 colitis, grade 2 pneumonitis, grade 3 arthralgia, grade 2 vasculitis and grade 2 nivolumab infusion-related allergic reaction. Additionally, patient MM18 had symptoms of myocarditis at the time of death with highly elevated cardiac troponin I levels. Bedside echocardiography showed an ejection fraction of 15%, which at baseline was 60%, but an autopsy was not conducted, and myocarditis was never confirmed pathologically. Patient MM06 had received first-line treatment with ipilimumab before entering the trial and was on substitution corticosteroids at the time of inclusion. Adrenal insufficiency was aggravated by an erysipelas infection with high fever, reaching grade 3 in Common Terminology Criteria for Adverse Events and resolving quickly after appropriate antibiotic therapy was initiated. As expected, local side effects were common with 77% of the patients who developed injection site reactions. These reactions were classified as granulomas (63%), redness (20%), pain (13%) and pruritus (13%) at the injection site. All local reactions were grade 1–2, most likely related to the Montanide adjuvant and typically transient. However, two patients (MM07 and MM20) decided to discontinue vaccination after eight and 11 injections, respectively, due to granulomas, tenderness and pain that limited instrumental activities of daily living but continued to receive nivolumab (Extended Data Fig. 2c). Vaccine-specific responses in blood were frequently detected First, all 30 patients were assessed for the presence of vaccine-specific responses in peripheral blood mononuclear cells (PBMCs) before, on and after vaccination using a modified interferon (IFN)-γ enzyme-linked immune absorbent spot (ELISPOT) assay. This assay is known to expand antigen-specific memory cells and to improve the detection and the correlative power of ELISPOT during treatment19–21. Prevaccine IDO-specific responses were detectable in ten (33%) patients, while prevaccine PD-L1-specific responses were detectable in eight patients (27%); overlapping (specific to both IDO and PD-L1) prevaccine responses were present in four (13.3%) patients. During vaccination, an increase of IDO-specific T cells or PD-L1-specific T cells in the blood was observed in 28 (93%) and 26 (86%) patients, respectively. In total, 93% of patients had an increase in either PD-L1- or IDO-specific responses on vaccination (Fig. 3a), with a significant (P < 0.0001) median increase from baseline to the post-vaccine response demonstrated for both IDO and PD-L1 (at different time points on treatment), confirming that vaccine-specific immune responses were induced in patients regardless of clinical response (Fig. 3b,c). Immune responses fluctuated in the blood over time (Extended Data Fig. 3a). An increase in IDO- and PD-L1-specific responses in the peripheral blood was also detectable directly ex vivo across the different clinical response groups (Extended Data Fig. 4).Fig. 3 Vaccine-specific responses in blood. a, IDO- and PD-L1-specific T cell responses in PBMCs at baseline and on vaccination as measured by the IFN-γ ELISPOT assay (n = 30). *Responses were calculated as the difference between the average numbers of spots in wells stimulated with IDO or PD-L1 peptide (triplicates) and those from the corresponding control (DMSO), and statistical analyses of ELISPOT responses were performed using a distribution-free resampling method (Moodie et al.50). DR (double response), response was not statistically confirmed due to replicate number, but the number of spots in peptide wells was two times higher than that in control wells (DMSO). NS, no significant response and no DR. For a detailed overview of responses at serial time points on vaccination, see Extended Data Fig. 4a. b, IDO- and PD-L1-specific T cell response in PBMCs in all treated patients measured by the IFN-γ ELISPOT assay at baseline and on vaccination. On-vaccination responses were selected from the ‘best’ ELISPOT response at different time points for each patient (series 3, 6, 12, 18 or 24) during vaccination (n = 30). Wilcoxon matched-pairs signed-rank test was used to compare responses to IDO or PD-L1 peptides in the vaccine between baseline and later time points. c, Representative example of ELISPOT wells with response in patient MM23 in serial PBMCs before and on treatment. d, IDO-specific CD4+ and CD8+ T cells were isolated and expanded from PBMCs stimulated in vitro with the IDO peptide and a low dose of IL-2 for 14–15 d before sorting using the Miltenyi Cytokine Secretion Assay—Cell Enrichment and Detection kit. To assess their cytolytic potential, IDO-specific T cells were stimulated with IDO peptide, and expression of CD107a, IFN-γ and TNF-α was assessed by flow cytometry (the example is from patient MM14). Sustained vaccine-specific responses were observed 3 and 6 months after the last vaccine, indicating induction of memory responses in nine patients with response to clinical treatment who surpassed follow-up at data lock (Extended Data Fig. 3b). Importantly, PD-L1- and IDO-specific responses were observed irrespective of HLA genotype (Supplementary Table 5). To verify the functionality of vaccine-induced T cells, IDO- or PD-L1-specific T cells were isolated and in vitro expanded from PBMCs of five patients. Phenotypic characterization by flow cytometry revealed that the isolated vaccine-specific T cells consisted of both CD4+ and CD8+ T cells. Also, both IDO- and PD-L1-specific CD4+ and CD8+ T cells showed pro-inflammatory properties, as they expressed the cytolytic marker CD107a and secreted the cytokines IFN-γ and tumor necrosis factor (TNF)-α (Fig. 3d and Extended Data Fig. 6a–d). Interestingly, we were also able to detect vaccine-specific CD4+ and CD8+ T cell responses in peripheral blood ex vivo (Extended Data Figs. 4 and 5). We observed a significant increase in the overall percentage of CD107a, CD137 and TNF-α expression in response to peptide stimulation in on- or post-treatment PBMC samples compared to that at baseline, further confirming the expansion and the diverse signature of vaccine-specific T cells (Extended Data Fig. 5). Vaccine-specific responses detected in the skin at the vaccination site To investigate whether vaccine-specific T cells have the potential to migrate to peripheral tissue, delayed-type hypersensitivity (DTH) tests were performed after six cycles of treatment on 15 patients to assess the presence of vaccine-reactive T cells in the skin. Supplementary Table 6 display an overview of skin-infiltrating lymphocyte (SKIL) cultures. We detected IDO-specific T cells in the skin of six of ten patients and PD-L1-specific T cells in nine of 11 patients (Extended Data Fig. 7a). Intracellular cytokine staining was performed on SKILs from five patients after stimulation with either PD-L1 or IDO peptide. Here, we detected mainly CD4+ peptide-reactive T cells that secreted TNF-α and upregulated CD107a, and a minor fraction also secreted IFN-γ. In one patient, we detected CD8+ PD-L1-reactive T cells (Extended Data Fig. 7b–d). Vaccine-induced T cells specifically recognize target cells To confirm the functionality of vaccine-expanded T cells, vaccine-specific T cell clones (clonal purity was confirmed by T cell receptor (TCR) sequencing) were isolated and expanded from patient PBMCs (Extended Data Fig. 8e). We showed that PD-L1-specific T cells were able to recognize PD-L1+ autologous tumor cells in a PD-L1 expression-dependent manner if the cancer cells also expressed HLA-II (Fig. 4a,b). Similarly, an HLA-DR-restricted IDO-specific CD4+ T cell clone was able to recognize an HLA-DR-matched IDO-expressing model cell line, MonoMac1, in an IDO-expression-dependent manner (Fig. 4e–g). As previously described, IDO- and PD-L1-specific T cells’ mode of action is not limited to targeting only cancer cells. We were able to show that vaccine-specific T cell clones also reacted against PD-L1- and IDO-expressing autologous immune cells (Fig. 4c,h). To provide myeloid cells with a tumor-associated phenotype we treated isolated CD14+ myeloid cells with tumor conditioned medium (TCM) derived from an established autologous tumor cell line. We observed that such TCM-treated CD14+ cells had increased expression of PD-L1 and IDO and were effectively recognized by autologous PD-L1- and IDO-specific CD4+ T cell clones (Fig. 4c,d,h,i).Fig. 4 PD-L1- and IDO-specific T cells from vaccinated patients react against PD-L1- and IDO-expressing target cells. a, Left, PD-L1-specific T cell culture (MM1636.05) reactivity against PD-L1 peptide or autologous tumor cells in the IFN-γ ELISPOT assay. Tumor cells were either not treated or pretreated with 200 U ml−1 IFN-γ for 48 h before the assay. Effector:target (E:T) ratio of 10:1 was used. Right, PD-L1 and HLA-II surface expression on melanoma cells with (green) or without (yellow) pretreatment with IFN-γ compared to an isotype control (gray) as assessed by flow cytometry. b, Left, PD-L1-specific T cell (MM1636.05) reactivity in the IFN-γ ELISPOT assay against autologous tumor cells pretreated with IFN-γ (500 U ml−1) and transfected with mock or PD-L1 small interfering (si)RNA 24 h after transfection. E:T ratio, 10:1. Right, PD-L1 surface expression on melanoma tumor cells (MM1636.05) assessed by flow cytometry 24 h after transfection with mock (blue) or PD-L1 (red) siRNA compared to the isotype control (gray). c, Reactivity of the CD4+ PD-L1-specific T cell clone (MM1636.14) against PD-L1 peptide or autologous CD14+ cells; E:T ratio, 10:1. CD14+ cells were isolated using magnetic bead sorting and used as targets in an ELISPOT assay directly or after pretreatment for 2 d with TCM derived from the autologous tumor cell line. d, Quantitative PCR with reverse transcription (RT–qPCR) analysis of PD-L1 (CD274) expression in sorted CD14+ cells before and after treatment with autologous TCM for 48 h. e, Reactivity of the IDO-specific CD4+ T cell clone (MM1636.23) against IDO peptide combined with HLA-DR (L243)-, HLA-DQ (SPV-L3)- or HLA-DP (B7/21)-blocking antibodies (aHLA-DR, aHLA-DQ and aHLA-DP) in an intracellular staining assay (ICS) for IFN-γ and TNF-α production. T cells were incubated with individual blocking antibodies (2 μg ml−1) for 30 min before adding IDO peptide. f, Reactivity of the IDO-specific CD4+ T cell clone (MM1636.23) against the HLA-DR-matched IDO-expressing cell line MonoMac1 transfected with mock or IDO siRNA in an ICS assay, E.T ratio, 4:1. siRNA transfection was performed 48 h before the experiment. g, RT–qPCR analysis of IDO1 expression in MonoMac1 cells 48 h after siRNA transfection. h, Reactivity of the CD4+ IDO-specific T cell clone (MM1636.14) against IDO peptide or autologous CD14+ cells; E:T ratio, 20:1. CD14+ cells were isolated using magnetic bead sorting and used as targets in an ELISPOT assay directly or after pretreatment with TCM derived from the autologous tumor cell line. i, RT–qPCR analysis of IDO1 expression in sorted CD14+ cells before and after treatment with autologous TCM for 48 h. Bars in RT–qPCR data (d,g,i) represent the mean of three (d,i) or six (g) technical replicates ±s.d.; P values were determined by two-tailed parametric t-tests. ELISPOT counts (a,b,c,h) represent the mean value of three technical replicates ±s.e.m.; response P values were determined using the distribution-free resampling (DFR) method. TNTC, too numerous to count. T cell clones in blood and tumors To track the role of treatment-induced T cell responses, TCR sequencing of the complementarity-determining region 3 (CDR3) was performed on five patients in peripheral blood (baseline and cycles 3, 6 and 12) and paired biopsies. These five patients (MM01, MM02, MM08, MM09 and MM13) were selected due to the availability of material and to investigate a balanced patient group with both responders and non-responders. Due to a limited number of available paired biopsies (either as a consequence of patient refusal or unexpectedly rapid and substantial clinical responses), statistical analysis could not be applied. Details on clinical response are shown in Fig. 2a. Additionally, PBMCs (on treatment) or SKILs were stimulated with the IDO/PD-L1 peptides, and then cytokine-producing T cells were sorted to track vaccine-induced T cells both in the periphery and at the tumor site. To identify enriched IDO/PD-L1-specific T cell clones, TCR rearrangements in sorted IDO/PD-L1-specific clones and TCR rearrangements in sorted IDO/PD-L1-specific T cell samples were compared to sequences from baseline PBMC samples for each patient. Clonal expansion of vaccine-specific TCR rearrangements from samples on vaccination were then tracked using a differential abundance framework. Cumulative IDO/PD-L1-specific T cell frequencies were tracked in post-treatment samples. We found no relation between clinical response and the enrichment of vaccine-specific clones, but an increase in IDO/PD-L1-specific T cell clones was observed at different time points in the periphery in all five patients (Extended Data Fig. 3c). We next investigated overall changes in the T cell repertoire in the blood. A modest increase in the peripheral T cell fraction was observed in the three responding patients at cycle 3, while the two non-responding patients had a clear decrease in T cell fraction (Extended Data Fig. 8a). We thereafter investigated TCR clonality and TCR repertoire richness, exploring the proportion of abundant clones and the number of unique rearrangements, respectively. A decreasing peripheral Simpson clonality and increasing TCR repertoire richness was observed in responding patients at cycle 3, which might indicate tumor trafficking upon treatment. The opposite pattern was observed in non-responding patients (Extended Data Fig. 8b,c). Peripherally expanded clones were associated with tumors and persisted until cycle 12 (latest time point analyzed). The largest peripheral expansion was observed at cycle 3, with the most significant increase observed in patient MM01 (CR). Responding patients had a larger fraction of peripherally expanded clones that were also found in tumors compared to that of non-responders. By tracking peripherally expanded clones detected at the tumor site, we observed that patient MM01 had a substantial increase after treatment, indicating tumor trafficking of peripheral expanded clones (Extended Data Fig. 8d). Influx at the tumor site of vaccine-enriched T cell clones Given the observation of increased T cell fraction and enrichment of IDO- and PD-L1-specific clones in the blood after treatment, we investigated whether the same trend was observed at the tumor site. Both TCR sequencing and immunohistochemistry (IHC) of paired biopsies from the five patients described above showed an increase in the T cell fraction with an influx of CD3+ and CD8+ T cells after treatment in the three responding patients (Fig. 5a–c). IHC was not possible for patient MM09 due to tissue loss.Fig. 5 Changes in the TME after treatment. Number of CD3+ and CD8+ T cells, TCR fraction, TCR clonality, TCR repertoire, biopsy expanded TCR clones and enrichment of IDO- and PD-L1-specific T cells at the tumor site. a, Number of CD3+ and CD8+ T cells per mm2 (sum in the validated area) at the tumor site detected by IHC of paired biopsies from four patients. b, Example of IHC of CD3+ and CD8+ T cells at the tumor site before and after treatment (cells per mm2) in one patient (MM01). c, T cell fraction at the tumor site at baseline and cycle 6 by TCR sequencing. The T cell fraction was calculated by taking the total number of T cell templates and dividing by the total number of nucleated cells. d, Tracking of vaccine-associated clones at baseline and cycle 6 in tumor biopsies. Cumulative frequencies of IDO and PD-L1 vaccine-specific TCR rearrangements are represented. e,f, TCR clonality and TCR repertoire richness in five patients at the tumor site at baseline and cycle 6. Simpson clonality measures how evenly TCR sequences are distributed among a set of T cells, where 0 indicates an even distribution of frequencies and 1 indicates an asymmetric distribution in which a few clones dominate. TCR repertoire richness reports the mean number of unique rearrangements. g, Bar chart representing baseline expanded biopsy clones from five patients (colored bars) and the detection of biopsy expanded clones also found in the blood at baseline and series 3, 6 and 12 (white and gray bars) by TCR sequencing. We thereafter investigated whether some of the IDO/PD-L1 vaccine-linked T cells were present at the tumor site. Vaccine-associated clones were tracked as the combined frequency of IDO- and PD-L1-specific T cell rearrangements. In biopsies, the frequencies at cycle 6 were compared to those at baseline and showed an increase in vaccine-specific T cells in four of five patients, irrespective of clinical response (Fig. 5d). TCR sequencing of PD-L1-specific SKILs (a more specific culture than IDO/PD-L1-specific isolations derived from PBMCs on vaccination) and paired biopsies showed that two of the top five PD-L1-specific SKIL clones were present at the tumor site both before and after treatment (Extended Data Fig. 7e,f). With a focus on the more abundant T cell clones, we investigated overall TCR clonality at the tumor site before and after treatment. In addition, we explored the number of unique TCR rearrangements, dissecting the lower-frequency clones. Patient MM01 had a significant increase in TCR clonality and a decrease in repertoire richness at the tumor site after therapy, indicating a focused tumor repertoire response of selected clones. All three responding patients had a decrease in TCR repertoire richness, which again might indicate a focused tumor response (Fig. 5e,f). Deeper analyses showed that the T cell clones that expanded at the tumor site after therapy were also present in the blood at baseline and increased significantly after treatment in four of five patients. The highest proportion was detected early at cycle 3. These data again support trafficking of peripherally expanded clones to the tumor site and could indicate that the T cell response to treatment is derived from pre-existing peripherally tumor-associated T cells (Fig. 5g). Signs of treatment-induced inflammation in the TME To dissect changes in the TME induced by T cell influx upon treatment in responding patients, RNA gene expression analyses using the nCounter PanCancer Immune Profiling Panel from NanoString were performed on paired biopsies from two responding patients (MM01 and MM13). Expression of genes related to adaptive immunity such as T cell activation, effector functions (genes encoding IFN-γ, TNF-α, IL-15 and IL-18) and cytotoxicity was increased in post-treatment biopsies (Extended Data Fig. 9a,b). Also, expression of genes related to checkpoint inhibitors such as those encoding T cell immunoglobulin and mucin-domain containing-3 (TIM-3), IDO, PD-L1, PD-L2, PD-1 and CTLA-4 increased after treatment, indicating activation of immune cells in the TME (Extended Data Fig. 9c). Additionally, IHC of paired biopsies from four patients (MM01, MM02, MM05 and MM13) showed an upregulation of PD-L1, IDO, MHC-I and MHC-II on tumor cells, indicating a treatment-induced pro-inflammatory response in the three responding patients, except for a decrease in MHC-II expression in patient MM13. By contrast, the non-responding patient MM02 had a reduction in T cell numbers present in the tumor after treatment and no expression of PD-L1, IDO or MHC-II and, interestingly, total loss of MHC-I, demonstrating tumor immune escape (Extended Data Fig. 10a). CD8+ T cells and their distance (µm) to PD-L1-expressing cells in baseline biopsies from five patients was investigated by IHC. Except for patient MM13 (PR), distance and clinical responses were associated: the two responders had reduced distance (<20 µm) between cells expressing these markers compared to non-responding patients (>80 µm). This observation indicates that responding patients not only have a higher intratumoral infiltration of CD8+ T cells, but that these cells can surround and attack PD-L1-expressing immune cells and tumor cells (Extended Data Fig. 10b). Discussion In this clinical trial, MM1636, 30 patients with metastatic melanoma were treated with a first-in-class immunomodulatory IDO/PD-L1-targeting peptide vaccine combined with nivolumab. The treatment led to an unprecedented high ORR of 80%, with 43% of patients reaching a CR, and a striking mPFS of 26 months (95% CI, 15.4–69 months) was reached. The vaccine represents a new treatment strategy to activate specific T cells that target cells contributing to immune suppression (including tumor cells), positively modulating the TME by inducing local inflammation. Indeed, we show that vaccine-specific T cells isolated and expanded from vaccinated patients recognize not only tumor cells in a target- and HLA-restricted manner but also myeloid cells polarized toward a tumor-associated phenotype. Hence, myeloid cells become targets for vaccine-activated T cells when they have a tumor-associated phenotype in the TME. These phenomena may further induce checkpoint molecules and rewire the TME toward an increasingly aPD1-permissive state. A drawback of our study is the single-center nonrandomized setup. Comparison between trials or between patients in trials and real-world patients is problematic due to multiple factors, such as period of conduction and other therapies available at the different periods of time. Nevertheless, the rate of investigator-assessed ORR in the phase 3 trial CheckMate 067 was 43.7% in the nivolumab monotherapy group and 57% in the nivolumab and ipilimumab group. CRs occurred in 8.9% and 11.5% of patients, respectively18. The mPFS of 26 months (95% CI, 15.4–69 months) in this trial is more than twice as long as that for patients treated with nivolumab and ipilimumab in CheckMate 067, for which an mPFS of 11.5 months (95% CI, 8.7–19.3 months) was reached. Patient baseline characteristics were in general comparable to those of patients with MM who have been treated in CheckMate 067, although patients in MM1636 were older (mean age, 70 years) and a larger fraction were positive for PD-L1 (57%)3,18,22. Among patients with PD-L1-negative tumors in MM1636, an ORR of 61.5% was still reached, which would be expected to be around 33% for first-line nivolumab monotherapy23. Some studies suggest that older patients might have a tendency to respond better to aPD1 therapy; however, this is still debated24–27. To address potential trial bias and the nonrandomized setup, patients in MM1636 were matched for age, performance status, sex, M stage, LDH level, PD-L1 status and BRAF status with a historical control group from the DAMMED, who were treated contemporarily (2015–2019) with aPD1 monotherapy as standard of care17. We found a significantly higher ORR and CRR in MM1636 compared to those of matched patients, who had an ORR of 43% and a CRR of 13%, comparable to patients treated in CheckMate 067. Restrictions of the synthetic control group are of course that it is partially historic and patient selection outside matching criteria cannot be ruled out28. Numerous contemporary clinical trials are exploring the combination of aPD1 therapy with other immunomodulating agents for advanced melanoma. Talimogene laherparepvec, an oncolytic virus, is approved by the Food and Drug Administration and the European Medicines Agency to treat advanced melanoma. A small phase 1b trial with 21 patients (MASTERKEY-265) combined talimogene laherparepvec and pembrolizumab to treat patients with advanced unresectable melanoma and reached an ORR of 62% and a CR of 33%29,30. Seventy-one percent of the patients in this trial had an M stage below M1c; this number was 40% in our trial. Furthermore, mainly patients with an M stage below M1c responded to treatment, which was not the case in MM1636. Results from a large randomized phase 3 trial are awaited (KEYNOTE-034). Epacadostat, an IDO inhibitor, was tested in combination with pembrolizumab in a nonrandomized phase 2 trial in 40 aPD1 treatment-naive patients with MM with promising results, reaching an ORR of 62%. Unfortunately, the phase 3 trial showed no indication that epacadostat provided improvement in PFS and OS31. Limitations of the phase 3 trial were the sparse information on pharmacodynamics as well as biomarker evaluation to improve the design. The IDO/PD-L1 vaccine is different from epacadostat, as it is not an IDO inhibitor but targets IDO- and PD-L1-expressing cells. Similar vaccines administered as monotherapy induced objective responses in lung cancer and basal cell carcinoma, while epacadostat as monotherapy in 52 patients resulted in no responses32,33. Sahin et al. recently published encouraging data from a first-in-human trial, in which a vaccine containing liposomal RNA targeting four unmutated tumor-associated antigens (NY-ESO-1, MAGE-A3, tyrosinase and TPTE) was administered alone or in combination with nivolumab in patients with advanced melanoma. Responses were observed in both the monotherapy group as well as the combination group in checkpoint inhibitor-experienced patients, suggesting the efficacy of non-mutant shared tumor antigen vaccines34. The overall safety and tolerability findings are comparable to those of aPD1 monotherapy. Injection site reactions were exclusive to the vaccine. However, these side effects were transient and mild in most patients and most likely due to the adjuvant Montanide. IDO- and PD-L1-specific CD8+ and CD4+ T cells exist among peripheral blood lymphocytes in healthy donors35–38 and expand in response to pro-inflammatory stimuli35,39. Furthermore, both IDO and PD-L1 are induced in cells as a counter response to the inflammatory response. This provides a mechanism that ensures immune homeostasis, which keeps IDO/PD-L1-specific T cells in check; therefore, the risk of triggering autoimmune-related adverse events by vaccination appears to be minimal. This was confirmed in the first clinical trials of IDO and PD-L1 vaccination (NCT01543464 (refs. 33,40) and NCT03042793 (ref. 41)). The spontaneous (baseline) immune response to the vaccines observed in the current study is in agreement with our previous observation in various patients with cancer35–38. Numerous vaccine-induced changes in the blood and at the tumor site were observed. Peripheral IDO- and/or PD-L1-specific T cells were detected in vitro in a modified ELISPOT in over 93% of patients on vaccination, unrelated to patient HLA type. Immune responses were persistent in patients who surpassed follow-up at data cutoff and were still detectable up to 6 months after the last vaccine, suggesting induction of memory T cells. We did not observe a correlation between vaccine-induced responses in blood and clinical responses. However, the detection of a highly significant increase in vaccine-specific T cell numbers after vaccination in almost all patients together with the very low number of patients with PD makes it difficult to expect such a correlation, especially because other aspects should be taken into account (for example, the loss of class I expression on tumors cells in patient MM02 with PD after vaccination). Frequencies of peripheral T cells induced against PD-L1 were overall higher than those against IDO. Importantly, however, we also observed that it was not the same patients who responded strongly to both IDO and PD-L1, nor did patients react with a similar response pattern to the two antigens; that is, different time points of response were observed. This suggests that each component of the vaccine plays different roles in the ongoing immune response in patients. The goal of IDO/PD-L1 vaccination was to modulate the TME to increase responsiveness to aPD1 therapy, as we have observed in animal models, in which immune conversion is demonstrated in the TME with an increased influx of T cells. Targeting both IDO and PD-L1 together enables synergy, as the TME is known to use different immune escape mechanisms and IDO and PD-L1 are often overexpressed by different cellular compartments. Ex vivo ELISPOT and flow cytometry assays confirmed the induction of immune response toward both epitopes and the higher immunogenicity of the PD-L1 epitope, although because of lower sensitivity failed to detect some of the responses. TCR sequencing of five patients confirmed enrichment of IDO/PD-L1-specific T cell clones in the blood at different time points after treatment. Furthermore, an increase in enriched IDO/PD-L1-specific clones was observed in four of five patients at the tumor site after treatment, irrespective of clinical response. Phenotypic characterization showed that vaccine-specific T cells, which were expanded in vitro with interleukin (IL)-2 from the blood of vaccinated individuals, were both CD4+ T cells and CD8+ T cells. This was confirmed by ex vivo phenotype description of vaccine-activated T cells. Vaccine-specific T cells expressed CD107a and CD137 and produced IFN-γ and TNF-α upon stimulation with the cognate target, indicating their cytolytic capacity. Overall, the data from immune monitoring supports the importance of both antigens in the generation of the frequent clinical responses in the study. Despite a limited number of paired biopsies (due to either patient refusal or the fact that a large fraction of responding patients had no assessable tumors after six cycles), we observed trends indicating treatment-induced general T cell influx in responding patients. It was shown that proliferation of CD8+ T cells in the tumor after aPD1 treatment is associated with radiographic reduction in tumor size42. Additionally, we showed that a large proportion of expanded peripheral TCR clones were associated with tumors and the most considerable amount of clonal expansion was observed early, at cycle 3. For the patient with a CR included in the TCR sequencing analyses (MM01), the number of peripheral expanded clones present at the tumor site increased after treatment compared to that at baseline, indicating tumor trafficking of peripheral expanded clones. Gene expression analyses (two paired biopsies) and IHC (five paired biopsies) further demonstrated that the combination treatment induced a pro-inflammatory TME in responding patients with signs of T cell activation and cytotoxicity and increased cytokine activity. This may lead to further upregulation of IDO, PD-L1, MHC-I and MHC-II on tumor cells, leading to more treatment targets. It was shown that, following vaccination with a cancer vaccine, PD-L1 expression is increased on tumor cells due to recruitment of tumor-specific T cells and upregulation of adaptive immune resistance pathways in the TME43. Treatment with nivolumab monotherapy enhances PD-L1 expression, and it is therefore problematic to discriminate the effect of the vaccine as compared to that of nivolumab44. IDO- and PD-L1-specific pro-inflammatory effector T cells were hypothesized to counteract the functions of IDO- and PD-L1-expressing immune-suppressive cells as a means to keep the immune balance between immune activation and inhibition45. However, because the expression of these molecules is induced as a counter-regulatory response to inflammatory mediators such as IFNs, they can also be expressed by, for example, activated myeloid cells or T cells. Thus, activation of PD-L1-specific T cells may result in depletion of activated T cells in the tumor and other sites. Importantly, it was recently described that PD-L1+ T cells mainly have tolerogenic effects on tumor immunity and exert tumor-promoting properties, suggesting that targeting this immune population is indeed also beneficial46. We have previously investigated the effect of activating PD-L1-specific T cells in vitro and in vivo and found that, overall, they supported the effector phase of an immune response by removing PD-L1-expressing regulatory immune cells that inhibit PD-1+ effector T cells47,48. The major role of the PD-1 pathway is believed to be the regulation of effector T cell responses. Thus, this protective pathway is more important after activation, rather than at the initial T cell-activation stage49. Accordingly, the presence of PD-L1-specific T cells during the activation phase of an immune response may not increase or support a pro-inflammatory immune response due to the expression of PD-L1 on potent antigen-presenting cells or on the T cells themselves14. Thus, the overall effects of PD-L1-specific T cells may vary depending on the expression of both PD-1 and PD-L1, that is, due to the microenvironment and the state of the immune response. In conclusion, here we report an impressive response rate, CR rate and mPFS for a first-in-class immune-modulating vaccine combined with nivolumab. This may be a first step toward a new treatment strategy for patients with MM. Limitations are the low number of patients treated at a single institution and the lack of a randomized design with aPD1 monotherapy as comparator. Studies in aPD1 therapy-resistant or -refractory melanoma are ongoing as well as biomarker analysis for selecting patients at a higher likelihood to benefit from the combination versus aPD1 monotherapy. A larger randomized trial will be essential to validate these findings and determine the specific contribution of the vaccine to clinical responses and changes in the TME. In December 2020, the Food and Drug Administration granted breakthrough therapy designation for the IO102/IO103 vaccine combined with aPD1 therapy in metastatic melanoma based on data from the MM1636 trial. Methods Trial design and treatment plan MM1636 is an investigator-initiated, nonrandomized, open-label, single-center phase 1/2 study. All patients were treated at the Department of Oncology, Herlev and Gentofte Hospital, University of Copenhagen, Herlev, Denmark. The study was conducted according to the Declaration of Helsinki and Good Clinical Practice (GCP) and monitored by the GCP unit in Copenhagen, Denmark. The protocol was approved by the ethical committee of the Capital region of Denmark (H-17000988), the Danish Medical Agencies (2017011073) and the Capital Region of Denmark Data Unit (P-2019-172). The study was registered at https://clinicaltrials.gov/ under the identifier NCT03047928 and at the EudraCT (no. 2016-0004527-23). This study initially aimed to include 30 aPD1 treatment-naive patients with MM. An amendment with the addition of two other cohorts with ten patients in each cohort was made to evaluate immune responses and clinical efficacy in aPD1 therapy-resistant patients (cohort B, de novo resistance and cohort C, acquired resistance) for a total of 50 patients. The amendment with cohorts B and C was approved at the inclusion of 18 patients in cohort A. The trial is still including patients in cohorts B and C. This article reports results from cohort A. The first six patients were treated as for phase 1, evaluating for safety and tolerability before the remaining 24 patients were included in phase 2. The IDO/PD-L1 vaccine was administered subcutaneously biweekly for the first 10 weeks and thereafter every 4 weeks for approximately 9 months. A maximum of 15 vaccines were administered. Nivolumab was administered according to the approved label (3 mg per kg biweekly) for 24 cycles. The 15th vaccine was administered together with the 24th nivolumab cycle of 3 mg per kg, and responding patients thereafter continued nivolumab monotherapy every 4th week (6 mg per kg) as a standard of care after investigator assessment. Treatment with nivolumab was discontinued at the maximum benefit (assessed by the investigator), after a maximum of 2 years of therapy, at progression or due to severe adverse events (Extended Data Fig. 1b, treatment plan). Vaccine composition Each vaccine was composed of 100 µg IO102, a 21-amino-acid peptide (DTLLKALLEIASCLEKALQVF) from IDO, and 100 µg IO103, a 19-amino-acid peptide (FMTYWHLLNAFTVTVPKDL) from the signal peptide of PD-L1 (PolyPeptide Laboratories). Peptides were dissolved separately in 50 µl DMSO, filtered for sterility and frozen at −20 °C (NUNC CyroTubes CryoLine System Internal Thread, Sigma-Aldrich). At <24 h before administration, peptides were thawed. The PD-L1 peptide was diluted in 400 µl sterile water and, immediately before injection, mixed with the IDO peptide solution and 500 µl Montanide ISA-51 (SEPPIC) to achieve a total volume of 1 ml. Patients Patients above 18 years of age with locally advanced or stage 4 melanoma according to the AJCC (seventh edition), at least one measurable lesion according to RECIST 1.1 and an ECOG PS of 0–1 were eligible. The main exclusion criteria were prior treatment with aPD1 therapy, CNS metastases >1 cm, severe comorbidities and active autoimmune disease. Enrollment was not restricted to PD-L1 status, but it was known before inclusion. Patients were included after informed consent. Patients MM42 and MM20 have confirmed their approval of PET–CT images and clinical images that have been included in the article. Key study assessments Safety and tolerability were evaluated based on changes in clinical laboratory analyses and reported adverse events. Adverse events were assessed according to Common Terminology Criteria for Adverse Events (version 5.0) and were graded from 1 to 5 for all treated patients up to 6 months after the last dose of the IDO/PD-L1 vaccine. Clinical efficacy was assessed using FDG PET–CT scans before treatment and every 3rd month until progression. Objective responses were categorized into CR, PR, stable disease or PD according to RECIST version 1.1. Clinical data were collected at CCIT-DK in the eCRF program OpenClinica version 1.0 and in Microsoft Excel version 2002 on a secure server. Blood samples for immunologic analyses were collected before treatment, before the third cycle, after the sixth, 12th, 18th and 24th cycles (on vaccination) and 3 and 6 months after the last vaccine. Two to three tumor needle biopsies (1.2 mm) were collected at baseline and after six cycles from the same tumor site, when accessible, to evaluate immune responses at the tumor site. DTH skin tests were performed and punch biopsies were taken from the DTH area after cycle 6 for the evaluation of SKILs reactive to PD-L1 and IDO (Supplementary Fig. 1, treatment plan). Statistical analysis of clinical outcome Survival curves were computed in GraphPad Prism software version 9.0.0 according to Kaplan–Meier method. Median follow-up time of enrollment was calculated using the reverse Kaplan–Meier method, also in GraphPad Prism software version 9.0.0. For binary outcomes, 95% two-sided CIs were constructed using the Clopper–Pearson method, also in GraphPad version 9.0.0. An independent board of certified and experienced oncoradiologists performed an external review to evaluate clinical response to address the potential bias of investigator site review. The external review took place at Rigshospitalet, Copenhagen University Hospital. This hospital did not participate in the MM1636 trial, and the external reviewer had no prior knowledge about the clinical trial or the trial therapy. Only PET–CT images were accessed, and arrows indicating target or non-target lesions appeared on baseline images as the only additional information. To address potential trial bias regarding treatment effect, we matched patients in MM1636 with patients from the DAMMED, a population-based database that retrospectively collects data on patients with metastatic melanoma in Denmark. Here, data from 938 patients treated with aPD1 monotherapy contemporaneously (January 2015–October 2019) were extracted. Two hundred and eighteen of these patients were eligible for comparison and matching (all parameters available) (Supplementary Table 1), and 74 patients from the DAMMED were found to match. Patients were matched for age (≤70 years, >70 years), sex, LDH levels (normal, elevated), M stage (M1a, M1b, M1c), BRAF status (wild type, mutated) and PD-L1 status (<1%, ≥1%). An exact matching algorithm was used in which patients in MM1636 were matched with patients from the DAMMED with the same combination of variables. Twenty-nine patients from MM1636 were matched with exact combinations of the six variables. One patient could not be matched. To secure balance of the calculations, control patients were weighted according to the number of patients for each MM1636 patient. Estimates for treatment effects were calculated by weighted logistic regression analyses and weighted Cox proportional hazard model. The R package ‘Matchlt’ was used for matching patients. As the method chosen for matching control patients to protocol patients, a weighted binary logistic regression model was used for comparing response rates in the two matched cohorts. Odds ratios and response rates, including their corresponding 95% CIs, were extracted from the regression models. All P values were two sided, and P values below 0.05 were considered statistically significant. SAS version 9.4M5 was used for the weighted logistic regression models. Processing of peripheral blood mononuclear cells Peripheral blood was collected from all patients in heparinized tubes and was processed within 4 h. In brief, PBMCs were isolated using Lymphoprep (Medinor) separation. PBMCs were counted on a Sysmex XP-300 analyzer and frozen in Human AB Serum (Sigma-Aldrich, H4522-100ML) with 10% DMSO using controlled-rate freezing (CoolCell, BioCision) in a −80 °C freezer and were moved the next day to a freezer at −140 °C until further processing. Needle biopsies at baseline and after the sixth vaccine Two to three needle biopsies (1.2 mm) were taken at baseline and after the sixth cycle of treatment when assessable from the same tumor lesion. One fragment was fixed with formalin and embedded in paraffin (FFPE); one to two fragments were used for expanding tumor-infiltrating T cells and establishing autologous tumor cell lines. Delayed-type hypersensitivity and generation of skin-infiltrating lymphocytes After six cycles of treatment, we performed intradermal injections of vaccine components without adjuvant and one control injection containing DMSO without peptide. Patients were injected with either a mixture of both IDO and PD-L1 peptides at all three injection sites or PD-L1 peptide, IDO peptide or a mixture of the two at the injection sites, respectively (Supplementary Table 6). Eight hours after injection, punch biopsies were resected from the three sites where the peptide was injected and transported immediately to the laboratory and cut into fragments of 1–2 mm3. SKILs were expanded to establish ‘young SKILs’ in CM medium consisting of RPMI 1640 with GlutaMAX, 25 mM HEPES, pH 7.2 (Gibco, 72400-021), IL-2 (100 or 6,000 IU ml−1) (Proleukin Novartis, 004184), 10% heat-inactivated Human AB Serum (Sigma-Aldrich, H4522-100ML), 100 U ml−1 penicillin, 1.25 µg ml−1 Fungizone (Bristol Myers Squibb, 49182) and 100 µg ml−1 streptomycin (Gibco, 15140-122). Half of the medium was replaced three times per week. Young SKILs from samples derived from IDO peptide, PD-L1 peptide and the mixture injection sites were further expanded in a small-scale version of the 14-d rapid expansion protocol as previously described51. Quantification of specific T cells in blood by ELISPOT For enumeration of vaccine-specific T cells in the peripheral blood, PBMCs from patients were stimulated with IDO or PD-L1 peptide in the presence of a low dose of IL-2 (120 U ml−1) for 7–13 d before being used in IFN-γ ELISPOT assays with 2.5–3.2 × 105 cells per well. Briefly, cells were placed in a 96-well PVDF membrane-bottomed ELISPOT plate (MultiScreen MSIPN4W50, Millipore) precoated with IFN-γ-capture antibody (clone 1-D1K, Mabtech). Diluted IDO or PD-L1 peptide stocks in DMSO were added at 5 µM; an equivalent amount of DMSO was added to control wells. PBMCs from each patient were set up in duplicate or triplicate for peptide and control stimulations. Cells were incubated in ELISPOT plates in the presence of the peptide for 16–18 h, after which they were washed off, and biotinylated secondary antibody (anti-human IFN-γ mAB, clone 7-B6-1, Mabtech) was added. After 2 h of incubation, unbound secondary antibody was washed off, and streptavidin-conjugated alkaline phosphatase (Mabtech) was added for 1 h. Next, unbound streptavidin-conjugated enzyme was washed off, and the assay was developed by adding BCIP/NBT substrate (Mabtech). ELISPOT plates were analyzed on the CTL ImmunoSpot S6 Ultimate V analyzer using ImmunoSpot software version 5.1. Responses were calculated as the difference between the average numbers of spots in wells stimulated with IDO or PD-L1 peptide and those from corresponding control wells. For detection of IDO and PD-L1 peptide responses ex vivo, PBMCs were thawed and rested overnight in medium containing DNase I (1 μg ml−1, Sigma-Aldrich, 11284932001) before being used in IFN-γ ELISPOT assays as described above. A total of 6–9 × 105 PBMCs per well were seeded. Statistical analysis of all ELISPOT responses was performed using the DFR method as described by Moodie et al. using RStudio software (RStudio Team, 2016, http://www.rstudio.com/)50. Vaccine-specific ELISPOT responses were defined as true if the difference between the spot count in control and peptide-stimulated wells was statistically significant according to DFR statistical analysis or, for samples performed in duplicate, if the spot count in peptide-stimulated wells was at least 2× the spot count in control wells50. Cancer cell lines and tumor conditioned medium Autologous melanoma cell lines were established from needle biopsies. Briefly, biopsies were chopped into small fragments and seeded in 24-well cultures in RPMI 1640 with GlutaMAX, 25 mM HEPES, pH 7.2 (Gibco, 72400-021), 10% heat-inactivated FBS (Life Technologies, 10500064), 100 U ml−1 penicillin, 1.25 µg ml−1 Fungizone (Bristol Myers Squibb, 49182) and 100 µg ml−1 streptomycin (Gibco, 15140-122). Established adherent melanoma tumor cell lines were cryopreserved at −140 °C in freezing medium containing FBS with 10% DMSO. PD-L1 and HLA-II expression on established tumor cell lines was assessed by flow cytometry staining with anti-PD-L1–PE-Cy7 (1:23 dilution, clone M1h1 (RUO), BD, 558017) and anti-HLA-II–FITC (1:23 dilution, clone Tu39 (RUO), BD, 555558) antibodies. To obtain TCM, established tumor cell lines were cultured in 175-cm2 Nunc cell culture flasks until 80–90% confluency was reached. The culture medium was then replaced with 20 ml fresh X-VIVO 15 with Gentamicin and Phenol Red (Lonza, BE02-060Q), medium with 5% heat-inactivated Human AB Serum (Sigma-Aldrich, H4522-100ML). After 24 h of incubation, TCM was collected and centrifuged to remove any resuspended cells, after which TCM was aliquoted, frozen and stored at −80 °C. The acute monocytic leukemia cell line MonoMac1 was obtained from the DSMZ—German Collection of Microorganisms and Cell Cultures (ACC 252) and cultured in RPMI 1640 with GlutaMAX, 25 mM HEPES, pH 7.2 (Gibco, 72400-021) and 10% heat-inactivated FBS. Isolation of autologous myeloid cells Autologous CD14+ cells were sorted from freshly thawed PBMCs using a magnetic bead-separation kit (Miltenyi Biotec, 130-050-201) according to the manufacturer’s instructions. Isolated CD14+ cells were used as targets in the IFN-γ ELISPOT assay directly after sorting or differentiated in vitro into tumor-associated macrophages by culturing with 1 ml fresh X-VIVO 15 medium with Gentamicin and Phenol Red (Lonza, BE02-060Q) and 5% heat-inactivated Human AB Serum, supplemented with 1 ml autologous TCM in 24-well plates for 2 d. Quantification of vaccine-specific T cells from DTH biopsy sites by ELISPOT SKILs expanded in vitro were rested in medium without IL-2 overnight before being used in the IFN-γ ELISPOT assay as described above to evaluate reactivity of skin-infiltrating T cells. Generation of IDO- and PD-L1-specific T cell cultures from PBMCs or SKILs IDO- or PD-L1-specific T cells were isolated from peptide-stimulated in vitro PBMC cultures on days 14–15 after stimulation or from SKIL cultures expanded in vitro. For specific T cell isolation, PBMCs or SKILs were stimulated with IDO or PD-L1 peptide, and cytokine-producing T cells were sorted using the IFN-γ or TNF-α Secretion Assay—Cell Enrichment and Detection kit (Miltenyi Biotec). Cytokine-production profile of PD-L1- and IDO-specific T cells by intracellular staining To assess the T cell cytokine-production profile, isolated and expanded IDO- and PD-L1-specific T cell cultures were stimulated for 5 h with peptide at 5 μM in a 96-well plate. One hour after the start of the incubation, anti-CD107a–PE (1:133 dilution, clone H4A3, BD Biosciences, 555801) antibody and BD GolgiPlug (1:1,000 dilution, BD Biosciences) were added. After a 5-h incubation, cells were stained using fluorescently labeled surface marker antibodies: anti-CD4–PerCP (1:5 dilution, clone SK3, 345770), anti-CD8–FITC (1:5 dilution, clone SK1, 345772) and anti-CD3–APC-H7 (1:20 dilution, clone SK7, 560275) (all from BD Biosciences). Dead cells were stained using FVS510 (1:250 dilution, BD Biosciences, 564406), followed by overnight fixation and permeabilization using eBioscience Fixation/Permeabilization buffers (00-5123-43, 00-5223-56) according to the manufacturer’s instructions. Cells were then stained intracellularly in eBioscience permeabilization buffer (00-8333-56) with anti-IFN-γ–APC (1:20 dilution, clone 25723.11, BD Biosciences, 341117) and anti-TNF-α–BV421 (1:100 dilution, clone Mab11, BD Biosciences, 562783) antibodies. Samples were analyzed on the FACSCanto II (BD Biosciences) using BD FACSDiva software version 8.0.2. The gating strategy is shown in Supplementary Fig. 3a. For assessing the reactivity against IDO expressing MonoMac1 cells, the appropriate amount of cancer cells was added to the IDO-specific T cells to obtain an effector:target ratio of 4:1 in the 96-well plate and T cell cytokine production was tested after 5 h of co-culture as described above. For HLA-blocking experiments, HLA-blocking antibodies (2 μg ml−1) were added directly into wells 30 min before the addition of peptide. Blocking antibodies used were HLA-DR (1:500 dilution, clone L243, Abcam, ab136320), HLA-DQ (1:500 dilution, clone SPV-L3, Abcam, ab23632) and HLA-DP (1:500 dilution, clone B7/21, Abcam, ab20897). To assess ex vivo T cell reactivity to IDO and PD-L1 peptides in patient PBMCs, cells were thawed and rested for 1–2 d in medium containing DNase I (1 μg ml−1, Sigma-Aldrich, 11284932001). PBMCs were then stimulated with peptide at 5 μM in a 96-well plate for 8 h. An hour after the addition of peptide, anti-CD107a–BV421 (3:50 dilution, clone H4A3, BD Biosciences, 328626) antibody and BD GolgiPlug (1:1,000 dilution, BD Biosciences) were added. Surface and intracellular staining was performed as described above. Antibodies used for surface staining were anti-CD3–PE-CF594 (0.8:30 dilution, clone UCHT1, BD Biosciences, 562280), anti-CD4–BV711 (1:30 dilution, clone SK3, BD Biosciences, 563028) and anti-CD8–Qdot605 (1:150 dilution, clone 3B5, Thermo Fisher, Q10009). Antibodies used for intracellular staining were anti-CD137–PE (1:20 dilution, clone 4B4-1 (RUO), BD Biosciences, 555956), anti-IFN-γ–PE-Cy7 (1.5:20 dilution, clone B27, BD Biosciences, 557643) and anti-TNF-α–APC (1:20 dilution, clone Mab11, BD Biosciences, 554514). Samples were acquired on the NovoCyte Quanteon (ACEA Biosciences) and analyzed using NovoExpress software version 1.4.1. To assess vaccine-specific T cell responses, background values observed in unstimulated PBMC samples were subtracted from values observed in peptide-stimulated conditions. Positive response value threshold was set at a difference of 0.2% from the background values. Based on this response cutoff, only TNF-α, CD107a and CD137 responses were detected in this assay. Statistical analysis comparing baseline with on-treatment or post-treatment cytokine profiles was performed using two-sided Wilcoxon matched-pairs signed-rank test. The gating strategy is shown in Supplementary Fig. 3b. siRNA-mediated PD-L1 and IDO1 silencing A stealth siRNA duplex for targeted silencing of PD-L1 (Invitrogen)52, a custom silencer select siRNA for targeted silencing of IDO1 (Ambion) and the recommended silencer select negative control (Ambion) siRNA for mock transfection were used. The stealth PD-L1 siRNA duplex consisted of the sense sequence 5′-CCUACUGGCAUU-UGCUGAACGCAUU-3′ and the antisense sequence 5′-AAUGCGUUCAGCAAAUGCCAGUAGG-3′. The three silencer IDO siRNA duplexes used were siRNA1 (sense sequence, 5′-ACAUCUGCCUGAUCUCAUATT-3′; antisense sequence, 5′-UAUGAGAUCAGGCAGAUGUTT-3′), siRNA2 (sense sequence, 5′-CCACGAUCAUGUGAACCCATT-3′; antisense sequence, 5′-UGGGUUCACAUGAUCG-UGGAT-3′) and siRNA3 (sense sequence, 5′-CGAUCAUGUGAACCCAAAATT-3′; antisense sequence, 5′-UUUUGGGUUCACAUGAUCGTG-3′). For PD-L1 or IDO1 silencing, cancer cells were electroporated with 0.025 nmol of each siRNA duplex as previously described53. For PD-L1-silencing experiments, cancer cells were treated with IFN-γ (500 U ml−1, PeproTech) 1 h after electroporation. Electroporated cells were used as target cells in ELISPOT and ICS assays 24 h or 48 h after siRNA electroporation. RT–qPCR Total RNA was extracted using the RNeasy Plus Mini kit (Qiagen, 74134) following the manufacturer’s instructions. RNA concentration was quantified using a NanoDrop 2000 (Thermo Fisher Scientific), and a total of 1,000 ng RNA was reverse transcribed with the High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems, 4368814) using 1,000 ng input RNA for transcription. Real-time qPCR analyses were performed using the TaqMan Gene Expression Assay on a Roche LightCycler 480 instrument. RT–qPCR assays were performed with a minimum of three technical replicates and analyzed using the ΔCt method as described in Bookout et al.54 with normalization of IDO1 expression (primer ID, Hs00984148_m1) or PD-L1 expression (primer ID, Hs001125296_m1) to the expression level of the housekeeping gene POLR2A (primer ID, Hs00172187_m1) and the control sample (mock). For low-concentration samples with no amplification, Ct was set to 40. Controls without reverse transcriptase were used as controls for specific amplifications. P values were determined using two-tailed parametric t-tests. BRAF-mutational status and PD-L1 status at baseline in all patients A library of historical FFPE biopsies were assessable for all patients and analyzed locally at Herlev and Gentofte University Hospital by experienced pathologists for BRAF status and PD-L1 expression on tumor cells. The BRAF analysis was carried out with real-time PCR using the EntroGen BRAF Mutation Analysis Kit II (BRAFX-RT64, CE-IVD) to specifically detect mutations corresponding to V600D, V600E and V600K in the BRAF protein. PD-L1 status was assessed using the monoclonal rabbit anti-PD-L1 antibody (clone 28.8, PD-L1 IHC 28-8 pharmDx) in FFPE biopsies following the manufacturer’s instructions. Patients were considered positive for PD-L1 with expression levels ≥1% and negative for PD-L1 with expression levels <1%. Human leukocyte antigen type Blood samples from all 30 patients were genotyped for class I (HLA-A, HLA-B, HLA-C) and three class II types (HLA-DRB1, HLA-DQA1, HLA-DQB1) using LinkSēq HLA Typing Kits (Thermo Fisher, 1580C). These test kits are based on real-time PCR using allele-specific exponential amplification (sequence-specific primers) followed by melting curve analyses. Immunohistochemistry simplex: Immunoscore CR IHC staining was performed using a qualified Ventana Benchmark XT with four different steps: (1) antigen retrieval; (2) staining with the following primary antibodies: anti-CD3 antibody (clone HDX2; provider, HalioDx; HD-FG-000013; 10931065/10636667; concentration, 0.25 µg ml−1), anti-CD8 antibody (clone HDX3; provider, HalioDx; HD-FG-000019; 10931069/10337710/10639301; concentration, 0.5 µg ml−1), anti-IDO monoclonal antibody (clone VINC3IDO; provider, Thermo Fisher Scientific; 14-9750-82; E25003-101; concentration, 0.05 µg ml−1), anti-HLA class 1 ABC antibody (clone, EMR8-5; provider, Abcam; ab70328; GR3248333/GR3186494; concentration, 0.5 µg ml−1), anti-HLA-DR/DP/DQ/DX antibody (clone CR3/43; provider, Santa Cruz; sc-53302; L1714; concentration, 1 µg ml−1) and anti-PD-L1 antibody (clone HDX3; provider, HalioDx; HD-FG-000035; 106312810/106312816; concentration, 3.3 µg ml−1); (3) detection with a secondary antibody using the ultraView Universal DAB Detection kit; and (4) counterstaining using Hematoxylin & Bluing Reagent (staining of cellular nuclei). Control slides were systematically included in each staining run to permit quality control of the obtained measurements. Following coverslipping, slides were scanned with the NanoZoomer-XR to generate digital images (20×). Two consecutive slides were specifically used to perform Immunoscore CR TL staining of CD3+ and CD8+ cells. Digital pathology of T lymphocytes The digital pathology for Immunoscore CR TL allowed the quantification of positive cells (in cells per mm2) into the core tumor and invasive margin if present. Each sample was analyzed using the HalioDx Digital Pathology Platform. Digital pathology and pathologist analysis of IDO Digital pathology for IDO allowed the quantification of stained area (in mm2) into the whole tumor. Each sample was analyzed using the HalioDx Digital Pathology Platform. Analysis of IDO+ cells was performed by a pathologist. Results were expressed as an H score from 0 to 300. The score is obtained by the formula 3 × percentage of cells with strong staining + 2 × percentage of cells with moderate staining + percentage of cells with weak staining. Digital pathologist analysis of MHC-I and MHC-II Analysis of MHC-I+ and MHC-II+ cells was performed by a pathologist. Results are expressed as a percentage of positive tumor cells and a percentage of positive cells in the stroma. Digital pathology Immunoscore immune checkpoint (CD8+ and PD-L1+) The Immunoscore CR IC test allowed the quantification of CD8++ cell density in the whole tumor and a CD8+-centered proximity index (which corresponds to the percentage of CD8++ cells that have at least one PD-L1+ cell in the neighborhood) at different cutoff distances (20 µm, 40 µm, 60 µm and 80 µm). Pathologist analysis of PD-L1 A pathologist performed analysis of PD-L1+ cells. Positivity of a viable tumor cell was considered when partial or complete cell membrane staining was observed (more than 10% of the tumor cell membrane). Results were expressed as a percentage. NanoString RNA profiling and Immunosign RNA was extracted from FFPE tissues using Qiagen RNeasy FFPE extraction kits (Qiagen). Annotations from the pathologist performing H&E staining were used to guide removal of normal tissue from the slides by macrodissection before nucleic acid extraction, which occurred after tissue deparaffinization and lysis. Each extracted RNA sample was independently quantified using a NanoDrop spectrophotometer (NanoDrop Technologies) and qualified (Agilent Bioanalyzer). Degradation was quantified as the percentage of RNA fragments smaller than 300 bp using the RNA 6000 Nano kit (Agilent Bioanalyzer). Good sample quality was defined as less than 50% of RNA fragments being 50–300 bp in size. RNA expression profiling was performed using the nCounter PanCancer Immune Profiling Panel from NanoString (NanoString Technologies). The PanCancer Immune Profiling Panel contains 776 probes and is supplemented with six genes to complete HalioDx Immunosign targets. Hybridization was performed according to the manufacturer’s instructions. Hybridized probes were then purified and immobilized on a streptavidin-coated cartridge using the nCounter Prep Station (NanoString Technologies). Data collection was carried out on the nCounter Digital Analyzer (NanoString Technologies) following the manufacturer’s instructions to count individual fluorescent barcodes and quantify target RNA molecules present in each sample. For each assay, a scan of 490 fields of view was performed. Raw data from the NanoString nCounter were processed using NanoString recommendations. Quality control enables to retain data of good quality with a binding density that ranges between 0.05 and 2.25. The linearity of positive controls was checked using the R2 value of regression between the counts and the concentrations of positive controls. Samples that showed R2 < 0.75 were flagged and removed from the analysis. The background was removed using a thresholding method at the mean + 2 s.d. of negative controls. Raw counts were normalized using a positive normalization factor. Samples showing positive normalization factors outside the range of 0.3–3 were removed from the analysis. A second normalization was performed using the housekeeping gene normalization factor. Only the most stable housekeeping genes were selected for this normalization step using the variance-versus-mean relationship. All samples showing a normalization factor outside the range of 0.1–10 were removed from the analysis. All statistical analyses were performed on normalized counts using R software (version 2.6.2, 2019-12-12). T cell receptor variable β chain sequencing To track longitudinal immune responses to therapy, genomic DNA was extracted from longitudinal pre- and post-treatment PBMCs (five patients), pre- and post-treatment biopsies (five patients) (both FFPE and treated with RNAlater) and IDO- and PD-L1-specific T cell cultures from PBMCs (five patients) or SKILs (one patient). Three clonal (two IDO- and one PD-L1-) specific cultures were also sequenced to confirm clonal purity of cultures. DNA from PBMCs or RNAlater-treated biopsies was extracted with the DNeasy Blood and Tissue kit (Qiagen, 69504), DNA from sorted IDO and PD-L1-specific T cells from either PBMCs or SKILs was extracted using the QIAamp DNA Micro kit (Qiagen, 565304), and DNA from PFFE biopsies was extracted using the Maxwell RSC DNA FFPE kit (Promega, AS1450). Immunosequencing of the CDR3 regions of human TCRβ chains was performed using the immunoSEQÒ Assay (Adaptive Biotechnologies). Extracted genomic DNA was amplified in a bias-controlled multiplex PCR, followed by high-throughput sequencing. Sequences were collapsed and filtered to identify and quantitate the absolute abundance of each unique TCRβ CDR3 region for further analysis as previously described55–57. Statistical analyses of TCRβ sequencing results Two quantitative components of diversity were compared across samples in this study. First, Simpson clonality was calculated on productive rearrangements by ∑i=1Rpi2, where R is the total number of rearrangements and pi is the productive frequency of rearrangement i. Values of Simpson clonality range from 0 to 1 and measure how evenly receptor sequences (rearrangements) are distributed among a set of T cells. Clonality values approaching 0 indicate a very even distribution of frequencies, whereas values approaching 1 indicate an increasingly asymmetric distribution in which a few clones are present at high frequencies. Second, sample richness was calculated as the number of unique productive rearrangements in a sample after computationally downsampling to a common number of T cells to control for variation in sample depth or T cell fraction. Repertoires were randomly sampled without replacement five times, and we report the mean number of unique rearrangements. The T cell fraction was calculated by taking the total number of T cell templates and dividing by the total number of nucleated cells. The total number of nucleated cells was derived from reference genes using the immunoSEQ Analyzer version 3. To identify enriched vaccine clones in each patient, rearrangement frequencies in their baseline PBMCs and each IDO/PD-L1-sorted T cell sample were compared using a binomial distribution framework as previously described58. In brief, for each clone, we performed a two-sided test to determine whether frequencies were the same in the patient’s periphery and a PD-L1- or IDO-specific T cell sample. The Benjamini–Hochberg procedure was used to control the false discovery rate at 0.01 (ref. 59). Clonal expansion in post-treatment samples was similarly assessed using this differential abundance framework, but an IDO/PD-L1-specific T cell sample was replaced with a post-treatment series sample. In biopsies, the six series frequencies were compared to those of baseline tissue. Lastly, vaccine-associated clones were tracked in each PBMC and tissue sample by summing the frequency of each rearrangement enriched in either PD-L1- or IDO-specific T cells. All statistical analyses were performed in R version 3.6.1. Reporting Summary Further information on research design is available in the Nature Research Reporting Summary linked to this article. Online content Any methods, additional references, Nature Research reporting summaries, source data, extended data, supplementary information, acknowledgements, peer review information; details of author contributions and competing interests; and statements of data and code availability are available at 10.1038/s41591-021-01544-x. Supplementary information Supplementary Information Supplementary Figs. 1–3 and Tables 1–6 Reporting Summary Extended data Extended Data Fig. 1 Mode of Action and treatment plan. a) Anticipated mechanism of action of the combination therapy of an IDO/PD-L1 derived peptide vaccine and nivolumab (anti-PD1). 1) The IDO/PD-L1 peptide vaccine is administered subcutaneously (s.c) and nivolumab is administered intravenous (IV). 2) The vaccines peptides are phagocytosed by an antigen presenting cell and presented to IDO and PD-L1 specific T cells, which are activated. 3) The activated T cells migrates to the tumor site where they attack both immune-suppressive cells and tumour cells expressing IDO and/or PD-L1 leading to cytokine production and a pro-inflammatory tumour microenvironment. 4/5) Enhanced tumour killing by both IDO/PD-L1 specific T cells and tumor specific cytotoxic T cells due to PD-1 blockade. Created with BioRender.com b) Treatment plan. After written informed content patients were screened. Before treatment start a baseline PET/CT scan was performed, baseline blood sample for research use and if assessable a baseline needle biopsy. Patients were treated with the IDO/PD-L1 peptide vaccine subcutaneously biweekly for the first 6 injections and thereafter every fourth week for a maximum of 15 vaccines. Nivolumab was administered in parallel biweekly (3mg/kg) up to 24 series. If patients needed subsequent nivolumab after ended vaccination regiment they were treated with 6 mg/kg every fourth week up to two years. Needle biopsy and delayed type hypersensitivity (DTH) was performed after 6 series of treatment if assessable. PET/CT scans was performed every third month. Extended Data Fig. 2 Progression free survival and overall survival in matched historical control group and vaccine injection site reaction. a) Kaplan-Meier curve of progression free survival with corresponding 95% confidence intervals in the matched historical control group (n=74). Patients were matched on BRAF-status, PD-L1-status, age, gender, M-stage and LDH level). b) Kaplan-Meier curve of overall survival with corresponding 95% confidence intervals in the matched historical control group (n=74). c) Images of injection site reaction in patient MM20 (CR) after 11 vaccination show redness, rash and granuloma at injection site. Extended Data Fig. 3 Vaccine specific responses in blood. a) Heatmap of specific (background has been subtracted) IDO and PD-L1 responses in PBMCs at baseline, series 3, 6, 12, 18 and 24 measured by in vitro IFN-γ Elispot assay show fluctuations in the blood during treatment (n=30). b) Heatmap of specific (background has been subtracted) IDO and PD-L1 responses in PBMCs at baseline and 3 and 6 months after last vaccine measured by in vitro IFN- γ Elispot assay (n=30). 2.5–3.2x105 cells per well were used. c) Vaccine associated clones were tracked in the blood by summing the frequency of each rearrangement enriched in either IDO or PD-L1 T cells. Extended Data Fig. 4 Ex vivo vaccine specific responses in blood. Heatmaps of detected specific (background has been subtracted) IDO (a) and PD-L1 (b) responses in PBMCs at baseline and on/after treatment as measured by IFNγ ELISPOT (n=25). C) Example of well images of ex vivo ELISPOT wells for three different patients (n=3). 6-9x105 cells per well were used. Extended Data Fig. 5 CD4 and CD8 vaccine specific T cell responses in blood. Top: heatmaps of IDO specific CD4 (left) and CD8 (right) T cell responses in PBMCs at baseline and on/post treatment. Bottom: heatmaps of PD-L1 specific CD4 (left) and CD8 (right) T cell responses. Responses quantified by flow cytometry by an increased expression of CD107α, CD137 and TNFα after 8h peptide stimulation. Values represent specific responses after background values have been subtracted (n=21). Statistical analysis were performed using two-sided Wilcoxon matched-paired rank test. Extended Data Fig. 6 Pro-inflammatory profiles of sorted CD4 and CD8 T cells from blood. a/b/c/d) Percentage of in vitro stimulated and sorted PD-L1 and IDO specific CD4 and CD8 T cells secreting cytokines. Extended Data Fig. 7 Vaccine specific responses in skin. a) IDO and PD-L1 specific T cell responses in SKILs after 6 series of treatment measured by IFN-γ Elispot assay (n=13). SKILs were grown from DTH injection with either IDO peptide, PD-L1 peptide or a mix as presented by different blue colours. * Responses were calculated as the difference between average numbers of spots in wells stimulated with IDO or PD-L1 peptide (triplicates) and corresponding control (DMSO) and statistical analyses of Elispot responses were performed using distribution-free resampling method (Moodie et al.). DR: Not statistically confirmed response due to replicate number but number of spots in peptide wells are two times higher than control wells (DMSO). NS: No significant response and no DR. b/c/d) Percentage of cytokine secreting/CD107a+ CD4+ and CD8+ IDO and PD-L1 specific T cells in response to in vitro peptide stimulation by flow cytometry. e) Skin infiltrating PD-L1 specific T cell clones also found in biopsy in patient MM01. TCR sequencing was performed on a PD-L1 specific T cell culture generated from DTH area on the lower back injected with PD-L1 peptide on patient MM01. Bars show the frequency of top 25 clones in the culture with which indicates a high Simpson clonality of 0.43. f) Tracking the frequency of the top five skin infiltrating PD-L1 specific clones in tumour before and after treatment. Extended Data Fig. 8 T cell changes in blood after treatment. T cell fraction, TCR clonality and repertoire richness in blood and peripherally expanded TCR clones in both responding and non-responding patients. a) T cell fraction in peripheral blood of five patients at baseline, series 3, 6 and 12 by TCR sequencing. T cell fraction was calculated by taking the total number of T cell templates and dividing by the total number of nucleated cells (n=5.) b) TCR clonality in peripheral blood of five patients at baseline, series 3, 6 and 12 by TCR sequencing (n=5). Simpson clonality measures how evenly TCR sequences are distributed amongst a set of T cells where 0 indicate even distribution of frequencies and 1 indicate an asymmetric distribution where a few clones dominate. c) TCR repertoire richness in peripheral blood of five patients at baseline, series 3, 6 and 12 by TCR sequencing (n=5). TCR repertoire richness report the mean number of unique rearrangements. d) Bar-chart chart representing dynamics of the expanded T cell clones. Coloured bars represent peripherally expanded clones in five patients at series 3, series 6 and series 12 as compared to the baseline PBMC samples (n=5). Light gray bar represents peripherally expanded clones that are present in baseline biopsy samples while dark gray bar represents peripherally expanded clones that are present in the post-treatment biopsy (after 6 series). e) Frequency of the dominant TCR β chain in clonal PD-L1 and IDO specific cultures as determined by CDR3 sequencing. Extended Data Fig. 9 Profiling of genes relevant to T cell activation, cytokines and exhaustion markers on pre- and post-treatment biopsies in two patients. a) RNA expression profiling of genes related to T cell activation was performed using NanoString nCounter (n=2). b) RNA expression profiling of genes related to cytokine activity was performed using NanoString nCounter (n=2). c) RNA expression profiling of genes related to checkpoint inhibitors was performed using NanoString nCounter (n=2). Extended Data Fig. 10 Treatment induced upregulation of PD-L1, IDO, MHC I and MHC II and distance between CD8 T cells and PD-L1 expressing cells. a) IHC on 4 paired biopsies stained for CD3 and CD8 T cells, PD-L1, MHC I and MHC II on tumor cells/mm2 (sum in the validated area) and IDO H-score (from 0 to 300). H-score is the expression of IDO on both immune and tumor cells: The score is obtained by the formula: 3 x percentage of cells with a strong staining + 2 x percentage of cells with a moderate staining + percentage of cells with a weak staining (n=4). b) Distance in µm between CD8+ T cells and PD-L1+ stained cells on five baseline biopsies detected by IHC (n=5). Extended data is available for this paper at 10.1038/s41591-021-01544-x. Supplementary information The online version contains supplementary material available at 10.1038/s41591-021-01544-x. Acknowledgements We thank all patients and their relatives for being a part of the trial. We thank Ö. Met, M. Jonassen, S. Ullitz Færch, B. Saxild, S. Wendt and C. Grønhøj for technical support. We thank M. Cumberbatch for input with translational analysis. We thank the nurses at clinic 5 and the head of the Oncology Department at Herlev and Gentofte Hospital, L. Sengeløv. The study was funded through a research funding agreement between IO Biotech and the CCIT-DK, Herlev Hospital and the Oncology Department at Herlev Hospital. Author contributions I.M.S. and M.H.A. conceived the study. J.W.K. wrote the protocol with input from I.M.S., M.H.A. and M.-B.Z. J.W.K. and C.L.L. were responsible for patient treatment and patient care with help from R.B.H. and consulting from I.M.S., E. Ellebaek and M.D. I.M.S., E. Ellebaek and M.D. were responsible for patient recruitment. J.W.K. and C.L.L. were responsible for coordinating trial procedures and collecting data and samples. J.W.K. and C.L.L. were clinical investigators, and I.M.S. was a sponsor. J.W.K. and C.L.L. analyzed and interpreted clinical data with support from I.M.S. and E. Ehrnrooth. T.W.K. performed statistical analyses. H.W.H. helped with clinical imaging. J.W.K., E.M., C.O.M., S.M.A., S.E.W.-B., M.O.H. and A.W.P. analyzed and interpreted translational data with support from M.H.A. and I.M.S. Funding was acquired by I.M.S., M.H.A. and M.-B.Z. J.W.K. wrote the manuscript with input from I.M.S., M.H.A., E.M. and A.W.P. All authors reviewed the manuscript, interpreted data and approved the final version. Data availability This clinical trial was registered at https://www.clinicaltrials.gov/ before patient enrollment (clinical trial identifier NCT03047928). Each vaccine was composed of 100 µg IO102, a 21-amino-acid peptide (DTLLKALLEIASCLEKALQVF) from the IDO peptide, and 100 µg IO103, a 19-amino-acid peptide (FMTYWHLLNAFTVTVPKDL) from the signal peptide of PD-L1. TCR sequencing data are available from Adaptive Biotechnologies. Upon request, the CCIT-DK office will provide a username and a password to access the designated data within approximately 2–4 weeks (https://www.herlevhospital.dk/ccit-denmark/find-us/Sider/Contact-information.aspx). All requests for the remaining data including raw data and analyzed data and materials will, within a reasonable time frame, be reviewed by the CCIT-DK office (https://www.herlevhospital.dk/ccit-denmark/find-us/Sider/Contact-information.aspx) to verify whether the request is subject to any intellectual property or obligations. Patient-related data not included in the paper were generated as part of clinical trials and may be subject to patient confidentiality. Any data and materials that can be shared will be released via a material-transfer agreement. The following database was used in the study: https://research.regionh.dk/da/publications/the-danish-metastatic-melanoma-database-dammed(32749d99-095f-4cae-b5de-769bae27f01e).html. Competing interests M.H.A. is named as an inventor on various patent applications relating to therapeutic uses of IDO and PD-L1 peptides. These patent applications are assigned to the company IO Biotech, which is developing immune-modulating cancer treatments. M.H.A. is a founder, shareholder and advisor for IO Biotech. I.M.S. is a cofounder, shareholder and advisor for IO Biotech. I.M.S. has an advisory board relationship with or lectured for Roche, Novartis, MSD, Celgene, Incyte, TILT Bio, Pfizer and BMS AstraZeneca and has received limited grants for translational research from BMS, Roche and Novartis. M.-B.Z. is the CEO, founder and shareholder at IO Biotech. E.M., A.W.P. and E. Ehrnrooth are employees at IO Biotech. E. Ellebaek has received honoraria from BMS, Pierre Fabre, Roche and Kyowa Kirin and travel support from MSD. M.D. has received honoraria from Genzyme, MSD, BMS, Roche and Novartis and travel support from Novartis, MSD, BMS, Roche and Pfizer. The remaining authors declare that they have no conflict of interest. Peer review information Nature Medicine thanks George Coukos, Michael Schell and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. Javier Carmona was the primary editor on this article and managed its editorial process and peer review in collaboration with the rest of the editorial team. Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. These authors contributed equally: Julie Westerlin Kjeldsen, Cathrine Lund Lorentzen. Change history 3/8/2022 A Correction to this paper has been published: 10.1038/s41591-022-01771-w

NIVOLUMAB

DrugsGivenReaction

CC BY

34887574

2021-12-09

What was the administration route of drug 'NIVOLUMAB'?

A phase 1/2 trial of an immune-modulatory vaccine against IDO/PD-L1 in combination with nivolumab in metastatic melanoma. Anti-programmed death (PD)-1 (aPD1) therapy is an effective treatment for metastatic melanoma (MM); however, over 50% of patients progress due to resistance. We tested a first-in-class immune-modulatory vaccine (IO102/IO103) against indoleamine 2,3-dioxygenase (IDO) and PD ligand 1 (PD-L1), targeting immunosuppressive cells and tumor cells expressing IDO and/or PD-L1 (IDO/PD-L1), combined with nivolumab. Thirty aPD1 therapy-naive patients with MM were treated in a phase 1/2 study ( https://clinicaltrials.gov/ , NCT03047928). The primary endpoint was feasibility and safety; the systemic toxicity profile was comparable to that of nivolumab monotherapy. Secondary endpoints were efficacy and immunogenicity; an objective response rate (ORR) of 80% (confidence interval (CI), 62.7-90.5%) was reached, with 43% (CI, 27.4-60.8%) complete responses. After a median follow-up of 22.9 months, the median progression-free survival (PFS) was 26 months (CI, 15.4-69 months). Median overall survival (OS) was not reached. Vaccine-specific responses assessed in vitro were detected in the blood of >93% of patients during vaccination. Vaccine-reactive T cells comprised CD4+ and CD8+ T cells with activity against IDO- and PD-L1-expressing cancer and immune cells. T cell influx of peripherally expanded T cells into tumor sites was observed in responding patients, and general enrichment of IDO- and PD-L1-specific clones after treatment was documented. These clinical efficacy and favorable safety data support further validation in a larger randomized trial to confirm the clinical potential of this immunomodulating approach. pmcMain Despite remarkable advances in the treatment of MM with immune checkpoint inhibitors (ICIs) targeting PD-1 and cytotoxic T lymphocyte antigen 4 (CTLA-4), around half of patients are resistant to ICI monotherapy1. The combination of anti-CTLA-4 (aCTLA-4) and aPD1 therapy is to date the most effective therapy resulting in a response rate of around 60%; however, 50% of the patients also develop severe adverse events2,3. Therefore, an equally effective but less toxic treatment is highly needed. Several approaches to enhance ICI efficacy are currently being investigated, such as other ICIs, T cell therapy with tumor-infiltrating T cells or innate immunity stimulators such as Toll-like receptor 9 agonists4–6. Treating cancer patients with vaccines that stimulate a targeted immune response is another attractive approach, with very few side effects observed thus far in combination immunotherapy studies7,8. Immune-modulatory vaccines targeting tumoral immune escape mechanisms offer a new, generalizable strategy compared to patient-specific neoantigen cancer vaccines7,9. The immune-modulatory vaccine strategy in this clinical trial is based on the finding of circulating cytotoxic T cells specific to IDO and PD-L1 in the blood of patients with cancer and, to a lesser extent, in healthy donors. IDO- and PD-L1-specific CD8+ T cells can directly recognize and kill IDO+ and/or PD-L1+ tumor cells and likewise recognize and kill non-malignant cells that express their cognate targets. Furthermore, IDO- and PD-L1-specific CD4+ T cells release pro-inflammatory cytokines in response to IDO- or PD-L1-expressing target cells. IDO and PD-L1 are expressed not only by melanoma cells but also by many other cell types in the tumor microenvironment (TME), which differentiates these antigens from traditional tumor antigens used in other studies10–15. Activation of IDO/PD-L1-specific T cells by vaccination can therefore restrict the range of immunosuppressive signals mediated by immunosuppressive cells and thereby revert the TME from an immune hostile to an immune friendly environment. In animal models of cancer, vaccinations with IDO epitopes resulted in anti-tumor therapeutic effects that were correlated with reductions in IDO expression in myeloid cell populations within the TME16. The IDO/PD-L1 immune-modulating vaccine may lead to a translatable strategy for improving the efficacy of aPD1 therapy through activation of specific T cells. We hypothesize that the IDO/PD-L1 vaccine attracts T cells into the tumor, which induces type 1 helper T (TH1) cell inflammation and reverts the TME into an immune-permissive site, thereby turning the tumor ‘hot’. This would also upregulate PD-L1 expression in cancer and immune cells, generating more susceptible targets to aPD1 therapy (Extended Data Fig. 1a). This theory was confirmed in a mouse model in which aPD1 therapy and IDO vaccination show synergistic effects16. In this phase 1/2 clinical trial, MM1636, patients with MM received a combination of the IDO/PD-L1 (IO102/IO103) peptide vaccine with the adjuvant Montanide and the aPD1 antibody nivolumab. Patients were included in three cohorts: 30 aPD1 therapy-naive patients (cohort A), ten aPD1 therapy-refractory patients (cohort B, de novo resistance) and ten patients who progressed after aPD1 therapy (cohort C, acquired resistance). Here, we report results from cohort A. The vaccine was given biweekly for the first six administrations and thereafter every 4th week. A maximum of 15 vaccines were administered. Nivolumab was given in parallel, biweekly (3 mg per kg) or every 4th week (6 mg per kg) for up to 2 years (Extended Data Fig. 1b). The primary objective was safety and feasibility. Secondary objectives were immunogenicity and clinical efficacy. Results Patients and treatment Thirty patients were enrolled from December 2017 to June 2020. None of the 30 patients dropped out of the study; all received at least three cycles of therapy (Supplementary Fig. 1). At the current database lock (5 October 2020), six patients were still on treatment in the trial. Of the 24 patients who were not on trial treatment at data cutoff, two are still receiving nivolumab monotherapy (6 mg per kg every 4 weeks). Reasons for stopping treatment for the remaining 22 patients included disease progression (37%), toxicity (20%), maximum benefit or complete response (CR) confirmed on two consecutive scans (17%) or completing 2 years of treatment (7%). For the 24 patients who were not on trial treatment at data cutoff, the mean number of vaccinations was 10.5 (range, 3–15 vaccinations). Thirteen of these 24 patients continued nivolumab (6 mg per kg, every 4 weeks) as a standard of care. Nine patients received subsequent therapy after progression (Supplementary Table 1). Baseline characteristics are shown in Table 1. The mean age was 70 years; 37% of patients had elevated lactate dehydrogenase (LDH) levels, 60% were stage M1c, 37% had BRAF mutations, and 43% were negative for PD-L1 (<1%). A total of three patients (10%) had received prior ipilimumab therapy. No patients had brain metastasis (Supplementary Table 2).Table 1 Baseline patient characteristics (n = 30) Characteristic Number (%) Mean age, years (range) 70 (46–85) Sex, male 16 (55%) ECOG PS 0 26 (87%) LDH levels  ≤ULN 19 (63%)  >ULN 11 (37%) M stage (AJCC-8)  M1a 6 (20%)  M1b 6 (20%)  M1c 18 (60%) Number of lesion sites  1 6 (20%)  2–3 17 (57%)  >3 7 (23%) Liver metastases present 10 (33%) BRAF status  Mutant 11 (37%)  Wild type 19 (63%) PD-L1  <1% 13 (43%)  >1% 17 (57%) Previous systemic therapy  Ipilimumab 3 (10%)  No 27 (90%) AJCC-8, eighth edition of the American Joint Committee on Cancer; ECOG PS, Eastern Cooperative Oncology Group performance status; ULN, upper limit of normal. Notable clinical responses to the combination therapy Thirty patients with MM were treated with the IDO/PD-L1 vaccine and nivolumab according to the trial protocol. By investigator review, the ORR reached 80% (CI, 62.7–90.5%), with 43% of patients (CI, 27.4–60.8%) achieving a CR and 37% (CI, 20.9–54.5%) reaching a partial response (PR) as the best overall response, while 20% experienced progressive disease (PD) according to Response Evaluation Criteria in Solid Tumors (RECIST) 1.1 (Fig. 1a). Two of the patients with a PR did not have a confirmatory scan stating PR on two consecutive scans. Early onset of response was frequent, with 22 of 30 patients having an objective response at the first evaluation (after 12 weeks on treatment). Median times to PR and CR were 75 d (range, 54–256 d) and 327 d (range, 73–490 d), respectively (Fig. 2a–c).Fig. 1 Clinical response. a, Pie charts with percent ORR, CR, PR and PD according to RECIST 1.1 by investigator review of all patients (n = 30), PD-L1+ patients (>1%, (n = 17)) and PD-L1− patients (<1%, n = 13)), respectively. Two-sided CIs (95%) were constructed using the Clopper–Pearson method. b, Treatment effect in MM1636 compared with a matched historical control group from the DAMMED database (n = 74). Patients in MM1636 (n = 29) were matched with the exact same combination variable according to age (≤70 years, >70 years), sex, LDH levels (normal, elevated), M stage (M1a, M1b, M1c), BRAF status (wild type, mutated) and PD-L1 status (<1%, ≥1%). Estimates for treatment effects were calculated by weighted logistic regression analyses and weighted Cox proportional hazard model. Bar height indicates the estimated response rate; tops of bars are centers for error bars. Odds ratios (OR), response rates and their corresponding 95% CIs were extracted from the regression model. All P values were two sided, and P values below 0.05 were considered statistically significant. c, Best change in the sum of target lesion size compared with that at baseline (n = 30). The horizontal line at −30 shows the threshold for defining an objective response in the absence of non-target disease progression or new lesions according to RECIST 1.1. Two patients with 100% reduction in target lesion size had non-target lesions present. White stars, six patients had normalization (<10 mm) of fluor-18-deoxyglucose (FDG)-negative lymph nodes (at baseline, lymph nodes were >1.5 cm and FDG+) and 100% reduction of non-lymph node lesions and are considered to have had a CR (green bar). Black star, one patient (MM29) was considered to have had a PR (blue bar), although he had not reached a −30% change in target lesion size. The patient had a single measurable 13-mm lung metastasis at baseline and multiple biopsy-verified cutaneous metastases on the left crus (not detectable by positron-emission tomography–computed tomography (PET–CT) at baseline). The best change in target lesion size was 10 mm, and a post-treatment biopsy from the cutaneous metastases showed no sign of malignancy. Thus, overall, the patient was classified as having a PR. d, Kaplan–Meier curve of the response duration in the 24 patients with an objective response. e, Kaplan–Meier curve of PFS in all 30 treated patients. f, Kaplan–Meier curve of OS in all 30 treated patients. Fig. 2 Clinical response. a, Swimmer plot showing response duration and time to response according to RECIST 1.1 for all treated patients (n = 30). Triangles indicate first evidence of PR, while squares indicate first evidence of CR. Closed circles indicate time of progression. Arrows indicate ongoing responses. Patient MM18 died due to nivolumab-induced side effects. b, Spider plot showing response kinetics in all treated patients (n = 30). Red squares indicate time of progression. c, PET–CT images of patient MM42 before and after treatment (after 12 series of treatment) showing FDG metabolism in target lesions. SC, subcutaneous. The ORR among PD-L1+ (>1% (clone 28.8)) patients (n = 17) was 94.1% (CI, 73–99.7%) and 61.5% (CI, 35.5–82.3%) in PD-L1− patients (n = 13) (Fig. 1a). Objective responses were observed in patients irrespective of human leukocyte antigen (HLA) genotype (Supplementary Fig. 2). Clinical response data were validated by blinded independent external review, in which an ORR of 76.6% (CI, 57.7–90.1%) was reported, with 53.3% of patients achieving a CR, 23.3% achieving a PR and 3.3% experiencing stable disease. Comparisons between investigator review and external review are outlined in Supplementary Table 3. To examine whether the very high response rate should be attributed to nivolumab or to the vaccine, we retrieved clinical response information from a matched historical control group using the Danish Metastatic Melanoma Database (DAMMED) from contemporaneously treated patients with stage 3–4 melanoma who received aPD1 monotherapy17. Patients were matched with the exact same combination variable according to age, sex, PD-L1 status, BRAF status, LDH level and M stage (those at stage M1d were excluded from the control group (no patients with brain metastasis)). Matched controls were identified for 29 patients, and the ORR of 79.3% (CI, 61.0–90.4%) observed in MM1636 was found to be significantly higher (P < 0.0012) than that in the matched control group in which an ORR of 41.7% (CI, 31.0–53.3%) was reached. Furthermore, of the 29 patients in MM1636, a significantly (P < 0.0017) higher percentage (41.4% (CI, 25.2–59.6%)) of patients achieved CR in MM1636 than that (12% (CI, 6.3–21.6%)) in the matched historical control group. ORR and complete response rate (CRR) in the matched historical control group were comparable to those of patients treated in randomized phase 3 pivotal trials with aPD1 monotherapy18 (Fig. 1b). The treatment leads to prolonged PFS At data cutoff, the median duration of response had not been reached, with 87% of all responding patients being free from progression at 12 months (Fig. 1d). Patients were followed for up to 35 months with a median follow-up time of 22.9 months (CI, 14.9–26.2 months). OS and PFS were calculated from the first day of treatment to death or progression or to the date of the last follow-up (5 October 2020). The median PFS (mPFS) for all treated patients was 26 months (CI, 15.4–69 months) and was not reached for responding patients (Fig. 1e). The median OS was not reached at the data cutoff. OS at 12 months was 81.6% (CI, 61.6–92%) (Fig. 1f). One patient (MM18) with a CR died from nivolumab-related severe adverse events; the remaining patients died because of metastatic melanoma. For comparison, the mPFS was 8.3 months (CI, 5.5 months–NR (not reached)) in the matched historical control group (n = 74), while the median OS was 23.2 months (CI, 23.2 months–NR). (Extended Data Fig. 2a,b). The combination of the IDO/PD-L1 vaccine and nivolumab was safe Treatment-related adverse events are listed for all 30 patients in Supplementary Table 4. Common treatment-related grade 1–2 toxicities were fatigue (47%), rash (47%), arthralgia (30%), diarrhea (23%), nausea (23%), dry skin (20%), pruritus (20%), infusion reaction (17%), xerostomia (17%) and myalgia (17%). Four patients (13%) experienced grade 3–4 adverse events: one patient with a grade 3 maculopapular rash (MM01), one patient with grade 3 adrenal insufficiency (MM06) and one patient with grade 3 arthralgia (MM22). Patient MM18 died from urosepsis with multi-organ failure and severe hyponatremia. This patient had experienced multiple immune-related adverse events with grade 3 colitis, grade 2 pneumonitis, grade 3 arthralgia, grade 2 vasculitis and grade 2 nivolumab infusion-related allergic reaction. Additionally, patient MM18 had symptoms of myocarditis at the time of death with highly elevated cardiac troponin I levels. Bedside echocardiography showed an ejection fraction of 15%, which at baseline was 60%, but an autopsy was not conducted, and myocarditis was never confirmed pathologically. Patient MM06 had received first-line treatment with ipilimumab before entering the trial and was on substitution corticosteroids at the time of inclusion. Adrenal insufficiency was aggravated by an erysipelas infection with high fever, reaching grade 3 in Common Terminology Criteria for Adverse Events and resolving quickly after appropriate antibiotic therapy was initiated. As expected, local side effects were common with 77% of the patients who developed injection site reactions. These reactions were classified as granulomas (63%), redness (20%), pain (13%) and pruritus (13%) at the injection site. All local reactions were grade 1–2, most likely related to the Montanide adjuvant and typically transient. However, two patients (MM07 and MM20) decided to discontinue vaccination after eight and 11 injections, respectively, due to granulomas, tenderness and pain that limited instrumental activities of daily living but continued to receive nivolumab (Extended Data Fig. 2c). Vaccine-specific responses in blood were frequently detected First, all 30 patients were assessed for the presence of vaccine-specific responses in peripheral blood mononuclear cells (PBMCs) before, on and after vaccination using a modified interferon (IFN)-γ enzyme-linked immune absorbent spot (ELISPOT) assay. This assay is known to expand antigen-specific memory cells and to improve the detection and the correlative power of ELISPOT during treatment19–21. Prevaccine IDO-specific responses were detectable in ten (33%) patients, while prevaccine PD-L1-specific responses were detectable in eight patients (27%); overlapping (specific to both IDO and PD-L1) prevaccine responses were present in four (13.3%) patients. During vaccination, an increase of IDO-specific T cells or PD-L1-specific T cells in the blood was observed in 28 (93%) and 26 (86%) patients, respectively. In total, 93% of patients had an increase in either PD-L1- or IDO-specific responses on vaccination (Fig. 3a), with a significant (P < 0.0001) median increase from baseline to the post-vaccine response demonstrated for both IDO and PD-L1 (at different time points on treatment), confirming that vaccine-specific immune responses were induced in patients regardless of clinical response (Fig. 3b,c). Immune responses fluctuated in the blood over time (Extended Data Fig. 3a). An increase in IDO- and PD-L1-specific responses in the peripheral blood was also detectable directly ex vivo across the different clinical response groups (Extended Data Fig. 4).Fig. 3 Vaccine-specific responses in blood. a, IDO- and PD-L1-specific T cell responses in PBMCs at baseline and on vaccination as measured by the IFN-γ ELISPOT assay (n = 30). *Responses were calculated as the difference between the average numbers of spots in wells stimulated with IDO or PD-L1 peptide (triplicates) and those from the corresponding control (DMSO), and statistical analyses of ELISPOT responses were performed using a distribution-free resampling method (Moodie et al.50). DR (double response), response was not statistically confirmed due to replicate number, but the number of spots in peptide wells was two times higher than that in control wells (DMSO). NS, no significant response and no DR. For a detailed overview of responses at serial time points on vaccination, see Extended Data Fig. 4a. b, IDO- and PD-L1-specific T cell response in PBMCs in all treated patients measured by the IFN-γ ELISPOT assay at baseline and on vaccination. On-vaccination responses were selected from the ‘best’ ELISPOT response at different time points for each patient (series 3, 6, 12, 18 or 24) during vaccination (n = 30). Wilcoxon matched-pairs signed-rank test was used to compare responses to IDO or PD-L1 peptides in the vaccine between baseline and later time points. c, Representative example of ELISPOT wells with response in patient MM23 in serial PBMCs before and on treatment. d, IDO-specific CD4+ and CD8+ T cells were isolated and expanded from PBMCs stimulated in vitro with the IDO peptide and a low dose of IL-2 for 14–15 d before sorting using the Miltenyi Cytokine Secretion Assay—Cell Enrichment and Detection kit. To assess their cytolytic potential, IDO-specific T cells were stimulated with IDO peptide, and expression of CD107a, IFN-γ and TNF-α was assessed by flow cytometry (the example is from patient MM14). Sustained vaccine-specific responses were observed 3 and 6 months after the last vaccine, indicating induction of memory responses in nine patients with response to clinical treatment who surpassed follow-up at data lock (Extended Data Fig. 3b). Importantly, PD-L1- and IDO-specific responses were observed irrespective of HLA genotype (Supplementary Table 5). To verify the functionality of vaccine-induced T cells, IDO- or PD-L1-specific T cells were isolated and in vitro expanded from PBMCs of five patients. Phenotypic characterization by flow cytometry revealed that the isolated vaccine-specific T cells consisted of both CD4+ and CD8+ T cells. Also, both IDO- and PD-L1-specific CD4+ and CD8+ T cells showed pro-inflammatory properties, as they expressed the cytolytic marker CD107a and secreted the cytokines IFN-γ and tumor necrosis factor (TNF)-α (Fig. 3d and Extended Data Fig. 6a–d). Interestingly, we were also able to detect vaccine-specific CD4+ and CD8+ T cell responses in peripheral blood ex vivo (Extended Data Figs. 4 and 5). We observed a significant increase in the overall percentage of CD107a, CD137 and TNF-α expression in response to peptide stimulation in on- or post-treatment PBMC samples compared to that at baseline, further confirming the expansion and the diverse signature of vaccine-specific T cells (Extended Data Fig. 5). Vaccine-specific responses detected in the skin at the vaccination site To investigate whether vaccine-specific T cells have the potential to migrate to peripheral tissue, delayed-type hypersensitivity (DTH) tests were performed after six cycles of treatment on 15 patients to assess the presence of vaccine-reactive T cells in the skin. Supplementary Table 6 display an overview of skin-infiltrating lymphocyte (SKIL) cultures. We detected IDO-specific T cells in the skin of six of ten patients and PD-L1-specific T cells in nine of 11 patients (Extended Data Fig. 7a). Intracellular cytokine staining was performed on SKILs from five patients after stimulation with either PD-L1 or IDO peptide. Here, we detected mainly CD4+ peptide-reactive T cells that secreted TNF-α and upregulated CD107a, and a minor fraction also secreted IFN-γ. In one patient, we detected CD8+ PD-L1-reactive T cells (Extended Data Fig. 7b–d). Vaccine-induced T cells specifically recognize target cells To confirm the functionality of vaccine-expanded T cells, vaccine-specific T cell clones (clonal purity was confirmed by T cell receptor (TCR) sequencing) were isolated and expanded from patient PBMCs (Extended Data Fig. 8e). We showed that PD-L1-specific T cells were able to recognize PD-L1+ autologous tumor cells in a PD-L1 expression-dependent manner if the cancer cells also expressed HLA-II (Fig. 4a,b). Similarly, an HLA-DR-restricted IDO-specific CD4+ T cell clone was able to recognize an HLA-DR-matched IDO-expressing model cell line, MonoMac1, in an IDO-expression-dependent manner (Fig. 4e–g). As previously described, IDO- and PD-L1-specific T cells’ mode of action is not limited to targeting only cancer cells. We were able to show that vaccine-specific T cell clones also reacted against PD-L1- and IDO-expressing autologous immune cells (Fig. 4c,h). To provide myeloid cells with a tumor-associated phenotype we treated isolated CD14+ myeloid cells with tumor conditioned medium (TCM) derived from an established autologous tumor cell line. We observed that such TCM-treated CD14+ cells had increased expression of PD-L1 and IDO and were effectively recognized by autologous PD-L1- and IDO-specific CD4+ T cell clones (Fig. 4c,d,h,i).Fig. 4 PD-L1- and IDO-specific T cells from vaccinated patients react against PD-L1- and IDO-expressing target cells. a, Left, PD-L1-specific T cell culture (MM1636.05) reactivity against PD-L1 peptide or autologous tumor cells in the IFN-γ ELISPOT assay. Tumor cells were either not treated or pretreated with 200 U ml−1 IFN-γ for 48 h before the assay. Effector:target (E:T) ratio of 10:1 was used. Right, PD-L1 and HLA-II surface expression on melanoma cells with (green) or without (yellow) pretreatment with IFN-γ compared to an isotype control (gray) as assessed by flow cytometry. b, Left, PD-L1-specific T cell (MM1636.05) reactivity in the IFN-γ ELISPOT assay against autologous tumor cells pretreated with IFN-γ (500 U ml−1) and transfected with mock or PD-L1 small interfering (si)RNA 24 h after transfection. E:T ratio, 10:1. Right, PD-L1 surface expression on melanoma tumor cells (MM1636.05) assessed by flow cytometry 24 h after transfection with mock (blue) or PD-L1 (red) siRNA compared to the isotype control (gray). c, Reactivity of the CD4+ PD-L1-specific T cell clone (MM1636.14) against PD-L1 peptide or autologous CD14+ cells; E:T ratio, 10:1. CD14+ cells were isolated using magnetic bead sorting and used as targets in an ELISPOT assay directly or after pretreatment for 2 d with TCM derived from the autologous tumor cell line. d, Quantitative PCR with reverse transcription (RT–qPCR) analysis of PD-L1 (CD274) expression in sorted CD14+ cells before and after treatment with autologous TCM for 48 h. e, Reactivity of the IDO-specific CD4+ T cell clone (MM1636.23) against IDO peptide combined with HLA-DR (L243)-, HLA-DQ (SPV-L3)- or HLA-DP (B7/21)-blocking antibodies (aHLA-DR, aHLA-DQ and aHLA-DP) in an intracellular staining assay (ICS) for IFN-γ and TNF-α production. T cells were incubated with individual blocking antibodies (2 μg ml−1) for 30 min before adding IDO peptide. f, Reactivity of the IDO-specific CD4+ T cell clone (MM1636.23) against the HLA-DR-matched IDO-expressing cell line MonoMac1 transfected with mock or IDO siRNA in an ICS assay, E.T ratio, 4:1. siRNA transfection was performed 48 h before the experiment. g, RT–qPCR analysis of IDO1 expression in MonoMac1 cells 48 h after siRNA transfection. h, Reactivity of the CD4+ IDO-specific T cell clone (MM1636.14) against IDO peptide or autologous CD14+ cells; E:T ratio, 20:1. CD14+ cells were isolated using magnetic bead sorting and used as targets in an ELISPOT assay directly or after pretreatment with TCM derived from the autologous tumor cell line. i, RT–qPCR analysis of IDO1 expression in sorted CD14+ cells before and after treatment with autologous TCM for 48 h. Bars in RT–qPCR data (d,g,i) represent the mean of three (d,i) or six (g) technical replicates ±s.d.; P values were determined by two-tailed parametric t-tests. ELISPOT counts (a,b,c,h) represent the mean value of three technical replicates ±s.e.m.; response P values were determined using the distribution-free resampling (DFR) method. TNTC, too numerous to count. T cell clones in blood and tumors To track the role of treatment-induced T cell responses, TCR sequencing of the complementarity-determining region 3 (CDR3) was performed on five patients in peripheral blood (baseline and cycles 3, 6 and 12) and paired biopsies. These five patients (MM01, MM02, MM08, MM09 and MM13) were selected due to the availability of material and to investigate a balanced patient group with both responders and non-responders. Due to a limited number of available paired biopsies (either as a consequence of patient refusal or unexpectedly rapid and substantial clinical responses), statistical analysis could not be applied. Details on clinical response are shown in Fig. 2a. Additionally, PBMCs (on treatment) or SKILs were stimulated with the IDO/PD-L1 peptides, and then cytokine-producing T cells were sorted to track vaccine-induced T cells both in the periphery and at the tumor site. To identify enriched IDO/PD-L1-specific T cell clones, TCR rearrangements in sorted IDO/PD-L1-specific clones and TCR rearrangements in sorted IDO/PD-L1-specific T cell samples were compared to sequences from baseline PBMC samples for each patient. Clonal expansion of vaccine-specific TCR rearrangements from samples on vaccination were then tracked using a differential abundance framework. Cumulative IDO/PD-L1-specific T cell frequencies were tracked in post-treatment samples. We found no relation between clinical response and the enrichment of vaccine-specific clones, but an increase in IDO/PD-L1-specific T cell clones was observed at different time points in the periphery in all five patients (Extended Data Fig. 3c). We next investigated overall changes in the T cell repertoire in the blood. A modest increase in the peripheral T cell fraction was observed in the three responding patients at cycle 3, while the two non-responding patients had a clear decrease in T cell fraction (Extended Data Fig. 8a). We thereafter investigated TCR clonality and TCR repertoire richness, exploring the proportion of abundant clones and the number of unique rearrangements, respectively. A decreasing peripheral Simpson clonality and increasing TCR repertoire richness was observed in responding patients at cycle 3, which might indicate tumor trafficking upon treatment. The opposite pattern was observed in non-responding patients (Extended Data Fig. 8b,c). Peripherally expanded clones were associated with tumors and persisted until cycle 12 (latest time point analyzed). The largest peripheral expansion was observed at cycle 3, with the most significant increase observed in patient MM01 (CR). Responding patients had a larger fraction of peripherally expanded clones that were also found in tumors compared to that of non-responders. By tracking peripherally expanded clones detected at the tumor site, we observed that patient MM01 had a substantial increase after treatment, indicating tumor trafficking of peripheral expanded clones (Extended Data Fig. 8d). Influx at the tumor site of vaccine-enriched T cell clones Given the observation of increased T cell fraction and enrichment of IDO- and PD-L1-specific clones in the blood after treatment, we investigated whether the same trend was observed at the tumor site. Both TCR sequencing and immunohistochemistry (IHC) of paired biopsies from the five patients described above showed an increase in the T cell fraction with an influx of CD3+ and CD8+ T cells after treatment in the three responding patients (Fig. 5a–c). IHC was not possible for patient MM09 due to tissue loss.Fig. 5 Changes in the TME after treatment. Number of CD3+ and CD8+ T cells, TCR fraction, TCR clonality, TCR repertoire, biopsy expanded TCR clones and enrichment of IDO- and PD-L1-specific T cells at the tumor site. a, Number of CD3+ and CD8+ T cells per mm2 (sum in the validated area) at the tumor site detected by IHC of paired biopsies from four patients. b, Example of IHC of CD3+ and CD8+ T cells at the tumor site before and after treatment (cells per mm2) in one patient (MM01). c, T cell fraction at the tumor site at baseline and cycle 6 by TCR sequencing. The T cell fraction was calculated by taking the total number of T cell templates and dividing by the total number of nucleated cells. d, Tracking of vaccine-associated clones at baseline and cycle 6 in tumor biopsies. Cumulative frequencies of IDO and PD-L1 vaccine-specific TCR rearrangements are represented. e,f, TCR clonality and TCR repertoire richness in five patients at the tumor site at baseline and cycle 6. Simpson clonality measures how evenly TCR sequences are distributed among a set of T cells, where 0 indicates an even distribution of frequencies and 1 indicates an asymmetric distribution in which a few clones dominate. TCR repertoire richness reports the mean number of unique rearrangements. g, Bar chart representing baseline expanded biopsy clones from five patients (colored bars) and the detection of biopsy expanded clones also found in the blood at baseline and series 3, 6 and 12 (white and gray bars) by TCR sequencing. We thereafter investigated whether some of the IDO/PD-L1 vaccine-linked T cells were present at the tumor site. Vaccine-associated clones were tracked as the combined frequency of IDO- and PD-L1-specific T cell rearrangements. In biopsies, the frequencies at cycle 6 were compared to those at baseline and showed an increase in vaccine-specific T cells in four of five patients, irrespective of clinical response (Fig. 5d). TCR sequencing of PD-L1-specific SKILs (a more specific culture than IDO/PD-L1-specific isolations derived from PBMCs on vaccination) and paired biopsies showed that two of the top five PD-L1-specific SKIL clones were present at the tumor site both before and after treatment (Extended Data Fig. 7e,f). With a focus on the more abundant T cell clones, we investigated overall TCR clonality at the tumor site before and after treatment. In addition, we explored the number of unique TCR rearrangements, dissecting the lower-frequency clones. Patient MM01 had a significant increase in TCR clonality and a decrease in repertoire richness at the tumor site after therapy, indicating a focused tumor repertoire response of selected clones. All three responding patients had a decrease in TCR repertoire richness, which again might indicate a focused tumor response (Fig. 5e,f). Deeper analyses showed that the T cell clones that expanded at the tumor site after therapy were also present in the blood at baseline and increased significantly after treatment in four of five patients. The highest proportion was detected early at cycle 3. These data again support trafficking of peripherally expanded clones to the tumor site and could indicate that the T cell response to treatment is derived from pre-existing peripherally tumor-associated T cells (Fig. 5g). Signs of treatment-induced inflammation in the TME To dissect changes in the TME induced by T cell influx upon treatment in responding patients, RNA gene expression analyses using the nCounter PanCancer Immune Profiling Panel from NanoString were performed on paired biopsies from two responding patients (MM01 and MM13). Expression of genes related to adaptive immunity such as T cell activation, effector functions (genes encoding IFN-γ, TNF-α, IL-15 and IL-18) and cytotoxicity was increased in post-treatment biopsies (Extended Data Fig. 9a,b). Also, expression of genes related to checkpoint inhibitors such as those encoding T cell immunoglobulin and mucin-domain containing-3 (TIM-3), IDO, PD-L1, PD-L2, PD-1 and CTLA-4 increased after treatment, indicating activation of immune cells in the TME (Extended Data Fig. 9c). Additionally, IHC of paired biopsies from four patients (MM01, MM02, MM05 and MM13) showed an upregulation of PD-L1, IDO, MHC-I and MHC-II on tumor cells, indicating a treatment-induced pro-inflammatory response in the three responding patients, except for a decrease in MHC-II expression in patient MM13. By contrast, the non-responding patient MM02 had a reduction in T cell numbers present in the tumor after treatment and no expression of PD-L1, IDO or MHC-II and, interestingly, total loss of MHC-I, demonstrating tumor immune escape (Extended Data Fig. 10a). CD8+ T cells and their distance (µm) to PD-L1-expressing cells in baseline biopsies from five patients was investigated by IHC. Except for patient MM13 (PR), distance and clinical responses were associated: the two responders had reduced distance (<20 µm) between cells expressing these markers compared to non-responding patients (>80 µm). This observation indicates that responding patients not only have a higher intratumoral infiltration of CD8+ T cells, but that these cells can surround and attack PD-L1-expressing immune cells and tumor cells (Extended Data Fig. 10b). Discussion In this clinical trial, MM1636, 30 patients with metastatic melanoma were treated with a first-in-class immunomodulatory IDO/PD-L1-targeting peptide vaccine combined with nivolumab. The treatment led to an unprecedented high ORR of 80%, with 43% of patients reaching a CR, and a striking mPFS of 26 months (95% CI, 15.4–69 months) was reached. The vaccine represents a new treatment strategy to activate specific T cells that target cells contributing to immune suppression (including tumor cells), positively modulating the TME by inducing local inflammation. Indeed, we show that vaccine-specific T cells isolated and expanded from vaccinated patients recognize not only tumor cells in a target- and HLA-restricted manner but also myeloid cells polarized toward a tumor-associated phenotype. Hence, myeloid cells become targets for vaccine-activated T cells when they have a tumor-associated phenotype in the TME. These phenomena may further induce checkpoint molecules and rewire the TME toward an increasingly aPD1-permissive state. A drawback of our study is the single-center nonrandomized setup. Comparison between trials or between patients in trials and real-world patients is problematic due to multiple factors, such as period of conduction and other therapies available at the different periods of time. Nevertheless, the rate of investigator-assessed ORR in the phase 3 trial CheckMate 067 was 43.7% in the nivolumab monotherapy group and 57% in the nivolumab and ipilimumab group. CRs occurred in 8.9% and 11.5% of patients, respectively18. The mPFS of 26 months (95% CI, 15.4–69 months) in this trial is more than twice as long as that for patients treated with nivolumab and ipilimumab in CheckMate 067, for which an mPFS of 11.5 months (95% CI, 8.7–19.3 months) was reached. Patient baseline characteristics were in general comparable to those of patients with MM who have been treated in CheckMate 067, although patients in MM1636 were older (mean age, 70 years) and a larger fraction were positive for PD-L1 (57%)3,18,22. Among patients with PD-L1-negative tumors in MM1636, an ORR of 61.5% was still reached, which would be expected to be around 33% for first-line nivolumab monotherapy23. Some studies suggest that older patients might have a tendency to respond better to aPD1 therapy; however, this is still debated24–27. To address potential trial bias and the nonrandomized setup, patients in MM1636 were matched for age, performance status, sex, M stage, LDH level, PD-L1 status and BRAF status with a historical control group from the DAMMED, who were treated contemporarily (2015–2019) with aPD1 monotherapy as standard of care17. We found a significantly higher ORR and CRR in MM1636 compared to those of matched patients, who had an ORR of 43% and a CRR of 13%, comparable to patients treated in CheckMate 067. Restrictions of the synthetic control group are of course that it is partially historic and patient selection outside matching criteria cannot be ruled out28. Numerous contemporary clinical trials are exploring the combination of aPD1 therapy with other immunomodulating agents for advanced melanoma. Talimogene laherparepvec, an oncolytic virus, is approved by the Food and Drug Administration and the European Medicines Agency to treat advanced melanoma. A small phase 1b trial with 21 patients (MASTERKEY-265) combined talimogene laherparepvec and pembrolizumab to treat patients with advanced unresectable melanoma and reached an ORR of 62% and a CR of 33%29,30. Seventy-one percent of the patients in this trial had an M stage below M1c; this number was 40% in our trial. Furthermore, mainly patients with an M stage below M1c responded to treatment, which was not the case in MM1636. Results from a large randomized phase 3 trial are awaited (KEYNOTE-034). Epacadostat, an IDO inhibitor, was tested in combination with pembrolizumab in a nonrandomized phase 2 trial in 40 aPD1 treatment-naive patients with MM with promising results, reaching an ORR of 62%. Unfortunately, the phase 3 trial showed no indication that epacadostat provided improvement in PFS and OS31. Limitations of the phase 3 trial were the sparse information on pharmacodynamics as well as biomarker evaluation to improve the design. The IDO/PD-L1 vaccine is different from epacadostat, as it is not an IDO inhibitor but targets IDO- and PD-L1-expressing cells. Similar vaccines administered as monotherapy induced objective responses in lung cancer and basal cell carcinoma, while epacadostat as monotherapy in 52 patients resulted in no responses32,33. Sahin et al. recently published encouraging data from a first-in-human trial, in which a vaccine containing liposomal RNA targeting four unmutated tumor-associated antigens (NY-ESO-1, MAGE-A3, tyrosinase and TPTE) was administered alone or in combination with nivolumab in patients with advanced melanoma. Responses were observed in both the monotherapy group as well as the combination group in checkpoint inhibitor-experienced patients, suggesting the efficacy of non-mutant shared tumor antigen vaccines34. The overall safety and tolerability findings are comparable to those of aPD1 monotherapy. Injection site reactions were exclusive to the vaccine. However, these side effects were transient and mild in most patients and most likely due to the adjuvant Montanide. IDO- and PD-L1-specific CD8+ and CD4+ T cells exist among peripheral blood lymphocytes in healthy donors35–38 and expand in response to pro-inflammatory stimuli35,39. Furthermore, both IDO and PD-L1 are induced in cells as a counter response to the inflammatory response. This provides a mechanism that ensures immune homeostasis, which keeps IDO/PD-L1-specific T cells in check; therefore, the risk of triggering autoimmune-related adverse events by vaccination appears to be minimal. This was confirmed in the first clinical trials of IDO and PD-L1 vaccination (NCT01543464 (refs. 33,40) and NCT03042793 (ref. 41)). The spontaneous (baseline) immune response to the vaccines observed in the current study is in agreement with our previous observation in various patients with cancer35–38. Numerous vaccine-induced changes in the blood and at the tumor site were observed. Peripheral IDO- and/or PD-L1-specific T cells were detected in vitro in a modified ELISPOT in over 93% of patients on vaccination, unrelated to patient HLA type. Immune responses were persistent in patients who surpassed follow-up at data cutoff and were still detectable up to 6 months after the last vaccine, suggesting induction of memory T cells. We did not observe a correlation between vaccine-induced responses in blood and clinical responses. However, the detection of a highly significant increase in vaccine-specific T cell numbers after vaccination in almost all patients together with the very low number of patients with PD makes it difficult to expect such a correlation, especially because other aspects should be taken into account (for example, the loss of class I expression on tumors cells in patient MM02 with PD after vaccination). Frequencies of peripheral T cells induced against PD-L1 were overall higher than those against IDO. Importantly, however, we also observed that it was not the same patients who responded strongly to both IDO and PD-L1, nor did patients react with a similar response pattern to the two antigens; that is, different time points of response were observed. This suggests that each component of the vaccine plays different roles in the ongoing immune response in patients. The goal of IDO/PD-L1 vaccination was to modulate the TME to increase responsiveness to aPD1 therapy, as we have observed in animal models, in which immune conversion is demonstrated in the TME with an increased influx of T cells. Targeting both IDO and PD-L1 together enables synergy, as the TME is known to use different immune escape mechanisms and IDO and PD-L1 are often overexpressed by different cellular compartments. Ex vivo ELISPOT and flow cytometry assays confirmed the induction of immune response toward both epitopes and the higher immunogenicity of the PD-L1 epitope, although because of lower sensitivity failed to detect some of the responses. TCR sequencing of five patients confirmed enrichment of IDO/PD-L1-specific T cell clones in the blood at different time points after treatment. Furthermore, an increase in enriched IDO/PD-L1-specific clones was observed in four of five patients at the tumor site after treatment, irrespective of clinical response. Phenotypic characterization showed that vaccine-specific T cells, which were expanded in vitro with interleukin (IL)-2 from the blood of vaccinated individuals, were both CD4+ T cells and CD8+ T cells. This was confirmed by ex vivo phenotype description of vaccine-activated T cells. Vaccine-specific T cells expressed CD107a and CD137 and produced IFN-γ and TNF-α upon stimulation with the cognate target, indicating their cytolytic capacity. Overall, the data from immune monitoring supports the importance of both antigens in the generation of the frequent clinical responses in the study. Despite a limited number of paired biopsies (due to either patient refusal or the fact that a large fraction of responding patients had no assessable tumors after six cycles), we observed trends indicating treatment-induced general T cell influx in responding patients. It was shown that proliferation of CD8+ T cells in the tumor after aPD1 treatment is associated with radiographic reduction in tumor size42. Additionally, we showed that a large proportion of expanded peripheral TCR clones were associated with tumors and the most considerable amount of clonal expansion was observed early, at cycle 3. For the patient with a CR included in the TCR sequencing analyses (MM01), the number of peripheral expanded clones present at the tumor site increased after treatment compared to that at baseline, indicating tumor trafficking of peripheral expanded clones. Gene expression analyses (two paired biopsies) and IHC (five paired biopsies) further demonstrated that the combination treatment induced a pro-inflammatory TME in responding patients with signs of T cell activation and cytotoxicity and increased cytokine activity. This may lead to further upregulation of IDO, PD-L1, MHC-I and MHC-II on tumor cells, leading to more treatment targets. It was shown that, following vaccination with a cancer vaccine, PD-L1 expression is increased on tumor cells due to recruitment of tumor-specific T cells and upregulation of adaptive immune resistance pathways in the TME43. Treatment with nivolumab monotherapy enhances PD-L1 expression, and it is therefore problematic to discriminate the effect of the vaccine as compared to that of nivolumab44. IDO- and PD-L1-specific pro-inflammatory effector T cells were hypothesized to counteract the functions of IDO- and PD-L1-expressing immune-suppressive cells as a means to keep the immune balance between immune activation and inhibition45. However, because the expression of these molecules is induced as a counter-regulatory response to inflammatory mediators such as IFNs, they can also be expressed by, for example, activated myeloid cells or T cells. Thus, activation of PD-L1-specific T cells may result in depletion of activated T cells in the tumor and other sites. Importantly, it was recently described that PD-L1+ T cells mainly have tolerogenic effects on tumor immunity and exert tumor-promoting properties, suggesting that targeting this immune population is indeed also beneficial46. We have previously investigated the effect of activating PD-L1-specific T cells in vitro and in vivo and found that, overall, they supported the effector phase of an immune response by removing PD-L1-expressing regulatory immune cells that inhibit PD-1+ effector T cells47,48. The major role of the PD-1 pathway is believed to be the regulation of effector T cell responses. Thus, this protective pathway is more important after activation, rather than at the initial T cell-activation stage49. Accordingly, the presence of PD-L1-specific T cells during the activation phase of an immune response may not increase or support a pro-inflammatory immune response due to the expression of PD-L1 on potent antigen-presenting cells or on the T cells themselves14. Thus, the overall effects of PD-L1-specific T cells may vary depending on the expression of both PD-1 and PD-L1, that is, due to the microenvironment and the state of the immune response. In conclusion, here we report an impressive response rate, CR rate and mPFS for a first-in-class immune-modulating vaccine combined with nivolumab. This may be a first step toward a new treatment strategy for patients with MM. Limitations are the low number of patients treated at a single institution and the lack of a randomized design with aPD1 monotherapy as comparator. Studies in aPD1 therapy-resistant or -refractory melanoma are ongoing as well as biomarker analysis for selecting patients at a higher likelihood to benefit from the combination versus aPD1 monotherapy. A larger randomized trial will be essential to validate these findings and determine the specific contribution of the vaccine to clinical responses and changes in the TME. In December 2020, the Food and Drug Administration granted breakthrough therapy designation for the IO102/IO103 vaccine combined with aPD1 therapy in metastatic melanoma based on data from the MM1636 trial. Methods Trial design and treatment plan MM1636 is an investigator-initiated, nonrandomized, open-label, single-center phase 1/2 study. All patients were treated at the Department of Oncology, Herlev and Gentofte Hospital, University of Copenhagen, Herlev, Denmark. The study was conducted according to the Declaration of Helsinki and Good Clinical Practice (GCP) and monitored by the GCP unit in Copenhagen, Denmark. The protocol was approved by the ethical committee of the Capital region of Denmark (H-17000988), the Danish Medical Agencies (2017011073) and the Capital Region of Denmark Data Unit (P-2019-172). The study was registered at https://clinicaltrials.gov/ under the identifier NCT03047928 and at the EudraCT (no. 2016-0004527-23). This study initially aimed to include 30 aPD1 treatment-naive patients with MM. An amendment with the addition of two other cohorts with ten patients in each cohort was made to evaluate immune responses and clinical efficacy in aPD1 therapy-resistant patients (cohort B, de novo resistance and cohort C, acquired resistance) for a total of 50 patients. The amendment with cohorts B and C was approved at the inclusion of 18 patients in cohort A. The trial is still including patients in cohorts B and C. This article reports results from cohort A. The first six patients were treated as for phase 1, evaluating for safety and tolerability before the remaining 24 patients were included in phase 2. The IDO/PD-L1 vaccine was administered subcutaneously biweekly for the first 10 weeks and thereafter every 4 weeks for approximately 9 months. A maximum of 15 vaccines were administered. Nivolumab was administered according to the approved label (3 mg per kg biweekly) for 24 cycles. The 15th vaccine was administered together with the 24th nivolumab cycle of 3 mg per kg, and responding patients thereafter continued nivolumab monotherapy every 4th week (6 mg per kg) as a standard of care after investigator assessment. Treatment with nivolumab was discontinued at the maximum benefit (assessed by the investigator), after a maximum of 2 years of therapy, at progression or due to severe adverse events (Extended Data Fig. 1b, treatment plan). Vaccine composition Each vaccine was composed of 100 µg IO102, a 21-amino-acid peptide (DTLLKALLEIASCLEKALQVF) from IDO, and 100 µg IO103, a 19-amino-acid peptide (FMTYWHLLNAFTVTVPKDL) from the signal peptide of PD-L1 (PolyPeptide Laboratories). Peptides were dissolved separately in 50 µl DMSO, filtered for sterility and frozen at −20 °C (NUNC CyroTubes CryoLine System Internal Thread, Sigma-Aldrich). At <24 h before administration, peptides were thawed. The PD-L1 peptide was diluted in 400 µl sterile water and, immediately before injection, mixed with the IDO peptide solution and 500 µl Montanide ISA-51 (SEPPIC) to achieve a total volume of 1 ml. Patients Patients above 18 years of age with locally advanced or stage 4 melanoma according to the AJCC (seventh edition), at least one measurable lesion according to RECIST 1.1 and an ECOG PS of 0–1 were eligible. The main exclusion criteria were prior treatment with aPD1 therapy, CNS metastases >1 cm, severe comorbidities and active autoimmune disease. Enrollment was not restricted to PD-L1 status, but it was known before inclusion. Patients were included after informed consent. Patients MM42 and MM20 have confirmed their approval of PET–CT images and clinical images that have been included in the article. Key study assessments Safety and tolerability were evaluated based on changes in clinical laboratory analyses and reported adverse events. Adverse events were assessed according to Common Terminology Criteria for Adverse Events (version 5.0) and were graded from 1 to 5 for all treated patients up to 6 months after the last dose of the IDO/PD-L1 vaccine. Clinical efficacy was assessed using FDG PET–CT scans before treatment and every 3rd month until progression. Objective responses were categorized into CR, PR, stable disease or PD according to RECIST version 1.1. Clinical data were collected at CCIT-DK in the eCRF program OpenClinica version 1.0 and in Microsoft Excel version 2002 on a secure server. Blood samples for immunologic analyses were collected before treatment, before the third cycle, after the sixth, 12th, 18th and 24th cycles (on vaccination) and 3 and 6 months after the last vaccine. Two to three tumor needle biopsies (1.2 mm) were collected at baseline and after six cycles from the same tumor site, when accessible, to evaluate immune responses at the tumor site. DTH skin tests were performed and punch biopsies were taken from the DTH area after cycle 6 for the evaluation of SKILs reactive to PD-L1 and IDO (Supplementary Fig. 1, treatment plan). Statistical analysis of clinical outcome Survival curves were computed in GraphPad Prism software version 9.0.0 according to Kaplan–Meier method. Median follow-up time of enrollment was calculated using the reverse Kaplan–Meier method, also in GraphPad Prism software version 9.0.0. For binary outcomes, 95% two-sided CIs were constructed using the Clopper–Pearson method, also in GraphPad version 9.0.0. An independent board of certified and experienced oncoradiologists performed an external review to evaluate clinical response to address the potential bias of investigator site review. The external review took place at Rigshospitalet, Copenhagen University Hospital. This hospital did not participate in the MM1636 trial, and the external reviewer had no prior knowledge about the clinical trial or the trial therapy. Only PET–CT images were accessed, and arrows indicating target or non-target lesions appeared on baseline images as the only additional information. To address potential trial bias regarding treatment effect, we matched patients in MM1636 with patients from the DAMMED, a population-based database that retrospectively collects data on patients with metastatic melanoma in Denmark. Here, data from 938 patients treated with aPD1 monotherapy contemporaneously (January 2015–October 2019) were extracted. Two hundred and eighteen of these patients were eligible for comparison and matching (all parameters available) (Supplementary Table 1), and 74 patients from the DAMMED were found to match. Patients were matched for age (≤70 years, >70 years), sex, LDH levels (normal, elevated), M stage (M1a, M1b, M1c), BRAF status (wild type, mutated) and PD-L1 status (<1%, ≥1%). An exact matching algorithm was used in which patients in MM1636 were matched with patients from the DAMMED with the same combination of variables. Twenty-nine patients from MM1636 were matched with exact combinations of the six variables. One patient could not be matched. To secure balance of the calculations, control patients were weighted according to the number of patients for each MM1636 patient. Estimates for treatment effects were calculated by weighted logistic regression analyses and weighted Cox proportional hazard model. The R package ‘Matchlt’ was used for matching patients. As the method chosen for matching control patients to protocol patients, a weighted binary logistic regression model was used for comparing response rates in the two matched cohorts. Odds ratios and response rates, including their corresponding 95% CIs, were extracted from the regression models. All P values were two sided, and P values below 0.05 were considered statistically significant. SAS version 9.4M5 was used for the weighted logistic regression models. Processing of peripheral blood mononuclear cells Peripheral blood was collected from all patients in heparinized tubes and was processed within 4 h. In brief, PBMCs were isolated using Lymphoprep (Medinor) separation. PBMCs were counted on a Sysmex XP-300 analyzer and frozen in Human AB Serum (Sigma-Aldrich, H4522-100ML) with 10% DMSO using controlled-rate freezing (CoolCell, BioCision) in a −80 °C freezer and were moved the next day to a freezer at −140 °C until further processing. Needle biopsies at baseline and after the sixth vaccine Two to three needle biopsies (1.2 mm) were taken at baseline and after the sixth cycle of treatment when assessable from the same tumor lesion. One fragment was fixed with formalin and embedded in paraffin (FFPE); one to two fragments were used for expanding tumor-infiltrating T cells and establishing autologous tumor cell lines. Delayed-type hypersensitivity and generation of skin-infiltrating lymphocytes After six cycles of treatment, we performed intradermal injections of vaccine components without adjuvant and one control injection containing DMSO without peptide. Patients were injected with either a mixture of both IDO and PD-L1 peptides at all three injection sites or PD-L1 peptide, IDO peptide or a mixture of the two at the injection sites, respectively (Supplementary Table 6). Eight hours after injection, punch biopsies were resected from the three sites where the peptide was injected and transported immediately to the laboratory and cut into fragments of 1–2 mm3. SKILs were expanded to establish ‘young SKILs’ in CM medium consisting of RPMI 1640 with GlutaMAX, 25 mM HEPES, pH 7.2 (Gibco, 72400-021), IL-2 (100 or 6,000 IU ml−1) (Proleukin Novartis, 004184), 10% heat-inactivated Human AB Serum (Sigma-Aldrich, H4522-100ML), 100 U ml−1 penicillin, 1.25 µg ml−1 Fungizone (Bristol Myers Squibb, 49182) and 100 µg ml−1 streptomycin (Gibco, 15140-122). Half of the medium was replaced three times per week. Young SKILs from samples derived from IDO peptide, PD-L1 peptide and the mixture injection sites were further expanded in a small-scale version of the 14-d rapid expansion protocol as previously described51. Quantification of specific T cells in blood by ELISPOT For enumeration of vaccine-specific T cells in the peripheral blood, PBMCs from patients were stimulated with IDO or PD-L1 peptide in the presence of a low dose of IL-2 (120 U ml−1) for 7–13 d before being used in IFN-γ ELISPOT assays with 2.5–3.2 × 105 cells per well. Briefly, cells were placed in a 96-well PVDF membrane-bottomed ELISPOT plate (MultiScreen MSIPN4W50, Millipore) precoated with IFN-γ-capture antibody (clone 1-D1K, Mabtech). Diluted IDO or PD-L1 peptide stocks in DMSO were added at 5 µM; an equivalent amount of DMSO was added to control wells. PBMCs from each patient were set up in duplicate or triplicate for peptide and control stimulations. Cells were incubated in ELISPOT plates in the presence of the peptide for 16–18 h, after which they were washed off, and biotinylated secondary antibody (anti-human IFN-γ mAB, clone 7-B6-1, Mabtech) was added. After 2 h of incubation, unbound secondary antibody was washed off, and streptavidin-conjugated alkaline phosphatase (Mabtech) was added for 1 h. Next, unbound streptavidin-conjugated enzyme was washed off, and the assay was developed by adding BCIP/NBT substrate (Mabtech). ELISPOT plates were analyzed on the CTL ImmunoSpot S6 Ultimate V analyzer using ImmunoSpot software version 5.1. Responses were calculated as the difference between the average numbers of spots in wells stimulated with IDO or PD-L1 peptide and those from corresponding control wells. For detection of IDO and PD-L1 peptide responses ex vivo, PBMCs were thawed and rested overnight in medium containing DNase I (1 μg ml−1, Sigma-Aldrich, 11284932001) before being used in IFN-γ ELISPOT assays as described above. A total of 6–9 × 105 PBMCs per well were seeded. Statistical analysis of all ELISPOT responses was performed using the DFR method as described by Moodie et al. using RStudio software (RStudio Team, 2016, http://www.rstudio.com/)50. Vaccine-specific ELISPOT responses were defined as true if the difference between the spot count in control and peptide-stimulated wells was statistically significant according to DFR statistical analysis or, for samples performed in duplicate, if the spot count in peptide-stimulated wells was at least 2× the spot count in control wells50. Cancer cell lines and tumor conditioned medium Autologous melanoma cell lines were established from needle biopsies. Briefly, biopsies were chopped into small fragments and seeded in 24-well cultures in RPMI 1640 with GlutaMAX, 25 mM HEPES, pH 7.2 (Gibco, 72400-021), 10% heat-inactivated FBS (Life Technologies, 10500064), 100 U ml−1 penicillin, 1.25 µg ml−1 Fungizone (Bristol Myers Squibb, 49182) and 100 µg ml−1 streptomycin (Gibco, 15140-122). Established adherent melanoma tumor cell lines were cryopreserved at −140 °C in freezing medium containing FBS with 10% DMSO. PD-L1 and HLA-II expression on established tumor cell lines was assessed by flow cytometry staining with anti-PD-L1–PE-Cy7 (1:23 dilution, clone M1h1 (RUO), BD, 558017) and anti-HLA-II–FITC (1:23 dilution, clone Tu39 (RUO), BD, 555558) antibodies. To obtain TCM, established tumor cell lines were cultured in 175-cm2 Nunc cell culture flasks until 80–90% confluency was reached. The culture medium was then replaced with 20 ml fresh X-VIVO 15 with Gentamicin and Phenol Red (Lonza, BE02-060Q), medium with 5% heat-inactivated Human AB Serum (Sigma-Aldrich, H4522-100ML). After 24 h of incubation, TCM was collected and centrifuged to remove any resuspended cells, after which TCM was aliquoted, frozen and stored at −80 °C. The acute monocytic leukemia cell line MonoMac1 was obtained from the DSMZ—German Collection of Microorganisms and Cell Cultures (ACC 252) and cultured in RPMI 1640 with GlutaMAX, 25 mM HEPES, pH 7.2 (Gibco, 72400-021) and 10% heat-inactivated FBS. Isolation of autologous myeloid cells Autologous CD14+ cells were sorted from freshly thawed PBMCs using a magnetic bead-separation kit (Miltenyi Biotec, 130-050-201) according to the manufacturer’s instructions. Isolated CD14+ cells were used as targets in the IFN-γ ELISPOT assay directly after sorting or differentiated in vitro into tumor-associated macrophages by culturing with 1 ml fresh X-VIVO 15 medium with Gentamicin and Phenol Red (Lonza, BE02-060Q) and 5% heat-inactivated Human AB Serum, supplemented with 1 ml autologous TCM in 24-well plates for 2 d. Quantification of vaccine-specific T cells from DTH biopsy sites by ELISPOT SKILs expanded in vitro were rested in medium without IL-2 overnight before being used in the IFN-γ ELISPOT assay as described above to evaluate reactivity of skin-infiltrating T cells. Generation of IDO- and PD-L1-specific T cell cultures from PBMCs or SKILs IDO- or PD-L1-specific T cells were isolated from peptide-stimulated in vitro PBMC cultures on days 14–15 after stimulation or from SKIL cultures expanded in vitro. For specific T cell isolation, PBMCs or SKILs were stimulated with IDO or PD-L1 peptide, and cytokine-producing T cells were sorted using the IFN-γ or TNF-α Secretion Assay—Cell Enrichment and Detection kit (Miltenyi Biotec). Cytokine-production profile of PD-L1- and IDO-specific T cells by intracellular staining To assess the T cell cytokine-production profile, isolated and expanded IDO- and PD-L1-specific T cell cultures were stimulated for 5 h with peptide at 5 μM in a 96-well plate. One hour after the start of the incubation, anti-CD107a–PE (1:133 dilution, clone H4A3, BD Biosciences, 555801) antibody and BD GolgiPlug (1:1,000 dilution, BD Biosciences) were added. After a 5-h incubation, cells were stained using fluorescently labeled surface marker antibodies: anti-CD4–PerCP (1:5 dilution, clone SK3, 345770), anti-CD8–FITC (1:5 dilution, clone SK1, 345772) and anti-CD3–APC-H7 (1:20 dilution, clone SK7, 560275) (all from BD Biosciences). Dead cells were stained using FVS510 (1:250 dilution, BD Biosciences, 564406), followed by overnight fixation and permeabilization using eBioscience Fixation/Permeabilization buffers (00-5123-43, 00-5223-56) according to the manufacturer’s instructions. Cells were then stained intracellularly in eBioscience permeabilization buffer (00-8333-56) with anti-IFN-γ–APC (1:20 dilution, clone 25723.11, BD Biosciences, 341117) and anti-TNF-α–BV421 (1:100 dilution, clone Mab11, BD Biosciences, 562783) antibodies. Samples were analyzed on the FACSCanto II (BD Biosciences) using BD FACSDiva software version 8.0.2. The gating strategy is shown in Supplementary Fig. 3a. For assessing the reactivity against IDO expressing MonoMac1 cells, the appropriate amount of cancer cells was added to the IDO-specific T cells to obtain an effector:target ratio of 4:1 in the 96-well plate and T cell cytokine production was tested after 5 h of co-culture as described above. For HLA-blocking experiments, HLA-blocking antibodies (2 μg ml−1) were added directly into wells 30 min before the addition of peptide. Blocking antibodies used were HLA-DR (1:500 dilution, clone L243, Abcam, ab136320), HLA-DQ (1:500 dilution, clone SPV-L3, Abcam, ab23632) and HLA-DP (1:500 dilution, clone B7/21, Abcam, ab20897). To assess ex vivo T cell reactivity to IDO and PD-L1 peptides in patient PBMCs, cells were thawed and rested for 1–2 d in medium containing DNase I (1 μg ml−1, Sigma-Aldrich, 11284932001). PBMCs were then stimulated with peptide at 5 μM in a 96-well plate for 8 h. An hour after the addition of peptide, anti-CD107a–BV421 (3:50 dilution, clone H4A3, BD Biosciences, 328626) antibody and BD GolgiPlug (1:1,000 dilution, BD Biosciences) were added. Surface and intracellular staining was performed as described above. Antibodies used for surface staining were anti-CD3–PE-CF594 (0.8:30 dilution, clone UCHT1, BD Biosciences, 562280), anti-CD4–BV711 (1:30 dilution, clone SK3, BD Biosciences, 563028) and anti-CD8–Qdot605 (1:150 dilution, clone 3B5, Thermo Fisher, Q10009). Antibodies used for intracellular staining were anti-CD137–PE (1:20 dilution, clone 4B4-1 (RUO), BD Biosciences, 555956), anti-IFN-γ–PE-Cy7 (1.5:20 dilution, clone B27, BD Biosciences, 557643) and anti-TNF-α–APC (1:20 dilution, clone Mab11, BD Biosciences, 554514). Samples were acquired on the NovoCyte Quanteon (ACEA Biosciences) and analyzed using NovoExpress software version 1.4.1. To assess vaccine-specific T cell responses, background values observed in unstimulated PBMC samples were subtracted from values observed in peptide-stimulated conditions. Positive response value threshold was set at a difference of 0.2% from the background values. Based on this response cutoff, only TNF-α, CD107a and CD137 responses were detected in this assay. Statistical analysis comparing baseline with on-treatment or post-treatment cytokine profiles was performed using two-sided Wilcoxon matched-pairs signed-rank test. The gating strategy is shown in Supplementary Fig. 3b. siRNA-mediated PD-L1 and IDO1 silencing A stealth siRNA duplex for targeted silencing of PD-L1 (Invitrogen)52, a custom silencer select siRNA for targeted silencing of IDO1 (Ambion) and the recommended silencer select negative control (Ambion) siRNA for mock transfection were used. The stealth PD-L1 siRNA duplex consisted of the sense sequence 5′-CCUACUGGCAUU-UGCUGAACGCAUU-3′ and the antisense sequence 5′-AAUGCGUUCAGCAAAUGCCAGUAGG-3′. The three silencer IDO siRNA duplexes used were siRNA1 (sense sequence, 5′-ACAUCUGCCUGAUCUCAUATT-3′; antisense sequence, 5′-UAUGAGAUCAGGCAGAUGUTT-3′), siRNA2 (sense sequence, 5′-CCACGAUCAUGUGAACCCATT-3′; antisense sequence, 5′-UGGGUUCACAUGAUCG-UGGAT-3′) and siRNA3 (sense sequence, 5′-CGAUCAUGUGAACCCAAAATT-3′; antisense sequence, 5′-UUUUGGGUUCACAUGAUCGTG-3′). For PD-L1 or IDO1 silencing, cancer cells were electroporated with 0.025 nmol of each siRNA duplex as previously described53. For PD-L1-silencing experiments, cancer cells were treated with IFN-γ (500 U ml−1, PeproTech) 1 h after electroporation. Electroporated cells were used as target cells in ELISPOT and ICS assays 24 h or 48 h after siRNA electroporation. RT–qPCR Total RNA was extracted using the RNeasy Plus Mini kit (Qiagen, 74134) following the manufacturer’s instructions. RNA concentration was quantified using a NanoDrop 2000 (Thermo Fisher Scientific), and a total of 1,000 ng RNA was reverse transcribed with the High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems, 4368814) using 1,000 ng input RNA for transcription. Real-time qPCR analyses were performed using the TaqMan Gene Expression Assay on a Roche LightCycler 480 instrument. RT–qPCR assays were performed with a minimum of three technical replicates and analyzed using the ΔCt method as described in Bookout et al.54 with normalization of IDO1 expression (primer ID, Hs00984148_m1) or PD-L1 expression (primer ID, Hs001125296_m1) to the expression level of the housekeeping gene POLR2A (primer ID, Hs00172187_m1) and the control sample (mock). For low-concentration samples with no amplification, Ct was set to 40. Controls without reverse transcriptase were used as controls for specific amplifications. P values were determined using two-tailed parametric t-tests. BRAF-mutational status and PD-L1 status at baseline in all patients A library of historical FFPE biopsies were assessable for all patients and analyzed locally at Herlev and Gentofte University Hospital by experienced pathologists for BRAF status and PD-L1 expression on tumor cells. The BRAF analysis was carried out with real-time PCR using the EntroGen BRAF Mutation Analysis Kit II (BRAFX-RT64, CE-IVD) to specifically detect mutations corresponding to V600D, V600E and V600K in the BRAF protein. PD-L1 status was assessed using the monoclonal rabbit anti-PD-L1 antibody (clone 28.8, PD-L1 IHC 28-8 pharmDx) in FFPE biopsies following the manufacturer’s instructions. Patients were considered positive for PD-L1 with expression levels ≥1% and negative for PD-L1 with expression levels <1%. Human leukocyte antigen type Blood samples from all 30 patients were genotyped for class I (HLA-A, HLA-B, HLA-C) and three class II types (HLA-DRB1, HLA-DQA1, HLA-DQB1) using LinkSēq HLA Typing Kits (Thermo Fisher, 1580C). These test kits are based on real-time PCR using allele-specific exponential amplification (sequence-specific primers) followed by melting curve analyses. Immunohistochemistry simplex: Immunoscore CR IHC staining was performed using a qualified Ventana Benchmark XT with four different steps: (1) antigen retrieval; (2) staining with the following primary antibodies: anti-CD3 antibody (clone HDX2; provider, HalioDx; HD-FG-000013; 10931065/10636667; concentration, 0.25 µg ml−1), anti-CD8 antibody (clone HDX3; provider, HalioDx; HD-FG-000019; 10931069/10337710/10639301; concentration, 0.5 µg ml−1), anti-IDO monoclonal antibody (clone VINC3IDO; provider, Thermo Fisher Scientific; 14-9750-82; E25003-101; concentration, 0.05 µg ml−1), anti-HLA class 1 ABC antibody (clone, EMR8-5; provider, Abcam; ab70328; GR3248333/GR3186494; concentration, 0.5 µg ml−1), anti-HLA-DR/DP/DQ/DX antibody (clone CR3/43; provider, Santa Cruz; sc-53302; L1714; concentration, 1 µg ml−1) and anti-PD-L1 antibody (clone HDX3; provider, HalioDx; HD-FG-000035; 106312810/106312816; concentration, 3.3 µg ml−1); (3) detection with a secondary antibody using the ultraView Universal DAB Detection kit; and (4) counterstaining using Hematoxylin & Bluing Reagent (staining of cellular nuclei). Control slides were systematically included in each staining run to permit quality control of the obtained measurements. Following coverslipping, slides were scanned with the NanoZoomer-XR to generate digital images (20×). Two consecutive slides were specifically used to perform Immunoscore CR TL staining of CD3+ and CD8+ cells. Digital pathology of T lymphocytes The digital pathology for Immunoscore CR TL allowed the quantification of positive cells (in cells per mm2) into the core tumor and invasive margin if present. Each sample was analyzed using the HalioDx Digital Pathology Platform. Digital pathology and pathologist analysis of IDO Digital pathology for IDO allowed the quantification of stained area (in mm2) into the whole tumor. Each sample was analyzed using the HalioDx Digital Pathology Platform. Analysis of IDO+ cells was performed by a pathologist. Results were expressed as an H score from 0 to 300. The score is obtained by the formula 3 × percentage of cells with strong staining + 2 × percentage of cells with moderate staining + percentage of cells with weak staining. Digital pathologist analysis of MHC-I and MHC-II Analysis of MHC-I+ and MHC-II+ cells was performed by a pathologist. Results are expressed as a percentage of positive tumor cells and a percentage of positive cells in the stroma. Digital pathology Immunoscore immune checkpoint (CD8+ and PD-L1+) The Immunoscore CR IC test allowed the quantification of CD8++ cell density in the whole tumor and a CD8+-centered proximity index (which corresponds to the percentage of CD8++ cells that have at least one PD-L1+ cell in the neighborhood) at different cutoff distances (20 µm, 40 µm, 60 µm and 80 µm). Pathologist analysis of PD-L1 A pathologist performed analysis of PD-L1+ cells. Positivity of a viable tumor cell was considered when partial or complete cell membrane staining was observed (more than 10% of the tumor cell membrane). Results were expressed as a percentage. NanoString RNA profiling and Immunosign RNA was extracted from FFPE tissues using Qiagen RNeasy FFPE extraction kits (Qiagen). Annotations from the pathologist performing H&E staining were used to guide removal of normal tissue from the slides by macrodissection before nucleic acid extraction, which occurred after tissue deparaffinization and lysis. Each extracted RNA sample was independently quantified using a NanoDrop spectrophotometer (NanoDrop Technologies) and qualified (Agilent Bioanalyzer). Degradation was quantified as the percentage of RNA fragments smaller than 300 bp using the RNA 6000 Nano kit (Agilent Bioanalyzer). Good sample quality was defined as less than 50% of RNA fragments being 50–300 bp in size. RNA expression profiling was performed using the nCounter PanCancer Immune Profiling Panel from NanoString (NanoString Technologies). The PanCancer Immune Profiling Panel contains 776 probes and is supplemented with six genes to complete HalioDx Immunosign targets. Hybridization was performed according to the manufacturer’s instructions. Hybridized probes were then purified and immobilized on a streptavidin-coated cartridge using the nCounter Prep Station (NanoString Technologies). Data collection was carried out on the nCounter Digital Analyzer (NanoString Technologies) following the manufacturer’s instructions to count individual fluorescent barcodes and quantify target RNA molecules present in each sample. For each assay, a scan of 490 fields of view was performed. Raw data from the NanoString nCounter were processed using NanoString recommendations. Quality control enables to retain data of good quality with a binding density that ranges between 0.05 and 2.25. The linearity of positive controls was checked using the R2 value of regression between the counts and the concentrations of positive controls. Samples that showed R2 < 0.75 were flagged and removed from the analysis. The background was removed using a thresholding method at the mean + 2 s.d. of negative controls. Raw counts were normalized using a positive normalization factor. Samples showing positive normalization factors outside the range of 0.3–3 were removed from the analysis. A second normalization was performed using the housekeeping gene normalization factor. Only the most stable housekeeping genes were selected for this normalization step using the variance-versus-mean relationship. All samples showing a normalization factor outside the range of 0.1–10 were removed from the analysis. All statistical analyses were performed on normalized counts using R software (version 2.6.2, 2019-12-12). T cell receptor variable β chain sequencing To track longitudinal immune responses to therapy, genomic DNA was extracted from longitudinal pre- and post-treatment PBMCs (five patients), pre- and post-treatment biopsies (five patients) (both FFPE and treated with RNAlater) and IDO- and PD-L1-specific T cell cultures from PBMCs (five patients) or SKILs (one patient). Three clonal (two IDO- and one PD-L1-) specific cultures were also sequenced to confirm clonal purity of cultures. DNA from PBMCs or RNAlater-treated biopsies was extracted with the DNeasy Blood and Tissue kit (Qiagen, 69504), DNA from sorted IDO and PD-L1-specific T cells from either PBMCs or SKILs was extracted using the QIAamp DNA Micro kit (Qiagen, 565304), and DNA from PFFE biopsies was extracted using the Maxwell RSC DNA FFPE kit (Promega, AS1450). Immunosequencing of the CDR3 regions of human TCRβ chains was performed using the immunoSEQÒ Assay (Adaptive Biotechnologies). Extracted genomic DNA was amplified in a bias-controlled multiplex PCR, followed by high-throughput sequencing. Sequences were collapsed and filtered to identify and quantitate the absolute abundance of each unique TCRβ CDR3 region for further analysis as previously described55–57. Statistical analyses of TCRβ sequencing results Two quantitative components of diversity were compared across samples in this study. First, Simpson clonality was calculated on productive rearrangements by ∑i=1Rpi2, where R is the total number of rearrangements and pi is the productive frequency of rearrangement i. Values of Simpson clonality range from 0 to 1 and measure how evenly receptor sequences (rearrangements) are distributed among a set of T cells. Clonality values approaching 0 indicate a very even distribution of frequencies, whereas values approaching 1 indicate an increasingly asymmetric distribution in which a few clones are present at high frequencies. Second, sample richness was calculated as the number of unique productive rearrangements in a sample after computationally downsampling to a common number of T cells to control for variation in sample depth or T cell fraction. Repertoires were randomly sampled without replacement five times, and we report the mean number of unique rearrangements. The T cell fraction was calculated by taking the total number of T cell templates and dividing by the total number of nucleated cells. The total number of nucleated cells was derived from reference genes using the immunoSEQ Analyzer version 3. To identify enriched vaccine clones in each patient, rearrangement frequencies in their baseline PBMCs and each IDO/PD-L1-sorted T cell sample were compared using a binomial distribution framework as previously described58. In brief, for each clone, we performed a two-sided test to determine whether frequencies were the same in the patient’s periphery and a PD-L1- or IDO-specific T cell sample. The Benjamini–Hochberg procedure was used to control the false discovery rate at 0.01 (ref. 59). Clonal expansion in post-treatment samples was similarly assessed using this differential abundance framework, but an IDO/PD-L1-specific T cell sample was replaced with a post-treatment series sample. In biopsies, the six series frequencies were compared to those of baseline tissue. Lastly, vaccine-associated clones were tracked in each PBMC and tissue sample by summing the frequency of each rearrangement enriched in either PD-L1- or IDO-specific T cells. All statistical analyses were performed in R version 3.6.1. Reporting Summary Further information on research design is available in the Nature Research Reporting Summary linked to this article. Online content Any methods, additional references, Nature Research reporting summaries, source data, extended data, supplementary information, acknowledgements, peer review information; details of author contributions and competing interests; and statements of data and code availability are available at 10.1038/s41591-021-01544-x. Supplementary information Supplementary Information Supplementary Figs. 1–3 and Tables 1–6 Reporting Summary Extended data Extended Data Fig. 1 Mode of Action and treatment plan. a) Anticipated mechanism of action of the combination therapy of an IDO/PD-L1 derived peptide vaccine and nivolumab (anti-PD1). 1) The IDO/PD-L1 peptide vaccine is administered subcutaneously (s.c) and nivolumab is administered intravenous (IV). 2) The vaccines peptides are phagocytosed by an antigen presenting cell and presented to IDO and PD-L1 specific T cells, which are activated. 3) The activated T cells migrates to the tumor site where they attack both immune-suppressive cells and tumour cells expressing IDO and/or PD-L1 leading to cytokine production and a pro-inflammatory tumour microenvironment. 4/5) Enhanced tumour killing by both IDO/PD-L1 specific T cells and tumor specific cytotoxic T cells due to PD-1 blockade. Created with BioRender.com b) Treatment plan. After written informed content patients were screened. Before treatment start a baseline PET/CT scan was performed, baseline blood sample for research use and if assessable a baseline needle biopsy. Patients were treated with the IDO/PD-L1 peptide vaccine subcutaneously biweekly for the first 6 injections and thereafter every fourth week for a maximum of 15 vaccines. Nivolumab was administered in parallel biweekly (3mg/kg) up to 24 series. If patients needed subsequent nivolumab after ended vaccination regiment they were treated with 6 mg/kg every fourth week up to two years. Needle biopsy and delayed type hypersensitivity (DTH) was performed after 6 series of treatment if assessable. PET/CT scans was performed every third month. Extended Data Fig. 2 Progression free survival and overall survival in matched historical control group and vaccine injection site reaction. a) Kaplan-Meier curve of progression free survival with corresponding 95% confidence intervals in the matched historical control group (n=74). Patients were matched on BRAF-status, PD-L1-status, age, gender, M-stage and LDH level). b) Kaplan-Meier curve of overall survival with corresponding 95% confidence intervals in the matched historical control group (n=74). c) Images of injection site reaction in patient MM20 (CR) after 11 vaccination show redness, rash and granuloma at injection site. Extended Data Fig. 3 Vaccine specific responses in blood. a) Heatmap of specific (background has been subtracted) IDO and PD-L1 responses in PBMCs at baseline, series 3, 6, 12, 18 and 24 measured by in vitro IFN-γ Elispot assay show fluctuations in the blood during treatment (n=30). b) Heatmap of specific (background has been subtracted) IDO and PD-L1 responses in PBMCs at baseline and 3 and 6 months after last vaccine measured by in vitro IFN- γ Elispot assay (n=30). 2.5–3.2x105 cells per well were used. c) Vaccine associated clones were tracked in the blood by summing the frequency of each rearrangement enriched in either IDO or PD-L1 T cells. Extended Data Fig. 4 Ex vivo vaccine specific responses in blood. Heatmaps of detected specific (background has been subtracted) IDO (a) and PD-L1 (b) responses in PBMCs at baseline and on/after treatment as measured by IFNγ ELISPOT (n=25). C) Example of well images of ex vivo ELISPOT wells for three different patients (n=3). 6-9x105 cells per well were used. Extended Data Fig. 5 CD4 and CD8 vaccine specific T cell responses in blood. Top: heatmaps of IDO specific CD4 (left) and CD8 (right) T cell responses in PBMCs at baseline and on/post treatment. Bottom: heatmaps of PD-L1 specific CD4 (left) and CD8 (right) T cell responses. Responses quantified by flow cytometry by an increased expression of CD107α, CD137 and TNFα after 8h peptide stimulation. Values represent specific responses after background values have been subtracted (n=21). Statistical analysis were performed using two-sided Wilcoxon matched-paired rank test. Extended Data Fig. 6 Pro-inflammatory profiles of sorted CD4 and CD8 T cells from blood. a/b/c/d) Percentage of in vitro stimulated and sorted PD-L1 and IDO specific CD4 and CD8 T cells secreting cytokines. Extended Data Fig. 7 Vaccine specific responses in skin. a) IDO and PD-L1 specific T cell responses in SKILs after 6 series of treatment measured by IFN-γ Elispot assay (n=13). SKILs were grown from DTH injection with either IDO peptide, PD-L1 peptide or a mix as presented by different blue colours. * Responses were calculated as the difference between average numbers of spots in wells stimulated with IDO or PD-L1 peptide (triplicates) and corresponding control (DMSO) and statistical analyses of Elispot responses were performed using distribution-free resampling method (Moodie et al.). DR: Not statistically confirmed response due to replicate number but number of spots in peptide wells are two times higher than control wells (DMSO). NS: No significant response and no DR. b/c/d) Percentage of cytokine secreting/CD107a+ CD4+ and CD8+ IDO and PD-L1 specific T cells in response to in vitro peptide stimulation by flow cytometry. e) Skin infiltrating PD-L1 specific T cell clones also found in biopsy in patient MM01. TCR sequencing was performed on a PD-L1 specific T cell culture generated from DTH area on the lower back injected with PD-L1 peptide on patient MM01. Bars show the frequency of top 25 clones in the culture with which indicates a high Simpson clonality of 0.43. f) Tracking the frequency of the top five skin infiltrating PD-L1 specific clones in tumour before and after treatment. Extended Data Fig. 8 T cell changes in blood after treatment. T cell fraction, TCR clonality and repertoire richness in blood and peripherally expanded TCR clones in both responding and non-responding patients. a) T cell fraction in peripheral blood of five patients at baseline, series 3, 6 and 12 by TCR sequencing. T cell fraction was calculated by taking the total number of T cell templates and dividing by the total number of nucleated cells (n=5.) b) TCR clonality in peripheral blood of five patients at baseline, series 3, 6 and 12 by TCR sequencing (n=5). Simpson clonality measures how evenly TCR sequences are distributed amongst a set of T cells where 0 indicate even distribution of frequencies and 1 indicate an asymmetric distribution where a few clones dominate. c) TCR repertoire richness in peripheral blood of five patients at baseline, series 3, 6 and 12 by TCR sequencing (n=5). TCR repertoire richness report the mean number of unique rearrangements. d) Bar-chart chart representing dynamics of the expanded T cell clones. Coloured bars represent peripherally expanded clones in five patients at series 3, series 6 and series 12 as compared to the baseline PBMC samples (n=5). Light gray bar represents peripherally expanded clones that are present in baseline biopsy samples while dark gray bar represents peripherally expanded clones that are present in the post-treatment biopsy (after 6 series). e) Frequency of the dominant TCR β chain in clonal PD-L1 and IDO specific cultures as determined by CDR3 sequencing. Extended Data Fig. 9 Profiling of genes relevant to T cell activation, cytokines and exhaustion markers on pre- and post-treatment biopsies in two patients. a) RNA expression profiling of genes related to T cell activation was performed using NanoString nCounter (n=2). b) RNA expression profiling of genes related to cytokine activity was performed using NanoString nCounter (n=2). c) RNA expression profiling of genes related to checkpoint inhibitors was performed using NanoString nCounter (n=2). Extended Data Fig. 10 Treatment induced upregulation of PD-L1, IDO, MHC I and MHC II and distance between CD8 T cells and PD-L1 expressing cells. a) IHC on 4 paired biopsies stained for CD3 and CD8 T cells, PD-L1, MHC I and MHC II on tumor cells/mm2 (sum in the validated area) and IDO H-score (from 0 to 300). H-score is the expression of IDO on both immune and tumor cells: The score is obtained by the formula: 3 x percentage of cells with a strong staining + 2 x percentage of cells with a moderate staining + percentage of cells with a weak staining (n=4). b) Distance in µm between CD8+ T cells and PD-L1+ stained cells on five baseline biopsies detected by IHC (n=5). Extended data is available for this paper at 10.1038/s41591-021-01544-x. Supplementary information The online version contains supplementary material available at 10.1038/s41591-021-01544-x. Acknowledgements We thank all patients and their relatives for being a part of the trial. We thank Ö. Met, M. Jonassen, S. Ullitz Færch, B. Saxild, S. Wendt and C. Grønhøj for technical support. We thank M. Cumberbatch for input with translational analysis. We thank the nurses at clinic 5 and the head of the Oncology Department at Herlev and Gentofte Hospital, L. Sengeløv. The study was funded through a research funding agreement between IO Biotech and the CCIT-DK, Herlev Hospital and the Oncology Department at Herlev Hospital. Author contributions I.M.S. and M.H.A. conceived the study. J.W.K. wrote the protocol with input from I.M.S., M.H.A. and M.-B.Z. J.W.K. and C.L.L. were responsible for patient treatment and patient care with help from R.B.H. and consulting from I.M.S., E. Ellebaek and M.D. I.M.S., E. Ellebaek and M.D. were responsible for patient recruitment. J.W.K. and C.L.L. were responsible for coordinating trial procedures and collecting data and samples. J.W.K. and C.L.L. were clinical investigators, and I.M.S. was a sponsor. J.W.K. and C.L.L. analyzed and interpreted clinical data with support from I.M.S. and E. Ehrnrooth. T.W.K. performed statistical analyses. H.W.H. helped with clinical imaging. J.W.K., E.M., C.O.M., S.M.A., S.E.W.-B., M.O.H. and A.W.P. analyzed and interpreted translational data with support from M.H.A. and I.M.S. Funding was acquired by I.M.S., M.H.A. and M.-B.Z. J.W.K. wrote the manuscript with input from I.M.S., M.H.A., E.M. and A.W.P. All authors reviewed the manuscript, interpreted data and approved the final version. Data availability This clinical trial was registered at https://www.clinicaltrials.gov/ before patient enrollment (clinical trial identifier NCT03047928). Each vaccine was composed of 100 µg IO102, a 21-amino-acid peptide (DTLLKALLEIASCLEKALQVF) from the IDO peptide, and 100 µg IO103, a 19-amino-acid peptide (FMTYWHLLNAFTVTVPKDL) from the signal peptide of PD-L1. TCR sequencing data are available from Adaptive Biotechnologies. Upon request, the CCIT-DK office will provide a username and a password to access the designated data within approximately 2–4 weeks (https://www.herlevhospital.dk/ccit-denmark/find-us/Sider/Contact-information.aspx). All requests for the remaining data including raw data and analyzed data and materials will, within a reasonable time frame, be reviewed by the CCIT-DK office (https://www.herlevhospital.dk/ccit-denmark/find-us/Sider/Contact-information.aspx) to verify whether the request is subject to any intellectual property or obligations. Patient-related data not included in the paper were generated as part of clinical trials and may be subject to patient confidentiality. Any data and materials that can be shared will be released via a material-transfer agreement. The following database was used in the study: https://research.regionh.dk/da/publications/the-danish-metastatic-melanoma-database-dammed(32749d99-095f-4cae-b5de-769bae27f01e).html. Competing interests M.H.A. is named as an inventor on various patent applications relating to therapeutic uses of IDO and PD-L1 peptides. These patent applications are assigned to the company IO Biotech, which is developing immune-modulating cancer treatments. M.H.A. is a founder, shareholder and advisor for IO Biotech. I.M.S. is a cofounder, shareholder and advisor for IO Biotech. I.M.S. has an advisory board relationship with or lectured for Roche, Novartis, MSD, Celgene, Incyte, TILT Bio, Pfizer and BMS AstraZeneca and has received limited grants for translational research from BMS, Roche and Novartis. M.-B.Z. is the CEO, founder and shareholder at IO Biotech. E.M., A.W.P. and E. Ehrnrooth are employees at IO Biotech. E. Ellebaek has received honoraria from BMS, Pierre Fabre, Roche and Kyowa Kirin and travel support from MSD. M.D. has received honoraria from Genzyme, MSD, BMS, Roche and Novartis and travel support from Novartis, MSD, BMS, Roche and Pfizer. The remaining authors declare that they have no conflict of interest. Peer review information Nature Medicine thanks George Coukos, Michael Schell and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. Javier Carmona was the primary editor on this article and managed its editorial process and peer review in collaboration with the rest of the editorial team. Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. These authors contributed equally: Julie Westerlin Kjeldsen, Cathrine Lund Lorentzen. Change history 3/8/2022 A Correction to this paper has been published: 10.1038/s41591-022-01771-w

Intravenous (not otherwise specified)

DrugAdministrationRoute

CC BY

34887574

2021-12-09

What was the dosage of drug 'NIVOLUMAB'?

A phase 1/2 trial of an immune-modulatory vaccine against IDO/PD-L1 in combination with nivolumab in metastatic melanoma. Anti-programmed death (PD)-1 (aPD1) therapy is an effective treatment for metastatic melanoma (MM); however, over 50% of patients progress due to resistance. We tested a first-in-class immune-modulatory vaccine (IO102/IO103) against indoleamine 2,3-dioxygenase (IDO) and PD ligand 1 (PD-L1), targeting immunosuppressive cells and tumor cells expressing IDO and/or PD-L1 (IDO/PD-L1), combined with nivolumab. Thirty aPD1 therapy-naive patients with MM were treated in a phase 1/2 study ( https://clinicaltrials.gov/ , NCT03047928). The primary endpoint was feasibility and safety; the systemic toxicity profile was comparable to that of nivolumab monotherapy. Secondary endpoints were efficacy and immunogenicity; an objective response rate (ORR) of 80% (confidence interval (CI), 62.7-90.5%) was reached, with 43% (CI, 27.4-60.8%) complete responses. After a median follow-up of 22.9 months, the median progression-free survival (PFS) was 26 months (CI, 15.4-69 months). Median overall survival (OS) was not reached. Vaccine-specific responses assessed in vitro were detected in the blood of >93% of patients during vaccination. Vaccine-reactive T cells comprised CD4+ and CD8+ T cells with activity against IDO- and PD-L1-expressing cancer and immune cells. T cell influx of peripherally expanded T cells into tumor sites was observed in responding patients, and general enrichment of IDO- and PD-L1-specific clones after treatment was documented. These clinical efficacy and favorable safety data support further validation in a larger randomized trial to confirm the clinical potential of this immunomodulating approach. pmcMain Despite remarkable advances in the treatment of MM with immune checkpoint inhibitors (ICIs) targeting PD-1 and cytotoxic T lymphocyte antigen 4 (CTLA-4), around half of patients are resistant to ICI monotherapy1. The combination of anti-CTLA-4 (aCTLA-4) and aPD1 therapy is to date the most effective therapy resulting in a response rate of around 60%; however, 50% of the patients also develop severe adverse events2,3. Therefore, an equally effective but less toxic treatment is highly needed. Several approaches to enhance ICI efficacy are currently being investigated, such as other ICIs, T cell therapy with tumor-infiltrating T cells or innate immunity stimulators such as Toll-like receptor 9 agonists4–6. Treating cancer patients with vaccines that stimulate a targeted immune response is another attractive approach, with very few side effects observed thus far in combination immunotherapy studies7,8. Immune-modulatory vaccines targeting tumoral immune escape mechanisms offer a new, generalizable strategy compared to patient-specific neoantigen cancer vaccines7,9. The immune-modulatory vaccine strategy in this clinical trial is based on the finding of circulating cytotoxic T cells specific to IDO and PD-L1 in the blood of patients with cancer and, to a lesser extent, in healthy donors. IDO- and PD-L1-specific CD8+ T cells can directly recognize and kill IDO+ and/or PD-L1+ tumor cells and likewise recognize and kill non-malignant cells that express their cognate targets. Furthermore, IDO- and PD-L1-specific CD4+ T cells release pro-inflammatory cytokines in response to IDO- or PD-L1-expressing target cells. IDO and PD-L1 are expressed not only by melanoma cells but also by many other cell types in the tumor microenvironment (TME), which differentiates these antigens from traditional tumor antigens used in other studies10–15. Activation of IDO/PD-L1-specific T cells by vaccination can therefore restrict the range of immunosuppressive signals mediated by immunosuppressive cells and thereby revert the TME from an immune hostile to an immune friendly environment. In animal models of cancer, vaccinations with IDO epitopes resulted in anti-tumor therapeutic effects that were correlated with reductions in IDO expression in myeloid cell populations within the TME16. The IDO/PD-L1 immune-modulating vaccine may lead to a translatable strategy for improving the efficacy of aPD1 therapy through activation of specific T cells. We hypothesize that the IDO/PD-L1 vaccine attracts T cells into the tumor, which induces type 1 helper T (TH1) cell inflammation and reverts the TME into an immune-permissive site, thereby turning the tumor ‘hot’. This would also upregulate PD-L1 expression in cancer and immune cells, generating more susceptible targets to aPD1 therapy (Extended Data Fig. 1a). This theory was confirmed in a mouse model in which aPD1 therapy and IDO vaccination show synergistic effects16. In this phase 1/2 clinical trial, MM1636, patients with MM received a combination of the IDO/PD-L1 (IO102/IO103) peptide vaccine with the adjuvant Montanide and the aPD1 antibody nivolumab. Patients were included in three cohorts: 30 aPD1 therapy-naive patients (cohort A), ten aPD1 therapy-refractory patients (cohort B, de novo resistance) and ten patients who progressed after aPD1 therapy (cohort C, acquired resistance). Here, we report results from cohort A. The vaccine was given biweekly for the first six administrations and thereafter every 4th week. A maximum of 15 vaccines were administered. Nivolumab was given in parallel, biweekly (3 mg per kg) or every 4th week (6 mg per kg) for up to 2 years (Extended Data Fig. 1b). The primary objective was safety and feasibility. Secondary objectives were immunogenicity and clinical efficacy. Results Patients and treatment Thirty patients were enrolled from December 2017 to June 2020. None of the 30 patients dropped out of the study; all received at least three cycles of therapy (Supplementary Fig. 1). At the current database lock (5 October 2020), six patients were still on treatment in the trial. Of the 24 patients who were not on trial treatment at data cutoff, two are still receiving nivolumab monotherapy (6 mg per kg every 4 weeks). Reasons for stopping treatment for the remaining 22 patients included disease progression (37%), toxicity (20%), maximum benefit or complete response (CR) confirmed on two consecutive scans (17%) or completing 2 years of treatment (7%). For the 24 patients who were not on trial treatment at data cutoff, the mean number of vaccinations was 10.5 (range, 3–15 vaccinations). Thirteen of these 24 patients continued nivolumab (6 mg per kg, every 4 weeks) as a standard of care. Nine patients received subsequent therapy after progression (Supplementary Table 1). Baseline characteristics are shown in Table 1. The mean age was 70 years; 37% of patients had elevated lactate dehydrogenase (LDH) levels, 60% were stage M1c, 37% had BRAF mutations, and 43% were negative for PD-L1 (<1%). A total of three patients (10%) had received prior ipilimumab therapy. No patients had brain metastasis (Supplementary Table 2).Table 1 Baseline patient characteristics (n = 30) Characteristic Number (%) Mean age, years (range) 70 (46–85) Sex, male 16 (55%) ECOG PS 0 26 (87%) LDH levels  ≤ULN 19 (63%)  >ULN 11 (37%) M stage (AJCC-8)  M1a 6 (20%)  M1b 6 (20%)  M1c 18 (60%) Number of lesion sites  1 6 (20%)  2–3 17 (57%)  >3 7 (23%) Liver metastases present 10 (33%) BRAF status  Mutant 11 (37%)  Wild type 19 (63%) PD-L1  <1% 13 (43%)  >1% 17 (57%) Previous systemic therapy  Ipilimumab 3 (10%)  No 27 (90%) AJCC-8, eighth edition of the American Joint Committee on Cancer; ECOG PS, Eastern Cooperative Oncology Group performance status; ULN, upper limit of normal. Notable clinical responses to the combination therapy Thirty patients with MM were treated with the IDO/PD-L1 vaccine and nivolumab according to the trial protocol. By investigator review, the ORR reached 80% (CI, 62.7–90.5%), with 43% of patients (CI, 27.4–60.8%) achieving a CR and 37% (CI, 20.9–54.5%) reaching a partial response (PR) as the best overall response, while 20% experienced progressive disease (PD) according to Response Evaluation Criteria in Solid Tumors (RECIST) 1.1 (Fig. 1a). Two of the patients with a PR did not have a confirmatory scan stating PR on two consecutive scans. Early onset of response was frequent, with 22 of 30 patients having an objective response at the first evaluation (after 12 weeks on treatment). Median times to PR and CR were 75 d (range, 54–256 d) and 327 d (range, 73–490 d), respectively (Fig. 2a–c).Fig. 1 Clinical response. a, Pie charts with percent ORR, CR, PR and PD according to RECIST 1.1 by investigator review of all patients (n = 30), PD-L1+ patients (>1%, (n = 17)) and PD-L1− patients (<1%, n = 13)), respectively. Two-sided CIs (95%) were constructed using the Clopper–Pearson method. b, Treatment effect in MM1636 compared with a matched historical control group from the DAMMED database (n = 74). Patients in MM1636 (n = 29) were matched with the exact same combination variable according to age (≤70 years, >70 years), sex, LDH levels (normal, elevated), M stage (M1a, M1b, M1c), BRAF status (wild type, mutated) and PD-L1 status (<1%, ≥1%). Estimates for treatment effects were calculated by weighted logistic regression analyses and weighted Cox proportional hazard model. Bar height indicates the estimated response rate; tops of bars are centers for error bars. Odds ratios (OR), response rates and their corresponding 95% CIs were extracted from the regression model. All P values were two sided, and P values below 0.05 were considered statistically significant. c, Best change in the sum of target lesion size compared with that at baseline (n = 30). The horizontal line at −30 shows the threshold for defining an objective response in the absence of non-target disease progression or new lesions according to RECIST 1.1. Two patients with 100% reduction in target lesion size had non-target lesions present. White stars, six patients had normalization (<10 mm) of fluor-18-deoxyglucose (FDG)-negative lymph nodes (at baseline, lymph nodes were >1.5 cm and FDG+) and 100% reduction of non-lymph node lesions and are considered to have had a CR (green bar). Black star, one patient (MM29) was considered to have had a PR (blue bar), although he had not reached a −30% change in target lesion size. The patient had a single measurable 13-mm lung metastasis at baseline and multiple biopsy-verified cutaneous metastases on the left crus (not detectable by positron-emission tomography–computed tomography (PET–CT) at baseline). The best change in target lesion size was 10 mm, and a post-treatment biopsy from the cutaneous metastases showed no sign of malignancy. Thus, overall, the patient was classified as having a PR. d, Kaplan–Meier curve of the response duration in the 24 patients with an objective response. e, Kaplan–Meier curve of PFS in all 30 treated patients. f, Kaplan–Meier curve of OS in all 30 treated patients. Fig. 2 Clinical response. a, Swimmer plot showing response duration and time to response according to RECIST 1.1 for all treated patients (n = 30). Triangles indicate first evidence of PR, while squares indicate first evidence of CR. Closed circles indicate time of progression. Arrows indicate ongoing responses. Patient MM18 died due to nivolumab-induced side effects. b, Spider plot showing response kinetics in all treated patients (n = 30). Red squares indicate time of progression. c, PET–CT images of patient MM42 before and after treatment (after 12 series of treatment) showing FDG metabolism in target lesions. SC, subcutaneous. The ORR among PD-L1+ (>1% (clone 28.8)) patients (n = 17) was 94.1% (CI, 73–99.7%) and 61.5% (CI, 35.5–82.3%) in PD-L1− patients (n = 13) (Fig. 1a). Objective responses were observed in patients irrespective of human leukocyte antigen (HLA) genotype (Supplementary Fig. 2). Clinical response data were validated by blinded independent external review, in which an ORR of 76.6% (CI, 57.7–90.1%) was reported, with 53.3% of patients achieving a CR, 23.3% achieving a PR and 3.3% experiencing stable disease. Comparisons between investigator review and external review are outlined in Supplementary Table 3. To examine whether the very high response rate should be attributed to nivolumab or to the vaccine, we retrieved clinical response information from a matched historical control group using the Danish Metastatic Melanoma Database (DAMMED) from contemporaneously treated patients with stage 3–4 melanoma who received aPD1 monotherapy17. Patients were matched with the exact same combination variable according to age, sex, PD-L1 status, BRAF status, LDH level and M stage (those at stage M1d were excluded from the control group (no patients with brain metastasis)). Matched controls were identified for 29 patients, and the ORR of 79.3% (CI, 61.0–90.4%) observed in MM1636 was found to be significantly higher (P < 0.0012) than that in the matched control group in which an ORR of 41.7% (CI, 31.0–53.3%) was reached. Furthermore, of the 29 patients in MM1636, a significantly (P < 0.0017) higher percentage (41.4% (CI, 25.2–59.6%)) of patients achieved CR in MM1636 than that (12% (CI, 6.3–21.6%)) in the matched historical control group. ORR and complete response rate (CRR) in the matched historical control group were comparable to those of patients treated in randomized phase 3 pivotal trials with aPD1 monotherapy18 (Fig. 1b). The treatment leads to prolonged PFS At data cutoff, the median duration of response had not been reached, with 87% of all responding patients being free from progression at 12 months (Fig. 1d). Patients were followed for up to 35 months with a median follow-up time of 22.9 months (CI, 14.9–26.2 months). OS and PFS were calculated from the first day of treatment to death or progression or to the date of the last follow-up (5 October 2020). The median PFS (mPFS) for all treated patients was 26 months (CI, 15.4–69 months) and was not reached for responding patients (Fig. 1e). The median OS was not reached at the data cutoff. OS at 12 months was 81.6% (CI, 61.6–92%) (Fig. 1f). One patient (MM18) with a CR died from nivolumab-related severe adverse events; the remaining patients died because of metastatic melanoma. For comparison, the mPFS was 8.3 months (CI, 5.5 months–NR (not reached)) in the matched historical control group (n = 74), while the median OS was 23.2 months (CI, 23.2 months–NR). (Extended Data Fig. 2a,b). The combination of the IDO/PD-L1 vaccine and nivolumab was safe Treatment-related adverse events are listed for all 30 patients in Supplementary Table 4. Common treatment-related grade 1–2 toxicities were fatigue (47%), rash (47%), arthralgia (30%), diarrhea (23%), nausea (23%), dry skin (20%), pruritus (20%), infusion reaction (17%), xerostomia (17%) and myalgia (17%). Four patients (13%) experienced grade 3–4 adverse events: one patient with a grade 3 maculopapular rash (MM01), one patient with grade 3 adrenal insufficiency (MM06) and one patient with grade 3 arthralgia (MM22). Patient MM18 died from urosepsis with multi-organ failure and severe hyponatremia. This patient had experienced multiple immune-related adverse events with grade 3 colitis, grade 2 pneumonitis, grade 3 arthralgia, grade 2 vasculitis and grade 2 nivolumab infusion-related allergic reaction. Additionally, patient MM18 had symptoms of myocarditis at the time of death with highly elevated cardiac troponin I levels. Bedside echocardiography showed an ejection fraction of 15%, which at baseline was 60%, but an autopsy was not conducted, and myocarditis was never confirmed pathologically. Patient MM06 had received first-line treatment with ipilimumab before entering the trial and was on substitution corticosteroids at the time of inclusion. Adrenal insufficiency was aggravated by an erysipelas infection with high fever, reaching grade 3 in Common Terminology Criteria for Adverse Events and resolving quickly after appropriate antibiotic therapy was initiated. As expected, local side effects were common with 77% of the patients who developed injection site reactions. These reactions were classified as granulomas (63%), redness (20%), pain (13%) and pruritus (13%) at the injection site. All local reactions were grade 1–2, most likely related to the Montanide adjuvant and typically transient. However, two patients (MM07 and MM20) decided to discontinue vaccination after eight and 11 injections, respectively, due to granulomas, tenderness and pain that limited instrumental activities of daily living but continued to receive nivolumab (Extended Data Fig. 2c). Vaccine-specific responses in blood were frequently detected First, all 30 patients were assessed for the presence of vaccine-specific responses in peripheral blood mononuclear cells (PBMCs) before, on and after vaccination using a modified interferon (IFN)-γ enzyme-linked immune absorbent spot (ELISPOT) assay. This assay is known to expand antigen-specific memory cells and to improve the detection and the correlative power of ELISPOT during treatment19–21. Prevaccine IDO-specific responses were detectable in ten (33%) patients, while prevaccine PD-L1-specific responses were detectable in eight patients (27%); overlapping (specific to both IDO and PD-L1) prevaccine responses were present in four (13.3%) patients. During vaccination, an increase of IDO-specific T cells or PD-L1-specific T cells in the blood was observed in 28 (93%) and 26 (86%) patients, respectively. In total, 93% of patients had an increase in either PD-L1- or IDO-specific responses on vaccination (Fig. 3a), with a significant (P < 0.0001) median increase from baseline to the post-vaccine response demonstrated for both IDO and PD-L1 (at different time points on treatment), confirming that vaccine-specific immune responses were induced in patients regardless of clinical response (Fig. 3b,c). Immune responses fluctuated in the blood over time (Extended Data Fig. 3a). An increase in IDO- and PD-L1-specific responses in the peripheral blood was also detectable directly ex vivo across the different clinical response groups (Extended Data Fig. 4).Fig. 3 Vaccine-specific responses in blood. a, IDO- and PD-L1-specific T cell responses in PBMCs at baseline and on vaccination as measured by the IFN-γ ELISPOT assay (n = 30). *Responses were calculated as the difference between the average numbers of spots in wells stimulated with IDO or PD-L1 peptide (triplicates) and those from the corresponding control (DMSO), and statistical analyses of ELISPOT responses were performed using a distribution-free resampling method (Moodie et al.50). DR (double response), response was not statistically confirmed due to replicate number, but the number of spots in peptide wells was two times higher than that in control wells (DMSO). NS, no significant response and no DR. For a detailed overview of responses at serial time points on vaccination, see Extended Data Fig. 4a. b, IDO- and PD-L1-specific T cell response in PBMCs in all treated patients measured by the IFN-γ ELISPOT assay at baseline and on vaccination. On-vaccination responses were selected from the ‘best’ ELISPOT response at different time points for each patient (series 3, 6, 12, 18 or 24) during vaccination (n = 30). Wilcoxon matched-pairs signed-rank test was used to compare responses to IDO or PD-L1 peptides in the vaccine between baseline and later time points. c, Representative example of ELISPOT wells with response in patient MM23 in serial PBMCs before and on treatment. d, IDO-specific CD4+ and CD8+ T cells were isolated and expanded from PBMCs stimulated in vitro with the IDO peptide and a low dose of IL-2 for 14–15 d before sorting using the Miltenyi Cytokine Secretion Assay—Cell Enrichment and Detection kit. To assess their cytolytic potential, IDO-specific T cells were stimulated with IDO peptide, and expression of CD107a, IFN-γ and TNF-α was assessed by flow cytometry (the example is from patient MM14). Sustained vaccine-specific responses were observed 3 and 6 months after the last vaccine, indicating induction of memory responses in nine patients with response to clinical treatment who surpassed follow-up at data lock (Extended Data Fig. 3b). Importantly, PD-L1- and IDO-specific responses were observed irrespective of HLA genotype (Supplementary Table 5). To verify the functionality of vaccine-induced T cells, IDO- or PD-L1-specific T cells were isolated and in vitro expanded from PBMCs of five patients. Phenotypic characterization by flow cytometry revealed that the isolated vaccine-specific T cells consisted of both CD4+ and CD8+ T cells. Also, both IDO- and PD-L1-specific CD4+ and CD8+ T cells showed pro-inflammatory properties, as they expressed the cytolytic marker CD107a and secreted the cytokines IFN-γ and tumor necrosis factor (TNF)-α (Fig. 3d and Extended Data Fig. 6a–d). Interestingly, we were also able to detect vaccine-specific CD4+ and CD8+ T cell responses in peripheral blood ex vivo (Extended Data Figs. 4 and 5). We observed a significant increase in the overall percentage of CD107a, CD137 and TNF-α expression in response to peptide stimulation in on- or post-treatment PBMC samples compared to that at baseline, further confirming the expansion and the diverse signature of vaccine-specific T cells (Extended Data Fig. 5). Vaccine-specific responses detected in the skin at the vaccination site To investigate whether vaccine-specific T cells have the potential to migrate to peripheral tissue, delayed-type hypersensitivity (DTH) tests were performed after six cycles of treatment on 15 patients to assess the presence of vaccine-reactive T cells in the skin. Supplementary Table 6 display an overview of skin-infiltrating lymphocyte (SKIL) cultures. We detected IDO-specific T cells in the skin of six of ten patients and PD-L1-specific T cells in nine of 11 patients (Extended Data Fig. 7a). Intracellular cytokine staining was performed on SKILs from five patients after stimulation with either PD-L1 or IDO peptide. Here, we detected mainly CD4+ peptide-reactive T cells that secreted TNF-α and upregulated CD107a, and a minor fraction also secreted IFN-γ. In one patient, we detected CD8+ PD-L1-reactive T cells (Extended Data Fig. 7b–d). Vaccine-induced T cells specifically recognize target cells To confirm the functionality of vaccine-expanded T cells, vaccine-specific T cell clones (clonal purity was confirmed by T cell receptor (TCR) sequencing) were isolated and expanded from patient PBMCs (Extended Data Fig. 8e). We showed that PD-L1-specific T cells were able to recognize PD-L1+ autologous tumor cells in a PD-L1 expression-dependent manner if the cancer cells also expressed HLA-II (Fig. 4a,b). Similarly, an HLA-DR-restricted IDO-specific CD4+ T cell clone was able to recognize an HLA-DR-matched IDO-expressing model cell line, MonoMac1, in an IDO-expression-dependent manner (Fig. 4e–g). As previously described, IDO- and PD-L1-specific T cells’ mode of action is not limited to targeting only cancer cells. We were able to show that vaccine-specific T cell clones also reacted against PD-L1- and IDO-expressing autologous immune cells (Fig. 4c,h). To provide myeloid cells with a tumor-associated phenotype we treated isolated CD14+ myeloid cells with tumor conditioned medium (TCM) derived from an established autologous tumor cell line. We observed that such TCM-treated CD14+ cells had increased expression of PD-L1 and IDO and were effectively recognized by autologous PD-L1- and IDO-specific CD4+ T cell clones (Fig. 4c,d,h,i).Fig. 4 PD-L1- and IDO-specific T cells from vaccinated patients react against PD-L1- and IDO-expressing target cells. a, Left, PD-L1-specific T cell culture (MM1636.05) reactivity against PD-L1 peptide or autologous tumor cells in the IFN-γ ELISPOT assay. Tumor cells were either not treated or pretreated with 200 U ml−1 IFN-γ for 48 h before the assay. Effector:target (E:T) ratio of 10:1 was used. Right, PD-L1 and HLA-II surface expression on melanoma cells with (green) or without (yellow) pretreatment with IFN-γ compared to an isotype control (gray) as assessed by flow cytometry. b, Left, PD-L1-specific T cell (MM1636.05) reactivity in the IFN-γ ELISPOT assay against autologous tumor cells pretreated with IFN-γ (500 U ml−1) and transfected with mock or PD-L1 small interfering (si)RNA 24 h after transfection. E:T ratio, 10:1. Right, PD-L1 surface expression on melanoma tumor cells (MM1636.05) assessed by flow cytometry 24 h after transfection with mock (blue) or PD-L1 (red) siRNA compared to the isotype control (gray). c, Reactivity of the CD4+ PD-L1-specific T cell clone (MM1636.14) against PD-L1 peptide or autologous CD14+ cells; E:T ratio, 10:1. CD14+ cells were isolated using magnetic bead sorting and used as targets in an ELISPOT assay directly or after pretreatment for 2 d with TCM derived from the autologous tumor cell line. d, Quantitative PCR with reverse transcription (RT–qPCR) analysis of PD-L1 (CD274) expression in sorted CD14+ cells before and after treatment with autologous TCM for 48 h. e, Reactivity of the IDO-specific CD4+ T cell clone (MM1636.23) against IDO peptide combined with HLA-DR (L243)-, HLA-DQ (SPV-L3)- or HLA-DP (B7/21)-blocking antibodies (aHLA-DR, aHLA-DQ and aHLA-DP) in an intracellular staining assay (ICS) for IFN-γ and TNF-α production. T cells were incubated with individual blocking antibodies (2 μg ml−1) for 30 min before adding IDO peptide. f, Reactivity of the IDO-specific CD4+ T cell clone (MM1636.23) against the HLA-DR-matched IDO-expressing cell line MonoMac1 transfected with mock or IDO siRNA in an ICS assay, E.T ratio, 4:1. siRNA transfection was performed 48 h before the experiment. g, RT–qPCR analysis of IDO1 expression in MonoMac1 cells 48 h after siRNA transfection. h, Reactivity of the CD4+ IDO-specific T cell clone (MM1636.14) against IDO peptide or autologous CD14+ cells; E:T ratio, 20:1. CD14+ cells were isolated using magnetic bead sorting and used as targets in an ELISPOT assay directly or after pretreatment with TCM derived from the autologous tumor cell line. i, RT–qPCR analysis of IDO1 expression in sorted CD14+ cells before and after treatment with autologous TCM for 48 h. Bars in RT–qPCR data (d,g,i) represent the mean of three (d,i) or six (g) technical replicates ±s.d.; P values were determined by two-tailed parametric t-tests. ELISPOT counts (a,b,c,h) represent the mean value of three technical replicates ±s.e.m.; response P values were determined using the distribution-free resampling (DFR) method. TNTC, too numerous to count. T cell clones in blood and tumors To track the role of treatment-induced T cell responses, TCR sequencing of the complementarity-determining region 3 (CDR3) was performed on five patients in peripheral blood (baseline and cycles 3, 6 and 12) and paired biopsies. These five patients (MM01, MM02, MM08, MM09 and MM13) were selected due to the availability of material and to investigate a balanced patient group with both responders and non-responders. Due to a limited number of available paired biopsies (either as a consequence of patient refusal or unexpectedly rapid and substantial clinical responses), statistical analysis could not be applied. Details on clinical response are shown in Fig. 2a. Additionally, PBMCs (on treatment) or SKILs were stimulated with the IDO/PD-L1 peptides, and then cytokine-producing T cells were sorted to track vaccine-induced T cells both in the periphery and at the tumor site. To identify enriched IDO/PD-L1-specific T cell clones, TCR rearrangements in sorted IDO/PD-L1-specific clones and TCR rearrangements in sorted IDO/PD-L1-specific T cell samples were compared to sequences from baseline PBMC samples for each patient. Clonal expansion of vaccine-specific TCR rearrangements from samples on vaccination were then tracked using a differential abundance framework. Cumulative IDO/PD-L1-specific T cell frequencies were tracked in post-treatment samples. We found no relation between clinical response and the enrichment of vaccine-specific clones, but an increase in IDO/PD-L1-specific T cell clones was observed at different time points in the periphery in all five patients (Extended Data Fig. 3c). We next investigated overall changes in the T cell repertoire in the blood. A modest increase in the peripheral T cell fraction was observed in the three responding patients at cycle 3, while the two non-responding patients had a clear decrease in T cell fraction (Extended Data Fig. 8a). We thereafter investigated TCR clonality and TCR repertoire richness, exploring the proportion of abundant clones and the number of unique rearrangements, respectively. A decreasing peripheral Simpson clonality and increasing TCR repertoire richness was observed in responding patients at cycle 3, which might indicate tumor trafficking upon treatment. The opposite pattern was observed in non-responding patients (Extended Data Fig. 8b,c). Peripherally expanded clones were associated with tumors and persisted until cycle 12 (latest time point analyzed). The largest peripheral expansion was observed at cycle 3, with the most significant increase observed in patient MM01 (CR). Responding patients had a larger fraction of peripherally expanded clones that were also found in tumors compared to that of non-responders. By tracking peripherally expanded clones detected at the tumor site, we observed that patient MM01 had a substantial increase after treatment, indicating tumor trafficking of peripheral expanded clones (Extended Data Fig. 8d). Influx at the tumor site of vaccine-enriched T cell clones Given the observation of increased T cell fraction and enrichment of IDO- and PD-L1-specific clones in the blood after treatment, we investigated whether the same trend was observed at the tumor site. Both TCR sequencing and immunohistochemistry (IHC) of paired biopsies from the five patients described above showed an increase in the T cell fraction with an influx of CD3+ and CD8+ T cells after treatment in the three responding patients (Fig. 5a–c). IHC was not possible for patient MM09 due to tissue loss.Fig. 5 Changes in the TME after treatment. Number of CD3+ and CD8+ T cells, TCR fraction, TCR clonality, TCR repertoire, biopsy expanded TCR clones and enrichment of IDO- and PD-L1-specific T cells at the tumor site. a, Number of CD3+ and CD8+ T cells per mm2 (sum in the validated area) at the tumor site detected by IHC of paired biopsies from four patients. b, Example of IHC of CD3+ and CD8+ T cells at the tumor site before and after treatment (cells per mm2) in one patient (MM01). c, T cell fraction at the tumor site at baseline and cycle 6 by TCR sequencing. The T cell fraction was calculated by taking the total number of T cell templates and dividing by the total number of nucleated cells. d, Tracking of vaccine-associated clones at baseline and cycle 6 in tumor biopsies. Cumulative frequencies of IDO and PD-L1 vaccine-specific TCR rearrangements are represented. e,f, TCR clonality and TCR repertoire richness in five patients at the tumor site at baseline and cycle 6. Simpson clonality measures how evenly TCR sequences are distributed among a set of T cells, where 0 indicates an even distribution of frequencies and 1 indicates an asymmetric distribution in which a few clones dominate. TCR repertoire richness reports the mean number of unique rearrangements. g, Bar chart representing baseline expanded biopsy clones from five patients (colored bars) and the detection of biopsy expanded clones also found in the blood at baseline and series 3, 6 and 12 (white and gray bars) by TCR sequencing. We thereafter investigated whether some of the IDO/PD-L1 vaccine-linked T cells were present at the tumor site. Vaccine-associated clones were tracked as the combined frequency of IDO- and PD-L1-specific T cell rearrangements. In biopsies, the frequencies at cycle 6 were compared to those at baseline and showed an increase in vaccine-specific T cells in four of five patients, irrespective of clinical response (Fig. 5d). TCR sequencing of PD-L1-specific SKILs (a more specific culture than IDO/PD-L1-specific isolations derived from PBMCs on vaccination) and paired biopsies showed that two of the top five PD-L1-specific SKIL clones were present at the tumor site both before and after treatment (Extended Data Fig. 7e,f). With a focus on the more abundant T cell clones, we investigated overall TCR clonality at the tumor site before and after treatment. In addition, we explored the number of unique TCR rearrangements, dissecting the lower-frequency clones. Patient MM01 had a significant increase in TCR clonality and a decrease in repertoire richness at the tumor site after therapy, indicating a focused tumor repertoire response of selected clones. All three responding patients had a decrease in TCR repertoire richness, which again might indicate a focused tumor response (Fig. 5e,f). Deeper analyses showed that the T cell clones that expanded at the tumor site after therapy were also present in the blood at baseline and increased significantly after treatment in four of five patients. The highest proportion was detected early at cycle 3. These data again support trafficking of peripherally expanded clones to the tumor site and could indicate that the T cell response to treatment is derived from pre-existing peripherally tumor-associated T cells (Fig. 5g). Signs of treatment-induced inflammation in the TME To dissect changes in the TME induced by T cell influx upon treatment in responding patients, RNA gene expression analyses using the nCounter PanCancer Immune Profiling Panel from NanoString were performed on paired biopsies from two responding patients (MM01 and MM13). Expression of genes related to adaptive immunity such as T cell activation, effector functions (genes encoding IFN-γ, TNF-α, IL-15 and IL-18) and cytotoxicity was increased in post-treatment biopsies (Extended Data Fig. 9a,b). Also, expression of genes related to checkpoint inhibitors such as those encoding T cell immunoglobulin and mucin-domain containing-3 (TIM-3), IDO, PD-L1, PD-L2, PD-1 and CTLA-4 increased after treatment, indicating activation of immune cells in the TME (Extended Data Fig. 9c). Additionally, IHC of paired biopsies from four patients (MM01, MM02, MM05 and MM13) showed an upregulation of PD-L1, IDO, MHC-I and MHC-II on tumor cells, indicating a treatment-induced pro-inflammatory response in the three responding patients, except for a decrease in MHC-II expression in patient MM13. By contrast, the non-responding patient MM02 had a reduction in T cell numbers present in the tumor after treatment and no expression of PD-L1, IDO or MHC-II and, interestingly, total loss of MHC-I, demonstrating tumor immune escape (Extended Data Fig. 10a). CD8+ T cells and their distance (µm) to PD-L1-expressing cells in baseline biopsies from five patients was investigated by IHC. Except for patient MM13 (PR), distance and clinical responses were associated: the two responders had reduced distance (<20 µm) between cells expressing these markers compared to non-responding patients (>80 µm). This observation indicates that responding patients not only have a higher intratumoral infiltration of CD8+ T cells, but that these cells can surround and attack PD-L1-expressing immune cells and tumor cells (Extended Data Fig. 10b). Discussion In this clinical trial, MM1636, 30 patients with metastatic melanoma were treated with a first-in-class immunomodulatory IDO/PD-L1-targeting peptide vaccine combined with nivolumab. The treatment led to an unprecedented high ORR of 80%, with 43% of patients reaching a CR, and a striking mPFS of 26 months (95% CI, 15.4–69 months) was reached. The vaccine represents a new treatment strategy to activate specific T cells that target cells contributing to immune suppression (including tumor cells), positively modulating the TME by inducing local inflammation. Indeed, we show that vaccine-specific T cells isolated and expanded from vaccinated patients recognize not only tumor cells in a target- and HLA-restricted manner but also myeloid cells polarized toward a tumor-associated phenotype. Hence, myeloid cells become targets for vaccine-activated T cells when they have a tumor-associated phenotype in the TME. These phenomena may further induce checkpoint molecules and rewire the TME toward an increasingly aPD1-permissive state. A drawback of our study is the single-center nonrandomized setup. Comparison between trials or between patients in trials and real-world patients is problematic due to multiple factors, such as period of conduction and other therapies available at the different periods of time. Nevertheless, the rate of investigator-assessed ORR in the phase 3 trial CheckMate 067 was 43.7% in the nivolumab monotherapy group and 57% in the nivolumab and ipilimumab group. CRs occurred in 8.9% and 11.5% of patients, respectively18. The mPFS of 26 months (95% CI, 15.4–69 months) in this trial is more than twice as long as that for patients treated with nivolumab and ipilimumab in CheckMate 067, for which an mPFS of 11.5 months (95% CI, 8.7–19.3 months) was reached. Patient baseline characteristics were in general comparable to those of patients with MM who have been treated in CheckMate 067, although patients in MM1636 were older (mean age, 70 years) and a larger fraction were positive for PD-L1 (57%)3,18,22. Among patients with PD-L1-negative tumors in MM1636, an ORR of 61.5% was still reached, which would be expected to be around 33% for first-line nivolumab monotherapy23. Some studies suggest that older patients might have a tendency to respond better to aPD1 therapy; however, this is still debated24–27. To address potential trial bias and the nonrandomized setup, patients in MM1636 were matched for age, performance status, sex, M stage, LDH level, PD-L1 status and BRAF status with a historical control group from the DAMMED, who were treated contemporarily (2015–2019) with aPD1 monotherapy as standard of care17. We found a significantly higher ORR and CRR in MM1636 compared to those of matched patients, who had an ORR of 43% and a CRR of 13%, comparable to patients treated in CheckMate 067. Restrictions of the synthetic control group are of course that it is partially historic and patient selection outside matching criteria cannot be ruled out28. Numerous contemporary clinical trials are exploring the combination of aPD1 therapy with other immunomodulating agents for advanced melanoma. Talimogene laherparepvec, an oncolytic virus, is approved by the Food and Drug Administration and the European Medicines Agency to treat advanced melanoma. A small phase 1b trial with 21 patients (MASTERKEY-265) combined talimogene laherparepvec and pembrolizumab to treat patients with advanced unresectable melanoma and reached an ORR of 62% and a CR of 33%29,30. Seventy-one percent of the patients in this trial had an M stage below M1c; this number was 40% in our trial. Furthermore, mainly patients with an M stage below M1c responded to treatment, which was not the case in MM1636. Results from a large randomized phase 3 trial are awaited (KEYNOTE-034). Epacadostat, an IDO inhibitor, was tested in combination with pembrolizumab in a nonrandomized phase 2 trial in 40 aPD1 treatment-naive patients with MM with promising results, reaching an ORR of 62%. Unfortunately, the phase 3 trial showed no indication that epacadostat provided improvement in PFS and OS31. Limitations of the phase 3 trial were the sparse information on pharmacodynamics as well as biomarker evaluation to improve the design. The IDO/PD-L1 vaccine is different from epacadostat, as it is not an IDO inhibitor but targets IDO- and PD-L1-expressing cells. Similar vaccines administered as monotherapy induced objective responses in lung cancer and basal cell carcinoma, while epacadostat as monotherapy in 52 patients resulted in no responses32,33. Sahin et al. recently published encouraging data from a first-in-human trial, in which a vaccine containing liposomal RNA targeting four unmutated tumor-associated antigens (NY-ESO-1, MAGE-A3, tyrosinase and TPTE) was administered alone or in combination with nivolumab in patients with advanced melanoma. Responses were observed in both the monotherapy group as well as the combination group in checkpoint inhibitor-experienced patients, suggesting the efficacy of non-mutant shared tumor antigen vaccines34. The overall safety and tolerability findings are comparable to those of aPD1 monotherapy. Injection site reactions were exclusive to the vaccine. However, these side effects were transient and mild in most patients and most likely due to the adjuvant Montanide. IDO- and PD-L1-specific CD8+ and CD4+ T cells exist among peripheral blood lymphocytes in healthy donors35–38 and expand in response to pro-inflammatory stimuli35,39. Furthermore, both IDO and PD-L1 are induced in cells as a counter response to the inflammatory response. This provides a mechanism that ensures immune homeostasis, which keeps IDO/PD-L1-specific T cells in check; therefore, the risk of triggering autoimmune-related adverse events by vaccination appears to be minimal. This was confirmed in the first clinical trials of IDO and PD-L1 vaccination (NCT01543464 (refs. 33,40) and NCT03042793 (ref. 41)). The spontaneous (baseline) immune response to the vaccines observed in the current study is in agreement with our previous observation in various patients with cancer35–38. Numerous vaccine-induced changes in the blood and at the tumor site were observed. Peripheral IDO- and/or PD-L1-specific T cells were detected in vitro in a modified ELISPOT in over 93% of patients on vaccination, unrelated to patient HLA type. Immune responses were persistent in patients who surpassed follow-up at data cutoff and were still detectable up to 6 months after the last vaccine, suggesting induction of memory T cells. We did not observe a correlation between vaccine-induced responses in blood and clinical responses. However, the detection of a highly significant increase in vaccine-specific T cell numbers after vaccination in almost all patients together with the very low number of patients with PD makes it difficult to expect such a correlation, especially because other aspects should be taken into account (for example, the loss of class I expression on tumors cells in patient MM02 with PD after vaccination). Frequencies of peripheral T cells induced against PD-L1 were overall higher than those against IDO. Importantly, however, we also observed that it was not the same patients who responded strongly to both IDO and PD-L1, nor did patients react with a similar response pattern to the two antigens; that is, different time points of response were observed. This suggests that each component of the vaccine plays different roles in the ongoing immune response in patients. The goal of IDO/PD-L1 vaccination was to modulate the TME to increase responsiveness to aPD1 therapy, as we have observed in animal models, in which immune conversion is demonstrated in the TME with an increased influx of T cells. Targeting both IDO and PD-L1 together enables synergy, as the TME is known to use different immune escape mechanisms and IDO and PD-L1 are often overexpressed by different cellular compartments. Ex vivo ELISPOT and flow cytometry assays confirmed the induction of immune response toward both epitopes and the higher immunogenicity of the PD-L1 epitope, although because of lower sensitivity failed to detect some of the responses. TCR sequencing of five patients confirmed enrichment of IDO/PD-L1-specific T cell clones in the blood at different time points after treatment. Furthermore, an increase in enriched IDO/PD-L1-specific clones was observed in four of five patients at the tumor site after treatment, irrespective of clinical response. Phenotypic characterization showed that vaccine-specific T cells, which were expanded in vitro with interleukin (IL)-2 from the blood of vaccinated individuals, were both CD4+ T cells and CD8+ T cells. This was confirmed by ex vivo phenotype description of vaccine-activated T cells. Vaccine-specific T cells expressed CD107a and CD137 and produced IFN-γ and TNF-α upon stimulation with the cognate target, indicating their cytolytic capacity. Overall, the data from immune monitoring supports the importance of both antigens in the generation of the frequent clinical responses in the study. Despite a limited number of paired biopsies (due to either patient refusal or the fact that a large fraction of responding patients had no assessable tumors after six cycles), we observed trends indicating treatment-induced general T cell influx in responding patients. It was shown that proliferation of CD8+ T cells in the tumor after aPD1 treatment is associated with radiographic reduction in tumor size42. Additionally, we showed that a large proportion of expanded peripheral TCR clones were associated with tumors and the most considerable amount of clonal expansion was observed early, at cycle 3. For the patient with a CR included in the TCR sequencing analyses (MM01), the number of peripheral expanded clones present at the tumor site increased after treatment compared to that at baseline, indicating tumor trafficking of peripheral expanded clones. Gene expression analyses (two paired biopsies) and IHC (five paired biopsies) further demonstrated that the combination treatment induced a pro-inflammatory TME in responding patients with signs of T cell activation and cytotoxicity and increased cytokine activity. This may lead to further upregulation of IDO, PD-L1, MHC-I and MHC-II on tumor cells, leading to more treatment targets. It was shown that, following vaccination with a cancer vaccine, PD-L1 expression is increased on tumor cells due to recruitment of tumor-specific T cells and upregulation of adaptive immune resistance pathways in the TME43. Treatment with nivolumab monotherapy enhances PD-L1 expression, and it is therefore problematic to discriminate the effect of the vaccine as compared to that of nivolumab44. IDO- and PD-L1-specific pro-inflammatory effector T cells were hypothesized to counteract the functions of IDO- and PD-L1-expressing immune-suppressive cells as a means to keep the immune balance between immune activation and inhibition45. However, because the expression of these molecules is induced as a counter-regulatory response to inflammatory mediators such as IFNs, they can also be expressed by, for example, activated myeloid cells or T cells. Thus, activation of PD-L1-specific T cells may result in depletion of activated T cells in the tumor and other sites. Importantly, it was recently described that PD-L1+ T cells mainly have tolerogenic effects on tumor immunity and exert tumor-promoting properties, suggesting that targeting this immune population is indeed also beneficial46. We have previously investigated the effect of activating PD-L1-specific T cells in vitro and in vivo and found that, overall, they supported the effector phase of an immune response by removing PD-L1-expressing regulatory immune cells that inhibit PD-1+ effector T cells47,48. The major role of the PD-1 pathway is believed to be the regulation of effector T cell responses. Thus, this protective pathway is more important after activation, rather than at the initial T cell-activation stage49. Accordingly, the presence of PD-L1-specific T cells during the activation phase of an immune response may not increase or support a pro-inflammatory immune response due to the expression of PD-L1 on potent antigen-presenting cells or on the T cells themselves14. Thus, the overall effects of PD-L1-specific T cells may vary depending on the expression of both PD-1 and PD-L1, that is, due to the microenvironment and the state of the immune response. In conclusion, here we report an impressive response rate, CR rate and mPFS for a first-in-class immune-modulating vaccine combined with nivolumab. This may be a first step toward a new treatment strategy for patients with MM. Limitations are the low number of patients treated at a single institution and the lack of a randomized design with aPD1 monotherapy as comparator. Studies in aPD1 therapy-resistant or -refractory melanoma are ongoing as well as biomarker analysis for selecting patients at a higher likelihood to benefit from the combination versus aPD1 monotherapy. A larger randomized trial will be essential to validate these findings and determine the specific contribution of the vaccine to clinical responses and changes in the TME. In December 2020, the Food and Drug Administration granted breakthrough therapy designation for the IO102/IO103 vaccine combined with aPD1 therapy in metastatic melanoma based on data from the MM1636 trial. Methods Trial design and treatment plan MM1636 is an investigator-initiated, nonrandomized, open-label, single-center phase 1/2 study. All patients were treated at the Department of Oncology, Herlev and Gentofte Hospital, University of Copenhagen, Herlev, Denmark. The study was conducted according to the Declaration of Helsinki and Good Clinical Practice (GCP) and monitored by the GCP unit in Copenhagen, Denmark. The protocol was approved by the ethical committee of the Capital region of Denmark (H-17000988), the Danish Medical Agencies (2017011073) and the Capital Region of Denmark Data Unit (P-2019-172). The study was registered at https://clinicaltrials.gov/ under the identifier NCT03047928 and at the EudraCT (no. 2016-0004527-23). This study initially aimed to include 30 aPD1 treatment-naive patients with MM. An amendment with the addition of two other cohorts with ten patients in each cohort was made to evaluate immune responses and clinical efficacy in aPD1 therapy-resistant patients (cohort B, de novo resistance and cohort C, acquired resistance) for a total of 50 patients. The amendment with cohorts B and C was approved at the inclusion of 18 patients in cohort A. The trial is still including patients in cohorts B and C. This article reports results from cohort A. The first six patients were treated as for phase 1, evaluating for safety and tolerability before the remaining 24 patients were included in phase 2. The IDO/PD-L1 vaccine was administered subcutaneously biweekly for the first 10 weeks and thereafter every 4 weeks for approximately 9 months. A maximum of 15 vaccines were administered. Nivolumab was administered according to the approved label (3 mg per kg biweekly) for 24 cycles. The 15th vaccine was administered together with the 24th nivolumab cycle of 3 mg per kg, and responding patients thereafter continued nivolumab monotherapy every 4th week (6 mg per kg) as a standard of care after investigator assessment. Treatment with nivolumab was discontinued at the maximum benefit (assessed by the investigator), after a maximum of 2 years of therapy, at progression or due to severe adverse events (Extended Data Fig. 1b, treatment plan). Vaccine composition Each vaccine was composed of 100 µg IO102, a 21-amino-acid peptide (DTLLKALLEIASCLEKALQVF) from IDO, and 100 µg IO103, a 19-amino-acid peptide (FMTYWHLLNAFTVTVPKDL) from the signal peptide of PD-L1 (PolyPeptide Laboratories). Peptides were dissolved separately in 50 µl DMSO, filtered for sterility and frozen at −20 °C (NUNC CyroTubes CryoLine System Internal Thread, Sigma-Aldrich). At <24 h before administration, peptides were thawed. The PD-L1 peptide was diluted in 400 µl sterile water and, immediately before injection, mixed with the IDO peptide solution and 500 µl Montanide ISA-51 (SEPPIC) to achieve a total volume of 1 ml. Patients Patients above 18 years of age with locally advanced or stage 4 melanoma according to the AJCC (seventh edition), at least one measurable lesion according to RECIST 1.1 and an ECOG PS of 0–1 were eligible. The main exclusion criteria were prior treatment with aPD1 therapy, CNS metastases >1 cm, severe comorbidities and active autoimmune disease. Enrollment was not restricted to PD-L1 status, but it was known before inclusion. Patients were included after informed consent. Patients MM42 and MM20 have confirmed their approval of PET–CT images and clinical images that have been included in the article. Key study assessments Safety and tolerability were evaluated based on changes in clinical laboratory analyses and reported adverse events. Adverse events were assessed according to Common Terminology Criteria for Adverse Events (version 5.0) and were graded from 1 to 5 for all treated patients up to 6 months after the last dose of the IDO/PD-L1 vaccine. Clinical efficacy was assessed using FDG PET–CT scans before treatment and every 3rd month until progression. Objective responses were categorized into CR, PR, stable disease or PD according to RECIST version 1.1. Clinical data were collected at CCIT-DK in the eCRF program OpenClinica version 1.0 and in Microsoft Excel version 2002 on a secure server. Blood samples for immunologic analyses were collected before treatment, before the third cycle, after the sixth, 12th, 18th and 24th cycles (on vaccination) and 3 and 6 months after the last vaccine. Two to three tumor needle biopsies (1.2 mm) were collected at baseline and after six cycles from the same tumor site, when accessible, to evaluate immune responses at the tumor site. DTH skin tests were performed and punch biopsies were taken from the DTH area after cycle 6 for the evaluation of SKILs reactive to PD-L1 and IDO (Supplementary Fig. 1, treatment plan). Statistical analysis of clinical outcome Survival curves were computed in GraphPad Prism software version 9.0.0 according to Kaplan–Meier method. Median follow-up time of enrollment was calculated using the reverse Kaplan–Meier method, also in GraphPad Prism software version 9.0.0. For binary outcomes, 95% two-sided CIs were constructed using the Clopper–Pearson method, also in GraphPad version 9.0.0. An independent board of certified and experienced oncoradiologists performed an external review to evaluate clinical response to address the potential bias of investigator site review. The external review took place at Rigshospitalet, Copenhagen University Hospital. This hospital did not participate in the MM1636 trial, and the external reviewer had no prior knowledge about the clinical trial or the trial therapy. Only PET–CT images were accessed, and arrows indicating target or non-target lesions appeared on baseline images as the only additional information. To address potential trial bias regarding treatment effect, we matched patients in MM1636 with patients from the DAMMED, a population-based database that retrospectively collects data on patients with metastatic melanoma in Denmark. Here, data from 938 patients treated with aPD1 monotherapy contemporaneously (January 2015–October 2019) were extracted. Two hundred and eighteen of these patients were eligible for comparison and matching (all parameters available) (Supplementary Table 1), and 74 patients from the DAMMED were found to match. Patients were matched for age (≤70 years, >70 years), sex, LDH levels (normal, elevated), M stage (M1a, M1b, M1c), BRAF status (wild type, mutated) and PD-L1 status (<1%, ≥1%). An exact matching algorithm was used in which patients in MM1636 were matched with patients from the DAMMED with the same combination of variables. Twenty-nine patients from MM1636 were matched with exact combinations of the six variables. One patient could not be matched. To secure balance of the calculations, control patients were weighted according to the number of patients for each MM1636 patient. Estimates for treatment effects were calculated by weighted logistic regression analyses and weighted Cox proportional hazard model. The R package ‘Matchlt’ was used for matching patients. As the method chosen for matching control patients to protocol patients, a weighted binary logistic regression model was used for comparing response rates in the two matched cohorts. Odds ratios and response rates, including their corresponding 95% CIs, were extracted from the regression models. All P values were two sided, and P values below 0.05 were considered statistically significant. SAS version 9.4M5 was used for the weighted logistic regression models. Processing of peripheral blood mononuclear cells Peripheral blood was collected from all patients in heparinized tubes and was processed within 4 h. In brief, PBMCs were isolated using Lymphoprep (Medinor) separation. PBMCs were counted on a Sysmex XP-300 analyzer and frozen in Human AB Serum (Sigma-Aldrich, H4522-100ML) with 10% DMSO using controlled-rate freezing (CoolCell, BioCision) in a −80 °C freezer and were moved the next day to a freezer at −140 °C until further processing. Needle biopsies at baseline and after the sixth vaccine Two to three needle biopsies (1.2 mm) were taken at baseline and after the sixth cycle of treatment when assessable from the same tumor lesion. One fragment was fixed with formalin and embedded in paraffin (FFPE); one to two fragments were used for expanding tumor-infiltrating T cells and establishing autologous tumor cell lines. Delayed-type hypersensitivity and generation of skin-infiltrating lymphocytes After six cycles of treatment, we performed intradermal injections of vaccine components without adjuvant and one control injection containing DMSO without peptide. Patients were injected with either a mixture of both IDO and PD-L1 peptides at all three injection sites or PD-L1 peptide, IDO peptide or a mixture of the two at the injection sites, respectively (Supplementary Table 6). Eight hours after injection, punch biopsies were resected from the three sites where the peptide was injected and transported immediately to the laboratory and cut into fragments of 1–2 mm3. SKILs were expanded to establish ‘young SKILs’ in CM medium consisting of RPMI 1640 with GlutaMAX, 25 mM HEPES, pH 7.2 (Gibco, 72400-021), IL-2 (100 or 6,000 IU ml−1) (Proleukin Novartis, 004184), 10% heat-inactivated Human AB Serum (Sigma-Aldrich, H4522-100ML), 100 U ml−1 penicillin, 1.25 µg ml−1 Fungizone (Bristol Myers Squibb, 49182) and 100 µg ml−1 streptomycin (Gibco, 15140-122). Half of the medium was replaced three times per week. Young SKILs from samples derived from IDO peptide, PD-L1 peptide and the mixture injection sites were further expanded in a small-scale version of the 14-d rapid expansion protocol as previously described51. Quantification of specific T cells in blood by ELISPOT For enumeration of vaccine-specific T cells in the peripheral blood, PBMCs from patients were stimulated with IDO or PD-L1 peptide in the presence of a low dose of IL-2 (120 U ml−1) for 7–13 d before being used in IFN-γ ELISPOT assays with 2.5–3.2 × 105 cells per well. Briefly, cells were placed in a 96-well PVDF membrane-bottomed ELISPOT plate (MultiScreen MSIPN4W50, Millipore) precoated with IFN-γ-capture antibody (clone 1-D1K, Mabtech). Diluted IDO or PD-L1 peptide stocks in DMSO were added at 5 µM; an equivalent amount of DMSO was added to control wells. PBMCs from each patient were set up in duplicate or triplicate for peptide and control stimulations. Cells were incubated in ELISPOT plates in the presence of the peptide for 16–18 h, after which they were washed off, and biotinylated secondary antibody (anti-human IFN-γ mAB, clone 7-B6-1, Mabtech) was added. After 2 h of incubation, unbound secondary antibody was washed off, and streptavidin-conjugated alkaline phosphatase (Mabtech) was added for 1 h. Next, unbound streptavidin-conjugated enzyme was washed off, and the assay was developed by adding BCIP/NBT substrate (Mabtech). ELISPOT plates were analyzed on the CTL ImmunoSpot S6 Ultimate V analyzer using ImmunoSpot software version 5.1. Responses were calculated as the difference between the average numbers of spots in wells stimulated with IDO or PD-L1 peptide and those from corresponding control wells. For detection of IDO and PD-L1 peptide responses ex vivo, PBMCs were thawed and rested overnight in medium containing DNase I (1 μg ml−1, Sigma-Aldrich, 11284932001) before being used in IFN-γ ELISPOT assays as described above. A total of 6–9 × 105 PBMCs per well were seeded. Statistical analysis of all ELISPOT responses was performed using the DFR method as described by Moodie et al. using RStudio software (RStudio Team, 2016, http://www.rstudio.com/)50. Vaccine-specific ELISPOT responses were defined as true if the difference between the spot count in control and peptide-stimulated wells was statistically significant according to DFR statistical analysis or, for samples performed in duplicate, if the spot count in peptide-stimulated wells was at least 2× the spot count in control wells50. Cancer cell lines and tumor conditioned medium Autologous melanoma cell lines were established from needle biopsies. Briefly, biopsies were chopped into small fragments and seeded in 24-well cultures in RPMI 1640 with GlutaMAX, 25 mM HEPES, pH 7.2 (Gibco, 72400-021), 10% heat-inactivated FBS (Life Technologies, 10500064), 100 U ml−1 penicillin, 1.25 µg ml−1 Fungizone (Bristol Myers Squibb, 49182) and 100 µg ml−1 streptomycin (Gibco, 15140-122). Established adherent melanoma tumor cell lines were cryopreserved at −140 °C in freezing medium containing FBS with 10% DMSO. PD-L1 and HLA-II expression on established tumor cell lines was assessed by flow cytometry staining with anti-PD-L1–PE-Cy7 (1:23 dilution, clone M1h1 (RUO), BD, 558017) and anti-HLA-II–FITC (1:23 dilution, clone Tu39 (RUO), BD, 555558) antibodies. To obtain TCM, established tumor cell lines were cultured in 175-cm2 Nunc cell culture flasks until 80–90% confluency was reached. The culture medium was then replaced with 20 ml fresh X-VIVO 15 with Gentamicin and Phenol Red (Lonza, BE02-060Q), medium with 5% heat-inactivated Human AB Serum (Sigma-Aldrich, H4522-100ML). After 24 h of incubation, TCM was collected and centrifuged to remove any resuspended cells, after which TCM was aliquoted, frozen and stored at −80 °C. The acute monocytic leukemia cell line MonoMac1 was obtained from the DSMZ—German Collection of Microorganisms and Cell Cultures (ACC 252) and cultured in RPMI 1640 with GlutaMAX, 25 mM HEPES, pH 7.2 (Gibco, 72400-021) and 10% heat-inactivated FBS. Isolation of autologous myeloid cells Autologous CD14+ cells were sorted from freshly thawed PBMCs using a magnetic bead-separation kit (Miltenyi Biotec, 130-050-201) according to the manufacturer’s instructions. Isolated CD14+ cells were used as targets in the IFN-γ ELISPOT assay directly after sorting or differentiated in vitro into tumor-associated macrophages by culturing with 1 ml fresh X-VIVO 15 medium with Gentamicin and Phenol Red (Lonza, BE02-060Q) and 5% heat-inactivated Human AB Serum, supplemented with 1 ml autologous TCM in 24-well plates for 2 d. Quantification of vaccine-specific T cells from DTH biopsy sites by ELISPOT SKILs expanded in vitro were rested in medium without IL-2 overnight before being used in the IFN-γ ELISPOT assay as described above to evaluate reactivity of skin-infiltrating T cells. Generation of IDO- and PD-L1-specific T cell cultures from PBMCs or SKILs IDO- or PD-L1-specific T cells were isolated from peptide-stimulated in vitro PBMC cultures on days 14–15 after stimulation or from SKIL cultures expanded in vitro. For specific T cell isolation, PBMCs or SKILs were stimulated with IDO or PD-L1 peptide, and cytokine-producing T cells were sorted using the IFN-γ or TNF-α Secretion Assay—Cell Enrichment and Detection kit (Miltenyi Biotec). Cytokine-production profile of PD-L1- and IDO-specific T cells by intracellular staining To assess the T cell cytokine-production profile, isolated and expanded IDO- and PD-L1-specific T cell cultures were stimulated for 5 h with peptide at 5 μM in a 96-well plate. One hour after the start of the incubation, anti-CD107a–PE (1:133 dilution, clone H4A3, BD Biosciences, 555801) antibody and BD GolgiPlug (1:1,000 dilution, BD Biosciences) were added. After a 5-h incubation, cells were stained using fluorescently labeled surface marker antibodies: anti-CD4–PerCP (1:5 dilution, clone SK3, 345770), anti-CD8–FITC (1:5 dilution, clone SK1, 345772) and anti-CD3–APC-H7 (1:20 dilution, clone SK7, 560275) (all from BD Biosciences). Dead cells were stained using FVS510 (1:250 dilution, BD Biosciences, 564406), followed by overnight fixation and permeabilization using eBioscience Fixation/Permeabilization buffers (00-5123-43, 00-5223-56) according to the manufacturer’s instructions. Cells were then stained intracellularly in eBioscience permeabilization buffer (00-8333-56) with anti-IFN-γ–APC (1:20 dilution, clone 25723.11, BD Biosciences, 341117) and anti-TNF-α–BV421 (1:100 dilution, clone Mab11, BD Biosciences, 562783) antibodies. Samples were analyzed on the FACSCanto II (BD Biosciences) using BD FACSDiva software version 8.0.2. The gating strategy is shown in Supplementary Fig. 3a. For assessing the reactivity against IDO expressing MonoMac1 cells, the appropriate amount of cancer cells was added to the IDO-specific T cells to obtain an effector:target ratio of 4:1 in the 96-well plate and T cell cytokine production was tested after 5 h of co-culture as described above. For HLA-blocking experiments, HLA-blocking antibodies (2 μg ml−1) were added directly into wells 30 min before the addition of peptide. Blocking antibodies used were HLA-DR (1:500 dilution, clone L243, Abcam, ab136320), HLA-DQ (1:500 dilution, clone SPV-L3, Abcam, ab23632) and HLA-DP (1:500 dilution, clone B7/21, Abcam, ab20897). To assess ex vivo T cell reactivity to IDO and PD-L1 peptides in patient PBMCs, cells were thawed and rested for 1–2 d in medium containing DNase I (1 μg ml−1, Sigma-Aldrich, 11284932001). PBMCs were then stimulated with peptide at 5 μM in a 96-well plate for 8 h. An hour after the addition of peptide, anti-CD107a–BV421 (3:50 dilution, clone H4A3, BD Biosciences, 328626) antibody and BD GolgiPlug (1:1,000 dilution, BD Biosciences) were added. Surface and intracellular staining was performed as described above. Antibodies used for surface staining were anti-CD3–PE-CF594 (0.8:30 dilution, clone UCHT1, BD Biosciences, 562280), anti-CD4–BV711 (1:30 dilution, clone SK3, BD Biosciences, 563028) and anti-CD8–Qdot605 (1:150 dilution, clone 3B5, Thermo Fisher, Q10009). Antibodies used for intracellular staining were anti-CD137–PE (1:20 dilution, clone 4B4-1 (RUO), BD Biosciences, 555956), anti-IFN-γ–PE-Cy7 (1.5:20 dilution, clone B27, BD Biosciences, 557643) and anti-TNF-α–APC (1:20 dilution, clone Mab11, BD Biosciences, 554514). Samples were acquired on the NovoCyte Quanteon (ACEA Biosciences) and analyzed using NovoExpress software version 1.4.1. To assess vaccine-specific T cell responses, background values observed in unstimulated PBMC samples were subtracted from values observed in peptide-stimulated conditions. Positive response value threshold was set at a difference of 0.2% from the background values. Based on this response cutoff, only TNF-α, CD107a and CD137 responses were detected in this assay. Statistical analysis comparing baseline with on-treatment or post-treatment cytokine profiles was performed using two-sided Wilcoxon matched-pairs signed-rank test. The gating strategy is shown in Supplementary Fig. 3b. siRNA-mediated PD-L1 and IDO1 silencing A stealth siRNA duplex for targeted silencing of PD-L1 (Invitrogen)52, a custom silencer select siRNA for targeted silencing of IDO1 (Ambion) and the recommended silencer select negative control (Ambion) siRNA for mock transfection were used. The stealth PD-L1 siRNA duplex consisted of the sense sequence 5′-CCUACUGGCAUU-UGCUGAACGCAUU-3′ and the antisense sequence 5′-AAUGCGUUCAGCAAAUGCCAGUAGG-3′. The three silencer IDO siRNA duplexes used were siRNA1 (sense sequence, 5′-ACAUCUGCCUGAUCUCAUATT-3′; antisense sequence, 5′-UAUGAGAUCAGGCAGAUGUTT-3′), siRNA2 (sense sequence, 5′-CCACGAUCAUGUGAACCCATT-3′; antisense sequence, 5′-UGGGUUCACAUGAUCG-UGGAT-3′) and siRNA3 (sense sequence, 5′-CGAUCAUGUGAACCCAAAATT-3′; antisense sequence, 5′-UUUUGGGUUCACAUGAUCGTG-3′). For PD-L1 or IDO1 silencing, cancer cells were electroporated with 0.025 nmol of each siRNA duplex as previously described53. For PD-L1-silencing experiments, cancer cells were treated with IFN-γ (500 U ml−1, PeproTech) 1 h after electroporation. Electroporated cells were used as target cells in ELISPOT and ICS assays 24 h or 48 h after siRNA electroporation. RT–qPCR Total RNA was extracted using the RNeasy Plus Mini kit (Qiagen, 74134) following the manufacturer’s instructions. RNA concentration was quantified using a NanoDrop 2000 (Thermo Fisher Scientific), and a total of 1,000 ng RNA was reverse transcribed with the High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems, 4368814) using 1,000 ng input RNA for transcription. Real-time qPCR analyses were performed using the TaqMan Gene Expression Assay on a Roche LightCycler 480 instrument. RT–qPCR assays were performed with a minimum of three technical replicates and analyzed using the ΔCt method as described in Bookout et al.54 with normalization of IDO1 expression (primer ID, Hs00984148_m1) or PD-L1 expression (primer ID, Hs001125296_m1) to the expression level of the housekeeping gene POLR2A (primer ID, Hs00172187_m1) and the control sample (mock). For low-concentration samples with no amplification, Ct was set to 40. Controls without reverse transcriptase were used as controls for specific amplifications. P values were determined using two-tailed parametric t-tests. BRAF-mutational status and PD-L1 status at baseline in all patients A library of historical FFPE biopsies were assessable for all patients and analyzed locally at Herlev and Gentofte University Hospital by experienced pathologists for BRAF status and PD-L1 expression on tumor cells. The BRAF analysis was carried out with real-time PCR using the EntroGen BRAF Mutation Analysis Kit II (BRAFX-RT64, CE-IVD) to specifically detect mutations corresponding to V600D, V600E and V600K in the BRAF protein. PD-L1 status was assessed using the monoclonal rabbit anti-PD-L1 antibody (clone 28.8, PD-L1 IHC 28-8 pharmDx) in FFPE biopsies following the manufacturer’s instructions. Patients were considered positive for PD-L1 with expression levels ≥1% and negative for PD-L1 with expression levels <1%. Human leukocyte antigen type Blood samples from all 30 patients were genotyped for class I (HLA-A, HLA-B, HLA-C) and three class II types (HLA-DRB1, HLA-DQA1, HLA-DQB1) using LinkSēq HLA Typing Kits (Thermo Fisher, 1580C). These test kits are based on real-time PCR using allele-specific exponential amplification (sequence-specific primers) followed by melting curve analyses. Immunohistochemistry simplex: Immunoscore CR IHC staining was performed using a qualified Ventana Benchmark XT with four different steps: (1) antigen retrieval; (2) staining with the following primary antibodies: anti-CD3 antibody (clone HDX2; provider, HalioDx; HD-FG-000013; 10931065/10636667; concentration, 0.25 µg ml−1), anti-CD8 antibody (clone HDX3; provider, HalioDx; HD-FG-000019; 10931069/10337710/10639301; concentration, 0.5 µg ml−1), anti-IDO monoclonal antibody (clone VINC3IDO; provider, Thermo Fisher Scientific; 14-9750-82; E25003-101; concentration, 0.05 µg ml−1), anti-HLA class 1 ABC antibody (clone, EMR8-5; provider, Abcam; ab70328; GR3248333/GR3186494; concentration, 0.5 µg ml−1), anti-HLA-DR/DP/DQ/DX antibody (clone CR3/43; provider, Santa Cruz; sc-53302; L1714; concentration, 1 µg ml−1) and anti-PD-L1 antibody (clone HDX3; provider, HalioDx; HD-FG-000035; 106312810/106312816; concentration, 3.3 µg ml−1); (3) detection with a secondary antibody using the ultraView Universal DAB Detection kit; and (4) counterstaining using Hematoxylin & Bluing Reagent (staining of cellular nuclei). Control slides were systematically included in each staining run to permit quality control of the obtained measurements. Following coverslipping, slides were scanned with the NanoZoomer-XR to generate digital images (20×). Two consecutive slides were specifically used to perform Immunoscore CR TL staining of CD3+ and CD8+ cells. Digital pathology of T lymphocytes The digital pathology for Immunoscore CR TL allowed the quantification of positive cells (in cells per mm2) into the core tumor and invasive margin if present. Each sample was analyzed using the HalioDx Digital Pathology Platform. Digital pathology and pathologist analysis of IDO Digital pathology for IDO allowed the quantification of stained area (in mm2) into the whole tumor. Each sample was analyzed using the HalioDx Digital Pathology Platform. Analysis of IDO+ cells was performed by a pathologist. Results were expressed as an H score from 0 to 300. The score is obtained by the formula 3 × percentage of cells with strong staining + 2 × percentage of cells with moderate staining + percentage of cells with weak staining. Digital pathologist analysis of MHC-I and MHC-II Analysis of MHC-I+ and MHC-II+ cells was performed by a pathologist. Results are expressed as a percentage of positive tumor cells and a percentage of positive cells in the stroma. Digital pathology Immunoscore immune checkpoint (CD8+ and PD-L1+) The Immunoscore CR IC test allowed the quantification of CD8++ cell density in the whole tumor and a CD8+-centered proximity index (which corresponds to the percentage of CD8++ cells that have at least one PD-L1+ cell in the neighborhood) at different cutoff distances (20 µm, 40 µm, 60 µm and 80 µm). Pathologist analysis of PD-L1 A pathologist performed analysis of PD-L1+ cells. Positivity of a viable tumor cell was considered when partial or complete cell membrane staining was observed (more than 10% of the tumor cell membrane). Results were expressed as a percentage. NanoString RNA profiling and Immunosign RNA was extracted from FFPE tissues using Qiagen RNeasy FFPE extraction kits (Qiagen). Annotations from the pathologist performing H&E staining were used to guide removal of normal tissue from the slides by macrodissection before nucleic acid extraction, which occurred after tissue deparaffinization and lysis. Each extracted RNA sample was independently quantified using a NanoDrop spectrophotometer (NanoDrop Technologies) and qualified (Agilent Bioanalyzer). Degradation was quantified as the percentage of RNA fragments smaller than 300 bp using the RNA 6000 Nano kit (Agilent Bioanalyzer). Good sample quality was defined as less than 50% of RNA fragments being 50–300 bp in size. RNA expression profiling was performed using the nCounter PanCancer Immune Profiling Panel from NanoString (NanoString Technologies). The PanCancer Immune Profiling Panel contains 776 probes and is supplemented with six genes to complete HalioDx Immunosign targets. Hybridization was performed according to the manufacturer’s instructions. Hybridized probes were then purified and immobilized on a streptavidin-coated cartridge using the nCounter Prep Station (NanoString Technologies). Data collection was carried out on the nCounter Digital Analyzer (NanoString Technologies) following the manufacturer’s instructions to count individual fluorescent barcodes and quantify target RNA molecules present in each sample. For each assay, a scan of 490 fields of view was performed. Raw data from the NanoString nCounter were processed using NanoString recommendations. Quality control enables to retain data of good quality with a binding density that ranges between 0.05 and 2.25. The linearity of positive controls was checked using the R2 value of regression between the counts and the concentrations of positive controls. Samples that showed R2 < 0.75 were flagged and removed from the analysis. The background was removed using a thresholding method at the mean + 2 s.d. of negative controls. Raw counts were normalized using a positive normalization factor. Samples showing positive normalization factors outside the range of 0.3–3 were removed from the analysis. A second normalization was performed using the housekeeping gene normalization factor. Only the most stable housekeeping genes were selected for this normalization step using the variance-versus-mean relationship. All samples showing a normalization factor outside the range of 0.1–10 were removed from the analysis. All statistical analyses were performed on normalized counts using R software (version 2.6.2, 2019-12-12). T cell receptor variable β chain sequencing To track longitudinal immune responses to therapy, genomic DNA was extracted from longitudinal pre- and post-treatment PBMCs (five patients), pre- and post-treatment biopsies (five patients) (both FFPE and treated with RNAlater) and IDO- and PD-L1-specific T cell cultures from PBMCs (five patients) or SKILs (one patient). Three clonal (two IDO- and one PD-L1-) specific cultures were also sequenced to confirm clonal purity of cultures. DNA from PBMCs or RNAlater-treated biopsies was extracted with the DNeasy Blood and Tissue kit (Qiagen, 69504), DNA from sorted IDO and PD-L1-specific T cells from either PBMCs or SKILs was extracted using the QIAamp DNA Micro kit (Qiagen, 565304), and DNA from PFFE biopsies was extracted using the Maxwell RSC DNA FFPE kit (Promega, AS1450). Immunosequencing of the CDR3 regions of human TCRβ chains was performed using the immunoSEQÒ Assay (Adaptive Biotechnologies). Extracted genomic DNA was amplified in a bias-controlled multiplex PCR, followed by high-throughput sequencing. Sequences were collapsed and filtered to identify and quantitate the absolute abundance of each unique TCRβ CDR3 region for further analysis as previously described55–57. Statistical analyses of TCRβ sequencing results Two quantitative components of diversity were compared across samples in this study. First, Simpson clonality was calculated on productive rearrangements by ∑i=1Rpi2, where R is the total number of rearrangements and pi is the productive frequency of rearrangement i. Values of Simpson clonality range from 0 to 1 and measure how evenly receptor sequences (rearrangements) are distributed among a set of T cells. Clonality values approaching 0 indicate a very even distribution of frequencies, whereas values approaching 1 indicate an increasingly asymmetric distribution in which a few clones are present at high frequencies. Second, sample richness was calculated as the number of unique productive rearrangements in a sample after computationally downsampling to a common number of T cells to control for variation in sample depth or T cell fraction. Repertoires were randomly sampled without replacement five times, and we report the mean number of unique rearrangements. The T cell fraction was calculated by taking the total number of T cell templates and dividing by the total number of nucleated cells. The total number of nucleated cells was derived from reference genes using the immunoSEQ Analyzer version 3. To identify enriched vaccine clones in each patient, rearrangement frequencies in their baseline PBMCs and each IDO/PD-L1-sorted T cell sample were compared using a binomial distribution framework as previously described58. In brief, for each clone, we performed a two-sided test to determine whether frequencies were the same in the patient’s periphery and a PD-L1- or IDO-specific T cell sample. The Benjamini–Hochberg procedure was used to control the false discovery rate at 0.01 (ref. 59). Clonal expansion in post-treatment samples was similarly assessed using this differential abundance framework, but an IDO/PD-L1-specific T cell sample was replaced with a post-treatment series sample. In biopsies, the six series frequencies were compared to those of baseline tissue. Lastly, vaccine-associated clones were tracked in each PBMC and tissue sample by summing the frequency of each rearrangement enriched in either PD-L1- or IDO-specific T cells. All statistical analyses were performed in R version 3.6.1. Reporting Summary Further information on research design is available in the Nature Research Reporting Summary linked to this article. Online content Any methods, additional references, Nature Research reporting summaries, source data, extended data, supplementary information, acknowledgements, peer review information; details of author contributions and competing interests; and statements of data and code availability are available at 10.1038/s41591-021-01544-x. Supplementary information Supplementary Information Supplementary Figs. 1–3 and Tables 1–6 Reporting Summary Extended data Extended Data Fig. 1 Mode of Action and treatment plan. a) Anticipated mechanism of action of the combination therapy of an IDO/PD-L1 derived peptide vaccine and nivolumab (anti-PD1). 1) The IDO/PD-L1 peptide vaccine is administered subcutaneously (s.c) and nivolumab is administered intravenous (IV). 2) The vaccines peptides are phagocytosed by an antigen presenting cell and presented to IDO and PD-L1 specific T cells, which are activated. 3) The activated T cells migrates to the tumor site where they attack both immune-suppressive cells and tumour cells expressing IDO and/or PD-L1 leading to cytokine production and a pro-inflammatory tumour microenvironment. 4/5) Enhanced tumour killing by both IDO/PD-L1 specific T cells and tumor specific cytotoxic T cells due to PD-1 blockade. Created with BioRender.com b) Treatment plan. After written informed content patients were screened. Before treatment start a baseline PET/CT scan was performed, baseline blood sample for research use and if assessable a baseline needle biopsy. Patients were treated with the IDO/PD-L1 peptide vaccine subcutaneously biweekly for the first 6 injections and thereafter every fourth week for a maximum of 15 vaccines. Nivolumab was administered in parallel biweekly (3mg/kg) up to 24 series. If patients needed subsequent nivolumab after ended vaccination regiment they were treated with 6 mg/kg every fourth week up to two years. Needle biopsy and delayed type hypersensitivity (DTH) was performed after 6 series of treatment if assessable. PET/CT scans was performed every third month. Extended Data Fig. 2 Progression free survival and overall survival in matched historical control group and vaccine injection site reaction. a) Kaplan-Meier curve of progression free survival with corresponding 95% confidence intervals in the matched historical control group (n=74). Patients were matched on BRAF-status, PD-L1-status, age, gender, M-stage and LDH level). b) Kaplan-Meier curve of overall survival with corresponding 95% confidence intervals in the matched historical control group (n=74). c) Images of injection site reaction in patient MM20 (CR) after 11 vaccination show redness, rash and granuloma at injection site. Extended Data Fig. 3 Vaccine specific responses in blood. a) Heatmap of specific (background has been subtracted) IDO and PD-L1 responses in PBMCs at baseline, series 3, 6, 12, 18 and 24 measured by in vitro IFN-γ Elispot assay show fluctuations in the blood during treatment (n=30). b) Heatmap of specific (background has been subtracted) IDO and PD-L1 responses in PBMCs at baseline and 3 and 6 months after last vaccine measured by in vitro IFN- γ Elispot assay (n=30). 2.5–3.2x105 cells per well were used. c) Vaccine associated clones were tracked in the blood by summing the frequency of each rearrangement enriched in either IDO or PD-L1 T cells. Extended Data Fig. 4 Ex vivo vaccine specific responses in blood. Heatmaps of detected specific (background has been subtracted) IDO (a) and PD-L1 (b) responses in PBMCs at baseline and on/after treatment as measured by IFNγ ELISPOT (n=25). C) Example of well images of ex vivo ELISPOT wells for three different patients (n=3). 6-9x105 cells per well were used. Extended Data Fig. 5 CD4 and CD8 vaccine specific T cell responses in blood. Top: heatmaps of IDO specific CD4 (left) and CD8 (right) T cell responses in PBMCs at baseline and on/post treatment. Bottom: heatmaps of PD-L1 specific CD4 (left) and CD8 (right) T cell responses. Responses quantified by flow cytometry by an increased expression of CD107α, CD137 and TNFα after 8h peptide stimulation. Values represent specific responses after background values have been subtracted (n=21). Statistical analysis were performed using two-sided Wilcoxon matched-paired rank test. Extended Data Fig. 6 Pro-inflammatory profiles of sorted CD4 and CD8 T cells from blood. a/b/c/d) Percentage of in vitro stimulated and sorted PD-L1 and IDO specific CD4 and CD8 T cells secreting cytokines. Extended Data Fig. 7 Vaccine specific responses in skin. a) IDO and PD-L1 specific T cell responses in SKILs after 6 series of treatment measured by IFN-γ Elispot assay (n=13). SKILs were grown from DTH injection with either IDO peptide, PD-L1 peptide or a mix as presented by different blue colours. * Responses were calculated as the difference between average numbers of spots in wells stimulated with IDO or PD-L1 peptide (triplicates) and corresponding control (DMSO) and statistical analyses of Elispot responses were performed using distribution-free resampling method (Moodie et al.). DR: Not statistically confirmed response due to replicate number but number of spots in peptide wells are two times higher than control wells (DMSO). NS: No significant response and no DR. b/c/d) Percentage of cytokine secreting/CD107a+ CD4+ and CD8+ IDO and PD-L1 specific T cells in response to in vitro peptide stimulation by flow cytometry. e) Skin infiltrating PD-L1 specific T cell clones also found in biopsy in patient MM01. TCR sequencing was performed on a PD-L1 specific T cell culture generated from DTH area on the lower back injected with PD-L1 peptide on patient MM01. Bars show the frequency of top 25 clones in the culture with which indicates a high Simpson clonality of 0.43. f) Tracking the frequency of the top five skin infiltrating PD-L1 specific clones in tumour before and after treatment. Extended Data Fig. 8 T cell changes in blood after treatment. T cell fraction, TCR clonality and repertoire richness in blood and peripherally expanded TCR clones in both responding and non-responding patients. a) T cell fraction in peripheral blood of five patients at baseline, series 3, 6 and 12 by TCR sequencing. T cell fraction was calculated by taking the total number of T cell templates and dividing by the total number of nucleated cells (n=5.) b) TCR clonality in peripheral blood of five patients at baseline, series 3, 6 and 12 by TCR sequencing (n=5). Simpson clonality measures how evenly TCR sequences are distributed amongst a set of T cells where 0 indicate even distribution of frequencies and 1 indicate an asymmetric distribution where a few clones dominate. c) TCR repertoire richness in peripheral blood of five patients at baseline, series 3, 6 and 12 by TCR sequencing (n=5). TCR repertoire richness report the mean number of unique rearrangements. d) Bar-chart chart representing dynamics of the expanded T cell clones. Coloured bars represent peripherally expanded clones in five patients at series 3, series 6 and series 12 as compared to the baseline PBMC samples (n=5). Light gray bar represents peripherally expanded clones that are present in baseline biopsy samples while dark gray bar represents peripherally expanded clones that are present in the post-treatment biopsy (after 6 series). e) Frequency of the dominant TCR β chain in clonal PD-L1 and IDO specific cultures as determined by CDR3 sequencing. Extended Data Fig. 9 Profiling of genes relevant to T cell activation, cytokines and exhaustion markers on pre- and post-treatment biopsies in two patients. a) RNA expression profiling of genes related to T cell activation was performed using NanoString nCounter (n=2). b) RNA expression profiling of genes related to cytokine activity was performed using NanoString nCounter (n=2). c) RNA expression profiling of genes related to checkpoint inhibitors was performed using NanoString nCounter (n=2). Extended Data Fig. 10 Treatment induced upregulation of PD-L1, IDO, MHC I and MHC II and distance between CD8 T cells and PD-L1 expressing cells. a) IHC on 4 paired biopsies stained for CD3 and CD8 T cells, PD-L1, MHC I and MHC II on tumor cells/mm2 (sum in the validated area) and IDO H-score (from 0 to 300). H-score is the expression of IDO on both immune and tumor cells: The score is obtained by the formula: 3 x percentage of cells with a strong staining + 2 x percentage of cells with a moderate staining + percentage of cells with a weak staining (n=4). b) Distance in µm between CD8+ T cells and PD-L1+ stained cells on five baseline biopsies detected by IHC (n=5). Extended data is available for this paper at 10.1038/s41591-021-01544-x. Supplementary information The online version contains supplementary material available at 10.1038/s41591-021-01544-x. Acknowledgements We thank all patients and their relatives for being a part of the trial. We thank Ö. Met, M. Jonassen, S. Ullitz Færch, B. Saxild, S. Wendt and C. Grønhøj for technical support. We thank M. Cumberbatch for input with translational analysis. We thank the nurses at clinic 5 and the head of the Oncology Department at Herlev and Gentofte Hospital, L. Sengeløv. The study was funded through a research funding agreement between IO Biotech and the CCIT-DK, Herlev Hospital and the Oncology Department at Herlev Hospital. Author contributions I.M.S. and M.H.A. conceived the study. J.W.K. wrote the protocol with input from I.M.S., M.H.A. and M.-B.Z. J.W.K. and C.L.L. were responsible for patient treatment and patient care with help from R.B.H. and consulting from I.M.S., E. Ellebaek and M.D. I.M.S., E. Ellebaek and M.D. were responsible for patient recruitment. J.W.K. and C.L.L. were responsible for coordinating trial procedures and collecting data and samples. J.W.K. and C.L.L. were clinical investigators, and I.M.S. was a sponsor. J.W.K. and C.L.L. analyzed and interpreted clinical data with support from I.M.S. and E. Ehrnrooth. T.W.K. performed statistical analyses. H.W.H. helped with clinical imaging. J.W.K., E.M., C.O.M., S.M.A., S.E.W.-B., M.O.H. and A.W.P. analyzed and interpreted translational data with support from M.H.A. and I.M.S. Funding was acquired by I.M.S., M.H.A. and M.-B.Z. J.W.K. wrote the manuscript with input from I.M.S., M.H.A., E.M. and A.W.P. All authors reviewed the manuscript, interpreted data and approved the final version. Data availability This clinical trial was registered at https://www.clinicaltrials.gov/ before patient enrollment (clinical trial identifier NCT03047928). Each vaccine was composed of 100 µg IO102, a 21-amino-acid peptide (DTLLKALLEIASCLEKALQVF) from the IDO peptide, and 100 µg IO103, a 19-amino-acid peptide (FMTYWHLLNAFTVTVPKDL) from the signal peptide of PD-L1. TCR sequencing data are available from Adaptive Biotechnologies. Upon request, the CCIT-DK office will provide a username and a password to access the designated data within approximately 2–4 weeks (https://www.herlevhospital.dk/ccit-denmark/find-us/Sider/Contact-information.aspx). All requests for the remaining data including raw data and analyzed data and materials will, within a reasonable time frame, be reviewed by the CCIT-DK office (https://www.herlevhospital.dk/ccit-denmark/find-us/Sider/Contact-information.aspx) to verify whether the request is subject to any intellectual property or obligations. Patient-related data not included in the paper were generated as part of clinical trials and may be subject to patient confidentiality. Any data and materials that can be shared will be released via a material-transfer agreement. The following database was used in the study: https://research.regionh.dk/da/publications/the-danish-metastatic-melanoma-database-dammed(32749d99-095f-4cae-b5de-769bae27f01e).html. Competing interests M.H.A. is named as an inventor on various patent applications relating to therapeutic uses of IDO and PD-L1 peptides. These patent applications are assigned to the company IO Biotech, which is developing immune-modulating cancer treatments. M.H.A. is a founder, shareholder and advisor for IO Biotech. I.M.S. is a cofounder, shareholder and advisor for IO Biotech. I.M.S. has an advisory board relationship with or lectured for Roche, Novartis, MSD, Celgene, Incyte, TILT Bio, Pfizer and BMS AstraZeneca and has received limited grants for translational research from BMS, Roche and Novartis. M.-B.Z. is the CEO, founder and shareholder at IO Biotech. E.M., A.W.P. and E. Ehrnrooth are employees at IO Biotech. E. Ellebaek has received honoraria from BMS, Pierre Fabre, Roche and Kyowa Kirin and travel support from MSD. M.D. has received honoraria from Genzyme, MSD, BMS, Roche and Novartis and travel support from Novartis, MSD, BMS, Roche and Pfizer. The remaining authors declare that they have no conflict of interest. Peer review information Nature Medicine thanks George Coukos, Michael Schell and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. Javier Carmona was the primary editor on this article and managed its editorial process and peer review in collaboration with the rest of the editorial team. Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. These authors contributed equally: Julie Westerlin Kjeldsen, Cathrine Lund Lorentzen. Change history 3/8/2022 A Correction to this paper has been published: 10.1038/s41591-022-01771-w

3 MILLIGRAM/KILOGRAM, Q2WK

DrugDosageText

CC BY

34887574

2021-12-09

What was the outcome of reaction 'Hyponatraemia'?

A phase 1/2 trial of an immune-modulatory vaccine against IDO/PD-L1 in combination with nivolumab in metastatic melanoma. Anti-programmed death (PD)-1 (aPD1) therapy is an effective treatment for metastatic melanoma (MM); however, over 50% of patients progress due to resistance. We tested a first-in-class immune-modulatory vaccine (IO102/IO103) against indoleamine 2,3-dioxygenase (IDO) and PD ligand 1 (PD-L1), targeting immunosuppressive cells and tumor cells expressing IDO and/or PD-L1 (IDO/PD-L1), combined with nivolumab. Thirty aPD1 therapy-naive patients with MM were treated in a phase 1/2 study ( https://clinicaltrials.gov/ , NCT03047928). The primary endpoint was feasibility and safety; the systemic toxicity profile was comparable to that of nivolumab monotherapy. Secondary endpoints were efficacy and immunogenicity; an objective response rate (ORR) of 80% (confidence interval (CI), 62.7-90.5%) was reached, with 43% (CI, 27.4-60.8%) complete responses. After a median follow-up of 22.9 months, the median progression-free survival (PFS) was 26 months (CI, 15.4-69 months). Median overall survival (OS) was not reached. Vaccine-specific responses assessed in vitro were detected in the blood of >93% of patients during vaccination. Vaccine-reactive T cells comprised CD4+ and CD8+ T cells with activity against IDO- and PD-L1-expressing cancer and immune cells. T cell influx of peripherally expanded T cells into tumor sites was observed in responding patients, and general enrichment of IDO- and PD-L1-specific clones after treatment was documented. These clinical efficacy and favorable safety data support further validation in a larger randomized trial to confirm the clinical potential of this immunomodulating approach. pmcMain Despite remarkable advances in the treatment of MM with immune checkpoint inhibitors (ICIs) targeting PD-1 and cytotoxic T lymphocyte antigen 4 (CTLA-4), around half of patients are resistant to ICI monotherapy1. The combination of anti-CTLA-4 (aCTLA-4) and aPD1 therapy is to date the most effective therapy resulting in a response rate of around 60%; however, 50% of the patients also develop severe adverse events2,3. Therefore, an equally effective but less toxic treatment is highly needed. Several approaches to enhance ICI efficacy are currently being investigated, such as other ICIs, T cell therapy with tumor-infiltrating T cells or innate immunity stimulators such as Toll-like receptor 9 agonists4–6. Treating cancer patients with vaccines that stimulate a targeted immune response is another attractive approach, with very few side effects observed thus far in combination immunotherapy studies7,8. Immune-modulatory vaccines targeting tumoral immune escape mechanisms offer a new, generalizable strategy compared to patient-specific neoantigen cancer vaccines7,9. The immune-modulatory vaccine strategy in this clinical trial is based on the finding of circulating cytotoxic T cells specific to IDO and PD-L1 in the blood of patients with cancer and, to a lesser extent, in healthy donors. IDO- and PD-L1-specific CD8+ T cells can directly recognize and kill IDO+ and/or PD-L1+ tumor cells and likewise recognize and kill non-malignant cells that express their cognate targets. Furthermore, IDO- and PD-L1-specific CD4+ T cells release pro-inflammatory cytokines in response to IDO- or PD-L1-expressing target cells. IDO and PD-L1 are expressed not only by melanoma cells but also by many other cell types in the tumor microenvironment (TME), which differentiates these antigens from traditional tumor antigens used in other studies10–15. Activation of IDO/PD-L1-specific T cells by vaccination can therefore restrict the range of immunosuppressive signals mediated by immunosuppressive cells and thereby revert the TME from an immune hostile to an immune friendly environment. In animal models of cancer, vaccinations with IDO epitopes resulted in anti-tumor therapeutic effects that were correlated with reductions in IDO expression in myeloid cell populations within the TME16. The IDO/PD-L1 immune-modulating vaccine may lead to a translatable strategy for improving the efficacy of aPD1 therapy through activation of specific T cells. We hypothesize that the IDO/PD-L1 vaccine attracts T cells into the tumor, which induces type 1 helper T (TH1) cell inflammation and reverts the TME into an immune-permissive site, thereby turning the tumor ‘hot’. This would also upregulate PD-L1 expression in cancer and immune cells, generating more susceptible targets to aPD1 therapy (Extended Data Fig. 1a). This theory was confirmed in a mouse model in which aPD1 therapy and IDO vaccination show synergistic effects16. In this phase 1/2 clinical trial, MM1636, patients with MM received a combination of the IDO/PD-L1 (IO102/IO103) peptide vaccine with the adjuvant Montanide and the aPD1 antibody nivolumab. Patients were included in three cohorts: 30 aPD1 therapy-naive patients (cohort A), ten aPD1 therapy-refractory patients (cohort B, de novo resistance) and ten patients who progressed after aPD1 therapy (cohort C, acquired resistance). Here, we report results from cohort A. The vaccine was given biweekly for the first six administrations and thereafter every 4th week. A maximum of 15 vaccines were administered. Nivolumab was given in parallel, biweekly (3 mg per kg) or every 4th week (6 mg per kg) for up to 2 years (Extended Data Fig. 1b). The primary objective was safety and feasibility. Secondary objectives were immunogenicity and clinical efficacy. Results Patients and treatment Thirty patients were enrolled from December 2017 to June 2020. None of the 30 patients dropped out of the study; all received at least three cycles of therapy (Supplementary Fig. 1). At the current database lock (5 October 2020), six patients were still on treatment in the trial. Of the 24 patients who were not on trial treatment at data cutoff, two are still receiving nivolumab monotherapy (6 mg per kg every 4 weeks). Reasons for stopping treatment for the remaining 22 patients included disease progression (37%), toxicity (20%), maximum benefit or complete response (CR) confirmed on two consecutive scans (17%) or completing 2 years of treatment (7%). For the 24 patients who were not on trial treatment at data cutoff, the mean number of vaccinations was 10.5 (range, 3–15 vaccinations). Thirteen of these 24 patients continued nivolumab (6 mg per kg, every 4 weeks) as a standard of care. Nine patients received subsequent therapy after progression (Supplementary Table 1). Baseline characteristics are shown in Table 1. The mean age was 70 years; 37% of patients had elevated lactate dehydrogenase (LDH) levels, 60% were stage M1c, 37% had BRAF mutations, and 43% were negative for PD-L1 (<1%). A total of three patients (10%) had received prior ipilimumab therapy. No patients had brain metastasis (Supplementary Table 2).Table 1 Baseline patient characteristics (n = 30) Characteristic Number (%) Mean age, years (range) 70 (46–85) Sex, male 16 (55%) ECOG PS 0 26 (87%) LDH levels  ≤ULN 19 (63%)  >ULN 11 (37%) M stage (AJCC-8)  M1a 6 (20%)  M1b 6 (20%)  M1c 18 (60%) Number of lesion sites  1 6 (20%)  2–3 17 (57%)  >3 7 (23%) Liver metastases present 10 (33%) BRAF status  Mutant 11 (37%)  Wild type 19 (63%) PD-L1  <1% 13 (43%)  >1% 17 (57%) Previous systemic therapy  Ipilimumab 3 (10%)  No 27 (90%) AJCC-8, eighth edition of the American Joint Committee on Cancer; ECOG PS, Eastern Cooperative Oncology Group performance status; ULN, upper limit of normal. Notable clinical responses to the combination therapy Thirty patients with MM were treated with the IDO/PD-L1 vaccine and nivolumab according to the trial protocol. By investigator review, the ORR reached 80% (CI, 62.7–90.5%), with 43% of patients (CI, 27.4–60.8%) achieving a CR and 37% (CI, 20.9–54.5%) reaching a partial response (PR) as the best overall response, while 20% experienced progressive disease (PD) according to Response Evaluation Criteria in Solid Tumors (RECIST) 1.1 (Fig. 1a). Two of the patients with a PR did not have a confirmatory scan stating PR on two consecutive scans. Early onset of response was frequent, with 22 of 30 patients having an objective response at the first evaluation (after 12 weeks on treatment). Median times to PR and CR were 75 d (range, 54–256 d) and 327 d (range, 73–490 d), respectively (Fig. 2a–c).Fig. 1 Clinical response. a, Pie charts with percent ORR, CR, PR and PD according to RECIST 1.1 by investigator review of all patients (n = 30), PD-L1+ patients (>1%, (n = 17)) and PD-L1− patients (<1%, n = 13)), respectively. Two-sided CIs (95%) were constructed using the Clopper–Pearson method. b, Treatment effect in MM1636 compared with a matched historical control group from the DAMMED database (n = 74). Patients in MM1636 (n = 29) were matched with the exact same combination variable according to age (≤70 years, >70 years), sex, LDH levels (normal, elevated), M stage (M1a, M1b, M1c), BRAF status (wild type, mutated) and PD-L1 status (<1%, ≥1%). Estimates for treatment effects were calculated by weighted logistic regression analyses and weighted Cox proportional hazard model. Bar height indicates the estimated response rate; tops of bars are centers for error bars. Odds ratios (OR), response rates and their corresponding 95% CIs were extracted from the regression model. All P values were two sided, and P values below 0.05 were considered statistically significant. c, Best change in the sum of target lesion size compared with that at baseline (n = 30). The horizontal line at −30 shows the threshold for defining an objective response in the absence of non-target disease progression or new lesions according to RECIST 1.1. Two patients with 100% reduction in target lesion size had non-target lesions present. White stars, six patients had normalization (<10 mm) of fluor-18-deoxyglucose (FDG)-negative lymph nodes (at baseline, lymph nodes were >1.5 cm and FDG+) and 100% reduction of non-lymph node lesions and are considered to have had a CR (green bar). Black star, one patient (MM29) was considered to have had a PR (blue bar), although he had not reached a −30% change in target lesion size. The patient had a single measurable 13-mm lung metastasis at baseline and multiple biopsy-verified cutaneous metastases on the left crus (not detectable by positron-emission tomography–computed tomography (PET–CT) at baseline). The best change in target lesion size was 10 mm, and a post-treatment biopsy from the cutaneous metastases showed no sign of malignancy. Thus, overall, the patient was classified as having a PR. d, Kaplan–Meier curve of the response duration in the 24 patients with an objective response. e, Kaplan–Meier curve of PFS in all 30 treated patients. f, Kaplan–Meier curve of OS in all 30 treated patients. Fig. 2 Clinical response. a, Swimmer plot showing response duration and time to response according to RECIST 1.1 for all treated patients (n = 30). Triangles indicate first evidence of PR, while squares indicate first evidence of CR. Closed circles indicate time of progression. Arrows indicate ongoing responses. Patient MM18 died due to nivolumab-induced side effects. b, Spider plot showing response kinetics in all treated patients (n = 30). Red squares indicate time of progression. c, PET–CT images of patient MM42 before and after treatment (after 12 series of treatment) showing FDG metabolism in target lesions. SC, subcutaneous. The ORR among PD-L1+ (>1% (clone 28.8)) patients (n = 17) was 94.1% (CI, 73–99.7%) and 61.5% (CI, 35.5–82.3%) in PD-L1− patients (n = 13) (Fig. 1a). Objective responses were observed in patients irrespective of human leukocyte antigen (HLA) genotype (Supplementary Fig. 2). Clinical response data were validated by blinded independent external review, in which an ORR of 76.6% (CI, 57.7–90.1%) was reported, with 53.3% of patients achieving a CR, 23.3% achieving a PR and 3.3% experiencing stable disease. Comparisons between investigator review and external review are outlined in Supplementary Table 3. To examine whether the very high response rate should be attributed to nivolumab or to the vaccine, we retrieved clinical response information from a matched historical control group using the Danish Metastatic Melanoma Database (DAMMED) from contemporaneously treated patients with stage 3–4 melanoma who received aPD1 monotherapy17. Patients were matched with the exact same combination variable according to age, sex, PD-L1 status, BRAF status, LDH level and M stage (those at stage M1d were excluded from the control group (no patients with brain metastasis)). Matched controls were identified for 29 patients, and the ORR of 79.3% (CI, 61.0–90.4%) observed in MM1636 was found to be significantly higher (P < 0.0012) than that in the matched control group in which an ORR of 41.7% (CI, 31.0–53.3%) was reached. Furthermore, of the 29 patients in MM1636, a significantly (P < 0.0017) higher percentage (41.4% (CI, 25.2–59.6%)) of patients achieved CR in MM1636 than that (12% (CI, 6.3–21.6%)) in the matched historical control group. ORR and complete response rate (CRR) in the matched historical control group were comparable to those of patients treated in randomized phase 3 pivotal trials with aPD1 monotherapy18 (Fig. 1b). The treatment leads to prolonged PFS At data cutoff, the median duration of response had not been reached, with 87% of all responding patients being free from progression at 12 months (Fig. 1d). Patients were followed for up to 35 months with a median follow-up time of 22.9 months (CI, 14.9–26.2 months). OS and PFS were calculated from the first day of treatment to death or progression or to the date of the last follow-up (5 October 2020). The median PFS (mPFS) for all treated patients was 26 months (CI, 15.4–69 months) and was not reached for responding patients (Fig. 1e). The median OS was not reached at the data cutoff. OS at 12 months was 81.6% (CI, 61.6–92%) (Fig. 1f). One patient (MM18) with a CR died from nivolumab-related severe adverse events; the remaining patients died because of metastatic melanoma. For comparison, the mPFS was 8.3 months (CI, 5.5 months–NR (not reached)) in the matched historical control group (n = 74), while the median OS was 23.2 months (CI, 23.2 months–NR). (Extended Data Fig. 2a,b). The combination of the IDO/PD-L1 vaccine and nivolumab was safe Treatment-related adverse events are listed for all 30 patients in Supplementary Table 4. Common treatment-related grade 1–2 toxicities were fatigue (47%), rash (47%), arthralgia (30%), diarrhea (23%), nausea (23%), dry skin (20%), pruritus (20%), infusion reaction (17%), xerostomia (17%) and myalgia (17%). Four patients (13%) experienced grade 3–4 adverse events: one patient with a grade 3 maculopapular rash (MM01), one patient with grade 3 adrenal insufficiency (MM06) and one patient with grade 3 arthralgia (MM22). Patient MM18 died from urosepsis with multi-organ failure and severe hyponatremia. This patient had experienced multiple immune-related adverse events with grade 3 colitis, grade 2 pneumonitis, grade 3 arthralgia, grade 2 vasculitis and grade 2 nivolumab infusion-related allergic reaction. Additionally, patient MM18 had symptoms of myocarditis at the time of death with highly elevated cardiac troponin I levels. Bedside echocardiography showed an ejection fraction of 15%, which at baseline was 60%, but an autopsy was not conducted, and myocarditis was never confirmed pathologically. Patient MM06 had received first-line treatment with ipilimumab before entering the trial and was on substitution corticosteroids at the time of inclusion. Adrenal insufficiency was aggravated by an erysipelas infection with high fever, reaching grade 3 in Common Terminology Criteria for Adverse Events and resolving quickly after appropriate antibiotic therapy was initiated. As expected, local side effects were common with 77% of the patients who developed injection site reactions. These reactions were classified as granulomas (63%), redness (20%), pain (13%) and pruritus (13%) at the injection site. All local reactions were grade 1–2, most likely related to the Montanide adjuvant and typically transient. However, two patients (MM07 and MM20) decided to discontinue vaccination after eight and 11 injections, respectively, due to granulomas, tenderness and pain that limited instrumental activities of daily living but continued to receive nivolumab (Extended Data Fig. 2c). Vaccine-specific responses in blood were frequently detected First, all 30 patients were assessed for the presence of vaccine-specific responses in peripheral blood mononuclear cells (PBMCs) before, on and after vaccination using a modified interferon (IFN)-γ enzyme-linked immune absorbent spot (ELISPOT) assay. This assay is known to expand antigen-specific memory cells and to improve the detection and the correlative power of ELISPOT during treatment19–21. Prevaccine IDO-specific responses were detectable in ten (33%) patients, while prevaccine PD-L1-specific responses were detectable in eight patients (27%); overlapping (specific to both IDO and PD-L1) prevaccine responses were present in four (13.3%) patients. During vaccination, an increase of IDO-specific T cells or PD-L1-specific T cells in the blood was observed in 28 (93%) and 26 (86%) patients, respectively. In total, 93% of patients had an increase in either PD-L1- or IDO-specific responses on vaccination (Fig. 3a), with a significant (P < 0.0001) median increase from baseline to the post-vaccine response demonstrated for both IDO and PD-L1 (at different time points on treatment), confirming that vaccine-specific immune responses were induced in patients regardless of clinical response (Fig. 3b,c). Immune responses fluctuated in the blood over time (Extended Data Fig. 3a). An increase in IDO- and PD-L1-specific responses in the peripheral blood was also detectable directly ex vivo across the different clinical response groups (Extended Data Fig. 4).Fig. 3 Vaccine-specific responses in blood. a, IDO- and PD-L1-specific T cell responses in PBMCs at baseline and on vaccination as measured by the IFN-γ ELISPOT assay (n = 30). *Responses were calculated as the difference between the average numbers of spots in wells stimulated with IDO or PD-L1 peptide (triplicates) and those from the corresponding control (DMSO), and statistical analyses of ELISPOT responses were performed using a distribution-free resampling method (Moodie et al.50). DR (double response), response was not statistically confirmed due to replicate number, but the number of spots in peptide wells was two times higher than that in control wells (DMSO). NS, no significant response and no DR. For a detailed overview of responses at serial time points on vaccination, see Extended Data Fig. 4a. b, IDO- and PD-L1-specific T cell response in PBMCs in all treated patients measured by the IFN-γ ELISPOT assay at baseline and on vaccination. On-vaccination responses were selected from the ‘best’ ELISPOT response at different time points for each patient (series 3, 6, 12, 18 or 24) during vaccination (n = 30). Wilcoxon matched-pairs signed-rank test was used to compare responses to IDO or PD-L1 peptides in the vaccine between baseline and later time points. c, Representative example of ELISPOT wells with response in patient MM23 in serial PBMCs before and on treatment. d, IDO-specific CD4+ and CD8+ T cells were isolated and expanded from PBMCs stimulated in vitro with the IDO peptide and a low dose of IL-2 for 14–15 d before sorting using the Miltenyi Cytokine Secretion Assay—Cell Enrichment and Detection kit. To assess their cytolytic potential, IDO-specific T cells were stimulated with IDO peptide, and expression of CD107a, IFN-γ and TNF-α was assessed by flow cytometry (the example is from patient MM14). Sustained vaccine-specific responses were observed 3 and 6 months after the last vaccine, indicating induction of memory responses in nine patients with response to clinical treatment who surpassed follow-up at data lock (Extended Data Fig. 3b). Importantly, PD-L1- and IDO-specific responses were observed irrespective of HLA genotype (Supplementary Table 5). To verify the functionality of vaccine-induced T cells, IDO- or PD-L1-specific T cells were isolated and in vitro expanded from PBMCs of five patients. Phenotypic characterization by flow cytometry revealed that the isolated vaccine-specific T cells consisted of both CD4+ and CD8+ T cells. Also, both IDO- and PD-L1-specific CD4+ and CD8+ T cells showed pro-inflammatory properties, as they expressed the cytolytic marker CD107a and secreted the cytokines IFN-γ and tumor necrosis factor (TNF)-α (Fig. 3d and Extended Data Fig. 6a–d). Interestingly, we were also able to detect vaccine-specific CD4+ and CD8+ T cell responses in peripheral blood ex vivo (Extended Data Figs. 4 and 5). We observed a significant increase in the overall percentage of CD107a, CD137 and TNF-α expression in response to peptide stimulation in on- or post-treatment PBMC samples compared to that at baseline, further confirming the expansion and the diverse signature of vaccine-specific T cells (Extended Data Fig. 5). Vaccine-specific responses detected in the skin at the vaccination site To investigate whether vaccine-specific T cells have the potential to migrate to peripheral tissue, delayed-type hypersensitivity (DTH) tests were performed after six cycles of treatment on 15 patients to assess the presence of vaccine-reactive T cells in the skin. Supplementary Table 6 display an overview of skin-infiltrating lymphocyte (SKIL) cultures. We detected IDO-specific T cells in the skin of six of ten patients and PD-L1-specific T cells in nine of 11 patients (Extended Data Fig. 7a). Intracellular cytokine staining was performed on SKILs from five patients after stimulation with either PD-L1 or IDO peptide. Here, we detected mainly CD4+ peptide-reactive T cells that secreted TNF-α and upregulated CD107a, and a minor fraction also secreted IFN-γ. In one patient, we detected CD8+ PD-L1-reactive T cells (Extended Data Fig. 7b–d). Vaccine-induced T cells specifically recognize target cells To confirm the functionality of vaccine-expanded T cells, vaccine-specific T cell clones (clonal purity was confirmed by T cell receptor (TCR) sequencing) were isolated and expanded from patient PBMCs (Extended Data Fig. 8e). We showed that PD-L1-specific T cells were able to recognize PD-L1+ autologous tumor cells in a PD-L1 expression-dependent manner if the cancer cells also expressed HLA-II (Fig. 4a,b). Similarly, an HLA-DR-restricted IDO-specific CD4+ T cell clone was able to recognize an HLA-DR-matched IDO-expressing model cell line, MonoMac1, in an IDO-expression-dependent manner (Fig. 4e–g). As previously described, IDO- and PD-L1-specific T cells’ mode of action is not limited to targeting only cancer cells. We were able to show that vaccine-specific T cell clones also reacted against PD-L1- and IDO-expressing autologous immune cells (Fig. 4c,h). To provide myeloid cells with a tumor-associated phenotype we treated isolated CD14+ myeloid cells with tumor conditioned medium (TCM) derived from an established autologous tumor cell line. We observed that such TCM-treated CD14+ cells had increased expression of PD-L1 and IDO and were effectively recognized by autologous PD-L1- and IDO-specific CD4+ T cell clones (Fig. 4c,d,h,i).Fig. 4 PD-L1- and IDO-specific T cells from vaccinated patients react against PD-L1- and IDO-expressing target cells. a, Left, PD-L1-specific T cell culture (MM1636.05) reactivity against PD-L1 peptide or autologous tumor cells in the IFN-γ ELISPOT assay. Tumor cells were either not treated or pretreated with 200 U ml−1 IFN-γ for 48 h before the assay. Effector:target (E:T) ratio of 10:1 was used. Right, PD-L1 and HLA-II surface expression on melanoma cells with (green) or without (yellow) pretreatment with IFN-γ compared to an isotype control (gray) as assessed by flow cytometry. b, Left, PD-L1-specific T cell (MM1636.05) reactivity in the IFN-γ ELISPOT assay against autologous tumor cells pretreated with IFN-γ (500 U ml−1) and transfected with mock or PD-L1 small interfering (si)RNA 24 h after transfection. E:T ratio, 10:1. Right, PD-L1 surface expression on melanoma tumor cells (MM1636.05) assessed by flow cytometry 24 h after transfection with mock (blue) or PD-L1 (red) siRNA compared to the isotype control (gray). c, Reactivity of the CD4+ PD-L1-specific T cell clone (MM1636.14) against PD-L1 peptide or autologous CD14+ cells; E:T ratio, 10:1. CD14+ cells were isolated using magnetic bead sorting and used as targets in an ELISPOT assay directly or after pretreatment for 2 d with TCM derived from the autologous tumor cell line. d, Quantitative PCR with reverse transcription (RT–qPCR) analysis of PD-L1 (CD274) expression in sorted CD14+ cells before and after treatment with autologous TCM for 48 h. e, Reactivity of the IDO-specific CD4+ T cell clone (MM1636.23) against IDO peptide combined with HLA-DR (L243)-, HLA-DQ (SPV-L3)- or HLA-DP (B7/21)-blocking antibodies (aHLA-DR, aHLA-DQ and aHLA-DP) in an intracellular staining assay (ICS) for IFN-γ and TNF-α production. T cells were incubated with individual blocking antibodies (2 μg ml−1) for 30 min before adding IDO peptide. f, Reactivity of the IDO-specific CD4+ T cell clone (MM1636.23) against the HLA-DR-matched IDO-expressing cell line MonoMac1 transfected with mock or IDO siRNA in an ICS assay, E.T ratio, 4:1. siRNA transfection was performed 48 h before the experiment. g, RT–qPCR analysis of IDO1 expression in MonoMac1 cells 48 h after siRNA transfection. h, Reactivity of the CD4+ IDO-specific T cell clone (MM1636.14) against IDO peptide or autologous CD14+ cells; E:T ratio, 20:1. CD14+ cells were isolated using magnetic bead sorting and used as targets in an ELISPOT assay directly or after pretreatment with TCM derived from the autologous tumor cell line. i, RT–qPCR analysis of IDO1 expression in sorted CD14+ cells before and after treatment with autologous TCM for 48 h. Bars in RT–qPCR data (d,g,i) represent the mean of three (d,i) or six (g) technical replicates ±s.d.; P values were determined by two-tailed parametric t-tests. ELISPOT counts (a,b,c,h) represent the mean value of three technical replicates ±s.e.m.; response P values were determined using the distribution-free resampling (DFR) method. TNTC, too numerous to count. T cell clones in blood and tumors To track the role of treatment-induced T cell responses, TCR sequencing of the complementarity-determining region 3 (CDR3) was performed on five patients in peripheral blood (baseline and cycles 3, 6 and 12) and paired biopsies. These five patients (MM01, MM02, MM08, MM09 and MM13) were selected due to the availability of material and to investigate a balanced patient group with both responders and non-responders. Due to a limited number of available paired biopsies (either as a consequence of patient refusal or unexpectedly rapid and substantial clinical responses), statistical analysis could not be applied. Details on clinical response are shown in Fig. 2a. Additionally, PBMCs (on treatment) or SKILs were stimulated with the IDO/PD-L1 peptides, and then cytokine-producing T cells were sorted to track vaccine-induced T cells both in the periphery and at the tumor site. To identify enriched IDO/PD-L1-specific T cell clones, TCR rearrangements in sorted IDO/PD-L1-specific clones and TCR rearrangements in sorted IDO/PD-L1-specific T cell samples were compared to sequences from baseline PBMC samples for each patient. Clonal expansion of vaccine-specific TCR rearrangements from samples on vaccination were then tracked using a differential abundance framework. Cumulative IDO/PD-L1-specific T cell frequencies were tracked in post-treatment samples. We found no relation between clinical response and the enrichment of vaccine-specific clones, but an increase in IDO/PD-L1-specific T cell clones was observed at different time points in the periphery in all five patients (Extended Data Fig. 3c). We next investigated overall changes in the T cell repertoire in the blood. A modest increase in the peripheral T cell fraction was observed in the three responding patients at cycle 3, while the two non-responding patients had a clear decrease in T cell fraction (Extended Data Fig. 8a). We thereafter investigated TCR clonality and TCR repertoire richness, exploring the proportion of abundant clones and the number of unique rearrangements, respectively. A decreasing peripheral Simpson clonality and increasing TCR repertoire richness was observed in responding patients at cycle 3, which might indicate tumor trafficking upon treatment. The opposite pattern was observed in non-responding patients (Extended Data Fig. 8b,c). Peripherally expanded clones were associated with tumors and persisted until cycle 12 (latest time point analyzed). The largest peripheral expansion was observed at cycle 3, with the most significant increase observed in patient MM01 (CR). Responding patients had a larger fraction of peripherally expanded clones that were also found in tumors compared to that of non-responders. By tracking peripherally expanded clones detected at the tumor site, we observed that patient MM01 had a substantial increase after treatment, indicating tumor trafficking of peripheral expanded clones (Extended Data Fig. 8d). Influx at the tumor site of vaccine-enriched T cell clones Given the observation of increased T cell fraction and enrichment of IDO- and PD-L1-specific clones in the blood after treatment, we investigated whether the same trend was observed at the tumor site. Both TCR sequencing and immunohistochemistry (IHC) of paired biopsies from the five patients described above showed an increase in the T cell fraction with an influx of CD3+ and CD8+ T cells after treatment in the three responding patients (Fig. 5a–c). IHC was not possible for patient MM09 due to tissue loss.Fig. 5 Changes in the TME after treatment. Number of CD3+ and CD8+ T cells, TCR fraction, TCR clonality, TCR repertoire, biopsy expanded TCR clones and enrichment of IDO- and PD-L1-specific T cells at the tumor site. a, Number of CD3+ and CD8+ T cells per mm2 (sum in the validated area) at the tumor site detected by IHC of paired biopsies from four patients. b, Example of IHC of CD3+ and CD8+ T cells at the tumor site before and after treatment (cells per mm2) in one patient (MM01). c, T cell fraction at the tumor site at baseline and cycle 6 by TCR sequencing. The T cell fraction was calculated by taking the total number of T cell templates and dividing by the total number of nucleated cells. d, Tracking of vaccine-associated clones at baseline and cycle 6 in tumor biopsies. Cumulative frequencies of IDO and PD-L1 vaccine-specific TCR rearrangements are represented. e,f, TCR clonality and TCR repertoire richness in five patients at the tumor site at baseline and cycle 6. Simpson clonality measures how evenly TCR sequences are distributed among a set of T cells, where 0 indicates an even distribution of frequencies and 1 indicates an asymmetric distribution in which a few clones dominate. TCR repertoire richness reports the mean number of unique rearrangements. g, Bar chart representing baseline expanded biopsy clones from five patients (colored bars) and the detection of biopsy expanded clones also found in the blood at baseline and series 3, 6 and 12 (white and gray bars) by TCR sequencing. We thereafter investigated whether some of the IDO/PD-L1 vaccine-linked T cells were present at the tumor site. Vaccine-associated clones were tracked as the combined frequency of IDO- and PD-L1-specific T cell rearrangements. In biopsies, the frequencies at cycle 6 were compared to those at baseline and showed an increase in vaccine-specific T cells in four of five patients, irrespective of clinical response (Fig. 5d). TCR sequencing of PD-L1-specific SKILs (a more specific culture than IDO/PD-L1-specific isolations derived from PBMCs on vaccination) and paired biopsies showed that two of the top five PD-L1-specific SKIL clones were present at the tumor site both before and after treatment (Extended Data Fig. 7e,f). With a focus on the more abundant T cell clones, we investigated overall TCR clonality at the tumor site before and after treatment. In addition, we explored the number of unique TCR rearrangements, dissecting the lower-frequency clones. Patient MM01 had a significant increase in TCR clonality and a decrease in repertoire richness at the tumor site after therapy, indicating a focused tumor repertoire response of selected clones. All three responding patients had a decrease in TCR repertoire richness, which again might indicate a focused tumor response (Fig. 5e,f). Deeper analyses showed that the T cell clones that expanded at the tumor site after therapy were also present in the blood at baseline and increased significantly after treatment in four of five patients. The highest proportion was detected early at cycle 3. These data again support trafficking of peripherally expanded clones to the tumor site and could indicate that the T cell response to treatment is derived from pre-existing peripherally tumor-associated T cells (Fig. 5g). Signs of treatment-induced inflammation in the TME To dissect changes in the TME induced by T cell influx upon treatment in responding patients, RNA gene expression analyses using the nCounter PanCancer Immune Profiling Panel from NanoString were performed on paired biopsies from two responding patients (MM01 and MM13). Expression of genes related to adaptive immunity such as T cell activation, effector functions (genes encoding IFN-γ, TNF-α, IL-15 and IL-18) and cytotoxicity was increased in post-treatment biopsies (Extended Data Fig. 9a,b). Also, expression of genes related to checkpoint inhibitors such as those encoding T cell immunoglobulin and mucin-domain containing-3 (TIM-3), IDO, PD-L1, PD-L2, PD-1 and CTLA-4 increased after treatment, indicating activation of immune cells in the TME (Extended Data Fig. 9c). Additionally, IHC of paired biopsies from four patients (MM01, MM02, MM05 and MM13) showed an upregulation of PD-L1, IDO, MHC-I and MHC-II on tumor cells, indicating a treatment-induced pro-inflammatory response in the three responding patients, except for a decrease in MHC-II expression in patient MM13. By contrast, the non-responding patient MM02 had a reduction in T cell numbers present in the tumor after treatment and no expression of PD-L1, IDO or MHC-II and, interestingly, total loss of MHC-I, demonstrating tumor immune escape (Extended Data Fig. 10a). CD8+ T cells and their distance (µm) to PD-L1-expressing cells in baseline biopsies from five patients was investigated by IHC. Except for patient MM13 (PR), distance and clinical responses were associated: the two responders had reduced distance (<20 µm) between cells expressing these markers compared to non-responding patients (>80 µm). This observation indicates that responding patients not only have a higher intratumoral infiltration of CD8+ T cells, but that these cells can surround and attack PD-L1-expressing immune cells and tumor cells (Extended Data Fig. 10b). Discussion In this clinical trial, MM1636, 30 patients with metastatic melanoma were treated with a first-in-class immunomodulatory IDO/PD-L1-targeting peptide vaccine combined with nivolumab. The treatment led to an unprecedented high ORR of 80%, with 43% of patients reaching a CR, and a striking mPFS of 26 months (95% CI, 15.4–69 months) was reached. The vaccine represents a new treatment strategy to activate specific T cells that target cells contributing to immune suppression (including tumor cells), positively modulating the TME by inducing local inflammation. Indeed, we show that vaccine-specific T cells isolated and expanded from vaccinated patients recognize not only tumor cells in a target- and HLA-restricted manner but also myeloid cells polarized toward a tumor-associated phenotype. Hence, myeloid cells become targets for vaccine-activated T cells when they have a tumor-associated phenotype in the TME. These phenomena may further induce checkpoint molecules and rewire the TME toward an increasingly aPD1-permissive state. A drawback of our study is the single-center nonrandomized setup. Comparison between trials or between patients in trials and real-world patients is problematic due to multiple factors, such as period of conduction and other therapies available at the different periods of time. Nevertheless, the rate of investigator-assessed ORR in the phase 3 trial CheckMate 067 was 43.7% in the nivolumab monotherapy group and 57% in the nivolumab and ipilimumab group. CRs occurred in 8.9% and 11.5% of patients, respectively18. The mPFS of 26 months (95% CI, 15.4–69 months) in this trial is more than twice as long as that for patients treated with nivolumab and ipilimumab in CheckMate 067, for which an mPFS of 11.5 months (95% CI, 8.7–19.3 months) was reached. Patient baseline characteristics were in general comparable to those of patients with MM who have been treated in CheckMate 067, although patients in MM1636 were older (mean age, 70 years) and a larger fraction were positive for PD-L1 (57%)3,18,22. Among patients with PD-L1-negative tumors in MM1636, an ORR of 61.5% was still reached, which would be expected to be around 33% for first-line nivolumab monotherapy23. Some studies suggest that older patients might have a tendency to respond better to aPD1 therapy; however, this is still debated24–27. To address potential trial bias and the nonrandomized setup, patients in MM1636 were matched for age, performance status, sex, M stage, LDH level, PD-L1 status and BRAF status with a historical control group from the DAMMED, who were treated contemporarily (2015–2019) with aPD1 monotherapy as standard of care17. We found a significantly higher ORR and CRR in MM1636 compared to those of matched patients, who had an ORR of 43% and a CRR of 13%, comparable to patients treated in CheckMate 067. Restrictions of the synthetic control group are of course that it is partially historic and patient selection outside matching criteria cannot be ruled out28. Numerous contemporary clinical trials are exploring the combination of aPD1 therapy with other immunomodulating agents for advanced melanoma. Talimogene laherparepvec, an oncolytic virus, is approved by the Food and Drug Administration and the European Medicines Agency to treat advanced melanoma. A small phase 1b trial with 21 patients (MASTERKEY-265) combined talimogene laherparepvec and pembrolizumab to treat patients with advanced unresectable melanoma and reached an ORR of 62% and a CR of 33%29,30. Seventy-one percent of the patients in this trial had an M stage below M1c; this number was 40% in our trial. Furthermore, mainly patients with an M stage below M1c responded to treatment, which was not the case in MM1636. Results from a large randomized phase 3 trial are awaited (KEYNOTE-034). Epacadostat, an IDO inhibitor, was tested in combination with pembrolizumab in a nonrandomized phase 2 trial in 40 aPD1 treatment-naive patients with MM with promising results, reaching an ORR of 62%. Unfortunately, the phase 3 trial showed no indication that epacadostat provided improvement in PFS and OS31. Limitations of the phase 3 trial were the sparse information on pharmacodynamics as well as biomarker evaluation to improve the design. The IDO/PD-L1 vaccine is different from epacadostat, as it is not an IDO inhibitor but targets IDO- and PD-L1-expressing cells. Similar vaccines administered as monotherapy induced objective responses in lung cancer and basal cell carcinoma, while epacadostat as monotherapy in 52 patients resulted in no responses32,33. Sahin et al. recently published encouraging data from a first-in-human trial, in which a vaccine containing liposomal RNA targeting four unmutated tumor-associated antigens (NY-ESO-1, MAGE-A3, tyrosinase and TPTE) was administered alone or in combination with nivolumab in patients with advanced melanoma. Responses were observed in both the monotherapy group as well as the combination group in checkpoint inhibitor-experienced patients, suggesting the efficacy of non-mutant shared tumor antigen vaccines34. The overall safety and tolerability findings are comparable to those of aPD1 monotherapy. Injection site reactions were exclusive to the vaccine. However, these side effects were transient and mild in most patients and most likely due to the adjuvant Montanide. IDO- and PD-L1-specific CD8+ and CD4+ T cells exist among peripheral blood lymphocytes in healthy donors35–38 and expand in response to pro-inflammatory stimuli35,39. Furthermore, both IDO and PD-L1 are induced in cells as a counter response to the inflammatory response. This provides a mechanism that ensures immune homeostasis, which keeps IDO/PD-L1-specific T cells in check; therefore, the risk of triggering autoimmune-related adverse events by vaccination appears to be minimal. This was confirmed in the first clinical trials of IDO and PD-L1 vaccination (NCT01543464 (refs. 33,40) and NCT03042793 (ref. 41)). The spontaneous (baseline) immune response to the vaccines observed in the current study is in agreement with our previous observation in various patients with cancer35–38. Numerous vaccine-induced changes in the blood and at the tumor site were observed. Peripheral IDO- and/or PD-L1-specific T cells were detected in vitro in a modified ELISPOT in over 93% of patients on vaccination, unrelated to patient HLA type. Immune responses were persistent in patients who surpassed follow-up at data cutoff and were still detectable up to 6 months after the last vaccine, suggesting induction of memory T cells. We did not observe a correlation between vaccine-induced responses in blood and clinical responses. However, the detection of a highly significant increase in vaccine-specific T cell numbers after vaccination in almost all patients together with the very low number of patients with PD makes it difficult to expect such a correlation, especially because other aspects should be taken into account (for example, the loss of class I expression on tumors cells in patient MM02 with PD after vaccination). Frequencies of peripheral T cells induced against PD-L1 were overall higher than those against IDO. Importantly, however, we also observed that it was not the same patients who responded strongly to both IDO and PD-L1, nor did patients react with a similar response pattern to the two antigens; that is, different time points of response were observed. This suggests that each component of the vaccine plays different roles in the ongoing immune response in patients. The goal of IDO/PD-L1 vaccination was to modulate the TME to increase responsiveness to aPD1 therapy, as we have observed in animal models, in which immune conversion is demonstrated in the TME with an increased influx of T cells. Targeting both IDO and PD-L1 together enables synergy, as the TME is known to use different immune escape mechanisms and IDO and PD-L1 are often overexpressed by different cellular compartments. Ex vivo ELISPOT and flow cytometry assays confirmed the induction of immune response toward both epitopes and the higher immunogenicity of the PD-L1 epitope, although because of lower sensitivity failed to detect some of the responses. TCR sequencing of five patients confirmed enrichment of IDO/PD-L1-specific T cell clones in the blood at different time points after treatment. Furthermore, an increase in enriched IDO/PD-L1-specific clones was observed in four of five patients at the tumor site after treatment, irrespective of clinical response. Phenotypic characterization showed that vaccine-specific T cells, which were expanded in vitro with interleukin (IL)-2 from the blood of vaccinated individuals, were both CD4+ T cells and CD8+ T cells. This was confirmed by ex vivo phenotype description of vaccine-activated T cells. Vaccine-specific T cells expressed CD107a and CD137 and produced IFN-γ and TNF-α upon stimulation with the cognate target, indicating their cytolytic capacity. Overall, the data from immune monitoring supports the importance of both antigens in the generation of the frequent clinical responses in the study. Despite a limited number of paired biopsies (due to either patient refusal or the fact that a large fraction of responding patients had no assessable tumors after six cycles), we observed trends indicating treatment-induced general T cell influx in responding patients. It was shown that proliferation of CD8+ T cells in the tumor after aPD1 treatment is associated with radiographic reduction in tumor size42. Additionally, we showed that a large proportion of expanded peripheral TCR clones were associated with tumors and the most considerable amount of clonal expansion was observed early, at cycle 3. For the patient with a CR included in the TCR sequencing analyses (MM01), the number of peripheral expanded clones present at the tumor site increased after treatment compared to that at baseline, indicating tumor trafficking of peripheral expanded clones. Gene expression analyses (two paired biopsies) and IHC (five paired biopsies) further demonstrated that the combination treatment induced a pro-inflammatory TME in responding patients with signs of T cell activation and cytotoxicity and increased cytokine activity. This may lead to further upregulation of IDO, PD-L1, MHC-I and MHC-II on tumor cells, leading to more treatment targets. It was shown that, following vaccination with a cancer vaccine, PD-L1 expression is increased on tumor cells due to recruitment of tumor-specific T cells and upregulation of adaptive immune resistance pathways in the TME43. Treatment with nivolumab monotherapy enhances PD-L1 expression, and it is therefore problematic to discriminate the effect of the vaccine as compared to that of nivolumab44. IDO- and PD-L1-specific pro-inflammatory effector T cells were hypothesized to counteract the functions of IDO- and PD-L1-expressing immune-suppressive cells as a means to keep the immune balance between immune activation and inhibition45. However, because the expression of these molecules is induced as a counter-regulatory response to inflammatory mediators such as IFNs, they can also be expressed by, for example, activated myeloid cells or T cells. Thus, activation of PD-L1-specific T cells may result in depletion of activated T cells in the tumor and other sites. Importantly, it was recently described that PD-L1+ T cells mainly have tolerogenic effects on tumor immunity and exert tumor-promoting properties, suggesting that targeting this immune population is indeed also beneficial46. We have previously investigated the effect of activating PD-L1-specific T cells in vitro and in vivo and found that, overall, they supported the effector phase of an immune response by removing PD-L1-expressing regulatory immune cells that inhibit PD-1+ effector T cells47,48. The major role of the PD-1 pathway is believed to be the regulation of effector T cell responses. Thus, this protective pathway is more important after activation, rather than at the initial T cell-activation stage49. Accordingly, the presence of PD-L1-specific T cells during the activation phase of an immune response may not increase or support a pro-inflammatory immune response due to the expression of PD-L1 on potent antigen-presenting cells or on the T cells themselves14. Thus, the overall effects of PD-L1-specific T cells may vary depending on the expression of both PD-1 and PD-L1, that is, due to the microenvironment and the state of the immune response. In conclusion, here we report an impressive response rate, CR rate and mPFS for a first-in-class immune-modulating vaccine combined with nivolumab. This may be a first step toward a new treatment strategy for patients with MM. Limitations are the low number of patients treated at a single institution and the lack of a randomized design with aPD1 monotherapy as comparator. Studies in aPD1 therapy-resistant or -refractory melanoma are ongoing as well as biomarker analysis for selecting patients at a higher likelihood to benefit from the combination versus aPD1 monotherapy. A larger randomized trial will be essential to validate these findings and determine the specific contribution of the vaccine to clinical responses and changes in the TME. In December 2020, the Food and Drug Administration granted breakthrough therapy designation for the IO102/IO103 vaccine combined with aPD1 therapy in metastatic melanoma based on data from the MM1636 trial. Methods Trial design and treatment plan MM1636 is an investigator-initiated, nonrandomized, open-label, single-center phase 1/2 study. All patients were treated at the Department of Oncology, Herlev and Gentofte Hospital, University of Copenhagen, Herlev, Denmark. The study was conducted according to the Declaration of Helsinki and Good Clinical Practice (GCP) and monitored by the GCP unit in Copenhagen, Denmark. The protocol was approved by the ethical committee of the Capital region of Denmark (H-17000988), the Danish Medical Agencies (2017011073) and the Capital Region of Denmark Data Unit (P-2019-172). The study was registered at https://clinicaltrials.gov/ under the identifier NCT03047928 and at the EudraCT (no. 2016-0004527-23). This study initially aimed to include 30 aPD1 treatment-naive patients with MM. An amendment with the addition of two other cohorts with ten patients in each cohort was made to evaluate immune responses and clinical efficacy in aPD1 therapy-resistant patients (cohort B, de novo resistance and cohort C, acquired resistance) for a total of 50 patients. The amendment with cohorts B and C was approved at the inclusion of 18 patients in cohort A. The trial is still including patients in cohorts B and C. This article reports results from cohort A. The first six patients were treated as for phase 1, evaluating for safety and tolerability before the remaining 24 patients were included in phase 2. The IDO/PD-L1 vaccine was administered subcutaneously biweekly for the first 10 weeks and thereafter every 4 weeks for approximately 9 months. A maximum of 15 vaccines were administered. Nivolumab was administered according to the approved label (3 mg per kg biweekly) for 24 cycles. The 15th vaccine was administered together with the 24th nivolumab cycle of 3 mg per kg, and responding patients thereafter continued nivolumab monotherapy every 4th week (6 mg per kg) as a standard of care after investigator assessment. Treatment with nivolumab was discontinued at the maximum benefit (assessed by the investigator), after a maximum of 2 years of therapy, at progression or due to severe adverse events (Extended Data Fig. 1b, treatment plan). Vaccine composition Each vaccine was composed of 100 µg IO102, a 21-amino-acid peptide (DTLLKALLEIASCLEKALQVF) from IDO, and 100 µg IO103, a 19-amino-acid peptide (FMTYWHLLNAFTVTVPKDL) from the signal peptide of PD-L1 (PolyPeptide Laboratories). Peptides were dissolved separately in 50 µl DMSO, filtered for sterility and frozen at −20 °C (NUNC CyroTubes CryoLine System Internal Thread, Sigma-Aldrich). At <24 h before administration, peptides were thawed. The PD-L1 peptide was diluted in 400 µl sterile water and, immediately before injection, mixed with the IDO peptide solution and 500 µl Montanide ISA-51 (SEPPIC) to achieve a total volume of 1 ml. Patients Patients above 18 years of age with locally advanced or stage 4 melanoma according to the AJCC (seventh edition), at least one measurable lesion according to RECIST 1.1 and an ECOG PS of 0–1 were eligible. The main exclusion criteria were prior treatment with aPD1 therapy, CNS metastases >1 cm, severe comorbidities and active autoimmune disease. Enrollment was not restricted to PD-L1 status, but it was known before inclusion. Patients were included after informed consent. Patients MM42 and MM20 have confirmed their approval of PET–CT images and clinical images that have been included in the article. Key study assessments Safety and tolerability were evaluated based on changes in clinical laboratory analyses and reported adverse events. Adverse events were assessed according to Common Terminology Criteria for Adverse Events (version 5.0) and were graded from 1 to 5 for all treated patients up to 6 months after the last dose of the IDO/PD-L1 vaccine. Clinical efficacy was assessed using FDG PET–CT scans before treatment and every 3rd month until progression. Objective responses were categorized into CR, PR, stable disease or PD according to RECIST version 1.1. Clinical data were collected at CCIT-DK in the eCRF program OpenClinica version 1.0 and in Microsoft Excel version 2002 on a secure server. Blood samples for immunologic analyses were collected before treatment, before the third cycle, after the sixth, 12th, 18th and 24th cycles (on vaccination) and 3 and 6 months after the last vaccine. Two to three tumor needle biopsies (1.2 mm) were collected at baseline and after six cycles from the same tumor site, when accessible, to evaluate immune responses at the tumor site. DTH skin tests were performed and punch biopsies were taken from the DTH area after cycle 6 for the evaluation of SKILs reactive to PD-L1 and IDO (Supplementary Fig. 1, treatment plan). Statistical analysis of clinical outcome Survival curves were computed in GraphPad Prism software version 9.0.0 according to Kaplan–Meier method. Median follow-up time of enrollment was calculated using the reverse Kaplan–Meier method, also in GraphPad Prism software version 9.0.0. For binary outcomes, 95% two-sided CIs were constructed using the Clopper–Pearson method, also in GraphPad version 9.0.0. An independent board of certified and experienced oncoradiologists performed an external review to evaluate clinical response to address the potential bias of investigator site review. The external review took place at Rigshospitalet, Copenhagen University Hospital. This hospital did not participate in the MM1636 trial, and the external reviewer had no prior knowledge about the clinical trial or the trial therapy. Only PET–CT images were accessed, and arrows indicating target or non-target lesions appeared on baseline images as the only additional information. To address potential trial bias regarding treatment effect, we matched patients in MM1636 with patients from the DAMMED, a population-based database that retrospectively collects data on patients with metastatic melanoma in Denmark. Here, data from 938 patients treated with aPD1 monotherapy contemporaneously (January 2015–October 2019) were extracted. Two hundred and eighteen of these patients were eligible for comparison and matching (all parameters available) (Supplementary Table 1), and 74 patients from the DAMMED were found to match. Patients were matched for age (≤70 years, >70 years), sex, LDH levels (normal, elevated), M stage (M1a, M1b, M1c), BRAF status (wild type, mutated) and PD-L1 status (<1%, ≥1%). An exact matching algorithm was used in which patients in MM1636 were matched with patients from the DAMMED with the same combination of variables. Twenty-nine patients from MM1636 were matched with exact combinations of the six variables. One patient could not be matched. To secure balance of the calculations, control patients were weighted according to the number of patients for each MM1636 patient. Estimates for treatment effects were calculated by weighted logistic regression analyses and weighted Cox proportional hazard model. The R package ‘Matchlt’ was used for matching patients. As the method chosen for matching control patients to protocol patients, a weighted binary logistic regression model was used for comparing response rates in the two matched cohorts. Odds ratios and response rates, including their corresponding 95% CIs, were extracted from the regression models. All P values were two sided, and P values below 0.05 were considered statistically significant. SAS version 9.4M5 was used for the weighted logistic regression models. Processing of peripheral blood mononuclear cells Peripheral blood was collected from all patients in heparinized tubes and was processed within 4 h. In brief, PBMCs were isolated using Lymphoprep (Medinor) separation. PBMCs were counted on a Sysmex XP-300 analyzer and frozen in Human AB Serum (Sigma-Aldrich, H4522-100ML) with 10% DMSO using controlled-rate freezing (CoolCell, BioCision) in a −80 °C freezer and were moved the next day to a freezer at −140 °C until further processing. Needle biopsies at baseline and after the sixth vaccine Two to three needle biopsies (1.2 mm) were taken at baseline and after the sixth cycle of treatment when assessable from the same tumor lesion. One fragment was fixed with formalin and embedded in paraffin (FFPE); one to two fragments were used for expanding tumor-infiltrating T cells and establishing autologous tumor cell lines. Delayed-type hypersensitivity and generation of skin-infiltrating lymphocytes After six cycles of treatment, we performed intradermal injections of vaccine components without adjuvant and one control injection containing DMSO without peptide. Patients were injected with either a mixture of both IDO and PD-L1 peptides at all three injection sites or PD-L1 peptide, IDO peptide or a mixture of the two at the injection sites, respectively (Supplementary Table 6). Eight hours after injection, punch biopsies were resected from the three sites where the peptide was injected and transported immediately to the laboratory and cut into fragments of 1–2 mm3. SKILs were expanded to establish ‘young SKILs’ in CM medium consisting of RPMI 1640 with GlutaMAX, 25 mM HEPES, pH 7.2 (Gibco, 72400-021), IL-2 (100 or 6,000 IU ml−1) (Proleukin Novartis, 004184), 10% heat-inactivated Human AB Serum (Sigma-Aldrich, H4522-100ML), 100 U ml−1 penicillin, 1.25 µg ml−1 Fungizone (Bristol Myers Squibb, 49182) and 100 µg ml−1 streptomycin (Gibco, 15140-122). Half of the medium was replaced three times per week. Young SKILs from samples derived from IDO peptide, PD-L1 peptide and the mixture injection sites were further expanded in a small-scale version of the 14-d rapid expansion protocol as previously described51. Quantification of specific T cells in blood by ELISPOT For enumeration of vaccine-specific T cells in the peripheral blood, PBMCs from patients were stimulated with IDO or PD-L1 peptide in the presence of a low dose of IL-2 (120 U ml−1) for 7–13 d before being used in IFN-γ ELISPOT assays with 2.5–3.2 × 105 cells per well. Briefly, cells were placed in a 96-well PVDF membrane-bottomed ELISPOT plate (MultiScreen MSIPN4W50, Millipore) precoated with IFN-γ-capture antibody (clone 1-D1K, Mabtech). Diluted IDO or PD-L1 peptide stocks in DMSO were added at 5 µM; an equivalent amount of DMSO was added to control wells. PBMCs from each patient were set up in duplicate or triplicate for peptide and control stimulations. Cells were incubated in ELISPOT plates in the presence of the peptide for 16–18 h, after which they were washed off, and biotinylated secondary antibody (anti-human IFN-γ mAB, clone 7-B6-1, Mabtech) was added. After 2 h of incubation, unbound secondary antibody was washed off, and streptavidin-conjugated alkaline phosphatase (Mabtech) was added for 1 h. Next, unbound streptavidin-conjugated enzyme was washed off, and the assay was developed by adding BCIP/NBT substrate (Mabtech). ELISPOT plates were analyzed on the CTL ImmunoSpot S6 Ultimate V analyzer using ImmunoSpot software version 5.1. Responses were calculated as the difference between the average numbers of spots in wells stimulated with IDO or PD-L1 peptide and those from corresponding control wells. For detection of IDO and PD-L1 peptide responses ex vivo, PBMCs were thawed and rested overnight in medium containing DNase I (1 μg ml−1, Sigma-Aldrich, 11284932001) before being used in IFN-γ ELISPOT assays as described above. A total of 6–9 × 105 PBMCs per well were seeded. Statistical analysis of all ELISPOT responses was performed using the DFR method as described by Moodie et al. using RStudio software (RStudio Team, 2016, http://www.rstudio.com/)50. Vaccine-specific ELISPOT responses were defined as true if the difference between the spot count in control and peptide-stimulated wells was statistically significant according to DFR statistical analysis or, for samples performed in duplicate, if the spot count in peptide-stimulated wells was at least 2× the spot count in control wells50. Cancer cell lines and tumor conditioned medium Autologous melanoma cell lines were established from needle biopsies. Briefly, biopsies were chopped into small fragments and seeded in 24-well cultures in RPMI 1640 with GlutaMAX, 25 mM HEPES, pH 7.2 (Gibco, 72400-021), 10% heat-inactivated FBS (Life Technologies, 10500064), 100 U ml−1 penicillin, 1.25 µg ml−1 Fungizone (Bristol Myers Squibb, 49182) and 100 µg ml−1 streptomycin (Gibco, 15140-122). Established adherent melanoma tumor cell lines were cryopreserved at −140 °C in freezing medium containing FBS with 10% DMSO. PD-L1 and HLA-II expression on established tumor cell lines was assessed by flow cytometry staining with anti-PD-L1–PE-Cy7 (1:23 dilution, clone M1h1 (RUO), BD, 558017) and anti-HLA-II–FITC (1:23 dilution, clone Tu39 (RUO), BD, 555558) antibodies. To obtain TCM, established tumor cell lines were cultured in 175-cm2 Nunc cell culture flasks until 80–90% confluency was reached. The culture medium was then replaced with 20 ml fresh X-VIVO 15 with Gentamicin and Phenol Red (Lonza, BE02-060Q), medium with 5% heat-inactivated Human AB Serum (Sigma-Aldrich, H4522-100ML). After 24 h of incubation, TCM was collected and centrifuged to remove any resuspended cells, after which TCM was aliquoted, frozen and stored at −80 °C. The acute monocytic leukemia cell line MonoMac1 was obtained from the DSMZ—German Collection of Microorganisms and Cell Cultures (ACC 252) and cultured in RPMI 1640 with GlutaMAX, 25 mM HEPES, pH 7.2 (Gibco, 72400-021) and 10% heat-inactivated FBS. Isolation of autologous myeloid cells Autologous CD14+ cells were sorted from freshly thawed PBMCs using a magnetic bead-separation kit (Miltenyi Biotec, 130-050-201) according to the manufacturer’s instructions. Isolated CD14+ cells were used as targets in the IFN-γ ELISPOT assay directly after sorting or differentiated in vitro into tumor-associated macrophages by culturing with 1 ml fresh X-VIVO 15 medium with Gentamicin and Phenol Red (Lonza, BE02-060Q) and 5% heat-inactivated Human AB Serum, supplemented with 1 ml autologous TCM in 24-well plates for 2 d. Quantification of vaccine-specific T cells from DTH biopsy sites by ELISPOT SKILs expanded in vitro were rested in medium without IL-2 overnight before being used in the IFN-γ ELISPOT assay as described above to evaluate reactivity of skin-infiltrating T cells. Generation of IDO- and PD-L1-specific T cell cultures from PBMCs or SKILs IDO- or PD-L1-specific T cells were isolated from peptide-stimulated in vitro PBMC cultures on days 14–15 after stimulation or from SKIL cultures expanded in vitro. For specific T cell isolation, PBMCs or SKILs were stimulated with IDO or PD-L1 peptide, and cytokine-producing T cells were sorted using the IFN-γ or TNF-α Secretion Assay—Cell Enrichment and Detection kit (Miltenyi Biotec). Cytokine-production profile of PD-L1- and IDO-specific T cells by intracellular staining To assess the T cell cytokine-production profile, isolated and expanded IDO- and PD-L1-specific T cell cultures were stimulated for 5 h with peptide at 5 μM in a 96-well plate. One hour after the start of the incubation, anti-CD107a–PE (1:133 dilution, clone H4A3, BD Biosciences, 555801) antibody and BD GolgiPlug (1:1,000 dilution, BD Biosciences) were added. After a 5-h incubation, cells were stained using fluorescently labeled surface marker antibodies: anti-CD4–PerCP (1:5 dilution, clone SK3, 345770), anti-CD8–FITC (1:5 dilution, clone SK1, 345772) and anti-CD3–APC-H7 (1:20 dilution, clone SK7, 560275) (all from BD Biosciences). Dead cells were stained using FVS510 (1:250 dilution, BD Biosciences, 564406), followed by overnight fixation and permeabilization using eBioscience Fixation/Permeabilization buffers (00-5123-43, 00-5223-56) according to the manufacturer’s instructions. Cells were then stained intracellularly in eBioscience permeabilization buffer (00-8333-56) with anti-IFN-γ–APC (1:20 dilution, clone 25723.11, BD Biosciences, 341117) and anti-TNF-α–BV421 (1:100 dilution, clone Mab11, BD Biosciences, 562783) antibodies. Samples were analyzed on the FACSCanto II (BD Biosciences) using BD FACSDiva software version 8.0.2. The gating strategy is shown in Supplementary Fig. 3a. For assessing the reactivity against IDO expressing MonoMac1 cells, the appropriate amount of cancer cells was added to the IDO-specific T cells to obtain an effector:target ratio of 4:1 in the 96-well plate and T cell cytokine production was tested after 5 h of co-culture as described above. For HLA-blocking experiments, HLA-blocking antibodies (2 μg ml−1) were added directly into wells 30 min before the addition of peptide. Blocking antibodies used were HLA-DR (1:500 dilution, clone L243, Abcam, ab136320), HLA-DQ (1:500 dilution, clone SPV-L3, Abcam, ab23632) and HLA-DP (1:500 dilution, clone B7/21, Abcam, ab20897). To assess ex vivo T cell reactivity to IDO and PD-L1 peptides in patient PBMCs, cells were thawed and rested for 1–2 d in medium containing DNase I (1 μg ml−1, Sigma-Aldrich, 11284932001). PBMCs were then stimulated with peptide at 5 μM in a 96-well plate for 8 h. An hour after the addition of peptide, anti-CD107a–BV421 (3:50 dilution, clone H4A3, BD Biosciences, 328626) antibody and BD GolgiPlug (1:1,000 dilution, BD Biosciences) were added. Surface and intracellular staining was performed as described above. Antibodies used for surface staining were anti-CD3–PE-CF594 (0.8:30 dilution, clone UCHT1, BD Biosciences, 562280), anti-CD4–BV711 (1:30 dilution, clone SK3, BD Biosciences, 563028) and anti-CD8–Qdot605 (1:150 dilution, clone 3B5, Thermo Fisher, Q10009). Antibodies used for intracellular staining were anti-CD137–PE (1:20 dilution, clone 4B4-1 (RUO), BD Biosciences, 555956), anti-IFN-γ–PE-Cy7 (1.5:20 dilution, clone B27, BD Biosciences, 557643) and anti-TNF-α–APC (1:20 dilution, clone Mab11, BD Biosciences, 554514). Samples were acquired on the NovoCyte Quanteon (ACEA Biosciences) and analyzed using NovoExpress software version 1.4.1. To assess vaccine-specific T cell responses, background values observed in unstimulated PBMC samples were subtracted from values observed in peptide-stimulated conditions. Positive response value threshold was set at a difference of 0.2% from the background values. Based on this response cutoff, only TNF-α, CD107a and CD137 responses were detected in this assay. Statistical analysis comparing baseline with on-treatment or post-treatment cytokine profiles was performed using two-sided Wilcoxon matched-pairs signed-rank test. The gating strategy is shown in Supplementary Fig. 3b. siRNA-mediated PD-L1 and IDO1 silencing A stealth siRNA duplex for targeted silencing of PD-L1 (Invitrogen)52, a custom silencer select siRNA for targeted silencing of IDO1 (Ambion) and the recommended silencer select negative control (Ambion) siRNA for mock transfection were used. The stealth PD-L1 siRNA duplex consisted of the sense sequence 5′-CCUACUGGCAUU-UGCUGAACGCAUU-3′ and the antisense sequence 5′-AAUGCGUUCAGCAAAUGCCAGUAGG-3′. The three silencer IDO siRNA duplexes used were siRNA1 (sense sequence, 5′-ACAUCUGCCUGAUCUCAUATT-3′; antisense sequence, 5′-UAUGAGAUCAGGCAGAUGUTT-3′), siRNA2 (sense sequence, 5′-CCACGAUCAUGUGAACCCATT-3′; antisense sequence, 5′-UGGGUUCACAUGAUCG-UGGAT-3′) and siRNA3 (sense sequence, 5′-CGAUCAUGUGAACCCAAAATT-3′; antisense sequence, 5′-UUUUGGGUUCACAUGAUCGTG-3′). For PD-L1 or IDO1 silencing, cancer cells were electroporated with 0.025 nmol of each siRNA duplex as previously described53. For PD-L1-silencing experiments, cancer cells were treated with IFN-γ (500 U ml−1, PeproTech) 1 h after electroporation. Electroporated cells were used as target cells in ELISPOT and ICS assays 24 h or 48 h after siRNA electroporation. RT–qPCR Total RNA was extracted using the RNeasy Plus Mini kit (Qiagen, 74134) following the manufacturer’s instructions. RNA concentration was quantified using a NanoDrop 2000 (Thermo Fisher Scientific), and a total of 1,000 ng RNA was reverse transcribed with the High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems, 4368814) using 1,000 ng input RNA for transcription. Real-time qPCR analyses were performed using the TaqMan Gene Expression Assay on a Roche LightCycler 480 instrument. RT–qPCR assays were performed with a minimum of three technical replicates and analyzed using the ΔCt method as described in Bookout et al.54 with normalization of IDO1 expression (primer ID, Hs00984148_m1) or PD-L1 expression (primer ID, Hs001125296_m1) to the expression level of the housekeeping gene POLR2A (primer ID, Hs00172187_m1) and the control sample (mock). For low-concentration samples with no amplification, Ct was set to 40. Controls without reverse transcriptase were used as controls for specific amplifications. P values were determined using two-tailed parametric t-tests. BRAF-mutational status and PD-L1 status at baseline in all patients A library of historical FFPE biopsies were assessable for all patients and analyzed locally at Herlev and Gentofte University Hospital by experienced pathologists for BRAF status and PD-L1 expression on tumor cells. The BRAF analysis was carried out with real-time PCR using the EntroGen BRAF Mutation Analysis Kit II (BRAFX-RT64, CE-IVD) to specifically detect mutations corresponding to V600D, V600E and V600K in the BRAF protein. PD-L1 status was assessed using the monoclonal rabbit anti-PD-L1 antibody (clone 28.8, PD-L1 IHC 28-8 pharmDx) in FFPE biopsies following the manufacturer’s instructions. Patients were considered positive for PD-L1 with expression levels ≥1% and negative for PD-L1 with expression levels <1%. Human leukocyte antigen type Blood samples from all 30 patients were genotyped for class I (HLA-A, HLA-B, HLA-C) and three class II types (HLA-DRB1, HLA-DQA1, HLA-DQB1) using LinkSēq HLA Typing Kits (Thermo Fisher, 1580C). These test kits are based on real-time PCR using allele-specific exponential amplification (sequence-specific primers) followed by melting curve analyses. Immunohistochemistry simplex: Immunoscore CR IHC staining was performed using a qualified Ventana Benchmark XT with four different steps: (1) antigen retrieval; (2) staining with the following primary antibodies: anti-CD3 antibody (clone HDX2; provider, HalioDx; HD-FG-000013; 10931065/10636667; concentration, 0.25 µg ml−1), anti-CD8 antibody (clone HDX3; provider, HalioDx; HD-FG-000019; 10931069/10337710/10639301; concentration, 0.5 µg ml−1), anti-IDO monoclonal antibody (clone VINC3IDO; provider, Thermo Fisher Scientific; 14-9750-82; E25003-101; concentration, 0.05 µg ml−1), anti-HLA class 1 ABC antibody (clone, EMR8-5; provider, Abcam; ab70328; GR3248333/GR3186494; concentration, 0.5 µg ml−1), anti-HLA-DR/DP/DQ/DX antibody (clone CR3/43; provider, Santa Cruz; sc-53302; L1714; concentration, 1 µg ml−1) and anti-PD-L1 antibody (clone HDX3; provider, HalioDx; HD-FG-000035; 106312810/106312816; concentration, 3.3 µg ml−1); (3) detection with a secondary antibody using the ultraView Universal DAB Detection kit; and (4) counterstaining using Hematoxylin & Bluing Reagent (staining of cellular nuclei). Control slides were systematically included in each staining run to permit quality control of the obtained measurements. Following coverslipping, slides were scanned with the NanoZoomer-XR to generate digital images (20×). Two consecutive slides were specifically used to perform Immunoscore CR TL staining of CD3+ and CD8+ cells. Digital pathology of T lymphocytes The digital pathology for Immunoscore CR TL allowed the quantification of positive cells (in cells per mm2) into the core tumor and invasive margin if present. Each sample was analyzed using the HalioDx Digital Pathology Platform. Digital pathology and pathologist analysis of IDO Digital pathology for IDO allowed the quantification of stained area (in mm2) into the whole tumor. Each sample was analyzed using the HalioDx Digital Pathology Platform. Analysis of IDO+ cells was performed by a pathologist. Results were expressed as an H score from 0 to 300. The score is obtained by the formula 3 × percentage of cells with strong staining + 2 × percentage of cells with moderate staining + percentage of cells with weak staining. Digital pathologist analysis of MHC-I and MHC-II Analysis of MHC-I+ and MHC-II+ cells was performed by a pathologist. Results are expressed as a percentage of positive tumor cells and a percentage of positive cells in the stroma. Digital pathology Immunoscore immune checkpoint (CD8+ and PD-L1+) The Immunoscore CR IC test allowed the quantification of CD8++ cell density in the whole tumor and a CD8+-centered proximity index (which corresponds to the percentage of CD8++ cells that have at least one PD-L1+ cell in the neighborhood) at different cutoff distances (20 µm, 40 µm, 60 µm and 80 µm). Pathologist analysis of PD-L1 A pathologist performed analysis of PD-L1+ cells. Positivity of a viable tumor cell was considered when partial or complete cell membrane staining was observed (more than 10% of the tumor cell membrane). Results were expressed as a percentage. NanoString RNA profiling and Immunosign RNA was extracted from FFPE tissues using Qiagen RNeasy FFPE extraction kits (Qiagen). Annotations from the pathologist performing H&E staining were used to guide removal of normal tissue from the slides by macrodissection before nucleic acid extraction, which occurred after tissue deparaffinization and lysis. Each extracted RNA sample was independently quantified using a NanoDrop spectrophotometer (NanoDrop Technologies) and qualified (Agilent Bioanalyzer). Degradation was quantified as the percentage of RNA fragments smaller than 300 bp using the RNA 6000 Nano kit (Agilent Bioanalyzer). Good sample quality was defined as less than 50% of RNA fragments being 50–300 bp in size. RNA expression profiling was performed using the nCounter PanCancer Immune Profiling Panel from NanoString (NanoString Technologies). The PanCancer Immune Profiling Panel contains 776 probes and is supplemented with six genes to complete HalioDx Immunosign targets. Hybridization was performed according to the manufacturer’s instructions. Hybridized probes were then purified and immobilized on a streptavidin-coated cartridge using the nCounter Prep Station (NanoString Technologies). Data collection was carried out on the nCounter Digital Analyzer (NanoString Technologies) following the manufacturer’s instructions to count individual fluorescent barcodes and quantify target RNA molecules present in each sample. For each assay, a scan of 490 fields of view was performed. Raw data from the NanoString nCounter were processed using NanoString recommendations. Quality control enables to retain data of good quality with a binding density that ranges between 0.05 and 2.25. The linearity of positive controls was checked using the R2 value of regression between the counts and the concentrations of positive controls. Samples that showed R2 < 0.75 were flagged and removed from the analysis. The background was removed using a thresholding method at the mean + 2 s.d. of negative controls. Raw counts were normalized using a positive normalization factor. Samples showing positive normalization factors outside the range of 0.3–3 were removed from the analysis. A second normalization was performed using the housekeeping gene normalization factor. Only the most stable housekeeping genes were selected for this normalization step using the variance-versus-mean relationship. All samples showing a normalization factor outside the range of 0.1–10 were removed from the analysis. All statistical analyses were performed on normalized counts using R software (version 2.6.2, 2019-12-12). T cell receptor variable β chain sequencing To track longitudinal immune responses to therapy, genomic DNA was extracted from longitudinal pre- and post-treatment PBMCs (five patients), pre- and post-treatment biopsies (five patients) (both FFPE and treated with RNAlater) and IDO- and PD-L1-specific T cell cultures from PBMCs (five patients) or SKILs (one patient). Three clonal (two IDO- and one PD-L1-) specific cultures were also sequenced to confirm clonal purity of cultures. DNA from PBMCs or RNAlater-treated biopsies was extracted with the DNeasy Blood and Tissue kit (Qiagen, 69504), DNA from sorted IDO and PD-L1-specific T cells from either PBMCs or SKILs was extracted using the QIAamp DNA Micro kit (Qiagen, 565304), and DNA from PFFE biopsies was extracted using the Maxwell RSC DNA FFPE kit (Promega, AS1450). Immunosequencing of the CDR3 regions of human TCRβ chains was performed using the immunoSEQÒ Assay (Adaptive Biotechnologies). Extracted genomic DNA was amplified in a bias-controlled multiplex PCR, followed by high-throughput sequencing. Sequences were collapsed and filtered to identify and quantitate the absolute abundance of each unique TCRβ CDR3 region for further analysis as previously described55–57. Statistical analyses of TCRβ sequencing results Two quantitative components of diversity were compared across samples in this study. First, Simpson clonality was calculated on productive rearrangements by ∑i=1Rpi2, where R is the total number of rearrangements and pi is the productive frequency of rearrangement i. Values of Simpson clonality range from 0 to 1 and measure how evenly receptor sequences (rearrangements) are distributed among a set of T cells. Clonality values approaching 0 indicate a very even distribution of frequencies, whereas values approaching 1 indicate an increasingly asymmetric distribution in which a few clones are present at high frequencies. Second, sample richness was calculated as the number of unique productive rearrangements in a sample after computationally downsampling to a common number of T cells to control for variation in sample depth or T cell fraction. Repertoires were randomly sampled without replacement five times, and we report the mean number of unique rearrangements. The T cell fraction was calculated by taking the total number of T cell templates and dividing by the total number of nucleated cells. The total number of nucleated cells was derived from reference genes using the immunoSEQ Analyzer version 3. To identify enriched vaccine clones in each patient, rearrangement frequencies in their baseline PBMCs and each IDO/PD-L1-sorted T cell sample were compared using a binomial distribution framework as previously described58. In brief, for each clone, we performed a two-sided test to determine whether frequencies were the same in the patient’s periphery and a PD-L1- or IDO-specific T cell sample. The Benjamini–Hochberg procedure was used to control the false discovery rate at 0.01 (ref. 59). Clonal expansion in post-treatment samples was similarly assessed using this differential abundance framework, but an IDO/PD-L1-specific T cell sample was replaced with a post-treatment series sample. In biopsies, the six series frequencies were compared to those of baseline tissue. Lastly, vaccine-associated clones were tracked in each PBMC and tissue sample by summing the frequency of each rearrangement enriched in either PD-L1- or IDO-specific T cells. All statistical analyses were performed in R version 3.6.1. Reporting Summary Further information on research design is available in the Nature Research Reporting Summary linked to this article. Online content Any methods, additional references, Nature Research reporting summaries, source data, extended data, supplementary information, acknowledgements, peer review information; details of author contributions and competing interests; and statements of data and code availability are available at 10.1038/s41591-021-01544-x. Supplementary information Supplementary Information Supplementary Figs. 1–3 and Tables 1–6 Reporting Summary Extended data Extended Data Fig. 1 Mode of Action and treatment plan. a) Anticipated mechanism of action of the combination therapy of an IDO/PD-L1 derived peptide vaccine and nivolumab (anti-PD1). 1) The IDO/PD-L1 peptide vaccine is administered subcutaneously (s.c) and nivolumab is administered intravenous (IV). 2) The vaccines peptides are phagocytosed by an antigen presenting cell and presented to IDO and PD-L1 specific T cells, which are activated. 3) The activated T cells migrates to the tumor site where they attack both immune-suppressive cells and tumour cells expressing IDO and/or PD-L1 leading to cytokine production and a pro-inflammatory tumour microenvironment. 4/5) Enhanced tumour killing by both IDO/PD-L1 specific T cells and tumor specific cytotoxic T cells due to PD-1 blockade. Created with BioRender.com b) Treatment plan. After written informed content patients were screened. Before treatment start a baseline PET/CT scan was performed, baseline blood sample for research use and if assessable a baseline needle biopsy. Patients were treated with the IDO/PD-L1 peptide vaccine subcutaneously biweekly for the first 6 injections and thereafter every fourth week for a maximum of 15 vaccines. Nivolumab was administered in parallel biweekly (3mg/kg) up to 24 series. If patients needed subsequent nivolumab after ended vaccination regiment they were treated with 6 mg/kg every fourth week up to two years. Needle biopsy and delayed type hypersensitivity (DTH) was performed after 6 series of treatment if assessable. PET/CT scans was performed every third month. Extended Data Fig. 2 Progression free survival and overall survival in matched historical control group and vaccine injection site reaction. a) Kaplan-Meier curve of progression free survival with corresponding 95% confidence intervals in the matched historical control group (n=74). Patients were matched on BRAF-status, PD-L1-status, age, gender, M-stage and LDH level). b) Kaplan-Meier curve of overall survival with corresponding 95% confidence intervals in the matched historical control group (n=74). c) Images of injection site reaction in patient MM20 (CR) after 11 vaccination show redness, rash and granuloma at injection site. Extended Data Fig. 3 Vaccine specific responses in blood. a) Heatmap of specific (background has been subtracted) IDO and PD-L1 responses in PBMCs at baseline, series 3, 6, 12, 18 and 24 measured by in vitro IFN-γ Elispot assay show fluctuations in the blood during treatment (n=30). b) Heatmap of specific (background has been subtracted) IDO and PD-L1 responses in PBMCs at baseline and 3 and 6 months after last vaccine measured by in vitro IFN- γ Elispot assay (n=30). 2.5–3.2x105 cells per well were used. c) Vaccine associated clones were tracked in the blood by summing the frequency of each rearrangement enriched in either IDO or PD-L1 T cells. Extended Data Fig. 4 Ex vivo vaccine specific responses in blood. Heatmaps of detected specific (background has been subtracted) IDO (a) and PD-L1 (b) responses in PBMCs at baseline and on/after treatment as measured by IFNγ ELISPOT (n=25). C) Example of well images of ex vivo ELISPOT wells for three different patients (n=3). 6-9x105 cells per well were used. Extended Data Fig. 5 CD4 and CD8 vaccine specific T cell responses in blood. Top: heatmaps of IDO specific CD4 (left) and CD8 (right) T cell responses in PBMCs at baseline and on/post treatment. Bottom: heatmaps of PD-L1 specific CD4 (left) and CD8 (right) T cell responses. Responses quantified by flow cytometry by an increased expression of CD107α, CD137 and TNFα after 8h peptide stimulation. Values represent specific responses after background values have been subtracted (n=21). Statistical analysis were performed using two-sided Wilcoxon matched-paired rank test. Extended Data Fig. 6 Pro-inflammatory profiles of sorted CD4 and CD8 T cells from blood. a/b/c/d) Percentage of in vitro stimulated and sorted PD-L1 and IDO specific CD4 and CD8 T cells secreting cytokines. Extended Data Fig. 7 Vaccine specific responses in skin. a) IDO and PD-L1 specific T cell responses in SKILs after 6 series of treatment measured by IFN-γ Elispot assay (n=13). SKILs were grown from DTH injection with either IDO peptide, PD-L1 peptide or a mix as presented by different blue colours. * Responses were calculated as the difference between average numbers of spots in wells stimulated with IDO or PD-L1 peptide (triplicates) and corresponding control (DMSO) and statistical analyses of Elispot responses were performed using distribution-free resampling method (Moodie et al.). DR: Not statistically confirmed response due to replicate number but number of spots in peptide wells are two times higher than control wells (DMSO). NS: No significant response and no DR. b/c/d) Percentage of cytokine secreting/CD107a+ CD4+ and CD8+ IDO and PD-L1 specific T cells in response to in vitro peptide stimulation by flow cytometry. e) Skin infiltrating PD-L1 specific T cell clones also found in biopsy in patient MM01. TCR sequencing was performed on a PD-L1 specific T cell culture generated from DTH area on the lower back injected with PD-L1 peptide on patient MM01. Bars show the frequency of top 25 clones in the culture with which indicates a high Simpson clonality of 0.43. f) Tracking the frequency of the top five skin infiltrating PD-L1 specific clones in tumour before and after treatment. Extended Data Fig. 8 T cell changes in blood after treatment. T cell fraction, TCR clonality and repertoire richness in blood and peripherally expanded TCR clones in both responding and non-responding patients. a) T cell fraction in peripheral blood of five patients at baseline, series 3, 6 and 12 by TCR sequencing. T cell fraction was calculated by taking the total number of T cell templates and dividing by the total number of nucleated cells (n=5.) b) TCR clonality in peripheral blood of five patients at baseline, series 3, 6 and 12 by TCR sequencing (n=5). Simpson clonality measures how evenly TCR sequences are distributed amongst a set of T cells where 0 indicate even distribution of frequencies and 1 indicate an asymmetric distribution where a few clones dominate. c) TCR repertoire richness in peripheral blood of five patients at baseline, series 3, 6 and 12 by TCR sequencing (n=5). TCR repertoire richness report the mean number of unique rearrangements. d) Bar-chart chart representing dynamics of the expanded T cell clones. Coloured bars represent peripherally expanded clones in five patients at series 3, series 6 and series 12 as compared to the baseline PBMC samples (n=5). Light gray bar represents peripherally expanded clones that are present in baseline biopsy samples while dark gray bar represents peripherally expanded clones that are present in the post-treatment biopsy (after 6 series). e) Frequency of the dominant TCR β chain in clonal PD-L1 and IDO specific cultures as determined by CDR3 sequencing. Extended Data Fig. 9 Profiling of genes relevant to T cell activation, cytokines and exhaustion markers on pre- and post-treatment biopsies in two patients. a) RNA expression profiling of genes related to T cell activation was performed using NanoString nCounter (n=2). b) RNA expression profiling of genes related to cytokine activity was performed using NanoString nCounter (n=2). c) RNA expression profiling of genes related to checkpoint inhibitors was performed using NanoString nCounter (n=2). Extended Data Fig. 10 Treatment induced upregulation of PD-L1, IDO, MHC I and MHC II and distance between CD8 T cells and PD-L1 expressing cells. a) IHC on 4 paired biopsies stained for CD3 and CD8 T cells, PD-L1, MHC I and MHC II on tumor cells/mm2 (sum in the validated area) and IDO H-score (from 0 to 300). H-score is the expression of IDO on both immune and tumor cells: The score is obtained by the formula: 3 x percentage of cells with a strong staining + 2 x percentage of cells with a moderate staining + percentage of cells with a weak staining (n=4). b) Distance in µm between CD8+ T cells and PD-L1+ stained cells on five baseline biopsies detected by IHC (n=5). Extended data is available for this paper at 10.1038/s41591-021-01544-x. Supplementary information The online version contains supplementary material available at 10.1038/s41591-021-01544-x. Acknowledgements We thank all patients and their relatives for being a part of the trial. We thank Ö. Met, M. Jonassen, S. Ullitz Færch, B. Saxild, S. Wendt and C. Grønhøj for technical support. We thank M. Cumberbatch for input with translational analysis. We thank the nurses at clinic 5 and the head of the Oncology Department at Herlev and Gentofte Hospital, L. Sengeløv. The study was funded through a research funding agreement between IO Biotech and the CCIT-DK, Herlev Hospital and the Oncology Department at Herlev Hospital. Author contributions I.M.S. and M.H.A. conceived the study. J.W.K. wrote the protocol with input from I.M.S., M.H.A. and M.-B.Z. J.W.K. and C.L.L. were responsible for patient treatment and patient care with help from R.B.H. and consulting from I.M.S., E. Ellebaek and M.D. I.M.S., E. Ellebaek and M.D. were responsible for patient recruitment. J.W.K. and C.L.L. were responsible for coordinating trial procedures and collecting data and samples. J.W.K. and C.L.L. were clinical investigators, and I.M.S. was a sponsor. J.W.K. and C.L.L. analyzed and interpreted clinical data with support from I.M.S. and E. Ehrnrooth. T.W.K. performed statistical analyses. H.W.H. helped with clinical imaging. J.W.K., E.M., C.O.M., S.M.A., S.E.W.-B., M.O.H. and A.W.P. analyzed and interpreted translational data with support from M.H.A. and I.M.S. Funding was acquired by I.M.S., M.H.A. and M.-B.Z. J.W.K. wrote the manuscript with input from I.M.S., M.H.A., E.M. and A.W.P. All authors reviewed the manuscript, interpreted data and approved the final version. Data availability This clinical trial was registered at https://www.clinicaltrials.gov/ before patient enrollment (clinical trial identifier NCT03047928). Each vaccine was composed of 100 µg IO102, a 21-amino-acid peptide (DTLLKALLEIASCLEKALQVF) from the IDO peptide, and 100 µg IO103, a 19-amino-acid peptide (FMTYWHLLNAFTVTVPKDL) from the signal peptide of PD-L1. TCR sequencing data are available from Adaptive Biotechnologies. Upon request, the CCIT-DK office will provide a username and a password to access the designated data within approximately 2–4 weeks (https://www.herlevhospital.dk/ccit-denmark/find-us/Sider/Contact-information.aspx). All requests for the remaining data including raw data and analyzed data and materials will, within a reasonable time frame, be reviewed by the CCIT-DK office (https://www.herlevhospital.dk/ccit-denmark/find-us/Sider/Contact-information.aspx) to verify whether the request is subject to any intellectual property or obligations. Patient-related data not included in the paper were generated as part of clinical trials and may be subject to patient confidentiality. Any data and materials that can be shared will be released via a material-transfer agreement. The following database was used in the study: https://research.regionh.dk/da/publications/the-danish-metastatic-melanoma-database-dammed(32749d99-095f-4cae-b5de-769bae27f01e).html. Competing interests M.H.A. is named as an inventor on various patent applications relating to therapeutic uses of IDO and PD-L1 peptides. These patent applications are assigned to the company IO Biotech, which is developing immune-modulating cancer treatments. M.H.A. is a founder, shareholder and advisor for IO Biotech. I.M.S. is a cofounder, shareholder and advisor for IO Biotech. I.M.S. has an advisory board relationship with or lectured for Roche, Novartis, MSD, Celgene, Incyte, TILT Bio, Pfizer and BMS AstraZeneca and has received limited grants for translational research from BMS, Roche and Novartis. M.-B.Z. is the CEO, founder and shareholder at IO Biotech. E.M., A.W.P. and E. Ehrnrooth are employees at IO Biotech. E. Ellebaek has received honoraria from BMS, Pierre Fabre, Roche and Kyowa Kirin and travel support from MSD. M.D. has received honoraria from Genzyme, MSD, BMS, Roche and Novartis and travel support from Novartis, MSD, BMS, Roche and Pfizer. The remaining authors declare that they have no conflict of interest. Peer review information Nature Medicine thanks George Coukos, Michael Schell and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. Javier Carmona was the primary editor on this article and managed its editorial process and peer review in collaboration with the rest of the editorial team. Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. These authors contributed equally: Julie Westerlin Kjeldsen, Cathrine Lund Lorentzen. Change history 3/8/2022 A Correction to this paper has been published: 10.1038/s41591-022-01771-w

Fatal

ReactionOutcome

CC BY

34887574

2021-12-09

What was the outcome of reaction 'Multiple organ dysfunction syndrome'?

A phase 1/2 trial of an immune-modulatory vaccine against IDO/PD-L1 in combination with nivolumab in metastatic melanoma. Anti-programmed death (PD)-1 (aPD1) therapy is an effective treatment for metastatic melanoma (MM); however, over 50% of patients progress due to resistance. We tested a first-in-class immune-modulatory vaccine (IO102/IO103) against indoleamine 2,3-dioxygenase (IDO) and PD ligand 1 (PD-L1), targeting immunosuppressive cells and tumor cells expressing IDO and/or PD-L1 (IDO/PD-L1), combined with nivolumab. Thirty aPD1 therapy-naive patients with MM were treated in a phase 1/2 study ( https://clinicaltrials.gov/ , NCT03047928). The primary endpoint was feasibility and safety; the systemic toxicity profile was comparable to that of nivolumab monotherapy. Secondary endpoints were efficacy and immunogenicity; an objective response rate (ORR) of 80% (confidence interval (CI), 62.7-90.5%) was reached, with 43% (CI, 27.4-60.8%) complete responses. After a median follow-up of 22.9 months, the median progression-free survival (PFS) was 26 months (CI, 15.4-69 months). Median overall survival (OS) was not reached. Vaccine-specific responses assessed in vitro were detected in the blood of >93% of patients during vaccination. Vaccine-reactive T cells comprised CD4+ and CD8+ T cells with activity against IDO- and PD-L1-expressing cancer and immune cells. T cell influx of peripherally expanded T cells into tumor sites was observed in responding patients, and general enrichment of IDO- and PD-L1-specific clones after treatment was documented. These clinical efficacy and favorable safety data support further validation in a larger randomized trial to confirm the clinical potential of this immunomodulating approach. pmcMain Despite remarkable advances in the treatment of MM with immune checkpoint inhibitors (ICIs) targeting PD-1 and cytotoxic T lymphocyte antigen 4 (CTLA-4), around half of patients are resistant to ICI monotherapy1. The combination of anti-CTLA-4 (aCTLA-4) and aPD1 therapy is to date the most effective therapy resulting in a response rate of around 60%; however, 50% of the patients also develop severe adverse events2,3. Therefore, an equally effective but less toxic treatment is highly needed. Several approaches to enhance ICI efficacy are currently being investigated, such as other ICIs, T cell therapy with tumor-infiltrating T cells or innate immunity stimulators such as Toll-like receptor 9 agonists4–6. Treating cancer patients with vaccines that stimulate a targeted immune response is another attractive approach, with very few side effects observed thus far in combination immunotherapy studies7,8. Immune-modulatory vaccines targeting tumoral immune escape mechanisms offer a new, generalizable strategy compared to patient-specific neoantigen cancer vaccines7,9. The immune-modulatory vaccine strategy in this clinical trial is based on the finding of circulating cytotoxic T cells specific to IDO and PD-L1 in the blood of patients with cancer and, to a lesser extent, in healthy donors. IDO- and PD-L1-specific CD8+ T cells can directly recognize and kill IDO+ and/or PD-L1+ tumor cells and likewise recognize and kill non-malignant cells that express their cognate targets. Furthermore, IDO- and PD-L1-specific CD4+ T cells release pro-inflammatory cytokines in response to IDO- or PD-L1-expressing target cells. IDO and PD-L1 are expressed not only by melanoma cells but also by many other cell types in the tumor microenvironment (TME), which differentiates these antigens from traditional tumor antigens used in other studies10–15. Activation of IDO/PD-L1-specific T cells by vaccination can therefore restrict the range of immunosuppressive signals mediated by immunosuppressive cells and thereby revert the TME from an immune hostile to an immune friendly environment. In animal models of cancer, vaccinations with IDO epitopes resulted in anti-tumor therapeutic effects that were correlated with reductions in IDO expression in myeloid cell populations within the TME16. The IDO/PD-L1 immune-modulating vaccine may lead to a translatable strategy for improving the efficacy of aPD1 therapy through activation of specific T cells. We hypothesize that the IDO/PD-L1 vaccine attracts T cells into the tumor, which induces type 1 helper T (TH1) cell inflammation and reverts the TME into an immune-permissive site, thereby turning the tumor ‘hot’. This would also upregulate PD-L1 expression in cancer and immune cells, generating more susceptible targets to aPD1 therapy (Extended Data Fig. 1a). This theory was confirmed in a mouse model in which aPD1 therapy and IDO vaccination show synergistic effects16. In this phase 1/2 clinical trial, MM1636, patients with MM received a combination of the IDO/PD-L1 (IO102/IO103) peptide vaccine with the adjuvant Montanide and the aPD1 antibody nivolumab. Patients were included in three cohorts: 30 aPD1 therapy-naive patients (cohort A), ten aPD1 therapy-refractory patients (cohort B, de novo resistance) and ten patients who progressed after aPD1 therapy (cohort C, acquired resistance). Here, we report results from cohort A. The vaccine was given biweekly for the first six administrations and thereafter every 4th week. A maximum of 15 vaccines were administered. Nivolumab was given in parallel, biweekly (3 mg per kg) or every 4th week (6 mg per kg) for up to 2 years (Extended Data Fig. 1b). The primary objective was safety and feasibility. Secondary objectives were immunogenicity and clinical efficacy. Results Patients and treatment Thirty patients were enrolled from December 2017 to June 2020. None of the 30 patients dropped out of the study; all received at least three cycles of therapy (Supplementary Fig. 1). At the current database lock (5 October 2020), six patients were still on treatment in the trial. Of the 24 patients who were not on trial treatment at data cutoff, two are still receiving nivolumab monotherapy (6 mg per kg every 4 weeks). Reasons for stopping treatment for the remaining 22 patients included disease progression (37%), toxicity (20%), maximum benefit or complete response (CR) confirmed on two consecutive scans (17%) or completing 2 years of treatment (7%). For the 24 patients who were not on trial treatment at data cutoff, the mean number of vaccinations was 10.5 (range, 3–15 vaccinations). Thirteen of these 24 patients continued nivolumab (6 mg per kg, every 4 weeks) as a standard of care. Nine patients received subsequent therapy after progression (Supplementary Table 1). Baseline characteristics are shown in Table 1. The mean age was 70 years; 37% of patients had elevated lactate dehydrogenase (LDH) levels, 60% were stage M1c, 37% had BRAF mutations, and 43% were negative for PD-L1 (<1%). A total of three patients (10%) had received prior ipilimumab therapy. No patients had brain metastasis (Supplementary Table 2).Table 1 Baseline patient characteristics (n = 30) Characteristic Number (%) Mean age, years (range) 70 (46–85) Sex, male 16 (55%) ECOG PS 0 26 (87%) LDH levels  ≤ULN 19 (63%)  >ULN 11 (37%) M stage (AJCC-8)  M1a 6 (20%)  M1b 6 (20%)  M1c 18 (60%) Number of lesion sites  1 6 (20%)  2–3 17 (57%)  >3 7 (23%) Liver metastases present 10 (33%) BRAF status  Mutant 11 (37%)  Wild type 19 (63%) PD-L1  <1% 13 (43%)  >1% 17 (57%) Previous systemic therapy  Ipilimumab 3 (10%)  No 27 (90%) AJCC-8, eighth edition of the American Joint Committee on Cancer; ECOG PS, Eastern Cooperative Oncology Group performance status; ULN, upper limit of normal. Notable clinical responses to the combination therapy Thirty patients with MM were treated with the IDO/PD-L1 vaccine and nivolumab according to the trial protocol. By investigator review, the ORR reached 80% (CI, 62.7–90.5%), with 43% of patients (CI, 27.4–60.8%) achieving a CR and 37% (CI, 20.9–54.5%) reaching a partial response (PR) as the best overall response, while 20% experienced progressive disease (PD) according to Response Evaluation Criteria in Solid Tumors (RECIST) 1.1 (Fig. 1a). Two of the patients with a PR did not have a confirmatory scan stating PR on two consecutive scans. Early onset of response was frequent, with 22 of 30 patients having an objective response at the first evaluation (after 12 weeks on treatment). Median times to PR and CR were 75 d (range, 54–256 d) and 327 d (range, 73–490 d), respectively (Fig. 2a–c).Fig. 1 Clinical response. a, Pie charts with percent ORR, CR, PR and PD according to RECIST 1.1 by investigator review of all patients (n = 30), PD-L1+ patients (>1%, (n = 17)) and PD-L1− patients (<1%, n = 13)), respectively. Two-sided CIs (95%) were constructed using the Clopper–Pearson method. b, Treatment effect in MM1636 compared with a matched historical control group from the DAMMED database (n = 74). Patients in MM1636 (n = 29) were matched with the exact same combination variable according to age (≤70 years, >70 years), sex, LDH levels (normal, elevated), M stage (M1a, M1b, M1c), BRAF status (wild type, mutated) and PD-L1 status (<1%, ≥1%). Estimates for treatment effects were calculated by weighted logistic regression analyses and weighted Cox proportional hazard model. Bar height indicates the estimated response rate; tops of bars are centers for error bars. Odds ratios (OR), response rates and their corresponding 95% CIs were extracted from the regression model. All P values were two sided, and P values below 0.05 were considered statistically significant. c, Best change in the sum of target lesion size compared with that at baseline (n = 30). The horizontal line at −30 shows the threshold for defining an objective response in the absence of non-target disease progression or new lesions according to RECIST 1.1. Two patients with 100% reduction in target lesion size had non-target lesions present. White stars, six patients had normalization (<10 mm) of fluor-18-deoxyglucose (FDG)-negative lymph nodes (at baseline, lymph nodes were >1.5 cm and FDG+) and 100% reduction of non-lymph node lesions and are considered to have had a CR (green bar). Black star, one patient (MM29) was considered to have had a PR (blue bar), although he had not reached a −30% change in target lesion size. The patient had a single measurable 13-mm lung metastasis at baseline and multiple biopsy-verified cutaneous metastases on the left crus (not detectable by positron-emission tomography–computed tomography (PET–CT) at baseline). The best change in target lesion size was 10 mm, and a post-treatment biopsy from the cutaneous metastases showed no sign of malignancy. Thus, overall, the patient was classified as having a PR. d, Kaplan–Meier curve of the response duration in the 24 patients with an objective response. e, Kaplan–Meier curve of PFS in all 30 treated patients. f, Kaplan–Meier curve of OS in all 30 treated patients. Fig. 2 Clinical response. a, Swimmer plot showing response duration and time to response according to RECIST 1.1 for all treated patients (n = 30). Triangles indicate first evidence of PR, while squares indicate first evidence of CR. Closed circles indicate time of progression. Arrows indicate ongoing responses. Patient MM18 died due to nivolumab-induced side effects. b, Spider plot showing response kinetics in all treated patients (n = 30). Red squares indicate time of progression. c, PET–CT images of patient MM42 before and after treatment (after 12 series of treatment) showing FDG metabolism in target lesions. SC, subcutaneous. The ORR among PD-L1+ (>1% (clone 28.8)) patients (n = 17) was 94.1% (CI, 73–99.7%) and 61.5% (CI, 35.5–82.3%) in PD-L1− patients (n = 13) (Fig. 1a). Objective responses were observed in patients irrespective of human leukocyte antigen (HLA) genotype (Supplementary Fig. 2). Clinical response data were validated by blinded independent external review, in which an ORR of 76.6% (CI, 57.7–90.1%) was reported, with 53.3% of patients achieving a CR, 23.3% achieving a PR and 3.3% experiencing stable disease. Comparisons between investigator review and external review are outlined in Supplementary Table 3. To examine whether the very high response rate should be attributed to nivolumab or to the vaccine, we retrieved clinical response information from a matched historical control group using the Danish Metastatic Melanoma Database (DAMMED) from contemporaneously treated patients with stage 3–4 melanoma who received aPD1 monotherapy17. Patients were matched with the exact same combination variable according to age, sex, PD-L1 status, BRAF status, LDH level and M stage (those at stage M1d were excluded from the control group (no patients with brain metastasis)). Matched controls were identified for 29 patients, and the ORR of 79.3% (CI, 61.0–90.4%) observed in MM1636 was found to be significantly higher (P < 0.0012) than that in the matched control group in which an ORR of 41.7% (CI, 31.0–53.3%) was reached. Furthermore, of the 29 patients in MM1636, a significantly (P < 0.0017) higher percentage (41.4% (CI, 25.2–59.6%)) of patients achieved CR in MM1636 than that (12% (CI, 6.3–21.6%)) in the matched historical control group. ORR and complete response rate (CRR) in the matched historical control group were comparable to those of patients treated in randomized phase 3 pivotal trials with aPD1 monotherapy18 (Fig. 1b). The treatment leads to prolonged PFS At data cutoff, the median duration of response had not been reached, with 87% of all responding patients being free from progression at 12 months (Fig. 1d). Patients were followed for up to 35 months with a median follow-up time of 22.9 months (CI, 14.9–26.2 months). OS and PFS were calculated from the first day of treatment to death or progression or to the date of the last follow-up (5 October 2020). The median PFS (mPFS) for all treated patients was 26 months (CI, 15.4–69 months) and was not reached for responding patients (Fig. 1e). The median OS was not reached at the data cutoff. OS at 12 months was 81.6% (CI, 61.6–92%) (Fig. 1f). One patient (MM18) with a CR died from nivolumab-related severe adverse events; the remaining patients died because of metastatic melanoma. For comparison, the mPFS was 8.3 months (CI, 5.5 months–NR (not reached)) in the matched historical control group (n = 74), while the median OS was 23.2 months (CI, 23.2 months–NR). (Extended Data Fig. 2a,b). The combination of the IDO/PD-L1 vaccine and nivolumab was safe Treatment-related adverse events are listed for all 30 patients in Supplementary Table 4. Common treatment-related grade 1–2 toxicities were fatigue (47%), rash (47%), arthralgia (30%), diarrhea (23%), nausea (23%), dry skin (20%), pruritus (20%), infusion reaction (17%), xerostomia (17%) and myalgia (17%). Four patients (13%) experienced grade 3–4 adverse events: one patient with a grade 3 maculopapular rash (MM01), one patient with grade 3 adrenal insufficiency (MM06) and one patient with grade 3 arthralgia (MM22). Patient MM18 died from urosepsis with multi-organ failure and severe hyponatremia. This patient had experienced multiple immune-related adverse events with grade 3 colitis, grade 2 pneumonitis, grade 3 arthralgia, grade 2 vasculitis and grade 2 nivolumab infusion-related allergic reaction. Additionally, patient MM18 had symptoms of myocarditis at the time of death with highly elevated cardiac troponin I levels. Bedside echocardiography showed an ejection fraction of 15%, which at baseline was 60%, but an autopsy was not conducted, and myocarditis was never confirmed pathologically. Patient MM06 had received first-line treatment with ipilimumab before entering the trial and was on substitution corticosteroids at the time of inclusion. Adrenal insufficiency was aggravated by an erysipelas infection with high fever, reaching grade 3 in Common Terminology Criteria for Adverse Events and resolving quickly after appropriate antibiotic therapy was initiated. As expected, local side effects were common with 77% of the patients who developed injection site reactions. These reactions were classified as granulomas (63%), redness (20%), pain (13%) and pruritus (13%) at the injection site. All local reactions were grade 1–2, most likely related to the Montanide adjuvant and typically transient. However, two patients (MM07 and MM20) decided to discontinue vaccination after eight and 11 injections, respectively, due to granulomas, tenderness and pain that limited instrumental activities of daily living but continued to receive nivolumab (Extended Data Fig. 2c). Vaccine-specific responses in blood were frequently detected First, all 30 patients were assessed for the presence of vaccine-specific responses in peripheral blood mononuclear cells (PBMCs) before, on and after vaccination using a modified interferon (IFN)-γ enzyme-linked immune absorbent spot (ELISPOT) assay. This assay is known to expand antigen-specific memory cells and to improve the detection and the correlative power of ELISPOT during treatment19–21. Prevaccine IDO-specific responses were detectable in ten (33%) patients, while prevaccine PD-L1-specific responses were detectable in eight patients (27%); overlapping (specific to both IDO and PD-L1) prevaccine responses were present in four (13.3%) patients. During vaccination, an increase of IDO-specific T cells or PD-L1-specific T cells in the blood was observed in 28 (93%) and 26 (86%) patients, respectively. In total, 93% of patients had an increase in either PD-L1- or IDO-specific responses on vaccination (Fig. 3a), with a significant (P < 0.0001) median increase from baseline to the post-vaccine response demonstrated for both IDO and PD-L1 (at different time points on treatment), confirming that vaccine-specific immune responses were induced in patients regardless of clinical response (Fig. 3b,c). Immune responses fluctuated in the blood over time (Extended Data Fig. 3a). An increase in IDO- and PD-L1-specific responses in the peripheral blood was also detectable directly ex vivo across the different clinical response groups (Extended Data Fig. 4).Fig. 3 Vaccine-specific responses in blood. a, IDO- and PD-L1-specific T cell responses in PBMCs at baseline and on vaccination as measured by the IFN-γ ELISPOT assay (n = 30). *Responses were calculated as the difference between the average numbers of spots in wells stimulated with IDO or PD-L1 peptide (triplicates) and those from the corresponding control (DMSO), and statistical analyses of ELISPOT responses were performed using a distribution-free resampling method (Moodie et al.50). DR (double response), response was not statistically confirmed due to replicate number, but the number of spots in peptide wells was two times higher than that in control wells (DMSO). NS, no significant response and no DR. For a detailed overview of responses at serial time points on vaccination, see Extended Data Fig. 4a. b, IDO- and PD-L1-specific T cell response in PBMCs in all treated patients measured by the IFN-γ ELISPOT assay at baseline and on vaccination. On-vaccination responses were selected from the ‘best’ ELISPOT response at different time points for each patient (series 3, 6, 12, 18 or 24) during vaccination (n = 30). Wilcoxon matched-pairs signed-rank test was used to compare responses to IDO or PD-L1 peptides in the vaccine between baseline and later time points. c, Representative example of ELISPOT wells with response in patient MM23 in serial PBMCs before and on treatment. d, IDO-specific CD4+ and CD8+ T cells were isolated and expanded from PBMCs stimulated in vitro with the IDO peptide and a low dose of IL-2 for 14–15 d before sorting using the Miltenyi Cytokine Secretion Assay—Cell Enrichment and Detection kit. To assess their cytolytic potential, IDO-specific T cells were stimulated with IDO peptide, and expression of CD107a, IFN-γ and TNF-α was assessed by flow cytometry (the example is from patient MM14). Sustained vaccine-specific responses were observed 3 and 6 months after the last vaccine, indicating induction of memory responses in nine patients with response to clinical treatment who surpassed follow-up at data lock (Extended Data Fig. 3b). Importantly, PD-L1- and IDO-specific responses were observed irrespective of HLA genotype (Supplementary Table 5). To verify the functionality of vaccine-induced T cells, IDO- or PD-L1-specific T cells were isolated and in vitro expanded from PBMCs of five patients. Phenotypic characterization by flow cytometry revealed that the isolated vaccine-specific T cells consisted of both CD4+ and CD8+ T cells. Also, both IDO- and PD-L1-specific CD4+ and CD8+ T cells showed pro-inflammatory properties, as they expressed the cytolytic marker CD107a and secreted the cytokines IFN-γ and tumor necrosis factor (TNF)-α (Fig. 3d and Extended Data Fig. 6a–d). Interestingly, we were also able to detect vaccine-specific CD4+ and CD8+ T cell responses in peripheral blood ex vivo (Extended Data Figs. 4 and 5). We observed a significant increase in the overall percentage of CD107a, CD137 and TNF-α expression in response to peptide stimulation in on- or post-treatment PBMC samples compared to that at baseline, further confirming the expansion and the diverse signature of vaccine-specific T cells (Extended Data Fig. 5). Vaccine-specific responses detected in the skin at the vaccination site To investigate whether vaccine-specific T cells have the potential to migrate to peripheral tissue, delayed-type hypersensitivity (DTH) tests were performed after six cycles of treatment on 15 patients to assess the presence of vaccine-reactive T cells in the skin. Supplementary Table 6 display an overview of skin-infiltrating lymphocyte (SKIL) cultures. We detected IDO-specific T cells in the skin of six of ten patients and PD-L1-specific T cells in nine of 11 patients (Extended Data Fig. 7a). Intracellular cytokine staining was performed on SKILs from five patients after stimulation with either PD-L1 or IDO peptide. Here, we detected mainly CD4+ peptide-reactive T cells that secreted TNF-α and upregulated CD107a, and a minor fraction also secreted IFN-γ. In one patient, we detected CD8+ PD-L1-reactive T cells (Extended Data Fig. 7b–d). Vaccine-induced T cells specifically recognize target cells To confirm the functionality of vaccine-expanded T cells, vaccine-specific T cell clones (clonal purity was confirmed by T cell receptor (TCR) sequencing) were isolated and expanded from patient PBMCs (Extended Data Fig. 8e). We showed that PD-L1-specific T cells were able to recognize PD-L1+ autologous tumor cells in a PD-L1 expression-dependent manner if the cancer cells also expressed HLA-II (Fig. 4a,b). Similarly, an HLA-DR-restricted IDO-specific CD4+ T cell clone was able to recognize an HLA-DR-matched IDO-expressing model cell line, MonoMac1, in an IDO-expression-dependent manner (Fig. 4e–g). As previously described, IDO- and PD-L1-specific T cells’ mode of action is not limited to targeting only cancer cells. We were able to show that vaccine-specific T cell clones also reacted against PD-L1- and IDO-expressing autologous immune cells (Fig. 4c,h). To provide myeloid cells with a tumor-associated phenotype we treated isolated CD14+ myeloid cells with tumor conditioned medium (TCM) derived from an established autologous tumor cell line. We observed that such TCM-treated CD14+ cells had increased expression of PD-L1 and IDO and were effectively recognized by autologous PD-L1- and IDO-specific CD4+ T cell clones (Fig. 4c,d,h,i).Fig. 4 PD-L1- and IDO-specific T cells from vaccinated patients react against PD-L1- and IDO-expressing target cells. a, Left, PD-L1-specific T cell culture (MM1636.05) reactivity against PD-L1 peptide or autologous tumor cells in the IFN-γ ELISPOT assay. Tumor cells were either not treated or pretreated with 200 U ml−1 IFN-γ for 48 h before the assay. Effector:target (E:T) ratio of 10:1 was used. Right, PD-L1 and HLA-II surface expression on melanoma cells with (green) or without (yellow) pretreatment with IFN-γ compared to an isotype control (gray) as assessed by flow cytometry. b, Left, PD-L1-specific T cell (MM1636.05) reactivity in the IFN-γ ELISPOT assay against autologous tumor cells pretreated with IFN-γ (500 U ml−1) and transfected with mock or PD-L1 small interfering (si)RNA 24 h after transfection. E:T ratio, 10:1. Right, PD-L1 surface expression on melanoma tumor cells (MM1636.05) assessed by flow cytometry 24 h after transfection with mock (blue) or PD-L1 (red) siRNA compared to the isotype control (gray). c, Reactivity of the CD4+ PD-L1-specific T cell clone (MM1636.14) against PD-L1 peptide or autologous CD14+ cells; E:T ratio, 10:1. CD14+ cells were isolated using magnetic bead sorting and used as targets in an ELISPOT assay directly or after pretreatment for 2 d with TCM derived from the autologous tumor cell line. d, Quantitative PCR with reverse transcription (RT–qPCR) analysis of PD-L1 (CD274) expression in sorted CD14+ cells before and after treatment with autologous TCM for 48 h. e, Reactivity of the IDO-specific CD4+ T cell clone (MM1636.23) against IDO peptide combined with HLA-DR (L243)-, HLA-DQ (SPV-L3)- or HLA-DP (B7/21)-blocking antibodies (aHLA-DR, aHLA-DQ and aHLA-DP) in an intracellular staining assay (ICS) for IFN-γ and TNF-α production. T cells were incubated with individual blocking antibodies (2 μg ml−1) for 30 min before adding IDO peptide. f, Reactivity of the IDO-specific CD4+ T cell clone (MM1636.23) against the HLA-DR-matched IDO-expressing cell line MonoMac1 transfected with mock or IDO siRNA in an ICS assay, E.T ratio, 4:1. siRNA transfection was performed 48 h before the experiment. g, RT–qPCR analysis of IDO1 expression in MonoMac1 cells 48 h after siRNA transfection. h, Reactivity of the CD4+ IDO-specific T cell clone (MM1636.14) against IDO peptide or autologous CD14+ cells; E:T ratio, 20:1. CD14+ cells were isolated using magnetic bead sorting and used as targets in an ELISPOT assay directly or after pretreatment with TCM derived from the autologous tumor cell line. i, RT–qPCR analysis of IDO1 expression in sorted CD14+ cells before and after treatment with autologous TCM for 48 h. Bars in RT–qPCR data (d,g,i) represent the mean of three (d,i) or six (g) technical replicates ±s.d.; P values were determined by two-tailed parametric t-tests. ELISPOT counts (a,b,c,h) represent the mean value of three technical replicates ±s.e.m.; response P values were determined using the distribution-free resampling (DFR) method. TNTC, too numerous to count. T cell clones in blood and tumors To track the role of treatment-induced T cell responses, TCR sequencing of the complementarity-determining region 3 (CDR3) was performed on five patients in peripheral blood (baseline and cycles 3, 6 and 12) and paired biopsies. These five patients (MM01, MM02, MM08, MM09 and MM13) were selected due to the availability of material and to investigate a balanced patient group with both responders and non-responders. Due to a limited number of available paired biopsies (either as a consequence of patient refusal or unexpectedly rapid and substantial clinical responses), statistical analysis could not be applied. Details on clinical response are shown in Fig. 2a. Additionally, PBMCs (on treatment) or SKILs were stimulated with the IDO/PD-L1 peptides, and then cytokine-producing T cells were sorted to track vaccine-induced T cells both in the periphery and at the tumor site. To identify enriched IDO/PD-L1-specific T cell clones, TCR rearrangements in sorted IDO/PD-L1-specific clones and TCR rearrangements in sorted IDO/PD-L1-specific T cell samples were compared to sequences from baseline PBMC samples for each patient. Clonal expansion of vaccine-specific TCR rearrangements from samples on vaccination were then tracked using a differential abundance framework. Cumulative IDO/PD-L1-specific T cell frequencies were tracked in post-treatment samples. We found no relation between clinical response and the enrichment of vaccine-specific clones, but an increase in IDO/PD-L1-specific T cell clones was observed at different time points in the periphery in all five patients (Extended Data Fig. 3c). We next investigated overall changes in the T cell repertoire in the blood. A modest increase in the peripheral T cell fraction was observed in the three responding patients at cycle 3, while the two non-responding patients had a clear decrease in T cell fraction (Extended Data Fig. 8a). We thereafter investigated TCR clonality and TCR repertoire richness, exploring the proportion of abundant clones and the number of unique rearrangements, respectively. A decreasing peripheral Simpson clonality and increasing TCR repertoire richness was observed in responding patients at cycle 3, which might indicate tumor trafficking upon treatment. The opposite pattern was observed in non-responding patients (Extended Data Fig. 8b,c). Peripherally expanded clones were associated with tumors and persisted until cycle 12 (latest time point analyzed). The largest peripheral expansion was observed at cycle 3, with the most significant increase observed in patient MM01 (CR). Responding patients had a larger fraction of peripherally expanded clones that were also found in tumors compared to that of non-responders. By tracking peripherally expanded clones detected at the tumor site, we observed that patient MM01 had a substantial increase after treatment, indicating tumor trafficking of peripheral expanded clones (Extended Data Fig. 8d). Influx at the tumor site of vaccine-enriched T cell clones Given the observation of increased T cell fraction and enrichment of IDO- and PD-L1-specific clones in the blood after treatment, we investigated whether the same trend was observed at the tumor site. Both TCR sequencing and immunohistochemistry (IHC) of paired biopsies from the five patients described above showed an increase in the T cell fraction with an influx of CD3+ and CD8+ T cells after treatment in the three responding patients (Fig. 5a–c). IHC was not possible for patient MM09 due to tissue loss.Fig. 5 Changes in the TME after treatment. Number of CD3+ and CD8+ T cells, TCR fraction, TCR clonality, TCR repertoire, biopsy expanded TCR clones and enrichment of IDO- and PD-L1-specific T cells at the tumor site. a, Number of CD3+ and CD8+ T cells per mm2 (sum in the validated area) at the tumor site detected by IHC of paired biopsies from four patients. b, Example of IHC of CD3+ and CD8+ T cells at the tumor site before and after treatment (cells per mm2) in one patient (MM01). c, T cell fraction at the tumor site at baseline and cycle 6 by TCR sequencing. The T cell fraction was calculated by taking the total number of T cell templates and dividing by the total number of nucleated cells. d, Tracking of vaccine-associated clones at baseline and cycle 6 in tumor biopsies. Cumulative frequencies of IDO and PD-L1 vaccine-specific TCR rearrangements are represented. e,f, TCR clonality and TCR repertoire richness in five patients at the tumor site at baseline and cycle 6. Simpson clonality measures how evenly TCR sequences are distributed among a set of T cells, where 0 indicates an even distribution of frequencies and 1 indicates an asymmetric distribution in which a few clones dominate. TCR repertoire richness reports the mean number of unique rearrangements. g, Bar chart representing baseline expanded biopsy clones from five patients (colored bars) and the detection of biopsy expanded clones also found in the blood at baseline and series 3, 6 and 12 (white and gray bars) by TCR sequencing. We thereafter investigated whether some of the IDO/PD-L1 vaccine-linked T cells were present at the tumor site. Vaccine-associated clones were tracked as the combined frequency of IDO- and PD-L1-specific T cell rearrangements. In biopsies, the frequencies at cycle 6 were compared to those at baseline and showed an increase in vaccine-specific T cells in four of five patients, irrespective of clinical response (Fig. 5d). TCR sequencing of PD-L1-specific SKILs (a more specific culture than IDO/PD-L1-specific isolations derived from PBMCs on vaccination) and paired biopsies showed that two of the top five PD-L1-specific SKIL clones were present at the tumor site both before and after treatment (Extended Data Fig. 7e,f). With a focus on the more abundant T cell clones, we investigated overall TCR clonality at the tumor site before and after treatment. In addition, we explored the number of unique TCR rearrangements, dissecting the lower-frequency clones. Patient MM01 had a significant increase in TCR clonality and a decrease in repertoire richness at the tumor site after therapy, indicating a focused tumor repertoire response of selected clones. All three responding patients had a decrease in TCR repertoire richness, which again might indicate a focused tumor response (Fig. 5e,f). Deeper analyses showed that the T cell clones that expanded at the tumor site after therapy were also present in the blood at baseline and increased significantly after treatment in four of five patients. The highest proportion was detected early at cycle 3. These data again support trafficking of peripherally expanded clones to the tumor site and could indicate that the T cell response to treatment is derived from pre-existing peripherally tumor-associated T cells (Fig. 5g). Signs of treatment-induced inflammation in the TME To dissect changes in the TME induced by T cell influx upon treatment in responding patients, RNA gene expression analyses using the nCounter PanCancer Immune Profiling Panel from NanoString were performed on paired biopsies from two responding patients (MM01 and MM13). Expression of genes related to adaptive immunity such as T cell activation, effector functions (genes encoding IFN-γ, TNF-α, IL-15 and IL-18) and cytotoxicity was increased in post-treatment biopsies (Extended Data Fig. 9a,b). Also, expression of genes related to checkpoint inhibitors such as those encoding T cell immunoglobulin and mucin-domain containing-3 (TIM-3), IDO, PD-L1, PD-L2, PD-1 and CTLA-4 increased after treatment, indicating activation of immune cells in the TME (Extended Data Fig. 9c). Additionally, IHC of paired biopsies from four patients (MM01, MM02, MM05 and MM13) showed an upregulation of PD-L1, IDO, MHC-I and MHC-II on tumor cells, indicating a treatment-induced pro-inflammatory response in the three responding patients, except for a decrease in MHC-II expression in patient MM13. By contrast, the non-responding patient MM02 had a reduction in T cell numbers present in the tumor after treatment and no expression of PD-L1, IDO or MHC-II and, interestingly, total loss of MHC-I, demonstrating tumor immune escape (Extended Data Fig. 10a). CD8+ T cells and their distance (µm) to PD-L1-expressing cells in baseline biopsies from five patients was investigated by IHC. Except for patient MM13 (PR), distance and clinical responses were associated: the two responders had reduced distance (<20 µm) between cells expressing these markers compared to non-responding patients (>80 µm). This observation indicates that responding patients not only have a higher intratumoral infiltration of CD8+ T cells, but that these cells can surround and attack PD-L1-expressing immune cells and tumor cells (Extended Data Fig. 10b). Discussion In this clinical trial, MM1636, 30 patients with metastatic melanoma were treated with a first-in-class immunomodulatory IDO/PD-L1-targeting peptide vaccine combined with nivolumab. The treatment led to an unprecedented high ORR of 80%, with 43% of patients reaching a CR, and a striking mPFS of 26 months (95% CI, 15.4–69 months) was reached. The vaccine represents a new treatment strategy to activate specific T cells that target cells contributing to immune suppression (including tumor cells), positively modulating the TME by inducing local inflammation. Indeed, we show that vaccine-specific T cells isolated and expanded from vaccinated patients recognize not only tumor cells in a target- and HLA-restricted manner but also myeloid cells polarized toward a tumor-associated phenotype. Hence, myeloid cells become targets for vaccine-activated T cells when they have a tumor-associated phenotype in the TME. These phenomena may further induce checkpoint molecules and rewire the TME toward an increasingly aPD1-permissive state. A drawback of our study is the single-center nonrandomized setup. Comparison between trials or between patients in trials and real-world patients is problematic due to multiple factors, such as period of conduction and other therapies available at the different periods of time. Nevertheless, the rate of investigator-assessed ORR in the phase 3 trial CheckMate 067 was 43.7% in the nivolumab monotherapy group and 57% in the nivolumab and ipilimumab group. CRs occurred in 8.9% and 11.5% of patients, respectively18. The mPFS of 26 months (95% CI, 15.4–69 months) in this trial is more than twice as long as that for patients treated with nivolumab and ipilimumab in CheckMate 067, for which an mPFS of 11.5 months (95% CI, 8.7–19.3 months) was reached. Patient baseline characteristics were in general comparable to those of patients with MM who have been treated in CheckMate 067, although patients in MM1636 were older (mean age, 70 years) and a larger fraction were positive for PD-L1 (57%)3,18,22. Among patients with PD-L1-negative tumors in MM1636, an ORR of 61.5% was still reached, which would be expected to be around 33% for first-line nivolumab monotherapy23. Some studies suggest that older patients might have a tendency to respond better to aPD1 therapy; however, this is still debated24–27. To address potential trial bias and the nonrandomized setup, patients in MM1636 were matched for age, performance status, sex, M stage, LDH level, PD-L1 status and BRAF status with a historical control group from the DAMMED, who were treated contemporarily (2015–2019) with aPD1 monotherapy as standard of care17. We found a significantly higher ORR and CRR in MM1636 compared to those of matched patients, who had an ORR of 43% and a CRR of 13%, comparable to patients treated in CheckMate 067. Restrictions of the synthetic control group are of course that it is partially historic and patient selection outside matching criteria cannot be ruled out28. Numerous contemporary clinical trials are exploring the combination of aPD1 therapy with other immunomodulating agents for advanced melanoma. Talimogene laherparepvec, an oncolytic virus, is approved by the Food and Drug Administration and the European Medicines Agency to treat advanced melanoma. A small phase 1b trial with 21 patients (MASTERKEY-265) combined talimogene laherparepvec and pembrolizumab to treat patients with advanced unresectable melanoma and reached an ORR of 62% and a CR of 33%29,30. Seventy-one percent of the patients in this trial had an M stage below M1c; this number was 40% in our trial. Furthermore, mainly patients with an M stage below M1c responded to treatment, which was not the case in MM1636. Results from a large randomized phase 3 trial are awaited (KEYNOTE-034). Epacadostat, an IDO inhibitor, was tested in combination with pembrolizumab in a nonrandomized phase 2 trial in 40 aPD1 treatment-naive patients with MM with promising results, reaching an ORR of 62%. Unfortunately, the phase 3 trial showed no indication that epacadostat provided improvement in PFS and OS31. Limitations of the phase 3 trial were the sparse information on pharmacodynamics as well as biomarker evaluation to improve the design. The IDO/PD-L1 vaccine is different from epacadostat, as it is not an IDO inhibitor but targets IDO- and PD-L1-expressing cells. Similar vaccines administered as monotherapy induced objective responses in lung cancer and basal cell carcinoma, while epacadostat as monotherapy in 52 patients resulted in no responses32,33. Sahin et al. recently published encouraging data from a first-in-human trial, in which a vaccine containing liposomal RNA targeting four unmutated tumor-associated antigens (NY-ESO-1, MAGE-A3, tyrosinase and TPTE) was administered alone or in combination with nivolumab in patients with advanced melanoma. Responses were observed in both the monotherapy group as well as the combination group in checkpoint inhibitor-experienced patients, suggesting the efficacy of non-mutant shared tumor antigen vaccines34. The overall safety and tolerability findings are comparable to those of aPD1 monotherapy. Injection site reactions were exclusive to the vaccine. However, these side effects were transient and mild in most patients and most likely due to the adjuvant Montanide. IDO- and PD-L1-specific CD8+ and CD4+ T cells exist among peripheral blood lymphocytes in healthy donors35–38 and expand in response to pro-inflammatory stimuli35,39. Furthermore, both IDO and PD-L1 are induced in cells as a counter response to the inflammatory response. This provides a mechanism that ensures immune homeostasis, which keeps IDO/PD-L1-specific T cells in check; therefore, the risk of triggering autoimmune-related adverse events by vaccination appears to be minimal. This was confirmed in the first clinical trials of IDO and PD-L1 vaccination (NCT01543464 (refs. 33,40) and NCT03042793 (ref. 41)). The spontaneous (baseline) immune response to the vaccines observed in the current study is in agreement with our previous observation in various patients with cancer35–38. Numerous vaccine-induced changes in the blood and at the tumor site were observed. Peripheral IDO- and/or PD-L1-specific T cells were detected in vitro in a modified ELISPOT in over 93% of patients on vaccination, unrelated to patient HLA type. Immune responses were persistent in patients who surpassed follow-up at data cutoff and were still detectable up to 6 months after the last vaccine, suggesting induction of memory T cells. We did not observe a correlation between vaccine-induced responses in blood and clinical responses. However, the detection of a highly significant increase in vaccine-specific T cell numbers after vaccination in almost all patients together with the very low number of patients with PD makes it difficult to expect such a correlation, especially because other aspects should be taken into account (for example, the loss of class I expression on tumors cells in patient MM02 with PD after vaccination). Frequencies of peripheral T cells induced against PD-L1 were overall higher than those against IDO. Importantly, however, we also observed that it was not the same patients who responded strongly to both IDO and PD-L1, nor did patients react with a similar response pattern to the two antigens; that is, different time points of response were observed. This suggests that each component of the vaccine plays different roles in the ongoing immune response in patients. The goal of IDO/PD-L1 vaccination was to modulate the TME to increase responsiveness to aPD1 therapy, as we have observed in animal models, in which immune conversion is demonstrated in the TME with an increased influx of T cells. Targeting both IDO and PD-L1 together enables synergy, as the TME is known to use different immune escape mechanisms and IDO and PD-L1 are often overexpressed by different cellular compartments. Ex vivo ELISPOT and flow cytometry assays confirmed the induction of immune response toward both epitopes and the higher immunogenicity of the PD-L1 epitope, although because of lower sensitivity failed to detect some of the responses. TCR sequencing of five patients confirmed enrichment of IDO/PD-L1-specific T cell clones in the blood at different time points after treatment. Furthermore, an increase in enriched IDO/PD-L1-specific clones was observed in four of five patients at the tumor site after treatment, irrespective of clinical response. Phenotypic characterization showed that vaccine-specific T cells, which were expanded in vitro with interleukin (IL)-2 from the blood of vaccinated individuals, were both CD4+ T cells and CD8+ T cells. This was confirmed by ex vivo phenotype description of vaccine-activated T cells. Vaccine-specific T cells expressed CD107a and CD137 and produced IFN-γ and TNF-α upon stimulation with the cognate target, indicating their cytolytic capacity. Overall, the data from immune monitoring supports the importance of both antigens in the generation of the frequent clinical responses in the study. Despite a limited number of paired biopsies (due to either patient refusal or the fact that a large fraction of responding patients had no assessable tumors after six cycles), we observed trends indicating treatment-induced general T cell influx in responding patients. It was shown that proliferation of CD8+ T cells in the tumor after aPD1 treatment is associated with radiographic reduction in tumor size42. Additionally, we showed that a large proportion of expanded peripheral TCR clones were associated with tumors and the most considerable amount of clonal expansion was observed early, at cycle 3. For the patient with a CR included in the TCR sequencing analyses (MM01), the number of peripheral expanded clones present at the tumor site increased after treatment compared to that at baseline, indicating tumor trafficking of peripheral expanded clones. Gene expression analyses (two paired biopsies) and IHC (five paired biopsies) further demonstrated that the combination treatment induced a pro-inflammatory TME in responding patients with signs of T cell activation and cytotoxicity and increased cytokine activity. This may lead to further upregulation of IDO, PD-L1, MHC-I and MHC-II on tumor cells, leading to more treatment targets. It was shown that, following vaccination with a cancer vaccine, PD-L1 expression is increased on tumor cells due to recruitment of tumor-specific T cells and upregulation of adaptive immune resistance pathways in the TME43. Treatment with nivolumab monotherapy enhances PD-L1 expression, and it is therefore problematic to discriminate the effect of the vaccine as compared to that of nivolumab44. IDO- and PD-L1-specific pro-inflammatory effector T cells were hypothesized to counteract the functions of IDO- and PD-L1-expressing immune-suppressive cells as a means to keep the immune balance between immune activation and inhibition45. However, because the expression of these molecules is induced as a counter-regulatory response to inflammatory mediators such as IFNs, they can also be expressed by, for example, activated myeloid cells or T cells. Thus, activation of PD-L1-specific T cells may result in depletion of activated T cells in the tumor and other sites. Importantly, it was recently described that PD-L1+ T cells mainly have tolerogenic effects on tumor immunity and exert tumor-promoting properties, suggesting that targeting this immune population is indeed also beneficial46. We have previously investigated the effect of activating PD-L1-specific T cells in vitro and in vivo and found that, overall, they supported the effector phase of an immune response by removing PD-L1-expressing regulatory immune cells that inhibit PD-1+ effector T cells47,48. The major role of the PD-1 pathway is believed to be the regulation of effector T cell responses. Thus, this protective pathway is more important after activation, rather than at the initial T cell-activation stage49. Accordingly, the presence of PD-L1-specific T cells during the activation phase of an immune response may not increase or support a pro-inflammatory immune response due to the expression of PD-L1 on potent antigen-presenting cells or on the T cells themselves14. Thus, the overall effects of PD-L1-specific T cells may vary depending on the expression of both PD-1 and PD-L1, that is, due to the microenvironment and the state of the immune response. In conclusion, here we report an impressive response rate, CR rate and mPFS for a first-in-class immune-modulating vaccine combined with nivolumab. This may be a first step toward a new treatment strategy for patients with MM. Limitations are the low number of patients treated at a single institution and the lack of a randomized design with aPD1 monotherapy as comparator. Studies in aPD1 therapy-resistant or -refractory melanoma are ongoing as well as biomarker analysis for selecting patients at a higher likelihood to benefit from the combination versus aPD1 monotherapy. A larger randomized trial will be essential to validate these findings and determine the specific contribution of the vaccine to clinical responses and changes in the TME. In December 2020, the Food and Drug Administration granted breakthrough therapy designation for the IO102/IO103 vaccine combined with aPD1 therapy in metastatic melanoma based on data from the MM1636 trial. Methods Trial design and treatment plan MM1636 is an investigator-initiated, nonrandomized, open-label, single-center phase 1/2 study. All patients were treated at the Department of Oncology, Herlev and Gentofte Hospital, University of Copenhagen, Herlev, Denmark. The study was conducted according to the Declaration of Helsinki and Good Clinical Practice (GCP) and monitored by the GCP unit in Copenhagen, Denmark. The protocol was approved by the ethical committee of the Capital region of Denmark (H-17000988), the Danish Medical Agencies (2017011073) and the Capital Region of Denmark Data Unit (P-2019-172). The study was registered at https://clinicaltrials.gov/ under the identifier NCT03047928 and at the EudraCT (no. 2016-0004527-23). This study initially aimed to include 30 aPD1 treatment-naive patients with MM. An amendment with the addition of two other cohorts with ten patients in each cohort was made to evaluate immune responses and clinical efficacy in aPD1 therapy-resistant patients (cohort B, de novo resistance and cohort C, acquired resistance) for a total of 50 patients. The amendment with cohorts B and C was approved at the inclusion of 18 patients in cohort A. The trial is still including patients in cohorts B and C. This article reports results from cohort A. The first six patients were treated as for phase 1, evaluating for safety and tolerability before the remaining 24 patients were included in phase 2. The IDO/PD-L1 vaccine was administered subcutaneously biweekly for the first 10 weeks and thereafter every 4 weeks for approximately 9 months. A maximum of 15 vaccines were administered. Nivolumab was administered according to the approved label (3 mg per kg biweekly) for 24 cycles. The 15th vaccine was administered together with the 24th nivolumab cycle of 3 mg per kg, and responding patients thereafter continued nivolumab monotherapy every 4th week (6 mg per kg) as a standard of care after investigator assessment. Treatment with nivolumab was discontinued at the maximum benefit (assessed by the investigator), after a maximum of 2 years of therapy, at progression or due to severe adverse events (Extended Data Fig. 1b, treatment plan). Vaccine composition Each vaccine was composed of 100 µg IO102, a 21-amino-acid peptide (DTLLKALLEIASCLEKALQVF) from IDO, and 100 µg IO103, a 19-amino-acid peptide (FMTYWHLLNAFTVTVPKDL) from the signal peptide of PD-L1 (PolyPeptide Laboratories). Peptides were dissolved separately in 50 µl DMSO, filtered for sterility and frozen at −20 °C (NUNC CyroTubes CryoLine System Internal Thread, Sigma-Aldrich). At <24 h before administration, peptides were thawed. The PD-L1 peptide was diluted in 400 µl sterile water and, immediately before injection, mixed with the IDO peptide solution and 500 µl Montanide ISA-51 (SEPPIC) to achieve a total volume of 1 ml. Patients Patients above 18 years of age with locally advanced or stage 4 melanoma according to the AJCC (seventh edition), at least one measurable lesion according to RECIST 1.1 and an ECOG PS of 0–1 were eligible. The main exclusion criteria were prior treatment with aPD1 therapy, CNS metastases >1 cm, severe comorbidities and active autoimmune disease. Enrollment was not restricted to PD-L1 status, but it was known before inclusion. Patients were included after informed consent. Patients MM42 and MM20 have confirmed their approval of PET–CT images and clinical images that have been included in the article. Key study assessments Safety and tolerability were evaluated based on changes in clinical laboratory analyses and reported adverse events. Adverse events were assessed according to Common Terminology Criteria for Adverse Events (version 5.0) and were graded from 1 to 5 for all treated patients up to 6 months after the last dose of the IDO/PD-L1 vaccine. Clinical efficacy was assessed using FDG PET–CT scans before treatment and every 3rd month until progression. Objective responses were categorized into CR, PR, stable disease or PD according to RECIST version 1.1. Clinical data were collected at CCIT-DK in the eCRF program OpenClinica version 1.0 and in Microsoft Excel version 2002 on a secure server. Blood samples for immunologic analyses were collected before treatment, before the third cycle, after the sixth, 12th, 18th and 24th cycles (on vaccination) and 3 and 6 months after the last vaccine. Two to three tumor needle biopsies (1.2 mm) were collected at baseline and after six cycles from the same tumor site, when accessible, to evaluate immune responses at the tumor site. DTH skin tests were performed and punch biopsies were taken from the DTH area after cycle 6 for the evaluation of SKILs reactive to PD-L1 and IDO (Supplementary Fig. 1, treatment plan). Statistical analysis of clinical outcome Survival curves were computed in GraphPad Prism software version 9.0.0 according to Kaplan–Meier method. Median follow-up time of enrollment was calculated using the reverse Kaplan–Meier method, also in GraphPad Prism software version 9.0.0. For binary outcomes, 95% two-sided CIs were constructed using the Clopper–Pearson method, also in GraphPad version 9.0.0. An independent board of certified and experienced oncoradiologists performed an external review to evaluate clinical response to address the potential bias of investigator site review. The external review took place at Rigshospitalet, Copenhagen University Hospital. This hospital did not participate in the MM1636 trial, and the external reviewer had no prior knowledge about the clinical trial or the trial therapy. Only PET–CT images were accessed, and arrows indicating target or non-target lesions appeared on baseline images as the only additional information. To address potential trial bias regarding treatment effect, we matched patients in MM1636 with patients from the DAMMED, a population-based database that retrospectively collects data on patients with metastatic melanoma in Denmark. Here, data from 938 patients treated with aPD1 monotherapy contemporaneously (January 2015–October 2019) were extracted. Two hundred and eighteen of these patients were eligible for comparison and matching (all parameters available) (Supplementary Table 1), and 74 patients from the DAMMED were found to match. Patients were matched for age (≤70 years, >70 years), sex, LDH levels (normal, elevated), M stage (M1a, M1b, M1c), BRAF status (wild type, mutated) and PD-L1 status (<1%, ≥1%). An exact matching algorithm was used in which patients in MM1636 were matched with patients from the DAMMED with the same combination of variables. Twenty-nine patients from MM1636 were matched with exact combinations of the six variables. One patient could not be matched. To secure balance of the calculations, control patients were weighted according to the number of patients for each MM1636 patient. Estimates for treatment effects were calculated by weighted logistic regression analyses and weighted Cox proportional hazard model. The R package ‘Matchlt’ was used for matching patients. As the method chosen for matching control patients to protocol patients, a weighted binary logistic regression model was used for comparing response rates in the two matched cohorts. Odds ratios and response rates, including their corresponding 95% CIs, were extracted from the regression models. All P values were two sided, and P values below 0.05 were considered statistically significant. SAS version 9.4M5 was used for the weighted logistic regression models. Processing of peripheral blood mononuclear cells Peripheral blood was collected from all patients in heparinized tubes and was processed within 4 h. In brief, PBMCs were isolated using Lymphoprep (Medinor) separation. PBMCs were counted on a Sysmex XP-300 analyzer and frozen in Human AB Serum (Sigma-Aldrich, H4522-100ML) with 10% DMSO using controlled-rate freezing (CoolCell, BioCision) in a −80 °C freezer and were moved the next day to a freezer at −140 °C until further processing. Needle biopsies at baseline and after the sixth vaccine Two to three needle biopsies (1.2 mm) were taken at baseline and after the sixth cycle of treatment when assessable from the same tumor lesion. One fragment was fixed with formalin and embedded in paraffin (FFPE); one to two fragments were used for expanding tumor-infiltrating T cells and establishing autologous tumor cell lines. Delayed-type hypersensitivity and generation of skin-infiltrating lymphocytes After six cycles of treatment, we performed intradermal injections of vaccine components without adjuvant and one control injection containing DMSO without peptide. Patients were injected with either a mixture of both IDO and PD-L1 peptides at all three injection sites or PD-L1 peptide, IDO peptide or a mixture of the two at the injection sites, respectively (Supplementary Table 6). Eight hours after injection, punch biopsies were resected from the three sites where the peptide was injected and transported immediately to the laboratory and cut into fragments of 1–2 mm3. SKILs were expanded to establish ‘young SKILs’ in CM medium consisting of RPMI 1640 with GlutaMAX, 25 mM HEPES, pH 7.2 (Gibco, 72400-021), IL-2 (100 or 6,000 IU ml−1) (Proleukin Novartis, 004184), 10% heat-inactivated Human AB Serum (Sigma-Aldrich, H4522-100ML), 100 U ml−1 penicillin, 1.25 µg ml−1 Fungizone (Bristol Myers Squibb, 49182) and 100 µg ml−1 streptomycin (Gibco, 15140-122). Half of the medium was replaced three times per week. Young SKILs from samples derived from IDO peptide, PD-L1 peptide and the mixture injection sites were further expanded in a small-scale version of the 14-d rapid expansion protocol as previously described51. Quantification of specific T cells in blood by ELISPOT For enumeration of vaccine-specific T cells in the peripheral blood, PBMCs from patients were stimulated with IDO or PD-L1 peptide in the presence of a low dose of IL-2 (120 U ml−1) for 7–13 d before being used in IFN-γ ELISPOT assays with 2.5–3.2 × 105 cells per well. Briefly, cells were placed in a 96-well PVDF membrane-bottomed ELISPOT plate (MultiScreen MSIPN4W50, Millipore) precoated with IFN-γ-capture antibody (clone 1-D1K, Mabtech). Diluted IDO or PD-L1 peptide stocks in DMSO were added at 5 µM; an equivalent amount of DMSO was added to control wells. PBMCs from each patient were set up in duplicate or triplicate for peptide and control stimulations. Cells were incubated in ELISPOT plates in the presence of the peptide for 16–18 h, after which they were washed off, and biotinylated secondary antibody (anti-human IFN-γ mAB, clone 7-B6-1, Mabtech) was added. After 2 h of incubation, unbound secondary antibody was washed off, and streptavidin-conjugated alkaline phosphatase (Mabtech) was added for 1 h. Next, unbound streptavidin-conjugated enzyme was washed off, and the assay was developed by adding BCIP/NBT substrate (Mabtech). ELISPOT plates were analyzed on the CTL ImmunoSpot S6 Ultimate V analyzer using ImmunoSpot software version 5.1. Responses were calculated as the difference between the average numbers of spots in wells stimulated with IDO or PD-L1 peptide and those from corresponding control wells. For detection of IDO and PD-L1 peptide responses ex vivo, PBMCs were thawed and rested overnight in medium containing DNase I (1 μg ml−1, Sigma-Aldrich, 11284932001) before being used in IFN-γ ELISPOT assays as described above. A total of 6–9 × 105 PBMCs per well were seeded. Statistical analysis of all ELISPOT responses was performed using the DFR method as described by Moodie et al. using RStudio software (RStudio Team, 2016, http://www.rstudio.com/)50. Vaccine-specific ELISPOT responses were defined as true if the difference between the spot count in control and peptide-stimulated wells was statistically significant according to DFR statistical analysis or, for samples performed in duplicate, if the spot count in peptide-stimulated wells was at least 2× the spot count in control wells50. Cancer cell lines and tumor conditioned medium Autologous melanoma cell lines were established from needle biopsies. Briefly, biopsies were chopped into small fragments and seeded in 24-well cultures in RPMI 1640 with GlutaMAX, 25 mM HEPES, pH 7.2 (Gibco, 72400-021), 10% heat-inactivated FBS (Life Technologies, 10500064), 100 U ml−1 penicillin, 1.25 µg ml−1 Fungizone (Bristol Myers Squibb, 49182) and 100 µg ml−1 streptomycin (Gibco, 15140-122). Established adherent melanoma tumor cell lines were cryopreserved at −140 °C in freezing medium containing FBS with 10% DMSO. PD-L1 and HLA-II expression on established tumor cell lines was assessed by flow cytometry staining with anti-PD-L1–PE-Cy7 (1:23 dilution, clone M1h1 (RUO), BD, 558017) and anti-HLA-II–FITC (1:23 dilution, clone Tu39 (RUO), BD, 555558) antibodies. To obtain TCM, established tumor cell lines were cultured in 175-cm2 Nunc cell culture flasks until 80–90% confluency was reached. The culture medium was then replaced with 20 ml fresh X-VIVO 15 with Gentamicin and Phenol Red (Lonza, BE02-060Q), medium with 5% heat-inactivated Human AB Serum (Sigma-Aldrich, H4522-100ML). After 24 h of incubation, TCM was collected and centrifuged to remove any resuspended cells, after which TCM was aliquoted, frozen and stored at −80 °C. The acute monocytic leukemia cell line MonoMac1 was obtained from the DSMZ—German Collection of Microorganisms and Cell Cultures (ACC 252) and cultured in RPMI 1640 with GlutaMAX, 25 mM HEPES, pH 7.2 (Gibco, 72400-021) and 10% heat-inactivated FBS. Isolation of autologous myeloid cells Autologous CD14+ cells were sorted from freshly thawed PBMCs using a magnetic bead-separation kit (Miltenyi Biotec, 130-050-201) according to the manufacturer’s instructions. Isolated CD14+ cells were used as targets in the IFN-γ ELISPOT assay directly after sorting or differentiated in vitro into tumor-associated macrophages by culturing with 1 ml fresh X-VIVO 15 medium with Gentamicin and Phenol Red (Lonza, BE02-060Q) and 5% heat-inactivated Human AB Serum, supplemented with 1 ml autologous TCM in 24-well plates for 2 d. Quantification of vaccine-specific T cells from DTH biopsy sites by ELISPOT SKILs expanded in vitro were rested in medium without IL-2 overnight before being used in the IFN-γ ELISPOT assay as described above to evaluate reactivity of skin-infiltrating T cells. Generation of IDO- and PD-L1-specific T cell cultures from PBMCs or SKILs IDO- or PD-L1-specific T cells were isolated from peptide-stimulated in vitro PBMC cultures on days 14–15 after stimulation or from SKIL cultures expanded in vitro. For specific T cell isolation, PBMCs or SKILs were stimulated with IDO or PD-L1 peptide, and cytokine-producing T cells were sorted using the IFN-γ or TNF-α Secretion Assay—Cell Enrichment and Detection kit (Miltenyi Biotec). Cytokine-production profile of PD-L1- and IDO-specific T cells by intracellular staining To assess the T cell cytokine-production profile, isolated and expanded IDO- and PD-L1-specific T cell cultures were stimulated for 5 h with peptide at 5 μM in a 96-well plate. One hour after the start of the incubation, anti-CD107a–PE (1:133 dilution, clone H4A3, BD Biosciences, 555801) antibody and BD GolgiPlug (1:1,000 dilution, BD Biosciences) were added. After a 5-h incubation, cells were stained using fluorescently labeled surface marker antibodies: anti-CD4–PerCP (1:5 dilution, clone SK3, 345770), anti-CD8–FITC (1:5 dilution, clone SK1, 345772) and anti-CD3–APC-H7 (1:20 dilution, clone SK7, 560275) (all from BD Biosciences). Dead cells were stained using FVS510 (1:250 dilution, BD Biosciences, 564406), followed by overnight fixation and permeabilization using eBioscience Fixation/Permeabilization buffers (00-5123-43, 00-5223-56) according to the manufacturer’s instructions. Cells were then stained intracellularly in eBioscience permeabilization buffer (00-8333-56) with anti-IFN-γ–APC (1:20 dilution, clone 25723.11, BD Biosciences, 341117) and anti-TNF-α–BV421 (1:100 dilution, clone Mab11, BD Biosciences, 562783) antibodies. Samples were analyzed on the FACSCanto II (BD Biosciences) using BD FACSDiva software version 8.0.2. The gating strategy is shown in Supplementary Fig. 3a. For assessing the reactivity against IDO expressing MonoMac1 cells, the appropriate amount of cancer cells was added to the IDO-specific T cells to obtain an effector:target ratio of 4:1 in the 96-well plate and T cell cytokine production was tested after 5 h of co-culture as described above. For HLA-blocking experiments, HLA-blocking antibodies (2 μg ml−1) were added directly into wells 30 min before the addition of peptide. Blocking antibodies used were HLA-DR (1:500 dilution, clone L243, Abcam, ab136320), HLA-DQ (1:500 dilution, clone SPV-L3, Abcam, ab23632) and HLA-DP (1:500 dilution, clone B7/21, Abcam, ab20897). To assess ex vivo T cell reactivity to IDO and PD-L1 peptides in patient PBMCs, cells were thawed and rested for 1–2 d in medium containing DNase I (1 μg ml−1, Sigma-Aldrich, 11284932001). PBMCs were then stimulated with peptide at 5 μM in a 96-well plate for 8 h. An hour after the addition of peptide, anti-CD107a–BV421 (3:50 dilution, clone H4A3, BD Biosciences, 328626) antibody and BD GolgiPlug (1:1,000 dilution, BD Biosciences) were added. Surface and intracellular staining was performed as described above. Antibodies used for surface staining were anti-CD3–PE-CF594 (0.8:30 dilution, clone UCHT1, BD Biosciences, 562280), anti-CD4–BV711 (1:30 dilution, clone SK3, BD Biosciences, 563028) and anti-CD8–Qdot605 (1:150 dilution, clone 3B5, Thermo Fisher, Q10009). Antibodies used for intracellular staining were anti-CD137–PE (1:20 dilution, clone 4B4-1 (RUO), BD Biosciences, 555956), anti-IFN-γ–PE-Cy7 (1.5:20 dilution, clone B27, BD Biosciences, 557643) and anti-TNF-α–APC (1:20 dilution, clone Mab11, BD Biosciences, 554514). Samples were acquired on the NovoCyte Quanteon (ACEA Biosciences) and analyzed using NovoExpress software version 1.4.1. To assess vaccine-specific T cell responses, background values observed in unstimulated PBMC samples were subtracted from values observed in peptide-stimulated conditions. Positive response value threshold was set at a difference of 0.2% from the background values. Based on this response cutoff, only TNF-α, CD107a and CD137 responses were detected in this assay. Statistical analysis comparing baseline with on-treatment or post-treatment cytokine profiles was performed using two-sided Wilcoxon matched-pairs signed-rank test. The gating strategy is shown in Supplementary Fig. 3b. siRNA-mediated PD-L1 and IDO1 silencing A stealth siRNA duplex for targeted silencing of PD-L1 (Invitrogen)52, a custom silencer select siRNA for targeted silencing of IDO1 (Ambion) and the recommended silencer select negative control (Ambion) siRNA for mock transfection were used. The stealth PD-L1 siRNA duplex consisted of the sense sequence 5′-CCUACUGGCAUU-UGCUGAACGCAUU-3′ and the antisense sequence 5′-AAUGCGUUCAGCAAAUGCCAGUAGG-3′. The three silencer IDO siRNA duplexes used were siRNA1 (sense sequence, 5′-ACAUCUGCCUGAUCUCAUATT-3′; antisense sequence, 5′-UAUGAGAUCAGGCAGAUGUTT-3′), siRNA2 (sense sequence, 5′-CCACGAUCAUGUGAACCCATT-3′; antisense sequence, 5′-UGGGUUCACAUGAUCG-UGGAT-3′) and siRNA3 (sense sequence, 5′-CGAUCAUGUGAACCCAAAATT-3′; antisense sequence, 5′-UUUUGGGUUCACAUGAUCGTG-3′). For PD-L1 or IDO1 silencing, cancer cells were electroporated with 0.025 nmol of each siRNA duplex as previously described53. For PD-L1-silencing experiments, cancer cells were treated with IFN-γ (500 U ml−1, PeproTech) 1 h after electroporation. Electroporated cells were used as target cells in ELISPOT and ICS assays 24 h or 48 h after siRNA electroporation. RT–qPCR Total RNA was extracted using the RNeasy Plus Mini kit (Qiagen, 74134) following the manufacturer’s instructions. RNA concentration was quantified using a NanoDrop 2000 (Thermo Fisher Scientific), and a total of 1,000 ng RNA was reverse transcribed with the High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems, 4368814) using 1,000 ng input RNA for transcription. Real-time qPCR analyses were performed using the TaqMan Gene Expression Assay on a Roche LightCycler 480 instrument. RT–qPCR assays were performed with a minimum of three technical replicates and analyzed using the ΔCt method as described in Bookout et al.54 with normalization of IDO1 expression (primer ID, Hs00984148_m1) or PD-L1 expression (primer ID, Hs001125296_m1) to the expression level of the housekeeping gene POLR2A (primer ID, Hs00172187_m1) and the control sample (mock). For low-concentration samples with no amplification, Ct was set to 40. Controls without reverse transcriptase were used as controls for specific amplifications. P values were determined using two-tailed parametric t-tests. BRAF-mutational status and PD-L1 status at baseline in all patients A library of historical FFPE biopsies were assessable for all patients and analyzed locally at Herlev and Gentofte University Hospital by experienced pathologists for BRAF status and PD-L1 expression on tumor cells. The BRAF analysis was carried out with real-time PCR using the EntroGen BRAF Mutation Analysis Kit II (BRAFX-RT64, CE-IVD) to specifically detect mutations corresponding to V600D, V600E and V600K in the BRAF protein. PD-L1 status was assessed using the monoclonal rabbit anti-PD-L1 antibody (clone 28.8, PD-L1 IHC 28-8 pharmDx) in FFPE biopsies following the manufacturer’s instructions. Patients were considered positive for PD-L1 with expression levels ≥1% and negative for PD-L1 with expression levels <1%. Human leukocyte antigen type Blood samples from all 30 patients were genotyped for class I (HLA-A, HLA-B, HLA-C) and three class II types (HLA-DRB1, HLA-DQA1, HLA-DQB1) using LinkSēq HLA Typing Kits (Thermo Fisher, 1580C). These test kits are based on real-time PCR using allele-specific exponential amplification (sequence-specific primers) followed by melting curve analyses. Immunohistochemistry simplex: Immunoscore CR IHC staining was performed using a qualified Ventana Benchmark XT with four different steps: (1) antigen retrieval; (2) staining with the following primary antibodies: anti-CD3 antibody (clone HDX2; provider, HalioDx; HD-FG-000013; 10931065/10636667; concentration, 0.25 µg ml−1), anti-CD8 antibody (clone HDX3; provider, HalioDx; HD-FG-000019; 10931069/10337710/10639301; concentration, 0.5 µg ml−1), anti-IDO monoclonal antibody (clone VINC3IDO; provider, Thermo Fisher Scientific; 14-9750-82; E25003-101; concentration, 0.05 µg ml−1), anti-HLA class 1 ABC antibody (clone, EMR8-5; provider, Abcam; ab70328; GR3248333/GR3186494; concentration, 0.5 µg ml−1), anti-HLA-DR/DP/DQ/DX antibody (clone CR3/43; provider, Santa Cruz; sc-53302; L1714; concentration, 1 µg ml−1) and anti-PD-L1 antibody (clone HDX3; provider, HalioDx; HD-FG-000035; 106312810/106312816; concentration, 3.3 µg ml−1); (3) detection with a secondary antibody using the ultraView Universal DAB Detection kit; and (4) counterstaining using Hematoxylin & Bluing Reagent (staining of cellular nuclei). Control slides were systematically included in each staining run to permit quality control of the obtained measurements. Following coverslipping, slides were scanned with the NanoZoomer-XR to generate digital images (20×). Two consecutive slides were specifically used to perform Immunoscore CR TL staining of CD3+ and CD8+ cells. Digital pathology of T lymphocytes The digital pathology for Immunoscore CR TL allowed the quantification of positive cells (in cells per mm2) into the core tumor and invasive margin if present. Each sample was analyzed using the HalioDx Digital Pathology Platform. Digital pathology and pathologist analysis of IDO Digital pathology for IDO allowed the quantification of stained area (in mm2) into the whole tumor. Each sample was analyzed using the HalioDx Digital Pathology Platform. Analysis of IDO+ cells was performed by a pathologist. Results were expressed as an H score from 0 to 300. The score is obtained by the formula 3 × percentage of cells with strong staining + 2 × percentage of cells with moderate staining + percentage of cells with weak staining. Digital pathologist analysis of MHC-I and MHC-II Analysis of MHC-I+ and MHC-II+ cells was performed by a pathologist. Results are expressed as a percentage of positive tumor cells and a percentage of positive cells in the stroma. Digital pathology Immunoscore immune checkpoint (CD8+ and PD-L1+) The Immunoscore CR IC test allowed the quantification of CD8++ cell density in the whole tumor and a CD8+-centered proximity index (which corresponds to the percentage of CD8++ cells that have at least one PD-L1+ cell in the neighborhood) at different cutoff distances (20 µm, 40 µm, 60 µm and 80 µm). Pathologist analysis of PD-L1 A pathologist performed analysis of PD-L1+ cells. Positivity of a viable tumor cell was considered when partial or complete cell membrane staining was observed (more than 10% of the tumor cell membrane). Results were expressed as a percentage. NanoString RNA profiling and Immunosign RNA was extracted from FFPE tissues using Qiagen RNeasy FFPE extraction kits (Qiagen). Annotations from the pathologist performing H&E staining were used to guide removal of normal tissue from the slides by macrodissection before nucleic acid extraction, which occurred after tissue deparaffinization and lysis. Each extracted RNA sample was independently quantified using a NanoDrop spectrophotometer (NanoDrop Technologies) and qualified (Agilent Bioanalyzer). Degradation was quantified as the percentage of RNA fragments smaller than 300 bp using the RNA 6000 Nano kit (Agilent Bioanalyzer). Good sample quality was defined as less than 50% of RNA fragments being 50–300 bp in size. RNA expression profiling was performed using the nCounter PanCancer Immune Profiling Panel from NanoString (NanoString Technologies). The PanCancer Immune Profiling Panel contains 776 probes and is supplemented with six genes to complete HalioDx Immunosign targets. Hybridization was performed according to the manufacturer’s instructions. Hybridized probes were then purified and immobilized on a streptavidin-coated cartridge using the nCounter Prep Station (NanoString Technologies). Data collection was carried out on the nCounter Digital Analyzer (NanoString Technologies) following the manufacturer’s instructions to count individual fluorescent barcodes and quantify target RNA molecules present in each sample. For each assay, a scan of 490 fields of view was performed. Raw data from the NanoString nCounter were processed using NanoString recommendations. Quality control enables to retain data of good quality with a binding density that ranges between 0.05 and 2.25. The linearity of positive controls was checked using the R2 value of regression between the counts and the concentrations of positive controls. Samples that showed R2 < 0.75 were flagged and removed from the analysis. The background was removed using a thresholding method at the mean + 2 s.d. of negative controls. Raw counts were normalized using a positive normalization factor. Samples showing positive normalization factors outside the range of 0.3–3 were removed from the analysis. A second normalization was performed using the housekeeping gene normalization factor. Only the most stable housekeeping genes were selected for this normalization step using the variance-versus-mean relationship. All samples showing a normalization factor outside the range of 0.1–10 were removed from the analysis. All statistical analyses were performed on normalized counts using R software (version 2.6.2, 2019-12-12). T cell receptor variable β chain sequencing To track longitudinal immune responses to therapy, genomic DNA was extracted from longitudinal pre- and post-treatment PBMCs (five patients), pre- and post-treatment biopsies (five patients) (both FFPE and treated with RNAlater) and IDO- and PD-L1-specific T cell cultures from PBMCs (five patients) or SKILs (one patient). Three clonal (two IDO- and one PD-L1-) specific cultures were also sequenced to confirm clonal purity of cultures. DNA from PBMCs or RNAlater-treated biopsies was extracted with the DNeasy Blood and Tissue kit (Qiagen, 69504), DNA from sorted IDO and PD-L1-specific T cells from either PBMCs or SKILs was extracted using the QIAamp DNA Micro kit (Qiagen, 565304), and DNA from PFFE biopsies was extracted using the Maxwell RSC DNA FFPE kit (Promega, AS1450). Immunosequencing of the CDR3 regions of human TCRβ chains was performed using the immunoSEQÒ Assay (Adaptive Biotechnologies). Extracted genomic DNA was amplified in a bias-controlled multiplex PCR, followed by high-throughput sequencing. Sequences were collapsed and filtered to identify and quantitate the absolute abundance of each unique TCRβ CDR3 region for further analysis as previously described55–57. Statistical analyses of TCRβ sequencing results Two quantitative components of diversity were compared across samples in this study. First, Simpson clonality was calculated on productive rearrangements by ∑i=1Rpi2, where R is the total number of rearrangements and pi is the productive frequency of rearrangement i. Values of Simpson clonality range from 0 to 1 and measure how evenly receptor sequences (rearrangements) are distributed among a set of T cells. Clonality values approaching 0 indicate a very even distribution of frequencies, whereas values approaching 1 indicate an increasingly asymmetric distribution in which a few clones are present at high frequencies. Second, sample richness was calculated as the number of unique productive rearrangements in a sample after computationally downsampling to a common number of T cells to control for variation in sample depth or T cell fraction. Repertoires were randomly sampled without replacement five times, and we report the mean number of unique rearrangements. The T cell fraction was calculated by taking the total number of T cell templates and dividing by the total number of nucleated cells. The total number of nucleated cells was derived from reference genes using the immunoSEQ Analyzer version 3. To identify enriched vaccine clones in each patient, rearrangement frequencies in their baseline PBMCs and each IDO/PD-L1-sorted T cell sample were compared using a binomial distribution framework as previously described58. In brief, for each clone, we performed a two-sided test to determine whether frequencies were the same in the patient’s periphery and a PD-L1- or IDO-specific T cell sample. The Benjamini–Hochberg procedure was used to control the false discovery rate at 0.01 (ref. 59). Clonal expansion in post-treatment samples was similarly assessed using this differential abundance framework, but an IDO/PD-L1-specific T cell sample was replaced with a post-treatment series sample. In biopsies, the six series frequencies were compared to those of baseline tissue. Lastly, vaccine-associated clones were tracked in each PBMC and tissue sample by summing the frequency of each rearrangement enriched in either PD-L1- or IDO-specific T cells. All statistical analyses were performed in R version 3.6.1. Reporting Summary Further information on research design is available in the Nature Research Reporting Summary linked to this article. Online content Any methods, additional references, Nature Research reporting summaries, source data, extended data, supplementary information, acknowledgements, peer review information; details of author contributions and competing interests; and statements of data and code availability are available at 10.1038/s41591-021-01544-x. Supplementary information Supplementary Information Supplementary Figs. 1–3 and Tables 1–6 Reporting Summary Extended data Extended Data Fig. 1 Mode of Action and treatment plan. a) Anticipated mechanism of action of the combination therapy of an IDO/PD-L1 derived peptide vaccine and nivolumab (anti-PD1). 1) The IDO/PD-L1 peptide vaccine is administered subcutaneously (s.c) and nivolumab is administered intravenous (IV). 2) The vaccines peptides are phagocytosed by an antigen presenting cell and presented to IDO and PD-L1 specific T cells, which are activated. 3) The activated T cells migrates to the tumor site where they attack both immune-suppressive cells and tumour cells expressing IDO and/or PD-L1 leading to cytokine production and a pro-inflammatory tumour microenvironment. 4/5) Enhanced tumour killing by both IDO/PD-L1 specific T cells and tumor specific cytotoxic T cells due to PD-1 blockade. Created with BioRender.com b) Treatment plan. After written informed content patients were screened. Before treatment start a baseline PET/CT scan was performed, baseline blood sample for research use and if assessable a baseline needle biopsy. Patients were treated with the IDO/PD-L1 peptide vaccine subcutaneously biweekly for the first 6 injections and thereafter every fourth week for a maximum of 15 vaccines. Nivolumab was administered in parallel biweekly (3mg/kg) up to 24 series. If patients needed subsequent nivolumab after ended vaccination regiment they were treated with 6 mg/kg every fourth week up to two years. Needle biopsy and delayed type hypersensitivity (DTH) was performed after 6 series of treatment if assessable. PET/CT scans was performed every third month. Extended Data Fig. 2 Progression free survival and overall survival in matched historical control group and vaccine injection site reaction. a) Kaplan-Meier curve of progression free survival with corresponding 95% confidence intervals in the matched historical control group (n=74). Patients were matched on BRAF-status, PD-L1-status, age, gender, M-stage and LDH level). b) Kaplan-Meier curve of overall survival with corresponding 95% confidence intervals in the matched historical control group (n=74). c) Images of injection site reaction in patient MM20 (CR) after 11 vaccination show redness, rash and granuloma at injection site. Extended Data Fig. 3 Vaccine specific responses in blood. a) Heatmap of specific (background has been subtracted) IDO and PD-L1 responses in PBMCs at baseline, series 3, 6, 12, 18 and 24 measured by in vitro IFN-γ Elispot assay show fluctuations in the blood during treatment (n=30). b) Heatmap of specific (background has been subtracted) IDO and PD-L1 responses in PBMCs at baseline and 3 and 6 months after last vaccine measured by in vitro IFN- γ Elispot assay (n=30). 2.5–3.2x105 cells per well were used. c) Vaccine associated clones were tracked in the blood by summing the frequency of each rearrangement enriched in either IDO or PD-L1 T cells. Extended Data Fig. 4 Ex vivo vaccine specific responses in blood. Heatmaps of detected specific (background has been subtracted) IDO (a) and PD-L1 (b) responses in PBMCs at baseline and on/after treatment as measured by IFNγ ELISPOT (n=25). C) Example of well images of ex vivo ELISPOT wells for three different patients (n=3). 6-9x105 cells per well were used. Extended Data Fig. 5 CD4 and CD8 vaccine specific T cell responses in blood. Top: heatmaps of IDO specific CD4 (left) and CD8 (right) T cell responses in PBMCs at baseline and on/post treatment. Bottom: heatmaps of PD-L1 specific CD4 (left) and CD8 (right) T cell responses. Responses quantified by flow cytometry by an increased expression of CD107α, CD137 and TNFα after 8h peptide stimulation. Values represent specific responses after background values have been subtracted (n=21). Statistical analysis were performed using two-sided Wilcoxon matched-paired rank test. Extended Data Fig. 6 Pro-inflammatory profiles of sorted CD4 and CD8 T cells from blood. a/b/c/d) Percentage of in vitro stimulated and sorted PD-L1 and IDO specific CD4 and CD8 T cells secreting cytokines. Extended Data Fig. 7 Vaccine specific responses in skin. a) IDO and PD-L1 specific T cell responses in SKILs after 6 series of treatment measured by IFN-γ Elispot assay (n=13). SKILs were grown from DTH injection with either IDO peptide, PD-L1 peptide or a mix as presented by different blue colours. * Responses were calculated as the difference between average numbers of spots in wells stimulated with IDO or PD-L1 peptide (triplicates) and corresponding control (DMSO) and statistical analyses of Elispot responses were performed using distribution-free resampling method (Moodie et al.). DR: Not statistically confirmed response due to replicate number but number of spots in peptide wells are two times higher than control wells (DMSO). NS: No significant response and no DR. b/c/d) Percentage of cytokine secreting/CD107a+ CD4+ and CD8+ IDO and PD-L1 specific T cells in response to in vitro peptide stimulation by flow cytometry. e) Skin infiltrating PD-L1 specific T cell clones also found in biopsy in patient MM01. TCR sequencing was performed on a PD-L1 specific T cell culture generated from DTH area on the lower back injected with PD-L1 peptide on patient MM01. Bars show the frequency of top 25 clones in the culture with which indicates a high Simpson clonality of 0.43. f) Tracking the frequency of the top five skin infiltrating PD-L1 specific clones in tumour before and after treatment. Extended Data Fig. 8 T cell changes in blood after treatment. T cell fraction, TCR clonality and repertoire richness in blood and peripherally expanded TCR clones in both responding and non-responding patients. a) T cell fraction in peripheral blood of five patients at baseline, series 3, 6 and 12 by TCR sequencing. T cell fraction was calculated by taking the total number of T cell templates and dividing by the total number of nucleated cells (n=5.) b) TCR clonality in peripheral blood of five patients at baseline, series 3, 6 and 12 by TCR sequencing (n=5). Simpson clonality measures how evenly TCR sequences are distributed amongst a set of T cells where 0 indicate even distribution of frequencies and 1 indicate an asymmetric distribution where a few clones dominate. c) TCR repertoire richness in peripheral blood of five patients at baseline, series 3, 6 and 12 by TCR sequencing (n=5). TCR repertoire richness report the mean number of unique rearrangements. d) Bar-chart chart representing dynamics of the expanded T cell clones. Coloured bars represent peripherally expanded clones in five patients at series 3, series 6 and series 12 as compared to the baseline PBMC samples (n=5). Light gray bar represents peripherally expanded clones that are present in baseline biopsy samples while dark gray bar represents peripherally expanded clones that are present in the post-treatment biopsy (after 6 series). e) Frequency of the dominant TCR β chain in clonal PD-L1 and IDO specific cultures as determined by CDR3 sequencing. Extended Data Fig. 9 Profiling of genes relevant to T cell activation, cytokines and exhaustion markers on pre- and post-treatment biopsies in two patients. a) RNA expression profiling of genes related to T cell activation was performed using NanoString nCounter (n=2). b) RNA expression profiling of genes related to cytokine activity was performed using NanoString nCounter (n=2). c) RNA expression profiling of genes related to checkpoint inhibitors was performed using NanoString nCounter (n=2). Extended Data Fig. 10 Treatment induced upregulation of PD-L1, IDO, MHC I and MHC II and distance between CD8 T cells and PD-L1 expressing cells. a) IHC on 4 paired biopsies stained for CD3 and CD8 T cells, PD-L1, MHC I and MHC II on tumor cells/mm2 (sum in the validated area) and IDO H-score (from 0 to 300). H-score is the expression of IDO on both immune and tumor cells: The score is obtained by the formula: 3 x percentage of cells with a strong staining + 2 x percentage of cells with a moderate staining + percentage of cells with a weak staining (n=4). b) Distance in µm between CD8+ T cells and PD-L1+ stained cells on five baseline biopsies detected by IHC (n=5). Extended data is available for this paper at 10.1038/s41591-021-01544-x. Supplementary information The online version contains supplementary material available at 10.1038/s41591-021-01544-x. Acknowledgements We thank all patients and their relatives for being a part of the trial. We thank Ö. Met, M. Jonassen, S. Ullitz Færch, B. Saxild, S. Wendt and C. Grønhøj for technical support. We thank M. Cumberbatch for input with translational analysis. We thank the nurses at clinic 5 and the head of the Oncology Department at Herlev and Gentofte Hospital, L. Sengeløv. The study was funded through a research funding agreement between IO Biotech and the CCIT-DK, Herlev Hospital and the Oncology Department at Herlev Hospital. Author contributions I.M.S. and M.H.A. conceived the study. J.W.K. wrote the protocol with input from I.M.S., M.H.A. and M.-B.Z. J.W.K. and C.L.L. were responsible for patient treatment and patient care with help from R.B.H. and consulting from I.M.S., E. Ellebaek and M.D. I.M.S., E. Ellebaek and M.D. were responsible for patient recruitment. J.W.K. and C.L.L. were responsible for coordinating trial procedures and collecting data and samples. J.W.K. and C.L.L. were clinical investigators, and I.M.S. was a sponsor. J.W.K. and C.L.L. analyzed and interpreted clinical data with support from I.M.S. and E. Ehrnrooth. T.W.K. performed statistical analyses. H.W.H. helped with clinical imaging. J.W.K., E.M., C.O.M., S.M.A., S.E.W.-B., M.O.H. and A.W.P. analyzed and interpreted translational data with support from M.H.A. and I.M.S. Funding was acquired by I.M.S., M.H.A. and M.-B.Z. J.W.K. wrote the manuscript with input from I.M.S., M.H.A., E.M. and A.W.P. All authors reviewed the manuscript, interpreted data and approved the final version. Data availability This clinical trial was registered at https://www.clinicaltrials.gov/ before patient enrollment (clinical trial identifier NCT03047928). Each vaccine was composed of 100 µg IO102, a 21-amino-acid peptide (DTLLKALLEIASCLEKALQVF) from the IDO peptide, and 100 µg IO103, a 19-amino-acid peptide (FMTYWHLLNAFTVTVPKDL) from the signal peptide of PD-L1. TCR sequencing data are available from Adaptive Biotechnologies. Upon request, the CCIT-DK office will provide a username and a password to access the designated data within approximately 2–4 weeks (https://www.herlevhospital.dk/ccit-denmark/find-us/Sider/Contact-information.aspx). All requests for the remaining data including raw data and analyzed data and materials will, within a reasonable time frame, be reviewed by the CCIT-DK office (https://www.herlevhospital.dk/ccit-denmark/find-us/Sider/Contact-information.aspx) to verify whether the request is subject to any intellectual property or obligations. Patient-related data not included in the paper were generated as part of clinical trials and may be subject to patient confidentiality. Any data and materials that can be shared will be released via a material-transfer agreement. The following database was used in the study: https://research.regionh.dk/da/publications/the-danish-metastatic-melanoma-database-dammed(32749d99-095f-4cae-b5de-769bae27f01e).html. Competing interests M.H.A. is named as an inventor on various patent applications relating to therapeutic uses of IDO and PD-L1 peptides. These patent applications are assigned to the company IO Biotech, which is developing immune-modulating cancer treatments. M.H.A. is a founder, shareholder and advisor for IO Biotech. I.M.S. is a cofounder, shareholder and advisor for IO Biotech. I.M.S. has an advisory board relationship with or lectured for Roche, Novartis, MSD, Celgene, Incyte, TILT Bio, Pfizer and BMS AstraZeneca and has received limited grants for translational research from BMS, Roche and Novartis. M.-B.Z. is the CEO, founder and shareholder at IO Biotech. E.M., A.W.P. and E. Ehrnrooth are employees at IO Biotech. E. Ellebaek has received honoraria from BMS, Pierre Fabre, Roche and Kyowa Kirin and travel support from MSD. M.D. has received honoraria from Genzyme, MSD, BMS, Roche and Novartis and travel support from Novartis, MSD, BMS, Roche and Pfizer. The remaining authors declare that they have no conflict of interest. Peer review information Nature Medicine thanks George Coukos, Michael Schell and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. Javier Carmona was the primary editor on this article and managed its editorial process and peer review in collaboration with the rest of the editorial team. Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. These authors contributed equally: Julie Westerlin Kjeldsen, Cathrine Lund Lorentzen. Change history 3/8/2022 A Correction to this paper has been published: 10.1038/s41591-022-01771-w

Fatal

ReactionOutcome

CC BY

34887574

2021-12-09

What was the outcome of reaction 'Urosepsis'?

A phase 1/2 trial of an immune-modulatory vaccine against IDO/PD-L1 in combination with nivolumab in metastatic melanoma. Anti-programmed death (PD)-1 (aPD1) therapy is an effective treatment for metastatic melanoma (MM); however, over 50% of patients progress due to resistance. We tested a first-in-class immune-modulatory vaccine (IO102/IO103) against indoleamine 2,3-dioxygenase (IDO) and PD ligand 1 (PD-L1), targeting immunosuppressive cells and tumor cells expressing IDO and/or PD-L1 (IDO/PD-L1), combined with nivolumab. Thirty aPD1 therapy-naive patients with MM were treated in a phase 1/2 study ( https://clinicaltrials.gov/ , NCT03047928). The primary endpoint was feasibility and safety; the systemic toxicity profile was comparable to that of nivolumab monotherapy. Secondary endpoints were efficacy and immunogenicity; an objective response rate (ORR) of 80% (confidence interval (CI), 62.7-90.5%) was reached, with 43% (CI, 27.4-60.8%) complete responses. After a median follow-up of 22.9 months, the median progression-free survival (PFS) was 26 months (CI, 15.4-69 months). Median overall survival (OS) was not reached. Vaccine-specific responses assessed in vitro were detected in the blood of >93% of patients during vaccination. Vaccine-reactive T cells comprised CD4+ and CD8+ T cells with activity against IDO- and PD-L1-expressing cancer and immune cells. T cell influx of peripherally expanded T cells into tumor sites was observed in responding patients, and general enrichment of IDO- and PD-L1-specific clones after treatment was documented. These clinical efficacy and favorable safety data support further validation in a larger randomized trial to confirm the clinical potential of this immunomodulating approach. pmcMain Despite remarkable advances in the treatment of MM with immune checkpoint inhibitors (ICIs) targeting PD-1 and cytotoxic T lymphocyte antigen 4 (CTLA-4), around half of patients are resistant to ICI monotherapy1. The combination of anti-CTLA-4 (aCTLA-4) and aPD1 therapy is to date the most effective therapy resulting in a response rate of around 60%; however, 50% of the patients also develop severe adverse events2,3. Therefore, an equally effective but less toxic treatment is highly needed. Several approaches to enhance ICI efficacy are currently being investigated, such as other ICIs, T cell therapy with tumor-infiltrating T cells or innate immunity stimulators such as Toll-like receptor 9 agonists4–6. Treating cancer patients with vaccines that stimulate a targeted immune response is another attractive approach, with very few side effects observed thus far in combination immunotherapy studies7,8. Immune-modulatory vaccines targeting tumoral immune escape mechanisms offer a new, generalizable strategy compared to patient-specific neoantigen cancer vaccines7,9. The immune-modulatory vaccine strategy in this clinical trial is based on the finding of circulating cytotoxic T cells specific to IDO and PD-L1 in the blood of patients with cancer and, to a lesser extent, in healthy donors. IDO- and PD-L1-specific CD8+ T cells can directly recognize and kill IDO+ and/or PD-L1+ tumor cells and likewise recognize and kill non-malignant cells that express their cognate targets. Furthermore, IDO- and PD-L1-specific CD4+ T cells release pro-inflammatory cytokines in response to IDO- or PD-L1-expressing target cells. IDO and PD-L1 are expressed not only by melanoma cells but also by many other cell types in the tumor microenvironment (TME), which differentiates these antigens from traditional tumor antigens used in other studies10–15. Activation of IDO/PD-L1-specific T cells by vaccination can therefore restrict the range of immunosuppressive signals mediated by immunosuppressive cells and thereby revert the TME from an immune hostile to an immune friendly environment. In animal models of cancer, vaccinations with IDO epitopes resulted in anti-tumor therapeutic effects that were correlated with reductions in IDO expression in myeloid cell populations within the TME16. The IDO/PD-L1 immune-modulating vaccine may lead to a translatable strategy for improving the efficacy of aPD1 therapy through activation of specific T cells. We hypothesize that the IDO/PD-L1 vaccine attracts T cells into the tumor, which induces type 1 helper T (TH1) cell inflammation and reverts the TME into an immune-permissive site, thereby turning the tumor ‘hot’. This would also upregulate PD-L1 expression in cancer and immune cells, generating more susceptible targets to aPD1 therapy (Extended Data Fig. 1a). This theory was confirmed in a mouse model in which aPD1 therapy and IDO vaccination show synergistic effects16. In this phase 1/2 clinical trial, MM1636, patients with MM received a combination of the IDO/PD-L1 (IO102/IO103) peptide vaccine with the adjuvant Montanide and the aPD1 antibody nivolumab. Patients were included in three cohorts: 30 aPD1 therapy-naive patients (cohort A), ten aPD1 therapy-refractory patients (cohort B, de novo resistance) and ten patients who progressed after aPD1 therapy (cohort C, acquired resistance). Here, we report results from cohort A. The vaccine was given biweekly for the first six administrations and thereafter every 4th week. A maximum of 15 vaccines were administered. Nivolumab was given in parallel, biweekly (3 mg per kg) or every 4th week (6 mg per kg) for up to 2 years (Extended Data Fig. 1b). The primary objective was safety and feasibility. Secondary objectives were immunogenicity and clinical efficacy. Results Patients and treatment Thirty patients were enrolled from December 2017 to June 2020. None of the 30 patients dropped out of the study; all received at least three cycles of therapy (Supplementary Fig. 1). At the current database lock (5 October 2020), six patients were still on treatment in the trial. Of the 24 patients who were not on trial treatment at data cutoff, two are still receiving nivolumab monotherapy (6 mg per kg every 4 weeks). Reasons for stopping treatment for the remaining 22 patients included disease progression (37%), toxicity (20%), maximum benefit or complete response (CR) confirmed on two consecutive scans (17%) or completing 2 years of treatment (7%). For the 24 patients who were not on trial treatment at data cutoff, the mean number of vaccinations was 10.5 (range, 3–15 vaccinations). Thirteen of these 24 patients continued nivolumab (6 mg per kg, every 4 weeks) as a standard of care. Nine patients received subsequent therapy after progression (Supplementary Table 1). Baseline characteristics are shown in Table 1. The mean age was 70 years; 37% of patients had elevated lactate dehydrogenase (LDH) levels, 60% were stage M1c, 37% had BRAF mutations, and 43% were negative for PD-L1 (<1%). A total of three patients (10%) had received prior ipilimumab therapy. No patients had brain metastasis (Supplementary Table 2).Table 1 Baseline patient characteristics (n = 30) Characteristic Number (%) Mean age, years (range) 70 (46–85) Sex, male 16 (55%) ECOG PS 0 26 (87%) LDH levels  ≤ULN 19 (63%)  >ULN 11 (37%) M stage (AJCC-8)  M1a 6 (20%)  M1b 6 (20%)  M1c 18 (60%) Number of lesion sites  1 6 (20%)  2–3 17 (57%)  >3 7 (23%) Liver metastases present 10 (33%) BRAF status  Mutant 11 (37%)  Wild type 19 (63%) PD-L1  <1% 13 (43%)  >1% 17 (57%) Previous systemic therapy  Ipilimumab 3 (10%)  No 27 (90%) AJCC-8, eighth edition of the American Joint Committee on Cancer; ECOG PS, Eastern Cooperative Oncology Group performance status; ULN, upper limit of normal. Notable clinical responses to the combination therapy Thirty patients with MM were treated with the IDO/PD-L1 vaccine and nivolumab according to the trial protocol. By investigator review, the ORR reached 80% (CI, 62.7–90.5%), with 43% of patients (CI, 27.4–60.8%) achieving a CR and 37% (CI, 20.9–54.5%) reaching a partial response (PR) as the best overall response, while 20% experienced progressive disease (PD) according to Response Evaluation Criteria in Solid Tumors (RECIST) 1.1 (Fig. 1a). Two of the patients with a PR did not have a confirmatory scan stating PR on two consecutive scans. Early onset of response was frequent, with 22 of 30 patients having an objective response at the first evaluation (after 12 weeks on treatment). Median times to PR and CR were 75 d (range, 54–256 d) and 327 d (range, 73–490 d), respectively (Fig. 2a–c).Fig. 1 Clinical response. a, Pie charts with percent ORR, CR, PR and PD according to RECIST 1.1 by investigator review of all patients (n = 30), PD-L1+ patients (>1%, (n = 17)) and PD-L1− patients (<1%, n = 13)), respectively. Two-sided CIs (95%) were constructed using the Clopper–Pearson method. b, Treatment effect in MM1636 compared with a matched historical control group from the DAMMED database (n = 74). Patients in MM1636 (n = 29) were matched with the exact same combination variable according to age (≤70 years, >70 years), sex, LDH levels (normal, elevated), M stage (M1a, M1b, M1c), BRAF status (wild type, mutated) and PD-L1 status (<1%, ≥1%). Estimates for treatment effects were calculated by weighted logistic regression analyses and weighted Cox proportional hazard model. Bar height indicates the estimated response rate; tops of bars are centers for error bars. Odds ratios (OR), response rates and their corresponding 95% CIs were extracted from the regression model. All P values were two sided, and P values below 0.05 were considered statistically significant. c, Best change in the sum of target lesion size compared with that at baseline (n = 30). The horizontal line at −30 shows the threshold for defining an objective response in the absence of non-target disease progression or new lesions according to RECIST 1.1. Two patients with 100% reduction in target lesion size had non-target lesions present. White stars, six patients had normalization (<10 mm) of fluor-18-deoxyglucose (FDG)-negative lymph nodes (at baseline, lymph nodes were >1.5 cm and FDG+) and 100% reduction of non-lymph node lesions and are considered to have had a CR (green bar). Black star, one patient (MM29) was considered to have had a PR (blue bar), although he had not reached a −30% change in target lesion size. The patient had a single measurable 13-mm lung metastasis at baseline and multiple biopsy-verified cutaneous metastases on the left crus (not detectable by positron-emission tomography–computed tomography (PET–CT) at baseline). The best change in target lesion size was 10 mm, and a post-treatment biopsy from the cutaneous metastases showed no sign of malignancy. Thus, overall, the patient was classified as having a PR. d, Kaplan–Meier curve of the response duration in the 24 patients with an objective response. e, Kaplan–Meier curve of PFS in all 30 treated patients. f, Kaplan–Meier curve of OS in all 30 treated patients. Fig. 2 Clinical response. a, Swimmer plot showing response duration and time to response according to RECIST 1.1 for all treated patients (n = 30). Triangles indicate first evidence of PR, while squares indicate first evidence of CR. Closed circles indicate time of progression. Arrows indicate ongoing responses. Patient MM18 died due to nivolumab-induced side effects. b, Spider plot showing response kinetics in all treated patients (n = 30). Red squares indicate time of progression. c, PET–CT images of patient MM42 before and after treatment (after 12 series of treatment) showing FDG metabolism in target lesions. SC, subcutaneous. The ORR among PD-L1+ (>1% (clone 28.8)) patients (n = 17) was 94.1% (CI, 73–99.7%) and 61.5% (CI, 35.5–82.3%) in PD-L1− patients (n = 13) (Fig. 1a). Objective responses were observed in patients irrespective of human leukocyte antigen (HLA) genotype (Supplementary Fig. 2). Clinical response data were validated by blinded independent external review, in which an ORR of 76.6% (CI, 57.7–90.1%) was reported, with 53.3% of patients achieving a CR, 23.3% achieving a PR and 3.3% experiencing stable disease. Comparisons between investigator review and external review are outlined in Supplementary Table 3. To examine whether the very high response rate should be attributed to nivolumab or to the vaccine, we retrieved clinical response information from a matched historical control group using the Danish Metastatic Melanoma Database (DAMMED) from contemporaneously treated patients with stage 3–4 melanoma who received aPD1 monotherapy17. Patients were matched with the exact same combination variable according to age, sex, PD-L1 status, BRAF status, LDH level and M stage (those at stage M1d were excluded from the control group (no patients with brain metastasis)). Matched controls were identified for 29 patients, and the ORR of 79.3% (CI, 61.0–90.4%) observed in MM1636 was found to be significantly higher (P < 0.0012) than that in the matched control group in which an ORR of 41.7% (CI, 31.0–53.3%) was reached. Furthermore, of the 29 patients in MM1636, a significantly (P < 0.0017) higher percentage (41.4% (CI, 25.2–59.6%)) of patients achieved CR in MM1636 than that (12% (CI, 6.3–21.6%)) in the matched historical control group. ORR and complete response rate (CRR) in the matched historical control group were comparable to those of patients treated in randomized phase 3 pivotal trials with aPD1 monotherapy18 (Fig. 1b). The treatment leads to prolonged PFS At data cutoff, the median duration of response had not been reached, with 87% of all responding patients being free from progression at 12 months (Fig. 1d). Patients were followed for up to 35 months with a median follow-up time of 22.9 months (CI, 14.9–26.2 months). OS and PFS were calculated from the first day of treatment to death or progression or to the date of the last follow-up (5 October 2020). The median PFS (mPFS) for all treated patients was 26 months (CI, 15.4–69 months) and was not reached for responding patients (Fig. 1e). The median OS was not reached at the data cutoff. OS at 12 months was 81.6% (CI, 61.6–92%) (Fig. 1f). One patient (MM18) with a CR died from nivolumab-related severe adverse events; the remaining patients died because of metastatic melanoma. For comparison, the mPFS was 8.3 months (CI, 5.5 months–NR (not reached)) in the matched historical control group (n = 74), while the median OS was 23.2 months (CI, 23.2 months–NR). (Extended Data Fig. 2a,b). The combination of the IDO/PD-L1 vaccine and nivolumab was safe Treatment-related adverse events are listed for all 30 patients in Supplementary Table 4. Common treatment-related grade 1–2 toxicities were fatigue (47%), rash (47%), arthralgia (30%), diarrhea (23%), nausea (23%), dry skin (20%), pruritus (20%), infusion reaction (17%), xerostomia (17%) and myalgia (17%). Four patients (13%) experienced grade 3–4 adverse events: one patient with a grade 3 maculopapular rash (MM01), one patient with grade 3 adrenal insufficiency (MM06) and one patient with grade 3 arthralgia (MM22). Patient MM18 died from urosepsis with multi-organ failure and severe hyponatremia. This patient had experienced multiple immune-related adverse events with grade 3 colitis, grade 2 pneumonitis, grade 3 arthralgia, grade 2 vasculitis and grade 2 nivolumab infusion-related allergic reaction. Additionally, patient MM18 had symptoms of myocarditis at the time of death with highly elevated cardiac troponin I levels. Bedside echocardiography showed an ejection fraction of 15%, which at baseline was 60%, but an autopsy was not conducted, and myocarditis was never confirmed pathologically. Patient MM06 had received first-line treatment with ipilimumab before entering the trial and was on substitution corticosteroids at the time of inclusion. Adrenal insufficiency was aggravated by an erysipelas infection with high fever, reaching grade 3 in Common Terminology Criteria for Adverse Events and resolving quickly after appropriate antibiotic therapy was initiated. As expected, local side effects were common with 77% of the patients who developed injection site reactions. These reactions were classified as granulomas (63%), redness (20%), pain (13%) and pruritus (13%) at the injection site. All local reactions were grade 1–2, most likely related to the Montanide adjuvant and typically transient. However, two patients (MM07 and MM20) decided to discontinue vaccination after eight and 11 injections, respectively, due to granulomas, tenderness and pain that limited instrumental activities of daily living but continued to receive nivolumab (Extended Data Fig. 2c). Vaccine-specific responses in blood were frequently detected First, all 30 patients were assessed for the presence of vaccine-specific responses in peripheral blood mononuclear cells (PBMCs) before, on and after vaccination using a modified interferon (IFN)-γ enzyme-linked immune absorbent spot (ELISPOT) assay. This assay is known to expand antigen-specific memory cells and to improve the detection and the correlative power of ELISPOT during treatment19–21. Prevaccine IDO-specific responses were detectable in ten (33%) patients, while prevaccine PD-L1-specific responses were detectable in eight patients (27%); overlapping (specific to both IDO and PD-L1) prevaccine responses were present in four (13.3%) patients. During vaccination, an increase of IDO-specific T cells or PD-L1-specific T cells in the blood was observed in 28 (93%) and 26 (86%) patients, respectively. In total, 93% of patients had an increase in either PD-L1- or IDO-specific responses on vaccination (Fig. 3a), with a significant (P < 0.0001) median increase from baseline to the post-vaccine response demonstrated for both IDO and PD-L1 (at different time points on treatment), confirming that vaccine-specific immune responses were induced in patients regardless of clinical response (Fig. 3b,c). Immune responses fluctuated in the blood over time (Extended Data Fig. 3a). An increase in IDO- and PD-L1-specific responses in the peripheral blood was also detectable directly ex vivo across the different clinical response groups (Extended Data Fig. 4).Fig. 3 Vaccine-specific responses in blood. a, IDO- and PD-L1-specific T cell responses in PBMCs at baseline and on vaccination as measured by the IFN-γ ELISPOT assay (n = 30). *Responses were calculated as the difference between the average numbers of spots in wells stimulated with IDO or PD-L1 peptide (triplicates) and those from the corresponding control (DMSO), and statistical analyses of ELISPOT responses were performed using a distribution-free resampling method (Moodie et al.50). DR (double response), response was not statistically confirmed due to replicate number, but the number of spots in peptide wells was two times higher than that in control wells (DMSO). NS, no significant response and no DR. For a detailed overview of responses at serial time points on vaccination, see Extended Data Fig. 4a. b, IDO- and PD-L1-specific T cell response in PBMCs in all treated patients measured by the IFN-γ ELISPOT assay at baseline and on vaccination. On-vaccination responses were selected from the ‘best’ ELISPOT response at different time points for each patient (series 3, 6, 12, 18 or 24) during vaccination (n = 30). Wilcoxon matched-pairs signed-rank test was used to compare responses to IDO or PD-L1 peptides in the vaccine between baseline and later time points. c, Representative example of ELISPOT wells with response in patient MM23 in serial PBMCs before and on treatment. d, IDO-specific CD4+ and CD8+ T cells were isolated and expanded from PBMCs stimulated in vitro with the IDO peptide and a low dose of IL-2 for 14–15 d before sorting using the Miltenyi Cytokine Secretion Assay—Cell Enrichment and Detection kit. To assess their cytolytic potential, IDO-specific T cells were stimulated with IDO peptide, and expression of CD107a, IFN-γ and TNF-α was assessed by flow cytometry (the example is from patient MM14). Sustained vaccine-specific responses were observed 3 and 6 months after the last vaccine, indicating induction of memory responses in nine patients with response to clinical treatment who surpassed follow-up at data lock (Extended Data Fig. 3b). Importantly, PD-L1- and IDO-specific responses were observed irrespective of HLA genotype (Supplementary Table 5). To verify the functionality of vaccine-induced T cells, IDO- or PD-L1-specific T cells were isolated and in vitro expanded from PBMCs of five patients. Phenotypic characterization by flow cytometry revealed that the isolated vaccine-specific T cells consisted of both CD4+ and CD8+ T cells. Also, both IDO- and PD-L1-specific CD4+ and CD8+ T cells showed pro-inflammatory properties, as they expressed the cytolytic marker CD107a and secreted the cytokines IFN-γ and tumor necrosis factor (TNF)-α (Fig. 3d and Extended Data Fig. 6a–d). Interestingly, we were also able to detect vaccine-specific CD4+ and CD8+ T cell responses in peripheral blood ex vivo (Extended Data Figs. 4 and 5). We observed a significant increase in the overall percentage of CD107a, CD137 and TNF-α expression in response to peptide stimulation in on- or post-treatment PBMC samples compared to that at baseline, further confirming the expansion and the diverse signature of vaccine-specific T cells (Extended Data Fig. 5). Vaccine-specific responses detected in the skin at the vaccination site To investigate whether vaccine-specific T cells have the potential to migrate to peripheral tissue, delayed-type hypersensitivity (DTH) tests were performed after six cycles of treatment on 15 patients to assess the presence of vaccine-reactive T cells in the skin. Supplementary Table 6 display an overview of skin-infiltrating lymphocyte (SKIL) cultures. We detected IDO-specific T cells in the skin of six of ten patients and PD-L1-specific T cells in nine of 11 patients (Extended Data Fig. 7a). Intracellular cytokine staining was performed on SKILs from five patients after stimulation with either PD-L1 or IDO peptide. Here, we detected mainly CD4+ peptide-reactive T cells that secreted TNF-α and upregulated CD107a, and a minor fraction also secreted IFN-γ. In one patient, we detected CD8+ PD-L1-reactive T cells (Extended Data Fig. 7b–d). Vaccine-induced T cells specifically recognize target cells To confirm the functionality of vaccine-expanded T cells, vaccine-specific T cell clones (clonal purity was confirmed by T cell receptor (TCR) sequencing) were isolated and expanded from patient PBMCs (Extended Data Fig. 8e). We showed that PD-L1-specific T cells were able to recognize PD-L1+ autologous tumor cells in a PD-L1 expression-dependent manner if the cancer cells also expressed HLA-II (Fig. 4a,b). Similarly, an HLA-DR-restricted IDO-specific CD4+ T cell clone was able to recognize an HLA-DR-matched IDO-expressing model cell line, MonoMac1, in an IDO-expression-dependent manner (Fig. 4e–g). As previously described, IDO- and PD-L1-specific T cells’ mode of action is not limited to targeting only cancer cells. We were able to show that vaccine-specific T cell clones also reacted against PD-L1- and IDO-expressing autologous immune cells (Fig. 4c,h). To provide myeloid cells with a tumor-associated phenotype we treated isolated CD14+ myeloid cells with tumor conditioned medium (TCM) derived from an established autologous tumor cell line. We observed that such TCM-treated CD14+ cells had increased expression of PD-L1 and IDO and were effectively recognized by autologous PD-L1- and IDO-specific CD4+ T cell clones (Fig. 4c,d,h,i).Fig. 4 PD-L1- and IDO-specific T cells from vaccinated patients react against PD-L1- and IDO-expressing target cells. a, Left, PD-L1-specific T cell culture (MM1636.05) reactivity against PD-L1 peptide or autologous tumor cells in the IFN-γ ELISPOT assay. Tumor cells were either not treated or pretreated with 200 U ml−1 IFN-γ for 48 h before the assay. Effector:target (E:T) ratio of 10:1 was used. Right, PD-L1 and HLA-II surface expression on melanoma cells with (green) or without (yellow) pretreatment with IFN-γ compared to an isotype control (gray) as assessed by flow cytometry. b, Left, PD-L1-specific T cell (MM1636.05) reactivity in the IFN-γ ELISPOT assay against autologous tumor cells pretreated with IFN-γ (500 U ml−1) and transfected with mock or PD-L1 small interfering (si)RNA 24 h after transfection. E:T ratio, 10:1. Right, PD-L1 surface expression on melanoma tumor cells (MM1636.05) assessed by flow cytometry 24 h after transfection with mock (blue) or PD-L1 (red) siRNA compared to the isotype control (gray). c, Reactivity of the CD4+ PD-L1-specific T cell clone (MM1636.14) against PD-L1 peptide or autologous CD14+ cells; E:T ratio, 10:1. CD14+ cells were isolated using magnetic bead sorting and used as targets in an ELISPOT assay directly or after pretreatment for 2 d with TCM derived from the autologous tumor cell line. d, Quantitative PCR with reverse transcription (RT–qPCR) analysis of PD-L1 (CD274) expression in sorted CD14+ cells before and after treatment with autologous TCM for 48 h. e, Reactivity of the IDO-specific CD4+ T cell clone (MM1636.23) against IDO peptide combined with HLA-DR (L243)-, HLA-DQ (SPV-L3)- or HLA-DP (B7/21)-blocking antibodies (aHLA-DR, aHLA-DQ and aHLA-DP) in an intracellular staining assay (ICS) for IFN-γ and TNF-α production. T cells were incubated with individual blocking antibodies (2 μg ml−1) for 30 min before adding IDO peptide. f, Reactivity of the IDO-specific CD4+ T cell clone (MM1636.23) against the HLA-DR-matched IDO-expressing cell line MonoMac1 transfected with mock or IDO siRNA in an ICS assay, E.T ratio, 4:1. siRNA transfection was performed 48 h before the experiment. g, RT–qPCR analysis of IDO1 expression in MonoMac1 cells 48 h after siRNA transfection. h, Reactivity of the CD4+ IDO-specific T cell clone (MM1636.14) against IDO peptide or autologous CD14+ cells; E:T ratio, 20:1. CD14+ cells were isolated using magnetic bead sorting and used as targets in an ELISPOT assay directly or after pretreatment with TCM derived from the autologous tumor cell line. i, RT–qPCR analysis of IDO1 expression in sorted CD14+ cells before and after treatment with autologous TCM for 48 h. Bars in RT–qPCR data (d,g,i) represent the mean of three (d,i) or six (g) technical replicates ±s.d.; P values were determined by two-tailed parametric t-tests. ELISPOT counts (a,b,c,h) represent the mean value of three technical replicates ±s.e.m.; response P values were determined using the distribution-free resampling (DFR) method. TNTC, too numerous to count. T cell clones in blood and tumors To track the role of treatment-induced T cell responses, TCR sequencing of the complementarity-determining region 3 (CDR3) was performed on five patients in peripheral blood (baseline and cycles 3, 6 and 12) and paired biopsies. These five patients (MM01, MM02, MM08, MM09 and MM13) were selected due to the availability of material and to investigate a balanced patient group with both responders and non-responders. Due to a limited number of available paired biopsies (either as a consequence of patient refusal or unexpectedly rapid and substantial clinical responses), statistical analysis could not be applied. Details on clinical response are shown in Fig. 2a. Additionally, PBMCs (on treatment) or SKILs were stimulated with the IDO/PD-L1 peptides, and then cytokine-producing T cells were sorted to track vaccine-induced T cells both in the periphery and at the tumor site. To identify enriched IDO/PD-L1-specific T cell clones, TCR rearrangements in sorted IDO/PD-L1-specific clones and TCR rearrangements in sorted IDO/PD-L1-specific T cell samples were compared to sequences from baseline PBMC samples for each patient. Clonal expansion of vaccine-specific TCR rearrangements from samples on vaccination were then tracked using a differential abundance framework. Cumulative IDO/PD-L1-specific T cell frequencies were tracked in post-treatment samples. We found no relation between clinical response and the enrichment of vaccine-specific clones, but an increase in IDO/PD-L1-specific T cell clones was observed at different time points in the periphery in all five patients (Extended Data Fig. 3c). We next investigated overall changes in the T cell repertoire in the blood. A modest increase in the peripheral T cell fraction was observed in the three responding patients at cycle 3, while the two non-responding patients had a clear decrease in T cell fraction (Extended Data Fig. 8a). We thereafter investigated TCR clonality and TCR repertoire richness, exploring the proportion of abundant clones and the number of unique rearrangements, respectively. A decreasing peripheral Simpson clonality and increasing TCR repertoire richness was observed in responding patients at cycle 3, which might indicate tumor trafficking upon treatment. The opposite pattern was observed in non-responding patients (Extended Data Fig. 8b,c). Peripherally expanded clones were associated with tumors and persisted until cycle 12 (latest time point analyzed). The largest peripheral expansion was observed at cycle 3, with the most significant increase observed in patient MM01 (CR). Responding patients had a larger fraction of peripherally expanded clones that were also found in tumors compared to that of non-responders. By tracking peripherally expanded clones detected at the tumor site, we observed that patient MM01 had a substantial increase after treatment, indicating tumor trafficking of peripheral expanded clones (Extended Data Fig. 8d). Influx at the tumor site of vaccine-enriched T cell clones Given the observation of increased T cell fraction and enrichment of IDO- and PD-L1-specific clones in the blood after treatment, we investigated whether the same trend was observed at the tumor site. Both TCR sequencing and immunohistochemistry (IHC) of paired biopsies from the five patients described above showed an increase in the T cell fraction with an influx of CD3+ and CD8+ T cells after treatment in the three responding patients (Fig. 5a–c). IHC was not possible for patient MM09 due to tissue loss.Fig. 5 Changes in the TME after treatment. Number of CD3+ and CD8+ T cells, TCR fraction, TCR clonality, TCR repertoire, biopsy expanded TCR clones and enrichment of IDO- and PD-L1-specific T cells at the tumor site. a, Number of CD3+ and CD8+ T cells per mm2 (sum in the validated area) at the tumor site detected by IHC of paired biopsies from four patients. b, Example of IHC of CD3+ and CD8+ T cells at the tumor site before and after treatment (cells per mm2) in one patient (MM01). c, T cell fraction at the tumor site at baseline and cycle 6 by TCR sequencing. The T cell fraction was calculated by taking the total number of T cell templates and dividing by the total number of nucleated cells. d, Tracking of vaccine-associated clones at baseline and cycle 6 in tumor biopsies. Cumulative frequencies of IDO and PD-L1 vaccine-specific TCR rearrangements are represented. e,f, TCR clonality and TCR repertoire richness in five patients at the tumor site at baseline and cycle 6. Simpson clonality measures how evenly TCR sequences are distributed among a set of T cells, where 0 indicates an even distribution of frequencies and 1 indicates an asymmetric distribution in which a few clones dominate. TCR repertoire richness reports the mean number of unique rearrangements. g, Bar chart representing baseline expanded biopsy clones from five patients (colored bars) and the detection of biopsy expanded clones also found in the blood at baseline and series 3, 6 and 12 (white and gray bars) by TCR sequencing. We thereafter investigated whether some of the IDO/PD-L1 vaccine-linked T cells were present at the tumor site. Vaccine-associated clones were tracked as the combined frequency of IDO- and PD-L1-specific T cell rearrangements. In biopsies, the frequencies at cycle 6 were compared to those at baseline and showed an increase in vaccine-specific T cells in four of five patients, irrespective of clinical response (Fig. 5d). TCR sequencing of PD-L1-specific SKILs (a more specific culture than IDO/PD-L1-specific isolations derived from PBMCs on vaccination) and paired biopsies showed that two of the top five PD-L1-specific SKIL clones were present at the tumor site both before and after treatment (Extended Data Fig. 7e,f). With a focus on the more abundant T cell clones, we investigated overall TCR clonality at the tumor site before and after treatment. In addition, we explored the number of unique TCR rearrangements, dissecting the lower-frequency clones. Patient MM01 had a significant increase in TCR clonality and a decrease in repertoire richness at the tumor site after therapy, indicating a focused tumor repertoire response of selected clones. All three responding patients had a decrease in TCR repertoire richness, which again might indicate a focused tumor response (Fig. 5e,f). Deeper analyses showed that the T cell clones that expanded at the tumor site after therapy were also present in the blood at baseline and increased significantly after treatment in four of five patients. The highest proportion was detected early at cycle 3. These data again support trafficking of peripherally expanded clones to the tumor site and could indicate that the T cell response to treatment is derived from pre-existing peripherally tumor-associated T cells (Fig. 5g). Signs of treatment-induced inflammation in the TME To dissect changes in the TME induced by T cell influx upon treatment in responding patients, RNA gene expression analyses using the nCounter PanCancer Immune Profiling Panel from NanoString were performed on paired biopsies from two responding patients (MM01 and MM13). Expression of genes related to adaptive immunity such as T cell activation, effector functions (genes encoding IFN-γ, TNF-α, IL-15 and IL-18) and cytotoxicity was increased in post-treatment biopsies (Extended Data Fig. 9a,b). Also, expression of genes related to checkpoint inhibitors such as those encoding T cell immunoglobulin and mucin-domain containing-3 (TIM-3), IDO, PD-L1, PD-L2, PD-1 and CTLA-4 increased after treatment, indicating activation of immune cells in the TME (Extended Data Fig. 9c). Additionally, IHC of paired biopsies from four patients (MM01, MM02, MM05 and MM13) showed an upregulation of PD-L1, IDO, MHC-I and MHC-II on tumor cells, indicating a treatment-induced pro-inflammatory response in the three responding patients, except for a decrease in MHC-II expression in patient MM13. By contrast, the non-responding patient MM02 had a reduction in T cell numbers present in the tumor after treatment and no expression of PD-L1, IDO or MHC-II and, interestingly, total loss of MHC-I, demonstrating tumor immune escape (Extended Data Fig. 10a). CD8+ T cells and their distance (µm) to PD-L1-expressing cells in baseline biopsies from five patients was investigated by IHC. Except for patient MM13 (PR), distance and clinical responses were associated: the two responders had reduced distance (<20 µm) between cells expressing these markers compared to non-responding patients (>80 µm). This observation indicates that responding patients not only have a higher intratumoral infiltration of CD8+ T cells, but that these cells can surround and attack PD-L1-expressing immune cells and tumor cells (Extended Data Fig. 10b). Discussion In this clinical trial, MM1636, 30 patients with metastatic melanoma were treated with a first-in-class immunomodulatory IDO/PD-L1-targeting peptide vaccine combined with nivolumab. The treatment led to an unprecedented high ORR of 80%, with 43% of patients reaching a CR, and a striking mPFS of 26 months (95% CI, 15.4–69 months) was reached. The vaccine represents a new treatment strategy to activate specific T cells that target cells contributing to immune suppression (including tumor cells), positively modulating the TME by inducing local inflammation. Indeed, we show that vaccine-specific T cells isolated and expanded from vaccinated patients recognize not only tumor cells in a target- and HLA-restricted manner but also myeloid cells polarized toward a tumor-associated phenotype. Hence, myeloid cells become targets for vaccine-activated T cells when they have a tumor-associated phenotype in the TME. These phenomena may further induce checkpoint molecules and rewire the TME toward an increasingly aPD1-permissive state. A drawback of our study is the single-center nonrandomized setup. Comparison between trials or between patients in trials and real-world patients is problematic due to multiple factors, such as period of conduction and other therapies available at the different periods of time. Nevertheless, the rate of investigator-assessed ORR in the phase 3 trial CheckMate 067 was 43.7% in the nivolumab monotherapy group and 57% in the nivolumab and ipilimumab group. CRs occurred in 8.9% and 11.5% of patients, respectively18. The mPFS of 26 months (95% CI, 15.4–69 months) in this trial is more than twice as long as that for patients treated with nivolumab and ipilimumab in CheckMate 067, for which an mPFS of 11.5 months (95% CI, 8.7–19.3 months) was reached. Patient baseline characteristics were in general comparable to those of patients with MM who have been treated in CheckMate 067, although patients in MM1636 were older (mean age, 70 years) and a larger fraction were positive for PD-L1 (57%)3,18,22. Among patients with PD-L1-negative tumors in MM1636, an ORR of 61.5% was still reached, which would be expected to be around 33% for first-line nivolumab monotherapy23. Some studies suggest that older patients might have a tendency to respond better to aPD1 therapy; however, this is still debated24–27. To address potential trial bias and the nonrandomized setup, patients in MM1636 were matched for age, performance status, sex, M stage, LDH level, PD-L1 status and BRAF status with a historical control group from the DAMMED, who were treated contemporarily (2015–2019) with aPD1 monotherapy as standard of care17. We found a significantly higher ORR and CRR in MM1636 compared to those of matched patients, who had an ORR of 43% and a CRR of 13%, comparable to patients treated in CheckMate 067. Restrictions of the synthetic control group are of course that it is partially historic and patient selection outside matching criteria cannot be ruled out28. Numerous contemporary clinical trials are exploring the combination of aPD1 therapy with other immunomodulating agents for advanced melanoma. Talimogene laherparepvec, an oncolytic virus, is approved by the Food and Drug Administration and the European Medicines Agency to treat advanced melanoma. A small phase 1b trial with 21 patients (MASTERKEY-265) combined talimogene laherparepvec and pembrolizumab to treat patients with advanced unresectable melanoma and reached an ORR of 62% and a CR of 33%29,30. Seventy-one percent of the patients in this trial had an M stage below M1c; this number was 40% in our trial. Furthermore, mainly patients with an M stage below M1c responded to treatment, which was not the case in MM1636. Results from a large randomized phase 3 trial are awaited (KEYNOTE-034). Epacadostat, an IDO inhibitor, was tested in combination with pembrolizumab in a nonrandomized phase 2 trial in 40 aPD1 treatment-naive patients with MM with promising results, reaching an ORR of 62%. Unfortunately, the phase 3 trial showed no indication that epacadostat provided improvement in PFS and OS31. Limitations of the phase 3 trial were the sparse information on pharmacodynamics as well as biomarker evaluation to improve the design. The IDO/PD-L1 vaccine is different from epacadostat, as it is not an IDO inhibitor but targets IDO- and PD-L1-expressing cells. Similar vaccines administered as monotherapy induced objective responses in lung cancer and basal cell carcinoma, while epacadostat as monotherapy in 52 patients resulted in no responses32,33. Sahin et al. recently published encouraging data from a first-in-human trial, in which a vaccine containing liposomal RNA targeting four unmutated tumor-associated antigens (NY-ESO-1, MAGE-A3, tyrosinase and TPTE) was administered alone or in combination with nivolumab in patients with advanced melanoma. Responses were observed in both the monotherapy group as well as the combination group in checkpoint inhibitor-experienced patients, suggesting the efficacy of non-mutant shared tumor antigen vaccines34. The overall safety and tolerability findings are comparable to those of aPD1 monotherapy. Injection site reactions were exclusive to the vaccine. However, these side effects were transient and mild in most patients and most likely due to the adjuvant Montanide. IDO- and PD-L1-specific CD8+ and CD4+ T cells exist among peripheral blood lymphocytes in healthy donors35–38 and expand in response to pro-inflammatory stimuli35,39. Furthermore, both IDO and PD-L1 are induced in cells as a counter response to the inflammatory response. This provides a mechanism that ensures immune homeostasis, which keeps IDO/PD-L1-specific T cells in check; therefore, the risk of triggering autoimmune-related adverse events by vaccination appears to be minimal. This was confirmed in the first clinical trials of IDO and PD-L1 vaccination (NCT01543464 (refs. 33,40) and NCT03042793 (ref. 41)). The spontaneous (baseline) immune response to the vaccines observed in the current study is in agreement with our previous observation in various patients with cancer35–38. Numerous vaccine-induced changes in the blood and at the tumor site were observed. Peripheral IDO- and/or PD-L1-specific T cells were detected in vitro in a modified ELISPOT in over 93% of patients on vaccination, unrelated to patient HLA type. Immune responses were persistent in patients who surpassed follow-up at data cutoff and were still detectable up to 6 months after the last vaccine, suggesting induction of memory T cells. We did not observe a correlation between vaccine-induced responses in blood and clinical responses. However, the detection of a highly significant increase in vaccine-specific T cell numbers after vaccination in almost all patients together with the very low number of patients with PD makes it difficult to expect such a correlation, especially because other aspects should be taken into account (for example, the loss of class I expression on tumors cells in patient MM02 with PD after vaccination). Frequencies of peripheral T cells induced against PD-L1 were overall higher than those against IDO. Importantly, however, we also observed that it was not the same patients who responded strongly to both IDO and PD-L1, nor did patients react with a similar response pattern to the two antigens; that is, different time points of response were observed. This suggests that each component of the vaccine plays different roles in the ongoing immune response in patients. The goal of IDO/PD-L1 vaccination was to modulate the TME to increase responsiveness to aPD1 therapy, as we have observed in animal models, in which immune conversion is demonstrated in the TME with an increased influx of T cells. Targeting both IDO and PD-L1 together enables synergy, as the TME is known to use different immune escape mechanisms and IDO and PD-L1 are often overexpressed by different cellular compartments. Ex vivo ELISPOT and flow cytometry assays confirmed the induction of immune response toward both epitopes and the higher immunogenicity of the PD-L1 epitope, although because of lower sensitivity failed to detect some of the responses. TCR sequencing of five patients confirmed enrichment of IDO/PD-L1-specific T cell clones in the blood at different time points after treatment. Furthermore, an increase in enriched IDO/PD-L1-specific clones was observed in four of five patients at the tumor site after treatment, irrespective of clinical response. Phenotypic characterization showed that vaccine-specific T cells, which were expanded in vitro with interleukin (IL)-2 from the blood of vaccinated individuals, were both CD4+ T cells and CD8+ T cells. This was confirmed by ex vivo phenotype description of vaccine-activated T cells. Vaccine-specific T cells expressed CD107a and CD137 and produced IFN-γ and TNF-α upon stimulation with the cognate target, indicating their cytolytic capacity. Overall, the data from immune monitoring supports the importance of both antigens in the generation of the frequent clinical responses in the study. Despite a limited number of paired biopsies (due to either patient refusal or the fact that a large fraction of responding patients had no assessable tumors after six cycles), we observed trends indicating treatment-induced general T cell influx in responding patients. It was shown that proliferation of CD8+ T cells in the tumor after aPD1 treatment is associated with radiographic reduction in tumor size42. Additionally, we showed that a large proportion of expanded peripheral TCR clones were associated with tumors and the most considerable amount of clonal expansion was observed early, at cycle 3. For the patient with a CR included in the TCR sequencing analyses (MM01), the number of peripheral expanded clones present at the tumor site increased after treatment compared to that at baseline, indicating tumor trafficking of peripheral expanded clones. Gene expression analyses (two paired biopsies) and IHC (five paired biopsies) further demonstrated that the combination treatment induced a pro-inflammatory TME in responding patients with signs of T cell activation and cytotoxicity and increased cytokine activity. This may lead to further upregulation of IDO, PD-L1, MHC-I and MHC-II on tumor cells, leading to more treatment targets. It was shown that, following vaccination with a cancer vaccine, PD-L1 expression is increased on tumor cells due to recruitment of tumor-specific T cells and upregulation of adaptive immune resistance pathways in the TME43. Treatment with nivolumab monotherapy enhances PD-L1 expression, and it is therefore problematic to discriminate the effect of the vaccine as compared to that of nivolumab44. IDO- and PD-L1-specific pro-inflammatory effector T cells were hypothesized to counteract the functions of IDO- and PD-L1-expressing immune-suppressive cells as a means to keep the immune balance between immune activation and inhibition45. However, because the expression of these molecules is induced as a counter-regulatory response to inflammatory mediators such as IFNs, they can also be expressed by, for example, activated myeloid cells or T cells. Thus, activation of PD-L1-specific T cells may result in depletion of activated T cells in the tumor and other sites. Importantly, it was recently described that PD-L1+ T cells mainly have tolerogenic effects on tumor immunity and exert tumor-promoting properties, suggesting that targeting this immune population is indeed also beneficial46. We have previously investigated the effect of activating PD-L1-specific T cells in vitro and in vivo and found that, overall, they supported the effector phase of an immune response by removing PD-L1-expressing regulatory immune cells that inhibit PD-1+ effector T cells47,48. The major role of the PD-1 pathway is believed to be the regulation of effector T cell responses. Thus, this protective pathway is more important after activation, rather than at the initial T cell-activation stage49. Accordingly, the presence of PD-L1-specific T cells during the activation phase of an immune response may not increase or support a pro-inflammatory immune response due to the expression of PD-L1 on potent antigen-presenting cells or on the T cells themselves14. Thus, the overall effects of PD-L1-specific T cells may vary depending on the expression of both PD-1 and PD-L1, that is, due to the microenvironment and the state of the immune response. In conclusion, here we report an impressive response rate, CR rate and mPFS for a first-in-class immune-modulating vaccine combined with nivolumab. This may be a first step toward a new treatment strategy for patients with MM. Limitations are the low number of patients treated at a single institution and the lack of a randomized design with aPD1 monotherapy as comparator. Studies in aPD1 therapy-resistant or -refractory melanoma are ongoing as well as biomarker analysis for selecting patients at a higher likelihood to benefit from the combination versus aPD1 monotherapy. A larger randomized trial will be essential to validate these findings and determine the specific contribution of the vaccine to clinical responses and changes in the TME. In December 2020, the Food and Drug Administration granted breakthrough therapy designation for the IO102/IO103 vaccine combined with aPD1 therapy in metastatic melanoma based on data from the MM1636 trial. Methods Trial design and treatment plan MM1636 is an investigator-initiated, nonrandomized, open-label, single-center phase 1/2 study. All patients were treated at the Department of Oncology, Herlev and Gentofte Hospital, University of Copenhagen, Herlev, Denmark. The study was conducted according to the Declaration of Helsinki and Good Clinical Practice (GCP) and monitored by the GCP unit in Copenhagen, Denmark. The protocol was approved by the ethical committee of the Capital region of Denmark (H-17000988), the Danish Medical Agencies (2017011073) and the Capital Region of Denmark Data Unit (P-2019-172). The study was registered at https://clinicaltrials.gov/ under the identifier NCT03047928 and at the EudraCT (no. 2016-0004527-23). This study initially aimed to include 30 aPD1 treatment-naive patients with MM. An amendment with the addition of two other cohorts with ten patients in each cohort was made to evaluate immune responses and clinical efficacy in aPD1 therapy-resistant patients (cohort B, de novo resistance and cohort C, acquired resistance) for a total of 50 patients. The amendment with cohorts B and C was approved at the inclusion of 18 patients in cohort A. The trial is still including patients in cohorts B and C. This article reports results from cohort A. The first six patients were treated as for phase 1, evaluating for safety and tolerability before the remaining 24 patients were included in phase 2. The IDO/PD-L1 vaccine was administered subcutaneously biweekly for the first 10 weeks and thereafter every 4 weeks for approximately 9 months. A maximum of 15 vaccines were administered. Nivolumab was administered according to the approved label (3 mg per kg biweekly) for 24 cycles. The 15th vaccine was administered together with the 24th nivolumab cycle of 3 mg per kg, and responding patients thereafter continued nivolumab monotherapy every 4th week (6 mg per kg) as a standard of care after investigator assessment. Treatment with nivolumab was discontinued at the maximum benefit (assessed by the investigator), after a maximum of 2 years of therapy, at progression or due to severe adverse events (Extended Data Fig. 1b, treatment plan). Vaccine composition Each vaccine was composed of 100 µg IO102, a 21-amino-acid peptide (DTLLKALLEIASCLEKALQVF) from IDO, and 100 µg IO103, a 19-amino-acid peptide (FMTYWHLLNAFTVTVPKDL) from the signal peptide of PD-L1 (PolyPeptide Laboratories). Peptides were dissolved separately in 50 µl DMSO, filtered for sterility and frozen at −20 °C (NUNC CyroTubes CryoLine System Internal Thread, Sigma-Aldrich). At <24 h before administration, peptides were thawed. The PD-L1 peptide was diluted in 400 µl sterile water and, immediately before injection, mixed with the IDO peptide solution and 500 µl Montanide ISA-51 (SEPPIC) to achieve a total volume of 1 ml. Patients Patients above 18 years of age with locally advanced or stage 4 melanoma according to the AJCC (seventh edition), at least one measurable lesion according to RECIST 1.1 and an ECOG PS of 0–1 were eligible. The main exclusion criteria were prior treatment with aPD1 therapy, CNS metastases >1 cm, severe comorbidities and active autoimmune disease. Enrollment was not restricted to PD-L1 status, but it was known before inclusion. Patients were included after informed consent. Patients MM42 and MM20 have confirmed their approval of PET–CT images and clinical images that have been included in the article. Key study assessments Safety and tolerability were evaluated based on changes in clinical laboratory analyses and reported adverse events. Adverse events were assessed according to Common Terminology Criteria for Adverse Events (version 5.0) and were graded from 1 to 5 for all treated patients up to 6 months after the last dose of the IDO/PD-L1 vaccine. Clinical efficacy was assessed using FDG PET–CT scans before treatment and every 3rd month until progression. Objective responses were categorized into CR, PR, stable disease or PD according to RECIST version 1.1. Clinical data were collected at CCIT-DK in the eCRF program OpenClinica version 1.0 and in Microsoft Excel version 2002 on a secure server. Blood samples for immunologic analyses were collected before treatment, before the third cycle, after the sixth, 12th, 18th and 24th cycles (on vaccination) and 3 and 6 months after the last vaccine. Two to three tumor needle biopsies (1.2 mm) were collected at baseline and after six cycles from the same tumor site, when accessible, to evaluate immune responses at the tumor site. DTH skin tests were performed and punch biopsies were taken from the DTH area after cycle 6 for the evaluation of SKILs reactive to PD-L1 and IDO (Supplementary Fig. 1, treatment plan). Statistical analysis of clinical outcome Survival curves were computed in GraphPad Prism software version 9.0.0 according to Kaplan–Meier method. Median follow-up time of enrollment was calculated using the reverse Kaplan–Meier method, also in GraphPad Prism software version 9.0.0. For binary outcomes, 95% two-sided CIs were constructed using the Clopper–Pearson method, also in GraphPad version 9.0.0. An independent board of certified and experienced oncoradiologists performed an external review to evaluate clinical response to address the potential bias of investigator site review. The external review took place at Rigshospitalet, Copenhagen University Hospital. This hospital did not participate in the MM1636 trial, and the external reviewer had no prior knowledge about the clinical trial or the trial therapy. Only PET–CT images were accessed, and arrows indicating target or non-target lesions appeared on baseline images as the only additional information. To address potential trial bias regarding treatment effect, we matched patients in MM1636 with patients from the DAMMED, a population-based database that retrospectively collects data on patients with metastatic melanoma in Denmark. Here, data from 938 patients treated with aPD1 monotherapy contemporaneously (January 2015–October 2019) were extracted. Two hundred and eighteen of these patients were eligible for comparison and matching (all parameters available) (Supplementary Table 1), and 74 patients from the DAMMED were found to match. Patients were matched for age (≤70 years, >70 years), sex, LDH levels (normal, elevated), M stage (M1a, M1b, M1c), BRAF status (wild type, mutated) and PD-L1 status (<1%, ≥1%). An exact matching algorithm was used in which patients in MM1636 were matched with patients from the DAMMED with the same combination of variables. Twenty-nine patients from MM1636 were matched with exact combinations of the six variables. One patient could not be matched. To secure balance of the calculations, control patients were weighted according to the number of patients for each MM1636 patient. Estimates for treatment effects were calculated by weighted logistic regression analyses and weighted Cox proportional hazard model. The R package ‘Matchlt’ was used for matching patients. As the method chosen for matching control patients to protocol patients, a weighted binary logistic regression model was used for comparing response rates in the two matched cohorts. Odds ratios and response rates, including their corresponding 95% CIs, were extracted from the regression models. All P values were two sided, and P values below 0.05 were considered statistically significant. SAS version 9.4M5 was used for the weighted logistic regression models. Processing of peripheral blood mononuclear cells Peripheral blood was collected from all patients in heparinized tubes and was processed within 4 h. In brief, PBMCs were isolated using Lymphoprep (Medinor) separation. PBMCs were counted on a Sysmex XP-300 analyzer and frozen in Human AB Serum (Sigma-Aldrich, H4522-100ML) with 10% DMSO using controlled-rate freezing (CoolCell, BioCision) in a −80 °C freezer and were moved the next day to a freezer at −140 °C until further processing. Needle biopsies at baseline and after the sixth vaccine Two to three needle biopsies (1.2 mm) were taken at baseline and after the sixth cycle of treatment when assessable from the same tumor lesion. One fragment was fixed with formalin and embedded in paraffin (FFPE); one to two fragments were used for expanding tumor-infiltrating T cells and establishing autologous tumor cell lines. Delayed-type hypersensitivity and generation of skin-infiltrating lymphocytes After six cycles of treatment, we performed intradermal injections of vaccine components without adjuvant and one control injection containing DMSO without peptide. Patients were injected with either a mixture of both IDO and PD-L1 peptides at all three injection sites or PD-L1 peptide, IDO peptide or a mixture of the two at the injection sites, respectively (Supplementary Table 6). Eight hours after injection, punch biopsies were resected from the three sites where the peptide was injected and transported immediately to the laboratory and cut into fragments of 1–2 mm3. SKILs were expanded to establish ‘young SKILs’ in CM medium consisting of RPMI 1640 with GlutaMAX, 25 mM HEPES, pH 7.2 (Gibco, 72400-021), IL-2 (100 or 6,000 IU ml−1) (Proleukin Novartis, 004184), 10% heat-inactivated Human AB Serum (Sigma-Aldrich, H4522-100ML), 100 U ml−1 penicillin, 1.25 µg ml−1 Fungizone (Bristol Myers Squibb, 49182) and 100 µg ml−1 streptomycin (Gibco, 15140-122). Half of the medium was replaced three times per week. Young SKILs from samples derived from IDO peptide, PD-L1 peptide and the mixture injection sites were further expanded in a small-scale version of the 14-d rapid expansion protocol as previously described51. Quantification of specific T cells in blood by ELISPOT For enumeration of vaccine-specific T cells in the peripheral blood, PBMCs from patients were stimulated with IDO or PD-L1 peptide in the presence of a low dose of IL-2 (120 U ml−1) for 7–13 d before being used in IFN-γ ELISPOT assays with 2.5–3.2 × 105 cells per well. Briefly, cells were placed in a 96-well PVDF membrane-bottomed ELISPOT plate (MultiScreen MSIPN4W50, Millipore) precoated with IFN-γ-capture antibody (clone 1-D1K, Mabtech). Diluted IDO or PD-L1 peptide stocks in DMSO were added at 5 µM; an equivalent amount of DMSO was added to control wells. PBMCs from each patient were set up in duplicate or triplicate for peptide and control stimulations. Cells were incubated in ELISPOT plates in the presence of the peptide for 16–18 h, after which they were washed off, and biotinylated secondary antibody (anti-human IFN-γ mAB, clone 7-B6-1, Mabtech) was added. After 2 h of incubation, unbound secondary antibody was washed off, and streptavidin-conjugated alkaline phosphatase (Mabtech) was added for 1 h. Next, unbound streptavidin-conjugated enzyme was washed off, and the assay was developed by adding BCIP/NBT substrate (Mabtech). ELISPOT plates were analyzed on the CTL ImmunoSpot S6 Ultimate V analyzer using ImmunoSpot software version 5.1. Responses were calculated as the difference between the average numbers of spots in wells stimulated with IDO or PD-L1 peptide and those from corresponding control wells. For detection of IDO and PD-L1 peptide responses ex vivo, PBMCs were thawed and rested overnight in medium containing DNase I (1 μg ml−1, Sigma-Aldrich, 11284932001) before being used in IFN-γ ELISPOT assays as described above. A total of 6–9 × 105 PBMCs per well were seeded. Statistical analysis of all ELISPOT responses was performed using the DFR method as described by Moodie et al. using RStudio software (RStudio Team, 2016, http://www.rstudio.com/)50. Vaccine-specific ELISPOT responses were defined as true if the difference between the spot count in control and peptide-stimulated wells was statistically significant according to DFR statistical analysis or, for samples performed in duplicate, if the spot count in peptide-stimulated wells was at least 2× the spot count in control wells50. Cancer cell lines and tumor conditioned medium Autologous melanoma cell lines were established from needle biopsies. Briefly, biopsies were chopped into small fragments and seeded in 24-well cultures in RPMI 1640 with GlutaMAX, 25 mM HEPES, pH 7.2 (Gibco, 72400-021), 10% heat-inactivated FBS (Life Technologies, 10500064), 100 U ml−1 penicillin, 1.25 µg ml−1 Fungizone (Bristol Myers Squibb, 49182) and 100 µg ml−1 streptomycin (Gibco, 15140-122). Established adherent melanoma tumor cell lines were cryopreserved at −140 °C in freezing medium containing FBS with 10% DMSO. PD-L1 and HLA-II expression on established tumor cell lines was assessed by flow cytometry staining with anti-PD-L1–PE-Cy7 (1:23 dilution, clone M1h1 (RUO), BD, 558017) and anti-HLA-II–FITC (1:23 dilution, clone Tu39 (RUO), BD, 555558) antibodies. To obtain TCM, established tumor cell lines were cultured in 175-cm2 Nunc cell culture flasks until 80–90% confluency was reached. The culture medium was then replaced with 20 ml fresh X-VIVO 15 with Gentamicin and Phenol Red (Lonza, BE02-060Q), medium with 5% heat-inactivated Human AB Serum (Sigma-Aldrich, H4522-100ML). After 24 h of incubation, TCM was collected and centrifuged to remove any resuspended cells, after which TCM was aliquoted, frozen and stored at −80 °C. The acute monocytic leukemia cell line MonoMac1 was obtained from the DSMZ—German Collection of Microorganisms and Cell Cultures (ACC 252) and cultured in RPMI 1640 with GlutaMAX, 25 mM HEPES, pH 7.2 (Gibco, 72400-021) and 10% heat-inactivated FBS. Isolation of autologous myeloid cells Autologous CD14+ cells were sorted from freshly thawed PBMCs using a magnetic bead-separation kit (Miltenyi Biotec, 130-050-201) according to the manufacturer’s instructions. Isolated CD14+ cells were used as targets in the IFN-γ ELISPOT assay directly after sorting or differentiated in vitro into tumor-associated macrophages by culturing with 1 ml fresh X-VIVO 15 medium with Gentamicin and Phenol Red (Lonza, BE02-060Q) and 5% heat-inactivated Human AB Serum, supplemented with 1 ml autologous TCM in 24-well plates for 2 d. Quantification of vaccine-specific T cells from DTH biopsy sites by ELISPOT SKILs expanded in vitro were rested in medium without IL-2 overnight before being used in the IFN-γ ELISPOT assay as described above to evaluate reactivity of skin-infiltrating T cells. Generation of IDO- and PD-L1-specific T cell cultures from PBMCs or SKILs IDO- or PD-L1-specific T cells were isolated from peptide-stimulated in vitro PBMC cultures on days 14–15 after stimulation or from SKIL cultures expanded in vitro. For specific T cell isolation, PBMCs or SKILs were stimulated with IDO or PD-L1 peptide, and cytokine-producing T cells were sorted using the IFN-γ or TNF-α Secretion Assay—Cell Enrichment and Detection kit (Miltenyi Biotec). Cytokine-production profile of PD-L1- and IDO-specific T cells by intracellular staining To assess the T cell cytokine-production profile, isolated and expanded IDO- and PD-L1-specific T cell cultures were stimulated for 5 h with peptide at 5 μM in a 96-well plate. One hour after the start of the incubation, anti-CD107a–PE (1:133 dilution, clone H4A3, BD Biosciences, 555801) antibody and BD GolgiPlug (1:1,000 dilution, BD Biosciences) were added. After a 5-h incubation, cells were stained using fluorescently labeled surface marker antibodies: anti-CD4–PerCP (1:5 dilution, clone SK3, 345770), anti-CD8–FITC (1:5 dilution, clone SK1, 345772) and anti-CD3–APC-H7 (1:20 dilution, clone SK7, 560275) (all from BD Biosciences). Dead cells were stained using FVS510 (1:250 dilution, BD Biosciences, 564406), followed by overnight fixation and permeabilization using eBioscience Fixation/Permeabilization buffers (00-5123-43, 00-5223-56) according to the manufacturer’s instructions. Cells were then stained intracellularly in eBioscience permeabilization buffer (00-8333-56) with anti-IFN-γ–APC (1:20 dilution, clone 25723.11, BD Biosciences, 341117) and anti-TNF-α–BV421 (1:100 dilution, clone Mab11, BD Biosciences, 562783) antibodies. Samples were analyzed on the FACSCanto II (BD Biosciences) using BD FACSDiva software version 8.0.2. The gating strategy is shown in Supplementary Fig. 3a. For assessing the reactivity against IDO expressing MonoMac1 cells, the appropriate amount of cancer cells was added to the IDO-specific T cells to obtain an effector:target ratio of 4:1 in the 96-well plate and T cell cytokine production was tested after 5 h of co-culture as described above. For HLA-blocking experiments, HLA-blocking antibodies (2 μg ml−1) were added directly into wells 30 min before the addition of peptide. Blocking antibodies used were HLA-DR (1:500 dilution, clone L243, Abcam, ab136320), HLA-DQ (1:500 dilution, clone SPV-L3, Abcam, ab23632) and HLA-DP (1:500 dilution, clone B7/21, Abcam, ab20897). To assess ex vivo T cell reactivity to IDO and PD-L1 peptides in patient PBMCs, cells were thawed and rested for 1–2 d in medium containing DNase I (1 μg ml−1, Sigma-Aldrich, 11284932001). PBMCs were then stimulated with peptide at 5 μM in a 96-well plate for 8 h. An hour after the addition of peptide, anti-CD107a–BV421 (3:50 dilution, clone H4A3, BD Biosciences, 328626) antibody and BD GolgiPlug (1:1,000 dilution, BD Biosciences) were added. Surface and intracellular staining was performed as described above. Antibodies used for surface staining were anti-CD3–PE-CF594 (0.8:30 dilution, clone UCHT1, BD Biosciences, 562280), anti-CD4–BV711 (1:30 dilution, clone SK3, BD Biosciences, 563028) and anti-CD8–Qdot605 (1:150 dilution, clone 3B5, Thermo Fisher, Q10009). Antibodies used for intracellular staining were anti-CD137–PE (1:20 dilution, clone 4B4-1 (RUO), BD Biosciences, 555956), anti-IFN-γ–PE-Cy7 (1.5:20 dilution, clone B27, BD Biosciences, 557643) and anti-TNF-α–APC (1:20 dilution, clone Mab11, BD Biosciences, 554514). Samples were acquired on the NovoCyte Quanteon (ACEA Biosciences) and analyzed using NovoExpress software version 1.4.1. To assess vaccine-specific T cell responses, background values observed in unstimulated PBMC samples were subtracted from values observed in peptide-stimulated conditions. Positive response value threshold was set at a difference of 0.2% from the background values. Based on this response cutoff, only TNF-α, CD107a and CD137 responses were detected in this assay. Statistical analysis comparing baseline with on-treatment or post-treatment cytokine profiles was performed using two-sided Wilcoxon matched-pairs signed-rank test. The gating strategy is shown in Supplementary Fig. 3b. siRNA-mediated PD-L1 and IDO1 silencing A stealth siRNA duplex for targeted silencing of PD-L1 (Invitrogen)52, a custom silencer select siRNA for targeted silencing of IDO1 (Ambion) and the recommended silencer select negative control (Ambion) siRNA for mock transfection were used. The stealth PD-L1 siRNA duplex consisted of the sense sequence 5′-CCUACUGGCAUU-UGCUGAACGCAUU-3′ and the antisense sequence 5′-AAUGCGUUCAGCAAAUGCCAGUAGG-3′. The three silencer IDO siRNA duplexes used were siRNA1 (sense sequence, 5′-ACAUCUGCCUGAUCUCAUATT-3′; antisense sequence, 5′-UAUGAGAUCAGGCAGAUGUTT-3′), siRNA2 (sense sequence, 5′-CCACGAUCAUGUGAACCCATT-3′; antisense sequence, 5′-UGGGUUCACAUGAUCG-UGGAT-3′) and siRNA3 (sense sequence, 5′-CGAUCAUGUGAACCCAAAATT-3′; antisense sequence, 5′-UUUUGGGUUCACAUGAUCGTG-3′). For PD-L1 or IDO1 silencing, cancer cells were electroporated with 0.025 nmol of each siRNA duplex as previously described53. For PD-L1-silencing experiments, cancer cells were treated with IFN-γ (500 U ml−1, PeproTech) 1 h after electroporation. Electroporated cells were used as target cells in ELISPOT and ICS assays 24 h or 48 h after siRNA electroporation. RT–qPCR Total RNA was extracted using the RNeasy Plus Mini kit (Qiagen, 74134) following the manufacturer’s instructions. RNA concentration was quantified using a NanoDrop 2000 (Thermo Fisher Scientific), and a total of 1,000 ng RNA was reverse transcribed with the High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems, 4368814) using 1,000 ng input RNA for transcription. Real-time qPCR analyses were performed using the TaqMan Gene Expression Assay on a Roche LightCycler 480 instrument. RT–qPCR assays were performed with a minimum of three technical replicates and analyzed using the ΔCt method as described in Bookout et al.54 with normalization of IDO1 expression (primer ID, Hs00984148_m1) or PD-L1 expression (primer ID, Hs001125296_m1) to the expression level of the housekeeping gene POLR2A (primer ID, Hs00172187_m1) and the control sample (mock). For low-concentration samples with no amplification, Ct was set to 40. Controls without reverse transcriptase were used as controls for specific amplifications. P values were determined using two-tailed parametric t-tests. BRAF-mutational status and PD-L1 status at baseline in all patients A library of historical FFPE biopsies were assessable for all patients and analyzed locally at Herlev and Gentofte University Hospital by experienced pathologists for BRAF status and PD-L1 expression on tumor cells. The BRAF analysis was carried out with real-time PCR using the EntroGen BRAF Mutation Analysis Kit II (BRAFX-RT64, CE-IVD) to specifically detect mutations corresponding to V600D, V600E and V600K in the BRAF protein. PD-L1 status was assessed using the monoclonal rabbit anti-PD-L1 antibody (clone 28.8, PD-L1 IHC 28-8 pharmDx) in FFPE biopsies following the manufacturer’s instructions. Patients were considered positive for PD-L1 with expression levels ≥1% and negative for PD-L1 with expression levels <1%. Human leukocyte antigen type Blood samples from all 30 patients were genotyped for class I (HLA-A, HLA-B, HLA-C) and three class II types (HLA-DRB1, HLA-DQA1, HLA-DQB1) using LinkSēq HLA Typing Kits (Thermo Fisher, 1580C). These test kits are based on real-time PCR using allele-specific exponential amplification (sequence-specific primers) followed by melting curve analyses. Immunohistochemistry simplex: Immunoscore CR IHC staining was performed using a qualified Ventana Benchmark XT with four different steps: (1) antigen retrieval; (2) staining with the following primary antibodies: anti-CD3 antibody (clone HDX2; provider, HalioDx; HD-FG-000013; 10931065/10636667; concentration, 0.25 µg ml−1), anti-CD8 antibody (clone HDX3; provider, HalioDx; HD-FG-000019; 10931069/10337710/10639301; concentration, 0.5 µg ml−1), anti-IDO monoclonal antibody (clone VINC3IDO; provider, Thermo Fisher Scientific; 14-9750-82; E25003-101; concentration, 0.05 µg ml−1), anti-HLA class 1 ABC antibody (clone, EMR8-5; provider, Abcam; ab70328; GR3248333/GR3186494; concentration, 0.5 µg ml−1), anti-HLA-DR/DP/DQ/DX antibody (clone CR3/43; provider, Santa Cruz; sc-53302; L1714; concentration, 1 µg ml−1) and anti-PD-L1 antibody (clone HDX3; provider, HalioDx; HD-FG-000035; 106312810/106312816; concentration, 3.3 µg ml−1); (3) detection with a secondary antibody using the ultraView Universal DAB Detection kit; and (4) counterstaining using Hematoxylin & Bluing Reagent (staining of cellular nuclei). Control slides were systematically included in each staining run to permit quality control of the obtained measurements. Following coverslipping, slides were scanned with the NanoZoomer-XR to generate digital images (20×). Two consecutive slides were specifically used to perform Immunoscore CR TL staining of CD3+ and CD8+ cells. Digital pathology of T lymphocytes The digital pathology for Immunoscore CR TL allowed the quantification of positive cells (in cells per mm2) into the core tumor and invasive margin if present. Each sample was analyzed using the HalioDx Digital Pathology Platform. Digital pathology and pathologist analysis of IDO Digital pathology for IDO allowed the quantification of stained area (in mm2) into the whole tumor. Each sample was analyzed using the HalioDx Digital Pathology Platform. Analysis of IDO+ cells was performed by a pathologist. Results were expressed as an H score from 0 to 300. The score is obtained by the formula 3 × percentage of cells with strong staining + 2 × percentage of cells with moderate staining + percentage of cells with weak staining. Digital pathologist analysis of MHC-I and MHC-II Analysis of MHC-I+ and MHC-II+ cells was performed by a pathologist. Results are expressed as a percentage of positive tumor cells and a percentage of positive cells in the stroma. Digital pathology Immunoscore immune checkpoint (CD8+ and PD-L1+) The Immunoscore CR IC test allowed the quantification of CD8++ cell density in the whole tumor and a CD8+-centered proximity index (which corresponds to the percentage of CD8++ cells that have at least one PD-L1+ cell in the neighborhood) at different cutoff distances (20 µm, 40 µm, 60 µm and 80 µm). Pathologist analysis of PD-L1 A pathologist performed analysis of PD-L1+ cells. Positivity of a viable tumor cell was considered when partial or complete cell membrane staining was observed (more than 10% of the tumor cell membrane). Results were expressed as a percentage. NanoString RNA profiling and Immunosign RNA was extracted from FFPE tissues using Qiagen RNeasy FFPE extraction kits (Qiagen). Annotations from the pathologist performing H&E staining were used to guide removal of normal tissue from the slides by macrodissection before nucleic acid extraction, which occurred after tissue deparaffinization and lysis. Each extracted RNA sample was independently quantified using a NanoDrop spectrophotometer (NanoDrop Technologies) and qualified (Agilent Bioanalyzer). Degradation was quantified as the percentage of RNA fragments smaller than 300 bp using the RNA 6000 Nano kit (Agilent Bioanalyzer). Good sample quality was defined as less than 50% of RNA fragments being 50–300 bp in size. RNA expression profiling was performed using the nCounter PanCancer Immune Profiling Panel from NanoString (NanoString Technologies). The PanCancer Immune Profiling Panel contains 776 probes and is supplemented with six genes to complete HalioDx Immunosign targets. Hybridization was performed according to the manufacturer’s instructions. Hybridized probes were then purified and immobilized on a streptavidin-coated cartridge using the nCounter Prep Station (NanoString Technologies). Data collection was carried out on the nCounter Digital Analyzer (NanoString Technologies) following the manufacturer’s instructions to count individual fluorescent barcodes and quantify target RNA molecules present in each sample. For each assay, a scan of 490 fields of view was performed. Raw data from the NanoString nCounter were processed using NanoString recommendations. Quality control enables to retain data of good quality with a binding density that ranges between 0.05 and 2.25. The linearity of positive controls was checked using the R2 value of regression between the counts and the concentrations of positive controls. Samples that showed R2 < 0.75 were flagged and removed from the analysis. The background was removed using a thresholding method at the mean + 2 s.d. of negative controls. Raw counts were normalized using a positive normalization factor. Samples showing positive normalization factors outside the range of 0.3–3 were removed from the analysis. A second normalization was performed using the housekeeping gene normalization factor. Only the most stable housekeeping genes were selected for this normalization step using the variance-versus-mean relationship. All samples showing a normalization factor outside the range of 0.1–10 were removed from the analysis. All statistical analyses were performed on normalized counts using R software (version 2.6.2, 2019-12-12). T cell receptor variable β chain sequencing To track longitudinal immune responses to therapy, genomic DNA was extracted from longitudinal pre- and post-treatment PBMCs (five patients), pre- and post-treatment biopsies (five patients) (both FFPE and treated with RNAlater) and IDO- and PD-L1-specific T cell cultures from PBMCs (five patients) or SKILs (one patient). Three clonal (two IDO- and one PD-L1-) specific cultures were also sequenced to confirm clonal purity of cultures. DNA from PBMCs or RNAlater-treated biopsies was extracted with the DNeasy Blood and Tissue kit (Qiagen, 69504), DNA from sorted IDO and PD-L1-specific T cells from either PBMCs or SKILs was extracted using the QIAamp DNA Micro kit (Qiagen, 565304), and DNA from PFFE biopsies was extracted using the Maxwell RSC DNA FFPE kit (Promega, AS1450). Immunosequencing of the CDR3 regions of human TCRβ chains was performed using the immunoSEQÒ Assay (Adaptive Biotechnologies). Extracted genomic DNA was amplified in a bias-controlled multiplex PCR, followed by high-throughput sequencing. Sequences were collapsed and filtered to identify and quantitate the absolute abundance of each unique TCRβ CDR3 region for further analysis as previously described55–57. Statistical analyses of TCRβ sequencing results Two quantitative components of diversity were compared across samples in this study. First, Simpson clonality was calculated on productive rearrangements by ∑i=1Rpi2, where R is the total number of rearrangements and pi is the productive frequency of rearrangement i. Values of Simpson clonality range from 0 to 1 and measure how evenly receptor sequences (rearrangements) are distributed among a set of T cells. Clonality values approaching 0 indicate a very even distribution of frequencies, whereas values approaching 1 indicate an increasingly asymmetric distribution in which a few clones are present at high frequencies. Second, sample richness was calculated as the number of unique productive rearrangements in a sample after computationally downsampling to a common number of T cells to control for variation in sample depth or T cell fraction. Repertoires were randomly sampled without replacement five times, and we report the mean number of unique rearrangements. The T cell fraction was calculated by taking the total number of T cell templates and dividing by the total number of nucleated cells. The total number of nucleated cells was derived from reference genes using the immunoSEQ Analyzer version 3. To identify enriched vaccine clones in each patient, rearrangement frequencies in their baseline PBMCs and each IDO/PD-L1-sorted T cell sample were compared using a binomial distribution framework as previously described58. In brief, for each clone, we performed a two-sided test to determine whether frequencies were the same in the patient’s periphery and a PD-L1- or IDO-specific T cell sample. The Benjamini–Hochberg procedure was used to control the false discovery rate at 0.01 (ref. 59). Clonal expansion in post-treatment samples was similarly assessed using this differential abundance framework, but an IDO/PD-L1-specific T cell sample was replaced with a post-treatment series sample. In biopsies, the six series frequencies were compared to those of baseline tissue. Lastly, vaccine-associated clones were tracked in each PBMC and tissue sample by summing the frequency of each rearrangement enriched in either PD-L1- or IDO-specific T cells. All statistical analyses were performed in R version 3.6.1. Reporting Summary Further information on research design is available in the Nature Research Reporting Summary linked to this article. Online content Any methods, additional references, Nature Research reporting summaries, source data, extended data, supplementary information, acknowledgements, peer review information; details of author contributions and competing interests; and statements of data and code availability are available at 10.1038/s41591-021-01544-x. Supplementary information Supplementary Information Supplementary Figs. 1–3 and Tables 1–6 Reporting Summary Extended data Extended Data Fig. 1 Mode of Action and treatment plan. a) Anticipated mechanism of action of the combination therapy of an IDO/PD-L1 derived peptide vaccine and nivolumab (anti-PD1). 1) The IDO/PD-L1 peptide vaccine is administered subcutaneously (s.c) and nivolumab is administered intravenous (IV). 2) The vaccines peptides are phagocytosed by an antigen presenting cell and presented to IDO and PD-L1 specific T cells, which are activated. 3) The activated T cells migrates to the tumor site where they attack both immune-suppressive cells and tumour cells expressing IDO and/or PD-L1 leading to cytokine production and a pro-inflammatory tumour microenvironment. 4/5) Enhanced tumour killing by both IDO/PD-L1 specific T cells and tumor specific cytotoxic T cells due to PD-1 blockade. Created with BioRender.com b) Treatment plan. After written informed content patients were screened. Before treatment start a baseline PET/CT scan was performed, baseline blood sample for research use and if assessable a baseline needle biopsy. Patients were treated with the IDO/PD-L1 peptide vaccine subcutaneously biweekly for the first 6 injections and thereafter every fourth week for a maximum of 15 vaccines. Nivolumab was administered in parallel biweekly (3mg/kg) up to 24 series. If patients needed subsequent nivolumab after ended vaccination regiment they were treated with 6 mg/kg every fourth week up to two years. Needle biopsy and delayed type hypersensitivity (DTH) was performed after 6 series of treatment if assessable. PET/CT scans was performed every third month. Extended Data Fig. 2 Progression free survival and overall survival in matched historical control group and vaccine injection site reaction. a) Kaplan-Meier curve of progression free survival with corresponding 95% confidence intervals in the matched historical control group (n=74). Patients were matched on BRAF-status, PD-L1-status, age, gender, M-stage and LDH level). b) Kaplan-Meier curve of overall survival with corresponding 95% confidence intervals in the matched historical control group (n=74). c) Images of injection site reaction in patient MM20 (CR) after 11 vaccination show redness, rash and granuloma at injection site. Extended Data Fig. 3 Vaccine specific responses in blood. a) Heatmap of specific (background has been subtracted) IDO and PD-L1 responses in PBMCs at baseline, series 3, 6, 12, 18 and 24 measured by in vitro IFN-γ Elispot assay show fluctuations in the blood during treatment (n=30). b) Heatmap of specific (background has been subtracted) IDO and PD-L1 responses in PBMCs at baseline and 3 and 6 months after last vaccine measured by in vitro IFN- γ Elispot assay (n=30). 2.5–3.2x105 cells per well were used. c) Vaccine associated clones were tracked in the blood by summing the frequency of each rearrangement enriched in either IDO or PD-L1 T cells. Extended Data Fig. 4 Ex vivo vaccine specific responses in blood. Heatmaps of detected specific (background has been subtracted) IDO (a) and PD-L1 (b) responses in PBMCs at baseline and on/after treatment as measured by IFNγ ELISPOT (n=25). C) Example of well images of ex vivo ELISPOT wells for three different patients (n=3). 6-9x105 cells per well were used. Extended Data Fig. 5 CD4 and CD8 vaccine specific T cell responses in blood. Top: heatmaps of IDO specific CD4 (left) and CD8 (right) T cell responses in PBMCs at baseline and on/post treatment. Bottom: heatmaps of PD-L1 specific CD4 (left) and CD8 (right) T cell responses. Responses quantified by flow cytometry by an increased expression of CD107α, CD137 and TNFα after 8h peptide stimulation. Values represent specific responses after background values have been subtracted (n=21). Statistical analysis were performed using two-sided Wilcoxon matched-paired rank test. Extended Data Fig. 6 Pro-inflammatory profiles of sorted CD4 and CD8 T cells from blood. a/b/c/d) Percentage of in vitro stimulated and sorted PD-L1 and IDO specific CD4 and CD8 T cells secreting cytokines. Extended Data Fig. 7 Vaccine specific responses in skin. a) IDO and PD-L1 specific T cell responses in SKILs after 6 series of treatment measured by IFN-γ Elispot assay (n=13). SKILs were grown from DTH injection with either IDO peptide, PD-L1 peptide or a mix as presented by different blue colours. * Responses were calculated as the difference between average numbers of spots in wells stimulated with IDO or PD-L1 peptide (triplicates) and corresponding control (DMSO) and statistical analyses of Elispot responses were performed using distribution-free resampling method (Moodie et al.). DR: Not statistically confirmed response due to replicate number but number of spots in peptide wells are two times higher than control wells (DMSO). NS: No significant response and no DR. b/c/d) Percentage of cytokine secreting/CD107a+ CD4+ and CD8+ IDO and PD-L1 specific T cells in response to in vitro peptide stimulation by flow cytometry. e) Skin infiltrating PD-L1 specific T cell clones also found in biopsy in patient MM01. TCR sequencing was performed on a PD-L1 specific T cell culture generated from DTH area on the lower back injected with PD-L1 peptide on patient MM01. Bars show the frequency of top 25 clones in the culture with which indicates a high Simpson clonality of 0.43. f) Tracking the frequency of the top five skin infiltrating PD-L1 specific clones in tumour before and after treatment. Extended Data Fig. 8 T cell changes in blood after treatment. T cell fraction, TCR clonality and repertoire richness in blood and peripherally expanded TCR clones in both responding and non-responding patients. a) T cell fraction in peripheral blood of five patients at baseline, series 3, 6 and 12 by TCR sequencing. T cell fraction was calculated by taking the total number of T cell templates and dividing by the total number of nucleated cells (n=5.) b) TCR clonality in peripheral blood of five patients at baseline, series 3, 6 and 12 by TCR sequencing (n=5). Simpson clonality measures how evenly TCR sequences are distributed amongst a set of T cells where 0 indicate even distribution of frequencies and 1 indicate an asymmetric distribution where a few clones dominate. c) TCR repertoire richness in peripheral blood of five patients at baseline, series 3, 6 and 12 by TCR sequencing (n=5). TCR repertoire richness report the mean number of unique rearrangements. d) Bar-chart chart representing dynamics of the expanded T cell clones. Coloured bars represent peripherally expanded clones in five patients at series 3, series 6 and series 12 as compared to the baseline PBMC samples (n=5). Light gray bar represents peripherally expanded clones that are present in baseline biopsy samples while dark gray bar represents peripherally expanded clones that are present in the post-treatment biopsy (after 6 series). e) Frequency of the dominant TCR β chain in clonal PD-L1 and IDO specific cultures as determined by CDR3 sequencing. Extended Data Fig. 9 Profiling of genes relevant to T cell activation, cytokines and exhaustion markers on pre- and post-treatment biopsies in two patients. a) RNA expression profiling of genes related to T cell activation was performed using NanoString nCounter (n=2). b) RNA expression profiling of genes related to cytokine activity was performed using NanoString nCounter (n=2). c) RNA expression profiling of genes related to checkpoint inhibitors was performed using NanoString nCounter (n=2). Extended Data Fig. 10 Treatment induced upregulation of PD-L1, IDO, MHC I and MHC II and distance between CD8 T cells and PD-L1 expressing cells. a) IHC on 4 paired biopsies stained for CD3 and CD8 T cells, PD-L1, MHC I and MHC II on tumor cells/mm2 (sum in the validated area) and IDO H-score (from 0 to 300). H-score is the expression of IDO on both immune and tumor cells: The score is obtained by the formula: 3 x percentage of cells with a strong staining + 2 x percentage of cells with a moderate staining + percentage of cells with a weak staining (n=4). b) Distance in µm between CD8+ T cells and PD-L1+ stained cells on five baseline biopsies detected by IHC (n=5). Extended data is available for this paper at 10.1038/s41591-021-01544-x. Supplementary information The online version contains supplementary material available at 10.1038/s41591-021-01544-x. Acknowledgements We thank all patients and their relatives for being a part of the trial. We thank Ö. Met, M. Jonassen, S. Ullitz Færch, B. Saxild, S. Wendt and C. Grønhøj for technical support. We thank M. Cumberbatch for input with translational analysis. We thank the nurses at clinic 5 and the head of the Oncology Department at Herlev and Gentofte Hospital, L. Sengeløv. The study was funded through a research funding agreement between IO Biotech and the CCIT-DK, Herlev Hospital and the Oncology Department at Herlev Hospital. Author contributions I.M.S. and M.H.A. conceived the study. J.W.K. wrote the protocol with input from I.M.S., M.H.A. and M.-B.Z. J.W.K. and C.L.L. were responsible for patient treatment and patient care with help from R.B.H. and consulting from I.M.S., E. Ellebaek and M.D. I.M.S., E. Ellebaek and M.D. were responsible for patient recruitment. J.W.K. and C.L.L. were responsible for coordinating trial procedures and collecting data and samples. J.W.K. and C.L.L. were clinical investigators, and I.M.S. was a sponsor. J.W.K. and C.L.L. analyzed and interpreted clinical data with support from I.M.S. and E. Ehrnrooth. T.W.K. performed statistical analyses. H.W.H. helped with clinical imaging. J.W.K., E.M., C.O.M., S.M.A., S.E.W.-B., M.O.H. and A.W.P. analyzed and interpreted translational data with support from M.H.A. and I.M.S. Funding was acquired by I.M.S., M.H.A. and M.-B.Z. J.W.K. wrote the manuscript with input from I.M.S., M.H.A., E.M. and A.W.P. All authors reviewed the manuscript, interpreted data and approved the final version. Data availability This clinical trial was registered at https://www.clinicaltrials.gov/ before patient enrollment (clinical trial identifier NCT03047928). Each vaccine was composed of 100 µg IO102, a 21-amino-acid peptide (DTLLKALLEIASCLEKALQVF) from the IDO peptide, and 100 µg IO103, a 19-amino-acid peptide (FMTYWHLLNAFTVTVPKDL) from the signal peptide of PD-L1. TCR sequencing data are available from Adaptive Biotechnologies. Upon request, the CCIT-DK office will provide a username and a password to access the designated data within approximately 2–4 weeks (https://www.herlevhospital.dk/ccit-denmark/find-us/Sider/Contact-information.aspx). All requests for the remaining data including raw data and analyzed data and materials will, within a reasonable time frame, be reviewed by the CCIT-DK office (https://www.herlevhospital.dk/ccit-denmark/find-us/Sider/Contact-information.aspx) to verify whether the request is subject to any intellectual property or obligations. Patient-related data not included in the paper were generated as part of clinical trials and may be subject to patient confidentiality. Any data and materials that can be shared will be released via a material-transfer agreement. The following database was used in the study: https://research.regionh.dk/da/publications/the-danish-metastatic-melanoma-database-dammed(32749d99-095f-4cae-b5de-769bae27f01e).html. Competing interests M.H.A. is named as an inventor on various patent applications relating to therapeutic uses of IDO and PD-L1 peptides. These patent applications are assigned to the company IO Biotech, which is developing immune-modulating cancer treatments. M.H.A. is a founder, shareholder and advisor for IO Biotech. I.M.S. is a cofounder, shareholder and advisor for IO Biotech. I.M.S. has an advisory board relationship with or lectured for Roche, Novartis, MSD, Celgene, Incyte, TILT Bio, Pfizer and BMS AstraZeneca and has received limited grants for translational research from BMS, Roche and Novartis. M.-B.Z. is the CEO, founder and shareholder at IO Biotech. E.M., A.W.P. and E. Ehrnrooth are employees at IO Biotech. E. Ellebaek has received honoraria from BMS, Pierre Fabre, Roche and Kyowa Kirin and travel support from MSD. M.D. has received honoraria from Genzyme, MSD, BMS, Roche and Novartis and travel support from Novartis, MSD, BMS, Roche and Pfizer. The remaining authors declare that they have no conflict of interest. Peer review information Nature Medicine thanks George Coukos, Michael Schell and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. Javier Carmona was the primary editor on this article and managed its editorial process and peer review in collaboration with the rest of the editorial team. Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. These authors contributed equally: Julie Westerlin Kjeldsen, Cathrine Lund Lorentzen. Change history 3/8/2022 A Correction to this paper has been published: 10.1038/s41591-022-01771-w

Fatal

ReactionOutcome

CC BY

34887574

2021-12-09

Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Immune thrombocytopenia'.

Sequential Immune Thrombocytopenia (ITP) and Thrombotic Thrombocytopenic Purpura (TTP) in an Elderly Male Patient with Primary Sjogren's Syndrome: When in Doubt, Use the PLASMIC Score. Thrombotic thrombocytopenic purpura (TTP) is a rare, life-threatening thrombotic microangiopathy due to an acquired autoantibody to ADAMTS13 that requires a boutique treatment, urgent plasma exchange. Thus, TTP is often termed a "cannot miss" diagnosis. We report a patient with TTP who had a history of immune thrombocytopenia (ITP), had atypical demographics for TTP, and had also met criteria for primary Sjogren's syndrome. This exceedingly rare combination presented a temptation to dismiss TTP as a diagnosis. Discussion. Our case further demonstrates the practical utility of using the PLASMIC score as a tool that can help identify patients with TTP even when the patient has statistically rare characteristics. pmc1. Introduction Immune thrombocytopenia (ITP) and thrombotic thrombocytopenic purpura (TTP) are two separate diseases that are characterized by low platelets. Immune thrombocytopenia (ITP) is an acquired hematologic disorder caused by immune-mediated destruction of platelets that can be classified as idiopathic or secondary to another disease process. ITP is a clinical diagnosis that rests on the presence of isolated thrombocytopenia while excluding other causes of thrombocytopenia and identifying conditions that may be causing secondary ITP [1]. Thrombotic thrombocytopenic purpura (TTP) is a rare thrombotic microangiopathy caused by an acquired autoantibody that leads to decreased activity of the von Willebrand factor-cleaving protease ADAMTS13 that results in hemolytic anemia and severe thrombocytopenia [2]. While the treatment courses for both ITP and TTP may include corticosteroids and rituximab, a key treatment difference is urgent therapeutic plasma exchange (TPE) for all patients with TTP, whereas TPE is not effective for ITP. In this article, we report a case of sequential ITP and TTP in a patient with primary Sjogren's syndrome who was diagnosed in part due to the utility of the PLASMIC scoring system. 2. Case Presentation A 72-year-old Caucasian man was diagnosed in 2019 with ITP after he presented to his local emergency department (ED) with bruising on his left arm, transient hematuria, melena, persistent epistaxis, and a platelet count of 4,000/µL. A complete blood count showed that his hemoglobin level was 14.0 g/dL and also confirmed the severe thrombocytopenia. Hemolysis labs showed LDH 192 U/L, total bilirubin 0.4 mg/dL, haptoglobin 52 mg/dL, and reticulocyte count 1.6%. His presentation was felt to be drug-induced ITP from apixaban that was started several months earlier. He received 1 mg/kg prednisone and was transfused 1 unit of platelets before being admitted. Despite steroids and 2 additional platelet transfusions over the next two days, his platelets remained 4,000/µL, prompting transfer to our facility for IVIG consideration (Figure 1). On arrival, his platelet count was 0k/µL. His peripheral smear showed thrombocytopenia with giant platelets, some rouleaux formation, occasional large granular lymphocytes with otherwise normal red and white cell morphology, and no fragmented cells or schistocytes. His workup was significant for elevated rheumatoid factor to 71 IU/ml, ANA positive at >1:640 with a speckled pattern, ESR of 53 mm/hr, negative lupus anticoagulant, negative cardiolipin antibody, negative beta-2 glycoprotein, negative hepatitis C, negative hepatitis B, and negative apixaban-dependent platelet antibody. Bone marrow biopsy showed normocellular age with normal megakaryocytes. The patient was treated with dexamethasone 40 mg × 4 days and IVIG 1 g/kg × 2. His platelet count increased to 12,000/µL. Following this therapy, his platelet count dropped to 5,000/µL. He received romiplostim 120 (1 mcg/kg) mcg daily. After 6 days of romiplostim, his platelet count was 9,000/µL, so he received rituximab (375 mg/m2; 850 mg infusion) as well as mycophenolate (1,000 mg twice daily). At discharge, his platelet count was 22,000/µL, and he was continued on romiplostim 2 mcg/kg, mycophenolate, and rituximab. One week later, his platelet count was 72,000/µL, and he received one additional rituximab infusion before being followed with observation. Over the next 13 months, his platelet count was between 74,000/µL and 157,000/µL. In September 2020, nearly 19 months after initially presenting with ITP, our patient presented to his local ED with hematuria and epistaxis again. His platelet count was 9,000/µL, hemoglobin was 12 g/dL, and total bilirubin was 5.3 mg/dL. A relapse of his ITP was diagnosed. He was initially treated with dexamethasone 8 mg and transferred to our facility for possible rituximab infusion. On arrival, his platelet count was 2,000/µL, hemoglobin was 11.1 g/dL (down from a baseline of 15), and schistocytes were seen on peripheral smear. He had a new acute kidney injury with a creatinine of 2.41 mg/dL (up from a baseline of 1 mg/dL) and an LDH of 2,399 U/L. A haptoglobin level was undetectable, and reticulocyte count was 2.1%. The new hemolytic anemia as well as the severe thrombocytopenia raised the possibility of TTP. The patient had several characteristics that were statistically uncommon for TTP patients. Namely, he was male, in his seventies, and Caucasian. Furthermore, he had an existing diagnosis of ITP as well as symptoms that were nearly identical to his previous presentation with ITP. However, ITP alone would not explain his hemolytic anemia. Thus, the PLASMIC score was used, and his score was high. This meant that the probability of TTP was high based on this scoring system. Given the high likelihood for TTP, he received urgent plasmapheresis (Figure 2). He also received prednisone (1 mg/kg; 110 mg). The diagnosis of TTP was confirmed during his first plasmapheresis by a severely low ADAMTS13 activity of 9% as well as the presence of an inhibitor. Given his positive rheumatoid factor, positive ANA, and multiple autoimmune conditions, rheumatology was also consulted. He had a positive SSA and positive Schirmer's test of the left eye. Thus, he met ACR/EULAR classification criteria for primary Sjogren's syndrome. Given his atypical demographics for Sjogren's and lack of classic sicca symptoms, a labial salivary gland biopsy was performed but was not consistent with Sjogren's syndrome. Rheumatology suggested he may have Sjogren's secondary to SLE given  the positive ANA, low C4, and association between SLE and TTP. However, his dsDNA, RNP, Smith, and APS were all negative. His hospital course was complicated by MSSA bacteremia.This was presumably due to his apheresis/dialysis central venous catheter, so it was removed. Both TTE and TEE showed no evidence of vegetations. He was started on cefepime before narrowing to cefazolin with an outpatient plan of daptomycin for 4 weeks of total antibiotic therapy. He received daily TPE x14. He also received rituximab (375 mg/m2; 800 mg infusion) once a week x4. Dose 1 of this series was after TPE #7, dose 2 was after TPE #14, and doses 3 and 4 were over the following 2 weeks. Eventually, his platelets plateaued to around 120,000/uL. This was close to his baseline of around 130,000–150,000/uL during the previous year after he had been treated for ITP. He was discharged and did well at home. A month after discharge, his platelets increased to 179,000/uL without any significant drops. 3. Discussion ITP is a diagnosis of exclusion made in patients with isolated thrombocytopenia due to immune-mediated destruction of platelets. The diagnosis of ITP includes careful evaluation of the patient's history and physical examination along with laboratory testing including review of the peripheral smear to rule out other potential causes of thrombocytopenia. The pathogenesis of ITP is incompletely understood. Antibody-mediated destruction is generally considered the most common cause, as 60–70% of patients have platelet-specific immunoglobulin G autoantibodies, most commonly to platelet glycoprotein IIb/IIIa complex [1]. Other reported mechanisms include T-cell-mediated destruction of platelets [3] and impaired megakaryopoiesis [4]. Despite these frequent antibody associations, current antibody tests have been consistently demonstrated to be poorly sensitive and poorly specific for ITP in nearly all studies that have been published. Thus, ITP is a clinical diagnosis since no widely validated antibody test exists as of this writing. In 80% of ITP cases, the cause is idiopathic. The other 20% of cases are secondary to conditions such as autoimmune syndromes, immunodeficiency syndromes, infections, lymphoid malignancies, and medications [5]. Treatment for newly diagnosed primary ITP is typically reserved for patients with clinically significant bleeding and severely low platelets (<30,000/microL), as the overall risk for bleeding in ITP is low. The primary first-line treatment of ITP is corticosteroids. Increased or normalized platelet counts are usually seen within two weeks, and this response to therapy is sometimes used in practice as a confirmatory test of sorts. TTP is also a disorder that is characterized by thrombocytopenia, but TTP includes microangiopathic hemolytic anemia (MAHA) in addition to severe thrombocytopenia. While the pentad of fever, neurological and renal abnormalities, MAHA, and thrombocytopenia is classically associated with TTP, the full pentad is only seen in 5% of cases and should not be relied upon as a sensitive set of criteria for diagnosis [2, 6]. In addition, while schistocytes or red cell fragments are often seen in the disease, their presence is neither sensitive nor specific for TTP [7, 8]. In our experience, it is often a pitfall to rely on the presence or absence of overt end-organ damage or schistocytes. TTP is characterized by widespread small-vessel platelets and von Willebrand factor-rich thrombi that are the result of a reduction in the activity of the enzyme ADAMTS13, a disintegrin and metalloproteinase involved in cleavage of large vWF multimers [9]. Severe ADAMTS13 deficiency (<10%) is very often seen in TTP [2] and is caused by an autoantibody (often reported by the lab as an inhibitor) that is commonly detected using a mixing study [10]. The primary treatment of TTP is urgent TPE followed by TPE daily until the platelet count normalizes. Before the discovery of TPE as the treatment, TTP was fatal in 90% of patients [9]. Today, with treatment, about 80–90% of patients survive [2]. Recently, the diagnostic accuracy of the PLASMIC score was validated by a systematic review involving nearly 1,000 patients [11]. The PLASMIC score ranges from 0–7 and gives one point each for the following: the absence of active cancer, the absence of solid organ and stem cell transplant, and five categories of laboratory results. A PLASMIC score of 5 or greater has a sensitivity of 99 percent and specificity of 57 percent for ADAMTS13 deficiency. Given the potentially life-threatening results of withholding therapy as well as the relative safety of TPE, treatment with urgent TPE is recommended for patients with a score of 5 or higher while awaiting results of ADAMTS13 testing. We hasten to add that the turnaround time of ADAMTS13 testing has dramatically improved recently in many settings due to the wider availability of in-hospital test kits. Thus, guidelines that recommend TPE for patients with a low pretest probability of TTP usually assume that a turnaround time of at least a few days is required for a reference lab to perform the test. This is essentially rendered moot when an in-hospital test can produce a result within 45 minutes to a few hours. In other words, a diagnosis of “can't rule out TTP” in patient with a low PLASMIC score can very often be ruled out by a rapid in-hospital test (and infusing plasma while waiting, if clinically indicated). Awareness of this innovation should be raised, especially when writing guidelines. The exact diagnosis (or diagnoses) of our patient is open to debate and speculation. Sequential or concomitant ITP and TTP have been previously described in the literature [12, 13] and have been associated with pregnancy [14, 15], HIV [16–18], and essential thrombocythemia [19]. The 2001 review by Baron et al. identified 11 cases of sequential or concurrent ITP and TTP without HIV and demonstrated a female predominance (9/11, 81%) and an association with pregnancy and autoimmune disorders including systemic lupus erythematosus (SLE), rheumatoid arthritis (RA), hypothyroidism, and Sjogren's syndrome secondary to RA [13]. Our patient met ACR diagnostic criteria for primary Sjogren's syndrome (pSS) during his most recent hospitalization. While secondary ITP in patients with pSS may be relatively common [20], secondary TTP in a male patient with pSS is exceedingly rare. The recent review by Carvalho et al. identified 18 patients with SS and TTP [21]. pSS was observed in 16/18 (88.9%) with a similar female predominance (15/18, 83%) to patients with both ITP and TTP. In most cases they reviewed, SS preceded TTP. This may fit the timeline of our patient, as he was presumed to have an autoimmune condition at his first presentation that was likely to be pSS and later diagnosed during the subsequent hospitalization. Due to the increased co-occurrence of both ITP and TTP in patients with autoimmune conditions, HIV, or pregnancy, it has been hypothesized that there may be some shared pathophysiologic factors such as redundancy of the immune system or a mixed immune thrombocytopenia syndrome that connects the two conditions [13, 22]. A recent article described 19 novel autoantibodies in Sjogren's syndrome [23] that may further explain the connection between ITP, TTP, and secondary autoimmune conditions. To our knowledge, this is the first case of sequential ITP and TTP associated with pSS, and it notably occurred in a man. One key lesson is the utility of the PLASMIC score, especially for patients that “do not read the textbook.” It may be very tempting to dismiss a diagnosis of TTP in a patient with ITP or any other condition that can explain isolated thrombocytopenia. More generally, it can also be tempting to dismiss a diagnosis of TTP in a patient who is not in the “usual” demographic for TTP, as TTP is over-represented in females, black females, and most often afflicts such patients who are in their thirties and forties [24]. While the specificity of the PLASMIC score is moderate, the sensitivity is high. High sensitivity tests and scoring systems can be particularly useful when your goal is not to miss a high stakes diagnosis, especially when the therapy is relatively safe. Given TTP's stakes, urgency, and bespoke treatment, this case underscores the importance of treating empirically with TPE as soon as possible when the PLASMIC score is high—even when one or more patient characteristics or diagnoses tempt you to reconsider. Data Availability No data sets were used other than the patient's medical record. Conflicts of Interest The authors declare that they have no conflicts of interest. Figure 1 ITP diagnosis and treatment timeline. Major dates in the patient's initial ITP presentation and treatment course occurring in 2019 are listed along the x-axis with accompanying hash marks and events above. Daily platelet counts (/µL) and LDH (U/L) where available are listed below respective dates. The above timeline includes major therapies with which the patient was treated, spanning across dates received including prednisone 1 mg/kg (blue bar), IVIG (red bar), dexamethasone 40 mg (yellow bar), and romiplostim 120 mcg (green bar). Overlapping bars imply multiple treatments given on same days. Figure 2 TTP diagnosis and treatment timeline. Major dates in patient's TTP presentation and treatment course occurring in 2020 are listed along the x-axis with accompanying hash marks and events above. The red scale break on the x-axis signifies large relative gap in time between dates. Daily platelet counts (/µL) and LDH (U/L) where available are listed below respective dates. The above timeline includes major therapies with which the patient was treated, spanning across dates received including dexamethasone 40 mg (red bar), prednisone 1 mg/kg (blue bar), and plasma exchange (TPE, green bar). Overlapping bars imply multiple treatments given on same days.

APIXABAN

DrugsGivenReaction

CC BY

34887925

2021

Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Drug resistance'.

Challenges in diagnosis and management of neonatal hyperparathyroidism in a resource-limited country: a case series from a Sudanese family. Neonatal hyperparathyroidism is a rare disease caused by a homozygous inactivating mutation in the calcium sensing receptor gene. It presents early in life with life threatening manifestations of hypercalcemia, if left untreated the condition may be lethal. This is the first case series reported from Sudan. Three Sudanese siblings presented with severe symptoms of hypercalcemia in the form of polyuria, failure to thrive and multiple bone fractures. Serum calcium and parathyroid hormone levels were very high with low phosphate and normal alkaline phosphatase levels. Ultrasonography and sestamibi scan were normal and did not assist in diagnosing their condition. Medical management was a great challenge due to unavailability of medications such as parentral bisphosphonates and calcimimetics. Parathyroidectomy was inevitable. Tissue biopsies revealed parathyroid hyperplasia and no adenoma. Gene sequencing revealed a homozygous missense mutation: c 2038 C T p (Arg680Cys) in two siblings, both parents were heterozygous for the same missense mutation. Our report reflects the challenges in diagnosis and management of neonatal hyperparathyroidism in resource limited countries. We also highlight the importance of genetic testing in the diagnosis and management of such cases in countries with high rates of consanguineous marriage. pmcIntroduction The calcium sensing receptor (CaSR) plays a crucial role in calcium homeostasis. The human CaSR gene (Online Mendelian Inheritance in Man (OMIM #601199) is located at chromosome 3q13.3-q21.1) [1]. More than 100 mutations in the CaSR gene are known to date. Loss of functional mutations results in hypercalcemia with hypocalciuria. There are two forms, a heterozygous mutation resulting in a benign asymptomatic form not requiring treatment, called familial benign hypocalciuric hypercalcemia (FBHH) (OMIM #145980) typically characterized by moderate elevations of serum calcium concentration, inappropriately low urinary calcium excretion, and high normal or mildly elevated parathyroid hormone (PTH) levels [2-5]. The other being a homozygous mutation resulting in a more rare but severe form called neonatal hyperparathyroidism (NHPT) (OMIM # 239200) requiring urgent treatment. The latter NHPT presents in the first 6 months of life with severe manifestations of hypercalcemia in the form of respiratory, skeletal, and psychomotor symptoms. PTH levels are high, and if left untreated the condition may be lethal [2-5]. We here report, and for the first time from Sudan, three siblings with this rare condition and discuss the problems that we faced in diagnosing and managing this disorder in a resource-limited country. Methods Case 1 {(1V/12), Figure 1}: a 7-month-old female presented with multiple bone fractures and was accidentally found to have hypercalcemia on routine investigations. She had a history of NICU admission for what was thought to be neonatal sepsis because of poor feeding and floppiness, chronic constipation, and failure to thrive but no history of urinary stones. Her birth was 2.8 kg. There were no dysmorphic features and her weight at presentation was 5 kg (-2.6 standard deviation (SD) below the mean for her age and sex). Parents are healthy first-degree cousins. The daughter of her maternal cousin was diagnosed with hyperparathyroidism in infancy which required parathyroidectomy. The biochemical, radiological, and histopathological findings are shown in Table 1. Her mother´s calcium level was 11.3 mg/dl (2.83 mmol/l), father´s level was 9.1 mg/dl (2.27 mmol/l). Vitamin D3 was not measured for the parents. She only showed a partial response to medical treatment of hypercalcemia, in the form of intravenous saline with furosemide 2 mg/kg/dose TDS, bisphosphonates in the form of oral alendronate 10 mg/day and calcitonin 4 IU/kg given subcutaneously every 6 hours. The lowest calcium level achieved with medical treatment was 17 mg/dl. Surgical intervention was inevitable and included total parathyroidectomy through bilateral neck incision and removal of the four parathyroid glands with re-implantation of one gland in the sternocleidomastoid muscle. Following surgery, she remained symptomatic with persistently high levels of calcium and PTH. The condition was not responsive to same medical treatment mentioned and required a second surgery to remove any remaining parathyroid tissue 8 weeks following the first surgery. Her calcium levels normalized immediately following the second surgery and after which she remained asymptomatic with normal calcium levels without any treatment. She managed to catch up normal growth for her age by the age of 36 months. Figure 1 family pedigree Table 1 biochemical, radiological and histopathological findings in case one and two Investigation Case 1 Case 2 Normal values Serum calcium mg/dl (mmol/L) 22.5 (5.5) 19.5 (4.86) 8.5 - 10.2 (2.1 - 2.6) Serum phosphorus mg/dl (mmol/L) 1.4 (0.452) 1.1 (0.35) 2.5 - 4.5 (1.12 - 1.4) Alkaline phosphatase IU/L 178 235 150 - 420 PTH pg/ml 229 838.5 10-55 UCCR mol/mol 0.026 0.024 0.09 - 2.2 Vitamin D3 ng/ml 38 Not done ≥30 X-ray of long bones Osteopenia, multiple bone fractures Osteopenia - Ultrasonography of parathyroid glands Enlarged parathyroid glands Enlarged parathyroid glands suggestive parathyroid adenoma - SestaMibi scan Not done Normal uptake scan - Histopathology of parathyroid gland Chief cell hyperplasia, water clear cell hyperplasia, no adipose tissue Chief cell hyperplasia, water clear cell hyperplasia, no adipose tissue - PTH: parathyroid hormone; UCCR: urinary calcium creatinine ratio Case 2 {(1V/13),(Figure 1)} a younger male sibling, presented at the age of 8 months with a history of chronic constipation, abdominal pain, polyuria, failure to thrive, and floppiness since birth. a younger male sibling, presented at the age of 8 months with a history of chronic constipation, abdominal pain, polyuria, failure to thrive, and floppiness since birth. He had no history of bone fractures nor urinary stones. There were no dysmorphic features and his weight was 6 kg (-3.5 SD). His biochemical and radiological findings are shown in Table 1. Hypercalcemia was resistant to available medical treatment in the form of intravenous saline and furosemide, oral bisphosphonates, and calcitonin. Surgical intervention was required and removal of all four parathyroid glands with reimplantation of the fourth one in the sternocleidomastoid muscle was done. Immediately following surgery, he developed hypocalcemia and required intravenous calcium then was shifted to oral calcium supplementations, 40 mg/kg/day for 6 weeks after which his calcium levels normalized. There after 6 weeks from his surgery he remained asymptomatic with normal calcium levels and required no further treatment. He managed to catch up to normal growth for his age at 30 months. Case 3 {(1V/14), (Figure 1)}: the younger sibling presented post-delivery with severe symptoms of hypercalcemia, bone fractures and respiratory distress that required ventilatory support. PTH and calcium levels were high. Management required parathyroidectomy. Further medical data was difficult to access as this patient was born abroad in Saudi Arabia (Figure 1). Results Biochemical and radiological findings are shown in (Table 1). Genetic analysis (Figure 2): genetic testing polymerase chain reaction (PCR) products covering all coding regions and splice sites of the CaSR gene were amplified using genomic DNA. After purification, Sanger sequencing was performed on an applied biosystem 3730 automated sequencer. Primer sequences are available upon request. Figure 2 chromatogram of CaSR gene mutation in family members In two affected siblings case 1 (1V/12), and case 2 (1V/13) shows a homozygous pathogenic mutation in the CaSR gene (NM_00388.3) which was detected by DNA sequence analysis. It is a missense mutation: c. 2038C ≥ T (p. (Arg680Cys)). Both parents were found to be heterozygous for the same missense mutation. No available genetic data for case 3 (1V/14), and no mutation was identified in the non-affected sister (Figure 1). The mutation has been described [5] and functional testing has proven pathogenicity [6-8]. Discussion NHPT is a rare condition and only a few cases including ours were reported in the literature to date [9-12]. To the best of our knowledge, this is the first reported cases from Sudan and sub-Saharan Africa. Sudan is a country with high rates of consanguineous marriage. In this report three infants were affected from one family. Apart from bone fractures and respiratory distress, symptoms of hypercalcemia in infants can be nonspecific and therefore diagnosis can be delayed as happened to our cases. Even if hypercalcemia is diagnosed, in developing countries, there may be difficulties in identifying the cause due to lack of facilities for PTH assay in primary and secondary care setups. Limited access to medical data on case {(1V/14), (Figure 1)} made it difficult to compare his management and diagnosis to his two elder siblings. Nevertheless, availability of facilities and medications assisted in early diagnosis and management of this case in comparison to his elder two siblings who were born in Sudan. Inactivating mutations in CaSR may affect a single allele (heterozygosis), resulting in the phenotype characteristic of FBHH, or both alleles (homozygosis or compound heterozygosis if no consanguinity exists), leading to NHPT, so that the degree of gene defect is responsible for the great difference in the phenotypic presentation [2,3]. In this report, two siblings had NHPT with homozygous mutations and variability in the degree of severity of symptoms In case 1 {(1V/12), (Figure 1)} the symptoms were severe in the face of high, but not too high PTH and extremely high calcium levels, while her younger sibling, case 2 {(1V/13), (Figure 1)} presented with milder symptoms, high calcium levels and extremely high levels of PTH. and extremely high calcium levels, while her younger sibling, case 2 {(1V/13), (Figure 1) presented with milder symptoms, high calcium levels and extremely high levels of PTH. These findings suggest no correlation between the levels of PTH, calcium and the severity of clinical symptoms. NHPT shows a puzzling range of serum calcium and PTH levels. In a recently published study, the levels of serum calcium and PTH were compared between patients with NHPT with homozygous mutations versus those with heterozygous mutations. They concluded that homozygotes for pathogenic CaSR variants show higher calcium and PTH levels than heterozygotes and calcium levels above 4.5 mmol/L among NHPT are frequent and unique only to most homozygotes [13]. Zajickova et al. reported that vitamin D levels may play an important role in the degree of severity of hypercalcemia in a patient with CaSR mutation, be it either the patient´s own vitamin D level or the lack of vitamin D in the mother during pregnancy [14]. Their patient with low 25 OH vitamin D3 levels benefited from vitamin D supplementation. In our report, 25 OH vitamin D3 level was normal in case 1 {(1V/12), (Figure 1)} and was not measured in case 2 {(1V/13), (Figure 1)} or their parents because of the cost. Typically, cases of CaSR inactivation mutation present with hypocalciuric hypercalcemia. A clinical finding which characterizes it thus helps distinguish them from primary hyperparathyroidism. In contrary to adults, cut of levels for diagnosing hypocalciuria in children vary with age making it more difficult to define the cut off levels in children. In comparison to a study done by Matos et al. [15] two of our patients showed an inappropriately low urinary calcium creatinine ratio (UCCR). In addition to the clinical picture, genetic studies are very useful. However, this is not available in developing countries and even if done abroad it is received late as was in our case, and therefore management including parathyroidectomy can be done based on clinical grounds. Medical emergency management should be initiated by the restoration of extravascular volume followed by furosemide. Calcitonin given subcutaneously can also give some short-term improvement in serum calcium. Calcitonin was not initially available in our setting and was sent to the family from relatives living abroad. Bisphosphonates are second line treatment they act by halting bone resorption. In two separate studies Al-Shafaney and Waller, reported the benefit of bisphosphonates in managing mild cases and helping to bridge the severe cases for surgery [9,16]. Parenteral bisphosphonates are better than oral forms however, unlike the situation now, they were not initially available to us. Nevertheless, the oral bisphosphonates that we used in our patients gave temporary partial suppression of calcium level even better than calcitonin. Cinacalcet is known to increase the sensitivity of CASR [14] and is reported to be effective in heterozygous mutation causing NHPT on the contrary to homozygous cases where four reported heterozygous cases did not require parathyroidectomy [17]. Unfortunately, parenteral bisphosphonates and cinacalcet were not available to us at that time which made managing such cases extremely challenging. Therefore, parathyroidectomy was considered before receiving the results of genetic studies. Pre-operative imaging using ultrasonography or magnetic resonance or sestamibi scan and or intraoperative measurements of PTH levels may help guide the extent of parathyroid resection, particularly in the case of multigland hyperparathyroidism but these imaging studies do not usually help in children with neonatal and familial hyperparathyroidism [11]. Al-Shanafey et al. published series of five cases with NHPT; all patients had the ultrasound, computed tomography (CT) scan and sestamibi nuclear scan but these studies could not identify the parathyroid glands in any of them [9]. In many developing countries radio nuclear imaging may not be available and neonatal ultrasonography needs professional experience. In our report preoperative imaging using sestamibi scan was highly expensive and not helpful and the ultrasonography gave a false impression of parathyroid adenoma. Surgical intervention is inevitable in patients with severe clinical manifestations of hypercalcemia not responding to medical therapy. Surgery in infants is quite challenging owing to the difficulty in parathyroid gland identification, requiring sufficient expertise. Nowadays, total parathyroidectomy with or without autotransplantation is increasingly being performed in most centers [9-12]. Parathyroidectomy could be subtotal or total with - implantation of one gland in the sternocleidomastoid muscle of the non-dominant arm [18,19]. In cases of NHPT, usually, parathyroidectomy of all 4 glands is required to achieve a cure and less radical removal may result in the persistence of hyperparathyroidism and hypercalcemia [9,11,19]. Some surgeons recommend parathyroid auto transplantation, yet frequently these grafts fail to work [9,19]. It has been reported that autotransplantation may lead to graft dependent hypercalcemia in 33% and a failure rate of 6% [9-11,20]. Al Shafaney et al. reported no recurrence following autotransplantation in his series of five patients [9]. Alagaratnam et al. reported in a series of cases with one case who remained moderately hypercalcemic with elevated PTH following subtotal parathyroidectomy and was managed conservatively [11]. Savas-Erdeve et al. have reported a single case of recurrence following autotransplantation which required removal [10]. The latter study resembling our first patient who required two surgeries to become symptom free and normalize her calcium levels. Failure of surgery to cure hyperparathyroidism may be due to either multi gland disease (e.g. hyperplasia) unidentified by preoperative imaging studies or failure to locate parathyroid glands in unusual areas (ectopic glands). In case {(1V/12), (Figure 1)} reimplantation was not successful. Postoperative symptomatic hypocalcemia is a common postoperative complication. Alagaratnam et al. reported in a series of cases, five cases requiring hypocalcemia treatment following total parathyroidectomy [11] as happened to case 2 {(1V/13), (Figure 1)}. Various operative adjuncts such as intra-operative PTH monitoring, radio-guided parathyroidectomy, frozen section, and methylene blue are commonly used to overcome this problem and improve the cure rate. Unfortunately, these techniques are unavailable in our settings. Hence, the best operative option for cases of severe NHPT in places similar to our setting is total parathyroidectomy without reimplantation. Conclusion Neonatal hyperparathyroidism (NHPT) is a rare and challenging disorder and dynamic management strategies are highly required. Hence, management of such cases is extremely challenging in resource limited countries. Genetic testing is of great value in populations with high rates of consanguineous marriage like Sudan. Limitations: i) limited resources in our setting contributed to the lack of ability to perform some investigations and give the standard methods of therapy; ii) limited access to medical data on case 3 made it difficult to compare his management and diagnosis to his two elder siblings. What is known about this topic Neonatal hyperparathyroidism is a rare disease caused by a homozygous inactivating mutation in the calcium sensing receptor gene; it presents early in life with life threatening manifestations of hypercalcemia, if left untreated the condition may be lethal. What this study adds To our knowledge this is the first case series report of neonatal sever hyperparathyroidism from Sudan, an African country with high rate of consanguineous marriage; Our findings suggest no correlation between the levels of parathyroid hormone and serum calcium levels and the severity of clinical symptoms in neonatal severe hyperparathyroidism; Although genetic testing is highly helpful in cases neonatal severe hyperparathyroidism, yet in developing countries where it is not available, management including parathyroidectomy can be done based on clinical grounds; findings in our study recommend that the best operative option for cases of neonatal severe hyperparathyroidism in places with limited resources similar to our setting is total parathyroidectomy without reimplantation. Acknowledgments The authors would like to send their gratitude to the patients and family members who participated in the study. Special thanks and acknowledgments to Professor Sten Drop for creating links for genetic testing. Special thanks and acknowledgment to Professor Anamika Boot for reviewing and providing articles and references. Competing interests The authors declare no competing interests. Authors' contributions Conceptualized the study design, data entry and analysis, writing and drafting initial script: Samar Hassan; genetic analysis and writing genetic part in the manuscript: Marlies Kempers and Dorien Lugtenberg; reviewing manuscript and assisting in data collection: Salwa Abdelbagi Musa, Asmahan Tajelsir Abdallah and Areej Ahmed Ibrahim; critical review and finalizing the manuscript: Mohamed Ahmed Abdullah. All the authors have read and agreed to the final manuscript. Cite this article: Samar Sabir Hassan et al. Challenges in diagnosis and management of neonatal hyperparathyroidism in a resource-limited country: a case series from a Sudanese family. Pan African Medical Journal. 2021;40(105). 10.11604/pamj.2021.40.105.29527

CALCITONIN, FUROSEMIDE

DrugsGivenReaction

CC BY

34887979

2021

What is the weight of the patient?

Challenges in diagnosis and management of neonatal hyperparathyroidism in a resource-limited country: a case series from a Sudanese family. Neonatal hyperparathyroidism is a rare disease caused by a homozygous inactivating mutation in the calcium sensing receptor gene. It presents early in life with life threatening manifestations of hypercalcemia, if left untreated the condition may be lethal. This is the first case series reported from Sudan. Three Sudanese siblings presented with severe symptoms of hypercalcemia in the form of polyuria, failure to thrive and multiple bone fractures. Serum calcium and parathyroid hormone levels were very high with low phosphate and normal alkaline phosphatase levels. Ultrasonography and sestamibi scan were normal and did not assist in diagnosing their condition. Medical management was a great challenge due to unavailability of medications such as parentral bisphosphonates and calcimimetics. Parathyroidectomy was inevitable. Tissue biopsies revealed parathyroid hyperplasia and no adenoma. Gene sequencing revealed a homozygous missense mutation: c 2038 C T p (Arg680Cys) in two siblings, both parents were heterozygous for the same missense mutation. Our report reflects the challenges in diagnosis and management of neonatal hyperparathyroidism in resource limited countries. We also highlight the importance of genetic testing in the diagnosis and management of such cases in countries with high rates of consanguineous marriage. pmcIntroduction The calcium sensing receptor (CaSR) plays a crucial role in calcium homeostasis. The human CaSR gene (Online Mendelian Inheritance in Man (OMIM #601199) is located at chromosome 3q13.3-q21.1) [1]. More than 100 mutations in the CaSR gene are known to date. Loss of functional mutations results in hypercalcemia with hypocalciuria. There are two forms, a heterozygous mutation resulting in a benign asymptomatic form not requiring treatment, called familial benign hypocalciuric hypercalcemia (FBHH) (OMIM #145980) typically characterized by moderate elevations of serum calcium concentration, inappropriately low urinary calcium excretion, and high normal or mildly elevated parathyroid hormone (PTH) levels [2-5]. The other being a homozygous mutation resulting in a more rare but severe form called neonatal hyperparathyroidism (NHPT) (OMIM # 239200) requiring urgent treatment. The latter NHPT presents in the first 6 months of life with severe manifestations of hypercalcemia in the form of respiratory, skeletal, and psychomotor symptoms. PTH levels are high, and if left untreated the condition may be lethal [2-5]. We here report, and for the first time from Sudan, three siblings with this rare condition and discuss the problems that we faced in diagnosing and managing this disorder in a resource-limited country. Methods Case 1 {(1V/12), Figure 1}: a 7-month-old female presented with multiple bone fractures and was accidentally found to have hypercalcemia on routine investigations. She had a history of NICU admission for what was thought to be neonatal sepsis because of poor feeding and floppiness, chronic constipation, and failure to thrive but no history of urinary stones. Her birth was 2.8 kg. There were no dysmorphic features and her weight at presentation was 5 kg (-2.6 standard deviation (SD) below the mean for her age and sex). Parents are healthy first-degree cousins. The daughter of her maternal cousin was diagnosed with hyperparathyroidism in infancy which required parathyroidectomy. The biochemical, radiological, and histopathological findings are shown in Table 1. Her mother´s calcium level was 11.3 mg/dl (2.83 mmol/l), father´s level was 9.1 mg/dl (2.27 mmol/l). Vitamin D3 was not measured for the parents. She only showed a partial response to medical treatment of hypercalcemia, in the form of intravenous saline with furosemide 2 mg/kg/dose TDS, bisphosphonates in the form of oral alendronate 10 mg/day and calcitonin 4 IU/kg given subcutaneously every 6 hours. The lowest calcium level achieved with medical treatment was 17 mg/dl. Surgical intervention was inevitable and included total parathyroidectomy through bilateral neck incision and removal of the four parathyroid glands with re-implantation of one gland in the sternocleidomastoid muscle. Following surgery, she remained symptomatic with persistently high levels of calcium and PTH. The condition was not responsive to same medical treatment mentioned and required a second surgery to remove any remaining parathyroid tissue 8 weeks following the first surgery. Her calcium levels normalized immediately following the second surgery and after which she remained asymptomatic with normal calcium levels without any treatment. She managed to catch up normal growth for her age by the age of 36 months. Figure 1 family pedigree Table 1 biochemical, radiological and histopathological findings in case one and two Investigation Case 1 Case 2 Normal values Serum calcium mg/dl (mmol/L) 22.5 (5.5) 19.5 (4.86) 8.5 - 10.2 (2.1 - 2.6) Serum phosphorus mg/dl (mmol/L) 1.4 (0.452) 1.1 (0.35) 2.5 - 4.5 (1.12 - 1.4) Alkaline phosphatase IU/L 178 235 150 - 420 PTH pg/ml 229 838.5 10-55 UCCR mol/mol 0.026 0.024 0.09 - 2.2 Vitamin D3 ng/ml 38 Not done ≥30 X-ray of long bones Osteopenia, multiple bone fractures Osteopenia - Ultrasonography of parathyroid glands Enlarged parathyroid glands Enlarged parathyroid glands suggestive parathyroid adenoma - SestaMibi scan Not done Normal uptake scan - Histopathology of parathyroid gland Chief cell hyperplasia, water clear cell hyperplasia, no adipose tissue Chief cell hyperplasia, water clear cell hyperplasia, no adipose tissue - PTH: parathyroid hormone; UCCR: urinary calcium creatinine ratio Case 2 {(1V/13),(Figure 1)} a younger male sibling, presented at the age of 8 months with a history of chronic constipation, abdominal pain, polyuria, failure to thrive, and floppiness since birth. a younger male sibling, presented at the age of 8 months with a history of chronic constipation, abdominal pain, polyuria, failure to thrive, and floppiness since birth. He had no history of bone fractures nor urinary stones. There were no dysmorphic features and his weight was 6 kg (-3.5 SD). His biochemical and radiological findings are shown in Table 1. Hypercalcemia was resistant to available medical treatment in the form of intravenous saline and furosemide, oral bisphosphonates, and calcitonin. Surgical intervention was required and removal of all four parathyroid glands with reimplantation of the fourth one in the sternocleidomastoid muscle was done. Immediately following surgery, he developed hypocalcemia and required intravenous calcium then was shifted to oral calcium supplementations, 40 mg/kg/day for 6 weeks after which his calcium levels normalized. There after 6 weeks from his surgery he remained asymptomatic with normal calcium levels and required no further treatment. He managed to catch up to normal growth for his age at 30 months. Case 3 {(1V/14), (Figure 1)}: the younger sibling presented post-delivery with severe symptoms of hypercalcemia, bone fractures and respiratory distress that required ventilatory support. PTH and calcium levels were high. Management required parathyroidectomy. Further medical data was difficult to access as this patient was born abroad in Saudi Arabia (Figure 1). Results Biochemical and radiological findings are shown in (Table 1). Genetic analysis (Figure 2): genetic testing polymerase chain reaction (PCR) products covering all coding regions and splice sites of the CaSR gene were amplified using genomic DNA. After purification, Sanger sequencing was performed on an applied biosystem 3730 automated sequencer. Primer sequences are available upon request. Figure 2 chromatogram of CaSR gene mutation in family members In two affected siblings case 1 (1V/12), and case 2 (1V/13) shows a homozygous pathogenic mutation in the CaSR gene (NM_00388.3) which was detected by DNA sequence analysis. It is a missense mutation: c. 2038C ≥ T (p. (Arg680Cys)). Both parents were found to be heterozygous for the same missense mutation. No available genetic data for case 3 (1V/14), and no mutation was identified in the non-affected sister (Figure 1). The mutation has been described [5] and functional testing has proven pathogenicity [6-8]. Discussion NHPT is a rare condition and only a few cases including ours were reported in the literature to date [9-12]. To the best of our knowledge, this is the first reported cases from Sudan and sub-Saharan Africa. Sudan is a country with high rates of consanguineous marriage. In this report three infants were affected from one family. Apart from bone fractures and respiratory distress, symptoms of hypercalcemia in infants can be nonspecific and therefore diagnosis can be delayed as happened to our cases. Even if hypercalcemia is diagnosed, in developing countries, there may be difficulties in identifying the cause due to lack of facilities for PTH assay in primary and secondary care setups. Limited access to medical data on case {(1V/14), (Figure 1)} made it difficult to compare his management and diagnosis to his two elder siblings. Nevertheless, availability of facilities and medications assisted in early diagnosis and management of this case in comparison to his elder two siblings who were born in Sudan. Inactivating mutations in CaSR may affect a single allele (heterozygosis), resulting in the phenotype characteristic of FBHH, or both alleles (homozygosis or compound heterozygosis if no consanguinity exists), leading to NHPT, so that the degree of gene defect is responsible for the great difference in the phenotypic presentation [2,3]. In this report, two siblings had NHPT with homozygous mutations and variability in the degree of severity of symptoms In case 1 {(1V/12), (Figure 1)} the symptoms were severe in the face of high, but not too high PTH and extremely high calcium levels, while her younger sibling, case 2 {(1V/13), (Figure 1)} presented with milder symptoms, high calcium levels and extremely high levels of PTH. and extremely high calcium levels, while her younger sibling, case 2 {(1V/13), (Figure 1) presented with milder symptoms, high calcium levels and extremely high levels of PTH. These findings suggest no correlation between the levels of PTH, calcium and the severity of clinical symptoms. NHPT shows a puzzling range of serum calcium and PTH levels. In a recently published study, the levels of serum calcium and PTH were compared between patients with NHPT with homozygous mutations versus those with heterozygous mutations. They concluded that homozygotes for pathogenic CaSR variants show higher calcium and PTH levels than heterozygotes and calcium levels above 4.5 mmol/L among NHPT are frequent and unique only to most homozygotes [13]. Zajickova et al. reported that vitamin D levels may play an important role in the degree of severity of hypercalcemia in a patient with CaSR mutation, be it either the patient´s own vitamin D level or the lack of vitamin D in the mother during pregnancy [14]. Their patient with low 25 OH vitamin D3 levels benefited from vitamin D supplementation. In our report, 25 OH vitamin D3 level was normal in case 1 {(1V/12), (Figure 1)} and was not measured in case 2 {(1V/13), (Figure 1)} or their parents because of the cost. Typically, cases of CaSR inactivation mutation present with hypocalciuric hypercalcemia. A clinical finding which characterizes it thus helps distinguish them from primary hyperparathyroidism. In contrary to adults, cut of levels for diagnosing hypocalciuria in children vary with age making it more difficult to define the cut off levels in children. In comparison to a study done by Matos et al. [15] two of our patients showed an inappropriately low urinary calcium creatinine ratio (UCCR). In addition to the clinical picture, genetic studies are very useful. However, this is not available in developing countries and even if done abroad it is received late as was in our case, and therefore management including parathyroidectomy can be done based on clinical grounds. Medical emergency management should be initiated by the restoration of extravascular volume followed by furosemide. Calcitonin given subcutaneously can also give some short-term improvement in serum calcium. Calcitonin was not initially available in our setting and was sent to the family from relatives living abroad. Bisphosphonates are second line treatment they act by halting bone resorption. In two separate studies Al-Shafaney and Waller, reported the benefit of bisphosphonates in managing mild cases and helping to bridge the severe cases for surgery [9,16]. Parenteral bisphosphonates are better than oral forms however, unlike the situation now, they were not initially available to us. Nevertheless, the oral bisphosphonates that we used in our patients gave temporary partial suppression of calcium level even better than calcitonin. Cinacalcet is known to increase the sensitivity of CASR [14] and is reported to be effective in heterozygous mutation causing NHPT on the contrary to homozygous cases where four reported heterozygous cases did not require parathyroidectomy [17]. Unfortunately, parenteral bisphosphonates and cinacalcet were not available to us at that time which made managing such cases extremely challenging. Therefore, parathyroidectomy was considered before receiving the results of genetic studies. Pre-operative imaging using ultrasonography or magnetic resonance or sestamibi scan and or intraoperative measurements of PTH levels may help guide the extent of parathyroid resection, particularly in the case of multigland hyperparathyroidism but these imaging studies do not usually help in children with neonatal and familial hyperparathyroidism [11]. Al-Shanafey et al. published series of five cases with NHPT; all patients had the ultrasound, computed tomography (CT) scan and sestamibi nuclear scan but these studies could not identify the parathyroid glands in any of them [9]. In many developing countries radio nuclear imaging may not be available and neonatal ultrasonography needs professional experience. In our report preoperative imaging using sestamibi scan was highly expensive and not helpful and the ultrasonography gave a false impression of parathyroid adenoma. Surgical intervention is inevitable in patients with severe clinical manifestations of hypercalcemia not responding to medical therapy. Surgery in infants is quite challenging owing to the difficulty in parathyroid gland identification, requiring sufficient expertise. Nowadays, total parathyroidectomy with or without autotransplantation is increasingly being performed in most centers [9-12]. Parathyroidectomy could be subtotal or total with - implantation of one gland in the sternocleidomastoid muscle of the non-dominant arm [18,19]. In cases of NHPT, usually, parathyroidectomy of all 4 glands is required to achieve a cure and less radical removal may result in the persistence of hyperparathyroidism and hypercalcemia [9,11,19]. Some surgeons recommend parathyroid auto transplantation, yet frequently these grafts fail to work [9,19]. It has been reported that autotransplantation may lead to graft dependent hypercalcemia in 33% and a failure rate of 6% [9-11,20]. Al Shafaney et al. reported no recurrence following autotransplantation in his series of five patients [9]. Alagaratnam et al. reported in a series of cases with one case who remained moderately hypercalcemic with elevated PTH following subtotal parathyroidectomy and was managed conservatively [11]. Savas-Erdeve et al. have reported a single case of recurrence following autotransplantation which required removal [10]. The latter study resembling our first patient who required two surgeries to become symptom free and normalize her calcium levels. Failure of surgery to cure hyperparathyroidism may be due to either multi gland disease (e.g. hyperplasia) unidentified by preoperative imaging studies or failure to locate parathyroid glands in unusual areas (ectopic glands). In case {(1V/12), (Figure 1)} reimplantation was not successful. Postoperative symptomatic hypocalcemia is a common postoperative complication. Alagaratnam et al. reported in a series of cases, five cases requiring hypocalcemia treatment following total parathyroidectomy [11] as happened to case 2 {(1V/13), (Figure 1)}. Various operative adjuncts such as intra-operative PTH monitoring, radio-guided parathyroidectomy, frozen section, and methylene blue are commonly used to overcome this problem and improve the cure rate. Unfortunately, these techniques are unavailable in our settings. Hence, the best operative option for cases of severe NHPT in places similar to our setting is total parathyroidectomy without reimplantation. Conclusion Neonatal hyperparathyroidism (NHPT) is a rare and challenging disorder and dynamic management strategies are highly required. Hence, management of such cases is extremely challenging in resource limited countries. Genetic testing is of great value in populations with high rates of consanguineous marriage like Sudan. Limitations: i) limited resources in our setting contributed to the lack of ability to perform some investigations and give the standard methods of therapy; ii) limited access to medical data on case 3 made it difficult to compare his management and diagnosis to his two elder siblings. What is known about this topic Neonatal hyperparathyroidism is a rare disease caused by a homozygous inactivating mutation in the calcium sensing receptor gene; it presents early in life with life threatening manifestations of hypercalcemia, if left untreated the condition may be lethal. What this study adds To our knowledge this is the first case series report of neonatal sever hyperparathyroidism from Sudan, an African country with high rate of consanguineous marriage; Our findings suggest no correlation between the levels of parathyroid hormone and serum calcium levels and the severity of clinical symptoms in neonatal severe hyperparathyroidism; Although genetic testing is highly helpful in cases neonatal severe hyperparathyroidism, yet in developing countries where it is not available, management including parathyroidectomy can be done based on clinical grounds; findings in our study recommend that the best operative option for cases of neonatal severe hyperparathyroidism in places with limited resources similar to our setting is total parathyroidectomy without reimplantation. Acknowledgments The authors would like to send their gratitude to the patients and family members who participated in the study. Special thanks and acknowledgments to Professor Sten Drop for creating links for genetic testing. Special thanks and acknowledgment to Professor Anamika Boot for reviewing and providing articles and references. Competing interests The authors declare no competing interests. Authors' contributions Conceptualized the study design, data entry and analysis, writing and drafting initial script: Samar Hassan; genetic analysis and writing genetic part in the manuscript: Marlies Kempers and Dorien Lugtenberg; reviewing manuscript and assisting in data collection: Salwa Abdelbagi Musa, Asmahan Tajelsir Abdallah and Areej Ahmed Ibrahim; critical review and finalizing the manuscript: Mohamed Ahmed Abdullah. All the authors have read and agreed to the final manuscript. Cite this article: Samar Sabir Hassan et al. Challenges in diagnosis and management of neonatal hyperparathyroidism in a resource-limited country: a case series from a Sudanese family. Pan African Medical Journal. 2021;40(105). 10.11604/pamj.2021.40.105.29527

6 kg.

Weight

CC BY

34887979

2021

Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Cervix disorder'.

A Rare Type of Uterine Rupture Following Over-the-Counter Use of Misoprostol in Second Trimester Abortion. The use of misoprostol in the second trimester by a woman with a uterine scar may lead to severe contractions and uterine rupture. We report a 24-year-old pregnant female patient who presented at the Emergency Department at a tertiary care hospital in Puducherry, India, in 2020 with haemorrhagic shock. She was at 16 weeks of gestation and had taken over the counter misoprostol for inducing an abortion. A quick initial resuscitation and urgent laparotomy were performed. An irreparable circumferentially avulsed uterus suspended only by round ligaments was noted. Haemostasis required internal artery ligation and immediate total hysterectomy. The patient was doing well upon follow-up six months after the surgery. Proper and supervised use of misoprostol in the appropriate dosage can avoid life-threatening consequences of uterine rupture. pmcTermination of pregnancy in the first and second trimesters in patients with prior uterine scarring is challenging.1 Misoprostol is a commonly used drug for induction of abortion even in unsupervised clinical settings.2 However, its dosing regimen has remained controversial.3,4 Over-the-counter use of misoprostol has resulted in documented or undocumented incidents of varying grades of uterine rupture. Uterine rupture is a surgical emergency that requires immediate care to avoid life-threatening medical, surgical and psychological consequences.4 In the current case, a patient with a previous caesarean scar experienced a unique type of uterine rupture which developed after over-the-counter use of misoprostol to induce second trimester abortion. Case Report A 24-year-old G3P1L1A1 (pregnant woman who had had a full-term delivery and one abortion and also had one living child) presented at the Emergency Department of a tertiary care hospital in Puducherry, India, in 2020 with haemorrhagic shock. She had taken an unknown dosage of misoprostol (unsupervised) to induce abortion at the 16th week of gestation. She had undergone a caesarean delivery two years before the current pregnancy. She self-administered an unknown dosage of misoprostol through the vaginal route at four-hour intervals. She noticed expulsion of the products of conception after two hours of the last dose of misoprostol. A local physician performed dilatation and curettage to treat heavy vaginal bleeding one day prior to presentation at the hospital. Her vaginal bleeding did not subside and she could have developed iatrogenic uterine perforation after dilatation and curettage. At admission, she was drowsy, extremely pale and afebrile. Her vitals included a pulse rate of 130/min, systemic blood pressure of 90/60 mmHg and respiratory rate of 28 breaths per minute. The remaining cardiovascular, respiratory and central nervous system findings were unremarkable. Abdominal examination revealed a diffuse tenderness and a well retracted uterus deviated towards the right-side of the abdomen. During speculum examination, a continuous fresh blood trickling was noted along with cervical tears at the 2 o’clock and 6 o’clock positions extending until the posterior fornix. Furthermore, the cervix appeared uneffaced and os patulous. Both fornices were boggy and tender; a transverse rent was noted in the posterior fornix through which the posterior wall of the uterus was felt. A bedside pelvic ultrasound scan revealed a post-abortal uterus with an empty cavity and a moderate amount of free fluid in the abdomen. Resuscitation was performed under guided haemodynamic monitoring and an immediate exploratory laparotomy was planned under general anaesthesia. Intraoperatively, 500mL of haemoperitoneum was noted with large clots in the pelvic cavity. There was a full-length scar rupture with the rent extending posteriorly. The posterior aspect of the uterus was avulsed completely above the level of the internal os and it was suspended only by the round ligament bilaterally; it appeared like a bucket handle tear [Figure 1]. The patient remained haemodynamically unstable and circumferential rent repair did not seem possible. There was a broad ligament haematoma of 5 × 5 cm on the left-side. The left uterine artery appeared avulsed and could not be traced. Bladder integrity was normal and there was no evidence of haematuria. Total abdominal hysterectomy along with left internal artery ligation was performed to achieve haemostasis. The patient was successfully resuscitated and extubated inside the operation theatre. Her postoperative investigations were unremarkable. The patient’s clinical condition improved completely and she was discharged from the hospital on the seventh postoperative day. The patient was doing well when she was last contacted nearly six months after her surgery. The patient provided written informed consent for publication of the anonymised data during her first follow-up. Discussion An unknown dose of misoprostol followed by dilatation and curettage in a previously scarred uterus could have led to uterine rupture in the current patient. Uterine rupture is primarily a clinical diagnosis, and hence, prompt surgical management is critical. Risk factors for uterine rupture include a previous uterine scar, short interpregnancy interval, multiparity, uterotonic drugs and obstructed labour.5 Additionally, the incidence of uterine rupture has been noted to be higher in scarred compared to unscarred uteri (0.28% versus 0.04%).1 A patient’s survival after uterine rupture depends on the time interval between the rupture and the intervention and urgent referral to a tertiary care centre. Rent repair is the key to treatment and internal iliac artery ligation is a lifesaving procedure in cases of uncontrolled obstetric haemorrhage. However, hysterectomy should not be delayed if the bleeding is intractable or the uterine rupture is irreparable. Second trimester pregnancy termination using misoprostol in an appropriate dosage in supervised settings for women with cesarean scar is safe and it has been associated with uterine rupture in only 0.3–0.4% of cases.1,6,7 A retrospective report did not observe uterine rupture after misoprostol induction for the termination of second trimester pregnancy in a scarred uterus.8 In the current patient, a characteristic bucket handle type uterine rupture was noted as was also reported by Abubekar et al.9 An unyielding cervix with misoprostol-induced strong uterine contractions in a scarred uterus may predispose to this type of uterine rupture.10 In addition, unsupervised vigorous dilatation and curettage could have either led to iatrogenic uterine perforation or further aggravated the tear induced by misoprostol. Conclusion Unsupervised over-the-counter use of misoprostol in a scarred uterus to induce second trimester abortion can have catastrophic consequences. Dilatation and curettage should be avoided before ruling out uterine rupture. An early referral to a higher centre may prevent severe maternal morbidity and mortality. Figure 1 Photographs of the uterus of a 24-year-old female patient following over-the-counter use of misoprostol for abortion induction. A: Anterior view of the uterus circumferentially avulsed at the level of the internal os. B: Lateral view showing the avulsed uterus. C: Avulsed uterus suspended only through round ligaments. AUTHORS’ CONTRIBUTION NJ, HS, JS and AKJ conceptualised and designed the work as well as extracted the data. NJ, JS and AKJ interpreted the results. JS performed the statistical analysis. All authors contributed to the drafting and approved the final version of the manuscript.

MISOPROSTOL

DrugsGivenReaction

CC BY-ND

34888091

2021-11

Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Exposure during pregnancy'.

A Rare Type of Uterine Rupture Following Over-the-Counter Use of Misoprostol in Second Trimester Abortion. The use of misoprostol in the second trimester by a woman with a uterine scar may lead to severe contractions and uterine rupture. We report a 24-year-old pregnant female patient who presented at the Emergency Department at a tertiary care hospital in Puducherry, India, in 2020 with haemorrhagic shock. She was at 16 weeks of gestation and had taken over the counter misoprostol for inducing an abortion. A quick initial resuscitation and urgent laparotomy were performed. An irreparable circumferentially avulsed uterus suspended only by round ligaments was noted. Haemostasis required internal artery ligation and immediate total hysterectomy. The patient was doing well upon follow-up six months after the surgery. Proper and supervised use of misoprostol in the appropriate dosage can avoid life-threatening consequences of uterine rupture. pmcTermination of pregnancy in the first and second trimesters in patients with prior uterine scarring is challenging.1 Misoprostol is a commonly used drug for induction of abortion even in unsupervised clinical settings.2 However, its dosing regimen has remained controversial.3,4 Over-the-counter use of misoprostol has resulted in documented or undocumented incidents of varying grades of uterine rupture. Uterine rupture is a surgical emergency that requires immediate care to avoid life-threatening medical, surgical and psychological consequences.4 In the current case, a patient with a previous caesarean scar experienced a unique type of uterine rupture which developed after over-the-counter use of misoprostol to induce second trimester abortion. Case Report A 24-year-old G3P1L1A1 (pregnant woman who had had a full-term delivery and one abortion and also had one living child) presented at the Emergency Department of a tertiary care hospital in Puducherry, India, in 2020 with haemorrhagic shock. She had taken an unknown dosage of misoprostol (unsupervised) to induce abortion at the 16th week of gestation. She had undergone a caesarean delivery two years before the current pregnancy. She self-administered an unknown dosage of misoprostol through the vaginal route at four-hour intervals. She noticed expulsion of the products of conception after two hours of the last dose of misoprostol. A local physician performed dilatation and curettage to treat heavy vaginal bleeding one day prior to presentation at the hospital. Her vaginal bleeding did not subside and she could have developed iatrogenic uterine perforation after dilatation and curettage. At admission, she was drowsy, extremely pale and afebrile. Her vitals included a pulse rate of 130/min, systemic blood pressure of 90/60 mmHg and respiratory rate of 28 breaths per minute. The remaining cardiovascular, respiratory and central nervous system findings were unremarkable. Abdominal examination revealed a diffuse tenderness and a well retracted uterus deviated towards the right-side of the abdomen. During speculum examination, a continuous fresh blood trickling was noted along with cervical tears at the 2 o’clock and 6 o’clock positions extending until the posterior fornix. Furthermore, the cervix appeared uneffaced and os patulous. Both fornices were boggy and tender; a transverse rent was noted in the posterior fornix through which the posterior wall of the uterus was felt. A bedside pelvic ultrasound scan revealed a post-abortal uterus with an empty cavity and a moderate amount of free fluid in the abdomen. Resuscitation was performed under guided haemodynamic monitoring and an immediate exploratory laparotomy was planned under general anaesthesia. Intraoperatively, 500mL of haemoperitoneum was noted with large clots in the pelvic cavity. There was a full-length scar rupture with the rent extending posteriorly. The posterior aspect of the uterus was avulsed completely above the level of the internal os and it was suspended only by the round ligament bilaterally; it appeared like a bucket handle tear [Figure 1]. The patient remained haemodynamically unstable and circumferential rent repair did not seem possible. There was a broad ligament haematoma of 5 × 5 cm on the left-side. The left uterine artery appeared avulsed and could not be traced. Bladder integrity was normal and there was no evidence of haematuria. Total abdominal hysterectomy along with left internal artery ligation was performed to achieve haemostasis. The patient was successfully resuscitated and extubated inside the operation theatre. Her postoperative investigations were unremarkable. The patient’s clinical condition improved completely and she was discharged from the hospital on the seventh postoperative day. The patient was doing well when she was last contacted nearly six months after her surgery. The patient provided written informed consent for publication of the anonymised data during her first follow-up. Discussion An unknown dose of misoprostol followed by dilatation and curettage in a previously scarred uterus could have led to uterine rupture in the current patient. Uterine rupture is primarily a clinical diagnosis, and hence, prompt surgical management is critical. Risk factors for uterine rupture include a previous uterine scar, short interpregnancy interval, multiparity, uterotonic drugs and obstructed labour.5 Additionally, the incidence of uterine rupture has been noted to be higher in scarred compared to unscarred uteri (0.28% versus 0.04%).1 A patient’s survival after uterine rupture depends on the time interval between the rupture and the intervention and urgent referral to a tertiary care centre. Rent repair is the key to treatment and internal iliac artery ligation is a lifesaving procedure in cases of uncontrolled obstetric haemorrhage. However, hysterectomy should not be delayed if the bleeding is intractable or the uterine rupture is irreparable. Second trimester pregnancy termination using misoprostol in an appropriate dosage in supervised settings for women with cesarean scar is safe and it has been associated with uterine rupture in only 0.3–0.4% of cases.1,6,7 A retrospective report did not observe uterine rupture after misoprostol induction for the termination of second trimester pregnancy in a scarred uterus.8 In the current patient, a characteristic bucket handle type uterine rupture was noted as was also reported by Abubekar et al.9 An unyielding cervix with misoprostol-induced strong uterine contractions in a scarred uterus may predispose to this type of uterine rupture.10 In addition, unsupervised vigorous dilatation and curettage could have either led to iatrogenic uterine perforation or further aggravated the tear induced by misoprostol. Conclusion Unsupervised over-the-counter use of misoprostol in a scarred uterus to induce second trimester abortion can have catastrophic consequences. Dilatation and curettage should be avoided before ruling out uterine rupture. An early referral to a higher centre may prevent severe maternal morbidity and mortality. Figure 1 Photographs of the uterus of a 24-year-old female patient following over-the-counter use of misoprostol for abortion induction. A: Anterior view of the uterus circumferentially avulsed at the level of the internal os. B: Lateral view showing the avulsed uterus. C: Avulsed uterus suspended only through round ligaments. AUTHORS’ CONTRIBUTION NJ, HS, JS and AKJ conceptualised and designed the work as well as extracted the data. NJ, JS and AKJ interpreted the results. JS performed the statistical analysis. All authors contributed to the drafting and approved the final version of the manuscript.

MISOPROSTOL

DrugsGivenReaction

CC BY-ND

34888091

2021-11

Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Haemoperitoneum'.

A Rare Type of Uterine Rupture Following Over-the-Counter Use of Misoprostol in Second Trimester Abortion. The use of misoprostol in the second trimester by a woman with a uterine scar may lead to severe contractions and uterine rupture. We report a 24-year-old pregnant female patient who presented at the Emergency Department at a tertiary care hospital in Puducherry, India, in 2020 with haemorrhagic shock. She was at 16 weeks of gestation and had taken over the counter misoprostol for inducing an abortion. A quick initial resuscitation and urgent laparotomy were performed. An irreparable circumferentially avulsed uterus suspended only by round ligaments was noted. Haemostasis required internal artery ligation and immediate total hysterectomy. The patient was doing well upon follow-up six months after the surgery. Proper and supervised use of misoprostol in the appropriate dosage can avoid life-threatening consequences of uterine rupture. pmcTermination of pregnancy in the first and second trimesters in patients with prior uterine scarring is challenging.1 Misoprostol is a commonly used drug for induction of abortion even in unsupervised clinical settings.2 However, its dosing regimen has remained controversial.3,4 Over-the-counter use of misoprostol has resulted in documented or undocumented incidents of varying grades of uterine rupture. Uterine rupture is a surgical emergency that requires immediate care to avoid life-threatening medical, surgical and psychological consequences.4 In the current case, a patient with a previous caesarean scar experienced a unique type of uterine rupture which developed after over-the-counter use of misoprostol to induce second trimester abortion. Case Report A 24-year-old G3P1L1A1 (pregnant woman who had had a full-term delivery and one abortion and also had one living child) presented at the Emergency Department of a tertiary care hospital in Puducherry, India, in 2020 with haemorrhagic shock. She had taken an unknown dosage of misoprostol (unsupervised) to induce abortion at the 16th week of gestation. She had undergone a caesarean delivery two years before the current pregnancy. She self-administered an unknown dosage of misoprostol through the vaginal route at four-hour intervals. She noticed expulsion of the products of conception after two hours of the last dose of misoprostol. A local physician performed dilatation and curettage to treat heavy vaginal bleeding one day prior to presentation at the hospital. Her vaginal bleeding did not subside and she could have developed iatrogenic uterine perforation after dilatation and curettage. At admission, she was drowsy, extremely pale and afebrile. Her vitals included a pulse rate of 130/min, systemic blood pressure of 90/60 mmHg and respiratory rate of 28 breaths per minute. The remaining cardiovascular, respiratory and central nervous system findings were unremarkable. Abdominal examination revealed a diffuse tenderness and a well retracted uterus deviated towards the right-side of the abdomen. During speculum examination, a continuous fresh blood trickling was noted along with cervical tears at the 2 o’clock and 6 o’clock positions extending until the posterior fornix. Furthermore, the cervix appeared uneffaced and os patulous. Both fornices were boggy and tender; a transverse rent was noted in the posterior fornix through which the posterior wall of the uterus was felt. A bedside pelvic ultrasound scan revealed a post-abortal uterus with an empty cavity and a moderate amount of free fluid in the abdomen. Resuscitation was performed under guided haemodynamic monitoring and an immediate exploratory laparotomy was planned under general anaesthesia. Intraoperatively, 500mL of haemoperitoneum was noted with large clots in the pelvic cavity. There was a full-length scar rupture with the rent extending posteriorly. The posterior aspect of the uterus was avulsed completely above the level of the internal os and it was suspended only by the round ligament bilaterally; it appeared like a bucket handle tear [Figure 1]. The patient remained haemodynamically unstable and circumferential rent repair did not seem possible. There was a broad ligament haematoma of 5 × 5 cm on the left-side. The left uterine artery appeared avulsed and could not be traced. Bladder integrity was normal and there was no evidence of haematuria. Total abdominal hysterectomy along with left internal artery ligation was performed to achieve haemostasis. The patient was successfully resuscitated and extubated inside the operation theatre. Her postoperative investigations were unremarkable. The patient’s clinical condition improved completely and she was discharged from the hospital on the seventh postoperative day. The patient was doing well when she was last contacted nearly six months after her surgery. The patient provided written informed consent for publication of the anonymised data during her first follow-up. Discussion An unknown dose of misoprostol followed by dilatation and curettage in a previously scarred uterus could have led to uterine rupture in the current patient. Uterine rupture is primarily a clinical diagnosis, and hence, prompt surgical management is critical. Risk factors for uterine rupture include a previous uterine scar, short interpregnancy interval, multiparity, uterotonic drugs and obstructed labour.5 Additionally, the incidence of uterine rupture has been noted to be higher in scarred compared to unscarred uteri (0.28% versus 0.04%).1 A patient’s survival after uterine rupture depends on the time interval between the rupture and the intervention and urgent referral to a tertiary care centre. Rent repair is the key to treatment and internal iliac artery ligation is a lifesaving procedure in cases of uncontrolled obstetric haemorrhage. However, hysterectomy should not be delayed if the bleeding is intractable or the uterine rupture is irreparable. Second trimester pregnancy termination using misoprostol in an appropriate dosage in supervised settings for women with cesarean scar is safe and it has been associated with uterine rupture in only 0.3–0.4% of cases.1,6,7 A retrospective report did not observe uterine rupture after misoprostol induction for the termination of second trimester pregnancy in a scarred uterus.8 In the current patient, a characteristic bucket handle type uterine rupture was noted as was also reported by Abubekar et al.9 An unyielding cervix with misoprostol-induced strong uterine contractions in a scarred uterus may predispose to this type of uterine rupture.10 In addition, unsupervised vigorous dilatation and curettage could have either led to iatrogenic uterine perforation or further aggravated the tear induced by misoprostol. Conclusion Unsupervised over-the-counter use of misoprostol in a scarred uterus to induce second trimester abortion can have catastrophic consequences. Dilatation and curettage should be avoided before ruling out uterine rupture. An early referral to a higher centre may prevent severe maternal morbidity and mortality. Figure 1 Photographs of the uterus of a 24-year-old female patient following over-the-counter use of misoprostol for abortion induction. A: Anterior view of the uterus circumferentially avulsed at the level of the internal os. B: Lateral view showing the avulsed uterus. C: Avulsed uterus suspended only through round ligaments. AUTHORS’ CONTRIBUTION NJ, HS, JS and AKJ conceptualised and designed the work as well as extracted the data. NJ, JS and AKJ interpreted the results. JS performed the statistical analysis. All authors contributed to the drafting and approved the final version of the manuscript.

MISOPROSTOL

DrugsGivenReaction

CC BY-ND

34888091

2021-11

Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Intentional product misuse'.

A Rare Type of Uterine Rupture Following Over-the-Counter Use of Misoprostol in Second Trimester Abortion. The use of misoprostol in the second trimester by a woman with a uterine scar may lead to severe contractions and uterine rupture. We report a 24-year-old pregnant female patient who presented at the Emergency Department at a tertiary care hospital in Puducherry, India, in 2020 with haemorrhagic shock. She was at 16 weeks of gestation and had taken over the counter misoprostol for inducing an abortion. A quick initial resuscitation and urgent laparotomy were performed. An irreparable circumferentially avulsed uterus suspended only by round ligaments was noted. Haemostasis required internal artery ligation and immediate total hysterectomy. The patient was doing well upon follow-up six months after the surgery. Proper and supervised use of misoprostol in the appropriate dosage can avoid life-threatening consequences of uterine rupture. pmcTermination of pregnancy in the first and second trimesters in patients with prior uterine scarring is challenging.1 Misoprostol is a commonly used drug for induction of abortion even in unsupervised clinical settings.2 However, its dosing regimen has remained controversial.3,4 Over-the-counter use of misoprostol has resulted in documented or undocumented incidents of varying grades of uterine rupture. Uterine rupture is a surgical emergency that requires immediate care to avoid life-threatening medical, surgical and psychological consequences.4 In the current case, a patient with a previous caesarean scar experienced a unique type of uterine rupture which developed after over-the-counter use of misoprostol to induce second trimester abortion. Case Report A 24-year-old G3P1L1A1 (pregnant woman who had had a full-term delivery and one abortion and also had one living child) presented at the Emergency Department of a tertiary care hospital in Puducherry, India, in 2020 with haemorrhagic shock. She had taken an unknown dosage of misoprostol (unsupervised) to induce abortion at the 16th week of gestation. She had undergone a caesarean delivery two years before the current pregnancy. She self-administered an unknown dosage of misoprostol through the vaginal route at four-hour intervals. She noticed expulsion of the products of conception after two hours of the last dose of misoprostol. A local physician performed dilatation and curettage to treat heavy vaginal bleeding one day prior to presentation at the hospital. Her vaginal bleeding did not subside and she could have developed iatrogenic uterine perforation after dilatation and curettage. At admission, she was drowsy, extremely pale and afebrile. Her vitals included a pulse rate of 130/min, systemic blood pressure of 90/60 mmHg and respiratory rate of 28 breaths per minute. The remaining cardiovascular, respiratory and central nervous system findings were unremarkable. Abdominal examination revealed a diffuse tenderness and a well retracted uterus deviated towards the right-side of the abdomen. During speculum examination, a continuous fresh blood trickling was noted along with cervical tears at the 2 o’clock and 6 o’clock positions extending until the posterior fornix. Furthermore, the cervix appeared uneffaced and os patulous. Both fornices were boggy and tender; a transverse rent was noted in the posterior fornix through which the posterior wall of the uterus was felt. A bedside pelvic ultrasound scan revealed a post-abortal uterus with an empty cavity and a moderate amount of free fluid in the abdomen. Resuscitation was performed under guided haemodynamic monitoring and an immediate exploratory laparotomy was planned under general anaesthesia. Intraoperatively, 500mL of haemoperitoneum was noted with large clots in the pelvic cavity. There was a full-length scar rupture with the rent extending posteriorly. The posterior aspect of the uterus was avulsed completely above the level of the internal os and it was suspended only by the round ligament bilaterally; it appeared like a bucket handle tear [Figure 1]. The patient remained haemodynamically unstable and circumferential rent repair did not seem possible. There was a broad ligament haematoma of 5 × 5 cm on the left-side. The left uterine artery appeared avulsed and could not be traced. Bladder integrity was normal and there was no evidence of haematuria. Total abdominal hysterectomy along with left internal artery ligation was performed to achieve haemostasis. The patient was successfully resuscitated and extubated inside the operation theatre. Her postoperative investigations were unremarkable. The patient’s clinical condition improved completely and she was discharged from the hospital on the seventh postoperative day. The patient was doing well when she was last contacted nearly six months after her surgery. The patient provided written informed consent for publication of the anonymised data during her first follow-up. Discussion An unknown dose of misoprostol followed by dilatation and curettage in a previously scarred uterus could have led to uterine rupture in the current patient. Uterine rupture is primarily a clinical diagnosis, and hence, prompt surgical management is critical. Risk factors for uterine rupture include a previous uterine scar, short interpregnancy interval, multiparity, uterotonic drugs and obstructed labour.5 Additionally, the incidence of uterine rupture has been noted to be higher in scarred compared to unscarred uteri (0.28% versus 0.04%).1 A patient’s survival after uterine rupture depends on the time interval between the rupture and the intervention and urgent referral to a tertiary care centre. Rent repair is the key to treatment and internal iliac artery ligation is a lifesaving procedure in cases of uncontrolled obstetric haemorrhage. However, hysterectomy should not be delayed if the bleeding is intractable or the uterine rupture is irreparable. Second trimester pregnancy termination using misoprostol in an appropriate dosage in supervised settings for women with cesarean scar is safe and it has been associated with uterine rupture in only 0.3–0.4% of cases.1,6,7 A retrospective report did not observe uterine rupture after misoprostol induction for the termination of second trimester pregnancy in a scarred uterus.8 In the current patient, a characteristic bucket handle type uterine rupture was noted as was also reported by Abubekar et al.9 An unyielding cervix with misoprostol-induced strong uterine contractions in a scarred uterus may predispose to this type of uterine rupture.10 In addition, unsupervised vigorous dilatation and curettage could have either led to iatrogenic uterine perforation or further aggravated the tear induced by misoprostol. Conclusion Unsupervised over-the-counter use of misoprostol in a scarred uterus to induce second trimester abortion can have catastrophic consequences. Dilatation and curettage should be avoided before ruling out uterine rupture. An early referral to a higher centre may prevent severe maternal morbidity and mortality. Figure 1 Photographs of the uterus of a 24-year-old female patient following over-the-counter use of misoprostol for abortion induction. A: Anterior view of the uterus circumferentially avulsed at the level of the internal os. B: Lateral view showing the avulsed uterus. C: Avulsed uterus suspended only through round ligaments. AUTHORS’ CONTRIBUTION NJ, HS, JS and AKJ conceptualised and designed the work as well as extracted the data. NJ, JS and AKJ interpreted the results. JS performed the statistical analysis. All authors contributed to the drafting and approved the final version of the manuscript.

MISOPROSTOL

DrugsGivenReaction

CC BY-ND

34888091

2021-11

Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Pelvic haematoma'.

A Rare Type of Uterine Rupture Following Over-the-Counter Use of Misoprostol in Second Trimester Abortion. The use of misoprostol in the second trimester by a woman with a uterine scar may lead to severe contractions and uterine rupture. We report a 24-year-old pregnant female patient who presented at the Emergency Department at a tertiary care hospital in Puducherry, India, in 2020 with haemorrhagic shock. She was at 16 weeks of gestation and had taken over the counter misoprostol for inducing an abortion. A quick initial resuscitation and urgent laparotomy were performed. An irreparable circumferentially avulsed uterus suspended only by round ligaments was noted. Haemostasis required internal artery ligation and immediate total hysterectomy. The patient was doing well upon follow-up six months after the surgery. Proper and supervised use of misoprostol in the appropriate dosage can avoid life-threatening consequences of uterine rupture. pmcTermination of pregnancy in the first and second trimesters in patients with prior uterine scarring is challenging.1 Misoprostol is a commonly used drug for induction of abortion even in unsupervised clinical settings.2 However, its dosing regimen has remained controversial.3,4 Over-the-counter use of misoprostol has resulted in documented or undocumented incidents of varying grades of uterine rupture. Uterine rupture is a surgical emergency that requires immediate care to avoid life-threatening medical, surgical and psychological consequences.4 In the current case, a patient with a previous caesarean scar experienced a unique type of uterine rupture which developed after over-the-counter use of misoprostol to induce second trimester abortion. Case Report A 24-year-old G3P1L1A1 (pregnant woman who had had a full-term delivery and one abortion and also had one living child) presented at the Emergency Department of a tertiary care hospital in Puducherry, India, in 2020 with haemorrhagic shock. She had taken an unknown dosage of misoprostol (unsupervised) to induce abortion at the 16th week of gestation. She had undergone a caesarean delivery two years before the current pregnancy. She self-administered an unknown dosage of misoprostol through the vaginal route at four-hour intervals. She noticed expulsion of the products of conception after two hours of the last dose of misoprostol. A local physician performed dilatation and curettage to treat heavy vaginal bleeding one day prior to presentation at the hospital. Her vaginal bleeding did not subside and she could have developed iatrogenic uterine perforation after dilatation and curettage. At admission, she was drowsy, extremely pale and afebrile. Her vitals included a pulse rate of 130/min, systemic blood pressure of 90/60 mmHg and respiratory rate of 28 breaths per minute. The remaining cardiovascular, respiratory and central nervous system findings were unremarkable. Abdominal examination revealed a diffuse tenderness and a well retracted uterus deviated towards the right-side of the abdomen. During speculum examination, a continuous fresh blood trickling was noted along with cervical tears at the 2 o’clock and 6 o’clock positions extending until the posterior fornix. Furthermore, the cervix appeared uneffaced and os patulous. Both fornices were boggy and tender; a transverse rent was noted in the posterior fornix through which the posterior wall of the uterus was felt. A bedside pelvic ultrasound scan revealed a post-abortal uterus with an empty cavity and a moderate amount of free fluid in the abdomen. Resuscitation was performed under guided haemodynamic monitoring and an immediate exploratory laparotomy was planned under general anaesthesia. Intraoperatively, 500mL of haemoperitoneum was noted with large clots in the pelvic cavity. There was a full-length scar rupture with the rent extending posteriorly. The posterior aspect of the uterus was avulsed completely above the level of the internal os and it was suspended only by the round ligament bilaterally; it appeared like a bucket handle tear [Figure 1]. The patient remained haemodynamically unstable and circumferential rent repair did not seem possible. There was a broad ligament haematoma of 5 × 5 cm on the left-side. The left uterine artery appeared avulsed and could not be traced. Bladder integrity was normal and there was no evidence of haematuria. Total abdominal hysterectomy along with left internal artery ligation was performed to achieve haemostasis. The patient was successfully resuscitated and extubated inside the operation theatre. Her postoperative investigations were unremarkable. The patient’s clinical condition improved completely and she was discharged from the hospital on the seventh postoperative day. The patient was doing well when she was last contacted nearly six months after her surgery. The patient provided written informed consent for publication of the anonymised data during her first follow-up. Discussion An unknown dose of misoprostol followed by dilatation and curettage in a previously scarred uterus could have led to uterine rupture in the current patient. Uterine rupture is primarily a clinical diagnosis, and hence, prompt surgical management is critical. Risk factors for uterine rupture include a previous uterine scar, short interpregnancy interval, multiparity, uterotonic drugs and obstructed labour.5 Additionally, the incidence of uterine rupture has been noted to be higher in scarred compared to unscarred uteri (0.28% versus 0.04%).1 A patient’s survival after uterine rupture depends on the time interval between the rupture and the intervention and urgent referral to a tertiary care centre. Rent repair is the key to treatment and internal iliac artery ligation is a lifesaving procedure in cases of uncontrolled obstetric haemorrhage. However, hysterectomy should not be delayed if the bleeding is intractable or the uterine rupture is irreparable. Second trimester pregnancy termination using misoprostol in an appropriate dosage in supervised settings for women with cesarean scar is safe and it has been associated with uterine rupture in only 0.3–0.4% of cases.1,6,7 A retrospective report did not observe uterine rupture after misoprostol induction for the termination of second trimester pregnancy in a scarred uterus.8 In the current patient, a characteristic bucket handle type uterine rupture was noted as was also reported by Abubekar et al.9 An unyielding cervix with misoprostol-induced strong uterine contractions in a scarred uterus may predispose to this type of uterine rupture.10 In addition, unsupervised vigorous dilatation and curettage could have either led to iatrogenic uterine perforation or further aggravated the tear induced by misoprostol. Conclusion Unsupervised over-the-counter use of misoprostol in a scarred uterus to induce second trimester abortion can have catastrophic consequences. Dilatation and curettage should be avoided before ruling out uterine rupture. An early referral to a higher centre may prevent severe maternal morbidity and mortality. Figure 1 Photographs of the uterus of a 24-year-old female patient following over-the-counter use of misoprostol for abortion induction. A: Anterior view of the uterus circumferentially avulsed at the level of the internal os. B: Lateral view showing the avulsed uterus. C: Avulsed uterus suspended only through round ligaments. AUTHORS’ CONTRIBUTION NJ, HS, JS and AKJ conceptualised and designed the work as well as extracted the data. NJ, JS and AKJ interpreted the results. JS performed the statistical analysis. All authors contributed to the drafting and approved the final version of the manuscript.

MISOPROSTOL

DrugsGivenReaction

CC BY-ND

34888091

2021-11

Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Shock haemorrhagic'.

A Rare Type of Uterine Rupture Following Over-the-Counter Use of Misoprostol in Second Trimester Abortion. The use of misoprostol in the second trimester by a woman with a uterine scar may lead to severe contractions and uterine rupture. We report a 24-year-old pregnant female patient who presented at the Emergency Department at a tertiary care hospital in Puducherry, India, in 2020 with haemorrhagic shock. She was at 16 weeks of gestation and had taken over the counter misoprostol for inducing an abortion. A quick initial resuscitation and urgent laparotomy were performed. An irreparable circumferentially avulsed uterus suspended only by round ligaments was noted. Haemostasis required internal artery ligation and immediate total hysterectomy. The patient was doing well upon follow-up six months after the surgery. Proper and supervised use of misoprostol in the appropriate dosage can avoid life-threatening consequences of uterine rupture. pmcTermination of pregnancy in the first and second trimesters in patients with prior uterine scarring is challenging.1 Misoprostol is a commonly used drug for induction of abortion even in unsupervised clinical settings.2 However, its dosing regimen has remained controversial.3,4 Over-the-counter use of misoprostol has resulted in documented or undocumented incidents of varying grades of uterine rupture. Uterine rupture is a surgical emergency that requires immediate care to avoid life-threatening medical, surgical and psychological consequences.4 In the current case, a patient with a previous caesarean scar experienced a unique type of uterine rupture which developed after over-the-counter use of misoprostol to induce second trimester abortion. Case Report A 24-year-old G3P1L1A1 (pregnant woman who had had a full-term delivery and one abortion and also had one living child) presented at the Emergency Department of a tertiary care hospital in Puducherry, India, in 2020 with haemorrhagic shock. She had taken an unknown dosage of misoprostol (unsupervised) to induce abortion at the 16th week of gestation. She had undergone a caesarean delivery two years before the current pregnancy. She self-administered an unknown dosage of misoprostol through the vaginal route at four-hour intervals. She noticed expulsion of the products of conception after two hours of the last dose of misoprostol. A local physician performed dilatation and curettage to treat heavy vaginal bleeding one day prior to presentation at the hospital. Her vaginal bleeding did not subside and she could have developed iatrogenic uterine perforation after dilatation and curettage. At admission, she was drowsy, extremely pale and afebrile. Her vitals included a pulse rate of 130/min, systemic blood pressure of 90/60 mmHg and respiratory rate of 28 breaths per minute. The remaining cardiovascular, respiratory and central nervous system findings were unremarkable. Abdominal examination revealed a diffuse tenderness and a well retracted uterus deviated towards the right-side of the abdomen. During speculum examination, a continuous fresh blood trickling was noted along with cervical tears at the 2 o’clock and 6 o’clock positions extending until the posterior fornix. Furthermore, the cervix appeared uneffaced and os patulous. Both fornices were boggy and tender; a transverse rent was noted in the posterior fornix through which the posterior wall of the uterus was felt. A bedside pelvic ultrasound scan revealed a post-abortal uterus with an empty cavity and a moderate amount of free fluid in the abdomen. Resuscitation was performed under guided haemodynamic monitoring and an immediate exploratory laparotomy was planned under general anaesthesia. Intraoperatively, 500mL of haemoperitoneum was noted with large clots in the pelvic cavity. There was a full-length scar rupture with the rent extending posteriorly. The posterior aspect of the uterus was avulsed completely above the level of the internal os and it was suspended only by the round ligament bilaterally; it appeared like a bucket handle tear [Figure 1]. The patient remained haemodynamically unstable and circumferential rent repair did not seem possible. There was a broad ligament haematoma of 5 × 5 cm on the left-side. The left uterine artery appeared avulsed and could not be traced. Bladder integrity was normal and there was no evidence of haematuria. Total abdominal hysterectomy along with left internal artery ligation was performed to achieve haemostasis. The patient was successfully resuscitated and extubated inside the operation theatre. Her postoperative investigations were unremarkable. The patient’s clinical condition improved completely and she was discharged from the hospital on the seventh postoperative day. The patient was doing well when she was last contacted nearly six months after her surgery. The patient provided written informed consent for publication of the anonymised data during her first follow-up. Discussion An unknown dose of misoprostol followed by dilatation and curettage in a previously scarred uterus could have led to uterine rupture in the current patient. Uterine rupture is primarily a clinical diagnosis, and hence, prompt surgical management is critical. Risk factors for uterine rupture include a previous uterine scar, short interpregnancy interval, multiparity, uterotonic drugs and obstructed labour.5 Additionally, the incidence of uterine rupture has been noted to be higher in scarred compared to unscarred uteri (0.28% versus 0.04%).1 A patient’s survival after uterine rupture depends on the time interval between the rupture and the intervention and urgent referral to a tertiary care centre. Rent repair is the key to treatment and internal iliac artery ligation is a lifesaving procedure in cases of uncontrolled obstetric haemorrhage. However, hysterectomy should not be delayed if the bleeding is intractable or the uterine rupture is irreparable. Second trimester pregnancy termination using misoprostol in an appropriate dosage in supervised settings for women with cesarean scar is safe and it has been associated with uterine rupture in only 0.3–0.4% of cases.1,6,7 A retrospective report did not observe uterine rupture after misoprostol induction for the termination of second trimester pregnancy in a scarred uterus.8 In the current patient, a characteristic bucket handle type uterine rupture was noted as was also reported by Abubekar et al.9 An unyielding cervix with misoprostol-induced strong uterine contractions in a scarred uterus may predispose to this type of uterine rupture.10 In addition, unsupervised vigorous dilatation and curettage could have either led to iatrogenic uterine perforation or further aggravated the tear induced by misoprostol. Conclusion Unsupervised over-the-counter use of misoprostol in a scarred uterus to induce second trimester abortion can have catastrophic consequences. Dilatation and curettage should be avoided before ruling out uterine rupture. An early referral to a higher centre may prevent severe maternal morbidity and mortality. Figure 1 Photographs of the uterus of a 24-year-old female patient following over-the-counter use of misoprostol for abortion induction. A: Anterior view of the uterus circumferentially avulsed at the level of the internal os. B: Lateral view showing the avulsed uterus. C: Avulsed uterus suspended only through round ligaments. AUTHORS’ CONTRIBUTION NJ, HS, JS and AKJ conceptualised and designed the work as well as extracted the data. NJ, JS and AKJ interpreted the results. JS performed the statistical analysis. All authors contributed to the drafting and approved the final version of the manuscript.

MISOPROSTOL

DrugsGivenReaction

CC BY-ND

34888091

2021-11

Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Uterine rupture'.

A Rare Type of Uterine Rupture Following Over-the-Counter Use of Misoprostol in Second Trimester Abortion. The use of misoprostol in the second trimester by a woman with a uterine scar may lead to severe contractions and uterine rupture. We report a 24-year-old pregnant female patient who presented at the Emergency Department at a tertiary care hospital in Puducherry, India, in 2020 with haemorrhagic shock. She was at 16 weeks of gestation and had taken over the counter misoprostol for inducing an abortion. A quick initial resuscitation and urgent laparotomy were performed. An irreparable circumferentially avulsed uterus suspended only by round ligaments was noted. Haemostasis required internal artery ligation and immediate total hysterectomy. The patient was doing well upon follow-up six months after the surgery. Proper and supervised use of misoprostol in the appropriate dosage can avoid life-threatening consequences of uterine rupture. pmcTermination of pregnancy in the first and second trimesters in patients with prior uterine scarring is challenging.1 Misoprostol is a commonly used drug for induction of abortion even in unsupervised clinical settings.2 However, its dosing regimen has remained controversial.3,4 Over-the-counter use of misoprostol has resulted in documented or undocumented incidents of varying grades of uterine rupture. Uterine rupture is a surgical emergency that requires immediate care to avoid life-threatening medical, surgical and psychological consequences.4 In the current case, a patient with a previous caesarean scar experienced a unique type of uterine rupture which developed after over-the-counter use of misoprostol to induce second trimester abortion. Case Report A 24-year-old G3P1L1A1 (pregnant woman who had had a full-term delivery and one abortion and also had one living child) presented at the Emergency Department of a tertiary care hospital in Puducherry, India, in 2020 with haemorrhagic shock. She had taken an unknown dosage of misoprostol (unsupervised) to induce abortion at the 16th week of gestation. She had undergone a caesarean delivery two years before the current pregnancy. She self-administered an unknown dosage of misoprostol through the vaginal route at four-hour intervals. She noticed expulsion of the products of conception after two hours of the last dose of misoprostol. A local physician performed dilatation and curettage to treat heavy vaginal bleeding one day prior to presentation at the hospital. Her vaginal bleeding did not subside and she could have developed iatrogenic uterine perforation after dilatation and curettage. At admission, she was drowsy, extremely pale and afebrile. Her vitals included a pulse rate of 130/min, systemic blood pressure of 90/60 mmHg and respiratory rate of 28 breaths per minute. The remaining cardiovascular, respiratory and central nervous system findings were unremarkable. Abdominal examination revealed a diffuse tenderness and a well retracted uterus deviated towards the right-side of the abdomen. During speculum examination, a continuous fresh blood trickling was noted along with cervical tears at the 2 o’clock and 6 o’clock positions extending until the posterior fornix. Furthermore, the cervix appeared uneffaced and os patulous. Both fornices were boggy and tender; a transverse rent was noted in the posterior fornix through which the posterior wall of the uterus was felt. A bedside pelvic ultrasound scan revealed a post-abortal uterus with an empty cavity and a moderate amount of free fluid in the abdomen. Resuscitation was performed under guided haemodynamic monitoring and an immediate exploratory laparotomy was planned under general anaesthesia. Intraoperatively, 500mL of haemoperitoneum was noted with large clots in the pelvic cavity. There was a full-length scar rupture with the rent extending posteriorly. The posterior aspect of the uterus was avulsed completely above the level of the internal os and it was suspended only by the round ligament bilaterally; it appeared like a bucket handle tear [Figure 1]. The patient remained haemodynamically unstable and circumferential rent repair did not seem possible. There was a broad ligament haematoma of 5 × 5 cm on the left-side. The left uterine artery appeared avulsed and could not be traced. Bladder integrity was normal and there was no evidence of haematuria. Total abdominal hysterectomy along with left internal artery ligation was performed to achieve haemostasis. The patient was successfully resuscitated and extubated inside the operation theatre. Her postoperative investigations were unremarkable. The patient’s clinical condition improved completely and she was discharged from the hospital on the seventh postoperative day. The patient was doing well when she was last contacted nearly six months after her surgery. The patient provided written informed consent for publication of the anonymised data during her first follow-up. Discussion An unknown dose of misoprostol followed by dilatation and curettage in a previously scarred uterus could have led to uterine rupture in the current patient. Uterine rupture is primarily a clinical diagnosis, and hence, prompt surgical management is critical. Risk factors for uterine rupture include a previous uterine scar, short interpregnancy interval, multiparity, uterotonic drugs and obstructed labour.5 Additionally, the incidence of uterine rupture has been noted to be higher in scarred compared to unscarred uteri (0.28% versus 0.04%).1 A patient’s survival after uterine rupture depends on the time interval between the rupture and the intervention and urgent referral to a tertiary care centre. Rent repair is the key to treatment and internal iliac artery ligation is a lifesaving procedure in cases of uncontrolled obstetric haemorrhage. However, hysterectomy should not be delayed if the bleeding is intractable or the uterine rupture is irreparable. Second trimester pregnancy termination using misoprostol in an appropriate dosage in supervised settings for women with cesarean scar is safe and it has been associated with uterine rupture in only 0.3–0.4% of cases.1,6,7 A retrospective report did not observe uterine rupture after misoprostol induction for the termination of second trimester pregnancy in a scarred uterus.8 In the current patient, a characteristic bucket handle type uterine rupture was noted as was also reported by Abubekar et al.9 An unyielding cervix with misoprostol-induced strong uterine contractions in a scarred uterus may predispose to this type of uterine rupture.10 In addition, unsupervised vigorous dilatation and curettage could have either led to iatrogenic uterine perforation or further aggravated the tear induced by misoprostol. Conclusion Unsupervised over-the-counter use of misoprostol in a scarred uterus to induce second trimester abortion can have catastrophic consequences. Dilatation and curettage should be avoided before ruling out uterine rupture. An early referral to a higher centre may prevent severe maternal morbidity and mortality. Figure 1 Photographs of the uterus of a 24-year-old female patient following over-the-counter use of misoprostol for abortion induction. A: Anterior view of the uterus circumferentially avulsed at the level of the internal os. B: Lateral view showing the avulsed uterus. C: Avulsed uterus suspended only through round ligaments. AUTHORS’ CONTRIBUTION NJ, HS, JS and AKJ conceptualised and designed the work as well as extracted the data. NJ, JS and AKJ interpreted the results. JS performed the statistical analysis. All authors contributed to the drafting and approved the final version of the manuscript.

MISOPROSTOL

DrugsGivenReaction

CC BY-ND

34888091

2021-11

What was the administration route of drug 'MISOPROSTOL'?

A Rare Type of Uterine Rupture Following Over-the-Counter Use of Misoprostol in Second Trimester Abortion. The use of misoprostol in the second trimester by a woman with a uterine scar may lead to severe contractions and uterine rupture. We report a 24-year-old pregnant female patient who presented at the Emergency Department at a tertiary care hospital in Puducherry, India, in 2020 with haemorrhagic shock. She was at 16 weeks of gestation and had taken over the counter misoprostol for inducing an abortion. A quick initial resuscitation and urgent laparotomy were performed. An irreparable circumferentially avulsed uterus suspended only by round ligaments was noted. Haemostasis required internal artery ligation and immediate total hysterectomy. The patient was doing well upon follow-up six months after the surgery. Proper and supervised use of misoprostol in the appropriate dosage can avoid life-threatening consequences of uterine rupture. pmcTermination of pregnancy in the first and second trimesters in patients with prior uterine scarring is challenging.1 Misoprostol is a commonly used drug for induction of abortion even in unsupervised clinical settings.2 However, its dosing regimen has remained controversial.3,4 Over-the-counter use of misoprostol has resulted in documented or undocumented incidents of varying grades of uterine rupture. Uterine rupture is a surgical emergency that requires immediate care to avoid life-threatening medical, surgical and psychological consequences.4 In the current case, a patient with a previous caesarean scar experienced a unique type of uterine rupture which developed after over-the-counter use of misoprostol to induce second trimester abortion. Case Report A 24-year-old G3P1L1A1 (pregnant woman who had had a full-term delivery and one abortion and also had one living child) presented at the Emergency Department of a tertiary care hospital in Puducherry, India, in 2020 with haemorrhagic shock. She had taken an unknown dosage of misoprostol (unsupervised) to induce abortion at the 16th week of gestation. She had undergone a caesarean delivery two years before the current pregnancy. She self-administered an unknown dosage of misoprostol through the vaginal route at four-hour intervals. She noticed expulsion of the products of conception after two hours of the last dose of misoprostol. A local physician performed dilatation and curettage to treat heavy vaginal bleeding one day prior to presentation at the hospital. Her vaginal bleeding did not subside and she could have developed iatrogenic uterine perforation after dilatation and curettage. At admission, she was drowsy, extremely pale and afebrile. Her vitals included a pulse rate of 130/min, systemic blood pressure of 90/60 mmHg and respiratory rate of 28 breaths per minute. The remaining cardiovascular, respiratory and central nervous system findings were unremarkable. Abdominal examination revealed a diffuse tenderness and a well retracted uterus deviated towards the right-side of the abdomen. During speculum examination, a continuous fresh blood trickling was noted along with cervical tears at the 2 o’clock and 6 o’clock positions extending until the posterior fornix. Furthermore, the cervix appeared uneffaced and os patulous. Both fornices were boggy and tender; a transverse rent was noted in the posterior fornix through which the posterior wall of the uterus was felt. A bedside pelvic ultrasound scan revealed a post-abortal uterus with an empty cavity and a moderate amount of free fluid in the abdomen. Resuscitation was performed under guided haemodynamic monitoring and an immediate exploratory laparotomy was planned under general anaesthesia. Intraoperatively, 500mL of haemoperitoneum was noted with large clots in the pelvic cavity. There was a full-length scar rupture with the rent extending posteriorly. The posterior aspect of the uterus was avulsed completely above the level of the internal os and it was suspended only by the round ligament bilaterally; it appeared like a bucket handle tear [Figure 1]. The patient remained haemodynamically unstable and circumferential rent repair did not seem possible. There was a broad ligament haematoma of 5 × 5 cm on the left-side. The left uterine artery appeared avulsed and could not be traced. Bladder integrity was normal and there was no evidence of haematuria. Total abdominal hysterectomy along with left internal artery ligation was performed to achieve haemostasis. The patient was successfully resuscitated and extubated inside the operation theatre. Her postoperative investigations were unremarkable. The patient’s clinical condition improved completely and she was discharged from the hospital on the seventh postoperative day. The patient was doing well when she was last contacted nearly six months after her surgery. The patient provided written informed consent for publication of the anonymised data during her first follow-up. Discussion An unknown dose of misoprostol followed by dilatation and curettage in a previously scarred uterus could have led to uterine rupture in the current patient. Uterine rupture is primarily a clinical diagnosis, and hence, prompt surgical management is critical. Risk factors for uterine rupture include a previous uterine scar, short interpregnancy interval, multiparity, uterotonic drugs and obstructed labour.5 Additionally, the incidence of uterine rupture has been noted to be higher in scarred compared to unscarred uteri (0.28% versus 0.04%).1 A patient’s survival after uterine rupture depends on the time interval between the rupture and the intervention and urgent referral to a tertiary care centre. Rent repair is the key to treatment and internal iliac artery ligation is a lifesaving procedure in cases of uncontrolled obstetric haemorrhage. However, hysterectomy should not be delayed if the bleeding is intractable or the uterine rupture is irreparable. Second trimester pregnancy termination using misoprostol in an appropriate dosage in supervised settings for women with cesarean scar is safe and it has been associated with uterine rupture in only 0.3–0.4% of cases.1,6,7 A retrospective report did not observe uterine rupture after misoprostol induction for the termination of second trimester pregnancy in a scarred uterus.8 In the current patient, a characteristic bucket handle type uterine rupture was noted as was also reported by Abubekar et al.9 An unyielding cervix with misoprostol-induced strong uterine contractions in a scarred uterus may predispose to this type of uterine rupture.10 In addition, unsupervised vigorous dilatation and curettage could have either led to iatrogenic uterine perforation or further aggravated the tear induced by misoprostol. Conclusion Unsupervised over-the-counter use of misoprostol in a scarred uterus to induce second trimester abortion can have catastrophic consequences. Dilatation and curettage should be avoided before ruling out uterine rupture. An early referral to a higher centre may prevent severe maternal morbidity and mortality. Figure 1 Photographs of the uterus of a 24-year-old female patient following over-the-counter use of misoprostol for abortion induction. A: Anterior view of the uterus circumferentially avulsed at the level of the internal os. B: Lateral view showing the avulsed uterus. C: Avulsed uterus suspended only through round ligaments. AUTHORS’ CONTRIBUTION NJ, HS, JS and AKJ conceptualised and designed the work as well as extracted the data. NJ, JS and AKJ interpreted the results. JS performed the statistical analysis. All authors contributed to the drafting and approved the final version of the manuscript.

Vaginal

DrugAdministrationRoute

CC BY-ND

34888091

2021-11

What was the outcome of reaction 'Cervix disorder'?

A Rare Type of Uterine Rupture Following Over-the-Counter Use of Misoprostol in Second Trimester Abortion. The use of misoprostol in the second trimester by a woman with a uterine scar may lead to severe contractions and uterine rupture. We report a 24-year-old pregnant female patient who presented at the Emergency Department at a tertiary care hospital in Puducherry, India, in 2020 with haemorrhagic shock. She was at 16 weeks of gestation and had taken over the counter misoprostol for inducing an abortion. A quick initial resuscitation and urgent laparotomy were performed. An irreparable circumferentially avulsed uterus suspended only by round ligaments was noted. Haemostasis required internal artery ligation and immediate total hysterectomy. The patient was doing well upon follow-up six months after the surgery. Proper and supervised use of misoprostol in the appropriate dosage can avoid life-threatening consequences of uterine rupture. pmcTermination of pregnancy in the first and second trimesters in patients with prior uterine scarring is challenging.1 Misoprostol is a commonly used drug for induction of abortion even in unsupervised clinical settings.2 However, its dosing regimen has remained controversial.3,4 Over-the-counter use of misoprostol has resulted in documented or undocumented incidents of varying grades of uterine rupture. Uterine rupture is a surgical emergency that requires immediate care to avoid life-threatening medical, surgical and psychological consequences.4 In the current case, a patient with a previous caesarean scar experienced a unique type of uterine rupture which developed after over-the-counter use of misoprostol to induce second trimester abortion. Case Report A 24-year-old G3P1L1A1 (pregnant woman who had had a full-term delivery and one abortion and also had one living child) presented at the Emergency Department of a tertiary care hospital in Puducherry, India, in 2020 with haemorrhagic shock. She had taken an unknown dosage of misoprostol (unsupervised) to induce abortion at the 16th week of gestation. She had undergone a caesarean delivery two years before the current pregnancy. She self-administered an unknown dosage of misoprostol through the vaginal route at four-hour intervals. She noticed expulsion of the products of conception after two hours of the last dose of misoprostol. A local physician performed dilatation and curettage to treat heavy vaginal bleeding one day prior to presentation at the hospital. Her vaginal bleeding did not subside and she could have developed iatrogenic uterine perforation after dilatation and curettage. At admission, she was drowsy, extremely pale and afebrile. Her vitals included a pulse rate of 130/min, systemic blood pressure of 90/60 mmHg and respiratory rate of 28 breaths per minute. The remaining cardiovascular, respiratory and central nervous system findings were unremarkable. Abdominal examination revealed a diffuse tenderness and a well retracted uterus deviated towards the right-side of the abdomen. During speculum examination, a continuous fresh blood trickling was noted along with cervical tears at the 2 o’clock and 6 o’clock positions extending until the posterior fornix. Furthermore, the cervix appeared uneffaced and os patulous. Both fornices were boggy and tender; a transverse rent was noted in the posterior fornix through which the posterior wall of the uterus was felt. A bedside pelvic ultrasound scan revealed a post-abortal uterus with an empty cavity and a moderate amount of free fluid in the abdomen. Resuscitation was performed under guided haemodynamic monitoring and an immediate exploratory laparotomy was planned under general anaesthesia. Intraoperatively, 500mL of haemoperitoneum was noted with large clots in the pelvic cavity. There was a full-length scar rupture with the rent extending posteriorly. The posterior aspect of the uterus was avulsed completely above the level of the internal os and it was suspended only by the round ligament bilaterally; it appeared like a bucket handle tear [Figure 1]. The patient remained haemodynamically unstable and circumferential rent repair did not seem possible. There was a broad ligament haematoma of 5 × 5 cm on the left-side. The left uterine artery appeared avulsed and could not be traced. Bladder integrity was normal and there was no evidence of haematuria. Total abdominal hysterectomy along with left internal artery ligation was performed to achieve haemostasis. The patient was successfully resuscitated and extubated inside the operation theatre. Her postoperative investigations were unremarkable. The patient’s clinical condition improved completely and she was discharged from the hospital on the seventh postoperative day. The patient was doing well when she was last contacted nearly six months after her surgery. The patient provided written informed consent for publication of the anonymised data during her first follow-up. Discussion An unknown dose of misoprostol followed by dilatation and curettage in a previously scarred uterus could have led to uterine rupture in the current patient. Uterine rupture is primarily a clinical diagnosis, and hence, prompt surgical management is critical. Risk factors for uterine rupture include a previous uterine scar, short interpregnancy interval, multiparity, uterotonic drugs and obstructed labour.5 Additionally, the incidence of uterine rupture has been noted to be higher in scarred compared to unscarred uteri (0.28% versus 0.04%).1 A patient’s survival after uterine rupture depends on the time interval between the rupture and the intervention and urgent referral to a tertiary care centre. Rent repair is the key to treatment and internal iliac artery ligation is a lifesaving procedure in cases of uncontrolled obstetric haemorrhage. However, hysterectomy should not be delayed if the bleeding is intractable or the uterine rupture is irreparable. Second trimester pregnancy termination using misoprostol in an appropriate dosage in supervised settings for women with cesarean scar is safe and it has been associated with uterine rupture in only 0.3–0.4% of cases.1,6,7 A retrospective report did not observe uterine rupture after misoprostol induction for the termination of second trimester pregnancy in a scarred uterus.8 In the current patient, a characteristic bucket handle type uterine rupture was noted as was also reported by Abubekar et al.9 An unyielding cervix with misoprostol-induced strong uterine contractions in a scarred uterus may predispose to this type of uterine rupture.10 In addition, unsupervised vigorous dilatation and curettage could have either led to iatrogenic uterine perforation or further aggravated the tear induced by misoprostol. Conclusion Unsupervised over-the-counter use of misoprostol in a scarred uterus to induce second trimester abortion can have catastrophic consequences. Dilatation and curettage should be avoided before ruling out uterine rupture. An early referral to a higher centre may prevent severe maternal morbidity and mortality. Figure 1 Photographs of the uterus of a 24-year-old female patient following over-the-counter use of misoprostol for abortion induction. A: Anterior view of the uterus circumferentially avulsed at the level of the internal os. B: Lateral view showing the avulsed uterus. C: Avulsed uterus suspended only through round ligaments. AUTHORS’ CONTRIBUTION NJ, HS, JS and AKJ conceptualised and designed the work as well as extracted the data. NJ, JS and AKJ interpreted the results. JS performed the statistical analysis. All authors contributed to the drafting and approved the final version of the manuscript.

Recovered

ReactionOutcome

CC BY-ND

34888091

2021-11

What was the outcome of reaction 'Haemoperitoneum'?

A Rare Type of Uterine Rupture Following Over-the-Counter Use of Misoprostol in Second Trimester Abortion. The use of misoprostol in the second trimester by a woman with a uterine scar may lead to severe contractions and uterine rupture. We report a 24-year-old pregnant female patient who presented at the Emergency Department at a tertiary care hospital in Puducherry, India, in 2020 with haemorrhagic shock. She was at 16 weeks of gestation and had taken over the counter misoprostol for inducing an abortion. A quick initial resuscitation and urgent laparotomy were performed. An irreparable circumferentially avulsed uterus suspended only by round ligaments was noted. Haemostasis required internal artery ligation and immediate total hysterectomy. The patient was doing well upon follow-up six months after the surgery. Proper and supervised use of misoprostol in the appropriate dosage can avoid life-threatening consequences of uterine rupture. pmcTermination of pregnancy in the first and second trimesters in patients with prior uterine scarring is challenging.1 Misoprostol is a commonly used drug for induction of abortion even in unsupervised clinical settings.2 However, its dosing regimen has remained controversial.3,4 Over-the-counter use of misoprostol has resulted in documented or undocumented incidents of varying grades of uterine rupture. Uterine rupture is a surgical emergency that requires immediate care to avoid life-threatening medical, surgical and psychological consequences.4 In the current case, a patient with a previous caesarean scar experienced a unique type of uterine rupture which developed after over-the-counter use of misoprostol to induce second trimester abortion. Case Report A 24-year-old G3P1L1A1 (pregnant woman who had had a full-term delivery and one abortion and also had one living child) presented at the Emergency Department of a tertiary care hospital in Puducherry, India, in 2020 with haemorrhagic shock. She had taken an unknown dosage of misoprostol (unsupervised) to induce abortion at the 16th week of gestation. She had undergone a caesarean delivery two years before the current pregnancy. She self-administered an unknown dosage of misoprostol through the vaginal route at four-hour intervals. She noticed expulsion of the products of conception after two hours of the last dose of misoprostol. A local physician performed dilatation and curettage to treat heavy vaginal bleeding one day prior to presentation at the hospital. Her vaginal bleeding did not subside and she could have developed iatrogenic uterine perforation after dilatation and curettage. At admission, she was drowsy, extremely pale and afebrile. Her vitals included a pulse rate of 130/min, systemic blood pressure of 90/60 mmHg and respiratory rate of 28 breaths per minute. The remaining cardiovascular, respiratory and central nervous system findings were unremarkable. Abdominal examination revealed a diffuse tenderness and a well retracted uterus deviated towards the right-side of the abdomen. During speculum examination, a continuous fresh blood trickling was noted along with cervical tears at the 2 o’clock and 6 o’clock positions extending until the posterior fornix. Furthermore, the cervix appeared uneffaced and os patulous. Both fornices were boggy and tender; a transverse rent was noted in the posterior fornix through which the posterior wall of the uterus was felt. A bedside pelvic ultrasound scan revealed a post-abortal uterus with an empty cavity and a moderate amount of free fluid in the abdomen. Resuscitation was performed under guided haemodynamic monitoring and an immediate exploratory laparotomy was planned under general anaesthesia. Intraoperatively, 500mL of haemoperitoneum was noted with large clots in the pelvic cavity. There was a full-length scar rupture with the rent extending posteriorly. The posterior aspect of the uterus was avulsed completely above the level of the internal os and it was suspended only by the round ligament bilaterally; it appeared like a bucket handle tear [Figure 1]. The patient remained haemodynamically unstable and circumferential rent repair did not seem possible. There was a broad ligament haematoma of 5 × 5 cm on the left-side. The left uterine artery appeared avulsed and could not be traced. Bladder integrity was normal and there was no evidence of haematuria. Total abdominal hysterectomy along with left internal artery ligation was performed to achieve haemostasis. The patient was successfully resuscitated and extubated inside the operation theatre. Her postoperative investigations were unremarkable. The patient’s clinical condition improved completely and she was discharged from the hospital on the seventh postoperative day. The patient was doing well when she was last contacted nearly six months after her surgery. The patient provided written informed consent for publication of the anonymised data during her first follow-up. Discussion An unknown dose of misoprostol followed by dilatation and curettage in a previously scarred uterus could have led to uterine rupture in the current patient. Uterine rupture is primarily a clinical diagnosis, and hence, prompt surgical management is critical. Risk factors for uterine rupture include a previous uterine scar, short interpregnancy interval, multiparity, uterotonic drugs and obstructed labour.5 Additionally, the incidence of uterine rupture has been noted to be higher in scarred compared to unscarred uteri (0.28% versus 0.04%).1 A patient’s survival after uterine rupture depends on the time interval between the rupture and the intervention and urgent referral to a tertiary care centre. Rent repair is the key to treatment and internal iliac artery ligation is a lifesaving procedure in cases of uncontrolled obstetric haemorrhage. However, hysterectomy should not be delayed if the bleeding is intractable or the uterine rupture is irreparable. Second trimester pregnancy termination using misoprostol in an appropriate dosage in supervised settings for women with cesarean scar is safe and it has been associated with uterine rupture in only 0.3–0.4% of cases.1,6,7 A retrospective report did not observe uterine rupture after misoprostol induction for the termination of second trimester pregnancy in a scarred uterus.8 In the current patient, a characteristic bucket handle type uterine rupture was noted as was also reported by Abubekar et al.9 An unyielding cervix with misoprostol-induced strong uterine contractions in a scarred uterus may predispose to this type of uterine rupture.10 In addition, unsupervised vigorous dilatation and curettage could have either led to iatrogenic uterine perforation or further aggravated the tear induced by misoprostol. Conclusion Unsupervised over-the-counter use of misoprostol in a scarred uterus to induce second trimester abortion can have catastrophic consequences. Dilatation and curettage should be avoided before ruling out uterine rupture. An early referral to a higher centre may prevent severe maternal morbidity and mortality. Figure 1 Photographs of the uterus of a 24-year-old female patient following over-the-counter use of misoprostol for abortion induction. A: Anterior view of the uterus circumferentially avulsed at the level of the internal os. B: Lateral view showing the avulsed uterus. C: Avulsed uterus suspended only through round ligaments. AUTHORS’ CONTRIBUTION NJ, HS, JS and AKJ conceptualised and designed the work as well as extracted the data. NJ, JS and AKJ interpreted the results. JS performed the statistical analysis. All authors contributed to the drafting and approved the final version of the manuscript.

Recovered

ReactionOutcome

CC BY-ND

34888091

2021-11

What was the outcome of reaction 'Pelvic haematoma'?

A Rare Type of Uterine Rupture Following Over-the-Counter Use of Misoprostol in Second Trimester Abortion. The use of misoprostol in the second trimester by a woman with a uterine scar may lead to severe contractions and uterine rupture. We report a 24-year-old pregnant female patient who presented at the Emergency Department at a tertiary care hospital in Puducherry, India, in 2020 with haemorrhagic shock. She was at 16 weeks of gestation and had taken over the counter misoprostol for inducing an abortion. A quick initial resuscitation and urgent laparotomy were performed. An irreparable circumferentially avulsed uterus suspended only by round ligaments was noted. Haemostasis required internal artery ligation and immediate total hysterectomy. The patient was doing well upon follow-up six months after the surgery. Proper and supervised use of misoprostol in the appropriate dosage can avoid life-threatening consequences of uterine rupture. pmcTermination of pregnancy in the first and second trimesters in patients with prior uterine scarring is challenging.1 Misoprostol is a commonly used drug for induction of abortion even in unsupervised clinical settings.2 However, its dosing regimen has remained controversial.3,4 Over-the-counter use of misoprostol has resulted in documented or undocumented incidents of varying grades of uterine rupture. Uterine rupture is a surgical emergency that requires immediate care to avoid life-threatening medical, surgical and psychological consequences.4 In the current case, a patient with a previous caesarean scar experienced a unique type of uterine rupture which developed after over-the-counter use of misoprostol to induce second trimester abortion. Case Report A 24-year-old G3P1L1A1 (pregnant woman who had had a full-term delivery and one abortion and also had one living child) presented at the Emergency Department of a tertiary care hospital in Puducherry, India, in 2020 with haemorrhagic shock. She had taken an unknown dosage of misoprostol (unsupervised) to induce abortion at the 16th week of gestation. She had undergone a caesarean delivery two years before the current pregnancy. She self-administered an unknown dosage of misoprostol through the vaginal route at four-hour intervals. She noticed expulsion of the products of conception after two hours of the last dose of misoprostol. A local physician performed dilatation and curettage to treat heavy vaginal bleeding one day prior to presentation at the hospital. Her vaginal bleeding did not subside and she could have developed iatrogenic uterine perforation after dilatation and curettage. At admission, she was drowsy, extremely pale and afebrile. Her vitals included a pulse rate of 130/min, systemic blood pressure of 90/60 mmHg and respiratory rate of 28 breaths per minute. The remaining cardiovascular, respiratory and central nervous system findings were unremarkable. Abdominal examination revealed a diffuse tenderness and a well retracted uterus deviated towards the right-side of the abdomen. During speculum examination, a continuous fresh blood trickling was noted along with cervical tears at the 2 o’clock and 6 o’clock positions extending until the posterior fornix. Furthermore, the cervix appeared uneffaced and os patulous. Both fornices were boggy and tender; a transverse rent was noted in the posterior fornix through which the posterior wall of the uterus was felt. A bedside pelvic ultrasound scan revealed a post-abortal uterus with an empty cavity and a moderate amount of free fluid in the abdomen. Resuscitation was performed under guided haemodynamic monitoring and an immediate exploratory laparotomy was planned under general anaesthesia. Intraoperatively, 500mL of haemoperitoneum was noted with large clots in the pelvic cavity. There was a full-length scar rupture with the rent extending posteriorly. The posterior aspect of the uterus was avulsed completely above the level of the internal os and it was suspended only by the round ligament bilaterally; it appeared like a bucket handle tear [Figure 1]. The patient remained haemodynamically unstable and circumferential rent repair did not seem possible. There was a broad ligament haematoma of 5 × 5 cm on the left-side. The left uterine artery appeared avulsed and could not be traced. Bladder integrity was normal and there was no evidence of haematuria. Total abdominal hysterectomy along with left internal artery ligation was performed to achieve haemostasis. The patient was successfully resuscitated and extubated inside the operation theatre. Her postoperative investigations were unremarkable. The patient’s clinical condition improved completely and she was discharged from the hospital on the seventh postoperative day. The patient was doing well when she was last contacted nearly six months after her surgery. The patient provided written informed consent for publication of the anonymised data during her first follow-up. Discussion An unknown dose of misoprostol followed by dilatation and curettage in a previously scarred uterus could have led to uterine rupture in the current patient. Uterine rupture is primarily a clinical diagnosis, and hence, prompt surgical management is critical. Risk factors for uterine rupture include a previous uterine scar, short interpregnancy interval, multiparity, uterotonic drugs and obstructed labour.5 Additionally, the incidence of uterine rupture has been noted to be higher in scarred compared to unscarred uteri (0.28% versus 0.04%).1 A patient’s survival after uterine rupture depends on the time interval between the rupture and the intervention and urgent referral to a tertiary care centre. Rent repair is the key to treatment and internal iliac artery ligation is a lifesaving procedure in cases of uncontrolled obstetric haemorrhage. However, hysterectomy should not be delayed if the bleeding is intractable or the uterine rupture is irreparable. Second trimester pregnancy termination using misoprostol in an appropriate dosage in supervised settings for women with cesarean scar is safe and it has been associated with uterine rupture in only 0.3–0.4% of cases.1,6,7 A retrospective report did not observe uterine rupture after misoprostol induction for the termination of second trimester pregnancy in a scarred uterus.8 In the current patient, a characteristic bucket handle type uterine rupture was noted as was also reported by Abubekar et al.9 An unyielding cervix with misoprostol-induced strong uterine contractions in a scarred uterus may predispose to this type of uterine rupture.10 In addition, unsupervised vigorous dilatation and curettage could have either led to iatrogenic uterine perforation or further aggravated the tear induced by misoprostol. Conclusion Unsupervised over-the-counter use of misoprostol in a scarred uterus to induce second trimester abortion can have catastrophic consequences. Dilatation and curettage should be avoided before ruling out uterine rupture. An early referral to a higher centre may prevent severe maternal morbidity and mortality. Figure 1 Photographs of the uterus of a 24-year-old female patient following over-the-counter use of misoprostol for abortion induction. A: Anterior view of the uterus circumferentially avulsed at the level of the internal os. B: Lateral view showing the avulsed uterus. C: Avulsed uterus suspended only through round ligaments. AUTHORS’ CONTRIBUTION NJ, HS, JS and AKJ conceptualised and designed the work as well as extracted the data. NJ, JS and AKJ interpreted the results. JS performed the statistical analysis. All authors contributed to the drafting and approved the final version of the manuscript.

Recovered

ReactionOutcome

CC BY-ND

34888091

2021-11

What was the outcome of reaction 'Shock haemorrhagic'?

A Rare Type of Uterine Rupture Following Over-the-Counter Use of Misoprostol in Second Trimester Abortion. The use of misoprostol in the second trimester by a woman with a uterine scar may lead to severe contractions and uterine rupture. We report a 24-year-old pregnant female patient who presented at the Emergency Department at a tertiary care hospital in Puducherry, India, in 2020 with haemorrhagic shock. She was at 16 weeks of gestation and had taken over the counter misoprostol for inducing an abortion. A quick initial resuscitation and urgent laparotomy were performed. An irreparable circumferentially avulsed uterus suspended only by round ligaments was noted. Haemostasis required internal artery ligation and immediate total hysterectomy. The patient was doing well upon follow-up six months after the surgery. Proper and supervised use of misoprostol in the appropriate dosage can avoid life-threatening consequences of uterine rupture. pmcTermination of pregnancy in the first and second trimesters in patients with prior uterine scarring is challenging.1 Misoprostol is a commonly used drug for induction of abortion even in unsupervised clinical settings.2 However, its dosing regimen has remained controversial.3,4 Over-the-counter use of misoprostol has resulted in documented or undocumented incidents of varying grades of uterine rupture. Uterine rupture is a surgical emergency that requires immediate care to avoid life-threatening medical, surgical and psychological consequences.4 In the current case, a patient with a previous caesarean scar experienced a unique type of uterine rupture which developed after over-the-counter use of misoprostol to induce second trimester abortion. Case Report A 24-year-old G3P1L1A1 (pregnant woman who had had a full-term delivery and one abortion and also had one living child) presented at the Emergency Department of a tertiary care hospital in Puducherry, India, in 2020 with haemorrhagic shock. She had taken an unknown dosage of misoprostol (unsupervised) to induce abortion at the 16th week of gestation. She had undergone a caesarean delivery two years before the current pregnancy. She self-administered an unknown dosage of misoprostol through the vaginal route at four-hour intervals. She noticed expulsion of the products of conception after two hours of the last dose of misoprostol. A local physician performed dilatation and curettage to treat heavy vaginal bleeding one day prior to presentation at the hospital. Her vaginal bleeding did not subside and she could have developed iatrogenic uterine perforation after dilatation and curettage. At admission, she was drowsy, extremely pale and afebrile. Her vitals included a pulse rate of 130/min, systemic blood pressure of 90/60 mmHg and respiratory rate of 28 breaths per minute. The remaining cardiovascular, respiratory and central nervous system findings were unremarkable. Abdominal examination revealed a diffuse tenderness and a well retracted uterus deviated towards the right-side of the abdomen. During speculum examination, a continuous fresh blood trickling was noted along with cervical tears at the 2 o’clock and 6 o’clock positions extending until the posterior fornix. Furthermore, the cervix appeared uneffaced and os patulous. Both fornices were boggy and tender; a transverse rent was noted in the posterior fornix through which the posterior wall of the uterus was felt. A bedside pelvic ultrasound scan revealed a post-abortal uterus with an empty cavity and a moderate amount of free fluid in the abdomen. Resuscitation was performed under guided haemodynamic monitoring and an immediate exploratory laparotomy was planned under general anaesthesia. Intraoperatively, 500mL of haemoperitoneum was noted with large clots in the pelvic cavity. There was a full-length scar rupture with the rent extending posteriorly. The posterior aspect of the uterus was avulsed completely above the level of the internal os and it was suspended only by the round ligament bilaterally; it appeared like a bucket handle tear [Figure 1]. The patient remained haemodynamically unstable and circumferential rent repair did not seem possible. There was a broad ligament haematoma of 5 × 5 cm on the left-side. The left uterine artery appeared avulsed and could not be traced. Bladder integrity was normal and there was no evidence of haematuria. Total abdominal hysterectomy along with left internal artery ligation was performed to achieve haemostasis. The patient was successfully resuscitated and extubated inside the operation theatre. Her postoperative investigations were unremarkable. The patient’s clinical condition improved completely and she was discharged from the hospital on the seventh postoperative day. The patient was doing well when she was last contacted nearly six months after her surgery. The patient provided written informed consent for publication of the anonymised data during her first follow-up. Discussion An unknown dose of misoprostol followed by dilatation and curettage in a previously scarred uterus could have led to uterine rupture in the current patient. Uterine rupture is primarily a clinical diagnosis, and hence, prompt surgical management is critical. Risk factors for uterine rupture include a previous uterine scar, short interpregnancy interval, multiparity, uterotonic drugs and obstructed labour.5 Additionally, the incidence of uterine rupture has been noted to be higher in scarred compared to unscarred uteri (0.28% versus 0.04%).1 A patient’s survival after uterine rupture depends on the time interval between the rupture and the intervention and urgent referral to a tertiary care centre. Rent repair is the key to treatment and internal iliac artery ligation is a lifesaving procedure in cases of uncontrolled obstetric haemorrhage. However, hysterectomy should not be delayed if the bleeding is intractable or the uterine rupture is irreparable. Second trimester pregnancy termination using misoprostol in an appropriate dosage in supervised settings for women with cesarean scar is safe and it has been associated with uterine rupture in only 0.3–0.4% of cases.1,6,7 A retrospective report did not observe uterine rupture after misoprostol induction for the termination of second trimester pregnancy in a scarred uterus.8 In the current patient, a characteristic bucket handle type uterine rupture was noted as was also reported by Abubekar et al.9 An unyielding cervix with misoprostol-induced strong uterine contractions in a scarred uterus may predispose to this type of uterine rupture.10 In addition, unsupervised vigorous dilatation and curettage could have either led to iatrogenic uterine perforation or further aggravated the tear induced by misoprostol. Conclusion Unsupervised over-the-counter use of misoprostol in a scarred uterus to induce second trimester abortion can have catastrophic consequences. Dilatation and curettage should be avoided before ruling out uterine rupture. An early referral to a higher centre may prevent severe maternal morbidity and mortality. Figure 1 Photographs of the uterus of a 24-year-old female patient following over-the-counter use of misoprostol for abortion induction. A: Anterior view of the uterus circumferentially avulsed at the level of the internal os. B: Lateral view showing the avulsed uterus. C: Avulsed uterus suspended only through round ligaments. AUTHORS’ CONTRIBUTION NJ, HS, JS and AKJ conceptualised and designed the work as well as extracted the data. NJ, JS and AKJ interpreted the results. JS performed the statistical analysis. All authors contributed to the drafting and approved the final version of the manuscript.

Recovered

ReactionOutcome

CC BY-ND

34888091

2021-11

What was the outcome of reaction 'Uterine rupture'?

A Rare Type of Uterine Rupture Following Over-the-Counter Use of Misoprostol in Second Trimester Abortion. The use of misoprostol in the second trimester by a woman with a uterine scar may lead to severe contractions and uterine rupture. We report a 24-year-old pregnant female patient who presented at the Emergency Department at a tertiary care hospital in Puducherry, India, in 2020 with haemorrhagic shock. She was at 16 weeks of gestation and had taken over the counter misoprostol for inducing an abortion. A quick initial resuscitation and urgent laparotomy were performed. An irreparable circumferentially avulsed uterus suspended only by round ligaments was noted. Haemostasis required internal artery ligation and immediate total hysterectomy. The patient was doing well upon follow-up six months after the surgery. Proper and supervised use of misoprostol in the appropriate dosage can avoid life-threatening consequences of uterine rupture. pmcTermination of pregnancy in the first and second trimesters in patients with prior uterine scarring is challenging.1 Misoprostol is a commonly used drug for induction of abortion even in unsupervised clinical settings.2 However, its dosing regimen has remained controversial.3,4 Over-the-counter use of misoprostol has resulted in documented or undocumented incidents of varying grades of uterine rupture. Uterine rupture is a surgical emergency that requires immediate care to avoid life-threatening medical, surgical and psychological consequences.4 In the current case, a patient with a previous caesarean scar experienced a unique type of uterine rupture which developed after over-the-counter use of misoprostol to induce second trimester abortion. Case Report A 24-year-old G3P1L1A1 (pregnant woman who had had a full-term delivery and one abortion and also had one living child) presented at the Emergency Department of a tertiary care hospital in Puducherry, India, in 2020 with haemorrhagic shock. She had taken an unknown dosage of misoprostol (unsupervised) to induce abortion at the 16th week of gestation. She had undergone a caesarean delivery two years before the current pregnancy. She self-administered an unknown dosage of misoprostol through the vaginal route at four-hour intervals. She noticed expulsion of the products of conception after two hours of the last dose of misoprostol. A local physician performed dilatation and curettage to treat heavy vaginal bleeding one day prior to presentation at the hospital. Her vaginal bleeding did not subside and she could have developed iatrogenic uterine perforation after dilatation and curettage. At admission, she was drowsy, extremely pale and afebrile. Her vitals included a pulse rate of 130/min, systemic blood pressure of 90/60 mmHg and respiratory rate of 28 breaths per minute. The remaining cardiovascular, respiratory and central nervous system findings were unremarkable. Abdominal examination revealed a diffuse tenderness and a well retracted uterus deviated towards the right-side of the abdomen. During speculum examination, a continuous fresh blood trickling was noted along with cervical tears at the 2 o’clock and 6 o’clock positions extending until the posterior fornix. Furthermore, the cervix appeared uneffaced and os patulous. Both fornices were boggy and tender; a transverse rent was noted in the posterior fornix through which the posterior wall of the uterus was felt. A bedside pelvic ultrasound scan revealed a post-abortal uterus with an empty cavity and a moderate amount of free fluid in the abdomen. Resuscitation was performed under guided haemodynamic monitoring and an immediate exploratory laparotomy was planned under general anaesthesia. Intraoperatively, 500mL of haemoperitoneum was noted with large clots in the pelvic cavity. There was a full-length scar rupture with the rent extending posteriorly. The posterior aspect of the uterus was avulsed completely above the level of the internal os and it was suspended only by the round ligament bilaterally; it appeared like a bucket handle tear [Figure 1]. The patient remained haemodynamically unstable and circumferential rent repair did not seem possible. There was a broad ligament haematoma of 5 × 5 cm on the left-side. The left uterine artery appeared avulsed and could not be traced. Bladder integrity was normal and there was no evidence of haematuria. Total abdominal hysterectomy along with left internal artery ligation was performed to achieve haemostasis. The patient was successfully resuscitated and extubated inside the operation theatre. Her postoperative investigations were unremarkable. The patient’s clinical condition improved completely and she was discharged from the hospital on the seventh postoperative day. The patient was doing well when she was last contacted nearly six months after her surgery. The patient provided written informed consent for publication of the anonymised data during her first follow-up. Discussion An unknown dose of misoprostol followed by dilatation and curettage in a previously scarred uterus could have led to uterine rupture in the current patient. Uterine rupture is primarily a clinical diagnosis, and hence, prompt surgical management is critical. Risk factors for uterine rupture include a previous uterine scar, short interpregnancy interval, multiparity, uterotonic drugs and obstructed labour.5 Additionally, the incidence of uterine rupture has been noted to be higher in scarred compared to unscarred uteri (0.28% versus 0.04%).1 A patient’s survival after uterine rupture depends on the time interval between the rupture and the intervention and urgent referral to a tertiary care centre. Rent repair is the key to treatment and internal iliac artery ligation is a lifesaving procedure in cases of uncontrolled obstetric haemorrhage. However, hysterectomy should not be delayed if the bleeding is intractable or the uterine rupture is irreparable. Second trimester pregnancy termination using misoprostol in an appropriate dosage in supervised settings for women with cesarean scar is safe and it has been associated with uterine rupture in only 0.3–0.4% of cases.1,6,7 A retrospective report did not observe uterine rupture after misoprostol induction for the termination of second trimester pregnancy in a scarred uterus.8 In the current patient, a characteristic bucket handle type uterine rupture was noted as was also reported by Abubekar et al.9 An unyielding cervix with misoprostol-induced strong uterine contractions in a scarred uterus may predispose to this type of uterine rupture.10 In addition, unsupervised vigorous dilatation and curettage could have either led to iatrogenic uterine perforation or further aggravated the tear induced by misoprostol. Conclusion Unsupervised over-the-counter use of misoprostol in a scarred uterus to induce second trimester abortion can have catastrophic consequences. Dilatation and curettage should be avoided before ruling out uterine rupture. An early referral to a higher centre may prevent severe maternal morbidity and mortality. Figure 1 Photographs of the uterus of a 24-year-old female patient following over-the-counter use of misoprostol for abortion induction. A: Anterior view of the uterus circumferentially avulsed at the level of the internal os. B: Lateral view showing the avulsed uterus. C: Avulsed uterus suspended only through round ligaments. AUTHORS’ CONTRIBUTION NJ, HS, JS and AKJ conceptualised and designed the work as well as extracted the data. NJ, JS and AKJ interpreted the results. JS performed the statistical analysis. All authors contributed to the drafting and approved the final version of the manuscript.

Recovered

ReactionOutcome

CC BY-ND

34888091

2021-11

Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Cavernous sinus thrombosis'.

Mucormycosis with extensive cranial nerve involvement as the first presentation of diabetes mellitus: A case report. Mucormycosis, a rare fungal infection, mainly affects individuals with diabetes mellitus and those who were immunocompromised and has a high mortality rate. Its most common presentation is similar to that of acute bacterial sinusitis with symptoms of nasal congestion, headache, and fever. The involvement of multiple cranial nerves in mucormycosis was rarely reported in the literature and indicates severe disease. Herein, we report the case of a 56-year-old man who was referred to the ophthalmology outpatient clinic for facial nerve palsy. He was treated with systemic steroids for 10 days with no improvement. On examination, he had a loss of vision and a frozen orbit due to involvement of cranial nerves II, III, IV, V, VI, and VII. An extensive workup revealed a hemoglobin A1C of 10%. However, he was never diagnosed with diabetes mellitus previously and denied any of the classical symptoms of diabetes mellitus. He underwent ethmoidectomy, maxillectomy, and drainage of an intraorbital abscess after appropriate imaging studies. Histopathology confirmed the diagnosis of mucormycosis, and the patient was started on systemic amphotericin B. This case emphasizes the importance of screening for diabetes mellitus. Early recognition of underlying diabetes mellitus in this patient may have prevented the development of mucormycosis along with its devastating complications. pmcIntroduction Mucormycosis is a rare opportunistic angio-invasive infection caused by fungi of the order Mucorales. It mainly affects individuals with diabetes mellitus and patients who have underlying immunosuppression, with a mortality rate reaching up to 80%. 1 According to reports from the United States, mucormycosis has a very low incidence of 1.7 in every million individuals. Diabetes mellitus is the most common risk factor for the development of mucormycosis across Asia, while hematological malignancies are the major risk factors across Western countries. 2 Although mucormycosis is a severe complication of poorly controlled diabetes mellitus, it is very rare for diabetes mellitus to be diagnosed after the development of mucormycosis. 3 Mucormycosis can involve many organ systems, including the cutaneous, pulmonary, and gastrointestinal systems, and in severe forms, it can spread to other organs. However, rhinocerebral involvement is the most frequent presentation. 4 Involvement of cranial nerve VII along with the widespread involvement of other cranial nerves was rarely reported in the literature. 5 Herein, we report a very unusual case of mucormycosis with extensive rhino-orbit cerebral involvement presenting with multiple cranial nerve palsies as a first presentation of diabetes mellitus in a previously healthy patient. Case Description A 56-year-old man was referred to our institution for a case of facial palsy. He was treated with systemic methylprednisolone for 10 days, without improvement. On examination, he had a visual acuity of 1.0 in the right eye and hand motion in the left. Motility examination revealed limitation of abduction of the right eye. The patient had a frozen orbit with a limitation of all gazes in the left eye. Fundus examination revealed a pale optic disk, afferent pupillary defect, and an ischemic retina, indicating central retinal artery occlusion. Examination findings were also consistent with oculomotor, trochlear, and abducens nerve paresis. Mucormycosis was a result of steroid therapy, as the patient was diagnosed 2 weeks previously with an isolated cranial nerve VII and was started with steroids in a peripheral ophthalmology clinic. At his first presentation in Jordan University Hospital, he had a frozen orbit with cranial nerve III, IV, and VI palsies. The patient was admitted, and the results of baseline investigations were normal, except for a high white blood cell count and hemoglobin A1C (HbA1c) of 10%. Systemic steroids were stopped, and the patient was started on insulin to correct his blood sugar. He was never diagnosed with diabetes mellitus before and denied any symptoms of diabetes mellitus. He was initially diagnosed with diabetes mellitus. He tested negative for ketone bodies and was treated in the ward. Results of other routine investigations, such as kidney function tests and complete blood count, were normal, except for high blood sugar and a high white blood cell count. The differential diagnosis of his current symptoms was cavernous sinus thrombosis, a space-occupying lesion in the cavernous sinus or orbital cellulitis. Computed tomography of the brain and orbit was performed and revealed extensive sinus disease with left complicated ethmoidal and maxillary sinusitis extending into the left orbit with abscess formation. There was also extensive orbital cellulitis and evidence of left cavernous sinus thrombosis secondary to infection (Figure 1). It is possible that the patient had undiagnosed diabetes mellitus as indicated by a high HbA1c and was worsened by steroid therapy The patient was started on systemic vancomycin and ceftriaxone, and a swab from the nasal cavity was sent for evaluation. Nasal swab revealed methicillin-resistant staphylococcus and positive fungal staining. The patient was started on systemic amphotericin B at a loading dose and continued with the therapeutic dose after 2 days of his presentation to our institution. Nasal endoscopy revealed diffuse white filaments and necrosis of the bone from which a biopsy was taken. Sinuses were drained, followed by ethmoidectomy and maxillectomy. A biopsy from the necrotic tissue was sent, which showed the presence of non-septate hyphae branching at right angles consistent with angio-invasive mucormycosis. However, workup to identify the exact causative organism of mucormycosis was not performed as it did not affect further management. The patient was kept on daily irrigation of the sinuses along with administration of systemic amphotericin B. On follow-up magnetic resonance imaging, the patient had an intraorbital collection (Figure 2). Subsequently, drainage of the orbital abscess, tarsorrhaphy, and excision of the necrotic conjunctiva were performed. The patient was kept on systemic amphotericin B and showed clinical improvement. He remained hospitalized until a repeat biopsy of the paranasal sinuses was negative for the fungi. He was discharged with residual paresis of all cranial nerves involved along with left eye ptosis and only hand motion in the left eye. He was followed up serially for 1 year and 4 months. Currently, the patient is afebrile with complete resolution of symptoms in the right eye and partial recovery of cranial nerve palsies. However, in the left eye, residual ptosis has remained and only hand motion is visible due to the previous central retinal artery occlusion related to his initial presentation. Discussion Mucormycosis is an opportunistic infection with a high mortality rate, reaching up to 80% without treatment. 6 Even with early diagnosis and aggressive therapy, its prognosis remains poor. 7 Diabetes mellitus is the most common risk factor for mucormycosis, particularly during ketoacidosis. Ketones induce the fungi to utilize and produce ketoreductase, which facilitates its growth through various mechanisms. 4 Roden et al. reviewed the characteristics of 929 patients with mucormycosis, and of 929 patients, 36% had diabetes mellitus, 17% had an underlying malignancy, and 19% had no underlying conditions. 3 In a review of literature of 851 patients, the median age was 51 ± 12 years, and 531 (63%) were men. Diabetes mellitus was the most common underlying condition (340/851, 40%; 71 (20%) had documented ketoacidosis), followed by hematological malignancy (275/851, 32%; 116 (42%) had acute myeloid leukemia) and solid organ transplantation (116/851, 14%; 67 (58%) had received a kidney transplant). Of the predisposing factors, the use of corticosteroids at the time of presentation was the most common (273/851, 33%), followed by neutropenia (169/851, 20%) and trauma (166/851, 20%). Rhino-orbito-cerebral mucormycosis was the most commonly observed manifestation (288/851, 34%), followed by cutaneous (187/851, 22%) and pulmonary mucormycosis (172/851, 20%. 8 In the present case, the initiation of glucocorticoids as a treatment for facial nerve palsy may have facilitated the initial development of mucormycosis. Glucocorticoids are well known to increase blood glucose levels and can exacerbate hyperglycemia in a patient with diabetes mellitus. 9 Our patient should have been tested for hyperglycemia before the initiation of glucocorticoid therapy (1) to rule out diabetes mellitus as the underlying cause of his facial palsy and (2) to measure the baseline blood sugar before the initiation of steroid therapy. 10 Clinically, mucormycosis is characterized by rhinitis with granular and purulent discharge, nasal ulceration, black spots of infarcted mucosa, and paranasal sinusitis. However, other common presenting symptoms include epistaxis, ophthalmoplegia with blindness, proptosis and orbital cellulitis, hemiplegia or stroke, and decreased mental function. 11 In advanced disease, symptoms include chemosis, ptosis, proptosis, ophthalmoplegia, blindness, and multiple cranial nerve palsies (function of cranial nerves II, III, V, VI, and VI may be lost or impaired).4,5 The classical clinical presentations are facial pain, an irregular black eschar on the palatal or nasal mucosa, and pus discharge from the eye and nose. In a retrospective chart review of 48 patients with 49 cases of acute fulminant-invasive fungal sinusitis over 19 years, mucormycosis and aspergillosis were found in 22 and 27 cases, respectively. Orbital (proptosis, periorbital edema, and ophthalmoplegia) and cranial nerve symptoms were seen at presentation more frequently in mucormycosis (6 [27%] and 9 [41%]) than in aspergillosis cases (3 [11%] and 7 [26%]; p = 0.079). Long-term orbital and cranial nerve sequelae were reported in 16 (72%) mucormycosis cases and 10 (37%) aspergillosis cases (p = 0.0210). These data suggest that the presence of orbital and neurological symptoms at presentation warrants more aggressive surgical intervention because of the likelihood of mucormycosis. 12 In another review of mucormycosis cases, all six patients had uncontrolled diabetes mellitus with fungal-invasive disease and had multiple cranial nerve involvement, including cranial nerves II to VII. 2 However, our patient had not been definitely diagnosed with diabetes mellitus before he had the fungal infection. That cohort chiefly complained of facial swelling, blood-stained nasal discharge, and crusting. All six patients were on prolonged antibiotic treatment with an incomplete response. Facial dysesthesia was present in all sides involved. Six patients had impaired vision, two had vision loss, and four only had a perception of light. These patients improved to finger counting during treatment. All patients had periorbital edema, ptosis, and 7th cranial nerve palsy of the lower motor type, and two patients had irritability and disoriented state and succumbed to the disease within 48 h of treatment. By contrast, our patient was fully oriented and had an initial visual acuity of hand motion with limited improvement. 13 A diagnosis of rhino-orbito-cerebral mucormycosis requires high clinical suspicion, and a definitive diagnosis requires a biopsy indicating the presence of the fungi. Furthermore, imaging is required to identify the degree of local involvement and spread across the rhino-orbital structures. 3 Mucormycosis is treated with surgical debridement of necrotic tissues, use of intravenous antifungal drugs, and elimination of predisposing factors. In retrospective studies, the combination of surgery and antifungal drugs showed higher survival rates than medical treatment alone. 3 Our case emphasizes the importance of screening for diabetes mellitus to prevent such a devastating complication. Our patient denied any of the classical symptoms of diabetes mellitus, and the only risk factor he had for diabetes mellitus was his age. Given the relatively long asymptomatic period of diabetes mellitus, the American Diabetes Association recommends screening for diabetes mellitus at age 45 years. 14 Early detection and treatment of this patient's underlying diabetes mellitus may have prevented the development of mucormycosis, especially in a country like Jordan where the prevalence of diabetes mellitus among men aged >25 years is as high as 33%. 15 The involvement of cranial nerve VII in mucormycosis is very rare, and its pathogenesis is not well understood. 13 We are unsure whether the involvement of cranial nerve VII in this patient was the first manifestation of mucormycosis or it was due to an independent pathology in which its treatment with glucocorticoids facilitated the development of mucormycosis. Furthermore, the widespread involvement of cranial nerves III, IV, and VI is also uncommon and indicates severe disease possibly due to cavernous sinus thrombosis. 3 Conclusions Mucormycosis is a rare and possibly fatal opportunistic infection. It should be suspected in any patient who presents with symptoms of sinusitis along with the widespread involvement of the orbit and cranial nerves. Workup should be initiated for an underlying systemic disease, such as diabetes mellitus, in any patient diagnosed with mucormycosis. In the present case, initiation of steroid therapy may have influenced the development of fungal infection and its progression, and this highlights the importance of screening. It is critical to exclude diabetes mellitus before starting steroid treatment and to screen a susceptible population perhaps at a younger age than recommended. Treatment with systemic antifungals, surgical debridement of necrotic tissues, and correction of underlying hyperglycemia should be initiated as soon as possible in diabetic patients with mucormycosis. Although this patient was 56 years old, screening for diabetes mellitus in younger populations earlier than 45 years is recommended, especially in the presence of opportunistic infections. Ethical statement Ethical approval was obtained from the Jordan University Board. Informed consent was also obtained from the patient. Conflict of interest All authors declare no conflict of interest Funding sources This work was not funded by any funding bodies, and all authors have read and approved the contents of this manuscript. Figure 1. Figure 2.

METHYLPREDNISOLONE

DrugsGivenReaction

CC BY

34888198

2021

Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Cellulitis orbital'.

Mucormycosis with extensive cranial nerve involvement as the first presentation of diabetes mellitus: A case report. Mucormycosis, a rare fungal infection, mainly affects individuals with diabetes mellitus and those who were immunocompromised and has a high mortality rate. Its most common presentation is similar to that of acute bacterial sinusitis with symptoms of nasal congestion, headache, and fever. The involvement of multiple cranial nerves in mucormycosis was rarely reported in the literature and indicates severe disease. Herein, we report the case of a 56-year-old man who was referred to the ophthalmology outpatient clinic for facial nerve palsy. He was treated with systemic steroids for 10 days with no improvement. On examination, he had a loss of vision and a frozen orbit due to involvement of cranial nerves II, III, IV, V, VI, and VII. An extensive workup revealed a hemoglobin A1C of 10%. However, he was never diagnosed with diabetes mellitus previously and denied any of the classical symptoms of diabetes mellitus. He underwent ethmoidectomy, maxillectomy, and drainage of an intraorbital abscess after appropriate imaging studies. Histopathology confirmed the diagnosis of mucormycosis, and the patient was started on systemic amphotericin B. This case emphasizes the importance of screening for diabetes mellitus. Early recognition of underlying diabetes mellitus in this patient may have prevented the development of mucormycosis along with its devastating complications. pmcIntroduction Mucormycosis is a rare opportunistic angio-invasive infection caused by fungi of the order Mucorales. It mainly affects individuals with diabetes mellitus and patients who have underlying immunosuppression, with a mortality rate reaching up to 80%. 1 According to reports from the United States, mucormycosis has a very low incidence of 1.7 in every million individuals. Diabetes mellitus is the most common risk factor for the development of mucormycosis across Asia, while hematological malignancies are the major risk factors across Western countries. 2 Although mucormycosis is a severe complication of poorly controlled diabetes mellitus, it is very rare for diabetes mellitus to be diagnosed after the development of mucormycosis. 3 Mucormycosis can involve many organ systems, including the cutaneous, pulmonary, and gastrointestinal systems, and in severe forms, it can spread to other organs. However, rhinocerebral involvement is the most frequent presentation. 4 Involvement of cranial nerve VII along with the widespread involvement of other cranial nerves was rarely reported in the literature. 5 Herein, we report a very unusual case of mucormycosis with extensive rhino-orbit cerebral involvement presenting with multiple cranial nerve palsies as a first presentation of diabetes mellitus in a previously healthy patient. Case Description A 56-year-old man was referred to our institution for a case of facial palsy. He was treated with systemic methylprednisolone for 10 days, without improvement. On examination, he had a visual acuity of 1.0 in the right eye and hand motion in the left. Motility examination revealed limitation of abduction of the right eye. The patient had a frozen orbit with a limitation of all gazes in the left eye. Fundus examination revealed a pale optic disk, afferent pupillary defect, and an ischemic retina, indicating central retinal artery occlusion. Examination findings were also consistent with oculomotor, trochlear, and abducens nerve paresis. Mucormycosis was a result of steroid therapy, as the patient was diagnosed 2 weeks previously with an isolated cranial nerve VII and was started with steroids in a peripheral ophthalmology clinic. At his first presentation in Jordan University Hospital, he had a frozen orbit with cranial nerve III, IV, and VI palsies. The patient was admitted, and the results of baseline investigations were normal, except for a high white blood cell count and hemoglobin A1C (HbA1c) of 10%. Systemic steroids were stopped, and the patient was started on insulin to correct his blood sugar. He was never diagnosed with diabetes mellitus before and denied any symptoms of diabetes mellitus. He was initially diagnosed with diabetes mellitus. He tested negative for ketone bodies and was treated in the ward. Results of other routine investigations, such as kidney function tests and complete blood count, were normal, except for high blood sugar and a high white blood cell count. The differential diagnosis of his current symptoms was cavernous sinus thrombosis, a space-occupying lesion in the cavernous sinus or orbital cellulitis. Computed tomography of the brain and orbit was performed and revealed extensive sinus disease with left complicated ethmoidal and maxillary sinusitis extending into the left orbit with abscess formation. There was also extensive orbital cellulitis and evidence of left cavernous sinus thrombosis secondary to infection (Figure 1). It is possible that the patient had undiagnosed diabetes mellitus as indicated by a high HbA1c and was worsened by steroid therapy The patient was started on systemic vancomycin and ceftriaxone, and a swab from the nasal cavity was sent for evaluation. Nasal swab revealed methicillin-resistant staphylococcus and positive fungal staining. The patient was started on systemic amphotericin B at a loading dose and continued with the therapeutic dose after 2 days of his presentation to our institution. Nasal endoscopy revealed diffuse white filaments and necrosis of the bone from which a biopsy was taken. Sinuses were drained, followed by ethmoidectomy and maxillectomy. A biopsy from the necrotic tissue was sent, which showed the presence of non-septate hyphae branching at right angles consistent with angio-invasive mucormycosis. However, workup to identify the exact causative organism of mucormycosis was not performed as it did not affect further management. The patient was kept on daily irrigation of the sinuses along with administration of systemic amphotericin B. On follow-up magnetic resonance imaging, the patient had an intraorbital collection (Figure 2). Subsequently, drainage of the orbital abscess, tarsorrhaphy, and excision of the necrotic conjunctiva were performed. The patient was kept on systemic amphotericin B and showed clinical improvement. He remained hospitalized until a repeat biopsy of the paranasal sinuses was negative for the fungi. He was discharged with residual paresis of all cranial nerves involved along with left eye ptosis and only hand motion in the left eye. He was followed up serially for 1 year and 4 months. Currently, the patient is afebrile with complete resolution of symptoms in the right eye and partial recovery of cranial nerve palsies. However, in the left eye, residual ptosis has remained and only hand motion is visible due to the previous central retinal artery occlusion related to his initial presentation. Discussion Mucormycosis is an opportunistic infection with a high mortality rate, reaching up to 80% without treatment. 6 Even with early diagnosis and aggressive therapy, its prognosis remains poor. 7 Diabetes mellitus is the most common risk factor for mucormycosis, particularly during ketoacidosis. Ketones induce the fungi to utilize and produce ketoreductase, which facilitates its growth through various mechanisms. 4 Roden et al. reviewed the characteristics of 929 patients with mucormycosis, and of 929 patients, 36% had diabetes mellitus, 17% had an underlying malignancy, and 19% had no underlying conditions. 3 In a review of literature of 851 patients, the median age was 51 ± 12 years, and 531 (63%) were men. Diabetes mellitus was the most common underlying condition (340/851, 40%; 71 (20%) had documented ketoacidosis), followed by hematological malignancy (275/851, 32%; 116 (42%) had acute myeloid leukemia) and solid organ transplantation (116/851, 14%; 67 (58%) had received a kidney transplant). Of the predisposing factors, the use of corticosteroids at the time of presentation was the most common (273/851, 33%), followed by neutropenia (169/851, 20%) and trauma (166/851, 20%). Rhino-orbito-cerebral mucormycosis was the most commonly observed manifestation (288/851, 34%), followed by cutaneous (187/851, 22%) and pulmonary mucormycosis (172/851, 20%. 8 In the present case, the initiation of glucocorticoids as a treatment for facial nerve palsy may have facilitated the initial development of mucormycosis. Glucocorticoids are well known to increase blood glucose levels and can exacerbate hyperglycemia in a patient with diabetes mellitus. 9 Our patient should have been tested for hyperglycemia before the initiation of glucocorticoid therapy (1) to rule out diabetes mellitus as the underlying cause of his facial palsy and (2) to measure the baseline blood sugar before the initiation of steroid therapy. 10 Clinically, mucormycosis is characterized by rhinitis with granular and purulent discharge, nasal ulceration, black spots of infarcted mucosa, and paranasal sinusitis. However, other common presenting symptoms include epistaxis, ophthalmoplegia with blindness, proptosis and orbital cellulitis, hemiplegia or stroke, and decreased mental function. 11 In advanced disease, symptoms include chemosis, ptosis, proptosis, ophthalmoplegia, blindness, and multiple cranial nerve palsies (function of cranial nerves II, III, V, VI, and VI may be lost or impaired).4,5 The classical clinical presentations are facial pain, an irregular black eschar on the palatal or nasal mucosa, and pus discharge from the eye and nose. In a retrospective chart review of 48 patients with 49 cases of acute fulminant-invasive fungal sinusitis over 19 years, mucormycosis and aspergillosis were found in 22 and 27 cases, respectively. Orbital (proptosis, periorbital edema, and ophthalmoplegia) and cranial nerve symptoms were seen at presentation more frequently in mucormycosis (6 [27%] and 9 [41%]) than in aspergillosis cases (3 [11%] and 7 [26%]; p = 0.079). Long-term orbital and cranial nerve sequelae were reported in 16 (72%) mucormycosis cases and 10 (37%) aspergillosis cases (p = 0.0210). These data suggest that the presence of orbital and neurological symptoms at presentation warrants more aggressive surgical intervention because of the likelihood of mucormycosis. 12 In another review of mucormycosis cases, all six patients had uncontrolled diabetes mellitus with fungal-invasive disease and had multiple cranial nerve involvement, including cranial nerves II to VII. 2 However, our patient had not been definitely diagnosed with diabetes mellitus before he had the fungal infection. That cohort chiefly complained of facial swelling, blood-stained nasal discharge, and crusting. All six patients were on prolonged antibiotic treatment with an incomplete response. Facial dysesthesia was present in all sides involved. Six patients had impaired vision, two had vision loss, and four only had a perception of light. These patients improved to finger counting during treatment. All patients had periorbital edema, ptosis, and 7th cranial nerve palsy of the lower motor type, and two patients had irritability and disoriented state and succumbed to the disease within 48 h of treatment. By contrast, our patient was fully oriented and had an initial visual acuity of hand motion with limited improvement. 13 A diagnosis of rhino-orbito-cerebral mucormycosis requires high clinical suspicion, and a definitive diagnosis requires a biopsy indicating the presence of the fungi. Furthermore, imaging is required to identify the degree of local involvement and spread across the rhino-orbital structures. 3 Mucormycosis is treated with surgical debridement of necrotic tissues, use of intravenous antifungal drugs, and elimination of predisposing factors. In retrospective studies, the combination of surgery and antifungal drugs showed higher survival rates than medical treatment alone. 3 Our case emphasizes the importance of screening for diabetes mellitus to prevent such a devastating complication. Our patient denied any of the classical symptoms of diabetes mellitus, and the only risk factor he had for diabetes mellitus was his age. Given the relatively long asymptomatic period of diabetes mellitus, the American Diabetes Association recommends screening for diabetes mellitus at age 45 years. 14 Early detection and treatment of this patient's underlying diabetes mellitus may have prevented the development of mucormycosis, especially in a country like Jordan where the prevalence of diabetes mellitus among men aged >25 years is as high as 33%. 15 The involvement of cranial nerve VII in mucormycosis is very rare, and its pathogenesis is not well understood. 13 We are unsure whether the involvement of cranial nerve VII in this patient was the first manifestation of mucormycosis or it was due to an independent pathology in which its treatment with glucocorticoids facilitated the development of mucormycosis. Furthermore, the widespread involvement of cranial nerves III, IV, and VI is also uncommon and indicates severe disease possibly due to cavernous sinus thrombosis. 3 Conclusions Mucormycosis is a rare and possibly fatal opportunistic infection. It should be suspected in any patient who presents with symptoms of sinusitis along with the widespread involvement of the orbit and cranial nerves. Workup should be initiated for an underlying systemic disease, such as diabetes mellitus, in any patient diagnosed with mucormycosis. In the present case, initiation of steroid therapy may have influenced the development of fungal infection and its progression, and this highlights the importance of screening. It is critical to exclude diabetes mellitus before starting steroid treatment and to screen a susceptible population perhaps at a younger age than recommended. Treatment with systemic antifungals, surgical debridement of necrotic tissues, and correction of underlying hyperglycemia should be initiated as soon as possible in diabetic patients with mucormycosis. Although this patient was 56 years old, screening for diabetes mellitus in younger populations earlier than 45 years is recommended, especially in the presence of opportunistic infections. Ethical statement Ethical approval was obtained from the Jordan University Board. Informed consent was also obtained from the patient. Conflict of interest All authors declare no conflict of interest Funding sources This work was not funded by any funding bodies, and all authors have read and approved the contents of this manuscript. Figure 1. Figure 2.

METHYLPREDNISOLONE

DrugsGivenReaction

CC BY

34888198

2021

Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Diabetes mellitus'.

Mucormycosis with extensive cranial nerve involvement as the first presentation of diabetes mellitus: A case report. Mucormycosis, a rare fungal infection, mainly affects individuals with diabetes mellitus and those who were immunocompromised and has a high mortality rate. Its most common presentation is similar to that of acute bacterial sinusitis with symptoms of nasal congestion, headache, and fever. The involvement of multiple cranial nerves in mucormycosis was rarely reported in the literature and indicates severe disease. Herein, we report the case of a 56-year-old man who was referred to the ophthalmology outpatient clinic for facial nerve palsy. He was treated with systemic steroids for 10 days with no improvement. On examination, he had a loss of vision and a frozen orbit due to involvement of cranial nerves II, III, IV, V, VI, and VII. An extensive workup revealed a hemoglobin A1C of 10%. However, he was never diagnosed with diabetes mellitus previously and denied any of the classical symptoms of diabetes mellitus. He underwent ethmoidectomy, maxillectomy, and drainage of an intraorbital abscess after appropriate imaging studies. Histopathology confirmed the diagnosis of mucormycosis, and the patient was started on systemic amphotericin B. This case emphasizes the importance of screening for diabetes mellitus. Early recognition of underlying diabetes mellitus in this patient may have prevented the development of mucormycosis along with its devastating complications. pmcIntroduction Mucormycosis is a rare opportunistic angio-invasive infection caused by fungi of the order Mucorales. It mainly affects individuals with diabetes mellitus and patients who have underlying immunosuppression, with a mortality rate reaching up to 80%. 1 According to reports from the United States, mucormycosis has a very low incidence of 1.7 in every million individuals. Diabetes mellitus is the most common risk factor for the development of mucormycosis across Asia, while hematological malignancies are the major risk factors across Western countries. 2 Although mucormycosis is a severe complication of poorly controlled diabetes mellitus, it is very rare for diabetes mellitus to be diagnosed after the development of mucormycosis. 3 Mucormycosis can involve many organ systems, including the cutaneous, pulmonary, and gastrointestinal systems, and in severe forms, it can spread to other organs. However, rhinocerebral involvement is the most frequent presentation. 4 Involvement of cranial nerve VII along with the widespread involvement of other cranial nerves was rarely reported in the literature. 5 Herein, we report a very unusual case of mucormycosis with extensive rhino-orbit cerebral involvement presenting with multiple cranial nerve palsies as a first presentation of diabetes mellitus in a previously healthy patient. Case Description A 56-year-old man was referred to our institution for a case of facial palsy. He was treated with systemic methylprednisolone for 10 days, without improvement. On examination, he had a visual acuity of 1.0 in the right eye and hand motion in the left. Motility examination revealed limitation of abduction of the right eye. The patient had a frozen orbit with a limitation of all gazes in the left eye. Fundus examination revealed a pale optic disk, afferent pupillary defect, and an ischemic retina, indicating central retinal artery occlusion. Examination findings were also consistent with oculomotor, trochlear, and abducens nerve paresis. Mucormycosis was a result of steroid therapy, as the patient was diagnosed 2 weeks previously with an isolated cranial nerve VII and was started with steroids in a peripheral ophthalmology clinic. At his first presentation in Jordan University Hospital, he had a frozen orbit with cranial nerve III, IV, and VI palsies. The patient was admitted, and the results of baseline investigations were normal, except for a high white blood cell count and hemoglobin A1C (HbA1c) of 10%. Systemic steroids were stopped, and the patient was started on insulin to correct his blood sugar. He was never diagnosed with diabetes mellitus before and denied any symptoms of diabetes mellitus. He was initially diagnosed with diabetes mellitus. He tested negative for ketone bodies and was treated in the ward. Results of other routine investigations, such as kidney function tests and complete blood count, were normal, except for high blood sugar and a high white blood cell count. The differential diagnosis of his current symptoms was cavernous sinus thrombosis, a space-occupying lesion in the cavernous sinus or orbital cellulitis. Computed tomography of the brain and orbit was performed and revealed extensive sinus disease with left complicated ethmoidal and maxillary sinusitis extending into the left orbit with abscess formation. There was also extensive orbital cellulitis and evidence of left cavernous sinus thrombosis secondary to infection (Figure 1). It is possible that the patient had undiagnosed diabetes mellitus as indicated by a high HbA1c and was worsened by steroid therapy The patient was started on systemic vancomycin and ceftriaxone, and a swab from the nasal cavity was sent for evaluation. Nasal swab revealed methicillin-resistant staphylococcus and positive fungal staining. The patient was started on systemic amphotericin B at a loading dose and continued with the therapeutic dose after 2 days of his presentation to our institution. Nasal endoscopy revealed diffuse white filaments and necrosis of the bone from which a biopsy was taken. Sinuses were drained, followed by ethmoidectomy and maxillectomy. A biopsy from the necrotic tissue was sent, which showed the presence of non-septate hyphae branching at right angles consistent with angio-invasive mucormycosis. However, workup to identify the exact causative organism of mucormycosis was not performed as it did not affect further management. The patient was kept on daily irrigation of the sinuses along with administration of systemic amphotericin B. On follow-up magnetic resonance imaging, the patient had an intraorbital collection (Figure 2). Subsequently, drainage of the orbital abscess, tarsorrhaphy, and excision of the necrotic conjunctiva were performed. The patient was kept on systemic amphotericin B and showed clinical improvement. He remained hospitalized until a repeat biopsy of the paranasal sinuses was negative for the fungi. He was discharged with residual paresis of all cranial nerves involved along with left eye ptosis and only hand motion in the left eye. He was followed up serially for 1 year and 4 months. Currently, the patient is afebrile with complete resolution of symptoms in the right eye and partial recovery of cranial nerve palsies. However, in the left eye, residual ptosis has remained and only hand motion is visible due to the previous central retinal artery occlusion related to his initial presentation. Discussion Mucormycosis is an opportunistic infection with a high mortality rate, reaching up to 80% without treatment. 6 Even with early diagnosis and aggressive therapy, its prognosis remains poor. 7 Diabetes mellitus is the most common risk factor for mucormycosis, particularly during ketoacidosis. Ketones induce the fungi to utilize and produce ketoreductase, which facilitates its growth through various mechanisms. 4 Roden et al. reviewed the characteristics of 929 patients with mucormycosis, and of 929 patients, 36% had diabetes mellitus, 17% had an underlying malignancy, and 19% had no underlying conditions. 3 In a review of literature of 851 patients, the median age was 51 ± 12 years, and 531 (63%) were men. Diabetes mellitus was the most common underlying condition (340/851, 40%; 71 (20%) had documented ketoacidosis), followed by hematological malignancy (275/851, 32%; 116 (42%) had acute myeloid leukemia) and solid organ transplantation (116/851, 14%; 67 (58%) had received a kidney transplant). Of the predisposing factors, the use of corticosteroids at the time of presentation was the most common (273/851, 33%), followed by neutropenia (169/851, 20%) and trauma (166/851, 20%). Rhino-orbito-cerebral mucormycosis was the most commonly observed manifestation (288/851, 34%), followed by cutaneous (187/851, 22%) and pulmonary mucormycosis (172/851, 20%. 8 In the present case, the initiation of glucocorticoids as a treatment for facial nerve palsy may have facilitated the initial development of mucormycosis. Glucocorticoids are well known to increase blood glucose levels and can exacerbate hyperglycemia in a patient with diabetes mellitus. 9 Our patient should have been tested for hyperglycemia before the initiation of glucocorticoid therapy (1) to rule out diabetes mellitus as the underlying cause of his facial palsy and (2) to measure the baseline blood sugar before the initiation of steroid therapy. 10 Clinically, mucormycosis is characterized by rhinitis with granular and purulent discharge, nasal ulceration, black spots of infarcted mucosa, and paranasal sinusitis. However, other common presenting symptoms include epistaxis, ophthalmoplegia with blindness, proptosis and orbital cellulitis, hemiplegia or stroke, and decreased mental function. 11 In advanced disease, symptoms include chemosis, ptosis, proptosis, ophthalmoplegia, blindness, and multiple cranial nerve palsies (function of cranial nerves II, III, V, VI, and VI may be lost or impaired).4,5 The classical clinical presentations are facial pain, an irregular black eschar on the palatal or nasal mucosa, and pus discharge from the eye and nose. In a retrospective chart review of 48 patients with 49 cases of acute fulminant-invasive fungal sinusitis over 19 years, mucormycosis and aspergillosis were found in 22 and 27 cases, respectively. Orbital (proptosis, periorbital edema, and ophthalmoplegia) and cranial nerve symptoms were seen at presentation more frequently in mucormycosis (6 [27%] and 9 [41%]) than in aspergillosis cases (3 [11%] and 7 [26%]; p = 0.079). Long-term orbital and cranial nerve sequelae were reported in 16 (72%) mucormycosis cases and 10 (37%) aspergillosis cases (p = 0.0210). These data suggest that the presence of orbital and neurological symptoms at presentation warrants more aggressive surgical intervention because of the likelihood of mucormycosis. 12 In another review of mucormycosis cases, all six patients had uncontrolled diabetes mellitus with fungal-invasive disease and had multiple cranial nerve involvement, including cranial nerves II to VII. 2 However, our patient had not been definitely diagnosed with diabetes mellitus before he had the fungal infection. That cohort chiefly complained of facial swelling, blood-stained nasal discharge, and crusting. All six patients were on prolonged antibiotic treatment with an incomplete response. Facial dysesthesia was present in all sides involved. Six patients had impaired vision, two had vision loss, and four only had a perception of light. These patients improved to finger counting during treatment. All patients had periorbital edema, ptosis, and 7th cranial nerve palsy of the lower motor type, and two patients had irritability and disoriented state and succumbed to the disease within 48 h of treatment. By contrast, our patient was fully oriented and had an initial visual acuity of hand motion with limited improvement. 13 A diagnosis of rhino-orbito-cerebral mucormycosis requires high clinical suspicion, and a definitive diagnosis requires a biopsy indicating the presence of the fungi. Furthermore, imaging is required to identify the degree of local involvement and spread across the rhino-orbital structures. 3 Mucormycosis is treated with surgical debridement of necrotic tissues, use of intravenous antifungal drugs, and elimination of predisposing factors. In retrospective studies, the combination of surgery and antifungal drugs showed higher survival rates than medical treatment alone. 3 Our case emphasizes the importance of screening for diabetes mellitus to prevent such a devastating complication. Our patient denied any of the classical symptoms of diabetes mellitus, and the only risk factor he had for diabetes mellitus was his age. Given the relatively long asymptomatic period of diabetes mellitus, the American Diabetes Association recommends screening for diabetes mellitus at age 45 years. 14 Early detection and treatment of this patient's underlying diabetes mellitus may have prevented the development of mucormycosis, especially in a country like Jordan where the prevalence of diabetes mellitus among men aged >25 years is as high as 33%. 15 The involvement of cranial nerve VII in mucormycosis is very rare, and its pathogenesis is not well understood. 13 We are unsure whether the involvement of cranial nerve VII in this patient was the first manifestation of mucormycosis or it was due to an independent pathology in which its treatment with glucocorticoids facilitated the development of mucormycosis. Furthermore, the widespread involvement of cranial nerves III, IV, and VI is also uncommon and indicates severe disease possibly due to cavernous sinus thrombosis. 3 Conclusions Mucormycosis is a rare and possibly fatal opportunistic infection. It should be suspected in any patient who presents with symptoms of sinusitis along with the widespread involvement of the orbit and cranial nerves. Workup should be initiated for an underlying systemic disease, such as diabetes mellitus, in any patient diagnosed with mucormycosis. In the present case, initiation of steroid therapy may have influenced the development of fungal infection and its progression, and this highlights the importance of screening. It is critical to exclude diabetes mellitus before starting steroid treatment and to screen a susceptible population perhaps at a younger age than recommended. Treatment with systemic antifungals, surgical debridement of necrotic tissues, and correction of underlying hyperglycemia should be initiated as soon as possible in diabetic patients with mucormycosis. Although this patient was 56 years old, screening for diabetes mellitus in younger populations earlier than 45 years is recommended, especially in the presence of opportunistic infections. Ethical statement Ethical approval was obtained from the Jordan University Board. Informed consent was also obtained from the patient. Conflict of interest All authors declare no conflict of interest Funding sources This work was not funded by any funding bodies, and all authors have read and approved the contents of this manuscript. Figure 1. Figure 2.

METHYLPREDNISOLONE

DrugsGivenReaction

CC BY

34888198

2021

Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Drug ineffective for unapproved indication'.

Mucormycosis with extensive cranial nerve involvement as the first presentation of diabetes mellitus: A case report. Mucormycosis, a rare fungal infection, mainly affects individuals with diabetes mellitus and those who were immunocompromised and has a high mortality rate. Its most common presentation is similar to that of acute bacterial sinusitis with symptoms of nasal congestion, headache, and fever. The involvement of multiple cranial nerves in mucormycosis was rarely reported in the literature and indicates severe disease. Herein, we report the case of a 56-year-old man who was referred to the ophthalmology outpatient clinic for facial nerve palsy. He was treated with systemic steroids for 10 days with no improvement. On examination, he had a loss of vision and a frozen orbit due to involvement of cranial nerves II, III, IV, V, VI, and VII. An extensive workup revealed a hemoglobin A1C of 10%. However, he was never diagnosed with diabetes mellitus previously and denied any of the classical symptoms of diabetes mellitus. He underwent ethmoidectomy, maxillectomy, and drainage of an intraorbital abscess after appropriate imaging studies. Histopathology confirmed the diagnosis of mucormycosis, and the patient was started on systemic amphotericin B. This case emphasizes the importance of screening for diabetes mellitus. Early recognition of underlying diabetes mellitus in this patient may have prevented the development of mucormycosis along with its devastating complications. pmcIntroduction Mucormycosis is a rare opportunistic angio-invasive infection caused by fungi of the order Mucorales. It mainly affects individuals with diabetes mellitus and patients who have underlying immunosuppression, with a mortality rate reaching up to 80%. 1 According to reports from the United States, mucormycosis has a very low incidence of 1.7 in every million individuals. Diabetes mellitus is the most common risk factor for the development of mucormycosis across Asia, while hematological malignancies are the major risk factors across Western countries. 2 Although mucormycosis is a severe complication of poorly controlled diabetes mellitus, it is very rare for diabetes mellitus to be diagnosed after the development of mucormycosis. 3 Mucormycosis can involve many organ systems, including the cutaneous, pulmonary, and gastrointestinal systems, and in severe forms, it can spread to other organs. However, rhinocerebral involvement is the most frequent presentation. 4 Involvement of cranial nerve VII along with the widespread involvement of other cranial nerves was rarely reported in the literature. 5 Herein, we report a very unusual case of mucormycosis with extensive rhino-orbit cerebral involvement presenting with multiple cranial nerve palsies as a first presentation of diabetes mellitus in a previously healthy patient. Case Description A 56-year-old man was referred to our institution for a case of facial palsy. He was treated with systemic methylprednisolone for 10 days, without improvement. On examination, he had a visual acuity of 1.0 in the right eye and hand motion in the left. Motility examination revealed limitation of abduction of the right eye. The patient had a frozen orbit with a limitation of all gazes in the left eye. Fundus examination revealed a pale optic disk, afferent pupillary defect, and an ischemic retina, indicating central retinal artery occlusion. Examination findings were also consistent with oculomotor, trochlear, and abducens nerve paresis. Mucormycosis was a result of steroid therapy, as the patient was diagnosed 2 weeks previously with an isolated cranial nerve VII and was started with steroids in a peripheral ophthalmology clinic. At his first presentation in Jordan University Hospital, he had a frozen orbit with cranial nerve III, IV, and VI palsies. The patient was admitted, and the results of baseline investigations were normal, except for a high white blood cell count and hemoglobin A1C (HbA1c) of 10%. Systemic steroids were stopped, and the patient was started on insulin to correct his blood sugar. He was never diagnosed with diabetes mellitus before and denied any symptoms of diabetes mellitus. He was initially diagnosed with diabetes mellitus. He tested negative for ketone bodies and was treated in the ward. Results of other routine investigations, such as kidney function tests and complete blood count, were normal, except for high blood sugar and a high white blood cell count. The differential diagnosis of his current symptoms was cavernous sinus thrombosis, a space-occupying lesion in the cavernous sinus or orbital cellulitis. Computed tomography of the brain and orbit was performed and revealed extensive sinus disease with left complicated ethmoidal and maxillary sinusitis extending into the left orbit with abscess formation. There was also extensive orbital cellulitis and evidence of left cavernous sinus thrombosis secondary to infection (Figure 1). It is possible that the patient had undiagnosed diabetes mellitus as indicated by a high HbA1c and was worsened by steroid therapy The patient was started on systemic vancomycin and ceftriaxone, and a swab from the nasal cavity was sent for evaluation. Nasal swab revealed methicillin-resistant staphylococcus and positive fungal staining. The patient was started on systemic amphotericin B at a loading dose and continued with the therapeutic dose after 2 days of his presentation to our institution. Nasal endoscopy revealed diffuse white filaments and necrosis of the bone from which a biopsy was taken. Sinuses were drained, followed by ethmoidectomy and maxillectomy. A biopsy from the necrotic tissue was sent, which showed the presence of non-septate hyphae branching at right angles consistent with angio-invasive mucormycosis. However, workup to identify the exact causative organism of mucormycosis was not performed as it did not affect further management. The patient was kept on daily irrigation of the sinuses along with administration of systemic amphotericin B. On follow-up magnetic resonance imaging, the patient had an intraorbital collection (Figure 2). Subsequently, drainage of the orbital abscess, tarsorrhaphy, and excision of the necrotic conjunctiva were performed. The patient was kept on systemic amphotericin B and showed clinical improvement. He remained hospitalized until a repeat biopsy of the paranasal sinuses was negative for the fungi. He was discharged with residual paresis of all cranial nerves involved along with left eye ptosis and only hand motion in the left eye. He was followed up serially for 1 year and 4 months. Currently, the patient is afebrile with complete resolution of symptoms in the right eye and partial recovery of cranial nerve palsies. However, in the left eye, residual ptosis has remained and only hand motion is visible due to the previous central retinal artery occlusion related to his initial presentation. Discussion Mucormycosis is an opportunistic infection with a high mortality rate, reaching up to 80% without treatment. 6 Even with early diagnosis and aggressive therapy, its prognosis remains poor. 7 Diabetes mellitus is the most common risk factor for mucormycosis, particularly during ketoacidosis. Ketones induce the fungi to utilize and produce ketoreductase, which facilitates its growth through various mechanisms. 4 Roden et al. reviewed the characteristics of 929 patients with mucormycosis, and of 929 patients, 36% had diabetes mellitus, 17% had an underlying malignancy, and 19% had no underlying conditions. 3 In a review of literature of 851 patients, the median age was 51 ± 12 years, and 531 (63%) were men. Diabetes mellitus was the most common underlying condition (340/851, 40%; 71 (20%) had documented ketoacidosis), followed by hematological malignancy (275/851, 32%; 116 (42%) had acute myeloid leukemia) and solid organ transplantation (116/851, 14%; 67 (58%) had received a kidney transplant). Of the predisposing factors, the use of corticosteroids at the time of presentation was the most common (273/851, 33%), followed by neutropenia (169/851, 20%) and trauma (166/851, 20%). Rhino-orbito-cerebral mucormycosis was the most commonly observed manifestation (288/851, 34%), followed by cutaneous (187/851, 22%) and pulmonary mucormycosis (172/851, 20%. 8 In the present case, the initiation of glucocorticoids as a treatment for facial nerve palsy may have facilitated the initial development of mucormycosis. Glucocorticoids are well known to increase blood glucose levels and can exacerbate hyperglycemia in a patient with diabetes mellitus. 9 Our patient should have been tested for hyperglycemia before the initiation of glucocorticoid therapy (1) to rule out diabetes mellitus as the underlying cause of his facial palsy and (2) to measure the baseline blood sugar before the initiation of steroid therapy. 10 Clinically, mucormycosis is characterized by rhinitis with granular and purulent discharge, nasal ulceration, black spots of infarcted mucosa, and paranasal sinusitis. However, other common presenting symptoms include epistaxis, ophthalmoplegia with blindness, proptosis and orbital cellulitis, hemiplegia or stroke, and decreased mental function. 11 In advanced disease, symptoms include chemosis, ptosis, proptosis, ophthalmoplegia, blindness, and multiple cranial nerve palsies (function of cranial nerves II, III, V, VI, and VI may be lost or impaired).4,5 The classical clinical presentations are facial pain, an irregular black eschar on the palatal or nasal mucosa, and pus discharge from the eye and nose. In a retrospective chart review of 48 patients with 49 cases of acute fulminant-invasive fungal sinusitis over 19 years, mucormycosis and aspergillosis were found in 22 and 27 cases, respectively. Orbital (proptosis, periorbital edema, and ophthalmoplegia) and cranial nerve symptoms were seen at presentation more frequently in mucormycosis (6 [27%] and 9 [41%]) than in aspergillosis cases (3 [11%] and 7 [26%]; p = 0.079). Long-term orbital and cranial nerve sequelae were reported in 16 (72%) mucormycosis cases and 10 (37%) aspergillosis cases (p = 0.0210). These data suggest that the presence of orbital and neurological symptoms at presentation warrants more aggressive surgical intervention because of the likelihood of mucormycosis. 12 In another review of mucormycosis cases, all six patients had uncontrolled diabetes mellitus with fungal-invasive disease and had multiple cranial nerve involvement, including cranial nerves II to VII. 2 However, our patient had not been definitely diagnosed with diabetes mellitus before he had the fungal infection. That cohort chiefly complained of facial swelling, blood-stained nasal discharge, and crusting. All six patients were on prolonged antibiotic treatment with an incomplete response. Facial dysesthesia was present in all sides involved. Six patients had impaired vision, two had vision loss, and four only had a perception of light. These patients improved to finger counting during treatment. All patients had periorbital edema, ptosis, and 7th cranial nerve palsy of the lower motor type, and two patients had irritability and disoriented state and succumbed to the disease within 48 h of treatment. By contrast, our patient was fully oriented and had an initial visual acuity of hand motion with limited improvement. 13 A diagnosis of rhino-orbito-cerebral mucormycosis requires high clinical suspicion, and a definitive diagnosis requires a biopsy indicating the presence of the fungi. Furthermore, imaging is required to identify the degree of local involvement and spread across the rhino-orbital structures. 3 Mucormycosis is treated with surgical debridement of necrotic tissues, use of intravenous antifungal drugs, and elimination of predisposing factors. In retrospective studies, the combination of surgery and antifungal drugs showed higher survival rates than medical treatment alone. 3 Our case emphasizes the importance of screening for diabetes mellitus to prevent such a devastating complication. Our patient denied any of the classical symptoms of diabetes mellitus, and the only risk factor he had for diabetes mellitus was his age. Given the relatively long asymptomatic period of diabetes mellitus, the American Diabetes Association recommends screening for diabetes mellitus at age 45 years. 14 Early detection and treatment of this patient's underlying diabetes mellitus may have prevented the development of mucormycosis, especially in a country like Jordan where the prevalence of diabetes mellitus among men aged >25 years is as high as 33%. 15 The involvement of cranial nerve VII in mucormycosis is very rare, and its pathogenesis is not well understood. 13 We are unsure whether the involvement of cranial nerve VII in this patient was the first manifestation of mucormycosis or it was due to an independent pathology in which its treatment with glucocorticoids facilitated the development of mucormycosis. Furthermore, the widespread involvement of cranial nerves III, IV, and VI is also uncommon and indicates severe disease possibly due to cavernous sinus thrombosis. 3 Conclusions Mucormycosis is a rare and possibly fatal opportunistic infection. It should be suspected in any patient who presents with symptoms of sinusitis along with the widespread involvement of the orbit and cranial nerves. Workup should be initiated for an underlying systemic disease, such as diabetes mellitus, in any patient diagnosed with mucormycosis. In the present case, initiation of steroid therapy may have influenced the development of fungal infection and its progression, and this highlights the importance of screening. It is critical to exclude diabetes mellitus before starting steroid treatment and to screen a susceptible population perhaps at a younger age than recommended. Treatment with systemic antifungals, surgical debridement of necrotic tissues, and correction of underlying hyperglycemia should be initiated as soon as possible in diabetic patients with mucormycosis. Although this patient was 56 years old, screening for diabetes mellitus in younger populations earlier than 45 years is recommended, especially in the presence of opportunistic infections. Ethical statement Ethical approval was obtained from the Jordan University Board. Informed consent was also obtained from the patient. Conflict of interest All authors declare no conflict of interest Funding sources This work was not funded by any funding bodies, and all authors have read and approved the contents of this manuscript. Figure 1. Figure 2.

METHYLPREDNISOLONE

DrugsGivenReaction

CC BY

34888198

2021

Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Mucormycosis'.

Mucormycosis with extensive cranial nerve involvement as the first presentation of diabetes mellitus: A case report. Mucormycosis, a rare fungal infection, mainly affects individuals with diabetes mellitus and those who were immunocompromised and has a high mortality rate. Its most common presentation is similar to that of acute bacterial sinusitis with symptoms of nasal congestion, headache, and fever. The involvement of multiple cranial nerves in mucormycosis was rarely reported in the literature and indicates severe disease. Herein, we report the case of a 56-year-old man who was referred to the ophthalmology outpatient clinic for facial nerve palsy. He was treated with systemic steroids for 10 days with no improvement. On examination, he had a loss of vision and a frozen orbit due to involvement of cranial nerves II, III, IV, V, VI, and VII. An extensive workup revealed a hemoglobin A1C of 10%. However, he was never diagnosed with diabetes mellitus previously and denied any of the classical symptoms of diabetes mellitus. He underwent ethmoidectomy, maxillectomy, and drainage of an intraorbital abscess after appropriate imaging studies. Histopathology confirmed the diagnosis of mucormycosis, and the patient was started on systemic amphotericin B. This case emphasizes the importance of screening for diabetes mellitus. Early recognition of underlying diabetes mellitus in this patient may have prevented the development of mucormycosis along with its devastating complications. pmcIntroduction Mucormycosis is a rare opportunistic angio-invasive infection caused by fungi of the order Mucorales. It mainly affects individuals with diabetes mellitus and patients who have underlying immunosuppression, with a mortality rate reaching up to 80%. 1 According to reports from the United States, mucormycosis has a very low incidence of 1.7 in every million individuals. Diabetes mellitus is the most common risk factor for the development of mucormycosis across Asia, while hematological malignancies are the major risk factors across Western countries. 2 Although mucormycosis is a severe complication of poorly controlled diabetes mellitus, it is very rare for diabetes mellitus to be diagnosed after the development of mucormycosis. 3 Mucormycosis can involve many organ systems, including the cutaneous, pulmonary, and gastrointestinal systems, and in severe forms, it can spread to other organs. However, rhinocerebral involvement is the most frequent presentation. 4 Involvement of cranial nerve VII along with the widespread involvement of other cranial nerves was rarely reported in the literature. 5 Herein, we report a very unusual case of mucormycosis with extensive rhino-orbit cerebral involvement presenting with multiple cranial nerve palsies as a first presentation of diabetes mellitus in a previously healthy patient. Case Description A 56-year-old man was referred to our institution for a case of facial palsy. He was treated with systemic methylprednisolone for 10 days, without improvement. On examination, he had a visual acuity of 1.0 in the right eye and hand motion in the left. Motility examination revealed limitation of abduction of the right eye. The patient had a frozen orbit with a limitation of all gazes in the left eye. Fundus examination revealed a pale optic disk, afferent pupillary defect, and an ischemic retina, indicating central retinal artery occlusion. Examination findings were also consistent with oculomotor, trochlear, and abducens nerve paresis. Mucormycosis was a result of steroid therapy, as the patient was diagnosed 2 weeks previously with an isolated cranial nerve VII and was started with steroids in a peripheral ophthalmology clinic. At his first presentation in Jordan University Hospital, he had a frozen orbit with cranial nerve III, IV, and VI palsies. The patient was admitted, and the results of baseline investigations were normal, except for a high white blood cell count and hemoglobin A1C (HbA1c) of 10%. Systemic steroids were stopped, and the patient was started on insulin to correct his blood sugar. He was never diagnosed with diabetes mellitus before and denied any symptoms of diabetes mellitus. He was initially diagnosed with diabetes mellitus. He tested negative for ketone bodies and was treated in the ward. Results of other routine investigations, such as kidney function tests and complete blood count, were normal, except for high blood sugar and a high white blood cell count. The differential diagnosis of his current symptoms was cavernous sinus thrombosis, a space-occupying lesion in the cavernous sinus or orbital cellulitis. Computed tomography of the brain and orbit was performed and revealed extensive sinus disease with left complicated ethmoidal and maxillary sinusitis extending into the left orbit with abscess formation. There was also extensive orbital cellulitis and evidence of left cavernous sinus thrombosis secondary to infection (Figure 1). It is possible that the patient had undiagnosed diabetes mellitus as indicated by a high HbA1c and was worsened by steroid therapy The patient was started on systemic vancomycin and ceftriaxone, and a swab from the nasal cavity was sent for evaluation. Nasal swab revealed methicillin-resistant staphylococcus and positive fungal staining. The patient was started on systemic amphotericin B at a loading dose and continued with the therapeutic dose after 2 days of his presentation to our institution. Nasal endoscopy revealed diffuse white filaments and necrosis of the bone from which a biopsy was taken. Sinuses were drained, followed by ethmoidectomy and maxillectomy. A biopsy from the necrotic tissue was sent, which showed the presence of non-septate hyphae branching at right angles consistent with angio-invasive mucormycosis. However, workup to identify the exact causative organism of mucormycosis was not performed as it did not affect further management. The patient was kept on daily irrigation of the sinuses along with administration of systemic amphotericin B. On follow-up magnetic resonance imaging, the patient had an intraorbital collection (Figure 2). Subsequently, drainage of the orbital abscess, tarsorrhaphy, and excision of the necrotic conjunctiva were performed. The patient was kept on systemic amphotericin B and showed clinical improvement. He remained hospitalized until a repeat biopsy of the paranasal sinuses was negative for the fungi. He was discharged with residual paresis of all cranial nerves involved along with left eye ptosis and only hand motion in the left eye. He was followed up serially for 1 year and 4 months. Currently, the patient is afebrile with complete resolution of symptoms in the right eye and partial recovery of cranial nerve palsies. However, in the left eye, residual ptosis has remained and only hand motion is visible due to the previous central retinal artery occlusion related to his initial presentation. Discussion Mucormycosis is an opportunistic infection with a high mortality rate, reaching up to 80% without treatment. 6 Even with early diagnosis and aggressive therapy, its prognosis remains poor. 7 Diabetes mellitus is the most common risk factor for mucormycosis, particularly during ketoacidosis. Ketones induce the fungi to utilize and produce ketoreductase, which facilitates its growth through various mechanisms. 4 Roden et al. reviewed the characteristics of 929 patients with mucormycosis, and of 929 patients, 36% had diabetes mellitus, 17% had an underlying malignancy, and 19% had no underlying conditions. 3 In a review of literature of 851 patients, the median age was 51 ± 12 years, and 531 (63%) were men. Diabetes mellitus was the most common underlying condition (340/851, 40%; 71 (20%) had documented ketoacidosis), followed by hematological malignancy (275/851, 32%; 116 (42%) had acute myeloid leukemia) and solid organ transplantation (116/851, 14%; 67 (58%) had received a kidney transplant). Of the predisposing factors, the use of corticosteroids at the time of presentation was the most common (273/851, 33%), followed by neutropenia (169/851, 20%) and trauma (166/851, 20%). Rhino-orbito-cerebral mucormycosis was the most commonly observed manifestation (288/851, 34%), followed by cutaneous (187/851, 22%) and pulmonary mucormycosis (172/851, 20%. 8 In the present case, the initiation of glucocorticoids as a treatment for facial nerve palsy may have facilitated the initial development of mucormycosis. Glucocorticoids are well known to increase blood glucose levels and can exacerbate hyperglycemia in a patient with diabetes mellitus. 9 Our patient should have been tested for hyperglycemia before the initiation of glucocorticoid therapy (1) to rule out diabetes mellitus as the underlying cause of his facial palsy and (2) to measure the baseline blood sugar before the initiation of steroid therapy. 10 Clinically, mucormycosis is characterized by rhinitis with granular and purulent discharge, nasal ulceration, black spots of infarcted mucosa, and paranasal sinusitis. However, other common presenting symptoms include epistaxis, ophthalmoplegia with blindness, proptosis and orbital cellulitis, hemiplegia or stroke, and decreased mental function. 11 In advanced disease, symptoms include chemosis, ptosis, proptosis, ophthalmoplegia, blindness, and multiple cranial nerve palsies (function of cranial nerves II, III, V, VI, and VI may be lost or impaired).4,5 The classical clinical presentations are facial pain, an irregular black eschar on the palatal or nasal mucosa, and pus discharge from the eye and nose. In a retrospective chart review of 48 patients with 49 cases of acute fulminant-invasive fungal sinusitis over 19 years, mucormycosis and aspergillosis were found in 22 and 27 cases, respectively. Orbital (proptosis, periorbital edema, and ophthalmoplegia) and cranial nerve symptoms were seen at presentation more frequently in mucormycosis (6 [27%] and 9 [41%]) than in aspergillosis cases (3 [11%] and 7 [26%]; p = 0.079). Long-term orbital and cranial nerve sequelae were reported in 16 (72%) mucormycosis cases and 10 (37%) aspergillosis cases (p = 0.0210). These data suggest that the presence of orbital and neurological symptoms at presentation warrants more aggressive surgical intervention because of the likelihood of mucormycosis. 12 In another review of mucormycosis cases, all six patients had uncontrolled diabetes mellitus with fungal-invasive disease and had multiple cranial nerve involvement, including cranial nerves II to VII. 2 However, our patient had not been definitely diagnosed with diabetes mellitus before he had the fungal infection. That cohort chiefly complained of facial swelling, blood-stained nasal discharge, and crusting. All six patients were on prolonged antibiotic treatment with an incomplete response. Facial dysesthesia was present in all sides involved. Six patients had impaired vision, two had vision loss, and four only had a perception of light. These patients improved to finger counting during treatment. All patients had periorbital edema, ptosis, and 7th cranial nerve palsy of the lower motor type, and two patients had irritability and disoriented state and succumbed to the disease within 48 h of treatment. By contrast, our patient was fully oriented and had an initial visual acuity of hand motion with limited improvement. 13 A diagnosis of rhino-orbito-cerebral mucormycosis requires high clinical suspicion, and a definitive diagnosis requires a biopsy indicating the presence of the fungi. Furthermore, imaging is required to identify the degree of local involvement and spread across the rhino-orbital structures. 3 Mucormycosis is treated with surgical debridement of necrotic tissues, use of intravenous antifungal drugs, and elimination of predisposing factors. In retrospective studies, the combination of surgery and antifungal drugs showed higher survival rates than medical treatment alone. 3 Our case emphasizes the importance of screening for diabetes mellitus to prevent such a devastating complication. Our patient denied any of the classical symptoms of diabetes mellitus, and the only risk factor he had for diabetes mellitus was his age. Given the relatively long asymptomatic period of diabetes mellitus, the American Diabetes Association recommends screening for diabetes mellitus at age 45 years. 14 Early detection and treatment of this patient's underlying diabetes mellitus may have prevented the development of mucormycosis, especially in a country like Jordan where the prevalence of diabetes mellitus among men aged >25 years is as high as 33%. 15 The involvement of cranial nerve VII in mucormycosis is very rare, and its pathogenesis is not well understood. 13 We are unsure whether the involvement of cranial nerve VII in this patient was the first manifestation of mucormycosis or it was due to an independent pathology in which its treatment with glucocorticoids facilitated the development of mucormycosis. Furthermore, the widespread involvement of cranial nerves III, IV, and VI is also uncommon and indicates severe disease possibly due to cavernous sinus thrombosis. 3 Conclusions Mucormycosis is a rare and possibly fatal opportunistic infection. It should be suspected in any patient who presents with symptoms of sinusitis along with the widespread involvement of the orbit and cranial nerves. Workup should be initiated for an underlying systemic disease, such as diabetes mellitus, in any patient diagnosed with mucormycosis. In the present case, initiation of steroid therapy may have influenced the development of fungal infection and its progression, and this highlights the importance of screening. It is critical to exclude diabetes mellitus before starting steroid treatment and to screen a susceptible population perhaps at a younger age than recommended. Treatment with systemic antifungals, surgical debridement of necrotic tissues, and correction of underlying hyperglycemia should be initiated as soon as possible in diabetic patients with mucormycosis. Although this patient was 56 years old, screening for diabetes mellitus in younger populations earlier than 45 years is recommended, especially in the presence of opportunistic infections. Ethical statement Ethical approval was obtained from the Jordan University Board. Informed consent was also obtained from the patient. Conflict of interest All authors declare no conflict of interest Funding sources This work was not funded by any funding bodies, and all authors have read and approved the contents of this manuscript. Figure 1. Figure 2.

METHYLPREDNISOLONE

DrugsGivenReaction

CC BY

34888198

2021

Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Off label use'.

Mucormycosis with extensive cranial nerve involvement as the first presentation of diabetes mellitus: A case report. Mucormycosis, a rare fungal infection, mainly affects individuals with diabetes mellitus and those who were immunocompromised and has a high mortality rate. Its most common presentation is similar to that of acute bacterial sinusitis with symptoms of nasal congestion, headache, and fever. The involvement of multiple cranial nerves in mucormycosis was rarely reported in the literature and indicates severe disease. Herein, we report the case of a 56-year-old man who was referred to the ophthalmology outpatient clinic for facial nerve palsy. He was treated with systemic steroids for 10 days with no improvement. On examination, he had a loss of vision and a frozen orbit due to involvement of cranial nerves II, III, IV, V, VI, and VII. An extensive workup revealed a hemoglobin A1C of 10%. However, he was never diagnosed with diabetes mellitus previously and denied any of the classical symptoms of diabetes mellitus. He underwent ethmoidectomy, maxillectomy, and drainage of an intraorbital abscess after appropriate imaging studies. Histopathology confirmed the diagnosis of mucormycosis, and the patient was started on systemic amphotericin B. This case emphasizes the importance of screening for diabetes mellitus. Early recognition of underlying diabetes mellitus in this patient may have prevented the development of mucormycosis along with its devastating complications. pmcIntroduction Mucormycosis is a rare opportunistic angio-invasive infection caused by fungi of the order Mucorales. It mainly affects individuals with diabetes mellitus and patients who have underlying immunosuppression, with a mortality rate reaching up to 80%. 1 According to reports from the United States, mucormycosis has a very low incidence of 1.7 in every million individuals. Diabetes mellitus is the most common risk factor for the development of mucormycosis across Asia, while hematological malignancies are the major risk factors across Western countries. 2 Although mucormycosis is a severe complication of poorly controlled diabetes mellitus, it is very rare for diabetes mellitus to be diagnosed after the development of mucormycosis. 3 Mucormycosis can involve many organ systems, including the cutaneous, pulmonary, and gastrointestinal systems, and in severe forms, it can spread to other organs. However, rhinocerebral involvement is the most frequent presentation. 4 Involvement of cranial nerve VII along with the widespread involvement of other cranial nerves was rarely reported in the literature. 5 Herein, we report a very unusual case of mucormycosis with extensive rhino-orbit cerebral involvement presenting with multiple cranial nerve palsies as a first presentation of diabetes mellitus in a previously healthy patient. Case Description A 56-year-old man was referred to our institution for a case of facial palsy. He was treated with systemic methylprednisolone for 10 days, without improvement. On examination, he had a visual acuity of 1.0 in the right eye and hand motion in the left. Motility examination revealed limitation of abduction of the right eye. The patient had a frozen orbit with a limitation of all gazes in the left eye. Fundus examination revealed a pale optic disk, afferent pupillary defect, and an ischemic retina, indicating central retinal artery occlusion. Examination findings were also consistent with oculomotor, trochlear, and abducens nerve paresis. Mucormycosis was a result of steroid therapy, as the patient was diagnosed 2 weeks previously with an isolated cranial nerve VII and was started with steroids in a peripheral ophthalmology clinic. At his first presentation in Jordan University Hospital, he had a frozen orbit with cranial nerve III, IV, and VI palsies. The patient was admitted, and the results of baseline investigations were normal, except for a high white blood cell count and hemoglobin A1C (HbA1c) of 10%. Systemic steroids were stopped, and the patient was started on insulin to correct his blood sugar. He was never diagnosed with diabetes mellitus before and denied any symptoms of diabetes mellitus. He was initially diagnosed with diabetes mellitus. He tested negative for ketone bodies and was treated in the ward. Results of other routine investigations, such as kidney function tests and complete blood count, were normal, except for high blood sugar and a high white blood cell count. The differential diagnosis of his current symptoms was cavernous sinus thrombosis, a space-occupying lesion in the cavernous sinus or orbital cellulitis. Computed tomography of the brain and orbit was performed and revealed extensive sinus disease with left complicated ethmoidal and maxillary sinusitis extending into the left orbit with abscess formation. There was also extensive orbital cellulitis and evidence of left cavernous sinus thrombosis secondary to infection (Figure 1). It is possible that the patient had undiagnosed diabetes mellitus as indicated by a high HbA1c and was worsened by steroid therapy The patient was started on systemic vancomycin and ceftriaxone, and a swab from the nasal cavity was sent for evaluation. Nasal swab revealed methicillin-resistant staphylococcus and positive fungal staining. The patient was started on systemic amphotericin B at a loading dose and continued with the therapeutic dose after 2 days of his presentation to our institution. Nasal endoscopy revealed diffuse white filaments and necrosis of the bone from which a biopsy was taken. Sinuses were drained, followed by ethmoidectomy and maxillectomy. A biopsy from the necrotic tissue was sent, which showed the presence of non-septate hyphae branching at right angles consistent with angio-invasive mucormycosis. However, workup to identify the exact causative organism of mucormycosis was not performed as it did not affect further management. The patient was kept on daily irrigation of the sinuses along with administration of systemic amphotericin B. On follow-up magnetic resonance imaging, the patient had an intraorbital collection (Figure 2). Subsequently, drainage of the orbital abscess, tarsorrhaphy, and excision of the necrotic conjunctiva were performed. The patient was kept on systemic amphotericin B and showed clinical improvement. He remained hospitalized until a repeat biopsy of the paranasal sinuses was negative for the fungi. He was discharged with residual paresis of all cranial nerves involved along with left eye ptosis and only hand motion in the left eye. He was followed up serially for 1 year and 4 months. Currently, the patient is afebrile with complete resolution of symptoms in the right eye and partial recovery of cranial nerve palsies. However, in the left eye, residual ptosis has remained and only hand motion is visible due to the previous central retinal artery occlusion related to his initial presentation. Discussion Mucormycosis is an opportunistic infection with a high mortality rate, reaching up to 80% without treatment. 6 Even with early diagnosis and aggressive therapy, its prognosis remains poor. 7 Diabetes mellitus is the most common risk factor for mucormycosis, particularly during ketoacidosis. Ketones induce the fungi to utilize and produce ketoreductase, which facilitates its growth through various mechanisms. 4 Roden et al. reviewed the characteristics of 929 patients with mucormycosis, and of 929 patients, 36% had diabetes mellitus, 17% had an underlying malignancy, and 19% had no underlying conditions. 3 In a review of literature of 851 patients, the median age was 51 ± 12 years, and 531 (63%) were men. Diabetes mellitus was the most common underlying condition (340/851, 40%; 71 (20%) had documented ketoacidosis), followed by hematological malignancy (275/851, 32%; 116 (42%) had acute myeloid leukemia) and solid organ transplantation (116/851, 14%; 67 (58%) had received a kidney transplant). Of the predisposing factors, the use of corticosteroids at the time of presentation was the most common (273/851, 33%), followed by neutropenia (169/851, 20%) and trauma (166/851, 20%). Rhino-orbito-cerebral mucormycosis was the most commonly observed manifestation (288/851, 34%), followed by cutaneous (187/851, 22%) and pulmonary mucormycosis (172/851, 20%. 8 In the present case, the initiation of glucocorticoids as a treatment for facial nerve palsy may have facilitated the initial development of mucormycosis. Glucocorticoids are well known to increase blood glucose levels and can exacerbate hyperglycemia in a patient with diabetes mellitus. 9 Our patient should have been tested for hyperglycemia before the initiation of glucocorticoid therapy (1) to rule out diabetes mellitus as the underlying cause of his facial palsy and (2) to measure the baseline blood sugar before the initiation of steroid therapy. 10 Clinically, mucormycosis is characterized by rhinitis with granular and purulent discharge, nasal ulceration, black spots of infarcted mucosa, and paranasal sinusitis. However, other common presenting symptoms include epistaxis, ophthalmoplegia with blindness, proptosis and orbital cellulitis, hemiplegia or stroke, and decreased mental function. 11 In advanced disease, symptoms include chemosis, ptosis, proptosis, ophthalmoplegia, blindness, and multiple cranial nerve palsies (function of cranial nerves II, III, V, VI, and VI may be lost or impaired).4,5 The classical clinical presentations are facial pain, an irregular black eschar on the palatal or nasal mucosa, and pus discharge from the eye and nose. In a retrospective chart review of 48 patients with 49 cases of acute fulminant-invasive fungal sinusitis over 19 years, mucormycosis and aspergillosis were found in 22 and 27 cases, respectively. Orbital (proptosis, periorbital edema, and ophthalmoplegia) and cranial nerve symptoms were seen at presentation more frequently in mucormycosis (6 [27%] and 9 [41%]) than in aspergillosis cases (3 [11%] and 7 [26%]; p = 0.079). Long-term orbital and cranial nerve sequelae were reported in 16 (72%) mucormycosis cases and 10 (37%) aspergillosis cases (p = 0.0210). These data suggest that the presence of orbital and neurological symptoms at presentation warrants more aggressive surgical intervention because of the likelihood of mucormycosis. 12 In another review of mucormycosis cases, all six patients had uncontrolled diabetes mellitus with fungal-invasive disease and had multiple cranial nerve involvement, including cranial nerves II to VII. 2 However, our patient had not been definitely diagnosed with diabetes mellitus before he had the fungal infection. That cohort chiefly complained of facial swelling, blood-stained nasal discharge, and crusting. All six patients were on prolonged antibiotic treatment with an incomplete response. Facial dysesthesia was present in all sides involved. Six patients had impaired vision, two had vision loss, and four only had a perception of light. These patients improved to finger counting during treatment. All patients had periorbital edema, ptosis, and 7th cranial nerve palsy of the lower motor type, and two patients had irritability and disoriented state and succumbed to the disease within 48 h of treatment. By contrast, our patient was fully oriented and had an initial visual acuity of hand motion with limited improvement. 13 A diagnosis of rhino-orbito-cerebral mucormycosis requires high clinical suspicion, and a definitive diagnosis requires a biopsy indicating the presence of the fungi. Furthermore, imaging is required to identify the degree of local involvement and spread across the rhino-orbital structures. 3 Mucormycosis is treated with surgical debridement of necrotic tissues, use of intravenous antifungal drugs, and elimination of predisposing factors. In retrospective studies, the combination of surgery and antifungal drugs showed higher survival rates than medical treatment alone. 3 Our case emphasizes the importance of screening for diabetes mellitus to prevent such a devastating complication. Our patient denied any of the classical symptoms of diabetes mellitus, and the only risk factor he had for diabetes mellitus was his age. Given the relatively long asymptomatic period of diabetes mellitus, the American Diabetes Association recommends screening for diabetes mellitus at age 45 years. 14 Early detection and treatment of this patient's underlying diabetes mellitus may have prevented the development of mucormycosis, especially in a country like Jordan where the prevalence of diabetes mellitus among men aged >25 years is as high as 33%. 15 The involvement of cranial nerve VII in mucormycosis is very rare, and its pathogenesis is not well understood. 13 We are unsure whether the involvement of cranial nerve VII in this patient was the first manifestation of mucormycosis or it was due to an independent pathology in which its treatment with glucocorticoids facilitated the development of mucormycosis. Furthermore, the widespread involvement of cranial nerves III, IV, and VI is also uncommon and indicates severe disease possibly due to cavernous sinus thrombosis. 3 Conclusions Mucormycosis is a rare and possibly fatal opportunistic infection. It should be suspected in any patient who presents with symptoms of sinusitis along with the widespread involvement of the orbit and cranial nerves. Workup should be initiated for an underlying systemic disease, such as diabetes mellitus, in any patient diagnosed with mucormycosis. In the present case, initiation of steroid therapy may have influenced the development of fungal infection and its progression, and this highlights the importance of screening. It is critical to exclude diabetes mellitus before starting steroid treatment and to screen a susceptible population perhaps at a younger age than recommended. Treatment with systemic antifungals, surgical debridement of necrotic tissues, and correction of underlying hyperglycemia should be initiated as soon as possible in diabetic patients with mucormycosis. Although this patient was 56 years old, screening for diabetes mellitus in younger populations earlier than 45 years is recommended, especially in the presence of opportunistic infections. Ethical statement Ethical approval was obtained from the Jordan University Board. Informed consent was also obtained from the patient. Conflict of interest All authors declare no conflict of interest Funding sources This work was not funded by any funding bodies, and all authors have read and approved the contents of this manuscript. Figure 1. Figure 2.

METHYLPREDNISOLONE

DrugsGivenReaction

CC BY

34888198

2021

Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Osteonecrosis'.

Mucormycosis with extensive cranial nerve involvement as the first presentation of diabetes mellitus: A case report. Mucormycosis, a rare fungal infection, mainly affects individuals with diabetes mellitus and those who were immunocompromised and has a high mortality rate. Its most common presentation is similar to that of acute bacterial sinusitis with symptoms of nasal congestion, headache, and fever. The involvement of multiple cranial nerves in mucormycosis was rarely reported in the literature and indicates severe disease. Herein, we report the case of a 56-year-old man who was referred to the ophthalmology outpatient clinic for facial nerve palsy. He was treated with systemic steroids for 10 days with no improvement. On examination, he had a loss of vision and a frozen orbit due to involvement of cranial nerves II, III, IV, V, VI, and VII. An extensive workup revealed a hemoglobin A1C of 10%. However, he was never diagnosed with diabetes mellitus previously and denied any of the classical symptoms of diabetes mellitus. He underwent ethmoidectomy, maxillectomy, and drainage of an intraorbital abscess after appropriate imaging studies. Histopathology confirmed the diagnosis of mucormycosis, and the patient was started on systemic amphotericin B. This case emphasizes the importance of screening for diabetes mellitus. Early recognition of underlying diabetes mellitus in this patient may have prevented the development of mucormycosis along with its devastating complications. pmcIntroduction Mucormycosis is a rare opportunistic angio-invasive infection caused by fungi of the order Mucorales. It mainly affects individuals with diabetes mellitus and patients who have underlying immunosuppression, with a mortality rate reaching up to 80%. 1 According to reports from the United States, mucormycosis has a very low incidence of 1.7 in every million individuals. Diabetes mellitus is the most common risk factor for the development of mucormycosis across Asia, while hematological malignancies are the major risk factors across Western countries. 2 Although mucormycosis is a severe complication of poorly controlled diabetes mellitus, it is very rare for diabetes mellitus to be diagnosed after the development of mucormycosis. 3 Mucormycosis can involve many organ systems, including the cutaneous, pulmonary, and gastrointestinal systems, and in severe forms, it can spread to other organs. However, rhinocerebral involvement is the most frequent presentation. 4 Involvement of cranial nerve VII along with the widespread involvement of other cranial nerves was rarely reported in the literature. 5 Herein, we report a very unusual case of mucormycosis with extensive rhino-orbit cerebral involvement presenting with multiple cranial nerve palsies as a first presentation of diabetes mellitus in a previously healthy patient. Case Description A 56-year-old man was referred to our institution for a case of facial palsy. He was treated with systemic methylprednisolone for 10 days, without improvement. On examination, he had a visual acuity of 1.0 in the right eye and hand motion in the left. Motility examination revealed limitation of abduction of the right eye. The patient had a frozen orbit with a limitation of all gazes in the left eye. Fundus examination revealed a pale optic disk, afferent pupillary defect, and an ischemic retina, indicating central retinal artery occlusion. Examination findings were also consistent with oculomotor, trochlear, and abducens nerve paresis. Mucormycosis was a result of steroid therapy, as the patient was diagnosed 2 weeks previously with an isolated cranial nerve VII and was started with steroids in a peripheral ophthalmology clinic. At his first presentation in Jordan University Hospital, he had a frozen orbit with cranial nerve III, IV, and VI palsies. The patient was admitted, and the results of baseline investigations were normal, except for a high white blood cell count and hemoglobin A1C (HbA1c) of 10%. Systemic steroids were stopped, and the patient was started on insulin to correct his blood sugar. He was never diagnosed with diabetes mellitus before and denied any symptoms of diabetes mellitus. He was initially diagnosed with diabetes mellitus. He tested negative for ketone bodies and was treated in the ward. Results of other routine investigations, such as kidney function tests and complete blood count, were normal, except for high blood sugar and a high white blood cell count. The differential diagnosis of his current symptoms was cavernous sinus thrombosis, a space-occupying lesion in the cavernous sinus or orbital cellulitis. Computed tomography of the brain and orbit was performed and revealed extensive sinus disease with left complicated ethmoidal and maxillary sinusitis extending into the left orbit with abscess formation. There was also extensive orbital cellulitis and evidence of left cavernous sinus thrombosis secondary to infection (Figure 1). It is possible that the patient had undiagnosed diabetes mellitus as indicated by a high HbA1c and was worsened by steroid therapy The patient was started on systemic vancomycin and ceftriaxone, and a swab from the nasal cavity was sent for evaluation. Nasal swab revealed methicillin-resistant staphylococcus and positive fungal staining. The patient was started on systemic amphotericin B at a loading dose and continued with the therapeutic dose after 2 days of his presentation to our institution. Nasal endoscopy revealed diffuse white filaments and necrosis of the bone from which a biopsy was taken. Sinuses were drained, followed by ethmoidectomy and maxillectomy. A biopsy from the necrotic tissue was sent, which showed the presence of non-septate hyphae branching at right angles consistent with angio-invasive mucormycosis. However, workup to identify the exact causative organism of mucormycosis was not performed as it did not affect further management. The patient was kept on daily irrigation of the sinuses along with administration of systemic amphotericin B. On follow-up magnetic resonance imaging, the patient had an intraorbital collection (Figure 2). Subsequently, drainage of the orbital abscess, tarsorrhaphy, and excision of the necrotic conjunctiva were performed. The patient was kept on systemic amphotericin B and showed clinical improvement. He remained hospitalized until a repeat biopsy of the paranasal sinuses was negative for the fungi. He was discharged with residual paresis of all cranial nerves involved along with left eye ptosis and only hand motion in the left eye. He was followed up serially for 1 year and 4 months. Currently, the patient is afebrile with complete resolution of symptoms in the right eye and partial recovery of cranial nerve palsies. However, in the left eye, residual ptosis has remained and only hand motion is visible due to the previous central retinal artery occlusion related to his initial presentation. Discussion Mucormycosis is an opportunistic infection with a high mortality rate, reaching up to 80% without treatment. 6 Even with early diagnosis and aggressive therapy, its prognosis remains poor. 7 Diabetes mellitus is the most common risk factor for mucormycosis, particularly during ketoacidosis. Ketones induce the fungi to utilize and produce ketoreductase, which facilitates its growth through various mechanisms. 4 Roden et al. reviewed the characteristics of 929 patients with mucormycosis, and of 929 patients, 36% had diabetes mellitus, 17% had an underlying malignancy, and 19% had no underlying conditions. 3 In a review of literature of 851 patients, the median age was 51 ± 12 years, and 531 (63%) were men. Diabetes mellitus was the most common underlying condition (340/851, 40%; 71 (20%) had documented ketoacidosis), followed by hematological malignancy (275/851, 32%; 116 (42%) had acute myeloid leukemia) and solid organ transplantation (116/851, 14%; 67 (58%) had received a kidney transplant). Of the predisposing factors, the use of corticosteroids at the time of presentation was the most common (273/851, 33%), followed by neutropenia (169/851, 20%) and trauma (166/851, 20%). Rhino-orbito-cerebral mucormycosis was the most commonly observed manifestation (288/851, 34%), followed by cutaneous (187/851, 22%) and pulmonary mucormycosis (172/851, 20%. 8 In the present case, the initiation of glucocorticoids as a treatment for facial nerve palsy may have facilitated the initial development of mucormycosis. Glucocorticoids are well known to increase blood glucose levels and can exacerbate hyperglycemia in a patient with diabetes mellitus. 9 Our patient should have been tested for hyperglycemia before the initiation of glucocorticoid therapy (1) to rule out diabetes mellitus as the underlying cause of his facial palsy and (2) to measure the baseline blood sugar before the initiation of steroid therapy. 10 Clinically, mucormycosis is characterized by rhinitis with granular and purulent discharge, nasal ulceration, black spots of infarcted mucosa, and paranasal sinusitis. However, other common presenting symptoms include epistaxis, ophthalmoplegia with blindness, proptosis and orbital cellulitis, hemiplegia or stroke, and decreased mental function. 11 In advanced disease, symptoms include chemosis, ptosis, proptosis, ophthalmoplegia, blindness, and multiple cranial nerve palsies (function of cranial nerves II, III, V, VI, and VI may be lost or impaired).4,5 The classical clinical presentations are facial pain, an irregular black eschar on the palatal or nasal mucosa, and pus discharge from the eye and nose. In a retrospective chart review of 48 patients with 49 cases of acute fulminant-invasive fungal sinusitis over 19 years, mucormycosis and aspergillosis were found in 22 and 27 cases, respectively. Orbital (proptosis, periorbital edema, and ophthalmoplegia) and cranial nerve symptoms were seen at presentation more frequently in mucormycosis (6 [27%] and 9 [41%]) than in aspergillosis cases (3 [11%] and 7 [26%]; p = 0.079). Long-term orbital and cranial nerve sequelae were reported in 16 (72%) mucormycosis cases and 10 (37%) aspergillosis cases (p = 0.0210). These data suggest that the presence of orbital and neurological symptoms at presentation warrants more aggressive surgical intervention because of the likelihood of mucormycosis. 12 In another review of mucormycosis cases, all six patients had uncontrolled diabetes mellitus with fungal-invasive disease and had multiple cranial nerve involvement, including cranial nerves II to VII. 2 However, our patient had not been definitely diagnosed with diabetes mellitus before he had the fungal infection. That cohort chiefly complained of facial swelling, blood-stained nasal discharge, and crusting. All six patients were on prolonged antibiotic treatment with an incomplete response. Facial dysesthesia was present in all sides involved. Six patients had impaired vision, two had vision loss, and four only had a perception of light. These patients improved to finger counting during treatment. All patients had periorbital edema, ptosis, and 7th cranial nerve palsy of the lower motor type, and two patients had irritability and disoriented state and succumbed to the disease within 48 h of treatment. By contrast, our patient was fully oriented and had an initial visual acuity of hand motion with limited improvement. 13 A diagnosis of rhino-orbito-cerebral mucormycosis requires high clinical suspicion, and a definitive diagnosis requires a biopsy indicating the presence of the fungi. Furthermore, imaging is required to identify the degree of local involvement and spread across the rhino-orbital structures. 3 Mucormycosis is treated with surgical debridement of necrotic tissues, use of intravenous antifungal drugs, and elimination of predisposing factors. In retrospective studies, the combination of surgery and antifungal drugs showed higher survival rates than medical treatment alone. 3 Our case emphasizes the importance of screening for diabetes mellitus to prevent such a devastating complication. Our patient denied any of the classical symptoms of diabetes mellitus, and the only risk factor he had for diabetes mellitus was his age. Given the relatively long asymptomatic period of diabetes mellitus, the American Diabetes Association recommends screening for diabetes mellitus at age 45 years. 14 Early detection and treatment of this patient's underlying diabetes mellitus may have prevented the development of mucormycosis, especially in a country like Jordan where the prevalence of diabetes mellitus among men aged >25 years is as high as 33%. 15 The involvement of cranial nerve VII in mucormycosis is very rare, and its pathogenesis is not well understood. 13 We are unsure whether the involvement of cranial nerve VII in this patient was the first manifestation of mucormycosis or it was due to an independent pathology in which its treatment with glucocorticoids facilitated the development of mucormycosis. Furthermore, the widespread involvement of cranial nerves III, IV, and VI is also uncommon and indicates severe disease possibly due to cavernous sinus thrombosis. 3 Conclusions Mucormycosis is a rare and possibly fatal opportunistic infection. It should be suspected in any patient who presents with symptoms of sinusitis along with the widespread involvement of the orbit and cranial nerves. Workup should be initiated for an underlying systemic disease, such as diabetes mellitus, in any patient diagnosed with mucormycosis. In the present case, initiation of steroid therapy may have influenced the development of fungal infection and its progression, and this highlights the importance of screening. It is critical to exclude diabetes mellitus before starting steroid treatment and to screen a susceptible population perhaps at a younger age than recommended. Treatment with systemic antifungals, surgical debridement of necrotic tissues, and correction of underlying hyperglycemia should be initiated as soon as possible in diabetic patients with mucormycosis. Although this patient was 56 years old, screening for diabetes mellitus in younger populations earlier than 45 years is recommended, especially in the presence of opportunistic infections. Ethical statement Ethical approval was obtained from the Jordan University Board. Informed consent was also obtained from the patient. Conflict of interest All authors declare no conflict of interest Funding sources This work was not funded by any funding bodies, and all authors have read and approved the contents of this manuscript. Figure 1. Figure 2.

METHYLPREDNISOLONE

DrugsGivenReaction

CC BY

34888198

2021

Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Periorbital abscess'.

Mucormycosis with extensive cranial nerve involvement as the first presentation of diabetes mellitus: A case report. Mucormycosis, a rare fungal infection, mainly affects individuals with diabetes mellitus and those who were immunocompromised and has a high mortality rate. Its most common presentation is similar to that of acute bacterial sinusitis with symptoms of nasal congestion, headache, and fever. The involvement of multiple cranial nerves in mucormycosis was rarely reported in the literature and indicates severe disease. Herein, we report the case of a 56-year-old man who was referred to the ophthalmology outpatient clinic for facial nerve palsy. He was treated with systemic steroids for 10 days with no improvement. On examination, he had a loss of vision and a frozen orbit due to involvement of cranial nerves II, III, IV, V, VI, and VII. An extensive workup revealed a hemoglobin A1C of 10%. However, he was never diagnosed with diabetes mellitus previously and denied any of the classical symptoms of diabetes mellitus. He underwent ethmoidectomy, maxillectomy, and drainage of an intraorbital abscess after appropriate imaging studies. Histopathology confirmed the diagnosis of mucormycosis, and the patient was started on systemic amphotericin B. This case emphasizes the importance of screening for diabetes mellitus. Early recognition of underlying diabetes mellitus in this patient may have prevented the development of mucormycosis along with its devastating complications. pmcIntroduction Mucormycosis is a rare opportunistic angio-invasive infection caused by fungi of the order Mucorales. It mainly affects individuals with diabetes mellitus and patients who have underlying immunosuppression, with a mortality rate reaching up to 80%. 1 According to reports from the United States, mucormycosis has a very low incidence of 1.7 in every million individuals. Diabetes mellitus is the most common risk factor for the development of mucormycosis across Asia, while hematological malignancies are the major risk factors across Western countries. 2 Although mucormycosis is a severe complication of poorly controlled diabetes mellitus, it is very rare for diabetes mellitus to be diagnosed after the development of mucormycosis. 3 Mucormycosis can involve many organ systems, including the cutaneous, pulmonary, and gastrointestinal systems, and in severe forms, it can spread to other organs. However, rhinocerebral involvement is the most frequent presentation. 4 Involvement of cranial nerve VII along with the widespread involvement of other cranial nerves was rarely reported in the literature. 5 Herein, we report a very unusual case of mucormycosis with extensive rhino-orbit cerebral involvement presenting with multiple cranial nerve palsies as a first presentation of diabetes mellitus in a previously healthy patient. Case Description A 56-year-old man was referred to our institution for a case of facial palsy. He was treated with systemic methylprednisolone for 10 days, without improvement. On examination, he had a visual acuity of 1.0 in the right eye and hand motion in the left. Motility examination revealed limitation of abduction of the right eye. The patient had a frozen orbit with a limitation of all gazes in the left eye. Fundus examination revealed a pale optic disk, afferent pupillary defect, and an ischemic retina, indicating central retinal artery occlusion. Examination findings were also consistent with oculomotor, trochlear, and abducens nerve paresis. Mucormycosis was a result of steroid therapy, as the patient was diagnosed 2 weeks previously with an isolated cranial nerve VII and was started with steroids in a peripheral ophthalmology clinic. At his first presentation in Jordan University Hospital, he had a frozen orbit with cranial nerve III, IV, and VI palsies. The patient was admitted, and the results of baseline investigations were normal, except for a high white blood cell count and hemoglobin A1C (HbA1c) of 10%. Systemic steroids were stopped, and the patient was started on insulin to correct his blood sugar. He was never diagnosed with diabetes mellitus before and denied any symptoms of diabetes mellitus. He was initially diagnosed with diabetes mellitus. He tested negative for ketone bodies and was treated in the ward. Results of other routine investigations, such as kidney function tests and complete blood count, were normal, except for high blood sugar and a high white blood cell count. The differential diagnosis of his current symptoms was cavernous sinus thrombosis, a space-occupying lesion in the cavernous sinus or orbital cellulitis. Computed tomography of the brain and orbit was performed and revealed extensive sinus disease with left complicated ethmoidal and maxillary sinusitis extending into the left orbit with abscess formation. There was also extensive orbital cellulitis and evidence of left cavernous sinus thrombosis secondary to infection (Figure 1). It is possible that the patient had undiagnosed diabetes mellitus as indicated by a high HbA1c and was worsened by steroid therapy The patient was started on systemic vancomycin and ceftriaxone, and a swab from the nasal cavity was sent for evaluation. Nasal swab revealed methicillin-resistant staphylococcus and positive fungal staining. The patient was started on systemic amphotericin B at a loading dose and continued with the therapeutic dose after 2 days of his presentation to our institution. Nasal endoscopy revealed diffuse white filaments and necrosis of the bone from which a biopsy was taken. Sinuses were drained, followed by ethmoidectomy and maxillectomy. A biopsy from the necrotic tissue was sent, which showed the presence of non-septate hyphae branching at right angles consistent with angio-invasive mucormycosis. However, workup to identify the exact causative organism of mucormycosis was not performed as it did not affect further management. The patient was kept on daily irrigation of the sinuses along with administration of systemic amphotericin B. On follow-up magnetic resonance imaging, the patient had an intraorbital collection (Figure 2). Subsequently, drainage of the orbital abscess, tarsorrhaphy, and excision of the necrotic conjunctiva were performed. The patient was kept on systemic amphotericin B and showed clinical improvement. He remained hospitalized until a repeat biopsy of the paranasal sinuses was negative for the fungi. He was discharged with residual paresis of all cranial nerves involved along with left eye ptosis and only hand motion in the left eye. He was followed up serially for 1 year and 4 months. Currently, the patient is afebrile with complete resolution of symptoms in the right eye and partial recovery of cranial nerve palsies. However, in the left eye, residual ptosis has remained and only hand motion is visible due to the previous central retinal artery occlusion related to his initial presentation. Discussion Mucormycosis is an opportunistic infection with a high mortality rate, reaching up to 80% without treatment. 6 Even with early diagnosis and aggressive therapy, its prognosis remains poor. 7 Diabetes mellitus is the most common risk factor for mucormycosis, particularly during ketoacidosis. Ketones induce the fungi to utilize and produce ketoreductase, which facilitates its growth through various mechanisms. 4 Roden et al. reviewed the characteristics of 929 patients with mucormycosis, and of 929 patients, 36% had diabetes mellitus, 17% had an underlying malignancy, and 19% had no underlying conditions. 3 In a review of literature of 851 patients, the median age was 51 ± 12 years, and 531 (63%) were men. Diabetes mellitus was the most common underlying condition (340/851, 40%; 71 (20%) had documented ketoacidosis), followed by hematological malignancy (275/851, 32%; 116 (42%) had acute myeloid leukemia) and solid organ transplantation (116/851, 14%; 67 (58%) had received a kidney transplant). Of the predisposing factors, the use of corticosteroids at the time of presentation was the most common (273/851, 33%), followed by neutropenia (169/851, 20%) and trauma (166/851, 20%). Rhino-orbito-cerebral mucormycosis was the most commonly observed manifestation (288/851, 34%), followed by cutaneous (187/851, 22%) and pulmonary mucormycosis (172/851, 20%. 8 In the present case, the initiation of glucocorticoids as a treatment for facial nerve palsy may have facilitated the initial development of mucormycosis. Glucocorticoids are well known to increase blood glucose levels and can exacerbate hyperglycemia in a patient with diabetes mellitus. 9 Our patient should have been tested for hyperglycemia before the initiation of glucocorticoid therapy (1) to rule out diabetes mellitus as the underlying cause of his facial palsy and (2) to measure the baseline blood sugar before the initiation of steroid therapy. 10 Clinically, mucormycosis is characterized by rhinitis with granular and purulent discharge, nasal ulceration, black spots of infarcted mucosa, and paranasal sinusitis. However, other common presenting symptoms include epistaxis, ophthalmoplegia with blindness, proptosis and orbital cellulitis, hemiplegia or stroke, and decreased mental function. 11 In advanced disease, symptoms include chemosis, ptosis, proptosis, ophthalmoplegia, blindness, and multiple cranial nerve palsies (function of cranial nerves II, III, V, VI, and VI may be lost or impaired).4,5 The classical clinical presentations are facial pain, an irregular black eschar on the palatal or nasal mucosa, and pus discharge from the eye and nose. In a retrospective chart review of 48 patients with 49 cases of acute fulminant-invasive fungal sinusitis over 19 years, mucormycosis and aspergillosis were found in 22 and 27 cases, respectively. Orbital (proptosis, periorbital edema, and ophthalmoplegia) and cranial nerve symptoms were seen at presentation more frequently in mucormycosis (6 [27%] and 9 [41%]) than in aspergillosis cases (3 [11%] and 7 [26%]; p = 0.079). Long-term orbital and cranial nerve sequelae were reported in 16 (72%) mucormycosis cases and 10 (37%) aspergillosis cases (p = 0.0210). These data suggest that the presence of orbital and neurological symptoms at presentation warrants more aggressive surgical intervention because of the likelihood of mucormycosis. 12 In another review of mucormycosis cases, all six patients had uncontrolled diabetes mellitus with fungal-invasive disease and had multiple cranial nerve involvement, including cranial nerves II to VII. 2 However, our patient had not been definitely diagnosed with diabetes mellitus before he had the fungal infection. That cohort chiefly complained of facial swelling, blood-stained nasal discharge, and crusting. All six patients were on prolonged antibiotic treatment with an incomplete response. Facial dysesthesia was present in all sides involved. Six patients had impaired vision, two had vision loss, and four only had a perception of light. These patients improved to finger counting during treatment. All patients had periorbital edema, ptosis, and 7th cranial nerve palsy of the lower motor type, and two patients had irritability and disoriented state and succumbed to the disease within 48 h of treatment. By contrast, our patient was fully oriented and had an initial visual acuity of hand motion with limited improvement. 13 A diagnosis of rhino-orbito-cerebral mucormycosis requires high clinical suspicion, and a definitive diagnosis requires a biopsy indicating the presence of the fungi. Furthermore, imaging is required to identify the degree of local involvement and spread across the rhino-orbital structures. 3 Mucormycosis is treated with surgical debridement of necrotic tissues, use of intravenous antifungal drugs, and elimination of predisposing factors. In retrospective studies, the combination of surgery and antifungal drugs showed higher survival rates than medical treatment alone. 3 Our case emphasizes the importance of screening for diabetes mellitus to prevent such a devastating complication. Our patient denied any of the classical symptoms of diabetes mellitus, and the only risk factor he had for diabetes mellitus was his age. Given the relatively long asymptomatic period of diabetes mellitus, the American Diabetes Association recommends screening for diabetes mellitus at age 45 years. 14 Early detection and treatment of this patient's underlying diabetes mellitus may have prevented the development of mucormycosis, especially in a country like Jordan where the prevalence of diabetes mellitus among men aged >25 years is as high as 33%. 15 The involvement of cranial nerve VII in mucormycosis is very rare, and its pathogenesis is not well understood. 13 We are unsure whether the involvement of cranial nerve VII in this patient was the first manifestation of mucormycosis or it was due to an independent pathology in which its treatment with glucocorticoids facilitated the development of mucormycosis. Furthermore, the widespread involvement of cranial nerves III, IV, and VI is also uncommon and indicates severe disease possibly due to cavernous sinus thrombosis. 3 Conclusions Mucormycosis is a rare and possibly fatal opportunistic infection. It should be suspected in any patient who presents with symptoms of sinusitis along with the widespread involvement of the orbit and cranial nerves. Workup should be initiated for an underlying systemic disease, such as diabetes mellitus, in any patient diagnosed with mucormycosis. In the present case, initiation of steroid therapy may have influenced the development of fungal infection and its progression, and this highlights the importance of screening. It is critical to exclude diabetes mellitus before starting steroid treatment and to screen a susceptible population perhaps at a younger age than recommended. Treatment with systemic antifungals, surgical debridement of necrotic tissues, and correction of underlying hyperglycemia should be initiated as soon as possible in diabetic patients with mucormycosis. Although this patient was 56 years old, screening for diabetes mellitus in younger populations earlier than 45 years is recommended, especially in the presence of opportunistic infections. Ethical statement Ethical approval was obtained from the Jordan University Board. Informed consent was also obtained from the patient. Conflict of interest All authors declare no conflict of interest Funding sources This work was not funded by any funding bodies, and all authors have read and approved the contents of this manuscript. Figure 1. Figure 2.

METHYLPREDNISOLONE

DrugsGivenReaction

CC BY

34888198

2021

Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Rhinocerebral mucormycosis'.

Mucormycosis with extensive cranial nerve involvement as the first presentation of diabetes mellitus: A case report. Mucormycosis, a rare fungal infection, mainly affects individuals with diabetes mellitus and those who were immunocompromised and has a high mortality rate. Its most common presentation is similar to that of acute bacterial sinusitis with symptoms of nasal congestion, headache, and fever. The involvement of multiple cranial nerves in mucormycosis was rarely reported in the literature and indicates severe disease. Herein, we report the case of a 56-year-old man who was referred to the ophthalmology outpatient clinic for facial nerve palsy. He was treated with systemic steroids for 10 days with no improvement. On examination, he had a loss of vision and a frozen orbit due to involvement of cranial nerves II, III, IV, V, VI, and VII. An extensive workup revealed a hemoglobin A1C of 10%. However, he was never diagnosed with diabetes mellitus previously and denied any of the classical symptoms of diabetes mellitus. He underwent ethmoidectomy, maxillectomy, and drainage of an intraorbital abscess after appropriate imaging studies. Histopathology confirmed the diagnosis of mucormycosis, and the patient was started on systemic amphotericin B. This case emphasizes the importance of screening for diabetes mellitus. Early recognition of underlying diabetes mellitus in this patient may have prevented the development of mucormycosis along with its devastating complications. pmcIntroduction Mucormycosis is a rare opportunistic angio-invasive infection caused by fungi of the order Mucorales. It mainly affects individuals with diabetes mellitus and patients who have underlying immunosuppression, with a mortality rate reaching up to 80%. 1 According to reports from the United States, mucormycosis has a very low incidence of 1.7 in every million individuals. Diabetes mellitus is the most common risk factor for the development of mucormycosis across Asia, while hematological malignancies are the major risk factors across Western countries. 2 Although mucormycosis is a severe complication of poorly controlled diabetes mellitus, it is very rare for diabetes mellitus to be diagnosed after the development of mucormycosis. 3 Mucormycosis can involve many organ systems, including the cutaneous, pulmonary, and gastrointestinal systems, and in severe forms, it can spread to other organs. However, rhinocerebral involvement is the most frequent presentation. 4 Involvement of cranial nerve VII along with the widespread involvement of other cranial nerves was rarely reported in the literature. 5 Herein, we report a very unusual case of mucormycosis with extensive rhino-orbit cerebral involvement presenting with multiple cranial nerve palsies as a first presentation of diabetes mellitus in a previously healthy patient. Case Description A 56-year-old man was referred to our institution for a case of facial palsy. He was treated with systemic methylprednisolone for 10 days, without improvement. On examination, he had a visual acuity of 1.0 in the right eye and hand motion in the left. Motility examination revealed limitation of abduction of the right eye. The patient had a frozen orbit with a limitation of all gazes in the left eye. Fundus examination revealed a pale optic disk, afferent pupillary defect, and an ischemic retina, indicating central retinal artery occlusion. Examination findings were also consistent with oculomotor, trochlear, and abducens nerve paresis. Mucormycosis was a result of steroid therapy, as the patient was diagnosed 2 weeks previously with an isolated cranial nerve VII and was started with steroids in a peripheral ophthalmology clinic. At his first presentation in Jordan University Hospital, he had a frozen orbit with cranial nerve III, IV, and VI palsies. The patient was admitted, and the results of baseline investigations were normal, except for a high white blood cell count and hemoglobin A1C (HbA1c) of 10%. Systemic steroids were stopped, and the patient was started on insulin to correct his blood sugar. He was never diagnosed with diabetes mellitus before and denied any symptoms of diabetes mellitus. He was initially diagnosed with diabetes mellitus. He tested negative for ketone bodies and was treated in the ward. Results of other routine investigations, such as kidney function tests and complete blood count, were normal, except for high blood sugar and a high white blood cell count. The differential diagnosis of his current symptoms was cavernous sinus thrombosis, a space-occupying lesion in the cavernous sinus or orbital cellulitis. Computed tomography of the brain and orbit was performed and revealed extensive sinus disease with left complicated ethmoidal and maxillary sinusitis extending into the left orbit with abscess formation. There was also extensive orbital cellulitis and evidence of left cavernous sinus thrombosis secondary to infection (Figure 1). It is possible that the patient had undiagnosed diabetes mellitus as indicated by a high HbA1c and was worsened by steroid therapy The patient was started on systemic vancomycin and ceftriaxone, and a swab from the nasal cavity was sent for evaluation. Nasal swab revealed methicillin-resistant staphylococcus and positive fungal staining. The patient was started on systemic amphotericin B at a loading dose and continued with the therapeutic dose after 2 days of his presentation to our institution. Nasal endoscopy revealed diffuse white filaments and necrosis of the bone from which a biopsy was taken. Sinuses were drained, followed by ethmoidectomy and maxillectomy. A biopsy from the necrotic tissue was sent, which showed the presence of non-septate hyphae branching at right angles consistent with angio-invasive mucormycosis. However, workup to identify the exact causative organism of mucormycosis was not performed as it did not affect further management. The patient was kept on daily irrigation of the sinuses along with administration of systemic amphotericin B. On follow-up magnetic resonance imaging, the patient had an intraorbital collection (Figure 2). Subsequently, drainage of the orbital abscess, tarsorrhaphy, and excision of the necrotic conjunctiva were performed. The patient was kept on systemic amphotericin B and showed clinical improvement. He remained hospitalized until a repeat biopsy of the paranasal sinuses was negative for the fungi. He was discharged with residual paresis of all cranial nerves involved along with left eye ptosis and only hand motion in the left eye. He was followed up serially for 1 year and 4 months. Currently, the patient is afebrile with complete resolution of symptoms in the right eye and partial recovery of cranial nerve palsies. However, in the left eye, residual ptosis has remained and only hand motion is visible due to the previous central retinal artery occlusion related to his initial presentation. Discussion Mucormycosis is an opportunistic infection with a high mortality rate, reaching up to 80% without treatment. 6 Even with early diagnosis and aggressive therapy, its prognosis remains poor. 7 Diabetes mellitus is the most common risk factor for mucormycosis, particularly during ketoacidosis. Ketones induce the fungi to utilize and produce ketoreductase, which facilitates its growth through various mechanisms. 4 Roden et al. reviewed the characteristics of 929 patients with mucormycosis, and of 929 patients, 36% had diabetes mellitus, 17% had an underlying malignancy, and 19% had no underlying conditions. 3 In a review of literature of 851 patients, the median age was 51 ± 12 years, and 531 (63%) were men. Diabetes mellitus was the most common underlying condition (340/851, 40%; 71 (20%) had documented ketoacidosis), followed by hematological malignancy (275/851, 32%; 116 (42%) had acute myeloid leukemia) and solid organ transplantation (116/851, 14%; 67 (58%) had received a kidney transplant). Of the predisposing factors, the use of corticosteroids at the time of presentation was the most common (273/851, 33%), followed by neutropenia (169/851, 20%) and trauma (166/851, 20%). Rhino-orbito-cerebral mucormycosis was the most commonly observed manifestation (288/851, 34%), followed by cutaneous (187/851, 22%) and pulmonary mucormycosis (172/851, 20%. 8 In the present case, the initiation of glucocorticoids as a treatment for facial nerve palsy may have facilitated the initial development of mucormycosis. Glucocorticoids are well known to increase blood glucose levels and can exacerbate hyperglycemia in a patient with diabetes mellitus. 9 Our patient should have been tested for hyperglycemia before the initiation of glucocorticoid therapy (1) to rule out diabetes mellitus as the underlying cause of his facial palsy and (2) to measure the baseline blood sugar before the initiation of steroid therapy. 10 Clinically, mucormycosis is characterized by rhinitis with granular and purulent discharge, nasal ulceration, black spots of infarcted mucosa, and paranasal sinusitis. However, other common presenting symptoms include epistaxis, ophthalmoplegia with blindness, proptosis and orbital cellulitis, hemiplegia or stroke, and decreased mental function. 11 In advanced disease, symptoms include chemosis, ptosis, proptosis, ophthalmoplegia, blindness, and multiple cranial nerve palsies (function of cranial nerves II, III, V, VI, and VI may be lost or impaired).4,5 The classical clinical presentations are facial pain, an irregular black eschar on the palatal or nasal mucosa, and pus discharge from the eye and nose. In a retrospective chart review of 48 patients with 49 cases of acute fulminant-invasive fungal sinusitis over 19 years, mucormycosis and aspergillosis were found in 22 and 27 cases, respectively. Orbital (proptosis, periorbital edema, and ophthalmoplegia) and cranial nerve symptoms were seen at presentation more frequently in mucormycosis (6 [27%] and 9 [41%]) than in aspergillosis cases (3 [11%] and 7 [26%]; p = 0.079). Long-term orbital and cranial nerve sequelae were reported in 16 (72%) mucormycosis cases and 10 (37%) aspergillosis cases (p = 0.0210). These data suggest that the presence of orbital and neurological symptoms at presentation warrants more aggressive surgical intervention because of the likelihood of mucormycosis. 12 In another review of mucormycosis cases, all six patients had uncontrolled diabetes mellitus with fungal-invasive disease and had multiple cranial nerve involvement, including cranial nerves II to VII. 2 However, our patient had not been definitely diagnosed with diabetes mellitus before he had the fungal infection. That cohort chiefly complained of facial swelling, blood-stained nasal discharge, and crusting. All six patients were on prolonged antibiotic treatment with an incomplete response. Facial dysesthesia was present in all sides involved. Six patients had impaired vision, two had vision loss, and four only had a perception of light. These patients improved to finger counting during treatment. All patients had periorbital edema, ptosis, and 7th cranial nerve palsy of the lower motor type, and two patients had irritability and disoriented state and succumbed to the disease within 48 h of treatment. By contrast, our patient was fully oriented and had an initial visual acuity of hand motion with limited improvement. 13 A diagnosis of rhino-orbito-cerebral mucormycosis requires high clinical suspicion, and a definitive diagnosis requires a biopsy indicating the presence of the fungi. Furthermore, imaging is required to identify the degree of local involvement and spread across the rhino-orbital structures. 3 Mucormycosis is treated with surgical debridement of necrotic tissues, use of intravenous antifungal drugs, and elimination of predisposing factors. In retrospective studies, the combination of surgery and antifungal drugs showed higher survival rates than medical treatment alone. 3 Our case emphasizes the importance of screening for diabetes mellitus to prevent such a devastating complication. Our patient denied any of the classical symptoms of diabetes mellitus, and the only risk factor he had for diabetes mellitus was his age. Given the relatively long asymptomatic period of diabetes mellitus, the American Diabetes Association recommends screening for diabetes mellitus at age 45 years. 14 Early detection and treatment of this patient's underlying diabetes mellitus may have prevented the development of mucormycosis, especially in a country like Jordan where the prevalence of diabetes mellitus among men aged >25 years is as high as 33%. 15 The involvement of cranial nerve VII in mucormycosis is very rare, and its pathogenesis is not well understood. 13 We are unsure whether the involvement of cranial nerve VII in this patient was the first manifestation of mucormycosis or it was due to an independent pathology in which its treatment with glucocorticoids facilitated the development of mucormycosis. Furthermore, the widespread involvement of cranial nerves III, IV, and VI is also uncommon and indicates severe disease possibly due to cavernous sinus thrombosis. 3 Conclusions Mucormycosis is a rare and possibly fatal opportunistic infection. It should be suspected in any patient who presents with symptoms of sinusitis along with the widespread involvement of the orbit and cranial nerves. Workup should be initiated for an underlying systemic disease, such as diabetes mellitus, in any patient diagnosed with mucormycosis. In the present case, initiation of steroid therapy may have influenced the development of fungal infection and its progression, and this highlights the importance of screening. It is critical to exclude diabetes mellitus before starting steroid treatment and to screen a susceptible population perhaps at a younger age than recommended. Treatment with systemic antifungals, surgical debridement of necrotic tissues, and correction of underlying hyperglycemia should be initiated as soon as possible in diabetic patients with mucormycosis. Although this patient was 56 years old, screening for diabetes mellitus in younger populations earlier than 45 years is recommended, especially in the presence of opportunistic infections. Ethical statement Ethical approval was obtained from the Jordan University Board. Informed consent was also obtained from the patient. Conflict of interest All authors declare no conflict of interest Funding sources This work was not funded by any funding bodies, and all authors have read and approved the contents of this manuscript. Figure 1. Figure 2.

METHYLPREDNISOLONE

DrugsGivenReaction

CC BY

34888198

2021

Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Sinusitis'.

Mucormycosis with extensive cranial nerve involvement as the first presentation of diabetes mellitus: A case report. Mucormycosis, a rare fungal infection, mainly affects individuals with diabetes mellitus and those who were immunocompromised and has a high mortality rate. Its most common presentation is similar to that of acute bacterial sinusitis with symptoms of nasal congestion, headache, and fever. The involvement of multiple cranial nerves in mucormycosis was rarely reported in the literature and indicates severe disease. Herein, we report the case of a 56-year-old man who was referred to the ophthalmology outpatient clinic for facial nerve palsy. He was treated with systemic steroids for 10 days with no improvement. On examination, he had a loss of vision and a frozen orbit due to involvement of cranial nerves II, III, IV, V, VI, and VII. An extensive workup revealed a hemoglobin A1C of 10%. However, he was never diagnosed with diabetes mellitus previously and denied any of the classical symptoms of diabetes mellitus. He underwent ethmoidectomy, maxillectomy, and drainage of an intraorbital abscess after appropriate imaging studies. Histopathology confirmed the diagnosis of mucormycosis, and the patient was started on systemic amphotericin B. This case emphasizes the importance of screening for diabetes mellitus. Early recognition of underlying diabetes mellitus in this patient may have prevented the development of mucormycosis along with its devastating complications. pmcIntroduction Mucormycosis is a rare opportunistic angio-invasive infection caused by fungi of the order Mucorales. It mainly affects individuals with diabetes mellitus and patients who have underlying immunosuppression, with a mortality rate reaching up to 80%. 1 According to reports from the United States, mucormycosis has a very low incidence of 1.7 in every million individuals. Diabetes mellitus is the most common risk factor for the development of mucormycosis across Asia, while hematological malignancies are the major risk factors across Western countries. 2 Although mucormycosis is a severe complication of poorly controlled diabetes mellitus, it is very rare for diabetes mellitus to be diagnosed after the development of mucormycosis. 3 Mucormycosis can involve many organ systems, including the cutaneous, pulmonary, and gastrointestinal systems, and in severe forms, it can spread to other organs. However, rhinocerebral involvement is the most frequent presentation. 4 Involvement of cranial nerve VII along with the widespread involvement of other cranial nerves was rarely reported in the literature. 5 Herein, we report a very unusual case of mucormycosis with extensive rhino-orbit cerebral involvement presenting with multiple cranial nerve palsies as a first presentation of diabetes mellitus in a previously healthy patient. Case Description A 56-year-old man was referred to our institution for a case of facial palsy. He was treated with systemic methylprednisolone for 10 days, without improvement. On examination, he had a visual acuity of 1.0 in the right eye and hand motion in the left. Motility examination revealed limitation of abduction of the right eye. The patient had a frozen orbit with a limitation of all gazes in the left eye. Fundus examination revealed a pale optic disk, afferent pupillary defect, and an ischemic retina, indicating central retinal artery occlusion. Examination findings were also consistent with oculomotor, trochlear, and abducens nerve paresis. Mucormycosis was a result of steroid therapy, as the patient was diagnosed 2 weeks previously with an isolated cranial nerve VII and was started with steroids in a peripheral ophthalmology clinic. At his first presentation in Jordan University Hospital, he had a frozen orbit with cranial nerve III, IV, and VI palsies. The patient was admitted, and the results of baseline investigations were normal, except for a high white blood cell count and hemoglobin A1C (HbA1c) of 10%. Systemic steroids were stopped, and the patient was started on insulin to correct his blood sugar. He was never diagnosed with diabetes mellitus before and denied any symptoms of diabetes mellitus. He was initially diagnosed with diabetes mellitus. He tested negative for ketone bodies and was treated in the ward. Results of other routine investigations, such as kidney function tests and complete blood count, were normal, except for high blood sugar and a high white blood cell count. The differential diagnosis of his current symptoms was cavernous sinus thrombosis, a space-occupying lesion in the cavernous sinus or orbital cellulitis. Computed tomography of the brain and orbit was performed and revealed extensive sinus disease with left complicated ethmoidal and maxillary sinusitis extending into the left orbit with abscess formation. There was also extensive orbital cellulitis and evidence of left cavernous sinus thrombosis secondary to infection (Figure 1). It is possible that the patient had undiagnosed diabetes mellitus as indicated by a high HbA1c and was worsened by steroid therapy The patient was started on systemic vancomycin and ceftriaxone, and a swab from the nasal cavity was sent for evaluation. Nasal swab revealed methicillin-resistant staphylococcus and positive fungal staining. The patient was started on systemic amphotericin B at a loading dose and continued with the therapeutic dose after 2 days of his presentation to our institution. Nasal endoscopy revealed diffuse white filaments and necrosis of the bone from which a biopsy was taken. Sinuses were drained, followed by ethmoidectomy and maxillectomy. A biopsy from the necrotic tissue was sent, which showed the presence of non-septate hyphae branching at right angles consistent with angio-invasive mucormycosis. However, workup to identify the exact causative organism of mucormycosis was not performed as it did not affect further management. The patient was kept on daily irrigation of the sinuses along with administration of systemic amphotericin B. On follow-up magnetic resonance imaging, the patient had an intraorbital collection (Figure 2). Subsequently, drainage of the orbital abscess, tarsorrhaphy, and excision of the necrotic conjunctiva were performed. The patient was kept on systemic amphotericin B and showed clinical improvement. He remained hospitalized until a repeat biopsy of the paranasal sinuses was negative for the fungi. He was discharged with residual paresis of all cranial nerves involved along with left eye ptosis and only hand motion in the left eye. He was followed up serially for 1 year and 4 months. Currently, the patient is afebrile with complete resolution of symptoms in the right eye and partial recovery of cranial nerve palsies. However, in the left eye, residual ptosis has remained and only hand motion is visible due to the previous central retinal artery occlusion related to his initial presentation. Discussion Mucormycosis is an opportunistic infection with a high mortality rate, reaching up to 80% without treatment. 6 Even with early diagnosis and aggressive therapy, its prognosis remains poor. 7 Diabetes mellitus is the most common risk factor for mucormycosis, particularly during ketoacidosis. Ketones induce the fungi to utilize and produce ketoreductase, which facilitates its growth through various mechanisms. 4 Roden et al. reviewed the characteristics of 929 patients with mucormycosis, and of 929 patients, 36% had diabetes mellitus, 17% had an underlying malignancy, and 19% had no underlying conditions. 3 In a review of literature of 851 patients, the median age was 51 ± 12 years, and 531 (63%) were men. Diabetes mellitus was the most common underlying condition (340/851, 40%; 71 (20%) had documented ketoacidosis), followed by hematological malignancy (275/851, 32%; 116 (42%) had acute myeloid leukemia) and solid organ transplantation (116/851, 14%; 67 (58%) had received a kidney transplant). Of the predisposing factors, the use of corticosteroids at the time of presentation was the most common (273/851, 33%), followed by neutropenia (169/851, 20%) and trauma (166/851, 20%). Rhino-orbito-cerebral mucormycosis was the most commonly observed manifestation (288/851, 34%), followed by cutaneous (187/851, 22%) and pulmonary mucormycosis (172/851, 20%. 8 In the present case, the initiation of glucocorticoids as a treatment for facial nerve palsy may have facilitated the initial development of mucormycosis. Glucocorticoids are well known to increase blood glucose levels and can exacerbate hyperglycemia in a patient with diabetes mellitus. 9 Our patient should have been tested for hyperglycemia before the initiation of glucocorticoid therapy (1) to rule out diabetes mellitus as the underlying cause of his facial palsy and (2) to measure the baseline blood sugar before the initiation of steroid therapy. 10 Clinically, mucormycosis is characterized by rhinitis with granular and purulent discharge, nasal ulceration, black spots of infarcted mucosa, and paranasal sinusitis. However, other common presenting symptoms include epistaxis, ophthalmoplegia with blindness, proptosis and orbital cellulitis, hemiplegia or stroke, and decreased mental function. 11 In advanced disease, symptoms include chemosis, ptosis, proptosis, ophthalmoplegia, blindness, and multiple cranial nerve palsies (function of cranial nerves II, III, V, VI, and VI may be lost or impaired).4,5 The classical clinical presentations are facial pain, an irregular black eschar on the palatal or nasal mucosa, and pus discharge from the eye and nose. In a retrospective chart review of 48 patients with 49 cases of acute fulminant-invasive fungal sinusitis over 19 years, mucormycosis and aspergillosis were found in 22 and 27 cases, respectively. Orbital (proptosis, periorbital edema, and ophthalmoplegia) and cranial nerve symptoms were seen at presentation more frequently in mucormycosis (6 [27%] and 9 [41%]) than in aspergillosis cases (3 [11%] and 7 [26%]; p = 0.079). Long-term orbital and cranial nerve sequelae were reported in 16 (72%) mucormycosis cases and 10 (37%) aspergillosis cases (p = 0.0210). These data suggest that the presence of orbital and neurological symptoms at presentation warrants more aggressive surgical intervention because of the likelihood of mucormycosis. 12 In another review of mucormycosis cases, all six patients had uncontrolled diabetes mellitus with fungal-invasive disease and had multiple cranial nerve involvement, including cranial nerves II to VII. 2 However, our patient had not been definitely diagnosed with diabetes mellitus before he had the fungal infection. That cohort chiefly complained of facial swelling, blood-stained nasal discharge, and crusting. All six patients were on prolonged antibiotic treatment with an incomplete response. Facial dysesthesia was present in all sides involved. Six patients had impaired vision, two had vision loss, and four only had a perception of light. These patients improved to finger counting during treatment. All patients had periorbital edema, ptosis, and 7th cranial nerve palsy of the lower motor type, and two patients had irritability and disoriented state and succumbed to the disease within 48 h of treatment. By contrast, our patient was fully oriented and had an initial visual acuity of hand motion with limited improvement. 13 A diagnosis of rhino-orbito-cerebral mucormycosis requires high clinical suspicion, and a definitive diagnosis requires a biopsy indicating the presence of the fungi. Furthermore, imaging is required to identify the degree of local involvement and spread across the rhino-orbital structures. 3 Mucormycosis is treated with surgical debridement of necrotic tissues, use of intravenous antifungal drugs, and elimination of predisposing factors. In retrospective studies, the combination of surgery and antifungal drugs showed higher survival rates than medical treatment alone. 3 Our case emphasizes the importance of screening for diabetes mellitus to prevent such a devastating complication. Our patient denied any of the classical symptoms of diabetes mellitus, and the only risk factor he had for diabetes mellitus was his age. Given the relatively long asymptomatic period of diabetes mellitus, the American Diabetes Association recommends screening for diabetes mellitus at age 45 years. 14 Early detection and treatment of this patient's underlying diabetes mellitus may have prevented the development of mucormycosis, especially in a country like Jordan where the prevalence of diabetes mellitus among men aged >25 years is as high as 33%. 15 The involvement of cranial nerve VII in mucormycosis is very rare, and its pathogenesis is not well understood. 13 We are unsure whether the involvement of cranial nerve VII in this patient was the first manifestation of mucormycosis or it was due to an independent pathology in which its treatment with glucocorticoids facilitated the development of mucormycosis. Furthermore, the widespread involvement of cranial nerves III, IV, and VI is also uncommon and indicates severe disease possibly due to cavernous sinus thrombosis. 3 Conclusions Mucormycosis is a rare and possibly fatal opportunistic infection. It should be suspected in any patient who presents with symptoms of sinusitis along with the widespread involvement of the orbit and cranial nerves. Workup should be initiated for an underlying systemic disease, such as diabetes mellitus, in any patient diagnosed with mucormycosis. In the present case, initiation of steroid therapy may have influenced the development of fungal infection and its progression, and this highlights the importance of screening. It is critical to exclude diabetes mellitus before starting steroid treatment and to screen a susceptible population perhaps at a younger age than recommended. Treatment with systemic antifungals, surgical debridement of necrotic tissues, and correction of underlying hyperglycemia should be initiated as soon as possible in diabetic patients with mucormycosis. Although this patient was 56 years old, screening for diabetes mellitus in younger populations earlier than 45 years is recommended, especially in the presence of opportunistic infections. Ethical statement Ethical approval was obtained from the Jordan University Board. Informed consent was also obtained from the patient. Conflict of interest All authors declare no conflict of interest Funding sources This work was not funded by any funding bodies, and all authors have read and approved the contents of this manuscript. Figure 1. Figure 2.

METHYLPREDNISOLONE

DrugsGivenReaction

CC BY

34888198

2021

What was the dosage of drug 'METHYLPREDNISOLONE'?

Mucormycosis with extensive cranial nerve involvement as the first presentation of diabetes mellitus: A case report. Mucormycosis, a rare fungal infection, mainly affects individuals with diabetes mellitus and those who were immunocompromised and has a high mortality rate. Its most common presentation is similar to that of acute bacterial sinusitis with symptoms of nasal congestion, headache, and fever. The involvement of multiple cranial nerves in mucormycosis was rarely reported in the literature and indicates severe disease. Herein, we report the case of a 56-year-old man who was referred to the ophthalmology outpatient clinic for facial nerve palsy. He was treated with systemic steroids for 10 days with no improvement. On examination, he had a loss of vision and a frozen orbit due to involvement of cranial nerves II, III, IV, V, VI, and VII. An extensive workup revealed a hemoglobin A1C of 10%. However, he was never diagnosed with diabetes mellitus previously and denied any of the classical symptoms of diabetes mellitus. He underwent ethmoidectomy, maxillectomy, and drainage of an intraorbital abscess after appropriate imaging studies. Histopathology confirmed the diagnosis of mucormycosis, and the patient was started on systemic amphotericin B. This case emphasizes the importance of screening for diabetes mellitus. Early recognition of underlying diabetes mellitus in this patient may have prevented the development of mucormycosis along with its devastating complications. pmcIntroduction Mucormycosis is a rare opportunistic angio-invasive infection caused by fungi of the order Mucorales. It mainly affects individuals with diabetes mellitus and patients who have underlying immunosuppression, with a mortality rate reaching up to 80%. 1 According to reports from the United States, mucormycosis has a very low incidence of 1.7 in every million individuals. Diabetes mellitus is the most common risk factor for the development of mucormycosis across Asia, while hematological malignancies are the major risk factors across Western countries. 2 Although mucormycosis is a severe complication of poorly controlled diabetes mellitus, it is very rare for diabetes mellitus to be diagnosed after the development of mucormycosis. 3 Mucormycosis can involve many organ systems, including the cutaneous, pulmonary, and gastrointestinal systems, and in severe forms, it can spread to other organs. However, rhinocerebral involvement is the most frequent presentation. 4 Involvement of cranial nerve VII along with the widespread involvement of other cranial nerves was rarely reported in the literature. 5 Herein, we report a very unusual case of mucormycosis with extensive rhino-orbit cerebral involvement presenting with multiple cranial nerve palsies as a first presentation of diabetes mellitus in a previously healthy patient. Case Description A 56-year-old man was referred to our institution for a case of facial palsy. He was treated with systemic methylprednisolone for 10 days, without improvement. On examination, he had a visual acuity of 1.0 in the right eye and hand motion in the left. Motility examination revealed limitation of abduction of the right eye. The patient had a frozen orbit with a limitation of all gazes in the left eye. Fundus examination revealed a pale optic disk, afferent pupillary defect, and an ischemic retina, indicating central retinal artery occlusion. Examination findings were also consistent with oculomotor, trochlear, and abducens nerve paresis. Mucormycosis was a result of steroid therapy, as the patient was diagnosed 2 weeks previously with an isolated cranial nerve VII and was started with steroids in a peripheral ophthalmology clinic. At his first presentation in Jordan University Hospital, he had a frozen orbit with cranial nerve III, IV, and VI palsies. The patient was admitted, and the results of baseline investigations were normal, except for a high white blood cell count and hemoglobin A1C (HbA1c) of 10%. Systemic steroids were stopped, and the patient was started on insulin to correct his blood sugar. He was never diagnosed with diabetes mellitus before and denied any symptoms of diabetes mellitus. He was initially diagnosed with diabetes mellitus. He tested negative for ketone bodies and was treated in the ward. Results of other routine investigations, such as kidney function tests and complete blood count, were normal, except for high blood sugar and a high white blood cell count. The differential diagnosis of his current symptoms was cavernous sinus thrombosis, a space-occupying lesion in the cavernous sinus or orbital cellulitis. Computed tomography of the brain and orbit was performed and revealed extensive sinus disease with left complicated ethmoidal and maxillary sinusitis extending into the left orbit with abscess formation. There was also extensive orbital cellulitis and evidence of left cavernous sinus thrombosis secondary to infection (Figure 1). It is possible that the patient had undiagnosed diabetes mellitus as indicated by a high HbA1c and was worsened by steroid therapy The patient was started on systemic vancomycin and ceftriaxone, and a swab from the nasal cavity was sent for evaluation. Nasal swab revealed methicillin-resistant staphylococcus and positive fungal staining. The patient was started on systemic amphotericin B at a loading dose and continued with the therapeutic dose after 2 days of his presentation to our institution. Nasal endoscopy revealed diffuse white filaments and necrosis of the bone from which a biopsy was taken. Sinuses were drained, followed by ethmoidectomy and maxillectomy. A biopsy from the necrotic tissue was sent, which showed the presence of non-septate hyphae branching at right angles consistent with angio-invasive mucormycosis. However, workup to identify the exact causative organism of mucormycosis was not performed as it did not affect further management. The patient was kept on daily irrigation of the sinuses along with administration of systemic amphotericin B. On follow-up magnetic resonance imaging, the patient had an intraorbital collection (Figure 2). Subsequently, drainage of the orbital abscess, tarsorrhaphy, and excision of the necrotic conjunctiva were performed. The patient was kept on systemic amphotericin B and showed clinical improvement. He remained hospitalized until a repeat biopsy of the paranasal sinuses was negative for the fungi. He was discharged with residual paresis of all cranial nerves involved along with left eye ptosis and only hand motion in the left eye. He was followed up serially for 1 year and 4 months. Currently, the patient is afebrile with complete resolution of symptoms in the right eye and partial recovery of cranial nerve palsies. However, in the left eye, residual ptosis has remained and only hand motion is visible due to the previous central retinal artery occlusion related to his initial presentation. Discussion Mucormycosis is an opportunistic infection with a high mortality rate, reaching up to 80% without treatment. 6 Even with early diagnosis and aggressive therapy, its prognosis remains poor. 7 Diabetes mellitus is the most common risk factor for mucormycosis, particularly during ketoacidosis. Ketones induce the fungi to utilize and produce ketoreductase, which facilitates its growth through various mechanisms. 4 Roden et al. reviewed the characteristics of 929 patients with mucormycosis, and of 929 patients, 36% had diabetes mellitus, 17% had an underlying malignancy, and 19% had no underlying conditions. 3 In a review of literature of 851 patients, the median age was 51 ± 12 years, and 531 (63%) were men. Diabetes mellitus was the most common underlying condition (340/851, 40%; 71 (20%) had documented ketoacidosis), followed by hematological malignancy (275/851, 32%; 116 (42%) had acute myeloid leukemia) and solid organ transplantation (116/851, 14%; 67 (58%) had received a kidney transplant). Of the predisposing factors, the use of corticosteroids at the time of presentation was the most common (273/851, 33%), followed by neutropenia (169/851, 20%) and trauma (166/851, 20%). Rhino-orbito-cerebral mucormycosis was the most commonly observed manifestation (288/851, 34%), followed by cutaneous (187/851, 22%) and pulmonary mucormycosis (172/851, 20%. 8 In the present case, the initiation of glucocorticoids as a treatment for facial nerve palsy may have facilitated the initial development of mucormycosis. Glucocorticoids are well known to increase blood glucose levels and can exacerbate hyperglycemia in a patient with diabetes mellitus. 9 Our patient should have been tested for hyperglycemia before the initiation of glucocorticoid therapy (1) to rule out diabetes mellitus as the underlying cause of his facial palsy and (2) to measure the baseline blood sugar before the initiation of steroid therapy. 10 Clinically, mucormycosis is characterized by rhinitis with granular and purulent discharge, nasal ulceration, black spots of infarcted mucosa, and paranasal sinusitis. However, other common presenting symptoms include epistaxis, ophthalmoplegia with blindness, proptosis and orbital cellulitis, hemiplegia or stroke, and decreased mental function. 11 In advanced disease, symptoms include chemosis, ptosis, proptosis, ophthalmoplegia, blindness, and multiple cranial nerve palsies (function of cranial nerves II, III, V, VI, and VI may be lost or impaired).4,5 The classical clinical presentations are facial pain, an irregular black eschar on the palatal or nasal mucosa, and pus discharge from the eye and nose. In a retrospective chart review of 48 patients with 49 cases of acute fulminant-invasive fungal sinusitis over 19 years, mucormycosis and aspergillosis were found in 22 and 27 cases, respectively. Orbital (proptosis, periorbital edema, and ophthalmoplegia) and cranial nerve symptoms were seen at presentation more frequently in mucormycosis (6 [27%] and 9 [41%]) than in aspergillosis cases (3 [11%] and 7 [26%]; p = 0.079). Long-term orbital and cranial nerve sequelae were reported in 16 (72%) mucormycosis cases and 10 (37%) aspergillosis cases (p = 0.0210). These data suggest that the presence of orbital and neurological symptoms at presentation warrants more aggressive surgical intervention because of the likelihood of mucormycosis. 12 In another review of mucormycosis cases, all six patients had uncontrolled diabetes mellitus with fungal-invasive disease and had multiple cranial nerve involvement, including cranial nerves II to VII. 2 However, our patient had not been definitely diagnosed with diabetes mellitus before he had the fungal infection. That cohort chiefly complained of facial swelling, blood-stained nasal discharge, and crusting. All six patients were on prolonged antibiotic treatment with an incomplete response. Facial dysesthesia was present in all sides involved. Six patients had impaired vision, two had vision loss, and four only had a perception of light. These patients improved to finger counting during treatment. All patients had periorbital edema, ptosis, and 7th cranial nerve palsy of the lower motor type, and two patients had irritability and disoriented state and succumbed to the disease within 48 h of treatment. By contrast, our patient was fully oriented and had an initial visual acuity of hand motion with limited improvement. 13 A diagnosis of rhino-orbito-cerebral mucormycosis requires high clinical suspicion, and a definitive diagnosis requires a biopsy indicating the presence of the fungi. Furthermore, imaging is required to identify the degree of local involvement and spread across the rhino-orbital structures. 3 Mucormycosis is treated with surgical debridement of necrotic tissues, use of intravenous antifungal drugs, and elimination of predisposing factors. In retrospective studies, the combination of surgery and antifungal drugs showed higher survival rates than medical treatment alone. 3 Our case emphasizes the importance of screening for diabetes mellitus to prevent such a devastating complication. Our patient denied any of the classical symptoms of diabetes mellitus, and the only risk factor he had for diabetes mellitus was his age. Given the relatively long asymptomatic period of diabetes mellitus, the American Diabetes Association recommends screening for diabetes mellitus at age 45 years. 14 Early detection and treatment of this patient's underlying diabetes mellitus may have prevented the development of mucormycosis, especially in a country like Jordan where the prevalence of diabetes mellitus among men aged >25 years is as high as 33%. 15 The involvement of cranial nerve VII in mucormycosis is very rare, and its pathogenesis is not well understood. 13 We are unsure whether the involvement of cranial nerve VII in this patient was the first manifestation of mucormycosis or it was due to an independent pathology in which its treatment with glucocorticoids facilitated the development of mucormycosis. Furthermore, the widespread involvement of cranial nerves III, IV, and VI is also uncommon and indicates severe disease possibly due to cavernous sinus thrombosis. 3 Conclusions Mucormycosis is a rare and possibly fatal opportunistic infection. It should be suspected in any patient who presents with symptoms of sinusitis along with the widespread involvement of the orbit and cranial nerves. Workup should be initiated for an underlying systemic disease, such as diabetes mellitus, in any patient diagnosed with mucormycosis. In the present case, initiation of steroid therapy may have influenced the development of fungal infection and its progression, and this highlights the importance of screening. It is critical to exclude diabetes mellitus before starting steroid treatment and to screen a susceptible population perhaps at a younger age than recommended. Treatment with systemic antifungals, surgical debridement of necrotic tissues, and correction of underlying hyperglycemia should be initiated as soon as possible in diabetic patients with mucormycosis. Although this patient was 56 years old, screening for diabetes mellitus in younger populations earlier than 45 years is recommended, especially in the presence of opportunistic infections. Ethical statement Ethical approval was obtained from the Jordan University Board. Informed consent was also obtained from the patient. Conflict of interest All authors declare no conflict of interest Funding sources This work was not funded by any funding bodies, and all authors have read and approved the contents of this manuscript. Figure 1. Figure 2.

UNK (SYSTEMIC)

DrugDosageText

CC BY

34888198

2021

What was the outcome of reaction 'Cellulitis orbital'?

Mucormycosis with extensive cranial nerve involvement as the first presentation of diabetes mellitus: A case report. Mucormycosis, a rare fungal infection, mainly affects individuals with diabetes mellitus and those who were immunocompromised and has a high mortality rate. Its most common presentation is similar to that of acute bacterial sinusitis with symptoms of nasal congestion, headache, and fever. The involvement of multiple cranial nerves in mucormycosis was rarely reported in the literature and indicates severe disease. Herein, we report the case of a 56-year-old man who was referred to the ophthalmology outpatient clinic for facial nerve palsy. He was treated with systemic steroids for 10 days with no improvement. On examination, he had a loss of vision and a frozen orbit due to involvement of cranial nerves II, III, IV, V, VI, and VII. An extensive workup revealed a hemoglobin A1C of 10%. However, he was never diagnosed with diabetes mellitus previously and denied any of the classical symptoms of diabetes mellitus. He underwent ethmoidectomy, maxillectomy, and drainage of an intraorbital abscess after appropriate imaging studies. Histopathology confirmed the diagnosis of mucormycosis, and the patient was started on systemic amphotericin B. This case emphasizes the importance of screening for diabetes mellitus. Early recognition of underlying diabetes mellitus in this patient may have prevented the development of mucormycosis along with its devastating complications. pmcIntroduction Mucormycosis is a rare opportunistic angio-invasive infection caused by fungi of the order Mucorales. It mainly affects individuals with diabetes mellitus and patients who have underlying immunosuppression, with a mortality rate reaching up to 80%. 1 According to reports from the United States, mucormycosis has a very low incidence of 1.7 in every million individuals. Diabetes mellitus is the most common risk factor for the development of mucormycosis across Asia, while hematological malignancies are the major risk factors across Western countries. 2 Although mucormycosis is a severe complication of poorly controlled diabetes mellitus, it is very rare for diabetes mellitus to be diagnosed after the development of mucormycosis. 3 Mucormycosis can involve many organ systems, including the cutaneous, pulmonary, and gastrointestinal systems, and in severe forms, it can spread to other organs. However, rhinocerebral involvement is the most frequent presentation. 4 Involvement of cranial nerve VII along with the widespread involvement of other cranial nerves was rarely reported in the literature. 5 Herein, we report a very unusual case of mucormycosis with extensive rhino-orbit cerebral involvement presenting with multiple cranial nerve palsies as a first presentation of diabetes mellitus in a previously healthy patient. Case Description A 56-year-old man was referred to our institution for a case of facial palsy. He was treated with systemic methylprednisolone for 10 days, without improvement. On examination, he had a visual acuity of 1.0 in the right eye and hand motion in the left. Motility examination revealed limitation of abduction of the right eye. The patient had a frozen orbit with a limitation of all gazes in the left eye. Fundus examination revealed a pale optic disk, afferent pupillary defect, and an ischemic retina, indicating central retinal artery occlusion. Examination findings were also consistent with oculomotor, trochlear, and abducens nerve paresis. Mucormycosis was a result of steroid therapy, as the patient was diagnosed 2 weeks previously with an isolated cranial nerve VII and was started with steroids in a peripheral ophthalmology clinic. At his first presentation in Jordan University Hospital, he had a frozen orbit with cranial nerve III, IV, and VI palsies. The patient was admitted, and the results of baseline investigations were normal, except for a high white blood cell count and hemoglobin A1C (HbA1c) of 10%. Systemic steroids were stopped, and the patient was started on insulin to correct his blood sugar. He was never diagnosed with diabetes mellitus before and denied any symptoms of diabetes mellitus. He was initially diagnosed with diabetes mellitus. He tested negative for ketone bodies and was treated in the ward. Results of other routine investigations, such as kidney function tests and complete blood count, were normal, except for high blood sugar and a high white blood cell count. The differential diagnosis of his current symptoms was cavernous sinus thrombosis, a space-occupying lesion in the cavernous sinus or orbital cellulitis. Computed tomography of the brain and orbit was performed and revealed extensive sinus disease with left complicated ethmoidal and maxillary sinusitis extending into the left orbit with abscess formation. There was also extensive orbital cellulitis and evidence of left cavernous sinus thrombosis secondary to infection (Figure 1). It is possible that the patient had undiagnosed diabetes mellitus as indicated by a high HbA1c and was worsened by steroid therapy The patient was started on systemic vancomycin and ceftriaxone, and a swab from the nasal cavity was sent for evaluation. Nasal swab revealed methicillin-resistant staphylococcus and positive fungal staining. The patient was started on systemic amphotericin B at a loading dose and continued with the therapeutic dose after 2 days of his presentation to our institution. Nasal endoscopy revealed diffuse white filaments and necrosis of the bone from which a biopsy was taken. Sinuses were drained, followed by ethmoidectomy and maxillectomy. A biopsy from the necrotic tissue was sent, which showed the presence of non-septate hyphae branching at right angles consistent with angio-invasive mucormycosis. However, workup to identify the exact causative organism of mucormycosis was not performed as it did not affect further management. The patient was kept on daily irrigation of the sinuses along with administration of systemic amphotericin B. On follow-up magnetic resonance imaging, the patient had an intraorbital collection (Figure 2). Subsequently, drainage of the orbital abscess, tarsorrhaphy, and excision of the necrotic conjunctiva were performed. The patient was kept on systemic amphotericin B and showed clinical improvement. He remained hospitalized until a repeat biopsy of the paranasal sinuses was negative for the fungi. He was discharged with residual paresis of all cranial nerves involved along with left eye ptosis and only hand motion in the left eye. He was followed up serially for 1 year and 4 months. Currently, the patient is afebrile with complete resolution of symptoms in the right eye and partial recovery of cranial nerve palsies. However, in the left eye, residual ptosis has remained and only hand motion is visible due to the previous central retinal artery occlusion related to his initial presentation. Discussion Mucormycosis is an opportunistic infection with a high mortality rate, reaching up to 80% without treatment. 6 Even with early diagnosis and aggressive therapy, its prognosis remains poor. 7 Diabetes mellitus is the most common risk factor for mucormycosis, particularly during ketoacidosis. Ketones induce the fungi to utilize and produce ketoreductase, which facilitates its growth through various mechanisms. 4 Roden et al. reviewed the characteristics of 929 patients with mucormycosis, and of 929 patients, 36% had diabetes mellitus, 17% had an underlying malignancy, and 19% had no underlying conditions. 3 In a review of literature of 851 patients, the median age was 51 ± 12 years, and 531 (63%) were men. Diabetes mellitus was the most common underlying condition (340/851, 40%; 71 (20%) had documented ketoacidosis), followed by hematological malignancy (275/851, 32%; 116 (42%) had acute myeloid leukemia) and solid organ transplantation (116/851, 14%; 67 (58%) had received a kidney transplant). Of the predisposing factors, the use of corticosteroids at the time of presentation was the most common (273/851, 33%), followed by neutropenia (169/851, 20%) and trauma (166/851, 20%). Rhino-orbito-cerebral mucormycosis was the most commonly observed manifestation (288/851, 34%), followed by cutaneous (187/851, 22%) and pulmonary mucormycosis (172/851, 20%. 8 In the present case, the initiation of glucocorticoids as a treatment for facial nerve palsy may have facilitated the initial development of mucormycosis. Glucocorticoids are well known to increase blood glucose levels and can exacerbate hyperglycemia in a patient with diabetes mellitus. 9 Our patient should have been tested for hyperglycemia before the initiation of glucocorticoid therapy (1) to rule out diabetes mellitus as the underlying cause of his facial palsy and (2) to measure the baseline blood sugar before the initiation of steroid therapy. 10 Clinically, mucormycosis is characterized by rhinitis with granular and purulent discharge, nasal ulceration, black spots of infarcted mucosa, and paranasal sinusitis. However, other common presenting symptoms include epistaxis, ophthalmoplegia with blindness, proptosis and orbital cellulitis, hemiplegia or stroke, and decreased mental function. 11 In advanced disease, symptoms include chemosis, ptosis, proptosis, ophthalmoplegia, blindness, and multiple cranial nerve palsies (function of cranial nerves II, III, V, VI, and VI may be lost or impaired).4,5 The classical clinical presentations are facial pain, an irregular black eschar on the palatal or nasal mucosa, and pus discharge from the eye and nose. In a retrospective chart review of 48 patients with 49 cases of acute fulminant-invasive fungal sinusitis over 19 years, mucormycosis and aspergillosis were found in 22 and 27 cases, respectively. Orbital (proptosis, periorbital edema, and ophthalmoplegia) and cranial nerve symptoms were seen at presentation more frequently in mucormycosis (6 [27%] and 9 [41%]) than in aspergillosis cases (3 [11%] and 7 [26%]; p = 0.079). Long-term orbital and cranial nerve sequelae were reported in 16 (72%) mucormycosis cases and 10 (37%) aspergillosis cases (p = 0.0210). These data suggest that the presence of orbital and neurological symptoms at presentation warrants more aggressive surgical intervention because of the likelihood of mucormycosis. 12 In another review of mucormycosis cases, all six patients had uncontrolled diabetes mellitus with fungal-invasive disease and had multiple cranial nerve involvement, including cranial nerves II to VII. 2 However, our patient had not been definitely diagnosed with diabetes mellitus before he had the fungal infection. That cohort chiefly complained of facial swelling, blood-stained nasal discharge, and crusting. All six patients were on prolonged antibiotic treatment with an incomplete response. Facial dysesthesia was present in all sides involved. Six patients had impaired vision, two had vision loss, and four only had a perception of light. These patients improved to finger counting during treatment. All patients had periorbital edema, ptosis, and 7th cranial nerve palsy of the lower motor type, and two patients had irritability and disoriented state and succumbed to the disease within 48 h of treatment. By contrast, our patient was fully oriented and had an initial visual acuity of hand motion with limited improvement. 13 A diagnosis of rhino-orbito-cerebral mucormycosis requires high clinical suspicion, and a definitive diagnosis requires a biopsy indicating the presence of the fungi. Furthermore, imaging is required to identify the degree of local involvement and spread across the rhino-orbital structures. 3 Mucormycosis is treated with surgical debridement of necrotic tissues, use of intravenous antifungal drugs, and elimination of predisposing factors. In retrospective studies, the combination of surgery and antifungal drugs showed higher survival rates than medical treatment alone. 3 Our case emphasizes the importance of screening for diabetes mellitus to prevent such a devastating complication. Our patient denied any of the classical symptoms of diabetes mellitus, and the only risk factor he had for diabetes mellitus was his age. Given the relatively long asymptomatic period of diabetes mellitus, the American Diabetes Association recommends screening for diabetes mellitus at age 45 years. 14 Early detection and treatment of this patient's underlying diabetes mellitus may have prevented the development of mucormycosis, especially in a country like Jordan where the prevalence of diabetes mellitus among men aged >25 years is as high as 33%. 15 The involvement of cranial nerve VII in mucormycosis is very rare, and its pathogenesis is not well understood. 13 We are unsure whether the involvement of cranial nerve VII in this patient was the first manifestation of mucormycosis or it was due to an independent pathology in which its treatment with glucocorticoids facilitated the development of mucormycosis. Furthermore, the widespread involvement of cranial nerves III, IV, and VI is also uncommon and indicates severe disease possibly due to cavernous sinus thrombosis. 3 Conclusions Mucormycosis is a rare and possibly fatal opportunistic infection. It should be suspected in any patient who presents with symptoms of sinusitis along with the widespread involvement of the orbit and cranial nerves. Workup should be initiated for an underlying systemic disease, such as diabetes mellitus, in any patient diagnosed with mucormycosis. In the present case, initiation of steroid therapy may have influenced the development of fungal infection and its progression, and this highlights the importance of screening. It is critical to exclude diabetes mellitus before starting steroid treatment and to screen a susceptible population perhaps at a younger age than recommended. Treatment with systemic antifungals, surgical debridement of necrotic tissues, and correction of underlying hyperglycemia should be initiated as soon as possible in diabetic patients with mucormycosis. Although this patient was 56 years old, screening for diabetes mellitus in younger populations earlier than 45 years is recommended, especially in the presence of opportunistic infections. Ethical statement Ethical approval was obtained from the Jordan University Board. Informed consent was also obtained from the patient. Conflict of interest All authors declare no conflict of interest Funding sources This work was not funded by any funding bodies, and all authors have read and approved the contents of this manuscript. Figure 1. Figure 2.

Recovering

ReactionOutcome

CC BY

34888198

2021

What was the outcome of reaction 'Mucormycosis'?

Mucormycosis with extensive cranial nerve involvement as the first presentation of diabetes mellitus: A case report. Mucormycosis, a rare fungal infection, mainly affects individuals with diabetes mellitus and those who were immunocompromised and has a high mortality rate. Its most common presentation is similar to that of acute bacterial sinusitis with symptoms of nasal congestion, headache, and fever. The involvement of multiple cranial nerves in mucormycosis was rarely reported in the literature and indicates severe disease. Herein, we report the case of a 56-year-old man who was referred to the ophthalmology outpatient clinic for facial nerve palsy. He was treated with systemic steroids for 10 days with no improvement. On examination, he had a loss of vision and a frozen orbit due to involvement of cranial nerves II, III, IV, V, VI, and VII. An extensive workup revealed a hemoglobin A1C of 10%. However, he was never diagnosed with diabetes mellitus previously and denied any of the classical symptoms of diabetes mellitus. He underwent ethmoidectomy, maxillectomy, and drainage of an intraorbital abscess after appropriate imaging studies. Histopathology confirmed the diagnosis of mucormycosis, and the patient was started on systemic amphotericin B. This case emphasizes the importance of screening for diabetes mellitus. Early recognition of underlying diabetes mellitus in this patient may have prevented the development of mucormycosis along with its devastating complications. pmcIntroduction Mucormycosis is a rare opportunistic angio-invasive infection caused by fungi of the order Mucorales. It mainly affects individuals with diabetes mellitus and patients who have underlying immunosuppression, with a mortality rate reaching up to 80%. 1 According to reports from the United States, mucormycosis has a very low incidence of 1.7 in every million individuals. Diabetes mellitus is the most common risk factor for the development of mucormycosis across Asia, while hematological malignancies are the major risk factors across Western countries. 2 Although mucormycosis is a severe complication of poorly controlled diabetes mellitus, it is very rare for diabetes mellitus to be diagnosed after the development of mucormycosis. 3 Mucormycosis can involve many organ systems, including the cutaneous, pulmonary, and gastrointestinal systems, and in severe forms, it can spread to other organs. However, rhinocerebral involvement is the most frequent presentation. 4 Involvement of cranial nerve VII along with the widespread involvement of other cranial nerves was rarely reported in the literature. 5 Herein, we report a very unusual case of mucormycosis with extensive rhino-orbit cerebral involvement presenting with multiple cranial nerve palsies as a first presentation of diabetes mellitus in a previously healthy patient. Case Description A 56-year-old man was referred to our institution for a case of facial palsy. He was treated with systemic methylprednisolone for 10 days, without improvement. On examination, he had a visual acuity of 1.0 in the right eye and hand motion in the left. Motility examination revealed limitation of abduction of the right eye. The patient had a frozen orbit with a limitation of all gazes in the left eye. Fundus examination revealed a pale optic disk, afferent pupillary defect, and an ischemic retina, indicating central retinal artery occlusion. Examination findings were also consistent with oculomotor, trochlear, and abducens nerve paresis. Mucormycosis was a result of steroid therapy, as the patient was diagnosed 2 weeks previously with an isolated cranial nerve VII and was started with steroids in a peripheral ophthalmology clinic. At his first presentation in Jordan University Hospital, he had a frozen orbit with cranial nerve III, IV, and VI palsies. The patient was admitted, and the results of baseline investigations were normal, except for a high white blood cell count and hemoglobin A1C (HbA1c) of 10%. Systemic steroids were stopped, and the patient was started on insulin to correct his blood sugar. He was never diagnosed with diabetes mellitus before and denied any symptoms of diabetes mellitus. He was initially diagnosed with diabetes mellitus. He tested negative for ketone bodies and was treated in the ward. Results of other routine investigations, such as kidney function tests and complete blood count, were normal, except for high blood sugar and a high white blood cell count. The differential diagnosis of his current symptoms was cavernous sinus thrombosis, a space-occupying lesion in the cavernous sinus or orbital cellulitis. Computed tomography of the brain and orbit was performed and revealed extensive sinus disease with left complicated ethmoidal and maxillary sinusitis extending into the left orbit with abscess formation. There was also extensive orbital cellulitis and evidence of left cavernous sinus thrombosis secondary to infection (Figure 1). It is possible that the patient had undiagnosed diabetes mellitus as indicated by a high HbA1c and was worsened by steroid therapy The patient was started on systemic vancomycin and ceftriaxone, and a swab from the nasal cavity was sent for evaluation. Nasal swab revealed methicillin-resistant staphylococcus and positive fungal staining. The patient was started on systemic amphotericin B at a loading dose and continued with the therapeutic dose after 2 days of his presentation to our institution. Nasal endoscopy revealed diffuse white filaments and necrosis of the bone from which a biopsy was taken. Sinuses were drained, followed by ethmoidectomy and maxillectomy. A biopsy from the necrotic tissue was sent, which showed the presence of non-septate hyphae branching at right angles consistent with angio-invasive mucormycosis. However, workup to identify the exact causative organism of mucormycosis was not performed as it did not affect further management. The patient was kept on daily irrigation of the sinuses along with administration of systemic amphotericin B. On follow-up magnetic resonance imaging, the patient had an intraorbital collection (Figure 2). Subsequently, drainage of the orbital abscess, tarsorrhaphy, and excision of the necrotic conjunctiva were performed. The patient was kept on systemic amphotericin B and showed clinical improvement. He remained hospitalized until a repeat biopsy of the paranasal sinuses was negative for the fungi. He was discharged with residual paresis of all cranial nerves involved along with left eye ptosis and only hand motion in the left eye. He was followed up serially for 1 year and 4 months. Currently, the patient is afebrile with complete resolution of symptoms in the right eye and partial recovery of cranial nerve palsies. However, in the left eye, residual ptosis has remained and only hand motion is visible due to the previous central retinal artery occlusion related to his initial presentation. Discussion Mucormycosis is an opportunistic infection with a high mortality rate, reaching up to 80% without treatment. 6 Even with early diagnosis and aggressive therapy, its prognosis remains poor. 7 Diabetes mellitus is the most common risk factor for mucormycosis, particularly during ketoacidosis. Ketones induce the fungi to utilize and produce ketoreductase, which facilitates its growth through various mechanisms. 4 Roden et al. reviewed the characteristics of 929 patients with mucormycosis, and of 929 patients, 36% had diabetes mellitus, 17% had an underlying malignancy, and 19% had no underlying conditions. 3 In a review of literature of 851 patients, the median age was 51 ± 12 years, and 531 (63%) were men. Diabetes mellitus was the most common underlying condition (340/851, 40%; 71 (20%) had documented ketoacidosis), followed by hematological malignancy (275/851, 32%; 116 (42%) had acute myeloid leukemia) and solid organ transplantation (116/851, 14%; 67 (58%) had received a kidney transplant). Of the predisposing factors, the use of corticosteroids at the time of presentation was the most common (273/851, 33%), followed by neutropenia (169/851, 20%) and trauma (166/851, 20%). Rhino-orbito-cerebral mucormycosis was the most commonly observed manifestation (288/851, 34%), followed by cutaneous (187/851, 22%) and pulmonary mucormycosis (172/851, 20%. 8 In the present case, the initiation of glucocorticoids as a treatment for facial nerve palsy may have facilitated the initial development of mucormycosis. Glucocorticoids are well known to increase blood glucose levels and can exacerbate hyperglycemia in a patient with diabetes mellitus. 9 Our patient should have been tested for hyperglycemia before the initiation of glucocorticoid therapy (1) to rule out diabetes mellitus as the underlying cause of his facial palsy and (2) to measure the baseline blood sugar before the initiation of steroid therapy. 10 Clinically, mucormycosis is characterized by rhinitis with granular and purulent discharge, nasal ulceration, black spots of infarcted mucosa, and paranasal sinusitis. However, other common presenting symptoms include epistaxis, ophthalmoplegia with blindness, proptosis and orbital cellulitis, hemiplegia or stroke, and decreased mental function. 11 In advanced disease, symptoms include chemosis, ptosis, proptosis, ophthalmoplegia, blindness, and multiple cranial nerve palsies (function of cranial nerves II, III, V, VI, and VI may be lost or impaired).4,5 The classical clinical presentations are facial pain, an irregular black eschar on the palatal or nasal mucosa, and pus discharge from the eye and nose. In a retrospective chart review of 48 patients with 49 cases of acute fulminant-invasive fungal sinusitis over 19 years, mucormycosis and aspergillosis were found in 22 and 27 cases, respectively. Orbital (proptosis, periorbital edema, and ophthalmoplegia) and cranial nerve symptoms were seen at presentation more frequently in mucormycosis (6 [27%] and 9 [41%]) than in aspergillosis cases (3 [11%] and 7 [26%]; p = 0.079). Long-term orbital and cranial nerve sequelae were reported in 16 (72%) mucormycosis cases and 10 (37%) aspergillosis cases (p = 0.0210). These data suggest that the presence of orbital and neurological symptoms at presentation warrants more aggressive surgical intervention because of the likelihood of mucormycosis. 12 In another review of mucormycosis cases, all six patients had uncontrolled diabetes mellitus with fungal-invasive disease and had multiple cranial nerve involvement, including cranial nerves II to VII. 2 However, our patient had not been definitely diagnosed with diabetes mellitus before he had the fungal infection. That cohort chiefly complained of facial swelling, blood-stained nasal discharge, and crusting. All six patients were on prolonged antibiotic treatment with an incomplete response. Facial dysesthesia was present in all sides involved. Six patients had impaired vision, two had vision loss, and four only had a perception of light. These patients improved to finger counting during treatment. All patients had periorbital edema, ptosis, and 7th cranial nerve palsy of the lower motor type, and two patients had irritability and disoriented state and succumbed to the disease within 48 h of treatment. By contrast, our patient was fully oriented and had an initial visual acuity of hand motion with limited improvement. 13 A diagnosis of rhino-orbito-cerebral mucormycosis requires high clinical suspicion, and a definitive diagnosis requires a biopsy indicating the presence of the fungi. Furthermore, imaging is required to identify the degree of local involvement and spread across the rhino-orbital structures. 3 Mucormycosis is treated with surgical debridement of necrotic tissues, use of intravenous antifungal drugs, and elimination of predisposing factors. In retrospective studies, the combination of surgery and antifungal drugs showed higher survival rates than medical treatment alone. 3 Our case emphasizes the importance of screening for diabetes mellitus to prevent such a devastating complication. Our patient denied any of the classical symptoms of diabetes mellitus, and the only risk factor he had for diabetes mellitus was his age. Given the relatively long asymptomatic period of diabetes mellitus, the American Diabetes Association recommends screening for diabetes mellitus at age 45 years. 14 Early detection and treatment of this patient's underlying diabetes mellitus may have prevented the development of mucormycosis, especially in a country like Jordan where the prevalence of diabetes mellitus among men aged >25 years is as high as 33%. 15 The involvement of cranial nerve VII in mucormycosis is very rare, and its pathogenesis is not well understood. 13 We are unsure whether the involvement of cranial nerve VII in this patient was the first manifestation of mucormycosis or it was due to an independent pathology in which its treatment with glucocorticoids facilitated the development of mucormycosis. Furthermore, the widespread involvement of cranial nerves III, IV, and VI is also uncommon and indicates severe disease possibly due to cavernous sinus thrombosis. 3 Conclusions Mucormycosis is a rare and possibly fatal opportunistic infection. It should be suspected in any patient who presents with symptoms of sinusitis along with the widespread involvement of the orbit and cranial nerves. Workup should be initiated for an underlying systemic disease, such as diabetes mellitus, in any patient diagnosed with mucormycosis. In the present case, initiation of steroid therapy may have influenced the development of fungal infection and its progression, and this highlights the importance of screening. It is critical to exclude diabetes mellitus before starting steroid treatment and to screen a susceptible population perhaps at a younger age than recommended. Treatment with systemic antifungals, surgical debridement of necrotic tissues, and correction of underlying hyperglycemia should be initiated as soon as possible in diabetic patients with mucormycosis. Although this patient was 56 years old, screening for diabetes mellitus in younger populations earlier than 45 years is recommended, especially in the presence of opportunistic infections. Ethical statement Ethical approval was obtained from the Jordan University Board. Informed consent was also obtained from the patient. Conflict of interest All authors declare no conflict of interest Funding sources This work was not funded by any funding bodies, and all authors have read and approved the contents of this manuscript. Figure 1. Figure 2.

Recovered with sequelae (consequent health issues)

ReactionOutcome

CC BY

34888198

2021

What was the outcome of reaction 'Osteonecrosis'?

Mucormycosis with extensive cranial nerve involvement as the first presentation of diabetes mellitus: A case report. Mucormycosis, a rare fungal infection, mainly affects individuals with diabetes mellitus and those who were immunocompromised and has a high mortality rate. Its most common presentation is similar to that of acute bacterial sinusitis with symptoms of nasal congestion, headache, and fever. The involvement of multiple cranial nerves in mucormycosis was rarely reported in the literature and indicates severe disease. Herein, we report the case of a 56-year-old man who was referred to the ophthalmology outpatient clinic for facial nerve palsy. He was treated with systemic steroids for 10 days with no improvement. On examination, he had a loss of vision and a frozen orbit due to involvement of cranial nerves II, III, IV, V, VI, and VII. An extensive workup revealed a hemoglobin A1C of 10%. However, he was never diagnosed with diabetes mellitus previously and denied any of the classical symptoms of diabetes mellitus. He underwent ethmoidectomy, maxillectomy, and drainage of an intraorbital abscess after appropriate imaging studies. Histopathology confirmed the diagnosis of mucormycosis, and the patient was started on systemic amphotericin B. This case emphasizes the importance of screening for diabetes mellitus. Early recognition of underlying diabetes mellitus in this patient may have prevented the development of mucormycosis along with its devastating complications. pmcIntroduction Mucormycosis is a rare opportunistic angio-invasive infection caused by fungi of the order Mucorales. It mainly affects individuals with diabetes mellitus and patients who have underlying immunosuppression, with a mortality rate reaching up to 80%. 1 According to reports from the United States, mucormycosis has a very low incidence of 1.7 in every million individuals. Diabetes mellitus is the most common risk factor for the development of mucormycosis across Asia, while hematological malignancies are the major risk factors across Western countries. 2 Although mucormycosis is a severe complication of poorly controlled diabetes mellitus, it is very rare for diabetes mellitus to be diagnosed after the development of mucormycosis. 3 Mucormycosis can involve many organ systems, including the cutaneous, pulmonary, and gastrointestinal systems, and in severe forms, it can spread to other organs. However, rhinocerebral involvement is the most frequent presentation. 4 Involvement of cranial nerve VII along with the widespread involvement of other cranial nerves was rarely reported in the literature. 5 Herein, we report a very unusual case of mucormycosis with extensive rhino-orbit cerebral involvement presenting with multiple cranial nerve palsies as a first presentation of diabetes mellitus in a previously healthy patient. Case Description A 56-year-old man was referred to our institution for a case of facial palsy. He was treated with systemic methylprednisolone for 10 days, without improvement. On examination, he had a visual acuity of 1.0 in the right eye and hand motion in the left. Motility examination revealed limitation of abduction of the right eye. The patient had a frozen orbit with a limitation of all gazes in the left eye. Fundus examination revealed a pale optic disk, afferent pupillary defect, and an ischemic retina, indicating central retinal artery occlusion. Examination findings were also consistent with oculomotor, trochlear, and abducens nerve paresis. Mucormycosis was a result of steroid therapy, as the patient was diagnosed 2 weeks previously with an isolated cranial nerve VII and was started with steroids in a peripheral ophthalmology clinic. At his first presentation in Jordan University Hospital, he had a frozen orbit with cranial nerve III, IV, and VI palsies. The patient was admitted, and the results of baseline investigations were normal, except for a high white blood cell count and hemoglobin A1C (HbA1c) of 10%. Systemic steroids were stopped, and the patient was started on insulin to correct his blood sugar. He was never diagnosed with diabetes mellitus before and denied any symptoms of diabetes mellitus. He was initially diagnosed with diabetes mellitus. He tested negative for ketone bodies and was treated in the ward. Results of other routine investigations, such as kidney function tests and complete blood count, were normal, except for high blood sugar and a high white blood cell count. The differential diagnosis of his current symptoms was cavernous sinus thrombosis, a space-occupying lesion in the cavernous sinus or orbital cellulitis. Computed tomography of the brain and orbit was performed and revealed extensive sinus disease with left complicated ethmoidal and maxillary sinusitis extending into the left orbit with abscess formation. There was also extensive orbital cellulitis and evidence of left cavernous sinus thrombosis secondary to infection (Figure 1). It is possible that the patient had undiagnosed diabetes mellitus as indicated by a high HbA1c and was worsened by steroid therapy The patient was started on systemic vancomycin and ceftriaxone, and a swab from the nasal cavity was sent for evaluation. Nasal swab revealed methicillin-resistant staphylococcus and positive fungal staining. The patient was started on systemic amphotericin B at a loading dose and continued with the therapeutic dose after 2 days of his presentation to our institution. Nasal endoscopy revealed diffuse white filaments and necrosis of the bone from which a biopsy was taken. Sinuses were drained, followed by ethmoidectomy and maxillectomy. A biopsy from the necrotic tissue was sent, which showed the presence of non-septate hyphae branching at right angles consistent with angio-invasive mucormycosis. However, workup to identify the exact causative organism of mucormycosis was not performed as it did not affect further management. The patient was kept on daily irrigation of the sinuses along with administration of systemic amphotericin B. On follow-up magnetic resonance imaging, the patient had an intraorbital collection (Figure 2). Subsequently, drainage of the orbital abscess, tarsorrhaphy, and excision of the necrotic conjunctiva were performed. The patient was kept on systemic amphotericin B and showed clinical improvement. He remained hospitalized until a repeat biopsy of the paranasal sinuses was negative for the fungi. He was discharged with residual paresis of all cranial nerves involved along with left eye ptosis and only hand motion in the left eye. He was followed up serially for 1 year and 4 months. Currently, the patient is afebrile with complete resolution of symptoms in the right eye and partial recovery of cranial nerve palsies. However, in the left eye, residual ptosis has remained and only hand motion is visible due to the previous central retinal artery occlusion related to his initial presentation. Discussion Mucormycosis is an opportunistic infection with a high mortality rate, reaching up to 80% without treatment. 6 Even with early diagnosis and aggressive therapy, its prognosis remains poor. 7 Diabetes mellitus is the most common risk factor for mucormycosis, particularly during ketoacidosis. Ketones induce the fungi to utilize and produce ketoreductase, which facilitates its growth through various mechanisms. 4 Roden et al. reviewed the characteristics of 929 patients with mucormycosis, and of 929 patients, 36% had diabetes mellitus, 17% had an underlying malignancy, and 19% had no underlying conditions. 3 In a review of literature of 851 patients, the median age was 51 ± 12 years, and 531 (63%) were men. Diabetes mellitus was the most common underlying condition (340/851, 40%; 71 (20%) had documented ketoacidosis), followed by hematological malignancy (275/851, 32%; 116 (42%) had acute myeloid leukemia) and solid organ transplantation (116/851, 14%; 67 (58%) had received a kidney transplant). Of the predisposing factors, the use of corticosteroids at the time of presentation was the most common (273/851, 33%), followed by neutropenia (169/851, 20%) and trauma (166/851, 20%). Rhino-orbito-cerebral mucormycosis was the most commonly observed manifestation (288/851, 34%), followed by cutaneous (187/851, 22%) and pulmonary mucormycosis (172/851, 20%. 8 In the present case, the initiation of glucocorticoids as a treatment for facial nerve palsy may have facilitated the initial development of mucormycosis. Glucocorticoids are well known to increase blood glucose levels and can exacerbate hyperglycemia in a patient with diabetes mellitus. 9 Our patient should have been tested for hyperglycemia before the initiation of glucocorticoid therapy (1) to rule out diabetes mellitus as the underlying cause of his facial palsy and (2) to measure the baseline blood sugar before the initiation of steroid therapy. 10 Clinically, mucormycosis is characterized by rhinitis with granular and purulent discharge, nasal ulceration, black spots of infarcted mucosa, and paranasal sinusitis. However, other common presenting symptoms include epistaxis, ophthalmoplegia with blindness, proptosis and orbital cellulitis, hemiplegia or stroke, and decreased mental function. 11 In advanced disease, symptoms include chemosis, ptosis, proptosis, ophthalmoplegia, blindness, and multiple cranial nerve palsies (function of cranial nerves II, III, V, VI, and VI may be lost or impaired).4,5 The classical clinical presentations are facial pain, an irregular black eschar on the palatal or nasal mucosa, and pus discharge from the eye and nose. In a retrospective chart review of 48 patients with 49 cases of acute fulminant-invasive fungal sinusitis over 19 years, mucormycosis and aspergillosis were found in 22 and 27 cases, respectively. Orbital (proptosis, periorbital edema, and ophthalmoplegia) and cranial nerve symptoms were seen at presentation more frequently in mucormycosis (6 [27%] and 9 [41%]) than in aspergillosis cases (3 [11%] and 7 [26%]; p = 0.079). Long-term orbital and cranial nerve sequelae were reported in 16 (72%) mucormycosis cases and 10 (37%) aspergillosis cases (p = 0.0210). These data suggest that the presence of orbital and neurological symptoms at presentation warrants more aggressive surgical intervention because of the likelihood of mucormycosis. 12 In another review of mucormycosis cases, all six patients had uncontrolled diabetes mellitus with fungal-invasive disease and had multiple cranial nerve involvement, including cranial nerves II to VII. 2 However, our patient had not been definitely diagnosed with diabetes mellitus before he had the fungal infection. That cohort chiefly complained of facial swelling, blood-stained nasal discharge, and crusting. All six patients were on prolonged antibiotic treatment with an incomplete response. Facial dysesthesia was present in all sides involved. Six patients had impaired vision, two had vision loss, and four only had a perception of light. These patients improved to finger counting during treatment. All patients had periorbital edema, ptosis, and 7th cranial nerve palsy of the lower motor type, and two patients had irritability and disoriented state and succumbed to the disease within 48 h of treatment. By contrast, our patient was fully oriented and had an initial visual acuity of hand motion with limited improvement. 13 A diagnosis of rhino-orbito-cerebral mucormycosis requires high clinical suspicion, and a definitive diagnosis requires a biopsy indicating the presence of the fungi. Furthermore, imaging is required to identify the degree of local involvement and spread across the rhino-orbital structures. 3 Mucormycosis is treated with surgical debridement of necrotic tissues, use of intravenous antifungal drugs, and elimination of predisposing factors. In retrospective studies, the combination of surgery and antifungal drugs showed higher survival rates than medical treatment alone. 3 Our case emphasizes the importance of screening for diabetes mellitus to prevent such a devastating complication. Our patient denied any of the classical symptoms of diabetes mellitus, and the only risk factor he had for diabetes mellitus was his age. Given the relatively long asymptomatic period of diabetes mellitus, the American Diabetes Association recommends screening for diabetes mellitus at age 45 years. 14 Early detection and treatment of this patient's underlying diabetes mellitus may have prevented the development of mucormycosis, especially in a country like Jordan where the prevalence of diabetes mellitus among men aged >25 years is as high as 33%. 15 The involvement of cranial nerve VII in mucormycosis is very rare, and its pathogenesis is not well understood. 13 We are unsure whether the involvement of cranial nerve VII in this patient was the first manifestation of mucormycosis or it was due to an independent pathology in which its treatment with glucocorticoids facilitated the development of mucormycosis. Furthermore, the widespread involvement of cranial nerves III, IV, and VI is also uncommon and indicates severe disease possibly due to cavernous sinus thrombosis. 3 Conclusions Mucormycosis is a rare and possibly fatal opportunistic infection. It should be suspected in any patient who presents with symptoms of sinusitis along with the widespread involvement of the orbit and cranial nerves. Workup should be initiated for an underlying systemic disease, such as diabetes mellitus, in any patient diagnosed with mucormycosis. In the present case, initiation of steroid therapy may have influenced the development of fungal infection and its progression, and this highlights the importance of screening. It is critical to exclude diabetes mellitus before starting steroid treatment and to screen a susceptible population perhaps at a younger age than recommended. Treatment with systemic antifungals, surgical debridement of necrotic tissues, and correction of underlying hyperglycemia should be initiated as soon as possible in diabetic patients with mucormycosis. Although this patient was 56 years old, screening for diabetes mellitus in younger populations earlier than 45 years is recommended, especially in the presence of opportunistic infections. Ethical statement Ethical approval was obtained from the Jordan University Board. Informed consent was also obtained from the patient. Conflict of interest All authors declare no conflict of interest Funding sources This work was not funded by any funding bodies, and all authors have read and approved the contents of this manuscript. Figure 1. Figure 2.

Recovering

ReactionOutcome

CC BY

34888198

2021

What was the outcome of reaction 'Periorbital abscess'?

Mucormycosis with extensive cranial nerve involvement as the first presentation of diabetes mellitus: A case report. Mucormycosis, a rare fungal infection, mainly affects individuals with diabetes mellitus and those who were immunocompromised and has a high mortality rate. Its most common presentation is similar to that of acute bacterial sinusitis with symptoms of nasal congestion, headache, and fever. The involvement of multiple cranial nerves in mucormycosis was rarely reported in the literature and indicates severe disease. Herein, we report the case of a 56-year-old man who was referred to the ophthalmology outpatient clinic for facial nerve palsy. He was treated with systemic steroids for 10 days with no improvement. On examination, he had a loss of vision and a frozen orbit due to involvement of cranial nerves II, III, IV, V, VI, and VII. An extensive workup revealed a hemoglobin A1C of 10%. However, he was never diagnosed with diabetes mellitus previously and denied any of the classical symptoms of diabetes mellitus. He underwent ethmoidectomy, maxillectomy, and drainage of an intraorbital abscess after appropriate imaging studies. Histopathology confirmed the diagnosis of mucormycosis, and the patient was started on systemic amphotericin B. This case emphasizes the importance of screening for diabetes mellitus. Early recognition of underlying diabetes mellitus in this patient may have prevented the development of mucormycosis along with its devastating complications. pmcIntroduction Mucormycosis is a rare opportunistic angio-invasive infection caused by fungi of the order Mucorales. It mainly affects individuals with diabetes mellitus and patients who have underlying immunosuppression, with a mortality rate reaching up to 80%. 1 According to reports from the United States, mucormycosis has a very low incidence of 1.7 in every million individuals. Diabetes mellitus is the most common risk factor for the development of mucormycosis across Asia, while hematological malignancies are the major risk factors across Western countries. 2 Although mucormycosis is a severe complication of poorly controlled diabetes mellitus, it is very rare for diabetes mellitus to be diagnosed after the development of mucormycosis. 3 Mucormycosis can involve many organ systems, including the cutaneous, pulmonary, and gastrointestinal systems, and in severe forms, it can spread to other organs. However, rhinocerebral involvement is the most frequent presentation. 4 Involvement of cranial nerve VII along with the widespread involvement of other cranial nerves was rarely reported in the literature. 5 Herein, we report a very unusual case of mucormycosis with extensive rhino-orbit cerebral involvement presenting with multiple cranial nerve palsies as a first presentation of diabetes mellitus in a previously healthy patient. Case Description A 56-year-old man was referred to our institution for a case of facial palsy. He was treated with systemic methylprednisolone for 10 days, without improvement. On examination, he had a visual acuity of 1.0 in the right eye and hand motion in the left. Motility examination revealed limitation of abduction of the right eye. The patient had a frozen orbit with a limitation of all gazes in the left eye. Fundus examination revealed a pale optic disk, afferent pupillary defect, and an ischemic retina, indicating central retinal artery occlusion. Examination findings were also consistent with oculomotor, trochlear, and abducens nerve paresis. Mucormycosis was a result of steroid therapy, as the patient was diagnosed 2 weeks previously with an isolated cranial nerve VII and was started with steroids in a peripheral ophthalmology clinic. At his first presentation in Jordan University Hospital, he had a frozen orbit with cranial nerve III, IV, and VI palsies. The patient was admitted, and the results of baseline investigations were normal, except for a high white blood cell count and hemoglobin A1C (HbA1c) of 10%. Systemic steroids were stopped, and the patient was started on insulin to correct his blood sugar. He was never diagnosed with diabetes mellitus before and denied any symptoms of diabetes mellitus. He was initially diagnosed with diabetes mellitus. He tested negative for ketone bodies and was treated in the ward. Results of other routine investigations, such as kidney function tests and complete blood count, were normal, except for high blood sugar and a high white blood cell count. The differential diagnosis of his current symptoms was cavernous sinus thrombosis, a space-occupying lesion in the cavernous sinus or orbital cellulitis. Computed tomography of the brain and orbit was performed and revealed extensive sinus disease with left complicated ethmoidal and maxillary sinusitis extending into the left orbit with abscess formation. There was also extensive orbital cellulitis and evidence of left cavernous sinus thrombosis secondary to infection (Figure 1). It is possible that the patient had undiagnosed diabetes mellitus as indicated by a high HbA1c and was worsened by steroid therapy The patient was started on systemic vancomycin and ceftriaxone, and a swab from the nasal cavity was sent for evaluation. Nasal swab revealed methicillin-resistant staphylococcus and positive fungal staining. The patient was started on systemic amphotericin B at a loading dose and continued with the therapeutic dose after 2 days of his presentation to our institution. Nasal endoscopy revealed diffuse white filaments and necrosis of the bone from which a biopsy was taken. Sinuses were drained, followed by ethmoidectomy and maxillectomy. A biopsy from the necrotic tissue was sent, which showed the presence of non-septate hyphae branching at right angles consistent with angio-invasive mucormycosis. However, workup to identify the exact causative organism of mucormycosis was not performed as it did not affect further management. The patient was kept on daily irrigation of the sinuses along with administration of systemic amphotericin B. On follow-up magnetic resonance imaging, the patient had an intraorbital collection (Figure 2). Subsequently, drainage of the orbital abscess, tarsorrhaphy, and excision of the necrotic conjunctiva were performed. The patient was kept on systemic amphotericin B and showed clinical improvement. He remained hospitalized until a repeat biopsy of the paranasal sinuses was negative for the fungi. He was discharged with residual paresis of all cranial nerves involved along with left eye ptosis and only hand motion in the left eye. He was followed up serially for 1 year and 4 months. Currently, the patient is afebrile with complete resolution of symptoms in the right eye and partial recovery of cranial nerve palsies. However, in the left eye, residual ptosis has remained and only hand motion is visible due to the previous central retinal artery occlusion related to his initial presentation. Discussion Mucormycosis is an opportunistic infection with a high mortality rate, reaching up to 80% without treatment. 6 Even with early diagnosis and aggressive therapy, its prognosis remains poor. 7 Diabetes mellitus is the most common risk factor for mucormycosis, particularly during ketoacidosis. Ketones induce the fungi to utilize and produce ketoreductase, which facilitates its growth through various mechanisms. 4 Roden et al. reviewed the characteristics of 929 patients with mucormycosis, and of 929 patients, 36% had diabetes mellitus, 17% had an underlying malignancy, and 19% had no underlying conditions. 3 In a review of literature of 851 patients, the median age was 51 ± 12 years, and 531 (63%) were men. Diabetes mellitus was the most common underlying condition (340/851, 40%; 71 (20%) had documented ketoacidosis), followed by hematological malignancy (275/851, 32%; 116 (42%) had acute myeloid leukemia) and solid organ transplantation (116/851, 14%; 67 (58%) had received a kidney transplant). Of the predisposing factors, the use of corticosteroids at the time of presentation was the most common (273/851, 33%), followed by neutropenia (169/851, 20%) and trauma (166/851, 20%). Rhino-orbito-cerebral mucormycosis was the most commonly observed manifestation (288/851, 34%), followed by cutaneous (187/851, 22%) and pulmonary mucormycosis (172/851, 20%. 8 In the present case, the initiation of glucocorticoids as a treatment for facial nerve palsy may have facilitated the initial development of mucormycosis. Glucocorticoids are well known to increase blood glucose levels and can exacerbate hyperglycemia in a patient with diabetes mellitus. 9 Our patient should have been tested for hyperglycemia before the initiation of glucocorticoid therapy (1) to rule out diabetes mellitus as the underlying cause of his facial palsy and (2) to measure the baseline blood sugar before the initiation of steroid therapy. 10 Clinically, mucormycosis is characterized by rhinitis with granular and purulent discharge, nasal ulceration, black spots of infarcted mucosa, and paranasal sinusitis. However, other common presenting symptoms include epistaxis, ophthalmoplegia with blindness, proptosis and orbital cellulitis, hemiplegia or stroke, and decreased mental function. 11 In advanced disease, symptoms include chemosis, ptosis, proptosis, ophthalmoplegia, blindness, and multiple cranial nerve palsies (function of cranial nerves II, III, V, VI, and VI may be lost or impaired).4,5 The classical clinical presentations are facial pain, an irregular black eschar on the palatal or nasal mucosa, and pus discharge from the eye and nose. In a retrospective chart review of 48 patients with 49 cases of acute fulminant-invasive fungal sinusitis over 19 years, mucormycosis and aspergillosis were found in 22 and 27 cases, respectively. Orbital (proptosis, periorbital edema, and ophthalmoplegia) and cranial nerve symptoms were seen at presentation more frequently in mucormycosis (6 [27%] and 9 [41%]) than in aspergillosis cases (3 [11%] and 7 [26%]; p = 0.079). Long-term orbital and cranial nerve sequelae were reported in 16 (72%) mucormycosis cases and 10 (37%) aspergillosis cases (p = 0.0210). These data suggest that the presence of orbital and neurological symptoms at presentation warrants more aggressive surgical intervention because of the likelihood of mucormycosis. 12 In another review of mucormycosis cases, all six patients had uncontrolled diabetes mellitus with fungal-invasive disease and had multiple cranial nerve involvement, including cranial nerves II to VII. 2 However, our patient had not been definitely diagnosed with diabetes mellitus before he had the fungal infection. That cohort chiefly complained of facial swelling, blood-stained nasal discharge, and crusting. All six patients were on prolonged antibiotic treatment with an incomplete response. Facial dysesthesia was present in all sides involved. Six patients had impaired vision, two had vision loss, and four only had a perception of light. These patients improved to finger counting during treatment. All patients had periorbital edema, ptosis, and 7th cranial nerve palsy of the lower motor type, and two patients had irritability and disoriented state and succumbed to the disease within 48 h of treatment. By contrast, our patient was fully oriented and had an initial visual acuity of hand motion with limited improvement. 13 A diagnosis of rhino-orbito-cerebral mucormycosis requires high clinical suspicion, and a definitive diagnosis requires a biopsy indicating the presence of the fungi. Furthermore, imaging is required to identify the degree of local involvement and spread across the rhino-orbital structures. 3 Mucormycosis is treated with surgical debridement of necrotic tissues, use of intravenous antifungal drugs, and elimination of predisposing factors. In retrospective studies, the combination of surgery and antifungal drugs showed higher survival rates than medical treatment alone. 3 Our case emphasizes the importance of screening for diabetes mellitus to prevent such a devastating complication. Our patient denied any of the classical symptoms of diabetes mellitus, and the only risk factor he had for diabetes mellitus was his age. Given the relatively long asymptomatic period of diabetes mellitus, the American Diabetes Association recommends screening for diabetes mellitus at age 45 years. 14 Early detection and treatment of this patient's underlying diabetes mellitus may have prevented the development of mucormycosis, especially in a country like Jordan where the prevalence of diabetes mellitus among men aged >25 years is as high as 33%. 15 The involvement of cranial nerve VII in mucormycosis is very rare, and its pathogenesis is not well understood. 13 We are unsure whether the involvement of cranial nerve VII in this patient was the first manifestation of mucormycosis or it was due to an independent pathology in which its treatment with glucocorticoids facilitated the development of mucormycosis. Furthermore, the widespread involvement of cranial nerves III, IV, and VI is also uncommon and indicates severe disease possibly due to cavernous sinus thrombosis. 3 Conclusions Mucormycosis is a rare and possibly fatal opportunistic infection. It should be suspected in any patient who presents with symptoms of sinusitis along with the widespread involvement of the orbit and cranial nerves. Workup should be initiated for an underlying systemic disease, such as diabetes mellitus, in any patient diagnosed with mucormycosis. In the present case, initiation of steroid therapy may have influenced the development of fungal infection and its progression, and this highlights the importance of screening. It is critical to exclude diabetes mellitus before starting steroid treatment and to screen a susceptible population perhaps at a younger age than recommended. Treatment with systemic antifungals, surgical debridement of necrotic tissues, and correction of underlying hyperglycemia should be initiated as soon as possible in diabetic patients with mucormycosis. Although this patient was 56 years old, screening for diabetes mellitus in younger populations earlier than 45 years is recommended, especially in the presence of opportunistic infections. Ethical statement Ethical approval was obtained from the Jordan University Board. Informed consent was also obtained from the patient. Conflict of interest All authors declare no conflict of interest Funding sources This work was not funded by any funding bodies, and all authors have read and approved the contents of this manuscript. Figure 1. Figure 2.

Recovering

ReactionOutcome

CC BY

34888198

2021

What was the outcome of reaction 'Rhinocerebral mucormycosis'?

Mucormycosis with extensive cranial nerve involvement as the first presentation of diabetes mellitus: A case report. Mucormycosis, a rare fungal infection, mainly affects individuals with diabetes mellitus and those who were immunocompromised and has a high mortality rate. Its most common presentation is similar to that of acute bacterial sinusitis with symptoms of nasal congestion, headache, and fever. The involvement of multiple cranial nerves in mucormycosis was rarely reported in the literature and indicates severe disease. Herein, we report the case of a 56-year-old man who was referred to the ophthalmology outpatient clinic for facial nerve palsy. He was treated with systemic steroids for 10 days with no improvement. On examination, he had a loss of vision and a frozen orbit due to involvement of cranial nerves II, III, IV, V, VI, and VII. An extensive workup revealed a hemoglobin A1C of 10%. However, he was never diagnosed with diabetes mellitus previously and denied any of the classical symptoms of diabetes mellitus. He underwent ethmoidectomy, maxillectomy, and drainage of an intraorbital abscess after appropriate imaging studies. Histopathology confirmed the diagnosis of mucormycosis, and the patient was started on systemic amphotericin B. This case emphasizes the importance of screening for diabetes mellitus. Early recognition of underlying diabetes mellitus in this patient may have prevented the development of mucormycosis along with its devastating complications. pmcIntroduction Mucormycosis is a rare opportunistic angio-invasive infection caused by fungi of the order Mucorales. It mainly affects individuals with diabetes mellitus and patients who have underlying immunosuppression, with a mortality rate reaching up to 80%. 1 According to reports from the United States, mucormycosis has a very low incidence of 1.7 in every million individuals. Diabetes mellitus is the most common risk factor for the development of mucormycosis across Asia, while hematological malignancies are the major risk factors across Western countries. 2 Although mucormycosis is a severe complication of poorly controlled diabetes mellitus, it is very rare for diabetes mellitus to be diagnosed after the development of mucormycosis. 3 Mucormycosis can involve many organ systems, including the cutaneous, pulmonary, and gastrointestinal systems, and in severe forms, it can spread to other organs. However, rhinocerebral involvement is the most frequent presentation. 4 Involvement of cranial nerve VII along with the widespread involvement of other cranial nerves was rarely reported in the literature. 5 Herein, we report a very unusual case of mucormycosis with extensive rhino-orbit cerebral involvement presenting with multiple cranial nerve palsies as a first presentation of diabetes mellitus in a previously healthy patient. Case Description A 56-year-old man was referred to our institution for a case of facial palsy. He was treated with systemic methylprednisolone for 10 days, without improvement. On examination, he had a visual acuity of 1.0 in the right eye and hand motion in the left. Motility examination revealed limitation of abduction of the right eye. The patient had a frozen orbit with a limitation of all gazes in the left eye. Fundus examination revealed a pale optic disk, afferent pupillary defect, and an ischemic retina, indicating central retinal artery occlusion. Examination findings were also consistent with oculomotor, trochlear, and abducens nerve paresis. Mucormycosis was a result of steroid therapy, as the patient was diagnosed 2 weeks previously with an isolated cranial nerve VII and was started with steroids in a peripheral ophthalmology clinic. At his first presentation in Jordan University Hospital, he had a frozen orbit with cranial nerve III, IV, and VI palsies. The patient was admitted, and the results of baseline investigations were normal, except for a high white blood cell count and hemoglobin A1C (HbA1c) of 10%. Systemic steroids were stopped, and the patient was started on insulin to correct his blood sugar. He was never diagnosed with diabetes mellitus before and denied any symptoms of diabetes mellitus. He was initially diagnosed with diabetes mellitus. He tested negative for ketone bodies and was treated in the ward. Results of other routine investigations, such as kidney function tests and complete blood count, were normal, except for high blood sugar and a high white blood cell count. The differential diagnosis of his current symptoms was cavernous sinus thrombosis, a space-occupying lesion in the cavernous sinus or orbital cellulitis. Computed tomography of the brain and orbit was performed and revealed extensive sinus disease with left complicated ethmoidal and maxillary sinusitis extending into the left orbit with abscess formation. There was also extensive orbital cellulitis and evidence of left cavernous sinus thrombosis secondary to infection (Figure 1). It is possible that the patient had undiagnosed diabetes mellitus as indicated by a high HbA1c and was worsened by steroid therapy The patient was started on systemic vancomycin and ceftriaxone, and a swab from the nasal cavity was sent for evaluation. Nasal swab revealed methicillin-resistant staphylococcus and positive fungal staining. The patient was started on systemic amphotericin B at a loading dose and continued with the therapeutic dose after 2 days of his presentation to our institution. Nasal endoscopy revealed diffuse white filaments and necrosis of the bone from which a biopsy was taken. Sinuses were drained, followed by ethmoidectomy and maxillectomy. A biopsy from the necrotic tissue was sent, which showed the presence of non-septate hyphae branching at right angles consistent with angio-invasive mucormycosis. However, workup to identify the exact causative organism of mucormycosis was not performed as it did not affect further management. The patient was kept on daily irrigation of the sinuses along with administration of systemic amphotericin B. On follow-up magnetic resonance imaging, the patient had an intraorbital collection (Figure 2). Subsequently, drainage of the orbital abscess, tarsorrhaphy, and excision of the necrotic conjunctiva were performed. The patient was kept on systemic amphotericin B and showed clinical improvement. He remained hospitalized until a repeat biopsy of the paranasal sinuses was negative for the fungi. He was discharged with residual paresis of all cranial nerves involved along with left eye ptosis and only hand motion in the left eye. He was followed up serially for 1 year and 4 months. Currently, the patient is afebrile with complete resolution of symptoms in the right eye and partial recovery of cranial nerve palsies. However, in the left eye, residual ptosis has remained and only hand motion is visible due to the previous central retinal artery occlusion related to his initial presentation. Discussion Mucormycosis is an opportunistic infection with a high mortality rate, reaching up to 80% without treatment. 6 Even with early diagnosis and aggressive therapy, its prognosis remains poor. 7 Diabetes mellitus is the most common risk factor for mucormycosis, particularly during ketoacidosis. Ketones induce the fungi to utilize and produce ketoreductase, which facilitates its growth through various mechanisms. 4 Roden et al. reviewed the characteristics of 929 patients with mucormycosis, and of 929 patients, 36% had diabetes mellitus, 17% had an underlying malignancy, and 19% had no underlying conditions. 3 In a review of literature of 851 patients, the median age was 51 ± 12 years, and 531 (63%) were men. Diabetes mellitus was the most common underlying condition (340/851, 40%; 71 (20%) had documented ketoacidosis), followed by hematological malignancy (275/851, 32%; 116 (42%) had acute myeloid leukemia) and solid organ transplantation (116/851, 14%; 67 (58%) had received a kidney transplant). Of the predisposing factors, the use of corticosteroids at the time of presentation was the most common (273/851, 33%), followed by neutropenia (169/851, 20%) and trauma (166/851, 20%). Rhino-orbito-cerebral mucormycosis was the most commonly observed manifestation (288/851, 34%), followed by cutaneous (187/851, 22%) and pulmonary mucormycosis (172/851, 20%. 8 In the present case, the initiation of glucocorticoids as a treatment for facial nerve palsy may have facilitated the initial development of mucormycosis. Glucocorticoids are well known to increase blood glucose levels and can exacerbate hyperglycemia in a patient with diabetes mellitus. 9 Our patient should have been tested for hyperglycemia before the initiation of glucocorticoid therapy (1) to rule out diabetes mellitus as the underlying cause of his facial palsy and (2) to measure the baseline blood sugar before the initiation of steroid therapy. 10 Clinically, mucormycosis is characterized by rhinitis with granular and purulent discharge, nasal ulceration, black spots of infarcted mucosa, and paranasal sinusitis. However, other common presenting symptoms include epistaxis, ophthalmoplegia with blindness, proptosis and orbital cellulitis, hemiplegia or stroke, and decreased mental function. 11 In advanced disease, symptoms include chemosis, ptosis, proptosis, ophthalmoplegia, blindness, and multiple cranial nerve palsies (function of cranial nerves II, III, V, VI, and VI may be lost or impaired).4,5 The classical clinical presentations are facial pain, an irregular black eschar on the palatal or nasal mucosa, and pus discharge from the eye and nose. In a retrospective chart review of 48 patients with 49 cases of acute fulminant-invasive fungal sinusitis over 19 years, mucormycosis and aspergillosis were found in 22 and 27 cases, respectively. Orbital (proptosis, periorbital edema, and ophthalmoplegia) and cranial nerve symptoms were seen at presentation more frequently in mucormycosis (6 [27%] and 9 [41%]) than in aspergillosis cases (3 [11%] and 7 [26%]; p = 0.079). Long-term orbital and cranial nerve sequelae were reported in 16 (72%) mucormycosis cases and 10 (37%) aspergillosis cases (p = 0.0210). These data suggest that the presence of orbital and neurological symptoms at presentation warrants more aggressive surgical intervention because of the likelihood of mucormycosis. 12 In another review of mucormycosis cases, all six patients had uncontrolled diabetes mellitus with fungal-invasive disease and had multiple cranial nerve involvement, including cranial nerves II to VII. 2 However, our patient had not been definitely diagnosed with diabetes mellitus before he had the fungal infection. That cohort chiefly complained of facial swelling, blood-stained nasal discharge, and crusting. All six patients were on prolonged antibiotic treatment with an incomplete response. Facial dysesthesia was present in all sides involved. Six patients had impaired vision, two had vision loss, and four only had a perception of light. These patients improved to finger counting during treatment. All patients had periorbital edema, ptosis, and 7th cranial nerve palsy of the lower motor type, and two patients had irritability and disoriented state and succumbed to the disease within 48 h of treatment. By contrast, our patient was fully oriented and had an initial visual acuity of hand motion with limited improvement. 13 A diagnosis of rhino-orbito-cerebral mucormycosis requires high clinical suspicion, and a definitive diagnosis requires a biopsy indicating the presence of the fungi. Furthermore, imaging is required to identify the degree of local involvement and spread across the rhino-orbital structures. 3 Mucormycosis is treated with surgical debridement of necrotic tissues, use of intravenous antifungal drugs, and elimination of predisposing factors. In retrospective studies, the combination of surgery and antifungal drugs showed higher survival rates than medical treatment alone. 3 Our case emphasizes the importance of screening for diabetes mellitus to prevent such a devastating complication. Our patient denied any of the classical symptoms of diabetes mellitus, and the only risk factor he had for diabetes mellitus was his age. Given the relatively long asymptomatic period of diabetes mellitus, the American Diabetes Association recommends screening for diabetes mellitus at age 45 years. 14 Early detection and treatment of this patient's underlying diabetes mellitus may have prevented the development of mucormycosis, especially in a country like Jordan where the prevalence of diabetes mellitus among men aged >25 years is as high as 33%. 15 The involvement of cranial nerve VII in mucormycosis is very rare, and its pathogenesis is not well understood. 13 We are unsure whether the involvement of cranial nerve VII in this patient was the first manifestation of mucormycosis or it was due to an independent pathology in which its treatment with glucocorticoids facilitated the development of mucormycosis. Furthermore, the widespread involvement of cranial nerves III, IV, and VI is also uncommon and indicates severe disease possibly due to cavernous sinus thrombosis. 3 Conclusions Mucormycosis is a rare and possibly fatal opportunistic infection. It should be suspected in any patient who presents with symptoms of sinusitis along with the widespread involvement of the orbit and cranial nerves. Workup should be initiated for an underlying systemic disease, such as diabetes mellitus, in any patient diagnosed with mucormycosis. In the present case, initiation of steroid therapy may have influenced the development of fungal infection and its progression, and this highlights the importance of screening. It is critical to exclude diabetes mellitus before starting steroid treatment and to screen a susceptible population perhaps at a younger age than recommended. Treatment with systemic antifungals, surgical debridement of necrotic tissues, and correction of underlying hyperglycemia should be initiated as soon as possible in diabetic patients with mucormycosis. Although this patient was 56 years old, screening for diabetes mellitus in younger populations earlier than 45 years is recommended, especially in the presence of opportunistic infections. Ethical statement Ethical approval was obtained from the Jordan University Board. Informed consent was also obtained from the patient. Conflict of interest All authors declare no conflict of interest Funding sources This work was not funded by any funding bodies, and all authors have read and approved the contents of this manuscript. Figure 1. Figure 2.

Recovering

ReactionOutcome

CC BY

34888198

2021

What was the outcome of reaction 'Sinusitis'?

Mucormycosis with extensive cranial nerve involvement as the first presentation of diabetes mellitus: A case report. Mucormycosis, a rare fungal infection, mainly affects individuals with diabetes mellitus and those who were immunocompromised and has a high mortality rate. Its most common presentation is similar to that of acute bacterial sinusitis with symptoms of nasal congestion, headache, and fever. The involvement of multiple cranial nerves in mucormycosis was rarely reported in the literature and indicates severe disease. Herein, we report the case of a 56-year-old man who was referred to the ophthalmology outpatient clinic for facial nerve palsy. He was treated with systemic steroids for 10 days with no improvement. On examination, he had a loss of vision and a frozen orbit due to involvement of cranial nerves II, III, IV, V, VI, and VII. An extensive workup revealed a hemoglobin A1C of 10%. However, he was never diagnosed with diabetes mellitus previously and denied any of the classical symptoms of diabetes mellitus. He underwent ethmoidectomy, maxillectomy, and drainage of an intraorbital abscess after appropriate imaging studies. Histopathology confirmed the diagnosis of mucormycosis, and the patient was started on systemic amphotericin B. This case emphasizes the importance of screening for diabetes mellitus. Early recognition of underlying diabetes mellitus in this patient may have prevented the development of mucormycosis along with its devastating complications. pmcIntroduction Mucormycosis is a rare opportunistic angio-invasive infection caused by fungi of the order Mucorales. It mainly affects individuals with diabetes mellitus and patients who have underlying immunosuppression, with a mortality rate reaching up to 80%. 1 According to reports from the United States, mucormycosis has a very low incidence of 1.7 in every million individuals. Diabetes mellitus is the most common risk factor for the development of mucormycosis across Asia, while hematological malignancies are the major risk factors across Western countries. 2 Although mucormycosis is a severe complication of poorly controlled diabetes mellitus, it is very rare for diabetes mellitus to be diagnosed after the development of mucormycosis. 3 Mucormycosis can involve many organ systems, including the cutaneous, pulmonary, and gastrointestinal systems, and in severe forms, it can spread to other organs. However, rhinocerebral involvement is the most frequent presentation. 4 Involvement of cranial nerve VII along with the widespread involvement of other cranial nerves was rarely reported in the literature. 5 Herein, we report a very unusual case of mucormycosis with extensive rhino-orbit cerebral involvement presenting with multiple cranial nerve palsies as a first presentation of diabetes mellitus in a previously healthy patient. Case Description A 56-year-old man was referred to our institution for a case of facial palsy. He was treated with systemic methylprednisolone for 10 days, without improvement. On examination, he had a visual acuity of 1.0 in the right eye and hand motion in the left. Motility examination revealed limitation of abduction of the right eye. The patient had a frozen orbit with a limitation of all gazes in the left eye. Fundus examination revealed a pale optic disk, afferent pupillary defect, and an ischemic retina, indicating central retinal artery occlusion. Examination findings were also consistent with oculomotor, trochlear, and abducens nerve paresis. Mucormycosis was a result of steroid therapy, as the patient was diagnosed 2 weeks previously with an isolated cranial nerve VII and was started with steroids in a peripheral ophthalmology clinic. At his first presentation in Jordan University Hospital, he had a frozen orbit with cranial nerve III, IV, and VI palsies. The patient was admitted, and the results of baseline investigations were normal, except for a high white blood cell count and hemoglobin A1C (HbA1c) of 10%. Systemic steroids were stopped, and the patient was started on insulin to correct his blood sugar. He was never diagnosed with diabetes mellitus before and denied any symptoms of diabetes mellitus. He was initially diagnosed with diabetes mellitus. He tested negative for ketone bodies and was treated in the ward. Results of other routine investigations, such as kidney function tests and complete blood count, were normal, except for high blood sugar and a high white blood cell count. The differential diagnosis of his current symptoms was cavernous sinus thrombosis, a space-occupying lesion in the cavernous sinus or orbital cellulitis. Computed tomography of the brain and orbit was performed and revealed extensive sinus disease with left complicated ethmoidal and maxillary sinusitis extending into the left orbit with abscess formation. There was also extensive orbital cellulitis and evidence of left cavernous sinus thrombosis secondary to infection (Figure 1). It is possible that the patient had undiagnosed diabetes mellitus as indicated by a high HbA1c and was worsened by steroid therapy The patient was started on systemic vancomycin and ceftriaxone, and a swab from the nasal cavity was sent for evaluation. Nasal swab revealed methicillin-resistant staphylococcus and positive fungal staining. The patient was started on systemic amphotericin B at a loading dose and continued with the therapeutic dose after 2 days of his presentation to our institution. Nasal endoscopy revealed diffuse white filaments and necrosis of the bone from which a biopsy was taken. Sinuses were drained, followed by ethmoidectomy and maxillectomy. A biopsy from the necrotic tissue was sent, which showed the presence of non-septate hyphae branching at right angles consistent with angio-invasive mucormycosis. However, workup to identify the exact causative organism of mucormycosis was not performed as it did not affect further management. The patient was kept on daily irrigation of the sinuses along with administration of systemic amphotericin B. On follow-up magnetic resonance imaging, the patient had an intraorbital collection (Figure 2). Subsequently, drainage of the orbital abscess, tarsorrhaphy, and excision of the necrotic conjunctiva were performed. The patient was kept on systemic amphotericin B and showed clinical improvement. He remained hospitalized until a repeat biopsy of the paranasal sinuses was negative for the fungi. He was discharged with residual paresis of all cranial nerves involved along with left eye ptosis and only hand motion in the left eye. He was followed up serially for 1 year and 4 months. Currently, the patient is afebrile with complete resolution of symptoms in the right eye and partial recovery of cranial nerve palsies. However, in the left eye, residual ptosis has remained and only hand motion is visible due to the previous central retinal artery occlusion related to his initial presentation. Discussion Mucormycosis is an opportunistic infection with a high mortality rate, reaching up to 80% without treatment. 6 Even with early diagnosis and aggressive therapy, its prognosis remains poor. 7 Diabetes mellitus is the most common risk factor for mucormycosis, particularly during ketoacidosis. Ketones induce the fungi to utilize and produce ketoreductase, which facilitates its growth through various mechanisms. 4 Roden et al. reviewed the characteristics of 929 patients with mucormycosis, and of 929 patients, 36% had diabetes mellitus, 17% had an underlying malignancy, and 19% had no underlying conditions. 3 In a review of literature of 851 patients, the median age was 51 ± 12 years, and 531 (63%) were men. Diabetes mellitus was the most common underlying condition (340/851, 40%; 71 (20%) had documented ketoacidosis), followed by hematological malignancy (275/851, 32%; 116 (42%) had acute myeloid leukemia) and solid organ transplantation (116/851, 14%; 67 (58%) had received a kidney transplant). Of the predisposing factors, the use of corticosteroids at the time of presentation was the most common (273/851, 33%), followed by neutropenia (169/851, 20%) and trauma (166/851, 20%). Rhino-orbito-cerebral mucormycosis was the most commonly observed manifestation (288/851, 34%), followed by cutaneous (187/851, 22%) and pulmonary mucormycosis (172/851, 20%. 8 In the present case, the initiation of glucocorticoids as a treatment for facial nerve palsy may have facilitated the initial development of mucormycosis. Glucocorticoids are well known to increase blood glucose levels and can exacerbate hyperglycemia in a patient with diabetes mellitus. 9 Our patient should have been tested for hyperglycemia before the initiation of glucocorticoid therapy (1) to rule out diabetes mellitus as the underlying cause of his facial palsy and (2) to measure the baseline blood sugar before the initiation of steroid therapy. 10 Clinically, mucormycosis is characterized by rhinitis with granular and purulent discharge, nasal ulceration, black spots of infarcted mucosa, and paranasal sinusitis. However, other common presenting symptoms include epistaxis, ophthalmoplegia with blindness, proptosis and orbital cellulitis, hemiplegia or stroke, and decreased mental function. 11 In advanced disease, symptoms include chemosis, ptosis, proptosis, ophthalmoplegia, blindness, and multiple cranial nerve palsies (function of cranial nerves II, III, V, VI, and VI may be lost or impaired).4,5 The classical clinical presentations are facial pain, an irregular black eschar on the palatal or nasal mucosa, and pus discharge from the eye and nose. In a retrospective chart review of 48 patients with 49 cases of acute fulminant-invasive fungal sinusitis over 19 years, mucormycosis and aspergillosis were found in 22 and 27 cases, respectively. Orbital (proptosis, periorbital edema, and ophthalmoplegia) and cranial nerve symptoms were seen at presentation more frequently in mucormycosis (6 [27%] and 9 [41%]) than in aspergillosis cases (3 [11%] and 7 [26%]; p = 0.079). Long-term orbital and cranial nerve sequelae were reported in 16 (72%) mucormycosis cases and 10 (37%) aspergillosis cases (p = 0.0210). These data suggest that the presence of orbital and neurological symptoms at presentation warrants more aggressive surgical intervention because of the likelihood of mucormycosis. 12 In another review of mucormycosis cases, all six patients had uncontrolled diabetes mellitus with fungal-invasive disease and had multiple cranial nerve involvement, including cranial nerves II to VII. 2 However, our patient had not been definitely diagnosed with diabetes mellitus before he had the fungal infection. That cohort chiefly complained of facial swelling, blood-stained nasal discharge, and crusting. All six patients were on prolonged antibiotic treatment with an incomplete response. Facial dysesthesia was present in all sides involved. Six patients had impaired vision, two had vision loss, and four only had a perception of light. These patients improved to finger counting during treatment. All patients had periorbital edema, ptosis, and 7th cranial nerve palsy of the lower motor type, and two patients had irritability and disoriented state and succumbed to the disease within 48 h of treatment. By contrast, our patient was fully oriented and had an initial visual acuity of hand motion with limited improvement. 13 A diagnosis of rhino-orbito-cerebral mucormycosis requires high clinical suspicion, and a definitive diagnosis requires a biopsy indicating the presence of the fungi. Furthermore, imaging is required to identify the degree of local involvement and spread across the rhino-orbital structures. 3 Mucormycosis is treated with surgical debridement of necrotic tissues, use of intravenous antifungal drugs, and elimination of predisposing factors. In retrospective studies, the combination of surgery and antifungal drugs showed higher survival rates than medical treatment alone. 3 Our case emphasizes the importance of screening for diabetes mellitus to prevent such a devastating complication. Our patient denied any of the classical symptoms of diabetes mellitus, and the only risk factor he had for diabetes mellitus was his age. Given the relatively long asymptomatic period of diabetes mellitus, the American Diabetes Association recommends screening for diabetes mellitus at age 45 years. 14 Early detection and treatment of this patient's underlying diabetes mellitus may have prevented the development of mucormycosis, especially in a country like Jordan where the prevalence of diabetes mellitus among men aged >25 years is as high as 33%. 15 The involvement of cranial nerve VII in mucormycosis is very rare, and its pathogenesis is not well understood. 13 We are unsure whether the involvement of cranial nerve VII in this patient was the first manifestation of mucormycosis or it was due to an independent pathology in which its treatment with glucocorticoids facilitated the development of mucormycosis. Furthermore, the widespread involvement of cranial nerves III, IV, and VI is also uncommon and indicates severe disease possibly due to cavernous sinus thrombosis. 3 Conclusions Mucormycosis is a rare and possibly fatal opportunistic infection. It should be suspected in any patient who presents with symptoms of sinusitis along with the widespread involvement of the orbit and cranial nerves. Workup should be initiated for an underlying systemic disease, such as diabetes mellitus, in any patient diagnosed with mucormycosis. In the present case, initiation of steroid therapy may have influenced the development of fungal infection and its progression, and this highlights the importance of screening. It is critical to exclude diabetes mellitus before starting steroid treatment and to screen a susceptible population perhaps at a younger age than recommended. Treatment with systemic antifungals, surgical debridement of necrotic tissues, and correction of underlying hyperglycemia should be initiated as soon as possible in diabetic patients with mucormycosis. Although this patient was 56 years old, screening for diabetes mellitus in younger populations earlier than 45 years is recommended, especially in the presence of opportunistic infections. Ethical statement Ethical approval was obtained from the Jordan University Board. Informed consent was also obtained from the patient. Conflict of interest All authors declare no conflict of interest Funding sources This work was not funded by any funding bodies, and all authors have read and approved the contents of this manuscript. Figure 1. Figure 2.

Recovering

ReactionOutcome

CC BY

34888198

2021

Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Disease progression'.

An Early-Onset Advanced Rectal Cancer Patient With Increased KRAS Gene Copy Number Showed A Primary Resistance to Cetuximab in Combination With Chemotherapy: A Case Report. Mutations in KRAS (codon 12/13), NRAS, BRAF V600E, and amplification of ERBB2 and MET account for 70-80% of anti-epidermal growth factor receptor (EGFR) monoclonal antibody primary resistance. However, the list of anti-EGFR monoclonal antibody primary resistance biomarkers is still incomplete. Herein, we report a case of wild-type RAS/BRAF metastatic colorectal cancer (CRC) with resistance to anti-EGFR monoclonal antibody and chemotherapy. Initially, mutation detection in postoperative tumor tissue by using amplification-refractory mutation system polymerase chain reaction indicated wild-type RAS/BRAF without point mutations, insertion deletions, or fusion mutations. Therefore, we recommended combined therapy of cetuximab and FOLFIRI after failure of platinum-based adjuvant chemotherapy, but the disease continued to progress. Next generation sequencing analysis of the postoperative tumor tissue revealed that KRAS copy number was increased and detected SMAD4, RNF43, and PREX2 mutations. This is the first case of advanced CRC with increased copy numbers of KRAS resistant to cetuximab and chemotherapy, which results in poor patient survival, and other mutated genes may be associated with the outcomes. Our findings indicate KRAS copy number alterations should also be examined, especially with anti-EGFR monoclonal antibody therapy in CRC, since it may be related with the primary resistance to these drugs. pmcIntroduction Colorectal cancer (CRC) is one of the most common causes of cancer-related deaths worldwide (1). The mean age of patients with CRC ranges from 49 to 60 years. Early onset of CRC generally refers to the onset in patients younger than 50 years of age at the time of diagnosis. It is characterized by a more advanced stage at diagnosis, poorer cell differentiation, higher prevalence of signet ring cell histology, and a primary tumor located on the left colon or rectum (2). The combination of biological monoclonal antibodies and chemotherapeutic cytotoxic drugs provides clinical benefits to patients with advanced or metastatic CRC (mCRC) (3). Bevacizumab is an anti-vascular endothelial growth factor monoclonal antibody (MoAb) used as a first-line treatment in RAS- or BRAF-mutated mCRC (4). Cetuximab and panitumumab can promote survival in patients with wild-type RAS and BRAF tumors, which target the epidermal growth factor receptor (EGFR) extracellular domain and subsequently inhibit the mitogen-activated protein kinase signaling pathway. The principal downstream effectors of EGFR activation are the RAS/RAF/MEK/ERK, PI3K/AKT/mTOR, and PLCγ/PKC pathways (5), which is a key regulator of cell proliferation, differentiation, division, and survival, and the metastatic potential of tumor cells (6). Mutations in any of the upstream genes may be transmitted to the protein through transcription or translation, resulting in abnormal activation of the signaling pathway (7). Alterations in RAS, BRAF, PIK3CA, EGFR, PTEN, and HER2 are key determinants of resistance to anti-EGFR MoAb therapies (8–10). However, little is known regarding patients harboring KRAS amplification, and data on the response to anti-EGFR treatment are lacking. Here, we present a patient with an initial diagnosis of KRAS-amplified locally advanced rectal adenocarcinoma who demonstrated no clinical response to anti-EGFR MoAb and chemotherapy, and the disease progressed rapidly. This case report indicates the potential of increased KRAS gene copy numbers (GCN) in primary resistance to anti-EGFR MoAb in patients with CRC. Case Report Case Presentation A 40-year-old man was admitted in our hospital with complaints of diarrhea for one month. Physical examination was normal, but electronic colonoscopy revealed a large polypoid mass in the rectum, situated approximately 9–15 cm from the anal verge. The lesion almost completely occluded the rectal lumen, allowing only a small endoscope to pass through. Pathology revealed an adenocarcinoma. Enhanced computed tomography (CT) of the abdomen revealed occupying lesions in the middle and upper rectum. Considering the possibility of rectal cancer, CT findings were consistent with T4aN1M0 ( Figure 1 ). Blood tests showed a slight increase in tumor markers (CA724, 7.79 U/mol; carcinoembryonic antigen, 9.16 ng/mol), creatinine (102.8 μmol/L), direct bilirubin (7.2 μmol/L), and C-reactive protein (13 mg/L). The patient underwent total resection of the rectal cancer and ileostomy on March 24, 2021. After surgery, all blood test results returned to normal. Figure 1 Imaging examinations performed before surgery. Enhanced CT scans on March 21, 2020 of abdomen revealed that occupying lesions in the middle and upper rectum, the intestinal lumen was narrowed, and the serosal layer was hairy. After enhancement, the lesion was uneven and enhanced, and the length of the lesion was about 5.7 cm, considering that was rectal cancer (T4aN1M0). Surgical pathology findings showed a moderate-to-poorly differentiated adenocarcinoma in the rectum (pT3N2bMx). The tumor volume was approximately 4 cm × 3 cm × 1.2 cm. Cancer infiltration was observed in vessels and nerves. Residual tumors can be detected on the edges of the surgical specimens. Meanwhile, 20 mesenteric lymph nodes were observed, and cancer embolus was distinguished in the local submucosal vessels. Immunohistochemistry results were as follows: pMMR, Ki-67 (+70%); CKpan (+), P53 (−), CgA (−), and Syn (−) ( Figure 2 ). Figure 2 Histopathological examination of tumor. (A) Hematoxylin and eosin (H&E) staining of colonic primary tumor shows glandular differentiation and invasion into the entire intestinal wall (×20). (B) Poorly differentiated adenocarcinoma on the surface of mucosa with multiple intravascular thrombus in submucosa, (×40). (C) Metastatic tumors can be seen in lymph node, (×40). Immunohistochemical staining of tumor cells. Ki-67 partial expression in tumor (+70%) (D, ×100), CKpan expression in tumor (E, ×100.), Immunostaining for MLH1 (+70%) (F, ×100), MSH2 (+70%) (G, ×100), MSH6 (+80%) (H, ×100)and PMS2 (+50%) (I, ×100) confirmed the tumor with proficient MMR. Enhanced pelvic Magnetic Resonance Imaging (MRI) was performed 21 days after the surgery and revealed multiple lymph node shadows with visible enhancement beside bilateral iliac vessels with a short diameter of approximately 0.6–0.7 cm, but postoperative inflammatory changes were not excluded. Therefore, five cycles of adjuvant chemotherapy (XELOX: oxaliplatin, 130 mg/m2 on day 1; capecitabine, 1,000 mg/m2 twice daily on days 1–14, orally) strategy was recommended from April 14 to July 28, 2020. The CT and MRI evaluations were progressive disease (PD) after chemotherapy, detecting significantly enlarged lymph nodes around the abdominal aorta (0.6-1.5cm) on August 12, 2020 ( Figure 3A ). Figure 3 Imaging examinations performed after chemotherapy. (A) After five cycles of XELOX, CT scans on August 12, 2020 revealed enlargement of the abdominal para-aortic and bilateral para-iliac lymph nodes, about 0.5–1.5 cm, with slightly enhanced. (B) After six cycles of FOLFIRI plus cetuximab treatment, CT scans on November 28, 2020 revealed significant enlargement of lymph nodes in the hilar area, adjacent to the abdominal aorta, and bilateral iliac vessels, with a maximum of about 2.3 cm and slightly enhancement. (C) MRI scans on November 29, 2020 show patchy shadows are seen on the right femoral head and greater trochanter, enhanced scans are seen to be enhanced, consider metastasis. (D) CT scans on December 22, 2020 showed hydronephrodilation of the right renal pelvis and space-occupying lesion about 16.2 mm × 9.3 mm in the initial segment of the right ureter. Since the first-line platinum-based regiment was unsuccessful, chemotherapy (FOLFIRI: 5-fluorouracil 400 mg/m2, IV bolus, folinic acid 400 mg/m2, and irinotecan 180 mg/m2 on day 1, followed by a continuous 46-h infusion of fluorouracil 2,400 mg/m2) in combination with cetuximab (500 mg/m² biweekly on day 1) was administered for six cycles from August 17 to November 12, 2020, considering that the patient harbored no RAS or BRAF alterations according to the mutation profile. After three cycles of chemotherapy and targeted therapy, the patient developed a grade 2 cutaneous toxicity, a single skin impetigo with a diameter greater than 10 mm, and the curative effect was evaluated as stable disease (SD), so we continued the therapy for three more cycles. On November 28, 2020, CT and MRI scans indicated PD, detecting a lymph node around the abdominal aorta enlarged to 2.3 cm ( Figure 3B ), tumor metastases to the right femoral head and greater trochanter ( Figure 3C ), and left cervical lymph nodes enlarged to 11 mm × 8 mm. Likewise, abdominal CT scans indicated PD again, observing a space-occupying lesion (1.6 cm × 0.9 cm) in the initial segment of the right ureter on December 22, 2020 ( Figure 3D ). Renal function tests suggested that uric acid and creatinine levels continuously increased to 470 µmol/L (normal: 210–430 µmol/L) and 226 µmol/L (normal: 57–97 µmol/L), respectively. The patient decided to stop the treatment and never returned to the hospital after four days of palliative radiotherapy. Mutation Analysis Wild-type KRAS, NRAS, and BRAF were amplified using amplification refractory mutation system polymerase chain reaction (ARMS-PCR) performed on March 20, 2020. Results revealed that the patient had wild-type RAS and BRAF. However, the tumor failed to respond to both treatment options and continued to progress. Molecular characterization of the blood and postoperative tumor DNA of the patient was performed using next-generation sequencing (NGS) on December 1, 2020. Sequence analyses are shown in Table 1 . We detected that the patient had mutations in three genes (SMAD4, RNF43, and PREX2), and GCN variations of KRAS increased to 7.8 gene copies before chemotherapy. In addition, the mutation abundance of the three mutant genes was observed using blood-based circulating tumor DNA (ctDNA) analysis: SMAD4 increased from 38.69 to 58.0%, RNF43 increased from 41.25 to 48.03%, and PREX2 increased from 39.22 to 52.99%, after the failure of second-line treatment. Likewise, the GCN of KRAS also increased to 8.31 gene copies, along with low abundance mutations in the other nine newly confirmed genes—CTNNB1, EPHA5, ETV1, FANCA, JAK2, MYC, PALB2, PIK3CA, and SPEN. Table 1 The results of sequence analyses of paraffin sections and ctDNA on December 1, 2020. Gene Mutation site Mutation abundance (paraffin sections) Mutation abundance (ctDNA) SMAD4 EXON:2 c.136A>T p.K46* 38.69% 58.0% RNF43 EXON:3 c.354dupC p.C119Lfs*6 41.25% 48.03% PREX2 EXON:28 c.3482G>T p.C1161F 39.22% 52.99% EXON:23 c.2558T>G p. L853R – 1.08% SPEN EXON:11 c.9239C>T p.P3080L – 1.66% PIK3CA EXON:14 c.2115A>C p.Q705H – 0.69% PALB2 EXON:4 c.212A>G p.E71G – 1.61% MYC EXON:3 c.1141A>T p.R381W – 0.78% JAK2 EXON:9 c.1121G>A p.R374K – 0.43% FANCA EXON:35 c.3458A>C p.D1153A – 0.45% ETV1 EXON:8 c.586T>G p.S196A – 2.82% EPHA5 EXON:14 c.2363T>C p.M788T – 0.78% EXON:14 c.2446G>C p.V816L – 0.64% CTNNB1 EXON:3 c.134C>G p.S45C – 3.76% Microsatellite Instability and Tumor Mutation Burden We analyzed MSI through surgically resected tumor tissues, and the results were reported as microsatellite stability (MSS)/MSI-low. The TMB detected in the tumor tissues was described as low level (2.95 mutations/MB). Furthermore, the TMB detected in ctDNA was reported as medium level (11.82 mutations/MB) after the failure of second-line therapy. Discussion Here, we present a patient with an early onset CRC with an initial diagnosis of moderate-to-poorly differentiated locally advanced rectal adenocarcinoma. Unlike the traditional RAS/BRAF mutant CRCs with poor prognosis, the patient had RAS/BRAF wild-type CRC accompanied with KRAS amplification and other less-reported gene mutations. Moreover, the patient’s prognosis was extremely poor, the disease progressed rapidly even with the use cetuximab in combination with chemotherapy. Therefore, we hypothesized that the disease in our patient with early onset CRC was more aggressive due to gene alterations, especially the increase in copy number of KRAS. Therefore, we report this case and hope to bring clinical benefits to clinicians. After the rapid progress of second-line treatment in our case, blood tests and postoperative tumor tissues were performed and analyzed using NGS and retrospective NGS, respectively. Results suggested an increase in KRAS copy number before treatment, accompanied by SMAD4, RNF43, and PREX2 mutations (mutation abundance was much more than 5%). After treatment failure, blood analysis indicated that the copy number of KRAS and clonal mutation abundance of the three genes continued to increase, indicating that the tumor developed from the original clone. KRAS amplification in CRC is a rare event, with an overall prevalence of 0.67–2% (11, 12). Previous studies have identified somatic mutations in KRAS as biomarkers for inherent resistance to EGFR-targeted drugs in patients with CRC (13), with a positive tissue mutation rate of 32–52.1% (7). However, alterations are responsive to the anti-EGFR inhibitors cetuximab or panitumumab (11, 14). GCN amplification of KRAS leads to the activation of RAS-RAF-ERK or PIK-AKT-mTOR pathways, even without additional activating mutations in the genes (15). In the presence of KRAS amplification, cetuximab can partially eliminate the phosphorylation of MEK and ERK but cannot induce growth arrest (13). A retrospective clinical study conducted by Valtorta et al. (14) detected that all four KRAS-amplified cases were found among the 53 patients who were resistant to anti-EGFR antibodies, while none of the 44 responders had tumors carrying this molecular alteration. Similarly, tumor biopsies of 10 patients who developed resistance to anti-EGFR showed the emergence of KRAS amplification in one case and acquisition of secondary KRAS mutations in six cases (13). Another retrospective study by Favazza et al. (11) reported that all eight patients with KRAS-amplified mCRC showed disease progression at the time of anti-EGFR therapy, and they concluded that KRAS amplification is responsible for the primary resistance to EGFR inhibitors. Meanwhile, RAS amplifications (involving KRAS, NRAS, and HRAS) were correlated with a younger median patient age at initial diagnosis and a history of inflammatory bowel disease (11). In the present case, KRAS amplification was present before chemotherapy, resulting in resistance to cetuximab. Therefore, patients with advanced mCRC should be monitored not only based on the RAS/BRAF mutation status but also RAS amplification, and patients with CRC that do not respond to anti-EGFR MoAbs should be excluded. Moreover, we suggest that genetic testing using NGS should be performed in younger patients to guide decision-making and prolong overall survival, because KRAS amplification is more likely to occur in younger patients. The mutation frequency and abundance of clonal mutations of SMAD4, RNF43, and PREX2 indicated that the tumor was progressing from the original clone. SMAD4, PREX2, and RNF43 are involved in the TGF-β, PI3K, and Wnt signaling pathways, respectively. These pathways are involved in cell death resistance, growth suppression, and sustained proliferative signaling, respectively. Several studies have reported chemoresistance and anti-EGFR resistance in patients with SMAD4-mutated mCRC (16–18), leading to poor prognosis. Furthermore, SMAD4 and RNF43 mutations were more commonly observed in early onset mCRC (≤55 years) (19). NGS could be considered as the best choice to analyze genetic tests for anti-EGFR therapy because it is superior to ARMS-PCR in copy number detection. Moreover, it saves the amount of samples, is cost- and time-efficient, and has great potential for clinical application to expand testing to include mutations in RAS and other less reported CRC-related genes (20). NGS analysis must be carried out in early onset CRCs, family history-related CRCs, and CRC-related cancers to identify cause-effective mutations, elucidate the clinical diagnosis, guide decision-making, and prevent the development of the disease in other family members. Furthermore, blood-based NGS testing suggested low-abundance mutations (<5%) of nine newly emerged genes, indicating the emergence of tumor subcloning and heterogeneity. These genes have been previously reported to be associated with poor prognosis (21–27), but few reports on the efficacy of chemotherapy or target drugs are available. This suggests that tumors with polygenic mutations tend to be more aggressive in terms of biological behavior, leading to poor survival rates. NGS was partially proven to have a higher sensitivity and specificity for detecting mutations with low abundance than ARMS-PCR. Blood-based NGS testing—ctDNA analysis—offers a convenient way to monitor tumor progression and treatment response. Since tumor mutational profiles are highly variable from person to person, a fixed content panel may be insufficient to track the treatment response in all patients. Compared to tumor tissue biopsies, blood-based ctDNA analyses are minimally invasive and accessible for the regular follow-up of cancer patients. The amount of ctDNA can be quantified, and genetic changes can be identified (28). The ctDNA analysis improves both the specificity and sensitivity of monitoring treatment response across several tumor types. It can identify tumor recurrence potentially earlier than imaging-based diagnosis. When augmented with tumor hotspot genes, it can track acquired drug-related mutations in patients (29). In conclusion, the present case report identified a rare early onset CRC accompanied by KRAS GCN amplification. The disease progressed rapidly, and the effects of chemotherapy and anti-EGFR therapy were poor. This suggests that KRAS amplification might be responsible for the primary resistance to anti-EGFR treatment in a small proportion of patients, and targeted drugs should not only be based on RAS and BRAF mutations in CRCs. Furthermore, NGS has the advantage of discovering genes that affect the efficacy of anti-EGFR therapies, and blood-based serial ctDNA analysis provides a convenient way to monitor the efficacy and resistance mechanism of anti-EGFR MoAbs. With improvements and developments in NGS technology, the identification of copy number variations may provide more implications for the diagnosis and treatment of CRC. Further validation in a larger population is needed to establish a predictive biomarker for resistance to anti-EGFR therapy. Data Availability Statement All data generated or analysed during this study are included in this published article. Additional data and materials related to the genetic tests, pathologic reports, treatment information, and images are available for review upon reasonable request. Ethics Statement Written informed consent was obtained from the patient for the publication of any potentially identifiable images or data included in this manuscript. Author Contributions TF and TL contributed equally to this work. All authors were involved in the drafting of the manuscript. TF, TL, and YW designed the clinical treatment for the patient. ZZ, HW, LX, JL, and CY performed the clinical treatment for the patients. CW and YT provided comments and edited the manuscript to become the final version for submission. All authors contributed to the article and approved the submitted version. Funding This study was supported by Health commission of Jilin Province (NO. 2017J063) and Department of Finance of Jilin Province (NO. JLSWSRCZ2020-0048). Conflict of Interest The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. Publisher’s Note All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher. Acknowledgments We would like to thank Editage (www.editage.cn) for English language editing.

CAPECITABINE, OXALIPLATIN

DrugsGivenReaction

CC BY

34888240

2021

Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Drug ineffective'.

An Early-Onset Advanced Rectal Cancer Patient With Increased KRAS Gene Copy Number Showed A Primary Resistance to Cetuximab in Combination With Chemotherapy: A Case Report. Mutations in KRAS (codon 12/13), NRAS, BRAF V600E, and amplification of ERBB2 and MET account for 70-80% of anti-epidermal growth factor receptor (EGFR) monoclonal antibody primary resistance. However, the list of anti-EGFR monoclonal antibody primary resistance biomarkers is still incomplete. Herein, we report a case of wild-type RAS/BRAF metastatic colorectal cancer (CRC) with resistance to anti-EGFR monoclonal antibody and chemotherapy. Initially, mutation detection in postoperative tumor tissue by using amplification-refractory mutation system polymerase chain reaction indicated wild-type RAS/BRAF without point mutations, insertion deletions, or fusion mutations. Therefore, we recommended combined therapy of cetuximab and FOLFIRI after failure of platinum-based adjuvant chemotherapy, but the disease continued to progress. Next generation sequencing analysis of the postoperative tumor tissue revealed that KRAS copy number was increased and detected SMAD4, RNF43, and PREX2 mutations. This is the first case of advanced CRC with increased copy numbers of KRAS resistant to cetuximab and chemotherapy, which results in poor patient survival, and other mutated genes may be associated with the outcomes. Our findings indicate KRAS copy number alterations should also be examined, especially with anti-EGFR monoclonal antibody therapy in CRC, since it may be related with the primary resistance to these drugs. pmcIntroduction Colorectal cancer (CRC) is one of the most common causes of cancer-related deaths worldwide (1). The mean age of patients with CRC ranges from 49 to 60 years. Early onset of CRC generally refers to the onset in patients younger than 50 years of age at the time of diagnosis. It is characterized by a more advanced stage at diagnosis, poorer cell differentiation, higher prevalence of signet ring cell histology, and a primary tumor located on the left colon or rectum (2). The combination of biological monoclonal antibodies and chemotherapeutic cytotoxic drugs provides clinical benefits to patients with advanced or metastatic CRC (mCRC) (3). Bevacizumab is an anti-vascular endothelial growth factor monoclonal antibody (MoAb) used as a first-line treatment in RAS- or BRAF-mutated mCRC (4). Cetuximab and panitumumab can promote survival in patients with wild-type RAS and BRAF tumors, which target the epidermal growth factor receptor (EGFR) extracellular domain and subsequently inhibit the mitogen-activated protein kinase signaling pathway. The principal downstream effectors of EGFR activation are the RAS/RAF/MEK/ERK, PI3K/AKT/mTOR, and PLCγ/PKC pathways (5), which is a key regulator of cell proliferation, differentiation, division, and survival, and the metastatic potential of tumor cells (6). Mutations in any of the upstream genes may be transmitted to the protein through transcription or translation, resulting in abnormal activation of the signaling pathway (7). Alterations in RAS, BRAF, PIK3CA, EGFR, PTEN, and HER2 are key determinants of resistance to anti-EGFR MoAb therapies (8–10). However, little is known regarding patients harboring KRAS amplification, and data on the response to anti-EGFR treatment are lacking. Here, we present a patient with an initial diagnosis of KRAS-amplified locally advanced rectal adenocarcinoma who demonstrated no clinical response to anti-EGFR MoAb and chemotherapy, and the disease progressed rapidly. This case report indicates the potential of increased KRAS gene copy numbers (GCN) in primary resistance to anti-EGFR MoAb in patients with CRC. Case Report Case Presentation A 40-year-old man was admitted in our hospital with complaints of diarrhea for one month. Physical examination was normal, but electronic colonoscopy revealed a large polypoid mass in the rectum, situated approximately 9–15 cm from the anal verge. The lesion almost completely occluded the rectal lumen, allowing only a small endoscope to pass through. Pathology revealed an adenocarcinoma. Enhanced computed tomography (CT) of the abdomen revealed occupying lesions in the middle and upper rectum. Considering the possibility of rectal cancer, CT findings were consistent with T4aN1M0 ( Figure 1 ). Blood tests showed a slight increase in tumor markers (CA724, 7.79 U/mol; carcinoembryonic antigen, 9.16 ng/mol), creatinine (102.8 μmol/L), direct bilirubin (7.2 μmol/L), and C-reactive protein (13 mg/L). The patient underwent total resection of the rectal cancer and ileostomy on March 24, 2021. After surgery, all blood test results returned to normal. Figure 1 Imaging examinations performed before surgery. Enhanced CT scans on March 21, 2020 of abdomen revealed that occupying lesions in the middle and upper rectum, the intestinal lumen was narrowed, and the serosal layer was hairy. After enhancement, the lesion was uneven and enhanced, and the length of the lesion was about 5.7 cm, considering that was rectal cancer (T4aN1M0). Surgical pathology findings showed a moderate-to-poorly differentiated adenocarcinoma in the rectum (pT3N2bMx). The tumor volume was approximately 4 cm × 3 cm × 1.2 cm. Cancer infiltration was observed in vessels and nerves. Residual tumors can be detected on the edges of the surgical specimens. Meanwhile, 20 mesenteric lymph nodes were observed, and cancer embolus was distinguished in the local submucosal vessels. Immunohistochemistry results were as follows: pMMR, Ki-67 (+70%); CKpan (+), P53 (−), CgA (−), and Syn (−) ( Figure 2 ). Figure 2 Histopathological examination of tumor. (A) Hematoxylin and eosin (H&E) staining of colonic primary tumor shows glandular differentiation and invasion into the entire intestinal wall (×20). (B) Poorly differentiated adenocarcinoma on the surface of mucosa with multiple intravascular thrombus in submucosa, (×40). (C) Metastatic tumors can be seen in lymph node, (×40). Immunohistochemical staining of tumor cells. Ki-67 partial expression in tumor (+70%) (D, ×100), CKpan expression in tumor (E, ×100.), Immunostaining for MLH1 (+70%) (F, ×100), MSH2 (+70%) (G, ×100), MSH6 (+80%) (H, ×100)and PMS2 (+50%) (I, ×100) confirmed the tumor with proficient MMR. Enhanced pelvic Magnetic Resonance Imaging (MRI) was performed 21 days after the surgery and revealed multiple lymph node shadows with visible enhancement beside bilateral iliac vessels with a short diameter of approximately 0.6–0.7 cm, but postoperative inflammatory changes were not excluded. Therefore, five cycles of adjuvant chemotherapy (XELOX: oxaliplatin, 130 mg/m2 on day 1; capecitabine, 1,000 mg/m2 twice daily on days 1–14, orally) strategy was recommended from April 14 to July 28, 2020. The CT and MRI evaluations were progressive disease (PD) after chemotherapy, detecting significantly enlarged lymph nodes around the abdominal aorta (0.6-1.5cm) on August 12, 2020 ( Figure 3A ). Figure 3 Imaging examinations performed after chemotherapy. (A) After five cycles of XELOX, CT scans on August 12, 2020 revealed enlargement of the abdominal para-aortic and bilateral para-iliac lymph nodes, about 0.5–1.5 cm, with slightly enhanced. (B) After six cycles of FOLFIRI plus cetuximab treatment, CT scans on November 28, 2020 revealed significant enlargement of lymph nodes in the hilar area, adjacent to the abdominal aorta, and bilateral iliac vessels, with a maximum of about 2.3 cm and slightly enhancement. (C) MRI scans on November 29, 2020 show patchy shadows are seen on the right femoral head and greater trochanter, enhanced scans are seen to be enhanced, consider metastasis. (D) CT scans on December 22, 2020 showed hydronephrodilation of the right renal pelvis and space-occupying lesion about 16.2 mm × 9.3 mm in the initial segment of the right ureter. Since the first-line platinum-based regiment was unsuccessful, chemotherapy (FOLFIRI: 5-fluorouracil 400 mg/m2, IV bolus, folinic acid 400 mg/m2, and irinotecan 180 mg/m2 on day 1, followed by a continuous 46-h infusion of fluorouracil 2,400 mg/m2) in combination with cetuximab (500 mg/m² biweekly on day 1) was administered for six cycles from August 17 to November 12, 2020, considering that the patient harbored no RAS or BRAF alterations according to the mutation profile. After three cycles of chemotherapy and targeted therapy, the patient developed a grade 2 cutaneous toxicity, a single skin impetigo with a diameter greater than 10 mm, and the curative effect was evaluated as stable disease (SD), so we continued the therapy for three more cycles. On November 28, 2020, CT and MRI scans indicated PD, detecting a lymph node around the abdominal aorta enlarged to 2.3 cm ( Figure 3B ), tumor metastases to the right femoral head and greater trochanter ( Figure 3C ), and left cervical lymph nodes enlarged to 11 mm × 8 mm. Likewise, abdominal CT scans indicated PD again, observing a space-occupying lesion (1.6 cm × 0.9 cm) in the initial segment of the right ureter on December 22, 2020 ( Figure 3D ). Renal function tests suggested that uric acid and creatinine levels continuously increased to 470 µmol/L (normal: 210–430 µmol/L) and 226 µmol/L (normal: 57–97 µmol/L), respectively. The patient decided to stop the treatment and never returned to the hospital after four days of palliative radiotherapy. Mutation Analysis Wild-type KRAS, NRAS, and BRAF were amplified using amplification refractory mutation system polymerase chain reaction (ARMS-PCR) performed on March 20, 2020. Results revealed that the patient had wild-type RAS and BRAF. However, the tumor failed to respond to both treatment options and continued to progress. Molecular characterization of the blood and postoperative tumor DNA of the patient was performed using next-generation sequencing (NGS) on December 1, 2020. Sequence analyses are shown in Table 1 . We detected that the patient had mutations in three genes (SMAD4, RNF43, and PREX2), and GCN variations of KRAS increased to 7.8 gene copies before chemotherapy. In addition, the mutation abundance of the three mutant genes was observed using blood-based circulating tumor DNA (ctDNA) analysis: SMAD4 increased from 38.69 to 58.0%, RNF43 increased from 41.25 to 48.03%, and PREX2 increased from 39.22 to 52.99%, after the failure of second-line treatment. Likewise, the GCN of KRAS also increased to 8.31 gene copies, along with low abundance mutations in the other nine newly confirmed genes—CTNNB1, EPHA5, ETV1, FANCA, JAK2, MYC, PALB2, PIK3CA, and SPEN. Table 1 The results of sequence analyses of paraffin sections and ctDNA on December 1, 2020. Gene Mutation site Mutation abundance (paraffin sections) Mutation abundance (ctDNA) SMAD4 EXON:2 c.136A>T p.K46* 38.69% 58.0% RNF43 EXON:3 c.354dupC p.C119Lfs*6 41.25% 48.03% PREX2 EXON:28 c.3482G>T p.C1161F 39.22% 52.99% EXON:23 c.2558T>G p. L853R – 1.08% SPEN EXON:11 c.9239C>T p.P3080L – 1.66% PIK3CA EXON:14 c.2115A>C p.Q705H – 0.69% PALB2 EXON:4 c.212A>G p.E71G – 1.61% MYC EXON:3 c.1141A>T p.R381W – 0.78% JAK2 EXON:9 c.1121G>A p.R374K – 0.43% FANCA EXON:35 c.3458A>C p.D1153A – 0.45% ETV1 EXON:8 c.586T>G p.S196A – 2.82% EPHA5 EXON:14 c.2363T>C p.M788T – 0.78% EXON:14 c.2446G>C p.V816L – 0.64% CTNNB1 EXON:3 c.134C>G p.S45C – 3.76% Microsatellite Instability and Tumor Mutation Burden We analyzed MSI through surgically resected tumor tissues, and the results were reported as microsatellite stability (MSS)/MSI-low. The TMB detected in the tumor tissues was described as low level (2.95 mutations/MB). Furthermore, the TMB detected in ctDNA was reported as medium level (11.82 mutations/MB) after the failure of second-line therapy. Discussion Here, we present a patient with an early onset CRC with an initial diagnosis of moderate-to-poorly differentiated locally advanced rectal adenocarcinoma. Unlike the traditional RAS/BRAF mutant CRCs with poor prognosis, the patient had RAS/BRAF wild-type CRC accompanied with KRAS amplification and other less-reported gene mutations. Moreover, the patient’s prognosis was extremely poor, the disease progressed rapidly even with the use cetuximab in combination with chemotherapy. Therefore, we hypothesized that the disease in our patient with early onset CRC was more aggressive due to gene alterations, especially the increase in copy number of KRAS. Therefore, we report this case and hope to bring clinical benefits to clinicians. After the rapid progress of second-line treatment in our case, blood tests and postoperative tumor tissues were performed and analyzed using NGS and retrospective NGS, respectively. Results suggested an increase in KRAS copy number before treatment, accompanied by SMAD4, RNF43, and PREX2 mutations (mutation abundance was much more than 5%). After treatment failure, blood analysis indicated that the copy number of KRAS and clonal mutation abundance of the three genes continued to increase, indicating that the tumor developed from the original clone. KRAS amplification in CRC is a rare event, with an overall prevalence of 0.67–2% (11, 12). Previous studies have identified somatic mutations in KRAS as biomarkers for inherent resistance to EGFR-targeted drugs in patients with CRC (13), with a positive tissue mutation rate of 32–52.1% (7). However, alterations are responsive to the anti-EGFR inhibitors cetuximab or panitumumab (11, 14). GCN amplification of KRAS leads to the activation of RAS-RAF-ERK or PIK-AKT-mTOR pathways, even without additional activating mutations in the genes (15). In the presence of KRAS amplification, cetuximab can partially eliminate the phosphorylation of MEK and ERK but cannot induce growth arrest (13). A retrospective clinical study conducted by Valtorta et al. (14) detected that all four KRAS-amplified cases were found among the 53 patients who were resistant to anti-EGFR antibodies, while none of the 44 responders had tumors carrying this molecular alteration. Similarly, tumor biopsies of 10 patients who developed resistance to anti-EGFR showed the emergence of KRAS amplification in one case and acquisition of secondary KRAS mutations in six cases (13). Another retrospective study by Favazza et al. (11) reported that all eight patients with KRAS-amplified mCRC showed disease progression at the time of anti-EGFR therapy, and they concluded that KRAS amplification is responsible for the primary resistance to EGFR inhibitors. Meanwhile, RAS amplifications (involving KRAS, NRAS, and HRAS) were correlated with a younger median patient age at initial diagnosis and a history of inflammatory bowel disease (11). In the present case, KRAS amplification was present before chemotherapy, resulting in resistance to cetuximab. Therefore, patients with advanced mCRC should be monitored not only based on the RAS/BRAF mutation status but also RAS amplification, and patients with CRC that do not respond to anti-EGFR MoAbs should be excluded. Moreover, we suggest that genetic testing using NGS should be performed in younger patients to guide decision-making and prolong overall survival, because KRAS amplification is more likely to occur in younger patients. The mutation frequency and abundance of clonal mutations of SMAD4, RNF43, and PREX2 indicated that the tumor was progressing from the original clone. SMAD4, PREX2, and RNF43 are involved in the TGF-β, PI3K, and Wnt signaling pathways, respectively. These pathways are involved in cell death resistance, growth suppression, and sustained proliferative signaling, respectively. Several studies have reported chemoresistance and anti-EGFR resistance in patients with SMAD4-mutated mCRC (16–18), leading to poor prognosis. Furthermore, SMAD4 and RNF43 mutations were more commonly observed in early onset mCRC (≤55 years) (19). NGS could be considered as the best choice to analyze genetic tests for anti-EGFR therapy because it is superior to ARMS-PCR in copy number detection. Moreover, it saves the amount of samples, is cost- and time-efficient, and has great potential for clinical application to expand testing to include mutations in RAS and other less reported CRC-related genes (20). NGS analysis must be carried out in early onset CRCs, family history-related CRCs, and CRC-related cancers to identify cause-effective mutations, elucidate the clinical diagnosis, guide decision-making, and prevent the development of the disease in other family members. Furthermore, blood-based NGS testing suggested low-abundance mutations (<5%) of nine newly emerged genes, indicating the emergence of tumor subcloning and heterogeneity. These genes have been previously reported to be associated with poor prognosis (21–27), but few reports on the efficacy of chemotherapy or target drugs are available. This suggests that tumors with polygenic mutations tend to be more aggressive in terms of biological behavior, leading to poor survival rates. NGS was partially proven to have a higher sensitivity and specificity for detecting mutations with low abundance than ARMS-PCR. Blood-based NGS testing—ctDNA analysis—offers a convenient way to monitor tumor progression and treatment response. Since tumor mutational profiles are highly variable from person to person, a fixed content panel may be insufficient to track the treatment response in all patients. Compared to tumor tissue biopsies, blood-based ctDNA analyses are minimally invasive and accessible for the regular follow-up of cancer patients. The amount of ctDNA can be quantified, and genetic changes can be identified (28). The ctDNA analysis improves both the specificity and sensitivity of monitoring treatment response across several tumor types. It can identify tumor recurrence potentially earlier than imaging-based diagnosis. When augmented with tumor hotspot genes, it can track acquired drug-related mutations in patients (29). In conclusion, the present case report identified a rare early onset CRC accompanied by KRAS GCN amplification. The disease progressed rapidly, and the effects of chemotherapy and anti-EGFR therapy were poor. This suggests that KRAS amplification might be responsible for the primary resistance to anti-EGFR treatment in a small proportion of patients, and targeted drugs should not only be based on RAS and BRAF mutations in CRCs. Furthermore, NGS has the advantage of discovering genes that affect the efficacy of anti-EGFR therapies, and blood-based serial ctDNA analysis provides a convenient way to monitor the efficacy and resistance mechanism of anti-EGFR MoAbs. With improvements and developments in NGS technology, the identification of copy number variations may provide more implications for the diagnosis and treatment of CRC. Further validation in a larger population is needed to establish a predictive biomarker for resistance to anti-EGFR therapy. Data Availability Statement All data generated or analysed during this study are included in this published article. Additional data and materials related to the genetic tests, pathologic reports, treatment information, and images are available for review upon reasonable request. Ethics Statement Written informed consent was obtained from the patient for the publication of any potentially identifiable images or data included in this manuscript. Author Contributions TF and TL contributed equally to this work. All authors were involved in the drafting of the manuscript. TF, TL, and YW designed the clinical treatment for the patient. ZZ, HW, LX, JL, and CY performed the clinical treatment for the patients. CW and YT provided comments and edited the manuscript to become the final version for submission. All authors contributed to the article and approved the submitted version. Funding This study was supported by Health commission of Jilin Province (NO. 2017J063) and Department of Finance of Jilin Province (NO. JLSWSRCZ2020-0048). Conflict of Interest The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. Publisher’s Note All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher. Acknowledgments We would like to thank Editage (www.editage.cn) for English language editing.

CAPECITABINE, OXALIPLATIN

DrugsGivenReaction

CC BY

34888240

2021

What was the administration route of drug 'CAPECITABINE'?

An Early-Onset Advanced Rectal Cancer Patient With Increased KRAS Gene Copy Number Showed A Primary Resistance to Cetuximab in Combination With Chemotherapy: A Case Report. Mutations in KRAS (codon 12/13), NRAS, BRAF V600E, and amplification of ERBB2 and MET account for 70-80% of anti-epidermal growth factor receptor (EGFR) monoclonal antibody primary resistance. However, the list of anti-EGFR monoclonal antibody primary resistance biomarkers is still incomplete. Herein, we report a case of wild-type RAS/BRAF metastatic colorectal cancer (CRC) with resistance to anti-EGFR monoclonal antibody and chemotherapy. Initially, mutation detection in postoperative tumor tissue by using amplification-refractory mutation system polymerase chain reaction indicated wild-type RAS/BRAF without point mutations, insertion deletions, or fusion mutations. Therefore, we recommended combined therapy of cetuximab and FOLFIRI after failure of platinum-based adjuvant chemotherapy, but the disease continued to progress. Next generation sequencing analysis of the postoperative tumor tissue revealed that KRAS copy number was increased and detected SMAD4, RNF43, and PREX2 mutations. This is the first case of advanced CRC with increased copy numbers of KRAS resistant to cetuximab and chemotherapy, which results in poor patient survival, and other mutated genes may be associated with the outcomes. Our findings indicate KRAS copy number alterations should also be examined, especially with anti-EGFR monoclonal antibody therapy in CRC, since it may be related with the primary resistance to these drugs. pmcIntroduction Colorectal cancer (CRC) is one of the most common causes of cancer-related deaths worldwide (1). The mean age of patients with CRC ranges from 49 to 60 years. Early onset of CRC generally refers to the onset in patients younger than 50 years of age at the time of diagnosis. It is characterized by a more advanced stage at diagnosis, poorer cell differentiation, higher prevalence of signet ring cell histology, and a primary tumor located on the left colon or rectum (2). The combination of biological monoclonal antibodies and chemotherapeutic cytotoxic drugs provides clinical benefits to patients with advanced or metastatic CRC (mCRC) (3). Bevacizumab is an anti-vascular endothelial growth factor monoclonal antibody (MoAb) used as a first-line treatment in RAS- or BRAF-mutated mCRC (4). Cetuximab and panitumumab can promote survival in patients with wild-type RAS and BRAF tumors, which target the epidermal growth factor receptor (EGFR) extracellular domain and subsequently inhibit the mitogen-activated protein kinase signaling pathway. The principal downstream effectors of EGFR activation are the RAS/RAF/MEK/ERK, PI3K/AKT/mTOR, and PLCγ/PKC pathways (5), which is a key regulator of cell proliferation, differentiation, division, and survival, and the metastatic potential of tumor cells (6). Mutations in any of the upstream genes may be transmitted to the protein through transcription or translation, resulting in abnormal activation of the signaling pathway (7). Alterations in RAS, BRAF, PIK3CA, EGFR, PTEN, and HER2 are key determinants of resistance to anti-EGFR MoAb therapies (8–10). However, little is known regarding patients harboring KRAS amplification, and data on the response to anti-EGFR treatment are lacking. Here, we present a patient with an initial diagnosis of KRAS-amplified locally advanced rectal adenocarcinoma who demonstrated no clinical response to anti-EGFR MoAb and chemotherapy, and the disease progressed rapidly. This case report indicates the potential of increased KRAS gene copy numbers (GCN) in primary resistance to anti-EGFR MoAb in patients with CRC. Case Report Case Presentation A 40-year-old man was admitted in our hospital with complaints of diarrhea for one month. Physical examination was normal, but electronic colonoscopy revealed a large polypoid mass in the rectum, situated approximately 9–15 cm from the anal verge. The lesion almost completely occluded the rectal lumen, allowing only a small endoscope to pass through. Pathology revealed an adenocarcinoma. Enhanced computed tomography (CT) of the abdomen revealed occupying lesions in the middle and upper rectum. Considering the possibility of rectal cancer, CT findings were consistent with T4aN1M0 ( Figure 1 ). Blood tests showed a slight increase in tumor markers (CA724, 7.79 U/mol; carcinoembryonic antigen, 9.16 ng/mol), creatinine (102.8 μmol/L), direct bilirubin (7.2 μmol/L), and C-reactive protein (13 mg/L). The patient underwent total resection of the rectal cancer and ileostomy on March 24, 2021. After surgery, all blood test results returned to normal. Figure 1 Imaging examinations performed before surgery. Enhanced CT scans on March 21, 2020 of abdomen revealed that occupying lesions in the middle and upper rectum, the intestinal lumen was narrowed, and the serosal layer was hairy. After enhancement, the lesion was uneven and enhanced, and the length of the lesion was about 5.7 cm, considering that was rectal cancer (T4aN1M0). Surgical pathology findings showed a moderate-to-poorly differentiated adenocarcinoma in the rectum (pT3N2bMx). The tumor volume was approximately 4 cm × 3 cm × 1.2 cm. Cancer infiltration was observed in vessels and nerves. Residual tumors can be detected on the edges of the surgical specimens. Meanwhile, 20 mesenteric lymph nodes were observed, and cancer embolus was distinguished in the local submucosal vessels. Immunohistochemistry results were as follows: pMMR, Ki-67 (+70%); CKpan (+), P53 (−), CgA (−), and Syn (−) ( Figure 2 ). Figure 2 Histopathological examination of tumor. (A) Hematoxylin and eosin (H&E) staining of colonic primary tumor shows glandular differentiation and invasion into the entire intestinal wall (×20). (B) Poorly differentiated adenocarcinoma on the surface of mucosa with multiple intravascular thrombus in submucosa, (×40). (C) Metastatic tumors can be seen in lymph node, (×40). Immunohistochemical staining of tumor cells. Ki-67 partial expression in tumor (+70%) (D, ×100), CKpan expression in tumor (E, ×100.), Immunostaining for MLH1 (+70%) (F, ×100), MSH2 (+70%) (G, ×100), MSH6 (+80%) (H, ×100)and PMS2 (+50%) (I, ×100) confirmed the tumor with proficient MMR. Enhanced pelvic Magnetic Resonance Imaging (MRI) was performed 21 days after the surgery and revealed multiple lymph node shadows with visible enhancement beside bilateral iliac vessels with a short diameter of approximately 0.6–0.7 cm, but postoperative inflammatory changes were not excluded. Therefore, five cycles of adjuvant chemotherapy (XELOX: oxaliplatin, 130 mg/m2 on day 1; capecitabine, 1,000 mg/m2 twice daily on days 1–14, orally) strategy was recommended from April 14 to July 28, 2020. The CT and MRI evaluations were progressive disease (PD) after chemotherapy, detecting significantly enlarged lymph nodes around the abdominal aorta (0.6-1.5cm) on August 12, 2020 ( Figure 3A ). Figure 3 Imaging examinations performed after chemotherapy. (A) After five cycles of XELOX, CT scans on August 12, 2020 revealed enlargement of the abdominal para-aortic and bilateral para-iliac lymph nodes, about 0.5–1.5 cm, with slightly enhanced. (B) After six cycles of FOLFIRI plus cetuximab treatment, CT scans on November 28, 2020 revealed significant enlargement of lymph nodes in the hilar area, adjacent to the abdominal aorta, and bilateral iliac vessels, with a maximum of about 2.3 cm and slightly enhancement. (C) MRI scans on November 29, 2020 show patchy shadows are seen on the right femoral head and greater trochanter, enhanced scans are seen to be enhanced, consider metastasis. (D) CT scans on December 22, 2020 showed hydronephrodilation of the right renal pelvis and space-occupying lesion about 16.2 mm × 9.3 mm in the initial segment of the right ureter. Since the first-line platinum-based regiment was unsuccessful, chemotherapy (FOLFIRI: 5-fluorouracil 400 mg/m2, IV bolus, folinic acid 400 mg/m2, and irinotecan 180 mg/m2 on day 1, followed by a continuous 46-h infusion of fluorouracil 2,400 mg/m2) in combination with cetuximab (500 mg/m² biweekly on day 1) was administered for six cycles from August 17 to November 12, 2020, considering that the patient harbored no RAS or BRAF alterations according to the mutation profile. After three cycles of chemotherapy and targeted therapy, the patient developed a grade 2 cutaneous toxicity, a single skin impetigo with a diameter greater than 10 mm, and the curative effect was evaluated as stable disease (SD), so we continued the therapy for three more cycles. On November 28, 2020, CT and MRI scans indicated PD, detecting a lymph node around the abdominal aorta enlarged to 2.3 cm ( Figure 3B ), tumor metastases to the right femoral head and greater trochanter ( Figure 3C ), and left cervical lymph nodes enlarged to 11 mm × 8 mm. Likewise, abdominal CT scans indicated PD again, observing a space-occupying lesion (1.6 cm × 0.9 cm) in the initial segment of the right ureter on December 22, 2020 ( Figure 3D ). Renal function tests suggested that uric acid and creatinine levels continuously increased to 470 µmol/L (normal: 210–430 µmol/L) and 226 µmol/L (normal: 57–97 µmol/L), respectively. The patient decided to stop the treatment and never returned to the hospital after four days of palliative radiotherapy. Mutation Analysis Wild-type KRAS, NRAS, and BRAF were amplified using amplification refractory mutation system polymerase chain reaction (ARMS-PCR) performed on March 20, 2020. Results revealed that the patient had wild-type RAS and BRAF. However, the tumor failed to respond to both treatment options and continued to progress. Molecular characterization of the blood and postoperative tumor DNA of the patient was performed using next-generation sequencing (NGS) on December 1, 2020. Sequence analyses are shown in Table 1 . We detected that the patient had mutations in three genes (SMAD4, RNF43, and PREX2), and GCN variations of KRAS increased to 7.8 gene copies before chemotherapy. In addition, the mutation abundance of the three mutant genes was observed using blood-based circulating tumor DNA (ctDNA) analysis: SMAD4 increased from 38.69 to 58.0%, RNF43 increased from 41.25 to 48.03%, and PREX2 increased from 39.22 to 52.99%, after the failure of second-line treatment. Likewise, the GCN of KRAS also increased to 8.31 gene copies, along with low abundance mutations in the other nine newly confirmed genes—CTNNB1, EPHA5, ETV1, FANCA, JAK2, MYC, PALB2, PIK3CA, and SPEN. Table 1 The results of sequence analyses of paraffin sections and ctDNA on December 1, 2020. Gene Mutation site Mutation abundance (paraffin sections) Mutation abundance (ctDNA) SMAD4 EXON:2 c.136A>T p.K46* 38.69% 58.0% RNF43 EXON:3 c.354dupC p.C119Lfs*6 41.25% 48.03% PREX2 EXON:28 c.3482G>T p.C1161F 39.22% 52.99% EXON:23 c.2558T>G p. L853R – 1.08% SPEN EXON:11 c.9239C>T p.P3080L – 1.66% PIK3CA EXON:14 c.2115A>C p.Q705H – 0.69% PALB2 EXON:4 c.212A>G p.E71G – 1.61% MYC EXON:3 c.1141A>T p.R381W – 0.78% JAK2 EXON:9 c.1121G>A p.R374K – 0.43% FANCA EXON:35 c.3458A>C p.D1153A – 0.45% ETV1 EXON:8 c.586T>G p.S196A – 2.82% EPHA5 EXON:14 c.2363T>C p.M788T – 0.78% EXON:14 c.2446G>C p.V816L – 0.64% CTNNB1 EXON:3 c.134C>G p.S45C – 3.76% Microsatellite Instability and Tumor Mutation Burden We analyzed MSI through surgically resected tumor tissues, and the results were reported as microsatellite stability (MSS)/MSI-low. The TMB detected in the tumor tissues was described as low level (2.95 mutations/MB). Furthermore, the TMB detected in ctDNA was reported as medium level (11.82 mutations/MB) after the failure of second-line therapy. Discussion Here, we present a patient with an early onset CRC with an initial diagnosis of moderate-to-poorly differentiated locally advanced rectal adenocarcinoma. Unlike the traditional RAS/BRAF mutant CRCs with poor prognosis, the patient had RAS/BRAF wild-type CRC accompanied with KRAS amplification and other less-reported gene mutations. Moreover, the patient’s prognosis was extremely poor, the disease progressed rapidly even with the use cetuximab in combination with chemotherapy. Therefore, we hypothesized that the disease in our patient with early onset CRC was more aggressive due to gene alterations, especially the increase in copy number of KRAS. Therefore, we report this case and hope to bring clinical benefits to clinicians. After the rapid progress of second-line treatment in our case, blood tests and postoperative tumor tissues were performed and analyzed using NGS and retrospective NGS, respectively. Results suggested an increase in KRAS copy number before treatment, accompanied by SMAD4, RNF43, and PREX2 mutations (mutation abundance was much more than 5%). After treatment failure, blood analysis indicated that the copy number of KRAS and clonal mutation abundance of the three genes continued to increase, indicating that the tumor developed from the original clone. KRAS amplification in CRC is a rare event, with an overall prevalence of 0.67–2% (11, 12). Previous studies have identified somatic mutations in KRAS as biomarkers for inherent resistance to EGFR-targeted drugs in patients with CRC (13), with a positive tissue mutation rate of 32–52.1% (7). However, alterations are responsive to the anti-EGFR inhibitors cetuximab or panitumumab (11, 14). GCN amplification of KRAS leads to the activation of RAS-RAF-ERK or PIK-AKT-mTOR pathways, even without additional activating mutations in the genes (15). In the presence of KRAS amplification, cetuximab can partially eliminate the phosphorylation of MEK and ERK but cannot induce growth arrest (13). A retrospective clinical study conducted by Valtorta et al. (14) detected that all four KRAS-amplified cases were found among the 53 patients who were resistant to anti-EGFR antibodies, while none of the 44 responders had tumors carrying this molecular alteration. Similarly, tumor biopsies of 10 patients who developed resistance to anti-EGFR showed the emergence of KRAS amplification in one case and acquisition of secondary KRAS mutations in six cases (13). Another retrospective study by Favazza et al. (11) reported that all eight patients with KRAS-amplified mCRC showed disease progression at the time of anti-EGFR therapy, and they concluded that KRAS amplification is responsible for the primary resistance to EGFR inhibitors. Meanwhile, RAS amplifications (involving KRAS, NRAS, and HRAS) were correlated with a younger median patient age at initial diagnosis and a history of inflammatory bowel disease (11). In the present case, KRAS amplification was present before chemotherapy, resulting in resistance to cetuximab. Therefore, patients with advanced mCRC should be monitored not only based on the RAS/BRAF mutation status but also RAS amplification, and patients with CRC that do not respond to anti-EGFR MoAbs should be excluded. Moreover, we suggest that genetic testing using NGS should be performed in younger patients to guide decision-making and prolong overall survival, because KRAS amplification is more likely to occur in younger patients. The mutation frequency and abundance of clonal mutations of SMAD4, RNF43, and PREX2 indicated that the tumor was progressing from the original clone. SMAD4, PREX2, and RNF43 are involved in the TGF-β, PI3K, and Wnt signaling pathways, respectively. These pathways are involved in cell death resistance, growth suppression, and sustained proliferative signaling, respectively. Several studies have reported chemoresistance and anti-EGFR resistance in patients with SMAD4-mutated mCRC (16–18), leading to poor prognosis. Furthermore, SMAD4 and RNF43 mutations were more commonly observed in early onset mCRC (≤55 years) (19). NGS could be considered as the best choice to analyze genetic tests for anti-EGFR therapy because it is superior to ARMS-PCR in copy number detection. Moreover, it saves the amount of samples, is cost- and time-efficient, and has great potential for clinical application to expand testing to include mutations in RAS and other less reported CRC-related genes (20). NGS analysis must be carried out in early onset CRCs, family history-related CRCs, and CRC-related cancers to identify cause-effective mutations, elucidate the clinical diagnosis, guide decision-making, and prevent the development of the disease in other family members. Furthermore, blood-based NGS testing suggested low-abundance mutations (<5%) of nine newly emerged genes, indicating the emergence of tumor subcloning and heterogeneity. These genes have been previously reported to be associated with poor prognosis (21–27), but few reports on the efficacy of chemotherapy or target drugs are available. This suggests that tumors with polygenic mutations tend to be more aggressive in terms of biological behavior, leading to poor survival rates. NGS was partially proven to have a higher sensitivity and specificity for detecting mutations with low abundance than ARMS-PCR. Blood-based NGS testing—ctDNA analysis—offers a convenient way to monitor tumor progression and treatment response. Since tumor mutational profiles are highly variable from person to person, a fixed content panel may be insufficient to track the treatment response in all patients. Compared to tumor tissue biopsies, blood-based ctDNA analyses are minimally invasive and accessible for the regular follow-up of cancer patients. The amount of ctDNA can be quantified, and genetic changes can be identified (28). The ctDNA analysis improves both the specificity and sensitivity of monitoring treatment response across several tumor types. It can identify tumor recurrence potentially earlier than imaging-based diagnosis. When augmented with tumor hotspot genes, it can track acquired drug-related mutations in patients (29). In conclusion, the present case report identified a rare early onset CRC accompanied by KRAS GCN amplification. The disease progressed rapidly, and the effects of chemotherapy and anti-EGFR therapy were poor. This suggests that KRAS amplification might be responsible for the primary resistance to anti-EGFR treatment in a small proportion of patients, and targeted drugs should not only be based on RAS and BRAF mutations in CRCs. Furthermore, NGS has the advantage of discovering genes that affect the efficacy of anti-EGFR therapies, and blood-based serial ctDNA analysis provides a convenient way to monitor the efficacy and resistance mechanism of anti-EGFR MoAbs. With improvements and developments in NGS technology, the identification of copy number variations may provide more implications for the diagnosis and treatment of CRC. Further validation in a larger population is needed to establish a predictive biomarker for resistance to anti-EGFR therapy. Data Availability Statement All data generated or analysed during this study are included in this published article. Additional data and materials related to the genetic tests, pathologic reports, treatment information, and images are available for review upon reasonable request. Ethics Statement Written informed consent was obtained from the patient for the publication of any potentially identifiable images or data included in this manuscript. Author Contributions TF and TL contributed equally to this work. All authors were involved in the drafting of the manuscript. TF, TL, and YW designed the clinical treatment for the patient. ZZ, HW, LX, JL, and CY performed the clinical treatment for the patients. CW and YT provided comments and edited the manuscript to become the final version for submission. All authors contributed to the article and approved the submitted version. Funding This study was supported by Health commission of Jilin Province (NO. 2017J063) and Department of Finance of Jilin Province (NO. JLSWSRCZ2020-0048). Conflict of Interest The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. Publisher’s Note All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher. Acknowledgments We would like to thank Editage (www.editage.cn) for English language editing.

Oral

DrugAdministrationRoute

CC BY

34888240

2021

What was the dosage of drug 'CAPECITABINE'?

An Early-Onset Advanced Rectal Cancer Patient With Increased KRAS Gene Copy Number Showed A Primary Resistance to Cetuximab in Combination With Chemotherapy: A Case Report. Mutations in KRAS (codon 12/13), NRAS, BRAF V600E, and amplification of ERBB2 and MET account for 70-80% of anti-epidermal growth factor receptor (EGFR) monoclonal antibody primary resistance. However, the list of anti-EGFR monoclonal antibody primary resistance biomarkers is still incomplete. Herein, we report a case of wild-type RAS/BRAF metastatic colorectal cancer (CRC) with resistance to anti-EGFR monoclonal antibody and chemotherapy. Initially, mutation detection in postoperative tumor tissue by using amplification-refractory mutation system polymerase chain reaction indicated wild-type RAS/BRAF without point mutations, insertion deletions, or fusion mutations. Therefore, we recommended combined therapy of cetuximab and FOLFIRI after failure of platinum-based adjuvant chemotherapy, but the disease continued to progress. Next generation sequencing analysis of the postoperative tumor tissue revealed that KRAS copy number was increased and detected SMAD4, RNF43, and PREX2 mutations. This is the first case of advanced CRC with increased copy numbers of KRAS resistant to cetuximab and chemotherapy, which results in poor patient survival, and other mutated genes may be associated with the outcomes. Our findings indicate KRAS copy number alterations should also be examined, especially with anti-EGFR monoclonal antibody therapy in CRC, since it may be related with the primary resistance to these drugs. pmcIntroduction Colorectal cancer (CRC) is one of the most common causes of cancer-related deaths worldwide (1). The mean age of patients with CRC ranges from 49 to 60 years. Early onset of CRC generally refers to the onset in patients younger than 50 years of age at the time of diagnosis. It is characterized by a more advanced stage at diagnosis, poorer cell differentiation, higher prevalence of signet ring cell histology, and a primary tumor located on the left colon or rectum (2). The combination of biological monoclonal antibodies and chemotherapeutic cytotoxic drugs provides clinical benefits to patients with advanced or metastatic CRC (mCRC) (3). Bevacizumab is an anti-vascular endothelial growth factor monoclonal antibody (MoAb) used as a first-line treatment in RAS- or BRAF-mutated mCRC (4). Cetuximab and panitumumab can promote survival in patients with wild-type RAS and BRAF tumors, which target the epidermal growth factor receptor (EGFR) extracellular domain and subsequently inhibit the mitogen-activated protein kinase signaling pathway. The principal downstream effectors of EGFR activation are the RAS/RAF/MEK/ERK, PI3K/AKT/mTOR, and PLCγ/PKC pathways (5), which is a key regulator of cell proliferation, differentiation, division, and survival, and the metastatic potential of tumor cells (6). Mutations in any of the upstream genes may be transmitted to the protein through transcription or translation, resulting in abnormal activation of the signaling pathway (7). Alterations in RAS, BRAF, PIK3CA, EGFR, PTEN, and HER2 are key determinants of resistance to anti-EGFR MoAb therapies (8–10). However, little is known regarding patients harboring KRAS amplification, and data on the response to anti-EGFR treatment are lacking. Here, we present a patient with an initial diagnosis of KRAS-amplified locally advanced rectal adenocarcinoma who demonstrated no clinical response to anti-EGFR MoAb and chemotherapy, and the disease progressed rapidly. This case report indicates the potential of increased KRAS gene copy numbers (GCN) in primary resistance to anti-EGFR MoAb in patients with CRC. Case Report Case Presentation A 40-year-old man was admitted in our hospital with complaints of diarrhea for one month. Physical examination was normal, but electronic colonoscopy revealed a large polypoid mass in the rectum, situated approximately 9–15 cm from the anal verge. The lesion almost completely occluded the rectal lumen, allowing only a small endoscope to pass through. Pathology revealed an adenocarcinoma. Enhanced computed tomography (CT) of the abdomen revealed occupying lesions in the middle and upper rectum. Considering the possibility of rectal cancer, CT findings were consistent with T4aN1M0 ( Figure 1 ). Blood tests showed a slight increase in tumor markers (CA724, 7.79 U/mol; carcinoembryonic antigen, 9.16 ng/mol), creatinine (102.8 μmol/L), direct bilirubin (7.2 μmol/L), and C-reactive protein (13 mg/L). The patient underwent total resection of the rectal cancer and ileostomy on March 24, 2021. After surgery, all blood test results returned to normal. Figure 1 Imaging examinations performed before surgery. Enhanced CT scans on March 21, 2020 of abdomen revealed that occupying lesions in the middle and upper rectum, the intestinal lumen was narrowed, and the serosal layer was hairy. After enhancement, the lesion was uneven and enhanced, and the length of the lesion was about 5.7 cm, considering that was rectal cancer (T4aN1M0). Surgical pathology findings showed a moderate-to-poorly differentiated adenocarcinoma in the rectum (pT3N2bMx). The tumor volume was approximately 4 cm × 3 cm × 1.2 cm. Cancer infiltration was observed in vessels and nerves. Residual tumors can be detected on the edges of the surgical specimens. Meanwhile, 20 mesenteric lymph nodes were observed, and cancer embolus was distinguished in the local submucosal vessels. Immunohistochemistry results were as follows: pMMR, Ki-67 (+70%); CKpan (+), P53 (−), CgA (−), and Syn (−) ( Figure 2 ). Figure 2 Histopathological examination of tumor. (A) Hematoxylin and eosin (H&E) staining of colonic primary tumor shows glandular differentiation and invasion into the entire intestinal wall (×20). (B) Poorly differentiated adenocarcinoma on the surface of mucosa with multiple intravascular thrombus in submucosa, (×40). (C) Metastatic tumors can be seen in lymph node, (×40). Immunohistochemical staining of tumor cells. Ki-67 partial expression in tumor (+70%) (D, ×100), CKpan expression in tumor (E, ×100.), Immunostaining for MLH1 (+70%) (F, ×100), MSH2 (+70%) (G, ×100), MSH6 (+80%) (H, ×100)and PMS2 (+50%) (I, ×100) confirmed the tumor with proficient MMR. Enhanced pelvic Magnetic Resonance Imaging (MRI) was performed 21 days after the surgery and revealed multiple lymph node shadows with visible enhancement beside bilateral iliac vessels with a short diameter of approximately 0.6–0.7 cm, but postoperative inflammatory changes were not excluded. Therefore, five cycles of adjuvant chemotherapy (XELOX: oxaliplatin, 130 mg/m2 on day 1; capecitabine, 1,000 mg/m2 twice daily on days 1–14, orally) strategy was recommended from April 14 to July 28, 2020. The CT and MRI evaluations were progressive disease (PD) after chemotherapy, detecting significantly enlarged lymph nodes around the abdominal aorta (0.6-1.5cm) on August 12, 2020 ( Figure 3A ). Figure 3 Imaging examinations performed after chemotherapy. (A) After five cycles of XELOX, CT scans on August 12, 2020 revealed enlargement of the abdominal para-aortic and bilateral para-iliac lymph nodes, about 0.5–1.5 cm, with slightly enhanced. (B) After six cycles of FOLFIRI plus cetuximab treatment, CT scans on November 28, 2020 revealed significant enlargement of lymph nodes in the hilar area, adjacent to the abdominal aorta, and bilateral iliac vessels, with a maximum of about 2.3 cm and slightly enhancement. (C) MRI scans on November 29, 2020 show patchy shadows are seen on the right femoral head and greater trochanter, enhanced scans are seen to be enhanced, consider metastasis. (D) CT scans on December 22, 2020 showed hydronephrodilation of the right renal pelvis and space-occupying lesion about 16.2 mm × 9.3 mm in the initial segment of the right ureter. Since the first-line platinum-based regiment was unsuccessful, chemotherapy (FOLFIRI: 5-fluorouracil 400 mg/m2, IV bolus, folinic acid 400 mg/m2, and irinotecan 180 mg/m2 on day 1, followed by a continuous 46-h infusion of fluorouracil 2,400 mg/m2) in combination with cetuximab (500 mg/m² biweekly on day 1) was administered for six cycles from August 17 to November 12, 2020, considering that the patient harbored no RAS or BRAF alterations according to the mutation profile. After three cycles of chemotherapy and targeted therapy, the patient developed a grade 2 cutaneous toxicity, a single skin impetigo with a diameter greater than 10 mm, and the curative effect was evaluated as stable disease (SD), so we continued the therapy for three more cycles. On November 28, 2020, CT and MRI scans indicated PD, detecting a lymph node around the abdominal aorta enlarged to 2.3 cm ( Figure 3B ), tumor metastases to the right femoral head and greater trochanter ( Figure 3C ), and left cervical lymph nodes enlarged to 11 mm × 8 mm. Likewise, abdominal CT scans indicated PD again, observing a space-occupying lesion (1.6 cm × 0.9 cm) in the initial segment of the right ureter on December 22, 2020 ( Figure 3D ). Renal function tests suggested that uric acid and creatinine levels continuously increased to 470 µmol/L (normal: 210–430 µmol/L) and 226 µmol/L (normal: 57–97 µmol/L), respectively. The patient decided to stop the treatment and never returned to the hospital after four days of palliative radiotherapy. Mutation Analysis Wild-type KRAS, NRAS, and BRAF were amplified using amplification refractory mutation system polymerase chain reaction (ARMS-PCR) performed on March 20, 2020. Results revealed that the patient had wild-type RAS and BRAF. However, the tumor failed to respond to both treatment options and continued to progress. Molecular characterization of the blood and postoperative tumor DNA of the patient was performed using next-generation sequencing (NGS) on December 1, 2020. Sequence analyses are shown in Table 1 . We detected that the patient had mutations in three genes (SMAD4, RNF43, and PREX2), and GCN variations of KRAS increased to 7.8 gene copies before chemotherapy. In addition, the mutation abundance of the three mutant genes was observed using blood-based circulating tumor DNA (ctDNA) analysis: SMAD4 increased from 38.69 to 58.0%, RNF43 increased from 41.25 to 48.03%, and PREX2 increased from 39.22 to 52.99%, after the failure of second-line treatment. Likewise, the GCN of KRAS also increased to 8.31 gene copies, along with low abundance mutations in the other nine newly confirmed genes—CTNNB1, EPHA5, ETV1, FANCA, JAK2, MYC, PALB2, PIK3CA, and SPEN. Table 1 The results of sequence analyses of paraffin sections and ctDNA on December 1, 2020. Gene Mutation site Mutation abundance (paraffin sections) Mutation abundance (ctDNA) SMAD4 EXON:2 c.136A>T p.K46* 38.69% 58.0% RNF43 EXON:3 c.354dupC p.C119Lfs*6 41.25% 48.03% PREX2 EXON:28 c.3482G>T p.C1161F 39.22% 52.99% EXON:23 c.2558T>G p. L853R – 1.08% SPEN EXON:11 c.9239C>T p.P3080L – 1.66% PIK3CA EXON:14 c.2115A>C p.Q705H – 0.69% PALB2 EXON:4 c.212A>G p.E71G – 1.61% MYC EXON:3 c.1141A>T p.R381W – 0.78% JAK2 EXON:9 c.1121G>A p.R374K – 0.43% FANCA EXON:35 c.3458A>C p.D1153A – 0.45% ETV1 EXON:8 c.586T>G p.S196A – 2.82% EPHA5 EXON:14 c.2363T>C p.M788T – 0.78% EXON:14 c.2446G>C p.V816L – 0.64% CTNNB1 EXON:3 c.134C>G p.S45C – 3.76% Microsatellite Instability and Tumor Mutation Burden We analyzed MSI through surgically resected tumor tissues, and the results were reported as microsatellite stability (MSS)/MSI-low. The TMB detected in the tumor tissues was described as low level (2.95 mutations/MB). Furthermore, the TMB detected in ctDNA was reported as medium level (11.82 mutations/MB) after the failure of second-line therapy. Discussion Here, we present a patient with an early onset CRC with an initial diagnosis of moderate-to-poorly differentiated locally advanced rectal adenocarcinoma. Unlike the traditional RAS/BRAF mutant CRCs with poor prognosis, the patient had RAS/BRAF wild-type CRC accompanied with KRAS amplification and other less-reported gene mutations. Moreover, the patient’s prognosis was extremely poor, the disease progressed rapidly even with the use cetuximab in combination with chemotherapy. Therefore, we hypothesized that the disease in our patient with early onset CRC was more aggressive due to gene alterations, especially the increase in copy number of KRAS. Therefore, we report this case and hope to bring clinical benefits to clinicians. After the rapid progress of second-line treatment in our case, blood tests and postoperative tumor tissues were performed and analyzed using NGS and retrospective NGS, respectively. Results suggested an increase in KRAS copy number before treatment, accompanied by SMAD4, RNF43, and PREX2 mutations (mutation abundance was much more than 5%). After treatment failure, blood analysis indicated that the copy number of KRAS and clonal mutation abundance of the three genes continued to increase, indicating that the tumor developed from the original clone. KRAS amplification in CRC is a rare event, with an overall prevalence of 0.67–2% (11, 12). Previous studies have identified somatic mutations in KRAS as biomarkers for inherent resistance to EGFR-targeted drugs in patients with CRC (13), with a positive tissue mutation rate of 32–52.1% (7). However, alterations are responsive to the anti-EGFR inhibitors cetuximab or panitumumab (11, 14). GCN amplification of KRAS leads to the activation of RAS-RAF-ERK or PIK-AKT-mTOR pathways, even without additional activating mutations in the genes (15). In the presence of KRAS amplification, cetuximab can partially eliminate the phosphorylation of MEK and ERK but cannot induce growth arrest (13). A retrospective clinical study conducted by Valtorta et al. (14) detected that all four KRAS-amplified cases were found among the 53 patients who were resistant to anti-EGFR antibodies, while none of the 44 responders had tumors carrying this molecular alteration. Similarly, tumor biopsies of 10 patients who developed resistance to anti-EGFR showed the emergence of KRAS amplification in one case and acquisition of secondary KRAS mutations in six cases (13). Another retrospective study by Favazza et al. (11) reported that all eight patients with KRAS-amplified mCRC showed disease progression at the time of anti-EGFR therapy, and they concluded that KRAS amplification is responsible for the primary resistance to EGFR inhibitors. Meanwhile, RAS amplifications (involving KRAS, NRAS, and HRAS) were correlated with a younger median patient age at initial diagnosis and a history of inflammatory bowel disease (11). In the present case, KRAS amplification was present before chemotherapy, resulting in resistance to cetuximab. Therefore, patients with advanced mCRC should be monitored not only based on the RAS/BRAF mutation status but also RAS amplification, and patients with CRC that do not respond to anti-EGFR MoAbs should be excluded. Moreover, we suggest that genetic testing using NGS should be performed in younger patients to guide decision-making and prolong overall survival, because KRAS amplification is more likely to occur in younger patients. The mutation frequency and abundance of clonal mutations of SMAD4, RNF43, and PREX2 indicated that the tumor was progressing from the original clone. SMAD4, PREX2, and RNF43 are involved in the TGF-β, PI3K, and Wnt signaling pathways, respectively. These pathways are involved in cell death resistance, growth suppression, and sustained proliferative signaling, respectively. Several studies have reported chemoresistance and anti-EGFR resistance in patients with SMAD4-mutated mCRC (16–18), leading to poor prognosis. Furthermore, SMAD4 and RNF43 mutations were more commonly observed in early onset mCRC (≤55 years) (19). NGS could be considered as the best choice to analyze genetic tests for anti-EGFR therapy because it is superior to ARMS-PCR in copy number detection. Moreover, it saves the amount of samples, is cost- and time-efficient, and has great potential for clinical application to expand testing to include mutations in RAS and other less reported CRC-related genes (20). NGS analysis must be carried out in early onset CRCs, family history-related CRCs, and CRC-related cancers to identify cause-effective mutations, elucidate the clinical diagnosis, guide decision-making, and prevent the development of the disease in other family members. Furthermore, blood-based NGS testing suggested low-abundance mutations (<5%) of nine newly emerged genes, indicating the emergence of tumor subcloning and heterogeneity. These genes have been previously reported to be associated with poor prognosis (21–27), but few reports on the efficacy of chemotherapy or target drugs are available. This suggests that tumors with polygenic mutations tend to be more aggressive in terms of biological behavior, leading to poor survival rates. NGS was partially proven to have a higher sensitivity and specificity for detecting mutations with low abundance than ARMS-PCR. Blood-based NGS testing—ctDNA analysis—offers a convenient way to monitor tumor progression and treatment response. Since tumor mutational profiles are highly variable from person to person, a fixed content panel may be insufficient to track the treatment response in all patients. Compared to tumor tissue biopsies, blood-based ctDNA analyses are minimally invasive and accessible for the regular follow-up of cancer patients. The amount of ctDNA can be quantified, and genetic changes can be identified (28). The ctDNA analysis improves both the specificity and sensitivity of monitoring treatment response across several tumor types. It can identify tumor recurrence potentially earlier than imaging-based diagnosis. When augmented with tumor hotspot genes, it can track acquired drug-related mutations in patients (29). In conclusion, the present case report identified a rare early onset CRC accompanied by KRAS GCN amplification. The disease progressed rapidly, and the effects of chemotherapy and anti-EGFR therapy were poor. This suggests that KRAS amplification might be responsible for the primary resistance to anti-EGFR treatment in a small proportion of patients, and targeted drugs should not only be based on RAS and BRAF mutations in CRCs. Furthermore, NGS has the advantage of discovering genes that affect the efficacy of anti-EGFR therapies, and blood-based serial ctDNA analysis provides a convenient way to monitor the efficacy and resistance mechanism of anti-EGFR MoAbs. With improvements and developments in NGS technology, the identification of copy number variations may provide more implications for the diagnosis and treatment of CRC. Further validation in a larger population is needed to establish a predictive biomarker for resistance to anti-EGFR therapy. Data Availability Statement All data generated or analysed during this study are included in this published article. Additional data and materials related to the genetic tests, pathologic reports, treatment information, and images are available for review upon reasonable request. Ethics Statement Written informed consent was obtained from the patient for the publication of any potentially identifiable images or data included in this manuscript. Author Contributions TF and TL contributed equally to this work. All authors were involved in the drafting of the manuscript. TF, TL, and YW designed the clinical treatment for the patient. ZZ, HW, LX, JL, and CY performed the clinical treatment for the patients. CW and YT provided comments and edited the manuscript to become the final version for submission. All authors contributed to the article and approved the submitted version. Funding This study was supported by Health commission of Jilin Province (NO. 2017J063) and Department of Finance of Jilin Province (NO. JLSWSRCZ2020-0048). Conflict of Interest The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. Publisher’s Note All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher. Acknowledgments We would like to thank Editage (www.editage.cn) for English language editing.

ON DAYS 1-14, FIVE CYCLES

DrugDosageText

CC BY

34888240

2021

What was the dosage of drug 'OXALIPLATIN'?

An Early-Onset Advanced Rectal Cancer Patient With Increased KRAS Gene Copy Number Showed A Primary Resistance to Cetuximab in Combination With Chemotherapy: A Case Report. Mutations in KRAS (codon 12/13), NRAS, BRAF V600E, and amplification of ERBB2 and MET account for 70-80% of anti-epidermal growth factor receptor (EGFR) monoclonal antibody primary resistance. However, the list of anti-EGFR monoclonal antibody primary resistance biomarkers is still incomplete. Herein, we report a case of wild-type RAS/BRAF metastatic colorectal cancer (CRC) with resistance to anti-EGFR monoclonal antibody and chemotherapy. Initially, mutation detection in postoperative tumor tissue by using amplification-refractory mutation system polymerase chain reaction indicated wild-type RAS/BRAF without point mutations, insertion deletions, or fusion mutations. Therefore, we recommended combined therapy of cetuximab and FOLFIRI after failure of platinum-based adjuvant chemotherapy, but the disease continued to progress. Next generation sequencing analysis of the postoperative tumor tissue revealed that KRAS copy number was increased and detected SMAD4, RNF43, and PREX2 mutations. This is the first case of advanced CRC with increased copy numbers of KRAS resistant to cetuximab and chemotherapy, which results in poor patient survival, and other mutated genes may be associated with the outcomes. Our findings indicate KRAS copy number alterations should also be examined, especially with anti-EGFR monoclonal antibody therapy in CRC, since it may be related with the primary resistance to these drugs. pmcIntroduction Colorectal cancer (CRC) is one of the most common causes of cancer-related deaths worldwide (1). The mean age of patients with CRC ranges from 49 to 60 years. Early onset of CRC generally refers to the onset in patients younger than 50 years of age at the time of diagnosis. It is characterized by a more advanced stage at diagnosis, poorer cell differentiation, higher prevalence of signet ring cell histology, and a primary tumor located on the left colon or rectum (2). The combination of biological monoclonal antibodies and chemotherapeutic cytotoxic drugs provides clinical benefits to patients with advanced or metastatic CRC (mCRC) (3). Bevacizumab is an anti-vascular endothelial growth factor monoclonal antibody (MoAb) used as a first-line treatment in RAS- or BRAF-mutated mCRC (4). Cetuximab and panitumumab can promote survival in patients with wild-type RAS and BRAF tumors, which target the epidermal growth factor receptor (EGFR) extracellular domain and subsequently inhibit the mitogen-activated protein kinase signaling pathway. The principal downstream effectors of EGFR activation are the RAS/RAF/MEK/ERK, PI3K/AKT/mTOR, and PLCγ/PKC pathways (5), which is a key regulator of cell proliferation, differentiation, division, and survival, and the metastatic potential of tumor cells (6). Mutations in any of the upstream genes may be transmitted to the protein through transcription or translation, resulting in abnormal activation of the signaling pathway (7). Alterations in RAS, BRAF, PIK3CA, EGFR, PTEN, and HER2 are key determinants of resistance to anti-EGFR MoAb therapies (8–10). However, little is known regarding patients harboring KRAS amplification, and data on the response to anti-EGFR treatment are lacking. Here, we present a patient with an initial diagnosis of KRAS-amplified locally advanced rectal adenocarcinoma who demonstrated no clinical response to anti-EGFR MoAb and chemotherapy, and the disease progressed rapidly. This case report indicates the potential of increased KRAS gene copy numbers (GCN) in primary resistance to anti-EGFR MoAb in patients with CRC. Case Report Case Presentation A 40-year-old man was admitted in our hospital with complaints of diarrhea for one month. Physical examination was normal, but electronic colonoscopy revealed a large polypoid mass in the rectum, situated approximately 9–15 cm from the anal verge. The lesion almost completely occluded the rectal lumen, allowing only a small endoscope to pass through. Pathology revealed an adenocarcinoma. Enhanced computed tomography (CT) of the abdomen revealed occupying lesions in the middle and upper rectum. Considering the possibility of rectal cancer, CT findings were consistent with T4aN1M0 ( Figure 1 ). Blood tests showed a slight increase in tumor markers (CA724, 7.79 U/mol; carcinoembryonic antigen, 9.16 ng/mol), creatinine (102.8 μmol/L), direct bilirubin (7.2 μmol/L), and C-reactive protein (13 mg/L). The patient underwent total resection of the rectal cancer and ileostomy on March 24, 2021. After surgery, all blood test results returned to normal. Figure 1 Imaging examinations performed before surgery. Enhanced CT scans on March 21, 2020 of abdomen revealed that occupying lesions in the middle and upper rectum, the intestinal lumen was narrowed, and the serosal layer was hairy. After enhancement, the lesion was uneven and enhanced, and the length of the lesion was about 5.7 cm, considering that was rectal cancer (T4aN1M0). Surgical pathology findings showed a moderate-to-poorly differentiated adenocarcinoma in the rectum (pT3N2bMx). The tumor volume was approximately 4 cm × 3 cm × 1.2 cm. Cancer infiltration was observed in vessels and nerves. Residual tumors can be detected on the edges of the surgical specimens. Meanwhile, 20 mesenteric lymph nodes were observed, and cancer embolus was distinguished in the local submucosal vessels. Immunohistochemistry results were as follows: pMMR, Ki-67 (+70%); CKpan (+), P53 (−), CgA (−), and Syn (−) ( Figure 2 ). Figure 2 Histopathological examination of tumor. (A) Hematoxylin and eosin (H&E) staining of colonic primary tumor shows glandular differentiation and invasion into the entire intestinal wall (×20). (B) Poorly differentiated adenocarcinoma on the surface of mucosa with multiple intravascular thrombus in submucosa, (×40). (C) Metastatic tumors can be seen in lymph node, (×40). Immunohistochemical staining of tumor cells. Ki-67 partial expression in tumor (+70%) (D, ×100), CKpan expression in tumor (E, ×100.), Immunostaining for MLH1 (+70%) (F, ×100), MSH2 (+70%) (G, ×100), MSH6 (+80%) (H, ×100)and PMS2 (+50%) (I, ×100) confirmed the tumor with proficient MMR. Enhanced pelvic Magnetic Resonance Imaging (MRI) was performed 21 days after the surgery and revealed multiple lymph node shadows with visible enhancement beside bilateral iliac vessels with a short diameter of approximately 0.6–0.7 cm, but postoperative inflammatory changes were not excluded. Therefore, five cycles of adjuvant chemotherapy (XELOX: oxaliplatin, 130 mg/m2 on day 1; capecitabine, 1,000 mg/m2 twice daily on days 1–14, orally) strategy was recommended from April 14 to July 28, 2020. The CT and MRI evaluations were progressive disease (PD) after chemotherapy, detecting significantly enlarged lymph nodes around the abdominal aorta (0.6-1.5cm) on August 12, 2020 ( Figure 3A ). Figure 3 Imaging examinations performed after chemotherapy. (A) After five cycles of XELOX, CT scans on August 12, 2020 revealed enlargement of the abdominal para-aortic and bilateral para-iliac lymph nodes, about 0.5–1.5 cm, with slightly enhanced. (B) After six cycles of FOLFIRI plus cetuximab treatment, CT scans on November 28, 2020 revealed significant enlargement of lymph nodes in the hilar area, adjacent to the abdominal aorta, and bilateral iliac vessels, with a maximum of about 2.3 cm and slightly enhancement. (C) MRI scans on November 29, 2020 show patchy shadows are seen on the right femoral head and greater trochanter, enhanced scans are seen to be enhanced, consider metastasis. (D) CT scans on December 22, 2020 showed hydronephrodilation of the right renal pelvis and space-occupying lesion about 16.2 mm × 9.3 mm in the initial segment of the right ureter. Since the first-line platinum-based regiment was unsuccessful, chemotherapy (FOLFIRI: 5-fluorouracil 400 mg/m2, IV bolus, folinic acid 400 mg/m2, and irinotecan 180 mg/m2 on day 1, followed by a continuous 46-h infusion of fluorouracil 2,400 mg/m2) in combination with cetuximab (500 mg/m² biweekly on day 1) was administered for six cycles from August 17 to November 12, 2020, considering that the patient harbored no RAS or BRAF alterations according to the mutation profile. After three cycles of chemotherapy and targeted therapy, the patient developed a grade 2 cutaneous toxicity, a single skin impetigo with a diameter greater than 10 mm, and the curative effect was evaluated as stable disease (SD), so we continued the therapy for three more cycles. On November 28, 2020, CT and MRI scans indicated PD, detecting a lymph node around the abdominal aorta enlarged to 2.3 cm ( Figure 3B ), tumor metastases to the right femoral head and greater trochanter ( Figure 3C ), and left cervical lymph nodes enlarged to 11 mm × 8 mm. Likewise, abdominal CT scans indicated PD again, observing a space-occupying lesion (1.6 cm × 0.9 cm) in the initial segment of the right ureter on December 22, 2020 ( Figure 3D ). Renal function tests suggested that uric acid and creatinine levels continuously increased to 470 µmol/L (normal: 210–430 µmol/L) and 226 µmol/L (normal: 57–97 µmol/L), respectively. The patient decided to stop the treatment and never returned to the hospital after four days of palliative radiotherapy. Mutation Analysis Wild-type KRAS, NRAS, and BRAF were amplified using amplification refractory mutation system polymerase chain reaction (ARMS-PCR) performed on March 20, 2020. Results revealed that the patient had wild-type RAS and BRAF. However, the tumor failed to respond to both treatment options and continued to progress. Molecular characterization of the blood and postoperative tumor DNA of the patient was performed using next-generation sequencing (NGS) on December 1, 2020. Sequence analyses are shown in Table 1 . We detected that the patient had mutations in three genes (SMAD4, RNF43, and PREX2), and GCN variations of KRAS increased to 7.8 gene copies before chemotherapy. In addition, the mutation abundance of the three mutant genes was observed using blood-based circulating tumor DNA (ctDNA) analysis: SMAD4 increased from 38.69 to 58.0%, RNF43 increased from 41.25 to 48.03%, and PREX2 increased from 39.22 to 52.99%, after the failure of second-line treatment. Likewise, the GCN of KRAS also increased to 8.31 gene copies, along with low abundance mutations in the other nine newly confirmed genes—CTNNB1, EPHA5, ETV1, FANCA, JAK2, MYC, PALB2, PIK3CA, and SPEN. Table 1 The results of sequence analyses of paraffin sections and ctDNA on December 1, 2020. Gene Mutation site Mutation abundance (paraffin sections) Mutation abundance (ctDNA) SMAD4 EXON:2 c.136A>T p.K46* 38.69% 58.0% RNF43 EXON:3 c.354dupC p.C119Lfs*6 41.25% 48.03% PREX2 EXON:28 c.3482G>T p.C1161F 39.22% 52.99% EXON:23 c.2558T>G p. L853R – 1.08% SPEN EXON:11 c.9239C>T p.P3080L – 1.66% PIK3CA EXON:14 c.2115A>C p.Q705H – 0.69% PALB2 EXON:4 c.212A>G p.E71G – 1.61% MYC EXON:3 c.1141A>T p.R381W – 0.78% JAK2 EXON:9 c.1121G>A p.R374K – 0.43% FANCA EXON:35 c.3458A>C p.D1153A – 0.45% ETV1 EXON:8 c.586T>G p.S196A – 2.82% EPHA5 EXON:14 c.2363T>C p.M788T – 0.78% EXON:14 c.2446G>C p.V816L – 0.64% CTNNB1 EXON:3 c.134C>G p.S45C – 3.76% Microsatellite Instability and Tumor Mutation Burden We analyzed MSI through surgically resected tumor tissues, and the results were reported as microsatellite stability (MSS)/MSI-low. The TMB detected in the tumor tissues was described as low level (2.95 mutations/MB). Furthermore, the TMB detected in ctDNA was reported as medium level (11.82 mutations/MB) after the failure of second-line therapy. Discussion Here, we present a patient with an early onset CRC with an initial diagnosis of moderate-to-poorly differentiated locally advanced rectal adenocarcinoma. Unlike the traditional RAS/BRAF mutant CRCs with poor prognosis, the patient had RAS/BRAF wild-type CRC accompanied with KRAS amplification and other less-reported gene mutations. Moreover, the patient’s prognosis was extremely poor, the disease progressed rapidly even with the use cetuximab in combination with chemotherapy. Therefore, we hypothesized that the disease in our patient with early onset CRC was more aggressive due to gene alterations, especially the increase in copy number of KRAS. Therefore, we report this case and hope to bring clinical benefits to clinicians. After the rapid progress of second-line treatment in our case, blood tests and postoperative tumor tissues were performed and analyzed using NGS and retrospective NGS, respectively. Results suggested an increase in KRAS copy number before treatment, accompanied by SMAD4, RNF43, and PREX2 mutations (mutation abundance was much more than 5%). After treatment failure, blood analysis indicated that the copy number of KRAS and clonal mutation abundance of the three genes continued to increase, indicating that the tumor developed from the original clone. KRAS amplification in CRC is a rare event, with an overall prevalence of 0.67–2% (11, 12). Previous studies have identified somatic mutations in KRAS as biomarkers for inherent resistance to EGFR-targeted drugs in patients with CRC (13), with a positive tissue mutation rate of 32–52.1% (7). However, alterations are responsive to the anti-EGFR inhibitors cetuximab or panitumumab (11, 14). GCN amplification of KRAS leads to the activation of RAS-RAF-ERK or PIK-AKT-mTOR pathways, even without additional activating mutations in the genes (15). In the presence of KRAS amplification, cetuximab can partially eliminate the phosphorylation of MEK and ERK but cannot induce growth arrest (13). A retrospective clinical study conducted by Valtorta et al. (14) detected that all four KRAS-amplified cases were found among the 53 patients who were resistant to anti-EGFR antibodies, while none of the 44 responders had tumors carrying this molecular alteration. Similarly, tumor biopsies of 10 patients who developed resistance to anti-EGFR showed the emergence of KRAS amplification in one case and acquisition of secondary KRAS mutations in six cases (13). Another retrospective study by Favazza et al. (11) reported that all eight patients with KRAS-amplified mCRC showed disease progression at the time of anti-EGFR therapy, and they concluded that KRAS amplification is responsible for the primary resistance to EGFR inhibitors. Meanwhile, RAS amplifications (involving KRAS, NRAS, and HRAS) were correlated with a younger median patient age at initial diagnosis and a history of inflammatory bowel disease (11). In the present case, KRAS amplification was present before chemotherapy, resulting in resistance to cetuximab. Therefore, patients with advanced mCRC should be monitored not only based on the RAS/BRAF mutation status but also RAS amplification, and patients with CRC that do not respond to anti-EGFR MoAbs should be excluded. Moreover, we suggest that genetic testing using NGS should be performed in younger patients to guide decision-making and prolong overall survival, because KRAS amplification is more likely to occur in younger patients. The mutation frequency and abundance of clonal mutations of SMAD4, RNF43, and PREX2 indicated that the tumor was progressing from the original clone. SMAD4, PREX2, and RNF43 are involved in the TGF-β, PI3K, and Wnt signaling pathways, respectively. These pathways are involved in cell death resistance, growth suppression, and sustained proliferative signaling, respectively. Several studies have reported chemoresistance and anti-EGFR resistance in patients with SMAD4-mutated mCRC (16–18), leading to poor prognosis. Furthermore, SMAD4 and RNF43 mutations were more commonly observed in early onset mCRC (≤55 years) (19). NGS could be considered as the best choice to analyze genetic tests for anti-EGFR therapy because it is superior to ARMS-PCR in copy number detection. Moreover, it saves the amount of samples, is cost- and time-efficient, and has great potential for clinical application to expand testing to include mutations in RAS and other less reported CRC-related genes (20). NGS analysis must be carried out in early onset CRCs, family history-related CRCs, and CRC-related cancers to identify cause-effective mutations, elucidate the clinical diagnosis, guide decision-making, and prevent the development of the disease in other family members. Furthermore, blood-based NGS testing suggested low-abundance mutations (<5%) of nine newly emerged genes, indicating the emergence of tumor subcloning and heterogeneity. These genes have been previously reported to be associated with poor prognosis (21–27), but few reports on the efficacy of chemotherapy or target drugs are available. This suggests that tumors with polygenic mutations tend to be more aggressive in terms of biological behavior, leading to poor survival rates. NGS was partially proven to have a higher sensitivity and specificity for detecting mutations with low abundance than ARMS-PCR. Blood-based NGS testing—ctDNA analysis—offers a convenient way to monitor tumor progression and treatment response. Since tumor mutational profiles are highly variable from person to person, a fixed content panel may be insufficient to track the treatment response in all patients. Compared to tumor tissue biopsies, blood-based ctDNA analyses are minimally invasive and accessible for the regular follow-up of cancer patients. The amount of ctDNA can be quantified, and genetic changes can be identified (28). The ctDNA analysis improves both the specificity and sensitivity of monitoring treatment response across several tumor types. It can identify tumor recurrence potentially earlier than imaging-based diagnosis. When augmented with tumor hotspot genes, it can track acquired drug-related mutations in patients (29). In conclusion, the present case report identified a rare early onset CRC accompanied by KRAS GCN amplification. The disease progressed rapidly, and the effects of chemotherapy and anti-EGFR therapy were poor. This suggests that KRAS amplification might be responsible for the primary resistance to anti-EGFR treatment in a small proportion of patients, and targeted drugs should not only be based on RAS and BRAF mutations in CRCs. Furthermore, NGS has the advantage of discovering genes that affect the efficacy of anti-EGFR therapies, and blood-based serial ctDNA analysis provides a convenient way to monitor the efficacy and resistance mechanism of anti-EGFR MoAbs. With improvements and developments in NGS technology, the identification of copy number variations may provide more implications for the diagnosis and treatment of CRC. Further validation in a larger population is needed to establish a predictive biomarker for resistance to anti-EGFR therapy. Data Availability Statement All data generated or analysed during this study are included in this published article. Additional data and materials related to the genetic tests, pathologic reports, treatment information, and images are available for review upon reasonable request. Ethics Statement Written informed consent was obtained from the patient for the publication of any potentially identifiable images or data included in this manuscript. Author Contributions TF and TL contributed equally to this work. All authors were involved in the drafting of the manuscript. TF, TL, and YW designed the clinical treatment for the patient. ZZ, HW, LX, JL, and CY performed the clinical treatment for the patients. CW and YT provided comments and edited the manuscript to become the final version for submission. All authors contributed to the article and approved the submitted version. Funding This study was supported by Health commission of Jilin Province (NO. 2017J063) and Department of Finance of Jilin Province (NO. JLSWSRCZ2020-0048). Conflict of Interest The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. Publisher’s Note All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher. Acknowledgments We would like to thank Editage (www.editage.cn) for English language editing.

ON DAY 1, FIVE CYCLES

DrugDosageText

CC BY

34888240

2021

What was the outcome of reaction 'Disease progression'?

An Early-Onset Advanced Rectal Cancer Patient With Increased KRAS Gene Copy Number Showed A Primary Resistance to Cetuximab in Combination With Chemotherapy: A Case Report. Mutations in KRAS (codon 12/13), NRAS, BRAF V600E, and amplification of ERBB2 and MET account for 70-80% of anti-epidermal growth factor receptor (EGFR) monoclonal antibody primary resistance. However, the list of anti-EGFR monoclonal antibody primary resistance biomarkers is still incomplete. Herein, we report a case of wild-type RAS/BRAF metastatic colorectal cancer (CRC) with resistance to anti-EGFR monoclonal antibody and chemotherapy. Initially, mutation detection in postoperative tumor tissue by using amplification-refractory mutation system polymerase chain reaction indicated wild-type RAS/BRAF without point mutations, insertion deletions, or fusion mutations. Therefore, we recommended combined therapy of cetuximab and FOLFIRI after failure of platinum-based adjuvant chemotherapy, but the disease continued to progress. Next generation sequencing analysis of the postoperative tumor tissue revealed that KRAS copy number was increased and detected SMAD4, RNF43, and PREX2 mutations. This is the first case of advanced CRC with increased copy numbers of KRAS resistant to cetuximab and chemotherapy, which results in poor patient survival, and other mutated genes may be associated with the outcomes. Our findings indicate KRAS copy number alterations should also be examined, especially with anti-EGFR monoclonal antibody therapy in CRC, since it may be related with the primary resistance to these drugs. pmcIntroduction Colorectal cancer (CRC) is one of the most common causes of cancer-related deaths worldwide (1). The mean age of patients with CRC ranges from 49 to 60 years. Early onset of CRC generally refers to the onset in patients younger than 50 years of age at the time of diagnosis. It is characterized by a more advanced stage at diagnosis, poorer cell differentiation, higher prevalence of signet ring cell histology, and a primary tumor located on the left colon or rectum (2). The combination of biological monoclonal antibodies and chemotherapeutic cytotoxic drugs provides clinical benefits to patients with advanced or metastatic CRC (mCRC) (3). Bevacizumab is an anti-vascular endothelial growth factor monoclonal antibody (MoAb) used as a first-line treatment in RAS- or BRAF-mutated mCRC (4). Cetuximab and panitumumab can promote survival in patients with wild-type RAS and BRAF tumors, which target the epidermal growth factor receptor (EGFR) extracellular domain and subsequently inhibit the mitogen-activated protein kinase signaling pathway. The principal downstream effectors of EGFR activation are the RAS/RAF/MEK/ERK, PI3K/AKT/mTOR, and PLCγ/PKC pathways (5), which is a key regulator of cell proliferation, differentiation, division, and survival, and the metastatic potential of tumor cells (6). Mutations in any of the upstream genes may be transmitted to the protein through transcription or translation, resulting in abnormal activation of the signaling pathway (7). Alterations in RAS, BRAF, PIK3CA, EGFR, PTEN, and HER2 are key determinants of resistance to anti-EGFR MoAb therapies (8–10). However, little is known regarding patients harboring KRAS amplification, and data on the response to anti-EGFR treatment are lacking. Here, we present a patient with an initial diagnosis of KRAS-amplified locally advanced rectal adenocarcinoma who demonstrated no clinical response to anti-EGFR MoAb and chemotherapy, and the disease progressed rapidly. This case report indicates the potential of increased KRAS gene copy numbers (GCN) in primary resistance to anti-EGFR MoAb in patients with CRC. Case Report Case Presentation A 40-year-old man was admitted in our hospital with complaints of diarrhea for one month. Physical examination was normal, but electronic colonoscopy revealed a large polypoid mass in the rectum, situated approximately 9–15 cm from the anal verge. The lesion almost completely occluded the rectal lumen, allowing only a small endoscope to pass through. Pathology revealed an adenocarcinoma. Enhanced computed tomography (CT) of the abdomen revealed occupying lesions in the middle and upper rectum. Considering the possibility of rectal cancer, CT findings were consistent with T4aN1M0 ( Figure 1 ). Blood tests showed a slight increase in tumor markers (CA724, 7.79 U/mol; carcinoembryonic antigen, 9.16 ng/mol), creatinine (102.8 μmol/L), direct bilirubin (7.2 μmol/L), and C-reactive protein (13 mg/L). The patient underwent total resection of the rectal cancer and ileostomy on March 24, 2021. After surgery, all blood test results returned to normal. Figure 1 Imaging examinations performed before surgery. Enhanced CT scans on March 21, 2020 of abdomen revealed that occupying lesions in the middle and upper rectum, the intestinal lumen was narrowed, and the serosal layer was hairy. After enhancement, the lesion was uneven and enhanced, and the length of the lesion was about 5.7 cm, considering that was rectal cancer (T4aN1M0). Surgical pathology findings showed a moderate-to-poorly differentiated adenocarcinoma in the rectum (pT3N2bMx). The tumor volume was approximately 4 cm × 3 cm × 1.2 cm. Cancer infiltration was observed in vessels and nerves. Residual tumors can be detected on the edges of the surgical specimens. Meanwhile, 20 mesenteric lymph nodes were observed, and cancer embolus was distinguished in the local submucosal vessels. Immunohistochemistry results were as follows: pMMR, Ki-67 (+70%); CKpan (+), P53 (−), CgA (−), and Syn (−) ( Figure 2 ). Figure 2 Histopathological examination of tumor. (A) Hematoxylin and eosin (H&E) staining of colonic primary tumor shows glandular differentiation and invasion into the entire intestinal wall (×20). (B) Poorly differentiated adenocarcinoma on the surface of mucosa with multiple intravascular thrombus in submucosa, (×40). (C) Metastatic tumors can be seen in lymph node, (×40). Immunohistochemical staining of tumor cells. Ki-67 partial expression in tumor (+70%) (D, ×100), CKpan expression in tumor (E, ×100.), Immunostaining for MLH1 (+70%) (F, ×100), MSH2 (+70%) (G, ×100), MSH6 (+80%) (H, ×100)and PMS2 (+50%) (I, ×100) confirmed the tumor with proficient MMR. Enhanced pelvic Magnetic Resonance Imaging (MRI) was performed 21 days after the surgery and revealed multiple lymph node shadows with visible enhancement beside bilateral iliac vessels with a short diameter of approximately 0.6–0.7 cm, but postoperative inflammatory changes were not excluded. Therefore, five cycles of adjuvant chemotherapy (XELOX: oxaliplatin, 130 mg/m2 on day 1; capecitabine, 1,000 mg/m2 twice daily on days 1–14, orally) strategy was recommended from April 14 to July 28, 2020. The CT and MRI evaluations were progressive disease (PD) after chemotherapy, detecting significantly enlarged lymph nodes around the abdominal aorta (0.6-1.5cm) on August 12, 2020 ( Figure 3A ). Figure 3 Imaging examinations performed after chemotherapy. (A) After five cycles of XELOX, CT scans on August 12, 2020 revealed enlargement of the abdominal para-aortic and bilateral para-iliac lymph nodes, about 0.5–1.5 cm, with slightly enhanced. (B) After six cycles of FOLFIRI plus cetuximab treatment, CT scans on November 28, 2020 revealed significant enlargement of lymph nodes in the hilar area, adjacent to the abdominal aorta, and bilateral iliac vessels, with a maximum of about 2.3 cm and slightly enhancement. (C) MRI scans on November 29, 2020 show patchy shadows are seen on the right femoral head and greater trochanter, enhanced scans are seen to be enhanced, consider metastasis. (D) CT scans on December 22, 2020 showed hydronephrodilation of the right renal pelvis and space-occupying lesion about 16.2 mm × 9.3 mm in the initial segment of the right ureter. Since the first-line platinum-based regiment was unsuccessful, chemotherapy (FOLFIRI: 5-fluorouracil 400 mg/m2, IV bolus, folinic acid 400 mg/m2, and irinotecan 180 mg/m2 on day 1, followed by a continuous 46-h infusion of fluorouracil 2,400 mg/m2) in combination with cetuximab (500 mg/m² biweekly on day 1) was administered for six cycles from August 17 to November 12, 2020, considering that the patient harbored no RAS or BRAF alterations according to the mutation profile. After three cycles of chemotherapy and targeted therapy, the patient developed a grade 2 cutaneous toxicity, a single skin impetigo with a diameter greater than 10 mm, and the curative effect was evaluated as stable disease (SD), so we continued the therapy for three more cycles. On November 28, 2020, CT and MRI scans indicated PD, detecting a lymph node around the abdominal aorta enlarged to 2.3 cm ( Figure 3B ), tumor metastases to the right femoral head and greater trochanter ( Figure 3C ), and left cervical lymph nodes enlarged to 11 mm × 8 mm. Likewise, abdominal CT scans indicated PD again, observing a space-occupying lesion (1.6 cm × 0.9 cm) in the initial segment of the right ureter on December 22, 2020 ( Figure 3D ). Renal function tests suggested that uric acid and creatinine levels continuously increased to 470 µmol/L (normal: 210–430 µmol/L) and 226 µmol/L (normal: 57–97 µmol/L), respectively. The patient decided to stop the treatment and never returned to the hospital after four days of palliative radiotherapy. Mutation Analysis Wild-type KRAS, NRAS, and BRAF were amplified using amplification refractory mutation system polymerase chain reaction (ARMS-PCR) performed on March 20, 2020. Results revealed that the patient had wild-type RAS and BRAF. However, the tumor failed to respond to both treatment options and continued to progress. Molecular characterization of the blood and postoperative tumor DNA of the patient was performed using next-generation sequencing (NGS) on December 1, 2020. Sequence analyses are shown in Table 1 . We detected that the patient had mutations in three genes (SMAD4, RNF43, and PREX2), and GCN variations of KRAS increased to 7.8 gene copies before chemotherapy. In addition, the mutation abundance of the three mutant genes was observed using blood-based circulating tumor DNA (ctDNA) analysis: SMAD4 increased from 38.69 to 58.0%, RNF43 increased from 41.25 to 48.03%, and PREX2 increased from 39.22 to 52.99%, after the failure of second-line treatment. Likewise, the GCN of KRAS also increased to 8.31 gene copies, along with low abundance mutations in the other nine newly confirmed genes—CTNNB1, EPHA5, ETV1, FANCA, JAK2, MYC, PALB2, PIK3CA, and SPEN. Table 1 The results of sequence analyses of paraffin sections and ctDNA on December 1, 2020. Gene Mutation site Mutation abundance (paraffin sections) Mutation abundance (ctDNA) SMAD4 EXON:2 c.136A>T p.K46* 38.69% 58.0% RNF43 EXON:3 c.354dupC p.C119Lfs*6 41.25% 48.03% PREX2 EXON:28 c.3482G>T p.C1161F 39.22% 52.99% EXON:23 c.2558T>G p. L853R – 1.08% SPEN EXON:11 c.9239C>T p.P3080L – 1.66% PIK3CA EXON:14 c.2115A>C p.Q705H – 0.69% PALB2 EXON:4 c.212A>G p.E71G – 1.61% MYC EXON:3 c.1141A>T p.R381W – 0.78% JAK2 EXON:9 c.1121G>A p.R374K – 0.43% FANCA EXON:35 c.3458A>C p.D1153A – 0.45% ETV1 EXON:8 c.586T>G p.S196A – 2.82% EPHA5 EXON:14 c.2363T>C p.M788T – 0.78% EXON:14 c.2446G>C p.V816L – 0.64% CTNNB1 EXON:3 c.134C>G p.S45C – 3.76% Microsatellite Instability and Tumor Mutation Burden We analyzed MSI through surgically resected tumor tissues, and the results were reported as microsatellite stability (MSS)/MSI-low. The TMB detected in the tumor tissues was described as low level (2.95 mutations/MB). Furthermore, the TMB detected in ctDNA was reported as medium level (11.82 mutations/MB) after the failure of second-line therapy. Discussion Here, we present a patient with an early onset CRC with an initial diagnosis of moderate-to-poorly differentiated locally advanced rectal adenocarcinoma. Unlike the traditional RAS/BRAF mutant CRCs with poor prognosis, the patient had RAS/BRAF wild-type CRC accompanied with KRAS amplification and other less-reported gene mutations. Moreover, the patient’s prognosis was extremely poor, the disease progressed rapidly even with the use cetuximab in combination with chemotherapy. Therefore, we hypothesized that the disease in our patient with early onset CRC was more aggressive due to gene alterations, especially the increase in copy number of KRAS. Therefore, we report this case and hope to bring clinical benefits to clinicians. After the rapid progress of second-line treatment in our case, blood tests and postoperative tumor tissues were performed and analyzed using NGS and retrospective NGS, respectively. Results suggested an increase in KRAS copy number before treatment, accompanied by SMAD4, RNF43, and PREX2 mutations (mutation abundance was much more than 5%). After treatment failure, blood analysis indicated that the copy number of KRAS and clonal mutation abundance of the three genes continued to increase, indicating that the tumor developed from the original clone. KRAS amplification in CRC is a rare event, with an overall prevalence of 0.67–2% (11, 12). Previous studies have identified somatic mutations in KRAS as biomarkers for inherent resistance to EGFR-targeted drugs in patients with CRC (13), with a positive tissue mutation rate of 32–52.1% (7). However, alterations are responsive to the anti-EGFR inhibitors cetuximab or panitumumab (11, 14). GCN amplification of KRAS leads to the activation of RAS-RAF-ERK or PIK-AKT-mTOR pathways, even without additional activating mutations in the genes (15). In the presence of KRAS amplification, cetuximab can partially eliminate the phosphorylation of MEK and ERK but cannot induce growth arrest (13). A retrospective clinical study conducted by Valtorta et al. (14) detected that all four KRAS-amplified cases were found among the 53 patients who were resistant to anti-EGFR antibodies, while none of the 44 responders had tumors carrying this molecular alteration. Similarly, tumor biopsies of 10 patients who developed resistance to anti-EGFR showed the emergence of KRAS amplification in one case and acquisition of secondary KRAS mutations in six cases (13). Another retrospective study by Favazza et al. (11) reported that all eight patients with KRAS-amplified mCRC showed disease progression at the time of anti-EGFR therapy, and they concluded that KRAS amplification is responsible for the primary resistance to EGFR inhibitors. Meanwhile, RAS amplifications (involving KRAS, NRAS, and HRAS) were correlated with a younger median patient age at initial diagnosis and a history of inflammatory bowel disease (11). In the present case, KRAS amplification was present before chemotherapy, resulting in resistance to cetuximab. Therefore, patients with advanced mCRC should be monitored not only based on the RAS/BRAF mutation status but also RAS amplification, and patients with CRC that do not respond to anti-EGFR MoAbs should be excluded. Moreover, we suggest that genetic testing using NGS should be performed in younger patients to guide decision-making and prolong overall survival, because KRAS amplification is more likely to occur in younger patients. The mutation frequency and abundance of clonal mutations of SMAD4, RNF43, and PREX2 indicated that the tumor was progressing from the original clone. SMAD4, PREX2, and RNF43 are involved in the TGF-β, PI3K, and Wnt signaling pathways, respectively. These pathways are involved in cell death resistance, growth suppression, and sustained proliferative signaling, respectively. Several studies have reported chemoresistance and anti-EGFR resistance in patients with SMAD4-mutated mCRC (16–18), leading to poor prognosis. Furthermore, SMAD4 and RNF43 mutations were more commonly observed in early onset mCRC (≤55 years) (19). NGS could be considered as the best choice to analyze genetic tests for anti-EGFR therapy because it is superior to ARMS-PCR in copy number detection. Moreover, it saves the amount of samples, is cost- and time-efficient, and has great potential for clinical application to expand testing to include mutations in RAS and other less reported CRC-related genes (20). NGS analysis must be carried out in early onset CRCs, family history-related CRCs, and CRC-related cancers to identify cause-effective mutations, elucidate the clinical diagnosis, guide decision-making, and prevent the development of the disease in other family members. Furthermore, blood-based NGS testing suggested low-abundance mutations (<5%) of nine newly emerged genes, indicating the emergence of tumor subcloning and heterogeneity. These genes have been previously reported to be associated with poor prognosis (21–27), but few reports on the efficacy of chemotherapy or target drugs are available. This suggests that tumors with polygenic mutations tend to be more aggressive in terms of biological behavior, leading to poor survival rates. NGS was partially proven to have a higher sensitivity and specificity for detecting mutations with low abundance than ARMS-PCR. Blood-based NGS testing—ctDNA analysis—offers a convenient way to monitor tumor progression and treatment response. Since tumor mutational profiles are highly variable from person to person, a fixed content panel may be insufficient to track the treatment response in all patients. Compared to tumor tissue biopsies, blood-based ctDNA analyses are minimally invasive and accessible for the regular follow-up of cancer patients. The amount of ctDNA can be quantified, and genetic changes can be identified (28). The ctDNA analysis improves both the specificity and sensitivity of monitoring treatment response across several tumor types. It can identify tumor recurrence potentially earlier than imaging-based diagnosis. When augmented with tumor hotspot genes, it can track acquired drug-related mutations in patients (29). In conclusion, the present case report identified a rare early onset CRC accompanied by KRAS GCN amplification. The disease progressed rapidly, and the effects of chemotherapy and anti-EGFR therapy were poor. This suggests that KRAS amplification might be responsible for the primary resistance to anti-EGFR treatment in a small proportion of patients, and targeted drugs should not only be based on RAS and BRAF mutations in CRCs. Furthermore, NGS has the advantage of discovering genes that affect the efficacy of anti-EGFR therapies, and blood-based serial ctDNA analysis provides a convenient way to monitor the efficacy and resistance mechanism of anti-EGFR MoAbs. With improvements and developments in NGS technology, the identification of copy number variations may provide more implications for the diagnosis and treatment of CRC. Further validation in a larger population is needed to establish a predictive biomarker for resistance to anti-EGFR therapy. Data Availability Statement All data generated or analysed during this study are included in this published article. Additional data and materials related to the genetic tests, pathologic reports, treatment information, and images are available for review upon reasonable request. Ethics Statement Written informed consent was obtained from the patient for the publication of any potentially identifiable images or data included in this manuscript. Author Contributions TF and TL contributed equally to this work. All authors were involved in the drafting of the manuscript. TF, TL, and YW designed the clinical treatment for the patient. ZZ, HW, LX, JL, and CY performed the clinical treatment for the patients. CW and YT provided comments and edited the manuscript to become the final version for submission. All authors contributed to the article and approved the submitted version. Funding This study was supported by Health commission of Jilin Province (NO. 2017J063) and Department of Finance of Jilin Province (NO. JLSWSRCZ2020-0048). Conflict of Interest The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest. Publisher’s Note All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher. Acknowledgments We would like to thank Editage (www.editage.cn) for English language editing.

Not recovered

ReactionOutcome

CC BY

34888240

2021

Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Drug ineffective'.

Aortic Root Thrombus Directly After Left Ventricular Assist Device Implantation. A 70-year-old female heart failure patient could not be weaned from temporary left ventricular mechanical support with Impella CP (Abiomed Inc, Danvers, MA) after myocardial infarction; therefore, she underwent left ventricular assist device implantation (HeartMate 3; Abbott, Chicago, IL). After uneventful surgery, the patient had an early postoperative thrombus in the aortic root, and surgical thrombectomy on extracorporeal circulation was performed on the seventh postoperative day. The patient recovered well and presented in good condition with no neurologic symptoms at the 6-month follow-up visit. Surgical excision of aortic root thrombus is a feasible option even for frail patients with a left ventricular assist device. pmcPatients with acute and chronic heart failure benefit from device innovations for permanent and temporary mechanical circulatory support. Although technological advancements continuously improved outcome with left ventricular assist devices (LVAD),1 modern temporary mechanical left ventricular support with Impella (Abiomed Inc, Danvers, MA) might increase the number of potential candidates for LVAD but also challenge heart teams with new complications2 as described in this case report. Case A 70-year-old female patient presented to an outside hospital with pain in the upper abdomen, both shoulders, and the left arm as well as clamminess. The electrocardiogram (ECG) showed a myocardial infarction of the posterior wall with ST-deviations in V2-V4. (Supplemental Fig. S1). Emergent cardiac catheterization found an ejection fraction of 15% and 3-vessel disease, and drug-eluting stents were placed into the left circumflex and marginal arteries (Supplemental Fig. S2). The patient had cardiogenic shock, and an intra-aortic balloon pump (IABP) (Cardiosave IABP Hybrid; MAQUET, Wayne, NJ, USA) was implanted via femoral artery access. The patient was transferred to another outside hospital at her own request. Full revascularization was obtained by stenting the right and left anterior descending coronary arteries. During this interventional procedure, the patient experienced hemodynamic instability caused by tachycardic atrial fibrillation (160 bpm); therefore, the IAPB was replaced by an Impella CP (Abiomed Inc) via femoral access on day 2. After 2 days, additional stents were placed into the right and left coronary arteries. Transthoracic echocardiography found left ventricular hypertrophy with severely impaired left ventricular ejection fraction of 15% and moderate mitral valve regurgitation. New-onset atrial fibrillation with rapid ventricular response was treated with amiodarone intravenously (1 g/d). Because of massive bleeding from the arterial femoral puncture site, the patient went into haemorrhagic shock and was transferred to a tertiary care centre. Immediate computed tomography found a retroperitoneal haematoma without active bleeding, and the patient was weaned successfully from Impella support on the sixth day. On the tenth day, the patient suffered from electrical storm with scar-related monomorphic premature ventricular contractions, repetitively inducing ventricular tachycardia requiring 25 electrical cardioversions and in-hospital resuscitation. During the electrophysiologic study, isolated potentials and fractioned signals in the area of the infero-lateral and antero-latero-basal scars were mapped (CARTO3; Biosense Webster, Inc, Irvine, CA), and radiofrequency catheter ablation was performed. At the end of the procedure, a reimplantation of an Impella CP was performed owing to left ventricular end-diastolic pressure of 42 mm Hg. Two days later, the patient showed a single episode of ventricular fibrillation under Impella support. After chest compressions for 5 minutes and 1 defibrillation with 200 joules, the patient had a normocardic sinus rhythm. Because weaning from the Impella device was unsuccessful due to development of pulmonary edema under flow reduction (Supplemental Fig. S3), the heart team opted for an early implant of an LVAD (bridge-to-destination) in combination with a mitral valve repair. Uneventful explantation of the Impella CP with simultaneous implantation of a Heart Mate 3 device (HeartMate 3; Abbott, Chicago, IL), and concomitant mitral valve ring annuloplasty (Medtronic Simulus Semi-Rigid Annuloplasty ring, 30 mm; Münchenbuchsee, Switzerland) were performed on day 21. Cardiopulmonary bypass (CPB) time was 131 minutes, aortic cross clamp time was 30 minutes. Anticoagulation therapy (heparin 32.500 IU/d) was started 24 hours postoperatively, and inotropic medication was stopped 36 hours after surgery. Under hemodynamic stability (pump speed, 5200 RPM; flow, 4.4 L/min; power, 3.5 watt), the patient was extubated on day 24. The postoperative transoesophageal echocardiography found a previously undetected patent foramen ovale and a new thrombus in the left coronary cusp of the aortic valve (Fig. 1). Although the dose of intravenous heparin was increased, the thrombus size could not be reduced during the following days.Figure 1 On the third day postimplant of the left ventricular assist device, a thrombus (white arrows) in the left coronary cusp of the aortic valve was observed on transesophageal echocardiography in the mid-esophageal aortic valve (A) long-axis and (B) short-axis views. Also highlighted are the left atrium (1), ascending aorta (2), right atrium (3), right ventricular outflow tract (4), right coronary cusp (rcc) and noncoronary cusp (ncc). Figure 1 On day 29, surgical thrombectomy was performed via re-sternotomy (CPB time, 49 minutes; aortic cross clamp time, 19 minutes). Closure of the patent foramen ovale was performed, and the thrombus in the left coronary cusp of the aortic valve (Fig. 2) was excised in total. No obvious pathologic condition of the native aortic valve was observed during the operation. Two days after thrombectomy, the patient presented with temporary neurologic symptoms of the right side (hanging right corner of the mouth, weakness of the right arm), which correlated with 2 left cerebellar infarctions in computed tomography. Four weeks later, neurologic dysfunctions could no longer be reproduced.Figure 2 The thrombus formation is visible in the aortic sinus. The thrombus with an approximate diameter of 1 cm (A, white arrow) is visible within the aortic root sinus. The outflow graft (1), aortic wall (2), and right ventricle (3) are also shown. Figure 2 Permanent oral anticoagulation therapy with phenprocoumon (target international normalized ratio, 2-3) was initiated in combination with 100 mg of acetylsalicylic acid (100 mg/d). One and 2 weeks after surgical thrombectomy, echocardiography-guided ramp tests were performed to optimize LVAD speed to improve left ventricular unloading while still enabling sporadic opening of the aortic valve to minimize rethrombosis. During cardiac rehabilitation, the patient's 6-minute walking distance improved from 120 m to 300 m, and the MacNew Heart Disease quality of life score3 improved from 5.9 to 6.6 points. A timeline of the case is shown in Supplemental Figure S4. Discussion Substantial efforts have been made to improve the outcomes of patients with LVAD; nevertheless, our case highlights the need for sustained peri- and postoperative alertness, even after uneventful implantation. A potential explanation for thrombus formation in the aortic root might be the prolonged support with Impella CP, which induces low flow between the blood inlet and outlet area and potentially prevents cusp movement. Data on the development of aortic root thrombus after Impella support and LVAD implantation are still scarce.4,5 Due to the lack of recommendations regarding the management of aortic root thrombus, different therapeutic approaches were discussed: (1) conservative management, (2) catheter-based local thrombolysis, and (3) reoperation with removal of thrombus. The rationale for reoperation in our case was the anticipated lower bleeding risk compared with lysis. Whether this approach is also associated with a decreased stroke risk is currently unknown. We admit that the opinions within the heart team were diverse, and the decision to perform redo surgery demanded multiple discussions preoperatively as well as postoperatively. Our case highlights that peri- and postinterventional echocardiography should include a focus on the aortic root because bridging with an Impella device preceding the LVAD implantation might potentially predispose patients to aortic root thrombus formation caused by Impella-related injury to the aortic valve and aortic root stasis. Heart teams need to be aware of early postoperative thrombus formation that might potentially be formed in the aortic root after LVAD implantation, especially when patients have undergone previous Impella support. In our case, redo surgery via aortotomy on CPB was successful even in a frail bridge-to-destination patient, but whether this approach is superior to conservative treatment is unknown. Postoperative echocardiography should be performed during the early postimplantation period to evaluate the aortic root. In addition, further studies are needed to evaluate the risk for aortic root thrombosis under mechanical left ventricular support with Impella devices. Further studies are needed to evaluate the risk for aortic root thrombosis under mechanical left ventricular support with Impella devices. Novel Teaching Points • Aortic root thrombus is a potential complication after treatment with a left ventricular Impella device. • Early postoperative echocardiography after LVAD implantation should evaluate for aortic root thrombus. • Surgical excision of aortic root thrombus is a feasible option even in frail LVAD patients. • LVAD patients are challenging and demand close monitoring, especially in the early and postoperative phase by specialized multidisciplinary heart teams. Funding Sources This research received no external funding. Disclosures David Santer received speaker honoraria and educational grants from Abbott and speaker honoraria from Nycomed GmbH. Michael Kühne reports personal fees from Bayer, Boehringer Ingelheim, Pfizer BMS, Daiichi Sankyo, Medtronic, Biotronik, Boston Scientific, and Johnson & Johnson and grants from Bayer, Pfizer, Boston Scientific, BMS, and Biotronik. The other authors report no financial support and no other potential conflict of interest relevant to this article. Appendix Supplementary materials Image, application 1 Ethics Statement: The reported research adhered to the Declaration of Helsinki. See page 1315 for disclosure information. To access the supplementary material accompanying this article, visit CJC Open at https://www.cjcopen.ca/ and at doi:10.1016/j.cjco.2021.05.016.

HEPARIN SODIUM

DrugsGivenReaction

CC BY-NC-ND

34888513

2021-10

What was the administration route of drug 'HEPARIN SODIUM'?

Aortic Root Thrombus Directly After Left Ventricular Assist Device Implantation. A 70-year-old female heart failure patient could not be weaned from temporary left ventricular mechanical support with Impella CP (Abiomed Inc, Danvers, MA) after myocardial infarction; therefore, she underwent left ventricular assist device implantation (HeartMate 3; Abbott, Chicago, IL). After uneventful surgery, the patient had an early postoperative thrombus in the aortic root, and surgical thrombectomy on extracorporeal circulation was performed on the seventh postoperative day. The patient recovered well and presented in good condition with no neurologic symptoms at the 6-month follow-up visit. Surgical excision of aortic root thrombus is a feasible option even for frail patients with a left ventricular assist device. pmcPatients with acute and chronic heart failure benefit from device innovations for permanent and temporary mechanical circulatory support. Although technological advancements continuously improved outcome with left ventricular assist devices (LVAD),1 modern temporary mechanical left ventricular support with Impella (Abiomed Inc, Danvers, MA) might increase the number of potential candidates for LVAD but also challenge heart teams with new complications2 as described in this case report. Case A 70-year-old female patient presented to an outside hospital with pain in the upper abdomen, both shoulders, and the left arm as well as clamminess. The electrocardiogram (ECG) showed a myocardial infarction of the posterior wall with ST-deviations in V2-V4. (Supplemental Fig. S1). Emergent cardiac catheterization found an ejection fraction of 15% and 3-vessel disease, and drug-eluting stents were placed into the left circumflex and marginal arteries (Supplemental Fig. S2). The patient had cardiogenic shock, and an intra-aortic balloon pump (IABP) (Cardiosave IABP Hybrid; MAQUET, Wayne, NJ, USA) was implanted via femoral artery access. The patient was transferred to another outside hospital at her own request. Full revascularization was obtained by stenting the right and left anterior descending coronary arteries. During this interventional procedure, the patient experienced hemodynamic instability caused by tachycardic atrial fibrillation (160 bpm); therefore, the IAPB was replaced by an Impella CP (Abiomed Inc) via femoral access on day 2. After 2 days, additional stents were placed into the right and left coronary arteries. Transthoracic echocardiography found left ventricular hypertrophy with severely impaired left ventricular ejection fraction of 15% and moderate mitral valve regurgitation. New-onset atrial fibrillation with rapid ventricular response was treated with amiodarone intravenously (1 g/d). Because of massive bleeding from the arterial femoral puncture site, the patient went into haemorrhagic shock and was transferred to a tertiary care centre. Immediate computed tomography found a retroperitoneal haematoma without active bleeding, and the patient was weaned successfully from Impella support on the sixth day. On the tenth day, the patient suffered from electrical storm with scar-related monomorphic premature ventricular contractions, repetitively inducing ventricular tachycardia requiring 25 electrical cardioversions and in-hospital resuscitation. During the electrophysiologic study, isolated potentials and fractioned signals in the area of the infero-lateral and antero-latero-basal scars were mapped (CARTO3; Biosense Webster, Inc, Irvine, CA), and radiofrequency catheter ablation was performed. At the end of the procedure, a reimplantation of an Impella CP was performed owing to left ventricular end-diastolic pressure of 42 mm Hg. Two days later, the patient showed a single episode of ventricular fibrillation under Impella support. After chest compressions for 5 minutes and 1 defibrillation with 200 joules, the patient had a normocardic sinus rhythm. Because weaning from the Impella device was unsuccessful due to development of pulmonary edema under flow reduction (Supplemental Fig. S3), the heart team opted for an early implant of an LVAD (bridge-to-destination) in combination with a mitral valve repair. Uneventful explantation of the Impella CP with simultaneous implantation of a Heart Mate 3 device (HeartMate 3; Abbott, Chicago, IL), and concomitant mitral valve ring annuloplasty (Medtronic Simulus Semi-Rigid Annuloplasty ring, 30 mm; Münchenbuchsee, Switzerland) were performed on day 21. Cardiopulmonary bypass (CPB) time was 131 minutes, aortic cross clamp time was 30 minutes. Anticoagulation therapy (heparin 32.500 IU/d) was started 24 hours postoperatively, and inotropic medication was stopped 36 hours after surgery. Under hemodynamic stability (pump speed, 5200 RPM; flow, 4.4 L/min; power, 3.5 watt), the patient was extubated on day 24. The postoperative transoesophageal echocardiography found a previously undetected patent foramen ovale and a new thrombus in the left coronary cusp of the aortic valve (Fig. 1). Although the dose of intravenous heparin was increased, the thrombus size could not be reduced during the following days.Figure 1 On the third day postimplant of the left ventricular assist device, a thrombus (white arrows) in the left coronary cusp of the aortic valve was observed on transesophageal echocardiography in the mid-esophageal aortic valve (A) long-axis and (B) short-axis views. Also highlighted are the left atrium (1), ascending aorta (2), right atrium (3), right ventricular outflow tract (4), right coronary cusp (rcc) and noncoronary cusp (ncc). Figure 1 On day 29, surgical thrombectomy was performed via re-sternotomy (CPB time, 49 minutes; aortic cross clamp time, 19 minutes). Closure of the patent foramen ovale was performed, and the thrombus in the left coronary cusp of the aortic valve (Fig. 2) was excised in total. No obvious pathologic condition of the native aortic valve was observed during the operation. Two days after thrombectomy, the patient presented with temporary neurologic symptoms of the right side (hanging right corner of the mouth, weakness of the right arm), which correlated with 2 left cerebellar infarctions in computed tomography. Four weeks later, neurologic dysfunctions could no longer be reproduced.Figure 2 The thrombus formation is visible in the aortic sinus. The thrombus with an approximate diameter of 1 cm (A, white arrow) is visible within the aortic root sinus. The outflow graft (1), aortic wall (2), and right ventricle (3) are also shown. Figure 2 Permanent oral anticoagulation therapy with phenprocoumon (target international normalized ratio, 2-3) was initiated in combination with 100 mg of acetylsalicylic acid (100 mg/d). One and 2 weeks after surgical thrombectomy, echocardiography-guided ramp tests were performed to optimize LVAD speed to improve left ventricular unloading while still enabling sporadic opening of the aortic valve to minimize rethrombosis. During cardiac rehabilitation, the patient's 6-minute walking distance improved from 120 m to 300 m, and the MacNew Heart Disease quality of life score3 improved from 5.9 to 6.6 points. A timeline of the case is shown in Supplemental Figure S4. Discussion Substantial efforts have been made to improve the outcomes of patients with LVAD; nevertheless, our case highlights the need for sustained peri- and postoperative alertness, even after uneventful implantation. A potential explanation for thrombus formation in the aortic root might be the prolonged support with Impella CP, which induces low flow between the blood inlet and outlet area and potentially prevents cusp movement. Data on the development of aortic root thrombus after Impella support and LVAD implantation are still scarce.4,5 Due to the lack of recommendations regarding the management of aortic root thrombus, different therapeutic approaches were discussed: (1) conservative management, (2) catheter-based local thrombolysis, and (3) reoperation with removal of thrombus. The rationale for reoperation in our case was the anticipated lower bleeding risk compared with lysis. Whether this approach is also associated with a decreased stroke risk is currently unknown. We admit that the opinions within the heart team were diverse, and the decision to perform redo surgery demanded multiple discussions preoperatively as well as postoperatively. Our case highlights that peri- and postinterventional echocardiography should include a focus on the aortic root because bridging with an Impella device preceding the LVAD implantation might potentially predispose patients to aortic root thrombus formation caused by Impella-related injury to the aortic valve and aortic root stasis. Heart teams need to be aware of early postoperative thrombus formation that might potentially be formed in the aortic root after LVAD implantation, especially when patients have undergone previous Impella support. In our case, redo surgery via aortotomy on CPB was successful even in a frail bridge-to-destination patient, but whether this approach is superior to conservative treatment is unknown. Postoperative echocardiography should be performed during the early postimplantation period to evaluate the aortic root. In addition, further studies are needed to evaluate the risk for aortic root thrombosis under mechanical left ventricular support with Impella devices. Further studies are needed to evaluate the risk for aortic root thrombosis under mechanical left ventricular support with Impella devices. Novel Teaching Points • Aortic root thrombus is a potential complication after treatment with a left ventricular Impella device. • Early postoperative echocardiography after LVAD implantation should evaluate for aortic root thrombus. • Surgical excision of aortic root thrombus is a feasible option even in frail LVAD patients. • LVAD patients are challenging and demand close monitoring, especially in the early and postoperative phase by specialized multidisciplinary heart teams. Funding Sources This research received no external funding. Disclosures David Santer received speaker honoraria and educational grants from Abbott and speaker honoraria from Nycomed GmbH. Michael Kühne reports personal fees from Bayer, Boehringer Ingelheim, Pfizer BMS, Daiichi Sankyo, Medtronic, Biotronik, Boston Scientific, and Johnson & Johnson and grants from Bayer, Pfizer, Boston Scientific, BMS, and Biotronik. The other authors report no financial support and no other potential conflict of interest relevant to this article. Appendix Supplementary materials Image, application 1 Ethics Statement: The reported research adhered to the Declaration of Helsinki. See page 1315 for disclosure information. To access the supplementary material accompanying this article, visit CJC Open at https://www.cjcopen.ca/ and at doi:10.1016/j.cjco.2021.05.016.

Intravenous (not otherwise specified)

DrugAdministrationRoute

CC BY-NC-ND

34888513

2021-10

Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Anaesthetic complication pulmonary'.

Esophageal achalasia detected by vomiting during induction of general anesthesia: a case report. BACKGROUND Esophageal achalasia is a rare disease with a high risk of aspiration during anesthesia induction. Here, we describe our experience involving a case of undiagnosed esophageal achalasia with profuse vomiting during anesthesia induction. METHODS A 58-year-old woman was scheduled for orthopedic surgery under general anesthesia. She vomited a large amount of watery contents during anesthesia induction, and planned surgery was postponed. After recovery from anesthesia, she informed us that she usually had to drink a large amount of water to get food into her stomach and purged watery vomit every night before sleep. However, she attributed it to her constitutional problem, not to a specific disease. She was subsequently diagnosed with esophageal achalasia and underwent Heller myotomy with Dor fundoplication before her re-scheduled orthopedic surgery. CONCLUSIONS A detailed history of dysphagia and regurgitation should be taken in preoperative examinations to prevent unexpected aspiration due to undiagnosed achalasia. pmcBackground Esophageal achalasia is a rare disease (incidence, 1 in 100,000) associated with a high risk of aspiration during the induction of general anesthesia due to difficulties in passing food [1]. Here, we present a patient with undiagnosed esophageal achalasia who vomited profusely during the induction of general anesthesia. Case presentation A 58-year-old woman (height 163 cm, weight 52 kg, body mass index 19.6) with a diagnosis of spinal canal stenosis was admitted to an orthopedic hospital to undergo posterior lumbar interbody fusion. Her medical history included anterior cervical fusion 12 years previously without any anesthetic complications such as allergy or bronchial asthma. No full-time anesthesiologist was on staff at the hospital, so a part-time anesthesiologist was in charge of the anesthesia. The patient had been fasting for 15 h, from dinner the night before surgery until entering the operating room (OR). Four hours before entering the OR, an intravenous infusion was started. The patient was administered ranitidine (150 mg orally) with a small amount of water 180 min before entering the OR. Immediately before starting the anesthesia induction, the anesthesiologist reconfirmed the patient’s medical history and whether she had any allergies or bronchial asthma. The patient did not provide any new medical information. Electrocardiography, pulse oximetry (SpO2), noninvasive blood pressure, and capnometry were applied. After preoxygenation with 100% oxygen for 3 min, general anesthesia was induced with fentanyl 50 μg and propofol 100 mg. After confirming the patient’s spontaneous breathing cessation, assisted ventilation was initiated with a face mask, followed by rocuronium 50 mg. Immediately after that, the patient threw up a massive amount of watery vomit. The vomit in her oral cavity was immediately suctioned, and then her trachea was intubated. A small amount of watery vomit was suctioned through the endotracheal tube. Lung compliance was low with manual ventilation, and wheezing was heard on auscultation with an obstructive pattern on capnography. Bronchospasm due to aspiration was suspected, and treatment was initiated. Under anesthesia (sevoflurane 2% in oxygen 100%), her SpO2 remained above 98%. Hydrocortisone 100 mg and aminophylline 250 mg were administered by drip infusion. Forty-five minutes after the event, her wheezing disappeared, and mostly normal lung compliance returned. The scheduled surgery was postponed, the patient was awakened from anesthesia with the aid of sugammadex to antagonize rocuronium, and her trachea was extubated. After returning to the ward, a more detailed medical history was obtained from the patient. She said that she usually had to drink a large amount of water to get food into her stomach. Therefore, she made it a daily habit to purge watery vomit before bed, as it would reflux when she lay down. Although she had an upper gastrointestinal endoscopy a few years prior, the physician had not pointed out anything unusual. Since then, she believed that this was a constitutional problem as opposed to a disease. The patient had not experienced any weight loss for the past few years. Although no clear image of pneumonia was seen on a chest X-ray at the ward, ampicillin and sulbactam were administered for prevention. The patient was discharged the next day without any medical problems. Based on an overall assessment of the episode, we believed that this patient had esophageal achalasia and advised her to visit a gastroenterologist. At the gastroenterology department in another hospital, fluid retention in the dilated esophagus was revealed by upper endoscopy and chest computed tomography (CT) (Fig. 1). The esophagogram showed a dilated esophagus with a diameter of 32 mm and functional narrowing of the esophagogastric junction (Fig. 2). Esophageal manometry showed the disappearance of primary peristaltic waves and the occurrence of simultaneous contraction waves. Based on these results, the patient was diagnosed with esophageal achalasia and underwent laparoscopic Heller myotomy with Dor fundoplication surgery. After the surgery, her subjective symptoms, including choking with food and daily vomiting before bed, disappeared. She was discharged on the sixth day after surgery. Three months later, she underwent posterior interbody fusion of the lumbar spine without any anesthetic complications.Fig. 1 Preoperative thoracic computed tomography showing fluid retention in the dilated esophagus. a Upper thoracic region. b Middle thoracic region Fig. 2 Preoperative esophagogram showed a dilated esophagus and functional narrowing of the esophagogastric junction. A mouthful of contrast medium was retained in the esophagus until the end of the imaging Discussion Esophageal achalasia is defined as an esophageal motility disorder of unknown etiology characterized by failure of lower esophageal sphincter (LES) relaxation and impaired peristalsis of the lower esophageal body [2]. Diagnosis is difficult because achalasia is a rare disease (incidence, 1 in 100,000 people) with nonspecific subjective symptoms [3]. The most frequent symptoms of achalasia include dysphagia of solids and liquids (> 90%), regurgitation of undigested food (76–91%), respiratory complications such as nocturnal cough (30%) and aspiration (8%), chest pains (25–64%), heartburn (18–52%), and weight loss (35–91%) [4]. Oral reflux in patients with achalasia does not originate from the stomach and thus does not contain acidic contents. Furthermore, if the volume of the oral reflux is mild, the patient may be unaware of the reflux. We assume two reasons for the aspiration during the first scheduled orthopedic surgery in this case. First, the patient was not aware of her rare disease. She had unusual habits such as washing food into her stomach with large amounts of water and purging watery vomit every night before bed. Nonetheless, she did not consider it unusual because her gastroenterologist told her that nothing was wrong. The diagnostic features of esophageal achalasia on upper gastrointestinal endoscopy include (1) dilatation of the esophageal lumen, (2) abnormal retention of food and fluid in the esophagus, (3) whitening and thickening of the esophageal mucosal surface, (4) functional narrowing of the esophagogastric junction, and (5) abnormal contraction waves of the esophagus [2]. In this case, preoperative endoscopy before Heller and Dor surgery showed fluid retention in the esophagus and functional narrowing of the esophagogastric junction, but the dilation of the esophageal lumen was mild. As a result, the gastroenterologist at the time was unable to detect achalasia, leading to her misconception that it was not a disease. Second, we could not obtain information regarding dysphagia and food regurgitation in the preoperative examination. Our routine preoperative examination for orthopedic patients without gastrointestinal complications includes a detailed medical history, allergies, and asthma, but not dysphagia or food regurgitation. In patients with esophageal achalasia, 37% have solid residue, and 14.8% retain water, even after fasting for 24 h before surgery [5]. Fortunately, because this patient had a habit of vomiting every night before bed, and the vomit did not contain any solids, she did not develop severe pneumonia. Achalasia is a rare disease, and only a few cases have been detected by aspiration during the induction of anesthesia [1, 6, 7]. Since achalasia is a chronic benign disease, patients may recognize it as a constitutional problem, as in this case. A detailed history of dysphagia and regurgitation should be taken from all patients during preoperative examinations to avoid aspiration risk, even if achalasia is not suspected. Pillow stains with drool while sleeping suggests the existence of achalasia. Although chest X-rays appear normal in the early phase of achalasia, the dilated esophagus creates new interfaces with the lung as the disease progresses, which makes achalasia-specific findings, including convex opacity overlapping the right mediastinum, air-fluid levels in the thorax esophagus, and small or absent gastric bubbles [8]. In this case, the esophageal dilatation was so mild that no suspicious contour was seen on the chest X-ray (Fig. 3). However, it should be worthwhile for physicians to get into the habit of looking for achalasia-specific findings on routine X-ray readings to detect undiagnosed achalasia. Furthermore, we would like to emphasize that achalasia should not be ruled out even if there are no findings specific to achalasia, especially in patients with symptoms such as dysphagia or regurgitation. If a chest CT is taken for the preoperative evaluation, physicians should also note the dilated image of the esophagus.Fig. 3 Preoperative chest X-ray did not show any abnormalities Conclusions Although esophageal achalasia is a rare disease, it is associated with a high risk of aspiration pneumonia if general anesthesia is induced without caution. A detailed history of dysphagia and regurgitation of solids and liquids in the preoperative examination is essential and practical for detecting the existence of achalasia and preventing unexpected aspiration due to undiagnosed achalasia. Additionally, getting into the habit of checking for achalasia-specific findings on chest X-rays and chest CT may reduce the occurrence of unexpected aspiration. Acknowledgements Not applicable. Authors’ contributions KA wrote and prepared the final draft of the manuscript. TK and YN helped draft and review the manuscript. The authors read and approved the final manuscript. Funding Not applicable. Declarations Ethics approval and consent to participate Not applicable. Consent for publication Written informed consent was obtained from the patient for publication of this case report and the accompanying images. Competing interests The authors declare that they have no competing interests. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

AMPICILLIN SODIUM\SULBACTAM SODIUM, FENTANYL, PROPOFOL, RANITIDINE HYDROCHLORIDE, ROCURONIUM BROMIDE, SEVOFLURANE, SUGAMMADEX SODIUM

DrugsGivenReaction

CC BY

34888750

2021-12-10

Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Bronchospasm'.

Esophageal achalasia detected by vomiting during induction of general anesthesia: a case report. BACKGROUND Esophageal achalasia is a rare disease with a high risk of aspiration during anesthesia induction. Here, we describe our experience involving a case of undiagnosed esophageal achalasia with profuse vomiting during anesthesia induction. METHODS A 58-year-old woman was scheduled for orthopedic surgery under general anesthesia. She vomited a large amount of watery contents during anesthesia induction, and planned surgery was postponed. After recovery from anesthesia, she informed us that she usually had to drink a large amount of water to get food into her stomach and purged watery vomit every night before sleep. However, she attributed it to her constitutional problem, not to a specific disease. She was subsequently diagnosed with esophageal achalasia and underwent Heller myotomy with Dor fundoplication before her re-scheduled orthopedic surgery. CONCLUSIONS A detailed history of dysphagia and regurgitation should be taken in preoperative examinations to prevent unexpected aspiration due to undiagnosed achalasia. pmcBackground Esophageal achalasia is a rare disease (incidence, 1 in 100,000) associated with a high risk of aspiration during the induction of general anesthesia due to difficulties in passing food [1]. Here, we present a patient with undiagnosed esophageal achalasia who vomited profusely during the induction of general anesthesia. Case presentation A 58-year-old woman (height 163 cm, weight 52 kg, body mass index 19.6) with a diagnosis of spinal canal stenosis was admitted to an orthopedic hospital to undergo posterior lumbar interbody fusion. Her medical history included anterior cervical fusion 12 years previously without any anesthetic complications such as allergy or bronchial asthma. No full-time anesthesiologist was on staff at the hospital, so a part-time anesthesiologist was in charge of the anesthesia. The patient had been fasting for 15 h, from dinner the night before surgery until entering the operating room (OR). Four hours before entering the OR, an intravenous infusion was started. The patient was administered ranitidine (150 mg orally) with a small amount of water 180 min before entering the OR. Immediately before starting the anesthesia induction, the anesthesiologist reconfirmed the patient’s medical history and whether she had any allergies or bronchial asthma. The patient did not provide any new medical information. Electrocardiography, pulse oximetry (SpO2), noninvasive blood pressure, and capnometry were applied. After preoxygenation with 100% oxygen for 3 min, general anesthesia was induced with fentanyl 50 μg and propofol 100 mg. After confirming the patient’s spontaneous breathing cessation, assisted ventilation was initiated with a face mask, followed by rocuronium 50 mg. Immediately after that, the patient threw up a massive amount of watery vomit. The vomit in her oral cavity was immediately suctioned, and then her trachea was intubated. A small amount of watery vomit was suctioned through the endotracheal tube. Lung compliance was low with manual ventilation, and wheezing was heard on auscultation with an obstructive pattern on capnography. Bronchospasm due to aspiration was suspected, and treatment was initiated. Under anesthesia (sevoflurane 2% in oxygen 100%), her SpO2 remained above 98%. Hydrocortisone 100 mg and aminophylline 250 mg were administered by drip infusion. Forty-five minutes after the event, her wheezing disappeared, and mostly normal lung compliance returned. The scheduled surgery was postponed, the patient was awakened from anesthesia with the aid of sugammadex to antagonize rocuronium, and her trachea was extubated. After returning to the ward, a more detailed medical history was obtained from the patient. She said that she usually had to drink a large amount of water to get food into her stomach. Therefore, she made it a daily habit to purge watery vomit before bed, as it would reflux when she lay down. Although she had an upper gastrointestinal endoscopy a few years prior, the physician had not pointed out anything unusual. Since then, she believed that this was a constitutional problem as opposed to a disease. The patient had not experienced any weight loss for the past few years. Although no clear image of pneumonia was seen on a chest X-ray at the ward, ampicillin and sulbactam were administered for prevention. The patient was discharged the next day without any medical problems. Based on an overall assessment of the episode, we believed that this patient had esophageal achalasia and advised her to visit a gastroenterologist. At the gastroenterology department in another hospital, fluid retention in the dilated esophagus was revealed by upper endoscopy and chest computed tomography (CT) (Fig. 1). The esophagogram showed a dilated esophagus with a diameter of 32 mm and functional narrowing of the esophagogastric junction (Fig. 2). Esophageal manometry showed the disappearance of primary peristaltic waves and the occurrence of simultaneous contraction waves. Based on these results, the patient was diagnosed with esophageal achalasia and underwent laparoscopic Heller myotomy with Dor fundoplication surgery. After the surgery, her subjective symptoms, including choking with food and daily vomiting before bed, disappeared. She was discharged on the sixth day after surgery. Three months later, she underwent posterior interbody fusion of the lumbar spine without any anesthetic complications.Fig. 1 Preoperative thoracic computed tomography showing fluid retention in the dilated esophagus. a Upper thoracic region. b Middle thoracic region Fig. 2 Preoperative esophagogram showed a dilated esophagus and functional narrowing of the esophagogastric junction. A mouthful of contrast medium was retained in the esophagus until the end of the imaging Discussion Esophageal achalasia is defined as an esophageal motility disorder of unknown etiology characterized by failure of lower esophageal sphincter (LES) relaxation and impaired peristalsis of the lower esophageal body [2]. Diagnosis is difficult because achalasia is a rare disease (incidence, 1 in 100,000 people) with nonspecific subjective symptoms [3]. The most frequent symptoms of achalasia include dysphagia of solids and liquids (> 90%), regurgitation of undigested food (76–91%), respiratory complications such as nocturnal cough (30%) and aspiration (8%), chest pains (25–64%), heartburn (18–52%), and weight loss (35–91%) [4]. Oral reflux in patients with achalasia does not originate from the stomach and thus does not contain acidic contents. Furthermore, if the volume of the oral reflux is mild, the patient may be unaware of the reflux. We assume two reasons for the aspiration during the first scheduled orthopedic surgery in this case. First, the patient was not aware of her rare disease. She had unusual habits such as washing food into her stomach with large amounts of water and purging watery vomit every night before bed. Nonetheless, she did not consider it unusual because her gastroenterologist told her that nothing was wrong. The diagnostic features of esophageal achalasia on upper gastrointestinal endoscopy include (1) dilatation of the esophageal lumen, (2) abnormal retention of food and fluid in the esophagus, (3) whitening and thickening of the esophageal mucosal surface, (4) functional narrowing of the esophagogastric junction, and (5) abnormal contraction waves of the esophagus [2]. In this case, preoperative endoscopy before Heller and Dor surgery showed fluid retention in the esophagus and functional narrowing of the esophagogastric junction, but the dilation of the esophageal lumen was mild. As a result, the gastroenterologist at the time was unable to detect achalasia, leading to her misconception that it was not a disease. Second, we could not obtain information regarding dysphagia and food regurgitation in the preoperative examination. Our routine preoperative examination for orthopedic patients without gastrointestinal complications includes a detailed medical history, allergies, and asthma, but not dysphagia or food regurgitation. In patients with esophageal achalasia, 37% have solid residue, and 14.8% retain water, even after fasting for 24 h before surgery [5]. Fortunately, because this patient had a habit of vomiting every night before bed, and the vomit did not contain any solids, she did not develop severe pneumonia. Achalasia is a rare disease, and only a few cases have been detected by aspiration during the induction of anesthesia [1, 6, 7]. Since achalasia is a chronic benign disease, patients may recognize it as a constitutional problem, as in this case. A detailed history of dysphagia and regurgitation should be taken from all patients during preoperative examinations to avoid aspiration risk, even if achalasia is not suspected. Pillow stains with drool while sleeping suggests the existence of achalasia. Although chest X-rays appear normal in the early phase of achalasia, the dilated esophagus creates new interfaces with the lung as the disease progresses, which makes achalasia-specific findings, including convex opacity overlapping the right mediastinum, air-fluid levels in the thorax esophagus, and small or absent gastric bubbles [8]. In this case, the esophageal dilatation was so mild that no suspicious contour was seen on the chest X-ray (Fig. 3). However, it should be worthwhile for physicians to get into the habit of looking for achalasia-specific findings on routine X-ray readings to detect undiagnosed achalasia. Furthermore, we would like to emphasize that achalasia should not be ruled out even if there are no findings specific to achalasia, especially in patients with symptoms such as dysphagia or regurgitation. If a chest CT is taken for the preoperative evaluation, physicians should also note the dilated image of the esophagus.Fig. 3 Preoperative chest X-ray did not show any abnormalities Conclusions Although esophageal achalasia is a rare disease, it is associated with a high risk of aspiration pneumonia if general anesthesia is induced without caution. A detailed history of dysphagia and regurgitation of solids and liquids in the preoperative examination is essential and practical for detecting the existence of achalasia and preventing unexpected aspiration due to undiagnosed achalasia. Additionally, getting into the habit of checking for achalasia-specific findings on chest X-rays and chest CT may reduce the occurrence of unexpected aspiration. Acknowledgements Not applicable. Authors’ contributions KA wrote and prepared the final draft of the manuscript. TK and YN helped draft and review the manuscript. The authors read and approved the final manuscript. Funding Not applicable. Declarations Ethics approval and consent to participate Not applicable. Consent for publication Written informed consent was obtained from the patient for publication of this case report and the accompanying images. Competing interests The authors declare that they have no competing interests. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

AMPICILLIN SODIUM\SULBACTAM SODIUM, FENTANYL, PROPOFOL, RANITIDINE HYDROCHLORIDE, ROCURONIUM BROMIDE, SEVOFLURANE, SUGAMMADEX SODIUM

DrugsGivenReaction

CC BY

34888750

2021-12-10

Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Product use issue'.

Esophageal achalasia detected by vomiting during induction of general anesthesia: a case report. BACKGROUND Esophageal achalasia is a rare disease with a high risk of aspiration during anesthesia induction. Here, we describe our experience involving a case of undiagnosed esophageal achalasia with profuse vomiting during anesthesia induction. METHODS A 58-year-old woman was scheduled for orthopedic surgery under general anesthesia. She vomited a large amount of watery contents during anesthesia induction, and planned surgery was postponed. After recovery from anesthesia, she informed us that she usually had to drink a large amount of water to get food into her stomach and purged watery vomit every night before sleep. However, she attributed it to her constitutional problem, not to a specific disease. She was subsequently diagnosed with esophageal achalasia and underwent Heller myotomy with Dor fundoplication before her re-scheduled orthopedic surgery. CONCLUSIONS A detailed history of dysphagia and regurgitation should be taken in preoperative examinations to prevent unexpected aspiration due to undiagnosed achalasia. pmcBackground Esophageal achalasia is a rare disease (incidence, 1 in 100,000) associated with a high risk of aspiration during the induction of general anesthesia due to difficulties in passing food [1]. Here, we present a patient with undiagnosed esophageal achalasia who vomited profusely during the induction of general anesthesia. Case presentation A 58-year-old woman (height 163 cm, weight 52 kg, body mass index 19.6) with a diagnosis of spinal canal stenosis was admitted to an orthopedic hospital to undergo posterior lumbar interbody fusion. Her medical history included anterior cervical fusion 12 years previously without any anesthetic complications such as allergy or bronchial asthma. No full-time anesthesiologist was on staff at the hospital, so a part-time anesthesiologist was in charge of the anesthesia. The patient had been fasting for 15 h, from dinner the night before surgery until entering the operating room (OR). Four hours before entering the OR, an intravenous infusion was started. The patient was administered ranitidine (150 mg orally) with a small amount of water 180 min before entering the OR. Immediately before starting the anesthesia induction, the anesthesiologist reconfirmed the patient’s medical history and whether she had any allergies or bronchial asthma. The patient did not provide any new medical information. Electrocardiography, pulse oximetry (SpO2), noninvasive blood pressure, and capnometry were applied. After preoxygenation with 100% oxygen for 3 min, general anesthesia was induced with fentanyl 50 μg and propofol 100 mg. After confirming the patient’s spontaneous breathing cessation, assisted ventilation was initiated with a face mask, followed by rocuronium 50 mg. Immediately after that, the patient threw up a massive amount of watery vomit. The vomit in her oral cavity was immediately suctioned, and then her trachea was intubated. A small amount of watery vomit was suctioned through the endotracheal tube. Lung compliance was low with manual ventilation, and wheezing was heard on auscultation with an obstructive pattern on capnography. Bronchospasm due to aspiration was suspected, and treatment was initiated. Under anesthesia (sevoflurane 2% in oxygen 100%), her SpO2 remained above 98%. Hydrocortisone 100 mg and aminophylline 250 mg were administered by drip infusion. Forty-five minutes after the event, her wheezing disappeared, and mostly normal lung compliance returned. The scheduled surgery was postponed, the patient was awakened from anesthesia with the aid of sugammadex to antagonize rocuronium, and her trachea was extubated. After returning to the ward, a more detailed medical history was obtained from the patient. She said that she usually had to drink a large amount of water to get food into her stomach. Therefore, she made it a daily habit to purge watery vomit before bed, as it would reflux when she lay down. Although she had an upper gastrointestinal endoscopy a few years prior, the physician had not pointed out anything unusual. Since then, she believed that this was a constitutional problem as opposed to a disease. The patient had not experienced any weight loss for the past few years. Although no clear image of pneumonia was seen on a chest X-ray at the ward, ampicillin and sulbactam were administered for prevention. The patient was discharged the next day without any medical problems. Based on an overall assessment of the episode, we believed that this patient had esophageal achalasia and advised her to visit a gastroenterologist. At the gastroenterology department in another hospital, fluid retention in the dilated esophagus was revealed by upper endoscopy and chest computed tomography (CT) (Fig. 1). The esophagogram showed a dilated esophagus with a diameter of 32 mm and functional narrowing of the esophagogastric junction (Fig. 2). Esophageal manometry showed the disappearance of primary peristaltic waves and the occurrence of simultaneous contraction waves. Based on these results, the patient was diagnosed with esophageal achalasia and underwent laparoscopic Heller myotomy with Dor fundoplication surgery. After the surgery, her subjective symptoms, including choking with food and daily vomiting before bed, disappeared. She was discharged on the sixth day after surgery. Three months later, she underwent posterior interbody fusion of the lumbar spine without any anesthetic complications.Fig. 1 Preoperative thoracic computed tomography showing fluid retention in the dilated esophagus. a Upper thoracic region. b Middle thoracic region Fig. 2 Preoperative esophagogram showed a dilated esophagus and functional narrowing of the esophagogastric junction. A mouthful of contrast medium was retained in the esophagus until the end of the imaging Discussion Esophageal achalasia is defined as an esophageal motility disorder of unknown etiology characterized by failure of lower esophageal sphincter (LES) relaxation and impaired peristalsis of the lower esophageal body [2]. Diagnosis is difficult because achalasia is a rare disease (incidence, 1 in 100,000 people) with nonspecific subjective symptoms [3]. The most frequent symptoms of achalasia include dysphagia of solids and liquids (> 90%), regurgitation of undigested food (76–91%), respiratory complications such as nocturnal cough (30%) and aspiration (8%), chest pains (25–64%), heartburn (18–52%), and weight loss (35–91%) [4]. Oral reflux in patients with achalasia does not originate from the stomach and thus does not contain acidic contents. Furthermore, if the volume of the oral reflux is mild, the patient may be unaware of the reflux. We assume two reasons for the aspiration during the first scheduled orthopedic surgery in this case. First, the patient was not aware of her rare disease. She had unusual habits such as washing food into her stomach with large amounts of water and purging watery vomit every night before bed. Nonetheless, she did not consider it unusual because her gastroenterologist told her that nothing was wrong. The diagnostic features of esophageal achalasia on upper gastrointestinal endoscopy include (1) dilatation of the esophageal lumen, (2) abnormal retention of food and fluid in the esophagus, (3) whitening and thickening of the esophageal mucosal surface, (4) functional narrowing of the esophagogastric junction, and (5) abnormal contraction waves of the esophagus [2]. In this case, preoperative endoscopy before Heller and Dor surgery showed fluid retention in the esophagus and functional narrowing of the esophagogastric junction, but the dilation of the esophageal lumen was mild. As a result, the gastroenterologist at the time was unable to detect achalasia, leading to her misconception that it was not a disease. Second, we could not obtain information regarding dysphagia and food regurgitation in the preoperative examination. Our routine preoperative examination for orthopedic patients without gastrointestinal complications includes a detailed medical history, allergies, and asthma, but not dysphagia or food regurgitation. In patients with esophageal achalasia, 37% have solid residue, and 14.8% retain water, even after fasting for 24 h before surgery [5]. Fortunately, because this patient had a habit of vomiting every night before bed, and the vomit did not contain any solids, she did not develop severe pneumonia. Achalasia is a rare disease, and only a few cases have been detected by aspiration during the induction of anesthesia [1, 6, 7]. Since achalasia is a chronic benign disease, patients may recognize it as a constitutional problem, as in this case. A detailed history of dysphagia and regurgitation should be taken from all patients during preoperative examinations to avoid aspiration risk, even if achalasia is not suspected. Pillow stains with drool while sleeping suggests the existence of achalasia. Although chest X-rays appear normal in the early phase of achalasia, the dilated esophagus creates new interfaces with the lung as the disease progresses, which makes achalasia-specific findings, including convex opacity overlapping the right mediastinum, air-fluid levels in the thorax esophagus, and small or absent gastric bubbles [8]. In this case, the esophageal dilatation was so mild that no suspicious contour was seen on the chest X-ray (Fig. 3). However, it should be worthwhile for physicians to get into the habit of looking for achalasia-specific findings on routine X-ray readings to detect undiagnosed achalasia. Furthermore, we would like to emphasize that achalasia should not be ruled out even if there are no findings specific to achalasia, especially in patients with symptoms such as dysphagia or regurgitation. If a chest CT is taken for the preoperative evaluation, physicians should also note the dilated image of the esophagus.Fig. 3 Preoperative chest X-ray did not show any abnormalities Conclusions Although esophageal achalasia is a rare disease, it is associated with a high risk of aspiration pneumonia if general anesthesia is induced without caution. A detailed history of dysphagia and regurgitation of solids and liquids in the preoperative examination is essential and practical for detecting the existence of achalasia and preventing unexpected aspiration due to undiagnosed achalasia. Additionally, getting into the habit of checking for achalasia-specific findings on chest X-rays and chest CT may reduce the occurrence of unexpected aspiration. Acknowledgements Not applicable. Authors’ contributions KA wrote and prepared the final draft of the manuscript. TK and YN helped draft and review the manuscript. The authors read and approved the final manuscript. Funding Not applicable. Declarations Ethics approval and consent to participate Not applicable. Consent for publication Written informed consent was obtained from the patient for publication of this case report and the accompanying images. Competing interests The authors declare that they have no competing interests. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

AMPICILLIN SODIUM\SULBACTAM SODIUM, FENTANYL, PROPOFOL, RANITIDINE HYDROCHLORIDE, ROCURONIUM BROMIDE, SEVOFLURANE, SUGAMMADEX SODIUM

DrugsGivenReaction

CC BY

34888750

2021-12-10

Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Vomiting'.

Esophageal achalasia detected by vomiting during induction of general anesthesia: a case report. BACKGROUND Esophageal achalasia is a rare disease with a high risk of aspiration during anesthesia induction. Here, we describe our experience involving a case of undiagnosed esophageal achalasia with profuse vomiting during anesthesia induction. METHODS A 58-year-old woman was scheduled for orthopedic surgery under general anesthesia. She vomited a large amount of watery contents during anesthesia induction, and planned surgery was postponed. After recovery from anesthesia, she informed us that she usually had to drink a large amount of water to get food into her stomach and purged watery vomit every night before sleep. However, she attributed it to her constitutional problem, not to a specific disease. She was subsequently diagnosed with esophageal achalasia and underwent Heller myotomy with Dor fundoplication before her re-scheduled orthopedic surgery. CONCLUSIONS A detailed history of dysphagia and regurgitation should be taken in preoperative examinations to prevent unexpected aspiration due to undiagnosed achalasia. pmcBackground Esophageal achalasia is a rare disease (incidence, 1 in 100,000) associated with a high risk of aspiration during the induction of general anesthesia due to difficulties in passing food [1]. Here, we present a patient with undiagnosed esophageal achalasia who vomited profusely during the induction of general anesthesia. Case presentation A 58-year-old woman (height 163 cm, weight 52 kg, body mass index 19.6) with a diagnosis of spinal canal stenosis was admitted to an orthopedic hospital to undergo posterior lumbar interbody fusion. Her medical history included anterior cervical fusion 12 years previously without any anesthetic complications such as allergy or bronchial asthma. No full-time anesthesiologist was on staff at the hospital, so a part-time anesthesiologist was in charge of the anesthesia. The patient had been fasting for 15 h, from dinner the night before surgery until entering the operating room (OR). Four hours before entering the OR, an intravenous infusion was started. The patient was administered ranitidine (150 mg orally) with a small amount of water 180 min before entering the OR. Immediately before starting the anesthesia induction, the anesthesiologist reconfirmed the patient’s medical history and whether she had any allergies or bronchial asthma. The patient did not provide any new medical information. Electrocardiography, pulse oximetry (SpO2), noninvasive blood pressure, and capnometry were applied. After preoxygenation with 100% oxygen for 3 min, general anesthesia was induced with fentanyl 50 μg and propofol 100 mg. After confirming the patient’s spontaneous breathing cessation, assisted ventilation was initiated with a face mask, followed by rocuronium 50 mg. Immediately after that, the patient threw up a massive amount of watery vomit. The vomit in her oral cavity was immediately suctioned, and then her trachea was intubated. A small amount of watery vomit was suctioned through the endotracheal tube. Lung compliance was low with manual ventilation, and wheezing was heard on auscultation with an obstructive pattern on capnography. Bronchospasm due to aspiration was suspected, and treatment was initiated. Under anesthesia (sevoflurane 2% in oxygen 100%), her SpO2 remained above 98%. Hydrocortisone 100 mg and aminophylline 250 mg were administered by drip infusion. Forty-five minutes after the event, her wheezing disappeared, and mostly normal lung compliance returned. The scheduled surgery was postponed, the patient was awakened from anesthesia with the aid of sugammadex to antagonize rocuronium, and her trachea was extubated. After returning to the ward, a more detailed medical history was obtained from the patient. She said that she usually had to drink a large amount of water to get food into her stomach. Therefore, she made it a daily habit to purge watery vomit before bed, as it would reflux when she lay down. Although she had an upper gastrointestinal endoscopy a few years prior, the physician had not pointed out anything unusual. Since then, she believed that this was a constitutional problem as opposed to a disease. The patient had not experienced any weight loss for the past few years. Although no clear image of pneumonia was seen on a chest X-ray at the ward, ampicillin and sulbactam were administered for prevention. The patient was discharged the next day without any medical problems. Based on an overall assessment of the episode, we believed that this patient had esophageal achalasia and advised her to visit a gastroenterologist. At the gastroenterology department in another hospital, fluid retention in the dilated esophagus was revealed by upper endoscopy and chest computed tomography (CT) (Fig. 1). The esophagogram showed a dilated esophagus with a diameter of 32 mm and functional narrowing of the esophagogastric junction (Fig. 2). Esophageal manometry showed the disappearance of primary peristaltic waves and the occurrence of simultaneous contraction waves. Based on these results, the patient was diagnosed with esophageal achalasia and underwent laparoscopic Heller myotomy with Dor fundoplication surgery. After the surgery, her subjective symptoms, including choking with food and daily vomiting before bed, disappeared. She was discharged on the sixth day after surgery. Three months later, she underwent posterior interbody fusion of the lumbar spine without any anesthetic complications.Fig. 1 Preoperative thoracic computed tomography showing fluid retention in the dilated esophagus. a Upper thoracic region. b Middle thoracic region Fig. 2 Preoperative esophagogram showed a dilated esophagus and functional narrowing of the esophagogastric junction. A mouthful of contrast medium was retained in the esophagus until the end of the imaging Discussion Esophageal achalasia is defined as an esophageal motility disorder of unknown etiology characterized by failure of lower esophageal sphincter (LES) relaxation and impaired peristalsis of the lower esophageal body [2]. Diagnosis is difficult because achalasia is a rare disease (incidence, 1 in 100,000 people) with nonspecific subjective symptoms [3]. The most frequent symptoms of achalasia include dysphagia of solids and liquids (> 90%), regurgitation of undigested food (76–91%), respiratory complications such as nocturnal cough (30%) and aspiration (8%), chest pains (25–64%), heartburn (18–52%), and weight loss (35–91%) [4]. Oral reflux in patients with achalasia does not originate from the stomach and thus does not contain acidic contents. Furthermore, if the volume of the oral reflux is mild, the patient may be unaware of the reflux. We assume two reasons for the aspiration during the first scheduled orthopedic surgery in this case. First, the patient was not aware of her rare disease. She had unusual habits such as washing food into her stomach with large amounts of water and purging watery vomit every night before bed. Nonetheless, she did not consider it unusual because her gastroenterologist told her that nothing was wrong. The diagnostic features of esophageal achalasia on upper gastrointestinal endoscopy include (1) dilatation of the esophageal lumen, (2) abnormal retention of food and fluid in the esophagus, (3) whitening and thickening of the esophageal mucosal surface, (4) functional narrowing of the esophagogastric junction, and (5) abnormal contraction waves of the esophagus [2]. In this case, preoperative endoscopy before Heller and Dor surgery showed fluid retention in the esophagus and functional narrowing of the esophagogastric junction, but the dilation of the esophageal lumen was mild. As a result, the gastroenterologist at the time was unable to detect achalasia, leading to her misconception that it was not a disease. Second, we could not obtain information regarding dysphagia and food regurgitation in the preoperative examination. Our routine preoperative examination for orthopedic patients without gastrointestinal complications includes a detailed medical history, allergies, and asthma, but not dysphagia or food regurgitation. In patients with esophageal achalasia, 37% have solid residue, and 14.8% retain water, even after fasting for 24 h before surgery [5]. Fortunately, because this patient had a habit of vomiting every night before bed, and the vomit did not contain any solids, she did not develop severe pneumonia. Achalasia is a rare disease, and only a few cases have been detected by aspiration during the induction of anesthesia [1, 6, 7]. Since achalasia is a chronic benign disease, patients may recognize it as a constitutional problem, as in this case. A detailed history of dysphagia and regurgitation should be taken from all patients during preoperative examinations to avoid aspiration risk, even if achalasia is not suspected. Pillow stains with drool while sleeping suggests the existence of achalasia. Although chest X-rays appear normal in the early phase of achalasia, the dilated esophagus creates new interfaces with the lung as the disease progresses, which makes achalasia-specific findings, including convex opacity overlapping the right mediastinum, air-fluid levels in the thorax esophagus, and small or absent gastric bubbles [8]. In this case, the esophageal dilatation was so mild that no suspicious contour was seen on the chest X-ray (Fig. 3). However, it should be worthwhile for physicians to get into the habit of looking for achalasia-specific findings on routine X-ray readings to detect undiagnosed achalasia. Furthermore, we would like to emphasize that achalasia should not be ruled out even if there are no findings specific to achalasia, especially in patients with symptoms such as dysphagia or regurgitation. If a chest CT is taken for the preoperative evaluation, physicians should also note the dilated image of the esophagus.Fig. 3 Preoperative chest X-ray did not show any abnormalities Conclusions Although esophageal achalasia is a rare disease, it is associated with a high risk of aspiration pneumonia if general anesthesia is induced without caution. A detailed history of dysphagia and regurgitation of solids and liquids in the preoperative examination is essential and practical for detecting the existence of achalasia and preventing unexpected aspiration due to undiagnosed achalasia. Additionally, getting into the habit of checking for achalasia-specific findings on chest X-rays and chest CT may reduce the occurrence of unexpected aspiration. Acknowledgements Not applicable. Authors’ contributions KA wrote and prepared the final draft of the manuscript. TK and YN helped draft and review the manuscript. The authors read and approved the final manuscript. Funding Not applicable. Declarations Ethics approval and consent to participate Not applicable. Consent for publication Written informed consent was obtained from the patient for publication of this case report and the accompanying images. Competing interests The authors declare that they have no competing interests. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

AMPICILLIN SODIUM\SULBACTAM SODIUM, FENTANYL, PROPOFOL, RANITIDINE HYDROCHLORIDE, ROCURONIUM BROMIDE, SEVOFLURANE, SUGAMMADEX SODIUM

DrugsGivenReaction

CC BY

34888750

2021-12-10

Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Wheezing'.

Esophageal achalasia detected by vomiting during induction of general anesthesia: a case report. BACKGROUND Esophageal achalasia is a rare disease with a high risk of aspiration during anesthesia induction. Here, we describe our experience involving a case of undiagnosed esophageal achalasia with profuse vomiting during anesthesia induction. METHODS A 58-year-old woman was scheduled for orthopedic surgery under general anesthesia. She vomited a large amount of watery contents during anesthesia induction, and planned surgery was postponed. After recovery from anesthesia, she informed us that she usually had to drink a large amount of water to get food into her stomach and purged watery vomit every night before sleep. However, she attributed it to her constitutional problem, not to a specific disease. She was subsequently diagnosed with esophageal achalasia and underwent Heller myotomy with Dor fundoplication before her re-scheduled orthopedic surgery. CONCLUSIONS A detailed history of dysphagia and regurgitation should be taken in preoperative examinations to prevent unexpected aspiration due to undiagnosed achalasia. pmcBackground Esophageal achalasia is a rare disease (incidence, 1 in 100,000) associated with a high risk of aspiration during the induction of general anesthesia due to difficulties in passing food [1]. Here, we present a patient with undiagnosed esophageal achalasia who vomited profusely during the induction of general anesthesia. Case presentation A 58-year-old woman (height 163 cm, weight 52 kg, body mass index 19.6) with a diagnosis of spinal canal stenosis was admitted to an orthopedic hospital to undergo posterior lumbar interbody fusion. Her medical history included anterior cervical fusion 12 years previously without any anesthetic complications such as allergy or bronchial asthma. No full-time anesthesiologist was on staff at the hospital, so a part-time anesthesiologist was in charge of the anesthesia. The patient had been fasting for 15 h, from dinner the night before surgery until entering the operating room (OR). Four hours before entering the OR, an intravenous infusion was started. The patient was administered ranitidine (150 mg orally) with a small amount of water 180 min before entering the OR. Immediately before starting the anesthesia induction, the anesthesiologist reconfirmed the patient’s medical history and whether she had any allergies or bronchial asthma. The patient did not provide any new medical information. Electrocardiography, pulse oximetry (SpO2), noninvasive blood pressure, and capnometry were applied. After preoxygenation with 100% oxygen for 3 min, general anesthesia was induced with fentanyl 50 μg and propofol 100 mg. After confirming the patient’s spontaneous breathing cessation, assisted ventilation was initiated with a face mask, followed by rocuronium 50 mg. Immediately after that, the patient threw up a massive amount of watery vomit. The vomit in her oral cavity was immediately suctioned, and then her trachea was intubated. A small amount of watery vomit was suctioned through the endotracheal tube. Lung compliance was low with manual ventilation, and wheezing was heard on auscultation with an obstructive pattern on capnography. Bronchospasm due to aspiration was suspected, and treatment was initiated. Under anesthesia (sevoflurane 2% in oxygen 100%), her SpO2 remained above 98%. Hydrocortisone 100 mg and aminophylline 250 mg were administered by drip infusion. Forty-five minutes after the event, her wheezing disappeared, and mostly normal lung compliance returned. The scheduled surgery was postponed, the patient was awakened from anesthesia with the aid of sugammadex to antagonize rocuronium, and her trachea was extubated. After returning to the ward, a more detailed medical history was obtained from the patient. She said that she usually had to drink a large amount of water to get food into her stomach. Therefore, she made it a daily habit to purge watery vomit before bed, as it would reflux when she lay down. Although she had an upper gastrointestinal endoscopy a few years prior, the physician had not pointed out anything unusual. Since then, she believed that this was a constitutional problem as opposed to a disease. The patient had not experienced any weight loss for the past few years. Although no clear image of pneumonia was seen on a chest X-ray at the ward, ampicillin and sulbactam were administered for prevention. The patient was discharged the next day without any medical problems. Based on an overall assessment of the episode, we believed that this patient had esophageal achalasia and advised her to visit a gastroenterologist. At the gastroenterology department in another hospital, fluid retention in the dilated esophagus was revealed by upper endoscopy and chest computed tomography (CT) (Fig. 1). The esophagogram showed a dilated esophagus with a diameter of 32 mm and functional narrowing of the esophagogastric junction (Fig. 2). Esophageal manometry showed the disappearance of primary peristaltic waves and the occurrence of simultaneous contraction waves. Based on these results, the patient was diagnosed with esophageal achalasia and underwent laparoscopic Heller myotomy with Dor fundoplication surgery. After the surgery, her subjective symptoms, including choking with food and daily vomiting before bed, disappeared. She was discharged on the sixth day after surgery. Three months later, she underwent posterior interbody fusion of the lumbar spine without any anesthetic complications.Fig. 1 Preoperative thoracic computed tomography showing fluid retention in the dilated esophagus. a Upper thoracic region. b Middle thoracic region Fig. 2 Preoperative esophagogram showed a dilated esophagus and functional narrowing of the esophagogastric junction. A mouthful of contrast medium was retained in the esophagus until the end of the imaging Discussion Esophageal achalasia is defined as an esophageal motility disorder of unknown etiology characterized by failure of lower esophageal sphincter (LES) relaxation and impaired peristalsis of the lower esophageal body [2]. Diagnosis is difficult because achalasia is a rare disease (incidence, 1 in 100,000 people) with nonspecific subjective symptoms [3]. The most frequent symptoms of achalasia include dysphagia of solids and liquids (> 90%), regurgitation of undigested food (76–91%), respiratory complications such as nocturnal cough (30%) and aspiration (8%), chest pains (25–64%), heartburn (18–52%), and weight loss (35–91%) [4]. Oral reflux in patients with achalasia does not originate from the stomach and thus does not contain acidic contents. Furthermore, if the volume of the oral reflux is mild, the patient may be unaware of the reflux. We assume two reasons for the aspiration during the first scheduled orthopedic surgery in this case. First, the patient was not aware of her rare disease. She had unusual habits such as washing food into her stomach with large amounts of water and purging watery vomit every night before bed. Nonetheless, she did not consider it unusual because her gastroenterologist told her that nothing was wrong. The diagnostic features of esophageal achalasia on upper gastrointestinal endoscopy include (1) dilatation of the esophageal lumen, (2) abnormal retention of food and fluid in the esophagus, (3) whitening and thickening of the esophageal mucosal surface, (4) functional narrowing of the esophagogastric junction, and (5) abnormal contraction waves of the esophagus [2]. In this case, preoperative endoscopy before Heller and Dor surgery showed fluid retention in the esophagus and functional narrowing of the esophagogastric junction, but the dilation of the esophageal lumen was mild. As a result, the gastroenterologist at the time was unable to detect achalasia, leading to her misconception that it was not a disease. Second, we could not obtain information regarding dysphagia and food regurgitation in the preoperative examination. Our routine preoperative examination for orthopedic patients without gastrointestinal complications includes a detailed medical history, allergies, and asthma, but not dysphagia or food regurgitation. In patients with esophageal achalasia, 37% have solid residue, and 14.8% retain water, even after fasting for 24 h before surgery [5]. Fortunately, because this patient had a habit of vomiting every night before bed, and the vomit did not contain any solids, she did not develop severe pneumonia. Achalasia is a rare disease, and only a few cases have been detected by aspiration during the induction of anesthesia [1, 6, 7]. Since achalasia is a chronic benign disease, patients may recognize it as a constitutional problem, as in this case. A detailed history of dysphagia and regurgitation should be taken from all patients during preoperative examinations to avoid aspiration risk, even if achalasia is not suspected. Pillow stains with drool while sleeping suggests the existence of achalasia. Although chest X-rays appear normal in the early phase of achalasia, the dilated esophagus creates new interfaces with the lung as the disease progresses, which makes achalasia-specific findings, including convex opacity overlapping the right mediastinum, air-fluid levels in the thorax esophagus, and small or absent gastric bubbles [8]. In this case, the esophageal dilatation was so mild that no suspicious contour was seen on the chest X-ray (Fig. 3). However, it should be worthwhile for physicians to get into the habit of looking for achalasia-specific findings on routine X-ray readings to detect undiagnosed achalasia. Furthermore, we would like to emphasize that achalasia should not be ruled out even if there are no findings specific to achalasia, especially in patients with symptoms such as dysphagia or regurgitation. If a chest CT is taken for the preoperative evaluation, physicians should also note the dilated image of the esophagus.Fig. 3 Preoperative chest X-ray did not show any abnormalities Conclusions Although esophageal achalasia is a rare disease, it is associated with a high risk of aspiration pneumonia if general anesthesia is induced without caution. A detailed history of dysphagia and regurgitation of solids and liquids in the preoperative examination is essential and practical for detecting the existence of achalasia and preventing unexpected aspiration due to undiagnosed achalasia. Additionally, getting into the habit of checking for achalasia-specific findings on chest X-rays and chest CT may reduce the occurrence of unexpected aspiration. Acknowledgements Not applicable. Authors’ contributions KA wrote and prepared the final draft of the manuscript. TK and YN helped draft and review the manuscript. The authors read and approved the final manuscript. Funding Not applicable. Declarations Ethics approval and consent to participate Not applicable. Consent for publication Written informed consent was obtained from the patient for publication of this case report and the accompanying images. Competing interests The authors declare that they have no competing interests. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

AMPICILLIN SODIUM\SULBACTAM SODIUM, FENTANYL, PROPOFOL, RANITIDINE HYDROCHLORIDE, ROCURONIUM BROMIDE, SEVOFLURANE, SUGAMMADEX SODIUM

DrugsGivenReaction

CC BY

34888750

2021-12-10

What is the weight of the patient?

Esophageal achalasia detected by vomiting during induction of general anesthesia: a case report. BACKGROUND Esophageal achalasia is a rare disease with a high risk of aspiration during anesthesia induction. Here, we describe our experience involving a case of undiagnosed esophageal achalasia with profuse vomiting during anesthesia induction. METHODS A 58-year-old woman was scheduled for orthopedic surgery under general anesthesia. She vomited a large amount of watery contents during anesthesia induction, and planned surgery was postponed. After recovery from anesthesia, she informed us that she usually had to drink a large amount of water to get food into her stomach and purged watery vomit every night before sleep. However, she attributed it to her constitutional problem, not to a specific disease. She was subsequently diagnosed with esophageal achalasia and underwent Heller myotomy with Dor fundoplication before her re-scheduled orthopedic surgery. CONCLUSIONS A detailed history of dysphagia and regurgitation should be taken in preoperative examinations to prevent unexpected aspiration due to undiagnosed achalasia. pmcBackground Esophageal achalasia is a rare disease (incidence, 1 in 100,000) associated with a high risk of aspiration during the induction of general anesthesia due to difficulties in passing food [1]. Here, we present a patient with undiagnosed esophageal achalasia who vomited profusely during the induction of general anesthesia. Case presentation A 58-year-old woman (height 163 cm, weight 52 kg, body mass index 19.6) with a diagnosis of spinal canal stenosis was admitted to an orthopedic hospital to undergo posterior lumbar interbody fusion. Her medical history included anterior cervical fusion 12 years previously without any anesthetic complications such as allergy or bronchial asthma. No full-time anesthesiologist was on staff at the hospital, so a part-time anesthesiologist was in charge of the anesthesia. The patient had been fasting for 15 h, from dinner the night before surgery until entering the operating room (OR). Four hours before entering the OR, an intravenous infusion was started. The patient was administered ranitidine (150 mg orally) with a small amount of water 180 min before entering the OR. Immediately before starting the anesthesia induction, the anesthesiologist reconfirmed the patient’s medical history and whether she had any allergies or bronchial asthma. The patient did not provide any new medical information. Electrocardiography, pulse oximetry (SpO2), noninvasive blood pressure, and capnometry were applied. After preoxygenation with 100% oxygen for 3 min, general anesthesia was induced with fentanyl 50 μg and propofol 100 mg. After confirming the patient’s spontaneous breathing cessation, assisted ventilation was initiated with a face mask, followed by rocuronium 50 mg. Immediately after that, the patient threw up a massive amount of watery vomit. The vomit in her oral cavity was immediately suctioned, and then her trachea was intubated. A small amount of watery vomit was suctioned through the endotracheal tube. Lung compliance was low with manual ventilation, and wheezing was heard on auscultation with an obstructive pattern on capnography. Bronchospasm due to aspiration was suspected, and treatment was initiated. Under anesthesia (sevoflurane 2% in oxygen 100%), her SpO2 remained above 98%. Hydrocortisone 100 mg and aminophylline 250 mg were administered by drip infusion. Forty-five minutes after the event, her wheezing disappeared, and mostly normal lung compliance returned. The scheduled surgery was postponed, the patient was awakened from anesthesia with the aid of sugammadex to antagonize rocuronium, and her trachea was extubated. After returning to the ward, a more detailed medical history was obtained from the patient. She said that she usually had to drink a large amount of water to get food into her stomach. Therefore, she made it a daily habit to purge watery vomit before bed, as it would reflux when she lay down. Although she had an upper gastrointestinal endoscopy a few years prior, the physician had not pointed out anything unusual. Since then, she believed that this was a constitutional problem as opposed to a disease. The patient had not experienced any weight loss for the past few years. Although no clear image of pneumonia was seen on a chest X-ray at the ward, ampicillin and sulbactam were administered for prevention. The patient was discharged the next day without any medical problems. Based on an overall assessment of the episode, we believed that this patient had esophageal achalasia and advised her to visit a gastroenterologist. At the gastroenterology department in another hospital, fluid retention in the dilated esophagus was revealed by upper endoscopy and chest computed tomography (CT) (Fig. 1). The esophagogram showed a dilated esophagus with a diameter of 32 mm and functional narrowing of the esophagogastric junction (Fig. 2). Esophageal manometry showed the disappearance of primary peristaltic waves and the occurrence of simultaneous contraction waves. Based on these results, the patient was diagnosed with esophageal achalasia and underwent laparoscopic Heller myotomy with Dor fundoplication surgery. After the surgery, her subjective symptoms, including choking with food and daily vomiting before bed, disappeared. She was discharged on the sixth day after surgery. Three months later, she underwent posterior interbody fusion of the lumbar spine without any anesthetic complications.Fig. 1 Preoperative thoracic computed tomography showing fluid retention in the dilated esophagus. a Upper thoracic region. b Middle thoracic region Fig. 2 Preoperative esophagogram showed a dilated esophagus and functional narrowing of the esophagogastric junction. A mouthful of contrast medium was retained in the esophagus until the end of the imaging Discussion Esophageal achalasia is defined as an esophageal motility disorder of unknown etiology characterized by failure of lower esophageal sphincter (LES) relaxation and impaired peristalsis of the lower esophageal body [2]. Diagnosis is difficult because achalasia is a rare disease (incidence, 1 in 100,000 people) with nonspecific subjective symptoms [3]. The most frequent symptoms of achalasia include dysphagia of solids and liquids (> 90%), regurgitation of undigested food (76–91%), respiratory complications such as nocturnal cough (30%) and aspiration (8%), chest pains (25–64%), heartburn (18–52%), and weight loss (35–91%) [4]. Oral reflux in patients with achalasia does not originate from the stomach and thus does not contain acidic contents. Furthermore, if the volume of the oral reflux is mild, the patient may be unaware of the reflux. We assume two reasons for the aspiration during the first scheduled orthopedic surgery in this case. First, the patient was not aware of her rare disease. She had unusual habits such as washing food into her stomach with large amounts of water and purging watery vomit every night before bed. Nonetheless, she did not consider it unusual because her gastroenterologist told her that nothing was wrong. The diagnostic features of esophageal achalasia on upper gastrointestinal endoscopy include (1) dilatation of the esophageal lumen, (2) abnormal retention of food and fluid in the esophagus, (3) whitening and thickening of the esophageal mucosal surface, (4) functional narrowing of the esophagogastric junction, and (5) abnormal contraction waves of the esophagus [2]. In this case, preoperative endoscopy before Heller and Dor surgery showed fluid retention in the esophagus and functional narrowing of the esophagogastric junction, but the dilation of the esophageal lumen was mild. As a result, the gastroenterologist at the time was unable to detect achalasia, leading to her misconception that it was not a disease. Second, we could not obtain information regarding dysphagia and food regurgitation in the preoperative examination. Our routine preoperative examination for orthopedic patients without gastrointestinal complications includes a detailed medical history, allergies, and asthma, but not dysphagia or food regurgitation. In patients with esophageal achalasia, 37% have solid residue, and 14.8% retain water, even after fasting for 24 h before surgery [5]. Fortunately, because this patient had a habit of vomiting every night before bed, and the vomit did not contain any solids, she did not develop severe pneumonia. Achalasia is a rare disease, and only a few cases have been detected by aspiration during the induction of anesthesia [1, 6, 7]. Since achalasia is a chronic benign disease, patients may recognize it as a constitutional problem, as in this case. A detailed history of dysphagia and regurgitation should be taken from all patients during preoperative examinations to avoid aspiration risk, even if achalasia is not suspected. Pillow stains with drool while sleeping suggests the existence of achalasia. Although chest X-rays appear normal in the early phase of achalasia, the dilated esophagus creates new interfaces with the lung as the disease progresses, which makes achalasia-specific findings, including convex opacity overlapping the right mediastinum, air-fluid levels in the thorax esophagus, and small or absent gastric bubbles [8]. In this case, the esophageal dilatation was so mild that no suspicious contour was seen on the chest X-ray (Fig. 3). However, it should be worthwhile for physicians to get into the habit of looking for achalasia-specific findings on routine X-ray readings to detect undiagnosed achalasia. Furthermore, we would like to emphasize that achalasia should not be ruled out even if there are no findings specific to achalasia, especially in patients with symptoms such as dysphagia or regurgitation. If a chest CT is taken for the preoperative evaluation, physicians should also note the dilated image of the esophagus.Fig. 3 Preoperative chest X-ray did not show any abnormalities Conclusions Although esophageal achalasia is a rare disease, it is associated with a high risk of aspiration pneumonia if general anesthesia is induced without caution. A detailed history of dysphagia and regurgitation of solids and liquids in the preoperative examination is essential and practical for detecting the existence of achalasia and preventing unexpected aspiration due to undiagnosed achalasia. Additionally, getting into the habit of checking for achalasia-specific findings on chest X-rays and chest CT may reduce the occurrence of unexpected aspiration. Acknowledgements Not applicable. Authors’ contributions KA wrote and prepared the final draft of the manuscript. TK and YN helped draft and review the manuscript. The authors read and approved the final manuscript. Funding Not applicable. Declarations Ethics approval and consent to participate Not applicable. Consent for publication Written informed consent was obtained from the patient for publication of this case report and the accompanying images. Competing interests The authors declare that they have no competing interests. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

52 kg.

Weight

CC BY

34888750

2021-12-10

What was the administration route of drug 'RANITIDINE HYDROCHLORIDE'?

Esophageal achalasia detected by vomiting during induction of general anesthesia: a case report. BACKGROUND Esophageal achalasia is a rare disease with a high risk of aspiration during anesthesia induction. Here, we describe our experience involving a case of undiagnosed esophageal achalasia with profuse vomiting during anesthesia induction. METHODS A 58-year-old woman was scheduled for orthopedic surgery under general anesthesia. She vomited a large amount of watery contents during anesthesia induction, and planned surgery was postponed. After recovery from anesthesia, she informed us that she usually had to drink a large amount of water to get food into her stomach and purged watery vomit every night before sleep. However, she attributed it to her constitutional problem, not to a specific disease. She was subsequently diagnosed with esophageal achalasia and underwent Heller myotomy with Dor fundoplication before her re-scheduled orthopedic surgery. CONCLUSIONS A detailed history of dysphagia and regurgitation should be taken in preoperative examinations to prevent unexpected aspiration due to undiagnosed achalasia. pmcBackground Esophageal achalasia is a rare disease (incidence, 1 in 100,000) associated with a high risk of aspiration during the induction of general anesthesia due to difficulties in passing food [1]. Here, we present a patient with undiagnosed esophageal achalasia who vomited profusely during the induction of general anesthesia. Case presentation A 58-year-old woman (height 163 cm, weight 52 kg, body mass index 19.6) with a diagnosis of spinal canal stenosis was admitted to an orthopedic hospital to undergo posterior lumbar interbody fusion. Her medical history included anterior cervical fusion 12 years previously without any anesthetic complications such as allergy or bronchial asthma. No full-time anesthesiologist was on staff at the hospital, so a part-time anesthesiologist was in charge of the anesthesia. The patient had been fasting for 15 h, from dinner the night before surgery until entering the operating room (OR). Four hours before entering the OR, an intravenous infusion was started. The patient was administered ranitidine (150 mg orally) with a small amount of water 180 min before entering the OR. Immediately before starting the anesthesia induction, the anesthesiologist reconfirmed the patient’s medical history and whether she had any allergies or bronchial asthma. The patient did not provide any new medical information. Electrocardiography, pulse oximetry (SpO2), noninvasive blood pressure, and capnometry were applied. After preoxygenation with 100% oxygen for 3 min, general anesthesia was induced with fentanyl 50 μg and propofol 100 mg. After confirming the patient’s spontaneous breathing cessation, assisted ventilation was initiated with a face mask, followed by rocuronium 50 mg. Immediately after that, the patient threw up a massive amount of watery vomit. The vomit in her oral cavity was immediately suctioned, and then her trachea was intubated. A small amount of watery vomit was suctioned through the endotracheal tube. Lung compliance was low with manual ventilation, and wheezing was heard on auscultation with an obstructive pattern on capnography. Bronchospasm due to aspiration was suspected, and treatment was initiated. Under anesthesia (sevoflurane 2% in oxygen 100%), her SpO2 remained above 98%. Hydrocortisone 100 mg and aminophylline 250 mg were administered by drip infusion. Forty-five minutes after the event, her wheezing disappeared, and mostly normal lung compliance returned. The scheduled surgery was postponed, the patient was awakened from anesthesia with the aid of sugammadex to antagonize rocuronium, and her trachea was extubated. After returning to the ward, a more detailed medical history was obtained from the patient. She said that she usually had to drink a large amount of water to get food into her stomach. Therefore, she made it a daily habit to purge watery vomit before bed, as it would reflux when she lay down. Although she had an upper gastrointestinal endoscopy a few years prior, the physician had not pointed out anything unusual. Since then, she believed that this was a constitutional problem as opposed to a disease. The patient had not experienced any weight loss for the past few years. Although no clear image of pneumonia was seen on a chest X-ray at the ward, ampicillin and sulbactam were administered for prevention. The patient was discharged the next day without any medical problems. Based on an overall assessment of the episode, we believed that this patient had esophageal achalasia and advised her to visit a gastroenterologist. At the gastroenterology department in another hospital, fluid retention in the dilated esophagus was revealed by upper endoscopy and chest computed tomography (CT) (Fig. 1). The esophagogram showed a dilated esophagus with a diameter of 32 mm and functional narrowing of the esophagogastric junction (Fig. 2). Esophageal manometry showed the disappearance of primary peristaltic waves and the occurrence of simultaneous contraction waves. Based on these results, the patient was diagnosed with esophageal achalasia and underwent laparoscopic Heller myotomy with Dor fundoplication surgery. After the surgery, her subjective symptoms, including choking with food and daily vomiting before bed, disappeared. She was discharged on the sixth day after surgery. Three months later, she underwent posterior interbody fusion of the lumbar spine without any anesthetic complications.Fig. 1 Preoperative thoracic computed tomography showing fluid retention in the dilated esophagus. a Upper thoracic region. b Middle thoracic region Fig. 2 Preoperative esophagogram showed a dilated esophagus and functional narrowing of the esophagogastric junction. A mouthful of contrast medium was retained in the esophagus until the end of the imaging Discussion Esophageal achalasia is defined as an esophageal motility disorder of unknown etiology characterized by failure of lower esophageal sphincter (LES) relaxation and impaired peristalsis of the lower esophageal body [2]. Diagnosis is difficult because achalasia is a rare disease (incidence, 1 in 100,000 people) with nonspecific subjective symptoms [3]. The most frequent symptoms of achalasia include dysphagia of solids and liquids (> 90%), regurgitation of undigested food (76–91%), respiratory complications such as nocturnal cough (30%) and aspiration (8%), chest pains (25–64%), heartburn (18–52%), and weight loss (35–91%) [4]. Oral reflux in patients with achalasia does not originate from the stomach and thus does not contain acidic contents. Furthermore, if the volume of the oral reflux is mild, the patient may be unaware of the reflux. We assume two reasons for the aspiration during the first scheduled orthopedic surgery in this case. First, the patient was not aware of her rare disease. She had unusual habits such as washing food into her stomach with large amounts of water and purging watery vomit every night before bed. Nonetheless, she did not consider it unusual because her gastroenterologist told her that nothing was wrong. The diagnostic features of esophageal achalasia on upper gastrointestinal endoscopy include (1) dilatation of the esophageal lumen, (2) abnormal retention of food and fluid in the esophagus, (3) whitening and thickening of the esophageal mucosal surface, (4) functional narrowing of the esophagogastric junction, and (5) abnormal contraction waves of the esophagus [2]. In this case, preoperative endoscopy before Heller and Dor surgery showed fluid retention in the esophagus and functional narrowing of the esophagogastric junction, but the dilation of the esophageal lumen was mild. As a result, the gastroenterologist at the time was unable to detect achalasia, leading to her misconception that it was not a disease. Second, we could not obtain information regarding dysphagia and food regurgitation in the preoperative examination. Our routine preoperative examination for orthopedic patients without gastrointestinal complications includes a detailed medical history, allergies, and asthma, but not dysphagia or food regurgitation. In patients with esophageal achalasia, 37% have solid residue, and 14.8% retain water, even after fasting for 24 h before surgery [5]. Fortunately, because this patient had a habit of vomiting every night before bed, and the vomit did not contain any solids, she did not develop severe pneumonia. Achalasia is a rare disease, and only a few cases have been detected by aspiration during the induction of anesthesia [1, 6, 7]. Since achalasia is a chronic benign disease, patients may recognize it as a constitutional problem, as in this case. A detailed history of dysphagia and regurgitation should be taken from all patients during preoperative examinations to avoid aspiration risk, even if achalasia is not suspected. Pillow stains with drool while sleeping suggests the existence of achalasia. Although chest X-rays appear normal in the early phase of achalasia, the dilated esophagus creates new interfaces with the lung as the disease progresses, which makes achalasia-specific findings, including convex opacity overlapping the right mediastinum, air-fluid levels in the thorax esophagus, and small or absent gastric bubbles [8]. In this case, the esophageal dilatation was so mild that no suspicious contour was seen on the chest X-ray (Fig. 3). However, it should be worthwhile for physicians to get into the habit of looking for achalasia-specific findings on routine X-ray readings to detect undiagnosed achalasia. Furthermore, we would like to emphasize that achalasia should not be ruled out even if there are no findings specific to achalasia, especially in patients with symptoms such as dysphagia or regurgitation. If a chest CT is taken for the preoperative evaluation, physicians should also note the dilated image of the esophagus.Fig. 3 Preoperative chest X-ray did not show any abnormalities Conclusions Although esophageal achalasia is a rare disease, it is associated with a high risk of aspiration pneumonia if general anesthesia is induced without caution. A detailed history of dysphagia and regurgitation of solids and liquids in the preoperative examination is essential and practical for detecting the existence of achalasia and preventing unexpected aspiration due to undiagnosed achalasia. Additionally, getting into the habit of checking for achalasia-specific findings on chest X-rays and chest CT may reduce the occurrence of unexpected aspiration. Acknowledgements Not applicable. Authors’ contributions KA wrote and prepared the final draft of the manuscript. TK and YN helped draft and review the manuscript. The authors read and approved the final manuscript. Funding Not applicable. Declarations Ethics approval and consent to participate Not applicable. Consent for publication Written informed consent was obtained from the patient for publication of this case report and the accompanying images. Competing interests The authors declare that they have no competing interests. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Oral

DrugAdministrationRoute

CC BY

34888750

2021-12-10

What was the outcome of reaction 'Vomiting'?

Esophageal achalasia detected by vomiting during induction of general anesthesia: a case report. BACKGROUND Esophageal achalasia is a rare disease with a high risk of aspiration during anesthesia induction. Here, we describe our experience involving a case of undiagnosed esophageal achalasia with profuse vomiting during anesthesia induction. METHODS A 58-year-old woman was scheduled for orthopedic surgery under general anesthesia. She vomited a large amount of watery contents during anesthesia induction, and planned surgery was postponed. After recovery from anesthesia, she informed us that she usually had to drink a large amount of water to get food into her stomach and purged watery vomit every night before sleep. However, she attributed it to her constitutional problem, not to a specific disease. She was subsequently diagnosed with esophageal achalasia and underwent Heller myotomy with Dor fundoplication before her re-scheduled orthopedic surgery. CONCLUSIONS A detailed history of dysphagia and regurgitation should be taken in preoperative examinations to prevent unexpected aspiration due to undiagnosed achalasia. pmcBackground Esophageal achalasia is a rare disease (incidence, 1 in 100,000) associated with a high risk of aspiration during the induction of general anesthesia due to difficulties in passing food [1]. Here, we present a patient with undiagnosed esophageal achalasia who vomited profusely during the induction of general anesthesia. Case presentation A 58-year-old woman (height 163 cm, weight 52 kg, body mass index 19.6) with a diagnosis of spinal canal stenosis was admitted to an orthopedic hospital to undergo posterior lumbar interbody fusion. Her medical history included anterior cervical fusion 12 years previously without any anesthetic complications such as allergy or bronchial asthma. No full-time anesthesiologist was on staff at the hospital, so a part-time anesthesiologist was in charge of the anesthesia. The patient had been fasting for 15 h, from dinner the night before surgery until entering the operating room (OR). Four hours before entering the OR, an intravenous infusion was started. The patient was administered ranitidine (150 mg orally) with a small amount of water 180 min before entering the OR. Immediately before starting the anesthesia induction, the anesthesiologist reconfirmed the patient’s medical history and whether she had any allergies or bronchial asthma. The patient did not provide any new medical information. Electrocardiography, pulse oximetry (SpO2), noninvasive blood pressure, and capnometry were applied. After preoxygenation with 100% oxygen for 3 min, general anesthesia was induced with fentanyl 50 μg and propofol 100 mg. After confirming the patient’s spontaneous breathing cessation, assisted ventilation was initiated with a face mask, followed by rocuronium 50 mg. Immediately after that, the patient threw up a massive amount of watery vomit. The vomit in her oral cavity was immediately suctioned, and then her trachea was intubated. A small amount of watery vomit was suctioned through the endotracheal tube. Lung compliance was low with manual ventilation, and wheezing was heard on auscultation with an obstructive pattern on capnography. Bronchospasm due to aspiration was suspected, and treatment was initiated. Under anesthesia (sevoflurane 2% in oxygen 100%), her SpO2 remained above 98%. Hydrocortisone 100 mg and aminophylline 250 mg were administered by drip infusion. Forty-five minutes after the event, her wheezing disappeared, and mostly normal lung compliance returned. The scheduled surgery was postponed, the patient was awakened from anesthesia with the aid of sugammadex to antagonize rocuronium, and her trachea was extubated. After returning to the ward, a more detailed medical history was obtained from the patient. She said that she usually had to drink a large amount of water to get food into her stomach. Therefore, she made it a daily habit to purge watery vomit before bed, as it would reflux when she lay down. Although she had an upper gastrointestinal endoscopy a few years prior, the physician had not pointed out anything unusual. Since then, she believed that this was a constitutional problem as opposed to a disease. The patient had not experienced any weight loss for the past few years. Although no clear image of pneumonia was seen on a chest X-ray at the ward, ampicillin and sulbactam were administered for prevention. The patient was discharged the next day without any medical problems. Based on an overall assessment of the episode, we believed that this patient had esophageal achalasia and advised her to visit a gastroenterologist. At the gastroenterology department in another hospital, fluid retention in the dilated esophagus was revealed by upper endoscopy and chest computed tomography (CT) (Fig. 1). The esophagogram showed a dilated esophagus with a diameter of 32 mm and functional narrowing of the esophagogastric junction (Fig. 2). Esophageal manometry showed the disappearance of primary peristaltic waves and the occurrence of simultaneous contraction waves. Based on these results, the patient was diagnosed with esophageal achalasia and underwent laparoscopic Heller myotomy with Dor fundoplication surgery. After the surgery, her subjective symptoms, including choking with food and daily vomiting before bed, disappeared. She was discharged on the sixth day after surgery. Three months later, she underwent posterior interbody fusion of the lumbar spine without any anesthetic complications.Fig. 1 Preoperative thoracic computed tomography showing fluid retention in the dilated esophagus. a Upper thoracic region. b Middle thoracic region Fig. 2 Preoperative esophagogram showed a dilated esophagus and functional narrowing of the esophagogastric junction. A mouthful of contrast medium was retained in the esophagus until the end of the imaging Discussion Esophageal achalasia is defined as an esophageal motility disorder of unknown etiology characterized by failure of lower esophageal sphincter (LES) relaxation and impaired peristalsis of the lower esophageal body [2]. Diagnosis is difficult because achalasia is a rare disease (incidence, 1 in 100,000 people) with nonspecific subjective symptoms [3]. The most frequent symptoms of achalasia include dysphagia of solids and liquids (> 90%), regurgitation of undigested food (76–91%), respiratory complications such as nocturnal cough (30%) and aspiration (8%), chest pains (25–64%), heartburn (18–52%), and weight loss (35–91%) [4]. Oral reflux in patients with achalasia does not originate from the stomach and thus does not contain acidic contents. Furthermore, if the volume of the oral reflux is mild, the patient may be unaware of the reflux. We assume two reasons for the aspiration during the first scheduled orthopedic surgery in this case. First, the patient was not aware of her rare disease. She had unusual habits such as washing food into her stomach with large amounts of water and purging watery vomit every night before bed. Nonetheless, she did not consider it unusual because her gastroenterologist told her that nothing was wrong. The diagnostic features of esophageal achalasia on upper gastrointestinal endoscopy include (1) dilatation of the esophageal lumen, (2) abnormal retention of food and fluid in the esophagus, (3) whitening and thickening of the esophageal mucosal surface, (4) functional narrowing of the esophagogastric junction, and (5) abnormal contraction waves of the esophagus [2]. In this case, preoperative endoscopy before Heller and Dor surgery showed fluid retention in the esophagus and functional narrowing of the esophagogastric junction, but the dilation of the esophageal lumen was mild. As a result, the gastroenterologist at the time was unable to detect achalasia, leading to her misconception that it was not a disease. Second, we could not obtain information regarding dysphagia and food regurgitation in the preoperative examination. Our routine preoperative examination for orthopedic patients without gastrointestinal complications includes a detailed medical history, allergies, and asthma, but not dysphagia or food regurgitation. In patients with esophageal achalasia, 37% have solid residue, and 14.8% retain water, even after fasting for 24 h before surgery [5]. Fortunately, because this patient had a habit of vomiting every night before bed, and the vomit did not contain any solids, she did not develop severe pneumonia. Achalasia is a rare disease, and only a few cases have been detected by aspiration during the induction of anesthesia [1, 6, 7]. Since achalasia is a chronic benign disease, patients may recognize it as a constitutional problem, as in this case. A detailed history of dysphagia and regurgitation should be taken from all patients during preoperative examinations to avoid aspiration risk, even if achalasia is not suspected. Pillow stains with drool while sleeping suggests the existence of achalasia. Although chest X-rays appear normal in the early phase of achalasia, the dilated esophagus creates new interfaces with the lung as the disease progresses, which makes achalasia-specific findings, including convex opacity overlapping the right mediastinum, air-fluid levels in the thorax esophagus, and small or absent gastric bubbles [8]. In this case, the esophageal dilatation was so mild that no suspicious contour was seen on the chest X-ray (Fig. 3). However, it should be worthwhile for physicians to get into the habit of looking for achalasia-specific findings on routine X-ray readings to detect undiagnosed achalasia. Furthermore, we would like to emphasize that achalasia should not be ruled out even if there are no findings specific to achalasia, especially in patients with symptoms such as dysphagia or regurgitation. If a chest CT is taken for the preoperative evaluation, physicians should also note the dilated image of the esophagus.Fig. 3 Preoperative chest X-ray did not show any abnormalities Conclusions Although esophageal achalasia is a rare disease, it is associated with a high risk of aspiration pneumonia if general anesthesia is induced without caution. A detailed history of dysphagia and regurgitation of solids and liquids in the preoperative examination is essential and practical for detecting the existence of achalasia and preventing unexpected aspiration due to undiagnosed achalasia. Additionally, getting into the habit of checking for achalasia-specific findings on chest X-rays and chest CT may reduce the occurrence of unexpected aspiration. Acknowledgements Not applicable. Authors’ contributions KA wrote and prepared the final draft of the manuscript. TK and YN helped draft and review the manuscript. The authors read and approved the final manuscript. Funding Not applicable. Declarations Ethics approval and consent to participate Not applicable. Consent for publication Written informed consent was obtained from the patient for publication of this case report and the accompanying images. Competing interests The authors declare that they have no competing interests. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Recovered

ReactionOutcome

CC BY

34888750

2021-12-10

What was the outcome of reaction 'Wheezing'?

Esophageal achalasia detected by vomiting during induction of general anesthesia: a case report. BACKGROUND Esophageal achalasia is a rare disease with a high risk of aspiration during anesthesia induction. Here, we describe our experience involving a case of undiagnosed esophageal achalasia with profuse vomiting during anesthesia induction. METHODS A 58-year-old woman was scheduled for orthopedic surgery under general anesthesia. She vomited a large amount of watery contents during anesthesia induction, and planned surgery was postponed. After recovery from anesthesia, she informed us that she usually had to drink a large amount of water to get food into her stomach and purged watery vomit every night before sleep. However, she attributed it to her constitutional problem, not to a specific disease. She was subsequently diagnosed with esophageal achalasia and underwent Heller myotomy with Dor fundoplication before her re-scheduled orthopedic surgery. CONCLUSIONS A detailed history of dysphagia and regurgitation should be taken in preoperative examinations to prevent unexpected aspiration due to undiagnosed achalasia. pmcBackground Esophageal achalasia is a rare disease (incidence, 1 in 100,000) associated with a high risk of aspiration during the induction of general anesthesia due to difficulties in passing food [1]. Here, we present a patient with undiagnosed esophageal achalasia who vomited profusely during the induction of general anesthesia. Case presentation A 58-year-old woman (height 163 cm, weight 52 kg, body mass index 19.6) with a diagnosis of spinal canal stenosis was admitted to an orthopedic hospital to undergo posterior lumbar interbody fusion. Her medical history included anterior cervical fusion 12 years previously without any anesthetic complications such as allergy or bronchial asthma. No full-time anesthesiologist was on staff at the hospital, so a part-time anesthesiologist was in charge of the anesthesia. The patient had been fasting for 15 h, from dinner the night before surgery until entering the operating room (OR). Four hours before entering the OR, an intravenous infusion was started. The patient was administered ranitidine (150 mg orally) with a small amount of water 180 min before entering the OR. Immediately before starting the anesthesia induction, the anesthesiologist reconfirmed the patient’s medical history and whether she had any allergies or bronchial asthma. The patient did not provide any new medical information. Electrocardiography, pulse oximetry (SpO2), noninvasive blood pressure, and capnometry were applied. After preoxygenation with 100% oxygen for 3 min, general anesthesia was induced with fentanyl 50 μg and propofol 100 mg. After confirming the patient’s spontaneous breathing cessation, assisted ventilation was initiated with a face mask, followed by rocuronium 50 mg. Immediately after that, the patient threw up a massive amount of watery vomit. The vomit in her oral cavity was immediately suctioned, and then her trachea was intubated. A small amount of watery vomit was suctioned through the endotracheal tube. Lung compliance was low with manual ventilation, and wheezing was heard on auscultation with an obstructive pattern on capnography. Bronchospasm due to aspiration was suspected, and treatment was initiated. Under anesthesia (sevoflurane 2% in oxygen 100%), her SpO2 remained above 98%. Hydrocortisone 100 mg and aminophylline 250 mg were administered by drip infusion. Forty-five minutes after the event, her wheezing disappeared, and mostly normal lung compliance returned. The scheduled surgery was postponed, the patient was awakened from anesthesia with the aid of sugammadex to antagonize rocuronium, and her trachea was extubated. After returning to the ward, a more detailed medical history was obtained from the patient. She said that she usually had to drink a large amount of water to get food into her stomach. Therefore, she made it a daily habit to purge watery vomit before bed, as it would reflux when she lay down. Although she had an upper gastrointestinal endoscopy a few years prior, the physician had not pointed out anything unusual. Since then, she believed that this was a constitutional problem as opposed to a disease. The patient had not experienced any weight loss for the past few years. Although no clear image of pneumonia was seen on a chest X-ray at the ward, ampicillin and sulbactam were administered for prevention. The patient was discharged the next day without any medical problems. Based on an overall assessment of the episode, we believed that this patient had esophageal achalasia and advised her to visit a gastroenterologist. At the gastroenterology department in another hospital, fluid retention in the dilated esophagus was revealed by upper endoscopy and chest computed tomography (CT) (Fig. 1). The esophagogram showed a dilated esophagus with a diameter of 32 mm and functional narrowing of the esophagogastric junction (Fig. 2). Esophageal manometry showed the disappearance of primary peristaltic waves and the occurrence of simultaneous contraction waves. Based on these results, the patient was diagnosed with esophageal achalasia and underwent laparoscopic Heller myotomy with Dor fundoplication surgery. After the surgery, her subjective symptoms, including choking with food and daily vomiting before bed, disappeared. She was discharged on the sixth day after surgery. Three months later, she underwent posterior interbody fusion of the lumbar spine without any anesthetic complications.Fig. 1 Preoperative thoracic computed tomography showing fluid retention in the dilated esophagus. a Upper thoracic region. b Middle thoracic region Fig. 2 Preoperative esophagogram showed a dilated esophagus and functional narrowing of the esophagogastric junction. A mouthful of contrast medium was retained in the esophagus until the end of the imaging Discussion Esophageal achalasia is defined as an esophageal motility disorder of unknown etiology characterized by failure of lower esophageal sphincter (LES) relaxation and impaired peristalsis of the lower esophageal body [2]. Diagnosis is difficult because achalasia is a rare disease (incidence, 1 in 100,000 people) with nonspecific subjective symptoms [3]. The most frequent symptoms of achalasia include dysphagia of solids and liquids (> 90%), regurgitation of undigested food (76–91%), respiratory complications such as nocturnal cough (30%) and aspiration (8%), chest pains (25–64%), heartburn (18–52%), and weight loss (35–91%) [4]. Oral reflux in patients with achalasia does not originate from the stomach and thus does not contain acidic contents. Furthermore, if the volume of the oral reflux is mild, the patient may be unaware of the reflux. We assume two reasons for the aspiration during the first scheduled orthopedic surgery in this case. First, the patient was not aware of her rare disease. She had unusual habits such as washing food into her stomach with large amounts of water and purging watery vomit every night before bed. Nonetheless, she did not consider it unusual because her gastroenterologist told her that nothing was wrong. The diagnostic features of esophageal achalasia on upper gastrointestinal endoscopy include (1) dilatation of the esophageal lumen, (2) abnormal retention of food and fluid in the esophagus, (3) whitening and thickening of the esophageal mucosal surface, (4) functional narrowing of the esophagogastric junction, and (5) abnormal contraction waves of the esophagus [2]. In this case, preoperative endoscopy before Heller and Dor surgery showed fluid retention in the esophagus and functional narrowing of the esophagogastric junction, but the dilation of the esophageal lumen was mild. As a result, the gastroenterologist at the time was unable to detect achalasia, leading to her misconception that it was not a disease. Second, we could not obtain information regarding dysphagia and food regurgitation in the preoperative examination. Our routine preoperative examination for orthopedic patients without gastrointestinal complications includes a detailed medical history, allergies, and asthma, but not dysphagia or food regurgitation. In patients with esophageal achalasia, 37% have solid residue, and 14.8% retain water, even after fasting for 24 h before surgery [5]. Fortunately, because this patient had a habit of vomiting every night before bed, and the vomit did not contain any solids, she did not develop severe pneumonia. Achalasia is a rare disease, and only a few cases have been detected by aspiration during the induction of anesthesia [1, 6, 7]. Since achalasia is a chronic benign disease, patients may recognize it as a constitutional problem, as in this case. A detailed history of dysphagia and regurgitation should be taken from all patients during preoperative examinations to avoid aspiration risk, even if achalasia is not suspected. Pillow stains with drool while sleeping suggests the existence of achalasia. Although chest X-rays appear normal in the early phase of achalasia, the dilated esophagus creates new interfaces with the lung as the disease progresses, which makes achalasia-specific findings, including convex opacity overlapping the right mediastinum, air-fluid levels in the thorax esophagus, and small or absent gastric bubbles [8]. In this case, the esophageal dilatation was so mild that no suspicious contour was seen on the chest X-ray (Fig. 3). However, it should be worthwhile for physicians to get into the habit of looking for achalasia-specific findings on routine X-ray readings to detect undiagnosed achalasia. Furthermore, we would like to emphasize that achalasia should not be ruled out even if there are no findings specific to achalasia, especially in patients with symptoms such as dysphagia or regurgitation. If a chest CT is taken for the preoperative evaluation, physicians should also note the dilated image of the esophagus.Fig. 3 Preoperative chest X-ray did not show any abnormalities Conclusions Although esophageal achalasia is a rare disease, it is associated with a high risk of aspiration pneumonia if general anesthesia is induced without caution. A detailed history of dysphagia and regurgitation of solids and liquids in the preoperative examination is essential and practical for detecting the existence of achalasia and preventing unexpected aspiration due to undiagnosed achalasia. Additionally, getting into the habit of checking for achalasia-specific findings on chest X-rays and chest CT may reduce the occurrence of unexpected aspiration. Acknowledgements Not applicable. Authors’ contributions KA wrote and prepared the final draft of the manuscript. TK and YN helped draft and review the manuscript. The authors read and approved the final manuscript. Funding Not applicable. Declarations Ethics approval and consent to participate Not applicable. Consent for publication Written informed consent was obtained from the patient for publication of this case report and the accompanying images. Competing interests The authors declare that they have no competing interests. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Recovered

ReactionOutcome

CC BY

34888750

2021-12-10

Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Drug interaction'.

Early intervention of acute liver injury related to venlafaxine: A case report. BACKGROUND Drug-induced liver injury (DILI) is the leading cause of acute liver injury (ALI), market withdrawal of a drug, and rejection of applications for marketing licenses. The incidence of DILI is very low, with a value between 1 and 19 per 100,000 patient years. All antidepressants may induce DILI even at low therapeutic doses. In this report, we present a case of ALI after venlafaxine administration. A 27-year-old Chinese Han woman was admitted for depression. Several serum liver function indices in this patient were abnormal after antidepressant treatment. The Roussel Uclaf Causality Assessment Method (RUCAM) causality assessment score was 8, and the R value was 31.18. The patient was diagnosed with hepatocellular ALI, which was derived from venlafaxine-related adverse events. METHODS First, all medications were stopped to block the progression of DILI. Then, a hepatoprotective strategy and proper psychological treatment were performed to recover the impaired hepatic function. RESULTS Liver function was fully recovered as indicated by liver function indices and ultrasound imaging. CONCLUSIONS The possibility of DILI should not be overlooked during the long-term use of antipsychotic drugs. In response, regular liver function monitoring should be performed in a timely manner to avoid missing diagnoses and delayed treatment. Furthermore, the necessary medical treatment needs to be conducted after the occurrence of ALI. pmc1 Introduction As a serotonin and norepinephrine reuptake inhibitor, venlafaxine (VEN) has been proven to be well tolerated with a rate of adverse drug reactions (ADRs) of less than 1/1000.[1] The most common ADRs[2,3] are mild, including nausea, vomiting, dizziness, diarrhea, dry mouth, decreased appetite, constipation, somnolence, and so on, all of which might be caused by serotonin toxicity or vulnerability to drug-drug interactions.[4] Notably, VEN-induced acute liver injury (ALI) has been less reported since its approval in 1993.[5–7] It has been confirmed that VEN can result in transient asymptomatic elevations in serum aminotransferase levels and has been linked to rare instances of clinically apparent ALI.[8] However, the mechanism by which VEN causes ALI is unknown. In this case report, we describe 1 patient with depression who was diagnosed with ALI after the administration of VEN. 2 Case presentation A 27-year-old Chinese Han woman was diagnosed with depression. She had no chronic or metabolic liver disease, blood transfusion, or history of alcohol consumption. The patient was admitted to our Department of Clinical Psychology on March 9, 2021, and low-dose VEN 50 mg (oral, qd) was started to give to her. Meanwhile, liver function was normal according to the test results (Table 1). The dosage of VEN was gradually increased to 225 mg/d on March 19 because of severe depression. The patient showed good tolerance without any symptoms. Additionally, she also received lorazepam 1 mg (oral, qn) and zolpidem tartrate 10 mg (oral, qn) for treating anxiety symptoms and dyssomnia during this period. After administering these treatments, depressive symptoms were well controlled. Table 1 The results of liver function tests and the reference values. Aminotransferase March 10 April 14 April 19 April 29 Reference value AST 13 IU/L 767 IU/L 24 IU/L 17 IU/L 5–50 IU/L ALT 12 IU/L 1777 IU/L 269 IU/L 37 IU/L 7–40 IU/L ALP 39 IU/L 171 IU/L 85.5 IU/L 57.0 IU/L 40–150 IU/L γ-GT 7 IU/L 147 IU/L 90 IU/L 48 IU/L 7–50 IU/L 5′-nucleotidease 2.9 U/L 37 U/L 12 U/L 5.3 U/L 2–11.4 U/L Before leaving the hospital, the liver function as well as her blood routine were reviewed on April 14, 2021. The results of routine blood tests were normal, while the symptoms of ALI (Table 1) were abnormal elevation of serum aspartate aminotransferase to 767 IU/L, alanine aminotransferase to 1777 IU/L, alkaline phosphatase to 171 IU/L, γ-glutamyltransferase to 147 IU/L, and 5’-nucleotidease to 37 U/L. Nevertheless, neither dilatation of the intrahepatic or extrahepatic bile ducts nor hepatomegaly or splenomegaly were detected by ultrasound of the liver (Fig. 1). These results indicate that ALI was still in its early stages. The results of Roussel Uclaf Causality Assessment Method (RUCAM) assessment (score: 8) illustrated that this ALI (hepatocellular, R = 31.18) was derived from VEN-related adverse events. Figure 1 The ultrasound of the liver on March 11, 2021 (A) and April 14, 2021 (B). There was no dilatation of the intrahepatic or extrahepatic bile ducts and hepatomegaly or splenomegaly appeared in 2 images. Based on these considerations, stopping all medications was adopted to block the progression of ALI. g of glutathione 1.08 g of magnesium isoglycyrrhizinate 150 mg (intravenous, qd) and silibinin meglumine 100 mg (oral, tid) were administered to the patient. Several indices of liver function were recovered based on the results from April 19, 2021 (Table 1). Therefore, the current hepatoprotective strategy, as well as proper psychological treatment, were held for this patient. The liver function fully recovered on April 29, 2021 (Table 1), and she was discharged from the hospital the next day. 3 Discussion ALI is a rare manifestation of VEN-induced ADRs. Despite the unknown mechanism, we attempted to reveal the pathogeny of VEN-related ALI in this case. It has been reported that the dosage of VEN-related liver function abnormalities is in the range of 25 to 300 mg/d with a median dose of 75 mg/d, and the upper limit of normal for concentration of VEN in the patients’ plasma is 400 ng/mL.[9] It is critical to note that this upper value may be exceeded in patients who receive only the conventional dosage. This individual difference increases the risk of dose-related adverse events. In view of these facts, we firstly performed the plasma drug concentration measurements and the result was only 93.9 ng/mL, which was even slightly lower than the recommended therapeutic concentration of Arbeitsgemeinschaft für Neuropsychopharmakologie und Pharmakopsychiatrie (100–400 ng/mL). Second, we attempted to determine the metabolic features of VEN in this patient. Cytochrome P450 (CYP) 2D6 in the liver is the predominant metabolic enzyme responsible for converting 90 percent of VEN to O-desmethylvenlafaxine (desvenlafaxine).[10] Additionally, CYP2C19 mediates the remaining 10% of this demethylation. Indeed, extensive CYP2D6 and CYP2C19 metabolizers always increase the clearance of VEN and maintain a low plasma concentration. To our surprise, results of genetic polymorphism testing showed that CYP2D6 was a normal metabolizer with rs1065852 CT, rs16947 CT, and rs5030865 GG, while CYP2C19 was an extensive metabolizer with rs4244285 GG, rs4986893 GG, and rs12248560 CC, which is in accordance with the low exposure of VEN. Third, the vulnerability of VEN to drug-drug interactions can also cause ADRs. In this case, the fast metabolic characteristics of VEN lead to a high abundance of cytochrome isoenzymes (mainly CYP2D6 and CYP2C19) as well as rapidly accumulated O-desmethylvenlafaxine, which are thought to be involved in VEN-associated liver injury. As a reversible profile of hepatic damage, accurate and timely liver function monitoring and the necessary measures taken after abnormalities in the liver are crucial to recover its function. Corticosteroids, such as methylprednisolone, are used as hepatic protectants against VEN-induced liver injury.[11] As another liver cell protective agent, magnesium isoglycyrate always plays a hepatoprotective effect by protecting the liver cell membrane, improving liver function, and its anti-inflammatory ability. It can inhibit the increase in serum transaminase, reduce the degeneration, necrosis, and inflammatory cell infiltration of liver cells, and hence repair the liver tissue and recover its activity.[12–15] In addition, VEN was shown to induce ROS formation to enhance the membrane permeability of mitochondria and lysosomes, triggering apoptosis or necrosis.[16] Thus, ROS scavenger agents have been identified as promising therapeutic strategies against drug/xenobitic-induced liver injuries.[17] 4 Conclusion During the administration of VEN or other antidepressants, liver function monitoring is crucial, and immediate drug withdrawal as well as necessary medical treatment should be taken in cases of ALI to obtain satisfactory post-treatment results. Furthermore, more samples and further explorations are still needed to better understand the etiology, early diagnosis, and treatment of VEN-induced ALI. We hope that our report will provide an effective reference for future research on this DILI. Author contributions Conceptualization: Lin Fang, Kun Yao. Funding acquisition: Kun Yao. Writing – original draft: Lin Fang, Shushan Wang. Writing – review & editing: Leiming Cao, Kun Yao. Abbreviations: ADRs = adverse drug reactions, ALI = acute liver injury, DILI = drug-induced liver injure, VEN = venlafaxine. How to cite this article: Fang L, Wang S, Cao L, Yao K. Early intervention of acute liver injury related to venlafaxine: a case report. Medicine. 2021;100:49(e28140). Written informed consent was obtained from the patient for the publication of this case report and accompanying images. This work was supported by Fund for development of Science and Technology of Nanjing Medical University (NMUB2020295). The authors have no conflicts of interest to disclose. The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request γ-GT = γ-glutamyltransferase, ALT = alanine aminotransferase, ALP = alkaline phosphatase, AST = aspartate aminotransferase.

LORAZEPAM, VENLAFAXINE HYDROCHLORIDE, ZOLPIDEM TARTRATE

DrugsGivenReaction

CC BY

34889278

2021-12-10

Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Drug-induced liver injury'.

Early intervention of acute liver injury related to venlafaxine: A case report. BACKGROUND Drug-induced liver injury (DILI) is the leading cause of acute liver injury (ALI), market withdrawal of a drug, and rejection of applications for marketing licenses. The incidence of DILI is very low, with a value between 1 and 19 per 100,000 patient years. All antidepressants may induce DILI even at low therapeutic doses. In this report, we present a case of ALI after venlafaxine administration. A 27-year-old Chinese Han woman was admitted for depression. Several serum liver function indices in this patient were abnormal after antidepressant treatment. The Roussel Uclaf Causality Assessment Method (RUCAM) causality assessment score was 8, and the R value was 31.18. The patient was diagnosed with hepatocellular ALI, which was derived from venlafaxine-related adverse events. METHODS First, all medications were stopped to block the progression of DILI. Then, a hepatoprotective strategy and proper psychological treatment were performed to recover the impaired hepatic function. RESULTS Liver function was fully recovered as indicated by liver function indices and ultrasound imaging. CONCLUSIONS The possibility of DILI should not be overlooked during the long-term use of antipsychotic drugs. In response, regular liver function monitoring should be performed in a timely manner to avoid missing diagnoses and delayed treatment. Furthermore, the necessary medical treatment needs to be conducted after the occurrence of ALI. pmc1 Introduction As a serotonin and norepinephrine reuptake inhibitor, venlafaxine (VEN) has been proven to be well tolerated with a rate of adverse drug reactions (ADRs) of less than 1/1000.[1] The most common ADRs[2,3] are mild, including nausea, vomiting, dizziness, diarrhea, dry mouth, decreased appetite, constipation, somnolence, and so on, all of which might be caused by serotonin toxicity or vulnerability to drug-drug interactions.[4] Notably, VEN-induced acute liver injury (ALI) has been less reported since its approval in 1993.[5–7] It has been confirmed that VEN can result in transient asymptomatic elevations in serum aminotransferase levels and has been linked to rare instances of clinically apparent ALI.[8] However, the mechanism by which VEN causes ALI is unknown. In this case report, we describe 1 patient with depression who was diagnosed with ALI after the administration of VEN. 2 Case presentation A 27-year-old Chinese Han woman was diagnosed with depression. She had no chronic or metabolic liver disease, blood transfusion, or history of alcohol consumption. The patient was admitted to our Department of Clinical Psychology on March 9, 2021, and low-dose VEN 50 mg (oral, qd) was started to give to her. Meanwhile, liver function was normal according to the test results (Table 1). The dosage of VEN was gradually increased to 225 mg/d on March 19 because of severe depression. The patient showed good tolerance without any symptoms. Additionally, she also received lorazepam 1 mg (oral, qn) and zolpidem tartrate 10 mg (oral, qn) for treating anxiety symptoms and dyssomnia during this period. After administering these treatments, depressive symptoms were well controlled. Table 1 The results of liver function tests and the reference values. Aminotransferase March 10 April 14 April 19 April 29 Reference value AST 13 IU/L 767 IU/L 24 IU/L 17 IU/L 5–50 IU/L ALT 12 IU/L 1777 IU/L 269 IU/L 37 IU/L 7–40 IU/L ALP 39 IU/L 171 IU/L 85.5 IU/L 57.0 IU/L 40–150 IU/L γ-GT 7 IU/L 147 IU/L 90 IU/L 48 IU/L 7–50 IU/L 5′-nucleotidease 2.9 U/L 37 U/L 12 U/L 5.3 U/L 2–11.4 U/L Before leaving the hospital, the liver function as well as her blood routine were reviewed on April 14, 2021. The results of routine blood tests were normal, while the symptoms of ALI (Table 1) were abnormal elevation of serum aspartate aminotransferase to 767 IU/L, alanine aminotransferase to 1777 IU/L, alkaline phosphatase to 171 IU/L, γ-glutamyltransferase to 147 IU/L, and 5’-nucleotidease to 37 U/L. Nevertheless, neither dilatation of the intrahepatic or extrahepatic bile ducts nor hepatomegaly or splenomegaly were detected by ultrasound of the liver (Fig. 1). These results indicate that ALI was still in its early stages. The results of Roussel Uclaf Causality Assessment Method (RUCAM) assessment (score: 8) illustrated that this ALI (hepatocellular, R = 31.18) was derived from VEN-related adverse events. Figure 1 The ultrasound of the liver on March 11, 2021 (A) and April 14, 2021 (B). There was no dilatation of the intrahepatic or extrahepatic bile ducts and hepatomegaly or splenomegaly appeared in 2 images. Based on these considerations, stopping all medications was adopted to block the progression of ALI. g of glutathione 1.08 g of magnesium isoglycyrrhizinate 150 mg (intravenous, qd) and silibinin meglumine 100 mg (oral, tid) were administered to the patient. Several indices of liver function were recovered based on the results from April 19, 2021 (Table 1). Therefore, the current hepatoprotective strategy, as well as proper psychological treatment, were held for this patient. The liver function fully recovered on April 29, 2021 (Table 1), and she was discharged from the hospital the next day. 3 Discussion ALI is a rare manifestation of VEN-induced ADRs. Despite the unknown mechanism, we attempted to reveal the pathogeny of VEN-related ALI in this case. It has been reported that the dosage of VEN-related liver function abnormalities is in the range of 25 to 300 mg/d with a median dose of 75 mg/d, and the upper limit of normal for concentration of VEN in the patients’ plasma is 400 ng/mL.[9] It is critical to note that this upper value may be exceeded in patients who receive only the conventional dosage. This individual difference increases the risk of dose-related adverse events. In view of these facts, we firstly performed the plasma drug concentration measurements and the result was only 93.9 ng/mL, which was even slightly lower than the recommended therapeutic concentration of Arbeitsgemeinschaft für Neuropsychopharmakologie und Pharmakopsychiatrie (100–400 ng/mL). Second, we attempted to determine the metabolic features of VEN in this patient. Cytochrome P450 (CYP) 2D6 in the liver is the predominant metabolic enzyme responsible for converting 90 percent of VEN to O-desmethylvenlafaxine (desvenlafaxine).[10] Additionally, CYP2C19 mediates the remaining 10% of this demethylation. Indeed, extensive CYP2D6 and CYP2C19 metabolizers always increase the clearance of VEN and maintain a low plasma concentration. To our surprise, results of genetic polymorphism testing showed that CYP2D6 was a normal metabolizer with rs1065852 CT, rs16947 CT, and rs5030865 GG, while CYP2C19 was an extensive metabolizer with rs4244285 GG, rs4986893 GG, and rs12248560 CC, which is in accordance with the low exposure of VEN. Third, the vulnerability of VEN to drug-drug interactions can also cause ADRs. In this case, the fast metabolic characteristics of VEN lead to a high abundance of cytochrome isoenzymes (mainly CYP2D6 and CYP2C19) as well as rapidly accumulated O-desmethylvenlafaxine, which are thought to be involved in VEN-associated liver injury. As a reversible profile of hepatic damage, accurate and timely liver function monitoring and the necessary measures taken after abnormalities in the liver are crucial to recover its function. Corticosteroids, such as methylprednisolone, are used as hepatic protectants against VEN-induced liver injury.[11] As another liver cell protective agent, magnesium isoglycyrate always plays a hepatoprotective effect by protecting the liver cell membrane, improving liver function, and its anti-inflammatory ability. It can inhibit the increase in serum transaminase, reduce the degeneration, necrosis, and inflammatory cell infiltration of liver cells, and hence repair the liver tissue and recover its activity.[12–15] In addition, VEN was shown to induce ROS formation to enhance the membrane permeability of mitochondria and lysosomes, triggering apoptosis or necrosis.[16] Thus, ROS scavenger agents have been identified as promising therapeutic strategies against drug/xenobitic-induced liver injuries.[17] 4 Conclusion During the administration of VEN or other antidepressants, liver function monitoring is crucial, and immediate drug withdrawal as well as necessary medical treatment should be taken in cases of ALI to obtain satisfactory post-treatment results. Furthermore, more samples and further explorations are still needed to better understand the etiology, early diagnosis, and treatment of VEN-induced ALI. We hope that our report will provide an effective reference for future research on this DILI. Author contributions Conceptualization: Lin Fang, Kun Yao. Funding acquisition: Kun Yao. Writing – original draft: Lin Fang, Shushan Wang. Writing – review & editing: Leiming Cao, Kun Yao. Abbreviations: ADRs = adverse drug reactions, ALI = acute liver injury, DILI = drug-induced liver injure, VEN = venlafaxine. How to cite this article: Fang L, Wang S, Cao L, Yao K. Early intervention of acute liver injury related to venlafaxine: a case report. Medicine. 2021;100:49(e28140). Written informed consent was obtained from the patient for the publication of this case report and accompanying images. This work was supported by Fund for development of Science and Technology of Nanjing Medical University (NMUB2020295). The authors have no conflicts of interest to disclose. The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request γ-GT = γ-glutamyltransferase, ALT = alanine aminotransferase, ALP = alkaline phosphatase, AST = aspartate aminotransferase.

LORAZEPAM, VENLAFAXINE HYDROCHLORIDE, ZOLPIDEM TARTRATE

DrugsGivenReaction

CC BY

34889278

2021-12-10

Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Hepatitis'.

Early intervention of acute liver injury related to venlafaxine: A case report. BACKGROUND Drug-induced liver injury (DILI) is the leading cause of acute liver injury (ALI), market withdrawal of a drug, and rejection of applications for marketing licenses. The incidence of DILI is very low, with a value between 1 and 19 per 100,000 patient years. All antidepressants may induce DILI even at low therapeutic doses. In this report, we present a case of ALI after venlafaxine administration. A 27-year-old Chinese Han woman was admitted for depression. Several serum liver function indices in this patient were abnormal after antidepressant treatment. The Roussel Uclaf Causality Assessment Method (RUCAM) causality assessment score was 8, and the R value was 31.18. The patient was diagnosed with hepatocellular ALI, which was derived from venlafaxine-related adverse events. METHODS First, all medications were stopped to block the progression of DILI. Then, a hepatoprotective strategy and proper psychological treatment were performed to recover the impaired hepatic function. RESULTS Liver function was fully recovered as indicated by liver function indices and ultrasound imaging. CONCLUSIONS The possibility of DILI should not be overlooked during the long-term use of antipsychotic drugs. In response, regular liver function monitoring should be performed in a timely manner to avoid missing diagnoses and delayed treatment. Furthermore, the necessary medical treatment needs to be conducted after the occurrence of ALI. pmc1 Introduction As a serotonin and norepinephrine reuptake inhibitor, venlafaxine (VEN) has been proven to be well tolerated with a rate of adverse drug reactions (ADRs) of less than 1/1000.[1] The most common ADRs[2,3] are mild, including nausea, vomiting, dizziness, diarrhea, dry mouth, decreased appetite, constipation, somnolence, and so on, all of which might be caused by serotonin toxicity or vulnerability to drug-drug interactions.[4] Notably, VEN-induced acute liver injury (ALI) has been less reported since its approval in 1993.[5–7] It has been confirmed that VEN can result in transient asymptomatic elevations in serum aminotransferase levels and has been linked to rare instances of clinically apparent ALI.[8] However, the mechanism by which VEN causes ALI is unknown. In this case report, we describe 1 patient with depression who was diagnosed with ALI after the administration of VEN. 2 Case presentation A 27-year-old Chinese Han woman was diagnosed with depression. She had no chronic or metabolic liver disease, blood transfusion, or history of alcohol consumption. The patient was admitted to our Department of Clinical Psychology on March 9, 2021, and low-dose VEN 50 mg (oral, qd) was started to give to her. Meanwhile, liver function was normal according to the test results (Table 1). The dosage of VEN was gradually increased to 225 mg/d on March 19 because of severe depression. The patient showed good tolerance without any symptoms. Additionally, she also received lorazepam 1 mg (oral, qn) and zolpidem tartrate 10 mg (oral, qn) for treating anxiety symptoms and dyssomnia during this period. After administering these treatments, depressive symptoms were well controlled. Table 1 The results of liver function tests and the reference values. Aminotransferase March 10 April 14 April 19 April 29 Reference value AST 13 IU/L 767 IU/L 24 IU/L 17 IU/L 5–50 IU/L ALT 12 IU/L 1777 IU/L 269 IU/L 37 IU/L 7–40 IU/L ALP 39 IU/L 171 IU/L 85.5 IU/L 57.0 IU/L 40–150 IU/L γ-GT 7 IU/L 147 IU/L 90 IU/L 48 IU/L 7–50 IU/L 5′-nucleotidease 2.9 U/L 37 U/L 12 U/L 5.3 U/L 2–11.4 U/L Before leaving the hospital, the liver function as well as her blood routine were reviewed on April 14, 2021. The results of routine blood tests were normal, while the symptoms of ALI (Table 1) were abnormal elevation of serum aspartate aminotransferase to 767 IU/L, alanine aminotransferase to 1777 IU/L, alkaline phosphatase to 171 IU/L, γ-glutamyltransferase to 147 IU/L, and 5’-nucleotidease to 37 U/L. Nevertheless, neither dilatation of the intrahepatic or extrahepatic bile ducts nor hepatomegaly or splenomegaly were detected by ultrasound of the liver (Fig. 1). These results indicate that ALI was still in its early stages. The results of Roussel Uclaf Causality Assessment Method (RUCAM) assessment (score: 8) illustrated that this ALI (hepatocellular, R = 31.18) was derived from VEN-related adverse events. Figure 1 The ultrasound of the liver on March 11, 2021 (A) and April 14, 2021 (B). There was no dilatation of the intrahepatic or extrahepatic bile ducts and hepatomegaly or splenomegaly appeared in 2 images. Based on these considerations, stopping all medications was adopted to block the progression of ALI. g of glutathione 1.08 g of magnesium isoglycyrrhizinate 150 mg (intravenous, qd) and silibinin meglumine 100 mg (oral, tid) were administered to the patient. Several indices of liver function were recovered based on the results from April 19, 2021 (Table 1). Therefore, the current hepatoprotective strategy, as well as proper psychological treatment, were held for this patient. The liver function fully recovered on April 29, 2021 (Table 1), and she was discharged from the hospital the next day. 3 Discussion ALI is a rare manifestation of VEN-induced ADRs. Despite the unknown mechanism, we attempted to reveal the pathogeny of VEN-related ALI in this case. It has been reported that the dosage of VEN-related liver function abnormalities is in the range of 25 to 300 mg/d with a median dose of 75 mg/d, and the upper limit of normal for concentration of VEN in the patients’ plasma is 400 ng/mL.[9] It is critical to note that this upper value may be exceeded in patients who receive only the conventional dosage. This individual difference increases the risk of dose-related adverse events. In view of these facts, we firstly performed the plasma drug concentration measurements and the result was only 93.9 ng/mL, which was even slightly lower than the recommended therapeutic concentration of Arbeitsgemeinschaft für Neuropsychopharmakologie und Pharmakopsychiatrie (100–400 ng/mL). Second, we attempted to determine the metabolic features of VEN in this patient. Cytochrome P450 (CYP) 2D6 in the liver is the predominant metabolic enzyme responsible for converting 90 percent of VEN to O-desmethylvenlafaxine (desvenlafaxine).[10] Additionally, CYP2C19 mediates the remaining 10% of this demethylation. Indeed, extensive CYP2D6 and CYP2C19 metabolizers always increase the clearance of VEN and maintain a low plasma concentration. To our surprise, results of genetic polymorphism testing showed that CYP2D6 was a normal metabolizer with rs1065852 CT, rs16947 CT, and rs5030865 GG, while CYP2C19 was an extensive metabolizer with rs4244285 GG, rs4986893 GG, and rs12248560 CC, which is in accordance with the low exposure of VEN. Third, the vulnerability of VEN to drug-drug interactions can also cause ADRs. In this case, the fast metabolic characteristics of VEN lead to a high abundance of cytochrome isoenzymes (mainly CYP2D6 and CYP2C19) as well as rapidly accumulated O-desmethylvenlafaxine, which are thought to be involved in VEN-associated liver injury. As a reversible profile of hepatic damage, accurate and timely liver function monitoring and the necessary measures taken after abnormalities in the liver are crucial to recover its function. Corticosteroids, such as methylprednisolone, are used as hepatic protectants against VEN-induced liver injury.[11] As another liver cell protective agent, magnesium isoglycyrate always plays a hepatoprotective effect by protecting the liver cell membrane, improving liver function, and its anti-inflammatory ability. It can inhibit the increase in serum transaminase, reduce the degeneration, necrosis, and inflammatory cell infiltration of liver cells, and hence repair the liver tissue and recover its activity.[12–15] In addition, VEN was shown to induce ROS formation to enhance the membrane permeability of mitochondria and lysosomes, triggering apoptosis or necrosis.[16] Thus, ROS scavenger agents have been identified as promising therapeutic strategies against drug/xenobitic-induced liver injuries.[17] 4 Conclusion During the administration of VEN or other antidepressants, liver function monitoring is crucial, and immediate drug withdrawal as well as necessary medical treatment should be taken in cases of ALI to obtain satisfactory post-treatment results. Furthermore, more samples and further explorations are still needed to better understand the etiology, early diagnosis, and treatment of VEN-induced ALI. We hope that our report will provide an effective reference for future research on this DILI. Author contributions Conceptualization: Lin Fang, Kun Yao. Funding acquisition: Kun Yao. Writing – original draft: Lin Fang, Shushan Wang. Writing – review & editing: Leiming Cao, Kun Yao. Abbreviations: ADRs = adverse drug reactions, ALI = acute liver injury, DILI = drug-induced liver injure, VEN = venlafaxine. How to cite this article: Fang L, Wang S, Cao L, Yao K. Early intervention of acute liver injury related to venlafaxine: a case report. Medicine. 2021;100:49(e28140). Written informed consent was obtained from the patient for the publication of this case report and accompanying images. This work was supported by Fund for development of Science and Technology of Nanjing Medical University (NMUB2020295). The authors have no conflicts of interest to disclose. The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request γ-GT = γ-glutamyltransferase, ALT = alanine aminotransferase, ALP = alkaline phosphatase, AST = aspartate aminotransferase.

LORAZEPAM, VENLAFAXINE HYDROCHLORIDE, ZOLPIDEM TARTRATE

DrugsGivenReaction

CC BY

34889278

2021-12-10

Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Liver injury'.

Early intervention of acute liver injury related to venlafaxine: A case report. BACKGROUND Drug-induced liver injury (DILI) is the leading cause of acute liver injury (ALI), market withdrawal of a drug, and rejection of applications for marketing licenses. The incidence of DILI is very low, with a value between 1 and 19 per 100,000 patient years. All antidepressants may induce DILI even at low therapeutic doses. In this report, we present a case of ALI after venlafaxine administration. A 27-year-old Chinese Han woman was admitted for depression. Several serum liver function indices in this patient were abnormal after antidepressant treatment. The Roussel Uclaf Causality Assessment Method (RUCAM) causality assessment score was 8, and the R value was 31.18. The patient was diagnosed with hepatocellular ALI, which was derived from venlafaxine-related adverse events. METHODS First, all medications were stopped to block the progression of DILI. Then, a hepatoprotective strategy and proper psychological treatment were performed to recover the impaired hepatic function. RESULTS Liver function was fully recovered as indicated by liver function indices and ultrasound imaging. CONCLUSIONS The possibility of DILI should not be overlooked during the long-term use of antipsychotic drugs. In response, regular liver function monitoring should be performed in a timely manner to avoid missing diagnoses and delayed treatment. Furthermore, the necessary medical treatment needs to be conducted after the occurrence of ALI. pmc1 Introduction As a serotonin and norepinephrine reuptake inhibitor, venlafaxine (VEN) has been proven to be well tolerated with a rate of adverse drug reactions (ADRs) of less than 1/1000.[1] The most common ADRs[2,3] are mild, including nausea, vomiting, dizziness, diarrhea, dry mouth, decreased appetite, constipation, somnolence, and so on, all of which might be caused by serotonin toxicity or vulnerability to drug-drug interactions.[4] Notably, VEN-induced acute liver injury (ALI) has been less reported since its approval in 1993.[5–7] It has been confirmed that VEN can result in transient asymptomatic elevations in serum aminotransferase levels and has been linked to rare instances of clinically apparent ALI.[8] However, the mechanism by which VEN causes ALI is unknown. In this case report, we describe 1 patient with depression who was diagnosed with ALI after the administration of VEN. 2 Case presentation A 27-year-old Chinese Han woman was diagnosed with depression. She had no chronic or metabolic liver disease, blood transfusion, or history of alcohol consumption. The patient was admitted to our Department of Clinical Psychology on March 9, 2021, and low-dose VEN 50 mg (oral, qd) was started to give to her. Meanwhile, liver function was normal according to the test results (Table 1). The dosage of VEN was gradually increased to 225 mg/d on March 19 because of severe depression. The patient showed good tolerance without any symptoms. Additionally, she also received lorazepam 1 mg (oral, qn) and zolpidem tartrate 10 mg (oral, qn) for treating anxiety symptoms and dyssomnia during this period. After administering these treatments, depressive symptoms were well controlled. Table 1 The results of liver function tests and the reference values. Aminotransferase March 10 April 14 April 19 April 29 Reference value AST 13 IU/L 767 IU/L 24 IU/L 17 IU/L 5–50 IU/L ALT 12 IU/L 1777 IU/L 269 IU/L 37 IU/L 7–40 IU/L ALP 39 IU/L 171 IU/L 85.5 IU/L 57.0 IU/L 40–150 IU/L γ-GT 7 IU/L 147 IU/L 90 IU/L 48 IU/L 7–50 IU/L 5′-nucleotidease 2.9 U/L 37 U/L 12 U/L 5.3 U/L 2–11.4 U/L Before leaving the hospital, the liver function as well as her blood routine were reviewed on April 14, 2021. The results of routine blood tests were normal, while the symptoms of ALI (Table 1) were abnormal elevation of serum aspartate aminotransferase to 767 IU/L, alanine aminotransferase to 1777 IU/L, alkaline phosphatase to 171 IU/L, γ-glutamyltransferase to 147 IU/L, and 5’-nucleotidease to 37 U/L. Nevertheless, neither dilatation of the intrahepatic or extrahepatic bile ducts nor hepatomegaly or splenomegaly were detected by ultrasound of the liver (Fig. 1). These results indicate that ALI was still in its early stages. The results of Roussel Uclaf Causality Assessment Method (RUCAM) assessment (score: 8) illustrated that this ALI (hepatocellular, R = 31.18) was derived from VEN-related adverse events. Figure 1 The ultrasound of the liver on March 11, 2021 (A) and April 14, 2021 (B). There was no dilatation of the intrahepatic or extrahepatic bile ducts and hepatomegaly or splenomegaly appeared in 2 images. Based on these considerations, stopping all medications was adopted to block the progression of ALI. g of glutathione 1.08 g of magnesium isoglycyrrhizinate 150 mg (intravenous, qd) and silibinin meglumine 100 mg (oral, tid) were administered to the patient. Several indices of liver function were recovered based on the results from April 19, 2021 (Table 1). Therefore, the current hepatoprotective strategy, as well as proper psychological treatment, were held for this patient. The liver function fully recovered on April 29, 2021 (Table 1), and she was discharged from the hospital the next day. 3 Discussion ALI is a rare manifestation of VEN-induced ADRs. Despite the unknown mechanism, we attempted to reveal the pathogeny of VEN-related ALI in this case. It has been reported that the dosage of VEN-related liver function abnormalities is in the range of 25 to 300 mg/d with a median dose of 75 mg/d, and the upper limit of normal for concentration of VEN in the patients’ plasma is 400 ng/mL.[9] It is critical to note that this upper value may be exceeded in patients who receive only the conventional dosage. This individual difference increases the risk of dose-related adverse events. In view of these facts, we firstly performed the plasma drug concentration measurements and the result was only 93.9 ng/mL, which was even slightly lower than the recommended therapeutic concentration of Arbeitsgemeinschaft für Neuropsychopharmakologie und Pharmakopsychiatrie (100–400 ng/mL). Second, we attempted to determine the metabolic features of VEN in this patient. Cytochrome P450 (CYP) 2D6 in the liver is the predominant metabolic enzyme responsible for converting 90 percent of VEN to O-desmethylvenlafaxine (desvenlafaxine).[10] Additionally, CYP2C19 mediates the remaining 10% of this demethylation. Indeed, extensive CYP2D6 and CYP2C19 metabolizers always increase the clearance of VEN and maintain a low plasma concentration. To our surprise, results of genetic polymorphism testing showed that CYP2D6 was a normal metabolizer with rs1065852 CT, rs16947 CT, and rs5030865 GG, while CYP2C19 was an extensive metabolizer with rs4244285 GG, rs4986893 GG, and rs12248560 CC, which is in accordance with the low exposure of VEN. Third, the vulnerability of VEN to drug-drug interactions can also cause ADRs. In this case, the fast metabolic characteristics of VEN lead to a high abundance of cytochrome isoenzymes (mainly CYP2D6 and CYP2C19) as well as rapidly accumulated O-desmethylvenlafaxine, which are thought to be involved in VEN-associated liver injury. As a reversible profile of hepatic damage, accurate and timely liver function monitoring and the necessary measures taken after abnormalities in the liver are crucial to recover its function. Corticosteroids, such as methylprednisolone, are used as hepatic protectants against VEN-induced liver injury.[11] As another liver cell protective agent, magnesium isoglycyrate always plays a hepatoprotective effect by protecting the liver cell membrane, improving liver function, and its anti-inflammatory ability. It can inhibit the increase in serum transaminase, reduce the degeneration, necrosis, and inflammatory cell infiltration of liver cells, and hence repair the liver tissue and recover its activity.[12–15] In addition, VEN was shown to induce ROS formation to enhance the membrane permeability of mitochondria and lysosomes, triggering apoptosis or necrosis.[16] Thus, ROS scavenger agents have been identified as promising therapeutic strategies against drug/xenobitic-induced liver injuries.[17] 4 Conclusion During the administration of VEN or other antidepressants, liver function monitoring is crucial, and immediate drug withdrawal as well as necessary medical treatment should be taken in cases of ALI to obtain satisfactory post-treatment results. Furthermore, more samples and further explorations are still needed to better understand the etiology, early diagnosis, and treatment of VEN-induced ALI. We hope that our report will provide an effective reference for future research on this DILI. Author contributions Conceptualization: Lin Fang, Kun Yao. Funding acquisition: Kun Yao. Writing – original draft: Lin Fang, Shushan Wang. Writing – review & editing: Leiming Cao, Kun Yao. Abbreviations: ADRs = adverse drug reactions, ALI = acute liver injury, DILI = drug-induced liver injure, VEN = venlafaxine. How to cite this article: Fang L, Wang S, Cao L, Yao K. Early intervention of acute liver injury related to venlafaxine: a case report. Medicine. 2021;100:49(e28140). Written informed consent was obtained from the patient for the publication of this case report and accompanying images. This work was supported by Fund for development of Science and Technology of Nanjing Medical University (NMUB2020295). The authors have no conflicts of interest to disclose. The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request γ-GT = γ-glutamyltransferase, ALT = alanine aminotransferase, ALP = alkaline phosphatase, AST = aspartate aminotransferase.

LORAZEPAM, VENLAFAXINE HYDROCHLORIDE, ZOLPIDEM TARTRATE

DrugsGivenReaction

CC BY

34889278

2021-12-10

What was the outcome of reaction 'Drug interaction'?

Early intervention of acute liver injury related to venlafaxine: A case report. BACKGROUND Drug-induced liver injury (DILI) is the leading cause of acute liver injury (ALI), market withdrawal of a drug, and rejection of applications for marketing licenses. The incidence of DILI is very low, with a value between 1 and 19 per 100,000 patient years. All antidepressants may induce DILI even at low therapeutic doses. In this report, we present a case of ALI after venlafaxine administration. A 27-year-old Chinese Han woman was admitted for depression. Several serum liver function indices in this patient were abnormal after antidepressant treatment. The Roussel Uclaf Causality Assessment Method (RUCAM) causality assessment score was 8, and the R value was 31.18. The patient was diagnosed with hepatocellular ALI, which was derived from venlafaxine-related adverse events. METHODS First, all medications were stopped to block the progression of DILI. Then, a hepatoprotective strategy and proper psychological treatment were performed to recover the impaired hepatic function. RESULTS Liver function was fully recovered as indicated by liver function indices and ultrasound imaging. CONCLUSIONS The possibility of DILI should not be overlooked during the long-term use of antipsychotic drugs. In response, regular liver function monitoring should be performed in a timely manner to avoid missing diagnoses and delayed treatment. Furthermore, the necessary medical treatment needs to be conducted after the occurrence of ALI. pmc1 Introduction As a serotonin and norepinephrine reuptake inhibitor, venlafaxine (VEN) has been proven to be well tolerated with a rate of adverse drug reactions (ADRs) of less than 1/1000.[1] The most common ADRs[2,3] are mild, including nausea, vomiting, dizziness, diarrhea, dry mouth, decreased appetite, constipation, somnolence, and so on, all of which might be caused by serotonin toxicity or vulnerability to drug-drug interactions.[4] Notably, VEN-induced acute liver injury (ALI) has been less reported since its approval in 1993.[5–7] It has been confirmed that VEN can result in transient asymptomatic elevations in serum aminotransferase levels and has been linked to rare instances of clinically apparent ALI.[8] However, the mechanism by which VEN causes ALI is unknown. In this case report, we describe 1 patient with depression who was diagnosed with ALI after the administration of VEN. 2 Case presentation A 27-year-old Chinese Han woman was diagnosed with depression. She had no chronic or metabolic liver disease, blood transfusion, or history of alcohol consumption. The patient was admitted to our Department of Clinical Psychology on March 9, 2021, and low-dose VEN 50 mg (oral, qd) was started to give to her. Meanwhile, liver function was normal according to the test results (Table 1). The dosage of VEN was gradually increased to 225 mg/d on March 19 because of severe depression. The patient showed good tolerance without any symptoms. Additionally, she also received lorazepam 1 mg (oral, qn) and zolpidem tartrate 10 mg (oral, qn) for treating anxiety symptoms and dyssomnia during this period. After administering these treatments, depressive symptoms were well controlled. Table 1 The results of liver function tests and the reference values. Aminotransferase March 10 April 14 April 19 April 29 Reference value AST 13 IU/L 767 IU/L 24 IU/L 17 IU/L 5–50 IU/L ALT 12 IU/L 1777 IU/L 269 IU/L 37 IU/L 7–40 IU/L ALP 39 IU/L 171 IU/L 85.5 IU/L 57.0 IU/L 40–150 IU/L γ-GT 7 IU/L 147 IU/L 90 IU/L 48 IU/L 7–50 IU/L 5′-nucleotidease 2.9 U/L 37 U/L 12 U/L 5.3 U/L 2–11.4 U/L Before leaving the hospital, the liver function as well as her blood routine were reviewed on April 14, 2021. The results of routine blood tests were normal, while the symptoms of ALI (Table 1) were abnormal elevation of serum aspartate aminotransferase to 767 IU/L, alanine aminotransferase to 1777 IU/L, alkaline phosphatase to 171 IU/L, γ-glutamyltransferase to 147 IU/L, and 5’-nucleotidease to 37 U/L. Nevertheless, neither dilatation of the intrahepatic or extrahepatic bile ducts nor hepatomegaly or splenomegaly were detected by ultrasound of the liver (Fig. 1). These results indicate that ALI was still in its early stages. The results of Roussel Uclaf Causality Assessment Method (RUCAM) assessment (score: 8) illustrated that this ALI (hepatocellular, R = 31.18) was derived from VEN-related adverse events. Figure 1 The ultrasound of the liver on March 11, 2021 (A) and April 14, 2021 (B). There was no dilatation of the intrahepatic or extrahepatic bile ducts and hepatomegaly or splenomegaly appeared in 2 images. Based on these considerations, stopping all medications was adopted to block the progression of ALI. g of glutathione 1.08 g of magnesium isoglycyrrhizinate 150 mg (intravenous, qd) and silibinin meglumine 100 mg (oral, tid) were administered to the patient. Several indices of liver function were recovered based on the results from April 19, 2021 (Table 1). Therefore, the current hepatoprotective strategy, as well as proper psychological treatment, were held for this patient. The liver function fully recovered on April 29, 2021 (Table 1), and she was discharged from the hospital the next day. 3 Discussion ALI is a rare manifestation of VEN-induced ADRs. Despite the unknown mechanism, we attempted to reveal the pathogeny of VEN-related ALI in this case. It has been reported that the dosage of VEN-related liver function abnormalities is in the range of 25 to 300 mg/d with a median dose of 75 mg/d, and the upper limit of normal for concentration of VEN in the patients’ plasma is 400 ng/mL.[9] It is critical to note that this upper value may be exceeded in patients who receive only the conventional dosage. This individual difference increases the risk of dose-related adverse events. In view of these facts, we firstly performed the plasma drug concentration measurements and the result was only 93.9 ng/mL, which was even slightly lower than the recommended therapeutic concentration of Arbeitsgemeinschaft für Neuropsychopharmakologie und Pharmakopsychiatrie (100–400 ng/mL). Second, we attempted to determine the metabolic features of VEN in this patient. Cytochrome P450 (CYP) 2D6 in the liver is the predominant metabolic enzyme responsible for converting 90 percent of VEN to O-desmethylvenlafaxine (desvenlafaxine).[10] Additionally, CYP2C19 mediates the remaining 10% of this demethylation. Indeed, extensive CYP2D6 and CYP2C19 metabolizers always increase the clearance of VEN and maintain a low plasma concentration. To our surprise, results of genetic polymorphism testing showed that CYP2D6 was a normal metabolizer with rs1065852 CT, rs16947 CT, and rs5030865 GG, while CYP2C19 was an extensive metabolizer with rs4244285 GG, rs4986893 GG, and rs12248560 CC, which is in accordance with the low exposure of VEN. Third, the vulnerability of VEN to drug-drug interactions can also cause ADRs. In this case, the fast metabolic characteristics of VEN lead to a high abundance of cytochrome isoenzymes (mainly CYP2D6 and CYP2C19) as well as rapidly accumulated O-desmethylvenlafaxine, which are thought to be involved in VEN-associated liver injury. As a reversible profile of hepatic damage, accurate and timely liver function monitoring and the necessary measures taken after abnormalities in the liver are crucial to recover its function. Corticosteroids, such as methylprednisolone, are used as hepatic protectants against VEN-induced liver injury.[11] As another liver cell protective agent, magnesium isoglycyrate always plays a hepatoprotective effect by protecting the liver cell membrane, improving liver function, and its anti-inflammatory ability. It can inhibit the increase in serum transaminase, reduce the degeneration, necrosis, and inflammatory cell infiltration of liver cells, and hence repair the liver tissue and recover its activity.[12–15] In addition, VEN was shown to induce ROS formation to enhance the membrane permeability of mitochondria and lysosomes, triggering apoptosis or necrosis.[16] Thus, ROS scavenger agents have been identified as promising therapeutic strategies against drug/xenobitic-induced liver injuries.[17] 4 Conclusion During the administration of VEN or other antidepressants, liver function monitoring is crucial, and immediate drug withdrawal as well as necessary medical treatment should be taken in cases of ALI to obtain satisfactory post-treatment results. Furthermore, more samples and further explorations are still needed to better understand the etiology, early diagnosis, and treatment of VEN-induced ALI. We hope that our report will provide an effective reference for future research on this DILI. Author contributions Conceptualization: Lin Fang, Kun Yao. Funding acquisition: Kun Yao. Writing – original draft: Lin Fang, Shushan Wang. Writing – review & editing: Leiming Cao, Kun Yao. Abbreviations: ADRs = adverse drug reactions, ALI = acute liver injury, DILI = drug-induced liver injure, VEN = venlafaxine. How to cite this article: Fang L, Wang S, Cao L, Yao K. Early intervention of acute liver injury related to venlafaxine: a case report. Medicine. 2021;100:49(e28140). Written informed consent was obtained from the patient for the publication of this case report and accompanying images. This work was supported by Fund for development of Science and Technology of Nanjing Medical University (NMUB2020295). The authors have no conflicts of interest to disclose. The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request γ-GT = γ-glutamyltransferase, ALT = alanine aminotransferase, ALP = alkaline phosphatase, AST = aspartate aminotransferase.

Recovered

ReactionOutcome

CC BY

34889278

2021-12-10

What was the outcome of reaction 'Drug-induced liver injury'?

Early intervention of acute liver injury related to venlafaxine: A case report. BACKGROUND Drug-induced liver injury (DILI) is the leading cause of acute liver injury (ALI), market withdrawal of a drug, and rejection of applications for marketing licenses. The incidence of DILI is very low, with a value between 1 and 19 per 100,000 patient years. All antidepressants may induce DILI even at low therapeutic doses. In this report, we present a case of ALI after venlafaxine administration. A 27-year-old Chinese Han woman was admitted for depression. Several serum liver function indices in this patient were abnormal after antidepressant treatment. The Roussel Uclaf Causality Assessment Method (RUCAM) causality assessment score was 8, and the R value was 31.18. The patient was diagnosed with hepatocellular ALI, which was derived from venlafaxine-related adverse events. METHODS First, all medications were stopped to block the progression of DILI. Then, a hepatoprotective strategy and proper psychological treatment were performed to recover the impaired hepatic function. RESULTS Liver function was fully recovered as indicated by liver function indices and ultrasound imaging. CONCLUSIONS The possibility of DILI should not be overlooked during the long-term use of antipsychotic drugs. In response, regular liver function monitoring should be performed in a timely manner to avoid missing diagnoses and delayed treatment. Furthermore, the necessary medical treatment needs to be conducted after the occurrence of ALI. pmc1 Introduction As a serotonin and norepinephrine reuptake inhibitor, venlafaxine (VEN) has been proven to be well tolerated with a rate of adverse drug reactions (ADRs) of less than 1/1000.[1] The most common ADRs[2,3] are mild, including nausea, vomiting, dizziness, diarrhea, dry mouth, decreased appetite, constipation, somnolence, and so on, all of which might be caused by serotonin toxicity or vulnerability to drug-drug interactions.[4] Notably, VEN-induced acute liver injury (ALI) has been less reported since its approval in 1993.[5–7] It has been confirmed that VEN can result in transient asymptomatic elevations in serum aminotransferase levels and has been linked to rare instances of clinically apparent ALI.[8] However, the mechanism by which VEN causes ALI is unknown. In this case report, we describe 1 patient with depression who was diagnosed with ALI after the administration of VEN. 2 Case presentation A 27-year-old Chinese Han woman was diagnosed with depression. She had no chronic or metabolic liver disease, blood transfusion, or history of alcohol consumption. The patient was admitted to our Department of Clinical Psychology on March 9, 2021, and low-dose VEN 50 mg (oral, qd) was started to give to her. Meanwhile, liver function was normal according to the test results (Table 1). The dosage of VEN was gradually increased to 225 mg/d on March 19 because of severe depression. The patient showed good tolerance without any symptoms. Additionally, she also received lorazepam 1 mg (oral, qn) and zolpidem tartrate 10 mg (oral, qn) for treating anxiety symptoms and dyssomnia during this period. After administering these treatments, depressive symptoms were well controlled. Table 1 The results of liver function tests and the reference values. Aminotransferase March 10 April 14 April 19 April 29 Reference value AST 13 IU/L 767 IU/L 24 IU/L 17 IU/L 5–50 IU/L ALT 12 IU/L 1777 IU/L 269 IU/L 37 IU/L 7–40 IU/L ALP 39 IU/L 171 IU/L 85.5 IU/L 57.0 IU/L 40–150 IU/L γ-GT 7 IU/L 147 IU/L 90 IU/L 48 IU/L 7–50 IU/L 5′-nucleotidease 2.9 U/L 37 U/L 12 U/L 5.3 U/L 2–11.4 U/L Before leaving the hospital, the liver function as well as her blood routine were reviewed on April 14, 2021. The results of routine blood tests were normal, while the symptoms of ALI (Table 1) were abnormal elevation of serum aspartate aminotransferase to 767 IU/L, alanine aminotransferase to 1777 IU/L, alkaline phosphatase to 171 IU/L, γ-glutamyltransferase to 147 IU/L, and 5’-nucleotidease to 37 U/L. Nevertheless, neither dilatation of the intrahepatic or extrahepatic bile ducts nor hepatomegaly or splenomegaly were detected by ultrasound of the liver (Fig. 1). These results indicate that ALI was still in its early stages. The results of Roussel Uclaf Causality Assessment Method (RUCAM) assessment (score: 8) illustrated that this ALI (hepatocellular, R = 31.18) was derived from VEN-related adverse events. Figure 1 The ultrasound of the liver on March 11, 2021 (A) and April 14, 2021 (B). There was no dilatation of the intrahepatic or extrahepatic bile ducts and hepatomegaly or splenomegaly appeared in 2 images. Based on these considerations, stopping all medications was adopted to block the progression of ALI. g of glutathione 1.08 g of magnesium isoglycyrrhizinate 150 mg (intravenous, qd) and silibinin meglumine 100 mg (oral, tid) were administered to the patient. Several indices of liver function were recovered based on the results from April 19, 2021 (Table 1). Therefore, the current hepatoprotective strategy, as well as proper psychological treatment, were held for this patient. The liver function fully recovered on April 29, 2021 (Table 1), and she was discharged from the hospital the next day. 3 Discussion ALI is a rare manifestation of VEN-induced ADRs. Despite the unknown mechanism, we attempted to reveal the pathogeny of VEN-related ALI in this case. It has been reported that the dosage of VEN-related liver function abnormalities is in the range of 25 to 300 mg/d with a median dose of 75 mg/d, and the upper limit of normal for concentration of VEN in the patients’ plasma is 400 ng/mL.[9] It is critical to note that this upper value may be exceeded in patients who receive only the conventional dosage. This individual difference increases the risk of dose-related adverse events. In view of these facts, we firstly performed the plasma drug concentration measurements and the result was only 93.9 ng/mL, which was even slightly lower than the recommended therapeutic concentration of Arbeitsgemeinschaft für Neuropsychopharmakologie und Pharmakopsychiatrie (100–400 ng/mL). Second, we attempted to determine the metabolic features of VEN in this patient. Cytochrome P450 (CYP) 2D6 in the liver is the predominant metabolic enzyme responsible for converting 90 percent of VEN to O-desmethylvenlafaxine (desvenlafaxine).[10] Additionally, CYP2C19 mediates the remaining 10% of this demethylation. Indeed, extensive CYP2D6 and CYP2C19 metabolizers always increase the clearance of VEN and maintain a low plasma concentration. To our surprise, results of genetic polymorphism testing showed that CYP2D6 was a normal metabolizer with rs1065852 CT, rs16947 CT, and rs5030865 GG, while CYP2C19 was an extensive metabolizer with rs4244285 GG, rs4986893 GG, and rs12248560 CC, which is in accordance with the low exposure of VEN. Third, the vulnerability of VEN to drug-drug interactions can also cause ADRs. In this case, the fast metabolic characteristics of VEN lead to a high abundance of cytochrome isoenzymes (mainly CYP2D6 and CYP2C19) as well as rapidly accumulated O-desmethylvenlafaxine, which are thought to be involved in VEN-associated liver injury. As a reversible profile of hepatic damage, accurate and timely liver function monitoring and the necessary measures taken after abnormalities in the liver are crucial to recover its function. Corticosteroids, such as methylprednisolone, are used as hepatic protectants against VEN-induced liver injury.[11] As another liver cell protective agent, magnesium isoglycyrate always plays a hepatoprotective effect by protecting the liver cell membrane, improving liver function, and its anti-inflammatory ability. It can inhibit the increase in serum transaminase, reduce the degeneration, necrosis, and inflammatory cell infiltration of liver cells, and hence repair the liver tissue and recover its activity.[12–15] In addition, VEN was shown to induce ROS formation to enhance the membrane permeability of mitochondria and lysosomes, triggering apoptosis or necrosis.[16] Thus, ROS scavenger agents have been identified as promising therapeutic strategies against drug/xenobitic-induced liver injuries.[17] 4 Conclusion During the administration of VEN or other antidepressants, liver function monitoring is crucial, and immediate drug withdrawal as well as necessary medical treatment should be taken in cases of ALI to obtain satisfactory post-treatment results. Furthermore, more samples and further explorations are still needed to better understand the etiology, early diagnosis, and treatment of VEN-induced ALI. We hope that our report will provide an effective reference for future research on this DILI. Author contributions Conceptualization: Lin Fang, Kun Yao. Funding acquisition: Kun Yao. Writing – original draft: Lin Fang, Shushan Wang. Writing – review & editing: Leiming Cao, Kun Yao. Abbreviations: ADRs = adverse drug reactions, ALI = acute liver injury, DILI = drug-induced liver injure, VEN = venlafaxine. How to cite this article: Fang L, Wang S, Cao L, Yao K. Early intervention of acute liver injury related to venlafaxine: a case report. Medicine. 2021;100:49(e28140). Written informed consent was obtained from the patient for the publication of this case report and accompanying images. This work was supported by Fund for development of Science and Technology of Nanjing Medical University (NMUB2020295). The authors have no conflicts of interest to disclose. The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request γ-GT = γ-glutamyltransferase, ALT = alanine aminotransferase, ALP = alkaline phosphatase, AST = aspartate aminotransferase.

Recovered

ReactionOutcome

CC BY

34889278

2021-12-10

What was the outcome of reaction 'Hepatitis'?

Early intervention of acute liver injury related to venlafaxine: A case report. BACKGROUND Drug-induced liver injury (DILI) is the leading cause of acute liver injury (ALI), market withdrawal of a drug, and rejection of applications for marketing licenses. The incidence of DILI is very low, with a value between 1 and 19 per 100,000 patient years. All antidepressants may induce DILI even at low therapeutic doses. In this report, we present a case of ALI after venlafaxine administration. A 27-year-old Chinese Han woman was admitted for depression. Several serum liver function indices in this patient were abnormal after antidepressant treatment. The Roussel Uclaf Causality Assessment Method (RUCAM) causality assessment score was 8, and the R value was 31.18. The patient was diagnosed with hepatocellular ALI, which was derived from venlafaxine-related adverse events. METHODS First, all medications were stopped to block the progression of DILI. Then, a hepatoprotective strategy and proper psychological treatment were performed to recover the impaired hepatic function. RESULTS Liver function was fully recovered as indicated by liver function indices and ultrasound imaging. CONCLUSIONS The possibility of DILI should not be overlooked during the long-term use of antipsychotic drugs. In response, regular liver function monitoring should be performed in a timely manner to avoid missing diagnoses and delayed treatment. Furthermore, the necessary medical treatment needs to be conducted after the occurrence of ALI. pmc1 Introduction As a serotonin and norepinephrine reuptake inhibitor, venlafaxine (VEN) has been proven to be well tolerated with a rate of adverse drug reactions (ADRs) of less than 1/1000.[1] The most common ADRs[2,3] are mild, including nausea, vomiting, dizziness, diarrhea, dry mouth, decreased appetite, constipation, somnolence, and so on, all of which might be caused by serotonin toxicity or vulnerability to drug-drug interactions.[4] Notably, VEN-induced acute liver injury (ALI) has been less reported since its approval in 1993.[5–7] It has been confirmed that VEN can result in transient asymptomatic elevations in serum aminotransferase levels and has been linked to rare instances of clinically apparent ALI.[8] However, the mechanism by which VEN causes ALI is unknown. In this case report, we describe 1 patient with depression who was diagnosed with ALI after the administration of VEN. 2 Case presentation A 27-year-old Chinese Han woman was diagnosed with depression. She had no chronic or metabolic liver disease, blood transfusion, or history of alcohol consumption. The patient was admitted to our Department of Clinical Psychology on March 9, 2021, and low-dose VEN 50 mg (oral, qd) was started to give to her. Meanwhile, liver function was normal according to the test results (Table 1). The dosage of VEN was gradually increased to 225 mg/d on March 19 because of severe depression. The patient showed good tolerance without any symptoms. Additionally, she also received lorazepam 1 mg (oral, qn) and zolpidem tartrate 10 mg (oral, qn) for treating anxiety symptoms and dyssomnia during this period. After administering these treatments, depressive symptoms were well controlled. Table 1 The results of liver function tests and the reference values. Aminotransferase March 10 April 14 April 19 April 29 Reference value AST 13 IU/L 767 IU/L 24 IU/L 17 IU/L 5–50 IU/L ALT 12 IU/L 1777 IU/L 269 IU/L 37 IU/L 7–40 IU/L ALP 39 IU/L 171 IU/L 85.5 IU/L 57.0 IU/L 40–150 IU/L γ-GT 7 IU/L 147 IU/L 90 IU/L 48 IU/L 7–50 IU/L 5′-nucleotidease 2.9 U/L 37 U/L 12 U/L 5.3 U/L 2–11.4 U/L Before leaving the hospital, the liver function as well as her blood routine were reviewed on April 14, 2021. The results of routine blood tests were normal, while the symptoms of ALI (Table 1) were abnormal elevation of serum aspartate aminotransferase to 767 IU/L, alanine aminotransferase to 1777 IU/L, alkaline phosphatase to 171 IU/L, γ-glutamyltransferase to 147 IU/L, and 5’-nucleotidease to 37 U/L. Nevertheless, neither dilatation of the intrahepatic or extrahepatic bile ducts nor hepatomegaly or splenomegaly were detected by ultrasound of the liver (Fig. 1). These results indicate that ALI was still in its early stages. The results of Roussel Uclaf Causality Assessment Method (RUCAM) assessment (score: 8) illustrated that this ALI (hepatocellular, R = 31.18) was derived from VEN-related adverse events. Figure 1 The ultrasound of the liver on March 11, 2021 (A) and April 14, 2021 (B). There was no dilatation of the intrahepatic or extrahepatic bile ducts and hepatomegaly or splenomegaly appeared in 2 images. Based on these considerations, stopping all medications was adopted to block the progression of ALI. g of glutathione 1.08 g of magnesium isoglycyrrhizinate 150 mg (intravenous, qd) and silibinin meglumine 100 mg (oral, tid) were administered to the patient. Several indices of liver function were recovered based on the results from April 19, 2021 (Table 1). Therefore, the current hepatoprotective strategy, as well as proper psychological treatment, were held for this patient. The liver function fully recovered on April 29, 2021 (Table 1), and she was discharged from the hospital the next day. 3 Discussion ALI is a rare manifestation of VEN-induced ADRs. Despite the unknown mechanism, we attempted to reveal the pathogeny of VEN-related ALI in this case. It has been reported that the dosage of VEN-related liver function abnormalities is in the range of 25 to 300 mg/d with a median dose of 75 mg/d, and the upper limit of normal for concentration of VEN in the patients’ plasma is 400 ng/mL.[9] It is critical to note that this upper value may be exceeded in patients who receive only the conventional dosage. This individual difference increases the risk of dose-related adverse events. In view of these facts, we firstly performed the plasma drug concentration measurements and the result was only 93.9 ng/mL, which was even slightly lower than the recommended therapeutic concentration of Arbeitsgemeinschaft für Neuropsychopharmakologie und Pharmakopsychiatrie (100–400 ng/mL). Second, we attempted to determine the metabolic features of VEN in this patient. Cytochrome P450 (CYP) 2D6 in the liver is the predominant metabolic enzyme responsible for converting 90 percent of VEN to O-desmethylvenlafaxine (desvenlafaxine).[10] Additionally, CYP2C19 mediates the remaining 10% of this demethylation. Indeed, extensive CYP2D6 and CYP2C19 metabolizers always increase the clearance of VEN and maintain a low plasma concentration. To our surprise, results of genetic polymorphism testing showed that CYP2D6 was a normal metabolizer with rs1065852 CT, rs16947 CT, and rs5030865 GG, while CYP2C19 was an extensive metabolizer with rs4244285 GG, rs4986893 GG, and rs12248560 CC, which is in accordance with the low exposure of VEN. Third, the vulnerability of VEN to drug-drug interactions can also cause ADRs. In this case, the fast metabolic characteristics of VEN lead to a high abundance of cytochrome isoenzymes (mainly CYP2D6 and CYP2C19) as well as rapidly accumulated O-desmethylvenlafaxine, which are thought to be involved in VEN-associated liver injury. As a reversible profile of hepatic damage, accurate and timely liver function monitoring and the necessary measures taken after abnormalities in the liver are crucial to recover its function. Corticosteroids, such as methylprednisolone, are used as hepatic protectants against VEN-induced liver injury.[11] As another liver cell protective agent, magnesium isoglycyrate always plays a hepatoprotective effect by protecting the liver cell membrane, improving liver function, and its anti-inflammatory ability. It can inhibit the increase in serum transaminase, reduce the degeneration, necrosis, and inflammatory cell infiltration of liver cells, and hence repair the liver tissue and recover its activity.[12–15] In addition, VEN was shown to induce ROS formation to enhance the membrane permeability of mitochondria and lysosomes, triggering apoptosis or necrosis.[16] Thus, ROS scavenger agents have been identified as promising therapeutic strategies against drug/xenobitic-induced liver injuries.[17] 4 Conclusion During the administration of VEN or other antidepressants, liver function monitoring is crucial, and immediate drug withdrawal as well as necessary medical treatment should be taken in cases of ALI to obtain satisfactory post-treatment results. Furthermore, more samples and further explorations are still needed to better understand the etiology, early diagnosis, and treatment of VEN-induced ALI. We hope that our report will provide an effective reference for future research on this DILI. Author contributions Conceptualization: Lin Fang, Kun Yao. Funding acquisition: Kun Yao. Writing – original draft: Lin Fang, Shushan Wang. Writing – review & editing: Leiming Cao, Kun Yao. Abbreviations: ADRs = adverse drug reactions, ALI = acute liver injury, DILI = drug-induced liver injure, VEN = venlafaxine. How to cite this article: Fang L, Wang S, Cao L, Yao K. Early intervention of acute liver injury related to venlafaxine: a case report. Medicine. 2021;100:49(e28140). Written informed consent was obtained from the patient for the publication of this case report and accompanying images. This work was supported by Fund for development of Science and Technology of Nanjing Medical University (NMUB2020295). The authors have no conflicts of interest to disclose. The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request γ-GT = γ-glutamyltransferase, ALT = alanine aminotransferase, ALP = alkaline phosphatase, AST = aspartate aminotransferase.

Recovered

ReactionOutcome

CC BY

34889278

2021-12-10

What was the outcome of reaction 'Liver injury'?

Early intervention of acute liver injury related to venlafaxine: A case report. BACKGROUND Drug-induced liver injury (DILI) is the leading cause of acute liver injury (ALI), market withdrawal of a drug, and rejection of applications for marketing licenses. The incidence of DILI is very low, with a value between 1 and 19 per 100,000 patient years. All antidepressants may induce DILI even at low therapeutic doses. In this report, we present a case of ALI after venlafaxine administration. A 27-year-old Chinese Han woman was admitted for depression. Several serum liver function indices in this patient were abnormal after antidepressant treatment. The Roussel Uclaf Causality Assessment Method (RUCAM) causality assessment score was 8, and the R value was 31.18. The patient was diagnosed with hepatocellular ALI, which was derived from venlafaxine-related adverse events. METHODS First, all medications were stopped to block the progression of DILI. Then, a hepatoprotective strategy and proper psychological treatment were performed to recover the impaired hepatic function. RESULTS Liver function was fully recovered as indicated by liver function indices and ultrasound imaging. CONCLUSIONS The possibility of DILI should not be overlooked during the long-term use of antipsychotic drugs. In response, regular liver function monitoring should be performed in a timely manner to avoid missing diagnoses and delayed treatment. Furthermore, the necessary medical treatment needs to be conducted after the occurrence of ALI. pmc1 Introduction As a serotonin and norepinephrine reuptake inhibitor, venlafaxine (VEN) has been proven to be well tolerated with a rate of adverse drug reactions (ADRs) of less than 1/1000.[1] The most common ADRs[2,3] are mild, including nausea, vomiting, dizziness, diarrhea, dry mouth, decreased appetite, constipation, somnolence, and so on, all of which might be caused by serotonin toxicity or vulnerability to drug-drug interactions.[4] Notably, VEN-induced acute liver injury (ALI) has been less reported since its approval in 1993.[5–7] It has been confirmed that VEN can result in transient asymptomatic elevations in serum aminotransferase levels and has been linked to rare instances of clinically apparent ALI.[8] However, the mechanism by which VEN causes ALI is unknown. In this case report, we describe 1 patient with depression who was diagnosed with ALI after the administration of VEN. 2 Case presentation A 27-year-old Chinese Han woman was diagnosed with depression. She had no chronic or metabolic liver disease, blood transfusion, or history of alcohol consumption. The patient was admitted to our Department of Clinical Psychology on March 9, 2021, and low-dose VEN 50 mg (oral, qd) was started to give to her. Meanwhile, liver function was normal according to the test results (Table 1). The dosage of VEN was gradually increased to 225 mg/d on March 19 because of severe depression. The patient showed good tolerance without any symptoms. Additionally, she also received lorazepam 1 mg (oral, qn) and zolpidem tartrate 10 mg (oral, qn) for treating anxiety symptoms and dyssomnia during this period. After administering these treatments, depressive symptoms were well controlled. Table 1 The results of liver function tests and the reference values. Aminotransferase March 10 April 14 April 19 April 29 Reference value AST 13 IU/L 767 IU/L 24 IU/L 17 IU/L 5–50 IU/L ALT 12 IU/L 1777 IU/L 269 IU/L 37 IU/L 7–40 IU/L ALP 39 IU/L 171 IU/L 85.5 IU/L 57.0 IU/L 40–150 IU/L γ-GT 7 IU/L 147 IU/L 90 IU/L 48 IU/L 7–50 IU/L 5′-nucleotidease 2.9 U/L 37 U/L 12 U/L 5.3 U/L 2–11.4 U/L Before leaving the hospital, the liver function as well as her blood routine were reviewed on April 14, 2021. The results of routine blood tests were normal, while the symptoms of ALI (Table 1) were abnormal elevation of serum aspartate aminotransferase to 767 IU/L, alanine aminotransferase to 1777 IU/L, alkaline phosphatase to 171 IU/L, γ-glutamyltransferase to 147 IU/L, and 5’-nucleotidease to 37 U/L. Nevertheless, neither dilatation of the intrahepatic or extrahepatic bile ducts nor hepatomegaly or splenomegaly were detected by ultrasound of the liver (Fig. 1). These results indicate that ALI was still in its early stages. The results of Roussel Uclaf Causality Assessment Method (RUCAM) assessment (score: 8) illustrated that this ALI (hepatocellular, R = 31.18) was derived from VEN-related adverse events. Figure 1 The ultrasound of the liver on March 11, 2021 (A) and April 14, 2021 (B). There was no dilatation of the intrahepatic or extrahepatic bile ducts and hepatomegaly or splenomegaly appeared in 2 images. Based on these considerations, stopping all medications was adopted to block the progression of ALI. g of glutathione 1.08 g of magnesium isoglycyrrhizinate 150 mg (intravenous, qd) and silibinin meglumine 100 mg (oral, tid) were administered to the patient. Several indices of liver function were recovered based on the results from April 19, 2021 (Table 1). Therefore, the current hepatoprotective strategy, as well as proper psychological treatment, were held for this patient. The liver function fully recovered on April 29, 2021 (Table 1), and she was discharged from the hospital the next day. 3 Discussion ALI is a rare manifestation of VEN-induced ADRs. Despite the unknown mechanism, we attempted to reveal the pathogeny of VEN-related ALI in this case. It has been reported that the dosage of VEN-related liver function abnormalities is in the range of 25 to 300 mg/d with a median dose of 75 mg/d, and the upper limit of normal for concentration of VEN in the patients’ plasma is 400 ng/mL.[9] It is critical to note that this upper value may be exceeded in patients who receive only the conventional dosage. This individual difference increases the risk of dose-related adverse events. In view of these facts, we firstly performed the plasma drug concentration measurements and the result was only 93.9 ng/mL, which was even slightly lower than the recommended therapeutic concentration of Arbeitsgemeinschaft für Neuropsychopharmakologie und Pharmakopsychiatrie (100–400 ng/mL). Second, we attempted to determine the metabolic features of VEN in this patient. Cytochrome P450 (CYP) 2D6 in the liver is the predominant metabolic enzyme responsible for converting 90 percent of VEN to O-desmethylvenlafaxine (desvenlafaxine).[10] Additionally, CYP2C19 mediates the remaining 10% of this demethylation. Indeed, extensive CYP2D6 and CYP2C19 metabolizers always increase the clearance of VEN and maintain a low plasma concentration. To our surprise, results of genetic polymorphism testing showed that CYP2D6 was a normal metabolizer with rs1065852 CT, rs16947 CT, and rs5030865 GG, while CYP2C19 was an extensive metabolizer with rs4244285 GG, rs4986893 GG, and rs12248560 CC, which is in accordance with the low exposure of VEN. Third, the vulnerability of VEN to drug-drug interactions can also cause ADRs. In this case, the fast metabolic characteristics of VEN lead to a high abundance of cytochrome isoenzymes (mainly CYP2D6 and CYP2C19) as well as rapidly accumulated O-desmethylvenlafaxine, which are thought to be involved in VEN-associated liver injury. As a reversible profile of hepatic damage, accurate and timely liver function monitoring and the necessary measures taken after abnormalities in the liver are crucial to recover its function. Corticosteroids, such as methylprednisolone, are used as hepatic protectants against VEN-induced liver injury.[11] As another liver cell protective agent, magnesium isoglycyrate always plays a hepatoprotective effect by protecting the liver cell membrane, improving liver function, and its anti-inflammatory ability. It can inhibit the increase in serum transaminase, reduce the degeneration, necrosis, and inflammatory cell infiltration of liver cells, and hence repair the liver tissue and recover its activity.[12–15] In addition, VEN was shown to induce ROS formation to enhance the membrane permeability of mitochondria and lysosomes, triggering apoptosis or necrosis.[16] Thus, ROS scavenger agents have been identified as promising therapeutic strategies against drug/xenobitic-induced liver injuries.[17] 4 Conclusion During the administration of VEN or other antidepressants, liver function monitoring is crucial, and immediate drug withdrawal as well as necessary medical treatment should be taken in cases of ALI to obtain satisfactory post-treatment results. Furthermore, more samples and further explorations are still needed to better understand the etiology, early diagnosis, and treatment of VEN-induced ALI. We hope that our report will provide an effective reference for future research on this DILI. Author contributions Conceptualization: Lin Fang, Kun Yao. Funding acquisition: Kun Yao. Writing – original draft: Lin Fang, Shushan Wang. Writing – review & editing: Leiming Cao, Kun Yao. Abbreviations: ADRs = adverse drug reactions, ALI = acute liver injury, DILI = drug-induced liver injure, VEN = venlafaxine. How to cite this article: Fang L, Wang S, Cao L, Yao K. Early intervention of acute liver injury related to venlafaxine: a case report. Medicine. 2021;100:49(e28140). Written informed consent was obtained from the patient for the publication of this case report and accompanying images. This work was supported by Fund for development of Science and Technology of Nanjing Medical University (NMUB2020295). The authors have no conflicts of interest to disclose. The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request γ-GT = γ-glutamyltransferase, ALT = alanine aminotransferase, ALP = alkaline phosphatase, AST = aspartate aminotransferase.

Recovered

ReactionOutcome

CC BY

34889278

2021-12-10

What was the administration route of drug 'NIVOLUMAB'?

Development of bullous pemphigoid following radiation therapy combined with nivolumab for renal cell carcinoma: A case report of abscopal toxicities. BACKGROUND Concern for immune-related adverse events from immunotherapy and radiation therapy are well-documented; however, side effects are mostly mild to moderate. However, high-grade, potentially life-threatening adverse events are increasing. While case reports regarding immunotherapy-related bullous pemphigoid (BP) have been rising, only 1 has described BP following concomitant use of both nivolumab and radiation therapy (RT). For that patient, nivolumab was used for 10 weeks prior to RT and development of PB followed 7 weeks later. This case presents a patient who tolerated nivolumab well for 38 months prior to developing BP less than 2 weeks after completing RT. We present the case of DH, a 67-year-old gentleman on nivolumab for metastatic renal cell carcinoma to the lung since May of 2017. Following progressing lung nodules, the patient had his nivolumab paused and completed a course of short-beam radiation therapy. After restarting nivolumab post-radiation, the patient presented with itchy rash and blisters on his arm, legs, and trunk. METHODS DH consulted dermatology following development of rash and was diagnosed with bullous dermatosis, likely bullous pemphigoid. Bullous pemphigoid following concomitant nivolumab (OPDIVO), despite prior tolerance and no history of autoimmune disease, was confirmed by biopsy a month later. METHODS Initial treatment was betamethasone 0.05% cream mixed 1:1 with powder to form paste applied twice daily. Given progressive symptoms and confirmatory biopsy of BP, nivolumab was held and 100 mg doxycycline and 80 mg prednisone daily was prescribed for a week, reduced to 60 mg during the second week. RESULTS A week following discontinuation of nivolumab and beginning of doxycycline and prednisone, the blistering and rash was almost entirely resolved. Four months later, nivolumab was restarted and the patient continued low-dose tapering of prednisone until December. Since completing prednisone, the patient has shown no recurrence of bullous pemphigoid and has not developed any other immune-related adverse events to nivolumab upon rechallenge. Follow-up through October 2021 demonstrates the patient's sites of disease, both in- and out-field, have remained responsive to treatment. CONCLUSIONS Treating physicians should be aware of off-target effects of radiotherapy for oligoprogressive disease, which may include abscopal toxicities and the development of new immune-related adverse effects. pmc1 Introduction Immune checkpoint inhibitors such as anti-programmed death-1/programmed death ligand-1 have emerged as an effective therapy for a variety of advanced malignancies. However, immune-related adverse events (irAE) are commonly reported and highly dependent on agent, dose, and exposure time.[1] Adding another dimension to these irAEs is the concomitant use of radiation therapy (RT). While RT has been traditionally viewed to be immunosuppressive, radiation-induced activation of the immune system has been increasingly recognized.[2] In this regard, these abscopal effects have been leveraged as a strategy to amplify the host's anti-tumor immune response during treatment. While the exact underlying mechanism of this abscopal effect remains unclear, it is speculated that it is the enhanced tumor antigen presentation and associated improved anti-tumor immune education afforded by RT that is responsible for the abscopal effect. Despite these positive abscopal effects in reducing cancer burden, the concern for additional irAEs remains. Cutaneous side effects have been noted following RT, but are often limited to radiation dermatitis with associated erythema, pruritis, and occasionally, focal desquamation.[2] Similarly, the use of immunotherapies have also been correlated with dermatologic irAEs. While most of these side effects are mild to moderate in severity, some manifestations can progress to high-grade and potentially life-threatening situations - such as bullous pemphigoid.[3,4] Since 2015, more than 40 cases of immunotherapy-related BP have been reported, involving either the skin or mucous membrane. Of these case reports, only 1 described the development of BP following concomitant use of both nivolumab and RT.[5] The increasing prevalence of these conditions not only raise concern for more serious toxicities, but also suggests the need for further investigation to elucidate possible mechanisms, risk factors, and management strategies for patients undergoing multi-modal treatment. In this report, we explore the case of a 67-year-old gentleman who developed BP shortly following RT to a lung metastasis, despite reporting no irAEs on nivolumab for 38 months of prior treatment. 2 Case presentation We present the case of a 67-year-old gentleman with no history of autoimmune disease, was diagnosed with clear cell renal cell carcinoma (RCC) stage T3b, Fuhrman grade 4/4, and underwent a right radical nephrectomy, right adrenalectomy, and vena cava tumor thrombectomy in November 2016. After a follow-up computed tomography in March 2017 demonstrated multiple lung nodules, the patient underwent a left upper lobe wedge resection in April 2017 which confirmed metastatic renal cell carcinoma to lung. Nivolumab 240 mg in sodium chloride 0.9% 100 mL infusion every 2 weeks was started in May of 2017 progressing to 480 mg nivolumab every 4 weeks on May 1, 2018. The patient reported no irAEs with nivolumab for the following 38 months. In April 2020, computed tomography demonstrated oligoprogressive disease with increase in size of 2 lung nodules, but otherwise stable disease. The patient was referred to radiation oncology and was treated with stereotactic body radiation therapy 5000 cGy in 5 fractions to both progressive lesions. Maximum dose to skin was 1481 cGy with less than 5 cc of skin receiving 1000 cGy or more (200 cGy per fraction). The associated dose to skin was minimal with only 5 cc of skin receiving 1000 cGy or more. Radiation was well-tolerated and completed on May 22, 2020. After completion of short-beam radiation therapy, nivolumab 480 mg in sodium chloride 0.9% 100 mL infusion was continued on schedule with the next infusion provided on May 26, 2020. Following this, the patient developed diffuse, mildly pruritic skin lesions with blisters and presented to dermatology 1 week later with multiple erythematous bullae on the trunk, as well as the upper and lower extremities. Some bullae coalesced into crusted plaques and others were hyperpigmented patches. At this presentation, less than 10% of the patient's body showed involvement characteristic of bullous dermatosis and he was appropriately characterized to have mild to moderate cutaneous toxicity. A topic steroid (betamethasone 0.05% cream) was prescribed to manage the rash and nivolumab was continued accordingly. Following the next 480 mg nivolumab in sodium chloride 0.9% 100 mL infusion 1 month later, the patient's bullous dermatosis worsened, with progressive involvement of the upper and lower extremities. The patient presented with progressive foot involvement and linear erosions of the face. Additionally, the patient had developed edema of the lower extremities resulting in a follow-up with dermatology on July 6 and punch biopsy on July 14. Sections of the biopsy demonstrated complete subepidermal detachment of the epidermis from the underlying dermis (Fig. 1). Within the papillary dermis there was a moderate band like inflammatory infiltrate composed of lymphocytes, a prominent number of eosinophils and rare neutrophils (Fig. 2). Direct immunofluorescence staining was performed on a separate skin biopsy and demonstrated linear staining for C3 and immunoglobulin-G along the basement membrane with no intercellular staining of the keratinocytes (Fig. 3). The histologic features in conjunction with the immunofluorescence staining supported a diagnosis of bullous pemphigoid. Figure 1 Subepidermal detachment of epidermis with band like inflammation within the papillary dermis (H&E, 40×). Figure 2 Inflammation within the papillary dermis consists of lymphocytes, large numbers of eosinophils and rare neutrophils (H&E, 200×). Figure 3 Linear staining for C3 along the dermal-epidermal junction. At the area of blister formation there is linear staining along both the roof and floor of the blister cavity. Similar staining was seen with IgG (direct immunofluorescence, 200×). IgG = immunoglobulin-G. Given the progressive symptoms despite topical corticosteroids and the biopsy proven development of bullous pemphigoid (BP), Nivolumab was held and the patient began treatment with 100 mg doxycycline and 80 mg prednisone daily for a week, reduced to 60 mg during the second week. On follow-up, the patient showed significant improvement and over the next 10 weeks was tapered down to 20 mg of prednisone. Four months following the initiation of treatment of BP, 480 mg nivolumab in sodium chloride 0.9% 146 mL infusion was restarted and the patient continued low-dose tapering of prednisone until December. Since completing the prednisone course, the patient has shown no recurrence of bullous pemphigoid and has not developed any other irAEs to nivolumab upon rechallenge. Follow-up through October of 2021 demonstrates the patient's sites of disease, both in- and out-field, have remained responsive to treatment. 3 Discussion While cutaneous side effects are common and well-documented within the context of immunotherapy[6,7] and radiation therapy alone,[8] they are generally mild, localized, and do not result in life-threatening conditions (only 0.3%-1.3%).[1] Nguyen et al[8] published 29 cases of BP following RT monotherapy. Of the 29 patients included, 84% had received RT for breast cancer and BP was localized to irradiated sites in 25 patients.[7] This is in stark contrast to the present case, wherein our patient developed a systemic BP less than 2 weeks following the addition of stereotactic radiation therapy to his previously well-tolerated regimen of nivolumab. Tanita et al[5] published a similar case on a patient with acral lentiginous melanoma who was treated with nivolumab and then intensity-modulated RT 10 weeks following nivolumab that also developed bullous pemphigoid. However, this case differs in several notable ways. First, their patient was treated with dacarbazine with interferon-B for 6 months prior to starting nivolumab. Second, the patient only demonstrated toleration of nivolumab treatment for 17 weeks compared to our patient's 38 months. Finally, the timeline differed as the development of bullous pemphigoid occurred 7 weeks after intensity-modulated RT irradiation compared to less than 2 weeks after short-beam radiation therapy irradiation in this present case.[5] In our presented case, the temporal relationship between nivolumab dosing, radiotherapy, and development of BP is most consistent with a possible abscopal toxicity from radiotherapy. While the mechanism by which immunotherapy and/or RT induces BP is currently unclear, it is reminiscent of the abscopal effect in RT.[2] RT is traditionally considered immunosuppressive as it has direct and indirect cytotoxic effects via deoxyribonucleic acid damage and the generation of free radicals respectively, on irradiated immune cells. However, the abscopal effect suggests that RT may have immunostimulating properties as well. In this theory, RT is thought to create an antitumor immune response in the tumor microenvironment by inducing a release of anti-tumor cytokines and chemokines. These changes in the tumor microenvironment leads to chemoattraction of dendritic cells and cytotoxic T lymphocytes. This upregulation of the immune system response, along with radiation-induced susceptibility of tumor cells, thus results in shrinkage of tumors distant to the site of initial radiation. When combined with new immunotherapies and biologics, this synergistic effect has improved survival outcomes for various types of advanced carcinomas.[2] Pouget et al[9] suggests that radiation exposure results in systemic inflammation and an overactive immune response as the result of a chronic stress state. This is characterized by the production of damage-associated molecular patterns, which contribute to the upregulation of proinflammatory cytokines and an overactive systemic response. In this regard, it is ostensible that the RT-induced anti-tumor inflammatory state can also have negative consequences, as the balance between a controlled vs a pathologic immunogenic response is altered. This cascade has been correlated with several immune-related off-target effects such as radiation-related thyroid autoimmunity, diabetes mellitus, gastritis,[10] and in this case, bullous pemphigoid. Further compounding this issue was the use of nivolumab in our patient. As nivolumab blocks programmed death-1/programmed death ligand-1, the addition of RT may have contributed to an overactive systemic response, leading to the development of off-target effects.[8] Understanding immune-mediated adverse events is vital for providing adequate care to patients receiving concomitant RT and immunotherapies. While radiotherapy to oligoprogressive disease for patients on immunotherapy remains an attractive approach, treating physicians should be aware of the off-target effects, which may include the development of new immune-related adverse effects. Therefore, further studies are required to elucidate the mechanism by which RT and immunotherapy can induce BP and other negative abscopal effects. Author contributions All authors participated in final approval of the version to be published and agree to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. LMH: drafting the work; data acquisition and analysis. BTB: drafting the work; data acquisition and analysis. DJD: interpretation of data; revising work for intellectual content. MJB: conception and design of work; interpretation of data; revising work for intellectual content. BAT: conception and design of work; interpretation of data; revising work for intellectual content. Conceptualization: Michael J. Baine, Benjamin A. Teply. Investigation: Linda My Huynh, Benjamin T. Bonebrake, Michael J. Baine, Benjamin A. Teply. Methodology: Dominick J. DiMaio, Michael J. Baine, Benjamin A. Teply. Project administration: Michael J. Baine, Benjamin A. Teply. Resources: Dominick J. DiMaio, Michael J. Baine, Benjamin A. Teply. Supervision: Michael J. Baine, Benjamin A. Teply. Writing – original draft: Linda My Huynh, Benjamin T. Bonebrake, Dominick J. DiMaio, Michael J. Baine, Benjamin A. Teply. Writing – review & editing: Linda My Huynh, Benjamin T. Bonebrake, Dominick J. DiMaio, Michael J. Baine, Benjamin A. Teply. Abbreviations: BP = bullous pemphigoid, irAEs = immune-related adverse events, RT = radiation therapy. How to cite this article: Huynh LM, Bonebrake BT, DiMaio DJ, Baine MJ, Teply BA. Development of bullous pemphigoid following radiation therapy combined with nivolumab for renal cell carcinoma: a case report of abscopal toxicities. Medicine. 2021;100:49(e28199). LMH and BTB contributed equally to this work. Written informed consent was obtained from the patient for publication of the case details and accompanying images. The authors have no funding and conflicts of interest to disclose. Data sharing not applicable to this article as no datasets were generated or analyzed during the current study.

Intravenous (not otherwise specified)

DrugAdministrationRoute

CC BY

34889301

2021-12-10

What was the dosage of drug 'BETAMETHASONE DIPROPIONATE'?

Development of bullous pemphigoid following radiation therapy combined with nivolumab for renal cell carcinoma: A case report of abscopal toxicities. BACKGROUND Concern for immune-related adverse events from immunotherapy and radiation therapy are well-documented; however, side effects are mostly mild to moderate. However, high-grade, potentially life-threatening adverse events are increasing. While case reports regarding immunotherapy-related bullous pemphigoid (BP) have been rising, only 1 has described BP following concomitant use of both nivolumab and radiation therapy (RT). For that patient, nivolumab was used for 10 weeks prior to RT and development of PB followed 7 weeks later. This case presents a patient who tolerated nivolumab well for 38 months prior to developing BP less than 2 weeks after completing RT. We present the case of DH, a 67-year-old gentleman on nivolumab for metastatic renal cell carcinoma to the lung since May of 2017. Following progressing lung nodules, the patient had his nivolumab paused and completed a course of short-beam radiation therapy. After restarting nivolumab post-radiation, the patient presented with itchy rash and blisters on his arm, legs, and trunk. METHODS DH consulted dermatology following development of rash and was diagnosed with bullous dermatosis, likely bullous pemphigoid. Bullous pemphigoid following concomitant nivolumab (OPDIVO), despite prior tolerance and no history of autoimmune disease, was confirmed by biopsy a month later. METHODS Initial treatment was betamethasone 0.05% cream mixed 1:1 with powder to form paste applied twice daily. Given progressive symptoms and confirmatory biopsy of BP, nivolumab was held and 100 mg doxycycline and 80 mg prednisone daily was prescribed for a week, reduced to 60 mg during the second week. RESULTS A week following discontinuation of nivolumab and beginning of doxycycline and prednisone, the blistering and rash was almost entirely resolved. Four months later, nivolumab was restarted and the patient continued low-dose tapering of prednisone until December. Since completing prednisone, the patient has shown no recurrence of bullous pemphigoid and has not developed any other immune-related adverse events to nivolumab upon rechallenge. Follow-up through October 2021 demonstrates the patient's sites of disease, both in- and out-field, have remained responsive to treatment. CONCLUSIONS Treating physicians should be aware of off-target effects of radiotherapy for oligoprogressive disease, which may include abscopal toxicities and the development of new immune-related adverse effects. pmc1 Introduction Immune checkpoint inhibitors such as anti-programmed death-1/programmed death ligand-1 have emerged as an effective therapy for a variety of advanced malignancies. However, immune-related adverse events (irAE) are commonly reported and highly dependent on agent, dose, and exposure time.[1] Adding another dimension to these irAEs is the concomitant use of radiation therapy (RT). While RT has been traditionally viewed to be immunosuppressive, radiation-induced activation of the immune system has been increasingly recognized.[2] In this regard, these abscopal effects have been leveraged as a strategy to amplify the host's anti-tumor immune response during treatment. While the exact underlying mechanism of this abscopal effect remains unclear, it is speculated that it is the enhanced tumor antigen presentation and associated improved anti-tumor immune education afforded by RT that is responsible for the abscopal effect. Despite these positive abscopal effects in reducing cancer burden, the concern for additional irAEs remains. Cutaneous side effects have been noted following RT, but are often limited to radiation dermatitis with associated erythema, pruritis, and occasionally, focal desquamation.[2] Similarly, the use of immunotherapies have also been correlated with dermatologic irAEs. While most of these side effects are mild to moderate in severity, some manifestations can progress to high-grade and potentially life-threatening situations - such as bullous pemphigoid.[3,4] Since 2015, more than 40 cases of immunotherapy-related BP have been reported, involving either the skin or mucous membrane. Of these case reports, only 1 described the development of BP following concomitant use of both nivolumab and RT.[5] The increasing prevalence of these conditions not only raise concern for more serious toxicities, but also suggests the need for further investigation to elucidate possible mechanisms, risk factors, and management strategies for patients undergoing multi-modal treatment. In this report, we explore the case of a 67-year-old gentleman who developed BP shortly following RT to a lung metastasis, despite reporting no irAEs on nivolumab for 38 months of prior treatment. 2 Case presentation We present the case of a 67-year-old gentleman with no history of autoimmune disease, was diagnosed with clear cell renal cell carcinoma (RCC) stage T3b, Fuhrman grade 4/4, and underwent a right radical nephrectomy, right adrenalectomy, and vena cava tumor thrombectomy in November 2016. After a follow-up computed tomography in March 2017 demonstrated multiple lung nodules, the patient underwent a left upper lobe wedge resection in April 2017 which confirmed metastatic renal cell carcinoma to lung. Nivolumab 240 mg in sodium chloride 0.9% 100 mL infusion every 2 weeks was started in May of 2017 progressing to 480 mg nivolumab every 4 weeks on May 1, 2018. The patient reported no irAEs with nivolumab for the following 38 months. In April 2020, computed tomography demonstrated oligoprogressive disease with increase in size of 2 lung nodules, but otherwise stable disease. The patient was referred to radiation oncology and was treated with stereotactic body radiation therapy 5000 cGy in 5 fractions to both progressive lesions. Maximum dose to skin was 1481 cGy with less than 5 cc of skin receiving 1000 cGy or more (200 cGy per fraction). The associated dose to skin was minimal with only 5 cc of skin receiving 1000 cGy or more. Radiation was well-tolerated and completed on May 22, 2020. After completion of short-beam radiation therapy, nivolumab 480 mg in sodium chloride 0.9% 100 mL infusion was continued on schedule with the next infusion provided on May 26, 2020. Following this, the patient developed diffuse, mildly pruritic skin lesions with blisters and presented to dermatology 1 week later with multiple erythematous bullae on the trunk, as well as the upper and lower extremities. Some bullae coalesced into crusted plaques and others were hyperpigmented patches. At this presentation, less than 10% of the patient's body showed involvement characteristic of bullous dermatosis and he was appropriately characterized to have mild to moderate cutaneous toxicity. A topic steroid (betamethasone 0.05% cream) was prescribed to manage the rash and nivolumab was continued accordingly. Following the next 480 mg nivolumab in sodium chloride 0.9% 100 mL infusion 1 month later, the patient's bullous dermatosis worsened, with progressive involvement of the upper and lower extremities. The patient presented with progressive foot involvement and linear erosions of the face. Additionally, the patient had developed edema of the lower extremities resulting in a follow-up with dermatology on July 6 and punch biopsy on July 14. Sections of the biopsy demonstrated complete subepidermal detachment of the epidermis from the underlying dermis (Fig. 1). Within the papillary dermis there was a moderate band like inflammatory infiltrate composed of lymphocytes, a prominent number of eosinophils and rare neutrophils (Fig. 2). Direct immunofluorescence staining was performed on a separate skin biopsy and demonstrated linear staining for C3 and immunoglobulin-G along the basement membrane with no intercellular staining of the keratinocytes (Fig. 3). The histologic features in conjunction with the immunofluorescence staining supported a diagnosis of bullous pemphigoid. Figure 1 Subepidermal detachment of epidermis with band like inflammation within the papillary dermis (H&E, 40×). Figure 2 Inflammation within the papillary dermis consists of lymphocytes, large numbers of eosinophils and rare neutrophils (H&E, 200×). Figure 3 Linear staining for C3 along the dermal-epidermal junction. At the area of blister formation there is linear staining along both the roof and floor of the blister cavity. Similar staining was seen with IgG (direct immunofluorescence, 200×). IgG = immunoglobulin-G. Given the progressive symptoms despite topical corticosteroids and the biopsy proven development of bullous pemphigoid (BP), Nivolumab was held and the patient began treatment with 100 mg doxycycline and 80 mg prednisone daily for a week, reduced to 60 mg during the second week. On follow-up, the patient showed significant improvement and over the next 10 weeks was tapered down to 20 mg of prednisone. Four months following the initiation of treatment of BP, 480 mg nivolumab in sodium chloride 0.9% 146 mL infusion was restarted and the patient continued low-dose tapering of prednisone until December. Since completing the prednisone course, the patient has shown no recurrence of bullous pemphigoid and has not developed any other irAEs to nivolumab upon rechallenge. Follow-up through October of 2021 demonstrates the patient's sites of disease, both in- and out-field, have remained responsive to treatment. 3 Discussion While cutaneous side effects are common and well-documented within the context of immunotherapy[6,7] and radiation therapy alone,[8] they are generally mild, localized, and do not result in life-threatening conditions (only 0.3%-1.3%).[1] Nguyen et al[8] published 29 cases of BP following RT monotherapy. Of the 29 patients included, 84% had received RT for breast cancer and BP was localized to irradiated sites in 25 patients.[7] This is in stark contrast to the present case, wherein our patient developed a systemic BP less than 2 weeks following the addition of stereotactic radiation therapy to his previously well-tolerated regimen of nivolumab. Tanita et al[5] published a similar case on a patient with acral lentiginous melanoma who was treated with nivolumab and then intensity-modulated RT 10 weeks following nivolumab that also developed bullous pemphigoid. However, this case differs in several notable ways. First, their patient was treated with dacarbazine with interferon-B for 6 months prior to starting nivolumab. Second, the patient only demonstrated toleration of nivolumab treatment for 17 weeks compared to our patient's 38 months. Finally, the timeline differed as the development of bullous pemphigoid occurred 7 weeks after intensity-modulated RT irradiation compared to less than 2 weeks after short-beam radiation therapy irradiation in this present case.[5] In our presented case, the temporal relationship between nivolumab dosing, radiotherapy, and development of BP is most consistent with a possible abscopal toxicity from radiotherapy. While the mechanism by which immunotherapy and/or RT induces BP is currently unclear, it is reminiscent of the abscopal effect in RT.[2] RT is traditionally considered immunosuppressive as it has direct and indirect cytotoxic effects via deoxyribonucleic acid damage and the generation of free radicals respectively, on irradiated immune cells. However, the abscopal effect suggests that RT may have immunostimulating properties as well. In this theory, RT is thought to create an antitumor immune response in the tumor microenvironment by inducing a release of anti-tumor cytokines and chemokines. These changes in the tumor microenvironment leads to chemoattraction of dendritic cells and cytotoxic T lymphocytes. This upregulation of the immune system response, along with radiation-induced susceptibility of tumor cells, thus results in shrinkage of tumors distant to the site of initial radiation. When combined with new immunotherapies and biologics, this synergistic effect has improved survival outcomes for various types of advanced carcinomas.[2] Pouget et al[9] suggests that radiation exposure results in systemic inflammation and an overactive immune response as the result of a chronic stress state. This is characterized by the production of damage-associated molecular patterns, which contribute to the upregulation of proinflammatory cytokines and an overactive systemic response. In this regard, it is ostensible that the RT-induced anti-tumor inflammatory state can also have negative consequences, as the balance between a controlled vs a pathologic immunogenic response is altered. This cascade has been correlated with several immune-related off-target effects such as radiation-related thyroid autoimmunity, diabetes mellitus, gastritis,[10] and in this case, bullous pemphigoid. Further compounding this issue was the use of nivolumab in our patient. As nivolumab blocks programmed death-1/programmed death ligand-1, the addition of RT may have contributed to an overactive systemic response, leading to the development of off-target effects.[8] Understanding immune-mediated adverse events is vital for providing adequate care to patients receiving concomitant RT and immunotherapies. While radiotherapy to oligoprogressive disease for patients on immunotherapy remains an attractive approach, treating physicians should be aware of the off-target effects, which may include the development of new immune-related adverse effects. Therefore, further studies are required to elucidate the mechanism by which RT and immunotherapy can induce BP and other negative abscopal effects. Author contributions All authors participated in final approval of the version to be published and agree to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. LMH: drafting the work; data acquisition and analysis. BTB: drafting the work; data acquisition and analysis. DJD: interpretation of data; revising work for intellectual content. MJB: conception and design of work; interpretation of data; revising work for intellectual content. BAT: conception and design of work; interpretation of data; revising work for intellectual content. Conceptualization: Michael J. Baine, Benjamin A. Teply. Investigation: Linda My Huynh, Benjamin T. Bonebrake, Michael J. Baine, Benjamin A. Teply. Methodology: Dominick J. DiMaio, Michael J. Baine, Benjamin A. Teply. Project administration: Michael J. Baine, Benjamin A. Teply. Resources: Dominick J. DiMaio, Michael J. Baine, Benjamin A. Teply. Supervision: Michael J. Baine, Benjamin A. Teply. Writing – original draft: Linda My Huynh, Benjamin T. Bonebrake, Dominick J. DiMaio, Michael J. Baine, Benjamin A. Teply. Writing – review & editing: Linda My Huynh, Benjamin T. Bonebrake, Dominick J. DiMaio, Michael J. Baine, Benjamin A. Teply. Abbreviations: BP = bullous pemphigoid, irAEs = immune-related adverse events, RT = radiation therapy. How to cite this article: Huynh LM, Bonebrake BT, DiMaio DJ, Baine MJ, Teply BA. Development of bullous pemphigoid following radiation therapy combined with nivolumab for renal cell carcinoma: a case report of abscopal toxicities. Medicine. 2021;100:49(e28199). LMH and BTB contributed equally to this work. Written informed consent was obtained from the patient for publication of the case details and accompanying images. The authors have no funding and conflicts of interest to disclose. Data sharing not applicable to this article as no datasets were generated or analyzed during the current study.

0.05 UNK

DrugDosageText

CC BY

34889301

2021-12-10

What was the dosage of drug 'SODIUM CHLORIDE'?

Development of bullous pemphigoid following radiation therapy combined with nivolumab for renal cell carcinoma: A case report of abscopal toxicities. BACKGROUND Concern for immune-related adverse events from immunotherapy and radiation therapy are well-documented; however, side effects are mostly mild to moderate. However, high-grade, potentially life-threatening adverse events are increasing. While case reports regarding immunotherapy-related bullous pemphigoid (BP) have been rising, only 1 has described BP following concomitant use of both nivolumab and radiation therapy (RT). For that patient, nivolumab was used for 10 weeks prior to RT and development of PB followed 7 weeks later. This case presents a patient who tolerated nivolumab well for 38 months prior to developing BP less than 2 weeks after completing RT. We present the case of DH, a 67-year-old gentleman on nivolumab for metastatic renal cell carcinoma to the lung since May of 2017. Following progressing lung nodules, the patient had his nivolumab paused and completed a course of short-beam radiation therapy. After restarting nivolumab post-radiation, the patient presented with itchy rash and blisters on his arm, legs, and trunk. METHODS DH consulted dermatology following development of rash and was diagnosed with bullous dermatosis, likely bullous pemphigoid. Bullous pemphigoid following concomitant nivolumab (OPDIVO), despite prior tolerance and no history of autoimmune disease, was confirmed by biopsy a month later. METHODS Initial treatment was betamethasone 0.05% cream mixed 1:1 with powder to form paste applied twice daily. Given progressive symptoms and confirmatory biopsy of BP, nivolumab was held and 100 mg doxycycline and 80 mg prednisone daily was prescribed for a week, reduced to 60 mg during the second week. RESULTS A week following discontinuation of nivolumab and beginning of doxycycline and prednisone, the blistering and rash was almost entirely resolved. Four months later, nivolumab was restarted and the patient continued low-dose tapering of prednisone until December. Since completing prednisone, the patient has shown no recurrence of bullous pemphigoid and has not developed any other immune-related adverse events to nivolumab upon rechallenge. Follow-up through October 2021 demonstrates the patient's sites of disease, both in- and out-field, have remained responsive to treatment. CONCLUSIONS Treating physicians should be aware of off-target effects of radiotherapy for oligoprogressive disease, which may include abscopal toxicities and the development of new immune-related adverse effects. pmc1 Introduction Immune checkpoint inhibitors such as anti-programmed death-1/programmed death ligand-1 have emerged as an effective therapy for a variety of advanced malignancies. However, immune-related adverse events (irAE) are commonly reported and highly dependent on agent, dose, and exposure time.[1] Adding another dimension to these irAEs is the concomitant use of radiation therapy (RT). While RT has been traditionally viewed to be immunosuppressive, radiation-induced activation of the immune system has been increasingly recognized.[2] In this regard, these abscopal effects have been leveraged as a strategy to amplify the host's anti-tumor immune response during treatment. While the exact underlying mechanism of this abscopal effect remains unclear, it is speculated that it is the enhanced tumor antigen presentation and associated improved anti-tumor immune education afforded by RT that is responsible for the abscopal effect. Despite these positive abscopal effects in reducing cancer burden, the concern for additional irAEs remains. Cutaneous side effects have been noted following RT, but are often limited to radiation dermatitis with associated erythema, pruritis, and occasionally, focal desquamation.[2] Similarly, the use of immunotherapies have also been correlated with dermatologic irAEs. While most of these side effects are mild to moderate in severity, some manifestations can progress to high-grade and potentially life-threatening situations - such as bullous pemphigoid.[3,4] Since 2015, more than 40 cases of immunotherapy-related BP have been reported, involving either the skin or mucous membrane. Of these case reports, only 1 described the development of BP following concomitant use of both nivolumab and RT.[5] The increasing prevalence of these conditions not only raise concern for more serious toxicities, but also suggests the need for further investigation to elucidate possible mechanisms, risk factors, and management strategies for patients undergoing multi-modal treatment. In this report, we explore the case of a 67-year-old gentleman who developed BP shortly following RT to a lung metastasis, despite reporting no irAEs on nivolumab for 38 months of prior treatment. 2 Case presentation We present the case of a 67-year-old gentleman with no history of autoimmune disease, was diagnosed with clear cell renal cell carcinoma (RCC) stage T3b, Fuhrman grade 4/4, and underwent a right radical nephrectomy, right adrenalectomy, and vena cava tumor thrombectomy in November 2016. After a follow-up computed tomography in March 2017 demonstrated multiple lung nodules, the patient underwent a left upper lobe wedge resection in April 2017 which confirmed metastatic renal cell carcinoma to lung. Nivolumab 240 mg in sodium chloride 0.9% 100 mL infusion every 2 weeks was started in May of 2017 progressing to 480 mg nivolumab every 4 weeks on May 1, 2018. The patient reported no irAEs with nivolumab for the following 38 months. In April 2020, computed tomography demonstrated oligoprogressive disease with increase in size of 2 lung nodules, but otherwise stable disease. The patient was referred to radiation oncology and was treated with stereotactic body radiation therapy 5000 cGy in 5 fractions to both progressive lesions. Maximum dose to skin was 1481 cGy with less than 5 cc of skin receiving 1000 cGy or more (200 cGy per fraction). The associated dose to skin was minimal with only 5 cc of skin receiving 1000 cGy or more. Radiation was well-tolerated and completed on May 22, 2020. After completion of short-beam radiation therapy, nivolumab 480 mg in sodium chloride 0.9% 100 mL infusion was continued on schedule with the next infusion provided on May 26, 2020. Following this, the patient developed diffuse, mildly pruritic skin lesions with blisters and presented to dermatology 1 week later with multiple erythematous bullae on the trunk, as well as the upper and lower extremities. Some bullae coalesced into crusted plaques and others were hyperpigmented patches. At this presentation, less than 10% of the patient's body showed involvement characteristic of bullous dermatosis and he was appropriately characterized to have mild to moderate cutaneous toxicity. A topic steroid (betamethasone 0.05% cream) was prescribed to manage the rash and nivolumab was continued accordingly. Following the next 480 mg nivolumab in sodium chloride 0.9% 100 mL infusion 1 month later, the patient's bullous dermatosis worsened, with progressive involvement of the upper and lower extremities. The patient presented with progressive foot involvement and linear erosions of the face. Additionally, the patient had developed edema of the lower extremities resulting in a follow-up with dermatology on July 6 and punch biopsy on July 14. Sections of the biopsy demonstrated complete subepidermal detachment of the epidermis from the underlying dermis (Fig. 1). Within the papillary dermis there was a moderate band like inflammatory infiltrate composed of lymphocytes, a prominent number of eosinophils and rare neutrophils (Fig. 2). Direct immunofluorescence staining was performed on a separate skin biopsy and demonstrated linear staining for C3 and immunoglobulin-G along the basement membrane with no intercellular staining of the keratinocytes (Fig. 3). The histologic features in conjunction with the immunofluorescence staining supported a diagnosis of bullous pemphigoid. Figure 1 Subepidermal detachment of epidermis with band like inflammation within the papillary dermis (H&E, 40×). Figure 2 Inflammation within the papillary dermis consists of lymphocytes, large numbers of eosinophils and rare neutrophils (H&E, 200×). Figure 3 Linear staining for C3 along the dermal-epidermal junction. At the area of blister formation there is linear staining along both the roof and floor of the blister cavity. Similar staining was seen with IgG (direct immunofluorescence, 200×). IgG = immunoglobulin-G. Given the progressive symptoms despite topical corticosteroids and the biopsy proven development of bullous pemphigoid (BP), Nivolumab was held and the patient began treatment with 100 mg doxycycline and 80 mg prednisone daily for a week, reduced to 60 mg during the second week. On follow-up, the patient showed significant improvement and over the next 10 weeks was tapered down to 20 mg of prednisone. Four months following the initiation of treatment of BP, 480 mg nivolumab in sodium chloride 0.9% 146 mL infusion was restarted and the patient continued low-dose tapering of prednisone until December. Since completing the prednisone course, the patient has shown no recurrence of bullous pemphigoid and has not developed any other irAEs to nivolumab upon rechallenge. Follow-up through October of 2021 demonstrates the patient's sites of disease, both in- and out-field, have remained responsive to treatment. 3 Discussion While cutaneous side effects are common and well-documented within the context of immunotherapy[6,7] and radiation therapy alone,[8] they are generally mild, localized, and do not result in life-threatening conditions (only 0.3%-1.3%).[1] Nguyen et al[8] published 29 cases of BP following RT monotherapy. Of the 29 patients included, 84% had received RT for breast cancer and BP was localized to irradiated sites in 25 patients.[7] This is in stark contrast to the present case, wherein our patient developed a systemic BP less than 2 weeks following the addition of stereotactic radiation therapy to his previously well-tolerated regimen of nivolumab. Tanita et al[5] published a similar case on a patient with acral lentiginous melanoma who was treated with nivolumab and then intensity-modulated RT 10 weeks following nivolumab that also developed bullous pemphigoid. However, this case differs in several notable ways. First, their patient was treated with dacarbazine with interferon-B for 6 months prior to starting nivolumab. Second, the patient only demonstrated toleration of nivolumab treatment for 17 weeks compared to our patient's 38 months. Finally, the timeline differed as the development of bullous pemphigoid occurred 7 weeks after intensity-modulated RT irradiation compared to less than 2 weeks after short-beam radiation therapy irradiation in this present case.[5] In our presented case, the temporal relationship between nivolumab dosing, radiotherapy, and development of BP is most consistent with a possible abscopal toxicity from radiotherapy. While the mechanism by which immunotherapy and/or RT induces BP is currently unclear, it is reminiscent of the abscopal effect in RT.[2] RT is traditionally considered immunosuppressive as it has direct and indirect cytotoxic effects via deoxyribonucleic acid damage and the generation of free radicals respectively, on irradiated immune cells. However, the abscopal effect suggests that RT may have immunostimulating properties as well. In this theory, RT is thought to create an antitumor immune response in the tumor microenvironment by inducing a release of anti-tumor cytokines and chemokines. These changes in the tumor microenvironment leads to chemoattraction of dendritic cells and cytotoxic T lymphocytes. This upregulation of the immune system response, along with radiation-induced susceptibility of tumor cells, thus results in shrinkage of tumors distant to the site of initial radiation. When combined with new immunotherapies and biologics, this synergistic effect has improved survival outcomes for various types of advanced carcinomas.[2] Pouget et al[9] suggests that radiation exposure results in systemic inflammation and an overactive immune response as the result of a chronic stress state. This is characterized by the production of damage-associated molecular patterns, which contribute to the upregulation of proinflammatory cytokines and an overactive systemic response. In this regard, it is ostensible that the RT-induced anti-tumor inflammatory state can also have negative consequences, as the balance between a controlled vs a pathologic immunogenic response is altered. This cascade has been correlated with several immune-related off-target effects such as radiation-related thyroid autoimmunity, diabetes mellitus, gastritis,[10] and in this case, bullous pemphigoid. Further compounding this issue was the use of nivolumab in our patient. As nivolumab blocks programmed death-1/programmed death ligand-1, the addition of RT may have contributed to an overactive systemic response, leading to the development of off-target effects.[8] Understanding immune-mediated adverse events is vital for providing adequate care to patients receiving concomitant RT and immunotherapies. While radiotherapy to oligoprogressive disease for patients on immunotherapy remains an attractive approach, treating physicians should be aware of the off-target effects, which may include the development of new immune-related adverse effects. Therefore, further studies are required to elucidate the mechanism by which RT and immunotherapy can induce BP and other negative abscopal effects. Author contributions All authors participated in final approval of the version to be published and agree to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. LMH: drafting the work; data acquisition and analysis. BTB: drafting the work; data acquisition and analysis. DJD: interpretation of data; revising work for intellectual content. MJB: conception and design of work; interpretation of data; revising work for intellectual content. BAT: conception and design of work; interpretation of data; revising work for intellectual content. Conceptualization: Michael J. Baine, Benjamin A. Teply. Investigation: Linda My Huynh, Benjamin T. Bonebrake, Michael J. Baine, Benjamin A. Teply. Methodology: Dominick J. DiMaio, Michael J. Baine, Benjamin A. Teply. Project administration: Michael J. Baine, Benjamin A. Teply. Resources: Dominick J. DiMaio, Michael J. Baine, Benjamin A. Teply. Supervision: Michael J. Baine, Benjamin A. Teply. Writing – original draft: Linda My Huynh, Benjamin T. Bonebrake, Dominick J. DiMaio, Michael J. Baine, Benjamin A. Teply. Writing – review & editing: Linda My Huynh, Benjamin T. Bonebrake, Dominick J. DiMaio, Michael J. Baine, Benjamin A. Teply. Abbreviations: BP = bullous pemphigoid, irAEs = immune-related adverse events, RT = radiation therapy. How to cite this article: Huynh LM, Bonebrake BT, DiMaio DJ, Baine MJ, Teply BA. Development of bullous pemphigoid following radiation therapy combined with nivolumab for renal cell carcinoma: a case report of abscopal toxicities. Medicine. 2021;100:49(e28199). LMH and BTB contributed equally to this work. Written informed consent was obtained from the patient for publication of the case details and accompanying images. The authors have no funding and conflicts of interest to disclose. Data sharing not applicable to this article as no datasets were generated or analyzed during the current study.

0.9 PERCENT 100 ML

DrugDosageText

CC BY

34889301

2021-12-10

Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Compression fracture'.

A Nationwide Study of GATA2 Deficiency in Norway-the Majority of Patients Have Undergone Allo-HSCT. OBJECTIVE GATA2 deficiency is a rare primary immunodeficiency that has become increasingly recognized due to improved molecular diagnostics and clinical awareness. The only cure for GATA2 deficiency is allogeneic hematopoietic stem cell transplantation (allo-HSCT). The inconsistency of genotype-phenotype correlations makes the decision regarding "who and when" to transplant challenging. Despite considerable morbidity and mortality, the reported proportion of patients with GATA2 deficiency that has undergone allo-HSCT is low (~ 35%). The purpose of this study was to explore if detailed clinical, genetic, and bone marrow characteristics could predict end-point outcome, i.e., death and allo-HSCT. METHODS All medical genetics departments in Norway were contacted to identify GATA2 deficient individuals. Clinical information, genetic variants, treatment, and outcome were subsequently retrieved from the patients' medical records. RESULTS Between 2013 and 2020, we identified 10 index cases or probands, four additional symptomatic patients, and no asymptomatic patients with germline GATA2 variants. These patients had a diverse clinical phenotype dominated by cytopenia (13/14), myeloid neoplasia (10/14), warts (8/14), and hearing loss (7/14). No valid genotype-phenotype correlations were found in our data set, and the phenotypes varied also within families. We found that 11/14 patients (79%), with known GATA2 deficiency, had already undergone allo-HSCT. In addition, one patient is awaiting allo-HSCT. The indications to perform allo-HSCT were myeloid neoplasia, disseminated viral infection, severe obliterating bronchiolitis, and/or HPV-associated in situ carcinoma. Two patients died, 8 months and 7 years after allo-HSCT, respectively. CONCLUSIONS Our main conclusion is that the majority of patients with symptomatic GATA2 deficiency will need allo-HSCT, and a close surveillance of these patients is important to find the "optimal window" for allo-HSCT. We advocate a more offensive approach to allo-HSCT than previously described. pmcIntroduction GATA2 deficiency is a rare primary immunodeficiency (PID), first described in 2011[1–3] that has become gradually more recognized due to improved molecular diagnostics and increased clinical awareness. GATA2, as a “master” transcription factor, plays a critical role in hematopoietic development[4]. Through cooperative processes that include other transcription factors, it controls the transition from hemogenic endothelium to hematopoietic stem cells and is required for survival and self-renewal of these cells[5]. GATA2 is also important for other tissue-forming stem cells, e.g., in the inner ear[6]. The heterozygous variants causing GATA2 deficiency are located both in coding, non-coding and enhancer regions[7]. The disease-causing loss-of-function variants can be localized across the gene. These variants can lead to defective DNA-binding capacity of the transcription factor and may cause disease through haploinsufficiency of the functional protein[5, 8]. Missense variants within the zink finger 2 (ZNF2) domain are the most frequent germline disease-causing GATA2 variants [9]. It has been estimated that approximately 1/3 of the patients have an autosomal dominant inherited disease-causing variant[10], whereas the remaining have a de novo GATA2 variant[7]. Of note, somatic variants in GATA2 are known to be drivers of myeloid neoplasia in adults. Such variants are diverse, may cause gain-of-function effects, and be located across the whole gene. This includes missense variants in the zink finger 1 (ZNF1) domain, which has not been observed in constitutional GATA2 deficiency[8]. Typically, GATA2 deficiency becomes clinically apparent in late childhood to early adulthood. The phenotype is heterogeneous, without any clear genotype–phenotype correlation, and with an incomplete clinical penetrance[11]. Symptoms may include recurrent or severe infections, warts, cytopenia (including monocytopenia), lymphedema, alveolar proteinosis, and malignant myeloid disease[9]. Infectious complications in GATA2 deficiency are likely due to deficiency of monocytes, NK cells, and B-lymphocytes as well as defective innate immune responses, including impaired type I interferon production[12]. This leads to both increased susceptibility to viral infections (e.g., human papilloma virus [HPV, warts] and herpes virus infections), non-tuberculous mycobacteria, and to more common bacterial respiratory infections. Hearing loss is a common clinical feature of GATA2 deficiency and is related to the critical role of GATA2 in vestibular morphogenesis of semicircular ducts and generation of the perilymphatic space around the inner ear’s semicircular canals[6, 13]. A substantial proportion of patients develop immunodeficiency, myelodysplastic syndrome (MDS), or acute myelogenous leukemia (AML) as initial manifestation[9, 14]. GATA2 deficiency is considered the most common hereditary predisposition to pediatric MDS[15]. The only cure for GATA2 deficiency is allogeneic hematopoietic stem cell transplantation (allo-HSCT) and results are encouraging[16–20]. However, the main challenge is deciding who and when to transplant due to the complexity and inconsistency of phenotype-genotype correlation in GATA2 deficiency[9]. To further elucidate this important issue, we present detailed clinical and molecular characteristics, treatment, and outcome of 14 Norwegian patients with germline GATA2 variants diagnosed between 2013 and 2020. The main aim of our study was to explore if detailed clinical, genetic, and bone marrow (BM) characteristics could predict end-point outcome such as death and allo-HSCT in patients with GATA2 deficiency. Methods Identification of Patients and Clinical Characteristics The first aim of this study was to obtain a complete overview of all patients with known GATA2 deficiency in Norway. For this purpose, a network of clinical immunologists, hematologists, pediatricians, and geneticists at Oslo University Hospital (OUH) collected clinical and laboratory data on patients with GATA2 deficiency at their institution. In addition, all medical genetics departments in Norway were contacted to identify any additional GATA2 deficient individuals. Clinical information, genetic variants, treatment, and outcome were subsequently retrieved from the patients’ medical records. Patients were enrolled into the study at OUH where most of the data was obtained, while supplemental data from Patient 3 was collected at the University Hospital of North Norway, Tromsø. Informed Consent Five of the patients had previously consented to be part of a genetic PID research project approved by the regional ethical committee (REC. 2014/1270–1), three patients were diagnosed with MDS in childhood and consented to be registered into the EWOG-MDS-2006 study (2015/1651/REC Nord), and all adult patients who underwent allo-HSCT had consented to publication of data (REC 11,909). In addition, due to the detailed clinical information published herein, all adult living patients signed an additional written informed consent for publication of their data and was given the opportunity to review the manuscript prior to publication. For children < 18 years, consent was given by their parents. This is in line with the recommendation given by the Ethical Constitutional board at OUH. Genetic Analyses Whole exome sequencing (WES) with in silico filtering for genes causing primary immunodeficiency disorders was performed in the probands and affected relatives as part of a routine laboratory service (Patients 2, 3, 8, 9, 10, 12, and 14) or on a research basis (Patients 4, 5, 6, and 7) as previously described (Supplemental methods)[21]. Patient 1 had severe cytopenia (Table 1), and the first attempt to extract DNA from peripheral blood was not successful. A skin biopsy was therefore performed to extract DNA from fibroblasts. In parallel, peripheral blood (from puncture of the fingertip) was applied directly to a Guthrie filter card, and by using multiple filter card punches, enough DNA was extracted to run next-generation sequencing (NGS) with an amplicon-based targeted panel for constitutional variants in PID genes (Supplemental methods). By using this rapid amplicon-based method, the molecular result was available within 3 working days[22]. DNA later extracted from fibroblasts confirmed the GATA2 variant by Sanger sequencing. Also, for Patient 13, who had advanced MDS with pancytopenia, the NGS results were available within 3 working days, with parental testing performed in parallel to evaluate as fast as possible the availability of a healthy unaffected matched related donor (MRD).Table 1 Clinical characteristics and outcome in patients with GATA2 deficiency Patient no Family Sex Current age Age at onset of symptoms/age at genetic diagnosis Infections Hearing loss Hematologic abnormalities Autoimmunity/immune dysregulation Miscellaneous HSCT, age Outcome Viral Bacterial 1 A (father of P2 and P3) M 44y 5y/41y HPV: warts HSV: disseminated disease Ear infections as a child Yes Hypoplastic BM: cytopenia, trilinear hypoplasia No No 41y Alive 2 A (son of P1) M 16y 7y/14y HPV: warts No No MDS-EB-1 No No 16y Alive 3 A (daughter of P1) F 13y 8y/9y HPV: warts No No No No No ND Alive 4a B (monozygotic twin to P5) F 45y 21y/38y HPV: warts, carcinoma in situ EBV: prolonged viremia Recurrent respiratory inf Suppurative skin infection after BCG * Yes No Progressive obliterating bronchiolitis, lupus-like syndrome Miscarriage 39y Alive 5a B (monozygotic twin to P4) F † (39y) 24y/38y HPV: warts, carcinoma VZV and EBV: prolonged viremia Recurrent respiratory inf Suppurative skin inf. after BCG* Yes MDS-MLD (hypoplastic) Progressive obliterating bronchiolitis, lupus-like syndrome DVT × 2 Squamous cell carcinoma in the cervix, rectum, and anus 39y Deceased 8 m post-HSCTb 6a C M 31y 11y/26y No Recurrent respiratory inf Yes MDS-MLD (hypoplastic) - Fever of unknown origin, recurrent pneumothorax 29y Alive 7a D F 23y 6y/17y HPV: warts Recurrent respiratory inf No MDS-MLD (hypoplastic) Interstitial lung disease Lymphedema, acne, rosacea, rash, fatigue 22y Alive 8 E F 56y 0y/53y No No No Hypoplastic BM No Lymphedema, premature graying ND Alive 9 F F 24y 15y/23y No No No AML with MDS-related changes Erythema nodosum DVT, PE, juvenile myoclonic epilepsy, epicanthic fold 23y Alive 10 G (sibling to P11) F 32y 6y/31y HPV: warts, cervix dysplasia Recurrent respiratory inf Yes MDS-MLD No Aneurysm of small vessels, hidradenitis suppurative, liver lesions: focal nodular hyperplasia 32y Alive 11 G (sibling to P10) M † (34y) 22y/PM No Recurrent skin and respiratory inf No MDS-MLD No Acne, rosacea, necrotizing fasciitis, pilonidal cysts, skin infections, ulcerations 27y Deceased 7y post-HSCTc 12 H F 19y 14y/14y No No Yesd MDS-RCC (hypoplastic) BPD/Asthma Born premature (week 26 + 5), BPD 14y Alive 13 I M 13y 9y/11 y HPV: warts No No MDS-EB1 Asthma Chronic skin abscesses, congenital ptosis 11y Alive 14 J F 31 23y/31y No No Yes MDS-SLD (hypoplastic) No Born prematurely (week 25), cerebral palsy, congenital hip dysplasia Planned Alive Abbreviations: AML, acute myeloid leukemia; BCG, bacille Calmette Geurin; BM, bone marrow; BPD, bronchopulmonary dysplasia, CT, computer tomography, HPV, human papilloma virus; HSCT, hematopoietic stem cell transplantation; Inf., infection; m, months; MDS, myelodysplastic syndrome; MDS-EB1, MDS with excess of blasts type 1; MDS-MLD, MDS with multilineage dysplasia; MDS-SLD, MDS with single lineage dysplasia; MDS-RCC, MDS subtype refractory cytopenia of childhood; ND, not done, PM, post mortem; VTE, venous thromboembolism; y, years, †; deceased aThese patients have previously been published in Stray-Pedersen, Sorte et al. 2016 (Patient 4 was 84.1, Patient 5 was 84.4, Patient 6 was 88.1, and Patient 7 was 86.1)[21] bThe patient was doing well after HSCT, but died unexpectedly of a cerebral hemorrhage cThe patient underwent lung transplantation for chronic lung GVHD 58 months after HSCT, and died of chronic lung rejection 26 months after bilateral lung transplantation dThe patient has reduced hearing, but this was confirmed after HSCT. Her hearing loss may be due to the disease-causing GATA2 variant, but may also be secondary to complications of HSCT therapy, e.g., aminoglycosides The molecular diagnosis in Patient 11 was confirmed post mortem using a BM sample collected prior to allo-HSCT (Table 2). Methods for testing for somatic occurring sequence variants on DNA extracted from whole blood or BM, and methods for testing chromosomal aberration on BM cells are described in Supplemental methods.Table 2 Constitutional and acquired genetic findings in patients with GATA2 deficiency Patient no Hematological abnormalities Constitutional heterozygous variants in GATA2, NM_001145661.1, predicted protein effect, domain, occurrence, novelty, and reference Somatic variants, predicted protein effect, VAF in BM/blood (prior to HSCT) Karyotype in BM (closest to HSCT)d  + 8  − 7 1 Hypoplastic bone marrow c.1062_1064del, p.Thr358del, in-frame exon 6, ZNF2, likely de novo, novel variant Unknown 46,XY[25/25] No No 2 MDS-EB1 c.1062_1064del, p.Thr358del, in-frame exon 6, ZNF2, paternal inherited, novel variant Acquired germline donor variant in GATA2 Post-HSCTb: c.1215G > T, p.(Lys405Asn) missense exon 7, outside and distal to ZNF2 domain, VAF: 49,5% BM NM_001145661.1 (GATA2): c.1168_1170del, p.(Lys390del), in-frame exon 7, in ZNF2, VAF: 40.2% BM NM_006758.2(U2AF1): c.470A>G, p.(Gln157Arg) VAF: 44,0% BM 46,XY,-7 + 8[20/20] Yes Yes 3 No c.1062_1064del, p.Thr358del, in-frame exon 6, ZNF2, paternal inherited, novel variant Unknown Unknown N.a N.a 4 a No c.1143 + 5G > A, p.Asn381fs*20, splice defect intron 6, ZNF2, de novo, novel, reported by usc)[21] Unknown Unknown N.a N.a 5 a MDS-MLD (hypoplastic) c.1143 + 5G > A, p.Asn381fs*20, splice defect intron 6, ZNF2, de novo, novel, reported by usc[21] None 46,XX[18/18] No No 6a MDS-MLD (hypoplastic) c.1078 T > A, p.Trp360Arg, missense exon 6, ZNF2, de novo, variant previously reported by others[23] Unknown 46,XY[25/25], but FISH MYC(8q24): + 8 in 14/303 metaphases Yes No 7a MDS-MLD (hypoplastic) c.1061C > T, p.Thr354Met, missense exon 6, ZNF2, de novo, but a recurrent GATA2 variant[21, 24, 25] NM_001042749.2(STAG2): c.2534-2A > G, predicted splice variant with loss of acceptor site, Chr.X, VAF: 11.7% blood 47,XX, + 8[4/10]/46,XX[6/10] Yes No 8 Hypoplastic bone marrow c.1017 + 1G > T, loss of donor splice site, splice defect intron 5, ZNF1, both parents deceased and not tested, novel variant Unknown 46,XX[25/25] No No 9 AML with MDS-related changes c.163C > T, p.Gln55*, nonsense exon 3, TAD domain, likely de novo (see pedigree), novel variant VAF:48.7% in BM, 49,4% in buccal swap NM_001754.4(RUNX1): c.593A > G,p.(Asp198Gly) VAF:15% BM NM_156039.3(CSF3R): c.2326C > T, p.(Gln776*) VAF: 12.5% BM NM_032458.2(PHF6): c.309C > G, p.(Tyr103Ter) VAF:12.0% BM NM_033632.3(FBXW7): c.1513C > T, p.(Arg505Cys), VAF:11.7% BM 46,XX, der(1;7)(q10;p10), + 1[11/20]/46,XX [9/20] No No 10 MDS-MLD c.1084C > T, p.Arg362*, nonsense exon 6, ZNF2, likely inherited, variant previously reported by others[15, 26–28] NM_001123385.1(BCOR): c.529_530del, p.(Ser177ProfsTer8), VAF: 23.0% BM 49∼50,XX, + 6, + 8, + 21? + 21[cp7/8]/46,XX[1/8] Yes No 11 MDS-MLD c.1084C > T, p.Arg362*, nonsense exon 6, ZNF2, likely inherited, variant previously reported by others[15, 26–28] NM_015338.5(ASXL1): c.2324 T > G, p.(Leu775Ter) VAF: 20.5% BM NM_001042749.1(STAG2): c.2990 T > A, p.(Leu997Ter) VAF: 9.6% BM 47∼48,XY, + 8[10/15],der(16)t(1;16)(q21;q24[10/15], + der(16)t(1;16)[1/15], + 21[6][cp11/15]/46,XY[3/15] Trisomy 8, evolving to unbalanced 1;16 translocation and later Trisomy 21 Yes No 12 MDS-RCC c.1098_1100delGGA, p.Asp367del, in-frame exon 6, ZNF2, de novo, novel variant None 46,XX,-7, + 8[15/20] Yes Yes 13 MDS-EB1 c.1021_1024insGCCG, p.Ala342Glyfs*43, frameshift exon 6, ZNF1 de novo, variant previously reported[29] NM_015338.5(ASXL1): c.1854dupA, p.(Ala619SerfsTer16), VAF:17.0%, BM NM_015559.2 (SETBP1): c.2612 T > C, p.(Ile871Thr), VAF: 16.3%, BM 45,XY,-7[12/12] No Yes 14 MDS-SLD (hypoplastic) c.1114G > A, p.(Ala372Thr), missense exon 6, ZNF2, variant previously reported [14] NM_001042749.1(STAG2): c.707del; p.(Asn236IlefsTer20) VAF: 5.1%, BM 46,XX[25/25] No No Abbreviations: AML, acute myeloid leukemia; BM, bone marrow; Chr, chromosome; HSCT, hematopoietic stem cell transplantation; MDS, myelodysplastic syndrome; MDS-EB1, MDS with excess blasts type 1; MDS-MLD, MDS with multilineage dysplasia; MDS-RCC, MDS subtype refractory cytopenia of childhood; MDS-SLD, MDS with single lineage dysplasia; N.a., Not applicable; TAD, N-terminal transactivation domain, ZNF2, Zinc finger 2 domain in GATA2 protein; VAF, variant allele frequency aThese patients have previously been published in Stray-Pedersen, Sorte et al. (2016) (Patient 4 was 84.1, Patient 5 was 84.4, Patient 6 was 88.1, and Patient 7 was 86.1)[21] bThis disease-related GATA2 variant was detected in a routine BM at day + 28 post-HSCT; it turned out to be donor-derived (from a MUD) cWES identified a potential splicing variant in GATA2 (c.1143 + 5G > A) in Patient 4. The variant was predicted (Alamut®) to inactivate the donor site of GATA2 exon 5. PCR amplification of GATA2 exon 4 to 7 on cDNA showed that most transcripts were normally spliced resulting in a main product of ~ 400 bps, as seen in the normal control. A slightly longer PCR product including 64 bps of intron 6 sequence via a cryptic donor site in intron 6 (NM_001145661.1), was observed in the sample from Patient 4, but not in the control (see Supplementary information). Sanger sequencing identified the GATA2 splicing variant in the proband’s deceased monozygotic twin (Patient 5). Details described in Supplemental Figure E6 in Stray-Pedersen, Sorte et al. (2016)[21] dNomenclature according to ISCN (The International System for Human Cytogenetic Nomenclature) 2020 guidelines Results Characteristics of Patients Between 2013 and 2020, ten index cases, or probands, and four additional symptomatic patients with germline GATA2 variants were identified (9 females, 5 male, Table 1). Five adult patients were diagnosed by infectious disease specialists (Patients 1, 4, 5, 10, and 11) where infections (mostly HPV infection/warts and recurrent bacterial airway infections) were prominent symptoms. Four additional adult patients were identified by hematologists (Patients 6, 8, 9, and 14), where three were referred with pancytopenia and one patient had AML (also with pancytopenia). Three patients were diagnosed by pediatricians, two patients with MDS (Patients 12 and 13) and one patient with extensive warts and NK-/B-cell deficiency (Patient 7). Additionally, two patients with GATA2 deficiency were identified after family screening (Patients 2 and 3). We did not detect any asymptomatic individuals with GATA2 deficiency in this study. The mean age for debut of symptoms, that we considered related to GATA2 deficiency, was 12 years (range 0–24 years, Supplemental Table S1). The median time from these symptoms to a diagnosis of GATA2 deficiency was 11 years, range 0–53 years (Supplemental Table S1). Retrospectively, hearing loss, warts, and skin manifestations were the most common early symptoms, which in some patients became apparent many years before the genetic diagnosis of GATA2 deficiency was made (Supplemental Table S1). A summary of the patients’ clinical characteristics is given in Table 1. Viral infections such as HPV-associated warts were common, affecting eight patients. In addition, two patients had disseminated BCG infections (after vaccination), and one patient had a life-threatening disseminated HSV infection (originating from genitalia and disseminating to CNS and liver). Two patients experienced prolonged EBV and/or Varicella zoster viremia. Furthermore, six patients had recurrent bacterial airway infections. Interestingly, one patient had early graying (Patient 8), with normal telomere length, and one patient had multiple aneurysms of small vessels (coronary arteries, axillary arteries, and an iliac artery; Patient 10), which both represent clinical characteristics not previously described in GATA2 deficiency. In Patient 10, Varicella zoster infection was excluded as a cause of vasculitis with negative VZV PCR in blood. In addition, two patients had obliterating bronchiolitis (Patients 4 and 5), which has been reported in only one previous patient with GATA2 deficiency [30]. Affected cell lineages and immunoglobulin levels prior to allo-HSCT are listed in Table 3. As expected, the majority of patients had decreased levels of monocytes (11/14) and one patient had increased levels of monocytes (Patient 12). In addition, decreased levels of B cells (10/11) and NK cells (9/11) were common findings (three patients did not have NK- and B cells measured before allo-HSCT).Table 3 Immunoglobulin levels and affected cell lineage in peripheral blood prior to HSCT Patient no Affected cell lineage (normal range) Immunoglobulins Time before HSCT (months) CD19 + , cells × 106/L (100–500) NK cells × 106 /L (100–400) CD3 +  T- cells × 106/L (800–2400) CD4 +  T- cells × 106/L (500–1400) CD8 +  T-cells × 106/L (200–2000) Monocytes  × 109/L (0.2–0.8) Neutrophils  × 109/L (1.5–7.3) IgG g/L (6.9–15.7) IgG2 g/L (1.69–15.7) 1 0 0 118 43 64 0.0 0.2 5.9 ND 5 2 10 10 550 200 300 0.3 2.9 9.0 ND 4.5 3 160 150 1930 780 950 0.2 2.4 8.4 ND NA 4  < 10 15 349 185 145 0.0 3.4 9.5 1,24 17 5 6 1 259 134 66 0.0 6.0 13.0 0.79 221 6 2 0 809 495 326 0.1 0.5 44.32 2.3 4 7 18 19 913 545 341 0.0 2.8 14.0 2.60 3 83 40 2 1335 546 761 0.1 1.9 9.7 ND NA 9 ND ND ND ND ND 0.0 4.2 16.2 - 1 10 70 206 247 134 108 0.1 1.7 18.5 0.79 6 11 ND ND ND ND ND 0.0 0.9 9.5 2.11 10 12 ND ND ND ND ND 1.7 2.6 8.1 ND 1 134 28 13 1073 633 417 0.2 0.3 12.8 ND 0.5 14 8 11 676 323 341 0.1 1.2 11.9 1.14 NA Abnormal values are given in bold 1On Prednisolone 20 mg a day when these samples were taken 2Hypergammaglobulinemia on IVIG due to IgG2 deficiency 3The values are from the time at diagnosis of GATA2 deficiency 3 years ago 4The reference values for Patient 13 who was 12 years old at the time of HSCT are CD19 200–600, NK 70–1200, CD3 800–3500, cd4 400–1200, cd8 200–1200 (given in cells × 106/L) and for IgG the normal reference value was 6.1–14.9 g/L ND, not done; NA, not applicable Germline GATA2 Variants and Somatic Variants in Other Genes Ten different GATA2 pathogenic, or likely pathogenic, variants were found in 14 patients (Table 2). All identified constitutional GATA2 variants, except one, were located in the ZNF2 domain, corresponding to or in close proximity to exons 5 and 6 (Table 2). Three nonsense variants (p.Ala342Glyfs*43, p.Arg362*, p.Asn381fs*20), one + 1 splicing variant, three missense variants (p.Thr354Met, p.Trp360Arg, and p.Ala372Thr), all previously reported to be disease-causing[14, 23–25], and two novel in-frame deletions (p.Thr358del, p.Asp367del) were found. The variant located in exon 3, outside the ZNF2 domain, was a nonsense variant (c.163C > T, p.Gln55*). It was initially identified by the NGS myeloid panel with variant allele frequency (VAF) 49% in the DNA from the patient’s BM and later verified to be germline with VAF 49% in a buccal DNA sample (Patient 9, Table 2). Patient 2 had a paternal inherited in-frame deletion, c.1062_1064del (p.Thr358del), in the ZNF2 domain, and a somatic in-frame deletion, c.1168_1170del (p.Lys390del), with a fairly high VAF, 40.2% in BM. As expected, these two in-frame deletions were no longer detectable after allo-HSCT. Surprisingly, in the first post-transplant BM sample at day + 28, we detected another acquired GATA2 variant, c.1215G > T (p.Lys405Asn) with VAF 49.5%. This missense mutation variant, affecting an amino acid located C-terminal to the ZNF2 domain, is a variant of unknown clinical significance. It is most likely a rare benign variant, which in retrospect was confirmed to be constitutional in the unrelated BM donor (Table 2). It is evaluated to ACMG category 3 minus, since altogether 5 heterozygote individuals with the same amino acid change p.Lys405Asn are reported in GnomAD (v.2.1.1)[31]. As far as we know, missense variants located outside the ZNF2 domain rarely represent constitutional susceptibility to development of myelodysplasia. One exception is the p.Ser447Arg located C-terminal of the ZNF2 domain[32], while no other missense variants outside the ZNF2 domain are currently defined as pathogenic or likely pathogenic in ClinVar (www.clinvar.com) as of year 2021. Karyotype abnormalities and somatic variants in other genes observed in the 14 patients are presented in Table 2. Trisomy 8 (n = 6), monosomy 7 (n = 3), STAG2 variants (n = 3), ASXL1 variants (n = 2), a combination of somatic variants in RUNX1/CSF3R/PHF6/FBXW7 (n = 1), and variants in the following MDS genes[33] were observed once in separate individuals: BCOR, SETBP1, U2AF1, and somatic GATA2. One adult GATA2 deficient patient who developed AML had an unbalanced translocation der(1;7) in the leukemic clone. Since GATA2 deficiency is considered the most common hereditary predisposition to pediatric MDS, we estimated the proportion of pediatric patients diagnosed with MDS in Norway that had a germline GATA2 variant in the same time period (2013–2020). We found that three out of 14 pediatric patients diagnosed with MDS (21%) had GATA2 deficiency. Of note, these are 14 pediatric MDS patients and not the same cohort of 14 GATA2 deficient patients described above (except three overlapping pediatric patients, Patients 2, 12, and 13, with MDS). Two of the 3 pediatric GATA2 deficient patients with MDS had both monosomy 7 and trisomy 8 in their bone marrow cells. Patients 4, 5, 8, 9, 10, and 11 from Family B, E, F, and G had frameshift or other definitive loss-of-function variants, while Patients 1, 2, 3, 6, 7, 12, 13, and 14 from Family A, C, D, H, I, and J had in-frame deletions or missense variants. No specific genotype–phenotype correlations were found in our data set, i.e., regarding debut of symptoms, type and distribution of infections, age of transition to MDS/AML, somatic occurring variants in blood and BM. The severity of the clinical presentations also varied within families. Families and Predictive Genetic Testing The pedigrees of the 10 families are presented in detail in Fig. 1. Patient 1 (Family A) had three apparently healthy children, when he was diagnosed with GATA2 deficiency. After genetic testing of first-degree relatives, we found that two of his children (Patients 2 and 3) had inherited the GATA2 variant. For Patient 2, initial clinical work-up revealed only mild cytopenia and warts. However, within 2 years of follow-up, he developed pancytopenia and transfusion dependency and was diagnosed with MDS-EB1. His sister, Patient 3, has warts as her only clinical manifestation, but will be followed up regularly for development of cytopenia/MDS.Fig. 1 Pedigrees of the ten families, including 14 patients, with known GATA2 deficiency. Solid symbols denote affected status. Individuals marked in gray are deceased and not tested for GATA2 deficiency but are suspected to carry the disease-causing variant. In family G, the mother of Patients 10 and 11 died at age 30 of acute respiratory distress syndrome, 27 years ago. She also had lymphedema since birth. In light of their mother’s medical history, the GATA2 variant is probably maternally inherited. The father is alive and healthy. In family J, the mother of Patient 14 had a combined B and T cell defect, warts, myelodysplastic syndrome, lymphedema, and recurrent respiratory tract infections. She died of vulval cancer at the age of 38. The maternal grandfather of Patient 14 died of acute leukemia at the age of 33. WT, wild-type In Family G, two siblings had the same germline GATA2 variant (Patients 10 and 11). Their mother died 27 years ago, at the age of 30, of acute respiratory distress syndrome (ARDS), of unknown etiology. She also had lymphedema since birth. In light of their mother’s medical history with lymphedema and ARDS, which could be secondary to complications related GATA2 deficiency, the GATA2 variant is probably maternally inherited. Their father is alive and healthy. The deceased mother of Patient 14 (Family J) had a combined B- and T cell defect, warts, MDS, lymphedema, and recurrent respiratory tract infections. At the age of 38 (years), she died of metastatic vulval cancer. The maternal grandfather of Patient 14 died of acute leukemia at the age of 33. Considering the family history, it is very likely that Patient 14 had inherited her germline GATA2 variant from her maternal grandfather via her mother. Both individuals died before GATA2 deficiency was acknowledged as a cause of PID. Her mother’s siblings are now offered genetic counselling/testing for GATA2 deficiency. Allo-HSCT in Patients with GATA2 Deficiency Twelve of 14 (86%) patients with GATA2 deficiency were found to have a clinical indication, cytogenetic findings, and/or molecular findings warranting to proceed to allo-HSCT. As of today, 11 patients have undergone allo-HSCT, whereas one is recently accepted for allo-HSCT (Patient 14). Clinical features that lead to the decision to perform allo-HSCT were previous life-threatening disseminated HSV infection (Patient 1), severe obliterating bronchiolitis and in situ carcinoma (Patients 4 and 5), MDS with cytogenetic abnormalities (monosomy 7) and/or excess of blasts with high likelihood of progression to leukemic transformation (Patients 2, 6, 7, 9, 12, 13, and 14), MDS and warts with high-grade dysplasia (Patient 10), and symptoms of severe immunodeficiency and MDS (Patient 11). Details on the allo-HSCT procedure, including conditioning, donor selection, stem cell source, donor/recipient cytomegalovirus status, donor chimerism, graft versus host disease (GVHD) prophylaxis, and the occurrence of GVHD, are presented in Table 4.Table 4 HSCT details for eleven patients with GATA2 deficiency Patient no Age at HSCT Donor Stem cell source HLA Match CD34 + , × 106/kg CMV status d/r Conditioning regimen & In vivo T-cell depletion Chimerism %Day + 28 GVHD prophylaxis GVHD Complications 1 41 y MUD PBSC 10/10 (11/12) 7.8 -/ +  RIC: Fludarabine 150 mg/ m2, Busulfan 8 mg/kg, ATG Thymoglobulin 4 mg/kg 99% Mtx + CsA No Hemorrhagic cystitis (BK-virus) 2 16 y MUD PBSC 10/10 (10/12) 9.7 -/ - MAC*: Busulfan for 4 days (TDM; Css 825 ng/ml), Cyclophosphamide 120 mg/kg, Melphalan 140 mg/m2, ATG Grafalon 3 × 10 mg/kg  > 99% Mtx + CsA No E. coli sepsis; BK-virus cystitis; mucositis (grade 3) 4 39 y MUD PBSC 10/10 (12/12) 10.6 + / +  RIC: Fludarabine 150 mg/ m2, Busulfan 8 mg/kg. ATG Thymoglobulin 4 mg/kg  > 99% Mtx + CsA Chronic: skin and gut CMV reactivation 5 39 y MUD PBSC 10/10 (11/12) 5.2 +/ - RIC: Fludarabine 90 mg/m2, 2 Gy TBI 99% Mtx + CsA No Enterococcus faecalis sepsis (2 months post-HSCT), prolonged cytopenia, died of intracerebral hemorrhage 8 months post-HSCT 6 29 y MRD PBSC HLA-id sibling 5.4 + / +  RIC: Fludarabine 150 mg/m2, Busulfan 8 mg/kg 98% Mtx + CsA Chronic: liver, oral mucosa and genitalia Cytopenia at day + 33, osteoporosis, compression fractures 7 22 y MUD PBSC 10/10 6.9 -/ - MAC: Fludarabine 150 mg/m2, Treosulfan 42 g/m2, ATG Thymoglobulin 4 mg/kg  > 99% Mtx + CsA Chronic (limited): skin Moraxella nonliquefaciens sepsis on day 0. E. coli sepsis day + 14. Oral mucositis grade IV. PTLD 7 weeks post-HSCT 9 23 y MUD PBSC 10/10 (11/12) 4.3 +/ - MAC: Fludarabine 160 mg/m2, Busulfan 12,8 mg/kg (i.v.), ATG Thymoglobulin 4 mg/kg  > 99% Mtx + CsA Acute GvHD grade I: skin None 10 32 y MUD PBSC 10/10 (11/12) 5.9 -/ - MAC: Fludarabine 150 mg/m2, Treosulfan 42 g/m2, ATG Thymoglobulin 4 mg/kg  > 99% Mtx + CsA Acute GvHD: serositis None 11 27 y MUD PBSC 10/10 (10/12) 10.2 -/ - MAC: Cyclophosphamide 100 mg/kg, Busulfan 16 mg/kg N.a Mtx + CsA Chronic (extensive): gut, eye, and lung Hemorrhagic cystitis, Herpes oesophagitis 12 14 y MUD BM 10/10 (11/12) TNC: 3.5 × 108/kg -/ - MAC*: Fludarabine 160 mg/m2, Treosulfan 42 g/m2,Thiotepa 8 mg/kg, ATG Grafalon 3 × 10 mg/kg day + 84: > 99% Mtx + CsA No Impetigo day + 40 13 11 y MMFD (father) PBSC, TCRab + /CD19 + depletion in vitro Haploidentical 10.3 + / +  MAC*: Fludarabine 160 mg/m2, Thiotepa 10 mg/kg, Melphalan 140 mg/m2, ATG Grafalon 3 × 10 mg/kg  > 99% MMF (until day + 28) No None Abbreviations: BM, bone marrow; CMV, cytomegalovirus; CsA, cyclosporine A; Css, concentration at steady-state; Cya, cyclosporine A; GVHD, graft versus host disease; i.v., intravenous; MAC, myeloablative conditioning; MMF, mycophenolate mofetil; MMFD, mismatched family donor; MRD, matched related donor; Mtx, methotrexate: MUD, matched unrelated donor; N.a., not available; PBSC, peripheral blood stem cells; RIC, reduced-intensity conditioning; TDM, therapeutic drug monitoring; Tx, transplantation *According to HSCT recommendations by the EWOG-MDS study group Clinical Outcome The clinical outcome of all 14 patients is presented in Table 1. Two adult patients (18%) died after allo-HSCT. Patient 5 had persistent thrombocytopenia and died of a cerebral hemorrhage 8 months post transplantation. Patient 11 developed respiratory failure due to cGVHD in the lungs and received a bilateral pulmonary transplant 5 years post allo-HSCT. However, he developed chronic pulmonary rejection and died 2 years after lung transplantation and 7 years after allo-HSCT. The mean follow-up of the nine patients still alive after allo-HSCT is 26 months (range 3–78 months). The incidence of aGvHD and cGVHD among the eleven transplanted patients was 25% and 33%, respectively, all occurring in patients > 18 years of age (Table 4). None of the pediatric patients had experienced aGVHD or cGVHD, serious infectious complications, or any serious or unexpected transplant-related acute or late toxicity. Their transplantation courses were uneventful and did not principally differ from MDS patients without germline disease-causing GATA2 variants. One patient is listed for allo-HSCT (Patient 14) and two patients are followed regularly in the out-patient clinic (Patients 3 and 8). Discussion This retrospective study describes clinical features and outcome of 14 patients from ten families diagnosed with GATA2 deficiency in Norway. The main findings were as follows: (i) We found a diverse clinical phenotype dominated by cytopenia (13/14), myeloid neoplasia (MDS/AML) (10/14), warts (8/14), and hearing loss (7/14). (ii) We observed two novel clinical features multiple aneurysms of small vessels (n = 1) and early graying (n = 1) that could be associated with GATA2 deficiency. (iii) The majority of patients (11/14) had already undergone allo-HSCT at the time of our analysis, illustrating the need for allo-HSCT in a large proportion of patients with GATA2 deficiency in Norway, and most likely in other countries. (iv) Genetic testing should be offered to first-degree relatives, particularly children, to identify individuals with GATA2 deficiency that need close surveillance. Genetic testing performed as soon as a clinical suspicion is raised, increases the likelihood of an early and correct molecular diagnosis. In Norway, exome-based panels including GATA2 are offered as a routine diagnostic laboratory service for constitutional hematological disorders and PID [21]. However, this approach will not detect all GATA2 variants. Germline variants located in the intronic transcriptional enhancer elements, the cis-acting E-box/GATA and ETS motifs within intron 5 (NM_001145661.1), may cause GATA2 deficiency[34]. Supplementary Sanger sequencing of the enhancer element sequences in intron 5[34] was only performed in one of the laboratories (see Supplementary Methods). Despite supplementary copy number variant calling from exome data, with chromosomal microarray and MLPA only in selected patients (Supplementary Methods), some structural variants may go undetected. In addition, synonymous disease-causing GATA2 variants resulting in selective loss of mutated RNA were recently reported [35]. Disease-causing intronic variants, small intragenic variants, such as structural variants and synonymous variants, may also have escaped detection and hence, individuals with GATA2 deficiency may have been overlooked. With the recent introduction of targeted NGS panels in the work-up of myeloid neoplasms searching for somatic GATA2 variants, unexpected germline variants can be identified, which may reveal the underlying constitutional cause of the myeloid disease and increase the prevalence of known GATA2 deficiency. Improving the NGS panels targeting both germline and somatic variants by deeper coverage of the whole genes including important non-exonic regions, better algorithms for detection of structural variants, and attention to rare synonymous variants may enhance the identification of GATA2 deficiency. In this case series, we found two clinical features that have not yet been described in GATA2 deficiency, namely small vessel aneurysm and early graying. Multiple small vessel aneurysms found in Patient 10 may be secondary to vasculitis, or could also represent a novel vascular feature associated with GATA2 deficiency. In fact, it has been suggested that alteration in GATA2 expression may be of importance for vascular integrity[36]. The observation of premature graying may be a coincidence. In the absence of telomere biology disorders as in the present patients, one may speculate if this feature may reflect a GATA2-linked autoimmune phenomenon which has gone undetected. One of the aims of this retrospective study was to describe the clinical characteristics, GATA2 variants, and other molecular variants representing risk factors for clonal evolution that could aid us in the difficult decision regarding: “Who and when to transplant”? The high proportion of patients (79%) that had already undergone allo-HSCT in our cohort was somewhat surprising. Donadieu et al. have published the largest cohort of patients with GATA2 deficiency, and found that only 28 patients (35%) of 79 patients had undergone allo-HSCT. However, this low percentage of allo-HSCT did not correspond with the severity of the disease in this cohort. At the age of 40, the authors reported a mortality rate of 35% and a hematological malignancy rate of 80%[14]. The high proportion of allo-HSCT in our study may have been influenced by increased awareness of negative prospective clinical markers of GATA2 deficiency. Based on the findings from our study and the high morbidity and mortality rate reported by Donadieu et al., it is clear that these patients need to be monitored closely. Ideally, allo-HSCT should be performed before they develop malignancies (both solid tumors and hematological malignancies)[37] or severe/recurrent infections causing organ failure. In our opinion, a history of disseminated viral infection, aggressive HPV infection (particular with dysplasia), or myeloid clonal disease is clear indication to consider allo-HSCT[14]. First-degree relatives with a severe outcome of the disease may further strengthen the indication for an early allo-HSCT in symptomatic patients with GATA2 deficiency. Overall, the decision to perform an allo-HSCT requires careful weighing of potential gain (restore immune function; diminish the risk of hematological malignancies) versus possible transplant complications, including GVHD and transplant-related mortality. This is particularly challenging given the lack of genotype–phenotype correlation. Keeping in mind that the observation time is short for some of the patients in our study, the survival rate after allo-HSCT was 82% (9/11). In patients with GATA2 deficiency, previous publications have reported 86% survival 2 years after HSCT (n = 22)[16], 73% and 62% survival 1 and 5 years after HSCT, respectively (n = 28)[14], 72%, 65%, and 54% survival 1, 2, and 4 years after HSCT, respectively (n = 21)[9], and 57% 3, 5 years after HSCT (n = 14)[38]. These cohorts are, however, not necessarily comparable in terms of severity of disease and conditioning regimen. Two children were diagnosed with GATA2 deficiency after family screening. The hematological surveillance of one of these children led to detection of hematological abnormalities consistent with MDS and, in the end, a timely allo-HSCT. We therefore recommend genetic testing of children of affected adults and hematological surveillance of individuals with known pathogenic germline GATA2 variants. This includes annual BM investigations with morphological and cytogenetic evaluations, and testing with NGS myeloid panel to screen for somatic occurring molecular drivers of malignancies. Monosomy 7 and trisomy 8 have been reported by others to be the major cytogenetic aberrations in hematopoietic cells of patients with GATA2 deficiency and MDS[15]. Advanced MDS disease and monosomy 7 have been related to worse outcome, especially for pediatric patients with GATA2 germline disease[20]. We found monosomy 7 only in ¼, but trisomy 8 in half of the karyotyped patients, which is in line with a previous published study by McReynolds et al. [39]. Two patients in our GATA2 cohort had monosomy 7 and trisomy 8, both were children. Some of the other somatic variants in our cohort occurred in genes previously reported to be mutated in GATA2 deficiency with MDS, such as ASXL1[40], STAG2, SEPTBP1, and RUNX1[41, 42]. Interestingly, one adult patient with GATA2 deficiency and MDS-related AML had a der(1;7) in the leukemic clone, a translocation that has recently been shown to be enriched in pediatric MDS patients with germline GATA2 mutations[43]. However, our number of patients are too small to determine possible genotype–phenotype correlations related to clonal disease progression. Previous reports have suggested that plasma levels of FLT3LG can be used as predictor of hematological disease in GATA2 deficiency, and used in clinical monitoring post-HSCT [25]. Unfortunately, we lack serum or plasma samples taken before and after HSCT in our cohort, and the role of FLT3LG as a disease progression marker could be explored in future studies. Our main conclusion of this study is that the majority of patients with symptomatic GATA2 deficiency will need allo-HSCT, and close surveillance of these patients is important to find the “optimal window” for allo-HSCT. We advocate a more offensive approach to allo-HSCT than previously described. Supplementary Information Below is the link to the electronic supplementary material.Supplementary file1 (DOCX 55 KB) Acknowledgements Our gratitude goes to the Section for Cancer Cytogenetics who karyotyped 9 of the 14 patients reported here (Table 2). Three of the reported families have been followed up at the Department of Medical Genetics, University Hospital of North Norway. We want to thank our colleagues, MD Gry Hoem, MD Hilde Yttervik, MD Specialist in medical genetics and pediatrics Marie Falkenberg Smeland, and MD PhD Øyvind Holsbø Hald at the Department of Medical Genetics, University Hospital of North Norway, Tromsø, Norway, for their clinical work and genetic counselling. The three pediatric patients are enrolled in the registry of the European Working Group of MDS in Childhood (EWOG-MDS; ClinicalTrials.gov Identifier: NCT00662090). Author Contribution SFJ, JB, AEM, PA, AS-P, TGD, and IN wrote the paper. The genetic analyses were performed by AS-P, MAK, HS, ØH, SS, and EL. SFJ, JB, AEM, EG, YF, CA, AB, IH, TF, BF, AS-P, TGD, and IN collected clinical data. SS did the bone marrow analyses. All authors critically revised the manuscript for important intellectual content and approved the final version of the manuscript. Funding Open access funding provided by University of Oslo (incl Oslo University Hospital). Data Availability Upon request. Code Availability Not applicable. Declarations Ethics Approval Five of the patients had previously consented to be part of a genetic PID research project approved by the regional ethical committee (REC. 2014/1270–1), three patients were diagnosed with MDS in childhood and consented to be registered into the EWOG-MDS-2006 study (2015/1651/REC Nord), and all adult patients who underwent allo-HSCT had consented to publication of data (REC 11909). Studies were performed according to the Declaration of Helsinki. Consent to Participate All adult living patients signed a written informed consent for publication. For children < 18 years, consent was given by their parents. Consent for Publication Obtained. Conflict of Interest The authors declare no competing interests. Abbreviations aGVHD Acute graft versus host disease allo-HSCT Allogeneic hematopoietic stem cell transplantation AML Acute myelogenous leukemia ARDS Acute respiratory distress syndrome BM Bone marrow cGVHD Chronic graft versus host disease GVHD Graft versus host disease HPV Human papilloma virus MDS Myelodysplastic syndrome MRD Matched related donor MUD Matched unrelated donors NGS Next-generation sequencing PID Primary immunodeficiency VAF Variant allele frequency WES Whole exome sequencing ZNF1 Zink finger 1 ZNF2 Zink finger 2 Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Asbjørg Stray-Pedersen, Tobias Gedde-Dahl and Ingvild Nordøy contributed equally

BUSULFAN, CYCLOSPORINE, FLUDARABINE PHOSPHATE, METHOTREXATE

DrugsGivenReaction

CC BY

34893945

2021-12-10

Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Cytopenia'.

A Nationwide Study of GATA2 Deficiency in Norway-the Majority of Patients Have Undergone Allo-HSCT. OBJECTIVE GATA2 deficiency is a rare primary immunodeficiency that has become increasingly recognized due to improved molecular diagnostics and clinical awareness. The only cure for GATA2 deficiency is allogeneic hematopoietic stem cell transplantation (allo-HSCT). The inconsistency of genotype-phenotype correlations makes the decision regarding "who and when" to transplant challenging. Despite considerable morbidity and mortality, the reported proportion of patients with GATA2 deficiency that has undergone allo-HSCT is low (~ 35%). The purpose of this study was to explore if detailed clinical, genetic, and bone marrow characteristics could predict end-point outcome, i.e., death and allo-HSCT. METHODS All medical genetics departments in Norway were contacted to identify GATA2 deficient individuals. Clinical information, genetic variants, treatment, and outcome were subsequently retrieved from the patients' medical records. RESULTS Between 2013 and 2020, we identified 10 index cases or probands, four additional symptomatic patients, and no asymptomatic patients with germline GATA2 variants. These patients had a diverse clinical phenotype dominated by cytopenia (13/14), myeloid neoplasia (10/14), warts (8/14), and hearing loss (7/14). No valid genotype-phenotype correlations were found in our data set, and the phenotypes varied also within families. We found that 11/14 patients (79%), with known GATA2 deficiency, had already undergone allo-HSCT. In addition, one patient is awaiting allo-HSCT. The indications to perform allo-HSCT were myeloid neoplasia, disseminated viral infection, severe obliterating bronchiolitis, and/or HPV-associated in situ carcinoma. Two patients died, 8 months and 7 years after allo-HSCT, respectively. CONCLUSIONS Our main conclusion is that the majority of patients with symptomatic GATA2 deficiency will need allo-HSCT, and a close surveillance of these patients is important to find the "optimal window" for allo-HSCT. We advocate a more offensive approach to allo-HSCT than previously described. pmcIntroduction GATA2 deficiency is a rare primary immunodeficiency (PID), first described in 2011[1–3] that has become gradually more recognized due to improved molecular diagnostics and increased clinical awareness. GATA2, as a “master” transcription factor, plays a critical role in hematopoietic development[4]. Through cooperative processes that include other transcription factors, it controls the transition from hemogenic endothelium to hematopoietic stem cells and is required for survival and self-renewal of these cells[5]. GATA2 is also important for other tissue-forming stem cells, e.g., in the inner ear[6]. The heterozygous variants causing GATA2 deficiency are located both in coding, non-coding and enhancer regions[7]. The disease-causing loss-of-function variants can be localized across the gene. These variants can lead to defective DNA-binding capacity of the transcription factor and may cause disease through haploinsufficiency of the functional protein[5, 8]. Missense variants within the zink finger 2 (ZNF2) domain are the most frequent germline disease-causing GATA2 variants [9]. It has been estimated that approximately 1/3 of the patients have an autosomal dominant inherited disease-causing variant[10], whereas the remaining have a de novo GATA2 variant[7]. Of note, somatic variants in GATA2 are known to be drivers of myeloid neoplasia in adults. Such variants are diverse, may cause gain-of-function effects, and be located across the whole gene. This includes missense variants in the zink finger 1 (ZNF1) domain, which has not been observed in constitutional GATA2 deficiency[8]. Typically, GATA2 deficiency becomes clinically apparent in late childhood to early adulthood. The phenotype is heterogeneous, without any clear genotype–phenotype correlation, and with an incomplete clinical penetrance[11]. Symptoms may include recurrent or severe infections, warts, cytopenia (including monocytopenia), lymphedema, alveolar proteinosis, and malignant myeloid disease[9]. Infectious complications in GATA2 deficiency are likely due to deficiency of monocytes, NK cells, and B-lymphocytes as well as defective innate immune responses, including impaired type I interferon production[12]. This leads to both increased susceptibility to viral infections (e.g., human papilloma virus [HPV, warts] and herpes virus infections), non-tuberculous mycobacteria, and to more common bacterial respiratory infections. Hearing loss is a common clinical feature of GATA2 deficiency and is related to the critical role of GATA2 in vestibular morphogenesis of semicircular ducts and generation of the perilymphatic space around the inner ear’s semicircular canals[6, 13]. A substantial proportion of patients develop immunodeficiency, myelodysplastic syndrome (MDS), or acute myelogenous leukemia (AML) as initial manifestation[9, 14]. GATA2 deficiency is considered the most common hereditary predisposition to pediatric MDS[15]. The only cure for GATA2 deficiency is allogeneic hematopoietic stem cell transplantation (allo-HSCT) and results are encouraging[16–20]. However, the main challenge is deciding who and when to transplant due to the complexity and inconsistency of phenotype-genotype correlation in GATA2 deficiency[9]. To further elucidate this important issue, we present detailed clinical and molecular characteristics, treatment, and outcome of 14 Norwegian patients with germline GATA2 variants diagnosed between 2013 and 2020. The main aim of our study was to explore if detailed clinical, genetic, and bone marrow (BM) characteristics could predict end-point outcome such as death and allo-HSCT in patients with GATA2 deficiency. Methods Identification of Patients and Clinical Characteristics The first aim of this study was to obtain a complete overview of all patients with known GATA2 deficiency in Norway. For this purpose, a network of clinical immunologists, hematologists, pediatricians, and geneticists at Oslo University Hospital (OUH) collected clinical and laboratory data on patients with GATA2 deficiency at their institution. In addition, all medical genetics departments in Norway were contacted to identify any additional GATA2 deficient individuals. Clinical information, genetic variants, treatment, and outcome were subsequently retrieved from the patients’ medical records. Patients were enrolled into the study at OUH where most of the data was obtained, while supplemental data from Patient 3 was collected at the University Hospital of North Norway, Tromsø. Informed Consent Five of the patients had previously consented to be part of a genetic PID research project approved by the regional ethical committee (REC. 2014/1270–1), three patients were diagnosed with MDS in childhood and consented to be registered into the EWOG-MDS-2006 study (2015/1651/REC Nord), and all adult patients who underwent allo-HSCT had consented to publication of data (REC 11,909). In addition, due to the detailed clinical information published herein, all adult living patients signed an additional written informed consent for publication of their data and was given the opportunity to review the manuscript prior to publication. For children < 18 years, consent was given by their parents. This is in line with the recommendation given by the Ethical Constitutional board at OUH. Genetic Analyses Whole exome sequencing (WES) with in silico filtering for genes causing primary immunodeficiency disorders was performed in the probands and affected relatives as part of a routine laboratory service (Patients 2, 3, 8, 9, 10, 12, and 14) or on a research basis (Patients 4, 5, 6, and 7) as previously described (Supplemental methods)[21]. Patient 1 had severe cytopenia (Table 1), and the first attempt to extract DNA from peripheral blood was not successful. A skin biopsy was therefore performed to extract DNA from fibroblasts. In parallel, peripheral blood (from puncture of the fingertip) was applied directly to a Guthrie filter card, and by using multiple filter card punches, enough DNA was extracted to run next-generation sequencing (NGS) with an amplicon-based targeted panel for constitutional variants in PID genes (Supplemental methods). By using this rapid amplicon-based method, the molecular result was available within 3 working days[22]. DNA later extracted from fibroblasts confirmed the GATA2 variant by Sanger sequencing. Also, for Patient 13, who had advanced MDS with pancytopenia, the NGS results were available within 3 working days, with parental testing performed in parallel to evaluate as fast as possible the availability of a healthy unaffected matched related donor (MRD).Table 1 Clinical characteristics and outcome in patients with GATA2 deficiency Patient no Family Sex Current age Age at onset of symptoms/age at genetic diagnosis Infections Hearing loss Hematologic abnormalities Autoimmunity/immune dysregulation Miscellaneous HSCT, age Outcome Viral Bacterial 1 A (father of P2 and P3) M 44y 5y/41y HPV: warts HSV: disseminated disease Ear infections as a child Yes Hypoplastic BM: cytopenia, trilinear hypoplasia No No 41y Alive 2 A (son of P1) M 16y 7y/14y HPV: warts No No MDS-EB-1 No No 16y Alive 3 A (daughter of P1) F 13y 8y/9y HPV: warts No No No No No ND Alive 4a B (monozygotic twin to P5) F 45y 21y/38y HPV: warts, carcinoma in situ EBV: prolonged viremia Recurrent respiratory inf Suppurative skin infection after BCG * Yes No Progressive obliterating bronchiolitis, lupus-like syndrome Miscarriage 39y Alive 5a B (monozygotic twin to P4) F † (39y) 24y/38y HPV: warts, carcinoma VZV and EBV: prolonged viremia Recurrent respiratory inf Suppurative skin inf. after BCG* Yes MDS-MLD (hypoplastic) Progressive obliterating bronchiolitis, lupus-like syndrome DVT × 2 Squamous cell carcinoma in the cervix, rectum, and anus 39y Deceased 8 m post-HSCTb 6a C M 31y 11y/26y No Recurrent respiratory inf Yes MDS-MLD (hypoplastic) - Fever of unknown origin, recurrent pneumothorax 29y Alive 7a D F 23y 6y/17y HPV: warts Recurrent respiratory inf No MDS-MLD (hypoplastic) Interstitial lung disease Lymphedema, acne, rosacea, rash, fatigue 22y Alive 8 E F 56y 0y/53y No No No Hypoplastic BM No Lymphedema, premature graying ND Alive 9 F F 24y 15y/23y No No No AML with MDS-related changes Erythema nodosum DVT, PE, juvenile myoclonic epilepsy, epicanthic fold 23y Alive 10 G (sibling to P11) F 32y 6y/31y HPV: warts, cervix dysplasia Recurrent respiratory inf Yes MDS-MLD No Aneurysm of small vessels, hidradenitis suppurative, liver lesions: focal nodular hyperplasia 32y Alive 11 G (sibling to P10) M † (34y) 22y/PM No Recurrent skin and respiratory inf No MDS-MLD No Acne, rosacea, necrotizing fasciitis, pilonidal cysts, skin infections, ulcerations 27y Deceased 7y post-HSCTc 12 H F 19y 14y/14y No No Yesd MDS-RCC (hypoplastic) BPD/Asthma Born premature (week 26 + 5), BPD 14y Alive 13 I M 13y 9y/11 y HPV: warts No No MDS-EB1 Asthma Chronic skin abscesses, congenital ptosis 11y Alive 14 J F 31 23y/31y No No Yes MDS-SLD (hypoplastic) No Born prematurely (week 25), cerebral palsy, congenital hip dysplasia Planned Alive Abbreviations: AML, acute myeloid leukemia; BCG, bacille Calmette Geurin; BM, bone marrow; BPD, bronchopulmonary dysplasia, CT, computer tomography, HPV, human papilloma virus; HSCT, hematopoietic stem cell transplantation; Inf., infection; m, months; MDS, myelodysplastic syndrome; MDS-EB1, MDS with excess of blasts type 1; MDS-MLD, MDS with multilineage dysplasia; MDS-SLD, MDS with single lineage dysplasia; MDS-RCC, MDS subtype refractory cytopenia of childhood; ND, not done, PM, post mortem; VTE, venous thromboembolism; y, years, †; deceased aThese patients have previously been published in Stray-Pedersen, Sorte et al. 2016 (Patient 4 was 84.1, Patient 5 was 84.4, Patient 6 was 88.1, and Patient 7 was 86.1)[21] bThe patient was doing well after HSCT, but died unexpectedly of a cerebral hemorrhage cThe patient underwent lung transplantation for chronic lung GVHD 58 months after HSCT, and died of chronic lung rejection 26 months after bilateral lung transplantation dThe patient has reduced hearing, but this was confirmed after HSCT. Her hearing loss may be due to the disease-causing GATA2 variant, but may also be secondary to complications of HSCT therapy, e.g., aminoglycosides The molecular diagnosis in Patient 11 was confirmed post mortem using a BM sample collected prior to allo-HSCT (Table 2). Methods for testing for somatic occurring sequence variants on DNA extracted from whole blood or BM, and methods for testing chromosomal aberration on BM cells are described in Supplemental methods.Table 2 Constitutional and acquired genetic findings in patients with GATA2 deficiency Patient no Hematological abnormalities Constitutional heterozygous variants in GATA2, NM_001145661.1, predicted protein effect, domain, occurrence, novelty, and reference Somatic variants, predicted protein effect, VAF in BM/blood (prior to HSCT) Karyotype in BM (closest to HSCT)d  + 8  − 7 1 Hypoplastic bone marrow c.1062_1064del, p.Thr358del, in-frame exon 6, ZNF2, likely de novo, novel variant Unknown 46,XY[25/25] No No 2 MDS-EB1 c.1062_1064del, p.Thr358del, in-frame exon 6, ZNF2, paternal inherited, novel variant Acquired germline donor variant in GATA2 Post-HSCTb: c.1215G > T, p.(Lys405Asn) missense exon 7, outside and distal to ZNF2 domain, VAF: 49,5% BM NM_001145661.1 (GATA2): c.1168_1170del, p.(Lys390del), in-frame exon 7, in ZNF2, VAF: 40.2% BM NM_006758.2(U2AF1): c.470A>G, p.(Gln157Arg) VAF: 44,0% BM 46,XY,-7 + 8[20/20] Yes Yes 3 No c.1062_1064del, p.Thr358del, in-frame exon 6, ZNF2, paternal inherited, novel variant Unknown Unknown N.a N.a 4 a No c.1143 + 5G > A, p.Asn381fs*20, splice defect intron 6, ZNF2, de novo, novel, reported by usc)[21] Unknown Unknown N.a N.a 5 a MDS-MLD (hypoplastic) c.1143 + 5G > A, p.Asn381fs*20, splice defect intron 6, ZNF2, de novo, novel, reported by usc[21] None 46,XX[18/18] No No 6a MDS-MLD (hypoplastic) c.1078 T > A, p.Trp360Arg, missense exon 6, ZNF2, de novo, variant previously reported by others[23] Unknown 46,XY[25/25], but FISH MYC(8q24): + 8 in 14/303 metaphases Yes No 7a MDS-MLD (hypoplastic) c.1061C > T, p.Thr354Met, missense exon 6, ZNF2, de novo, but a recurrent GATA2 variant[21, 24, 25] NM_001042749.2(STAG2): c.2534-2A > G, predicted splice variant with loss of acceptor site, Chr.X, VAF: 11.7% blood 47,XX, + 8[4/10]/46,XX[6/10] Yes No 8 Hypoplastic bone marrow c.1017 + 1G > T, loss of donor splice site, splice defect intron 5, ZNF1, both parents deceased and not tested, novel variant Unknown 46,XX[25/25] No No 9 AML with MDS-related changes c.163C > T, p.Gln55*, nonsense exon 3, TAD domain, likely de novo (see pedigree), novel variant VAF:48.7% in BM, 49,4% in buccal swap NM_001754.4(RUNX1): c.593A > G,p.(Asp198Gly) VAF:15% BM NM_156039.3(CSF3R): c.2326C > T, p.(Gln776*) VAF: 12.5% BM NM_032458.2(PHF6): c.309C > G, p.(Tyr103Ter) VAF:12.0% BM NM_033632.3(FBXW7): c.1513C > T, p.(Arg505Cys), VAF:11.7% BM 46,XX, der(1;7)(q10;p10), + 1[11/20]/46,XX [9/20] No No 10 MDS-MLD c.1084C > T, p.Arg362*, nonsense exon 6, ZNF2, likely inherited, variant previously reported by others[15, 26–28] NM_001123385.1(BCOR): c.529_530del, p.(Ser177ProfsTer8), VAF: 23.0% BM 49∼50,XX, + 6, + 8, + 21? + 21[cp7/8]/46,XX[1/8] Yes No 11 MDS-MLD c.1084C > T, p.Arg362*, nonsense exon 6, ZNF2, likely inherited, variant previously reported by others[15, 26–28] NM_015338.5(ASXL1): c.2324 T > G, p.(Leu775Ter) VAF: 20.5% BM NM_001042749.1(STAG2): c.2990 T > A, p.(Leu997Ter) VAF: 9.6% BM 47∼48,XY, + 8[10/15],der(16)t(1;16)(q21;q24[10/15], + der(16)t(1;16)[1/15], + 21[6][cp11/15]/46,XY[3/15] Trisomy 8, evolving to unbalanced 1;16 translocation and later Trisomy 21 Yes No 12 MDS-RCC c.1098_1100delGGA, p.Asp367del, in-frame exon 6, ZNF2, de novo, novel variant None 46,XX,-7, + 8[15/20] Yes Yes 13 MDS-EB1 c.1021_1024insGCCG, p.Ala342Glyfs*43, frameshift exon 6, ZNF1 de novo, variant previously reported[29] NM_015338.5(ASXL1): c.1854dupA, p.(Ala619SerfsTer16), VAF:17.0%, BM NM_015559.2 (SETBP1): c.2612 T > C, p.(Ile871Thr), VAF: 16.3%, BM 45,XY,-7[12/12] No Yes 14 MDS-SLD (hypoplastic) c.1114G > A, p.(Ala372Thr), missense exon 6, ZNF2, variant previously reported [14] NM_001042749.1(STAG2): c.707del; p.(Asn236IlefsTer20) VAF: 5.1%, BM 46,XX[25/25] No No Abbreviations: AML, acute myeloid leukemia; BM, bone marrow; Chr, chromosome; HSCT, hematopoietic stem cell transplantation; MDS, myelodysplastic syndrome; MDS-EB1, MDS with excess blasts type 1; MDS-MLD, MDS with multilineage dysplasia; MDS-RCC, MDS subtype refractory cytopenia of childhood; MDS-SLD, MDS with single lineage dysplasia; N.a., Not applicable; TAD, N-terminal transactivation domain, ZNF2, Zinc finger 2 domain in GATA2 protein; VAF, variant allele frequency aThese patients have previously been published in Stray-Pedersen, Sorte et al. (2016) (Patient 4 was 84.1, Patient 5 was 84.4, Patient 6 was 88.1, and Patient 7 was 86.1)[21] bThis disease-related GATA2 variant was detected in a routine BM at day + 28 post-HSCT; it turned out to be donor-derived (from a MUD) cWES identified a potential splicing variant in GATA2 (c.1143 + 5G > A) in Patient 4. The variant was predicted (Alamut®) to inactivate the donor site of GATA2 exon 5. PCR amplification of GATA2 exon 4 to 7 on cDNA showed that most transcripts were normally spliced resulting in a main product of ~ 400 bps, as seen in the normal control. A slightly longer PCR product including 64 bps of intron 6 sequence via a cryptic donor site in intron 6 (NM_001145661.1), was observed in the sample from Patient 4, but not in the control (see Supplementary information). Sanger sequencing identified the GATA2 splicing variant in the proband’s deceased monozygotic twin (Patient 5). Details described in Supplemental Figure E6 in Stray-Pedersen, Sorte et al. (2016)[21] dNomenclature according to ISCN (The International System for Human Cytogenetic Nomenclature) 2020 guidelines Results Characteristics of Patients Between 2013 and 2020, ten index cases, or probands, and four additional symptomatic patients with germline GATA2 variants were identified (9 females, 5 male, Table 1). Five adult patients were diagnosed by infectious disease specialists (Patients 1, 4, 5, 10, and 11) where infections (mostly HPV infection/warts and recurrent bacterial airway infections) were prominent symptoms. Four additional adult patients were identified by hematologists (Patients 6, 8, 9, and 14), where three were referred with pancytopenia and one patient had AML (also with pancytopenia). Three patients were diagnosed by pediatricians, two patients with MDS (Patients 12 and 13) and one patient with extensive warts and NK-/B-cell deficiency (Patient 7). Additionally, two patients with GATA2 deficiency were identified after family screening (Patients 2 and 3). We did not detect any asymptomatic individuals with GATA2 deficiency in this study. The mean age for debut of symptoms, that we considered related to GATA2 deficiency, was 12 years (range 0–24 years, Supplemental Table S1). The median time from these symptoms to a diagnosis of GATA2 deficiency was 11 years, range 0–53 years (Supplemental Table S1). Retrospectively, hearing loss, warts, and skin manifestations were the most common early symptoms, which in some patients became apparent many years before the genetic diagnosis of GATA2 deficiency was made (Supplemental Table S1). A summary of the patients’ clinical characteristics is given in Table 1. Viral infections such as HPV-associated warts were common, affecting eight patients. In addition, two patients had disseminated BCG infections (after vaccination), and one patient had a life-threatening disseminated HSV infection (originating from genitalia and disseminating to CNS and liver). Two patients experienced prolonged EBV and/or Varicella zoster viremia. Furthermore, six patients had recurrent bacterial airway infections. Interestingly, one patient had early graying (Patient 8), with normal telomere length, and one patient had multiple aneurysms of small vessels (coronary arteries, axillary arteries, and an iliac artery; Patient 10), which both represent clinical characteristics not previously described in GATA2 deficiency. In Patient 10, Varicella zoster infection was excluded as a cause of vasculitis with negative VZV PCR in blood. In addition, two patients had obliterating bronchiolitis (Patients 4 and 5), which has been reported in only one previous patient with GATA2 deficiency [30]. Affected cell lineages and immunoglobulin levels prior to allo-HSCT are listed in Table 3. As expected, the majority of patients had decreased levels of monocytes (11/14) and one patient had increased levels of monocytes (Patient 12). In addition, decreased levels of B cells (10/11) and NK cells (9/11) were common findings (three patients did not have NK- and B cells measured before allo-HSCT).Table 3 Immunoglobulin levels and affected cell lineage in peripheral blood prior to HSCT Patient no Affected cell lineage (normal range) Immunoglobulins Time before HSCT (months) CD19 + , cells × 106/L (100–500) NK cells × 106 /L (100–400) CD3 +  T- cells × 106/L (800–2400) CD4 +  T- cells × 106/L (500–1400) CD8 +  T-cells × 106/L (200–2000) Monocytes  × 109/L (0.2–0.8) Neutrophils  × 109/L (1.5–7.3) IgG g/L (6.9–15.7) IgG2 g/L (1.69–15.7) 1 0 0 118 43 64 0.0 0.2 5.9 ND 5 2 10 10 550 200 300 0.3 2.9 9.0 ND 4.5 3 160 150 1930 780 950 0.2 2.4 8.4 ND NA 4  < 10 15 349 185 145 0.0 3.4 9.5 1,24 17 5 6 1 259 134 66 0.0 6.0 13.0 0.79 221 6 2 0 809 495 326 0.1 0.5 44.32 2.3 4 7 18 19 913 545 341 0.0 2.8 14.0 2.60 3 83 40 2 1335 546 761 0.1 1.9 9.7 ND NA 9 ND ND ND ND ND 0.0 4.2 16.2 - 1 10 70 206 247 134 108 0.1 1.7 18.5 0.79 6 11 ND ND ND ND ND 0.0 0.9 9.5 2.11 10 12 ND ND ND ND ND 1.7 2.6 8.1 ND 1 134 28 13 1073 633 417 0.2 0.3 12.8 ND 0.5 14 8 11 676 323 341 0.1 1.2 11.9 1.14 NA Abnormal values are given in bold 1On Prednisolone 20 mg a day when these samples were taken 2Hypergammaglobulinemia on IVIG due to IgG2 deficiency 3The values are from the time at diagnosis of GATA2 deficiency 3 years ago 4The reference values for Patient 13 who was 12 years old at the time of HSCT are CD19 200–600, NK 70–1200, CD3 800–3500, cd4 400–1200, cd8 200–1200 (given in cells × 106/L) and for IgG the normal reference value was 6.1–14.9 g/L ND, not done; NA, not applicable Germline GATA2 Variants and Somatic Variants in Other Genes Ten different GATA2 pathogenic, or likely pathogenic, variants were found in 14 patients (Table 2). All identified constitutional GATA2 variants, except one, were located in the ZNF2 domain, corresponding to or in close proximity to exons 5 and 6 (Table 2). Three nonsense variants (p.Ala342Glyfs*43, p.Arg362*, p.Asn381fs*20), one + 1 splicing variant, three missense variants (p.Thr354Met, p.Trp360Arg, and p.Ala372Thr), all previously reported to be disease-causing[14, 23–25], and two novel in-frame deletions (p.Thr358del, p.Asp367del) were found. The variant located in exon 3, outside the ZNF2 domain, was a nonsense variant (c.163C > T, p.Gln55*). It was initially identified by the NGS myeloid panel with variant allele frequency (VAF) 49% in the DNA from the patient’s BM and later verified to be germline with VAF 49% in a buccal DNA sample (Patient 9, Table 2). Patient 2 had a paternal inherited in-frame deletion, c.1062_1064del (p.Thr358del), in the ZNF2 domain, and a somatic in-frame deletion, c.1168_1170del (p.Lys390del), with a fairly high VAF, 40.2% in BM. As expected, these two in-frame deletions were no longer detectable after allo-HSCT. Surprisingly, in the first post-transplant BM sample at day + 28, we detected another acquired GATA2 variant, c.1215G > T (p.Lys405Asn) with VAF 49.5%. This missense mutation variant, affecting an amino acid located C-terminal to the ZNF2 domain, is a variant of unknown clinical significance. It is most likely a rare benign variant, which in retrospect was confirmed to be constitutional in the unrelated BM donor (Table 2). It is evaluated to ACMG category 3 minus, since altogether 5 heterozygote individuals with the same amino acid change p.Lys405Asn are reported in GnomAD (v.2.1.1)[31]. As far as we know, missense variants located outside the ZNF2 domain rarely represent constitutional susceptibility to development of myelodysplasia. One exception is the p.Ser447Arg located C-terminal of the ZNF2 domain[32], while no other missense variants outside the ZNF2 domain are currently defined as pathogenic or likely pathogenic in ClinVar (www.clinvar.com) as of year 2021. Karyotype abnormalities and somatic variants in other genes observed in the 14 patients are presented in Table 2. Trisomy 8 (n = 6), monosomy 7 (n = 3), STAG2 variants (n = 3), ASXL1 variants (n = 2), a combination of somatic variants in RUNX1/CSF3R/PHF6/FBXW7 (n = 1), and variants in the following MDS genes[33] were observed once in separate individuals: BCOR, SETBP1, U2AF1, and somatic GATA2. One adult GATA2 deficient patient who developed AML had an unbalanced translocation der(1;7) in the leukemic clone. Since GATA2 deficiency is considered the most common hereditary predisposition to pediatric MDS, we estimated the proportion of pediatric patients diagnosed with MDS in Norway that had a germline GATA2 variant in the same time period (2013–2020). We found that three out of 14 pediatric patients diagnosed with MDS (21%) had GATA2 deficiency. Of note, these are 14 pediatric MDS patients and not the same cohort of 14 GATA2 deficient patients described above (except three overlapping pediatric patients, Patients 2, 12, and 13, with MDS). Two of the 3 pediatric GATA2 deficient patients with MDS had both monosomy 7 and trisomy 8 in their bone marrow cells. Patients 4, 5, 8, 9, 10, and 11 from Family B, E, F, and G had frameshift or other definitive loss-of-function variants, while Patients 1, 2, 3, 6, 7, 12, 13, and 14 from Family A, C, D, H, I, and J had in-frame deletions or missense variants. No specific genotype–phenotype correlations were found in our data set, i.e., regarding debut of symptoms, type and distribution of infections, age of transition to MDS/AML, somatic occurring variants in blood and BM. The severity of the clinical presentations also varied within families. Families and Predictive Genetic Testing The pedigrees of the 10 families are presented in detail in Fig. 1. Patient 1 (Family A) had three apparently healthy children, when he was diagnosed with GATA2 deficiency. After genetic testing of first-degree relatives, we found that two of his children (Patients 2 and 3) had inherited the GATA2 variant. For Patient 2, initial clinical work-up revealed only mild cytopenia and warts. However, within 2 years of follow-up, he developed pancytopenia and transfusion dependency and was diagnosed with MDS-EB1. His sister, Patient 3, has warts as her only clinical manifestation, but will be followed up regularly for development of cytopenia/MDS.Fig. 1 Pedigrees of the ten families, including 14 patients, with known GATA2 deficiency. Solid symbols denote affected status. Individuals marked in gray are deceased and not tested for GATA2 deficiency but are suspected to carry the disease-causing variant. In family G, the mother of Patients 10 and 11 died at age 30 of acute respiratory distress syndrome, 27 years ago. She also had lymphedema since birth. In light of their mother’s medical history, the GATA2 variant is probably maternally inherited. The father is alive and healthy. In family J, the mother of Patient 14 had a combined B and T cell defect, warts, myelodysplastic syndrome, lymphedema, and recurrent respiratory tract infections. She died of vulval cancer at the age of 38. The maternal grandfather of Patient 14 died of acute leukemia at the age of 33. WT, wild-type In Family G, two siblings had the same germline GATA2 variant (Patients 10 and 11). Their mother died 27 years ago, at the age of 30, of acute respiratory distress syndrome (ARDS), of unknown etiology. She also had lymphedema since birth. In light of their mother’s medical history with lymphedema and ARDS, which could be secondary to complications related GATA2 deficiency, the GATA2 variant is probably maternally inherited. Their father is alive and healthy. The deceased mother of Patient 14 (Family J) had a combined B- and T cell defect, warts, MDS, lymphedema, and recurrent respiratory tract infections. At the age of 38 (years), she died of metastatic vulval cancer. The maternal grandfather of Patient 14 died of acute leukemia at the age of 33. Considering the family history, it is very likely that Patient 14 had inherited her germline GATA2 variant from her maternal grandfather via her mother. Both individuals died before GATA2 deficiency was acknowledged as a cause of PID. Her mother’s siblings are now offered genetic counselling/testing for GATA2 deficiency. Allo-HSCT in Patients with GATA2 Deficiency Twelve of 14 (86%) patients with GATA2 deficiency were found to have a clinical indication, cytogenetic findings, and/or molecular findings warranting to proceed to allo-HSCT. As of today, 11 patients have undergone allo-HSCT, whereas one is recently accepted for allo-HSCT (Patient 14). Clinical features that lead to the decision to perform allo-HSCT were previous life-threatening disseminated HSV infection (Patient 1), severe obliterating bronchiolitis and in situ carcinoma (Patients 4 and 5), MDS with cytogenetic abnormalities (monosomy 7) and/or excess of blasts with high likelihood of progression to leukemic transformation (Patients 2, 6, 7, 9, 12, 13, and 14), MDS and warts with high-grade dysplasia (Patient 10), and symptoms of severe immunodeficiency and MDS (Patient 11). Details on the allo-HSCT procedure, including conditioning, donor selection, stem cell source, donor/recipient cytomegalovirus status, donor chimerism, graft versus host disease (GVHD) prophylaxis, and the occurrence of GVHD, are presented in Table 4.Table 4 HSCT details for eleven patients with GATA2 deficiency Patient no Age at HSCT Donor Stem cell source HLA Match CD34 + , × 106/kg CMV status d/r Conditioning regimen & In vivo T-cell depletion Chimerism %Day + 28 GVHD prophylaxis GVHD Complications 1 41 y MUD PBSC 10/10 (11/12) 7.8 -/ +  RIC: Fludarabine 150 mg/ m2, Busulfan 8 mg/kg, ATG Thymoglobulin 4 mg/kg 99% Mtx + CsA No Hemorrhagic cystitis (BK-virus) 2 16 y MUD PBSC 10/10 (10/12) 9.7 -/ - MAC*: Busulfan for 4 days (TDM; Css 825 ng/ml), Cyclophosphamide 120 mg/kg, Melphalan 140 mg/m2, ATG Grafalon 3 × 10 mg/kg  > 99% Mtx + CsA No E. coli sepsis; BK-virus cystitis; mucositis (grade 3) 4 39 y MUD PBSC 10/10 (12/12) 10.6 + / +  RIC: Fludarabine 150 mg/ m2, Busulfan 8 mg/kg. ATG Thymoglobulin 4 mg/kg  > 99% Mtx + CsA Chronic: skin and gut CMV reactivation 5 39 y MUD PBSC 10/10 (11/12) 5.2 +/ - RIC: Fludarabine 90 mg/m2, 2 Gy TBI 99% Mtx + CsA No Enterococcus faecalis sepsis (2 months post-HSCT), prolonged cytopenia, died of intracerebral hemorrhage 8 months post-HSCT 6 29 y MRD PBSC HLA-id sibling 5.4 + / +  RIC: Fludarabine 150 mg/m2, Busulfan 8 mg/kg 98% Mtx + CsA Chronic: liver, oral mucosa and genitalia Cytopenia at day + 33, osteoporosis, compression fractures 7 22 y MUD PBSC 10/10 6.9 -/ - MAC: Fludarabine 150 mg/m2, Treosulfan 42 g/m2, ATG Thymoglobulin 4 mg/kg  > 99% Mtx + CsA Chronic (limited): skin Moraxella nonliquefaciens sepsis on day 0. E. coli sepsis day + 14. Oral mucositis grade IV. PTLD 7 weeks post-HSCT 9 23 y MUD PBSC 10/10 (11/12) 4.3 +/ - MAC: Fludarabine 160 mg/m2, Busulfan 12,8 mg/kg (i.v.), ATG Thymoglobulin 4 mg/kg  > 99% Mtx + CsA Acute GvHD grade I: skin None 10 32 y MUD PBSC 10/10 (11/12) 5.9 -/ - MAC: Fludarabine 150 mg/m2, Treosulfan 42 g/m2, ATG Thymoglobulin 4 mg/kg  > 99% Mtx + CsA Acute GvHD: serositis None 11 27 y MUD PBSC 10/10 (10/12) 10.2 -/ - MAC: Cyclophosphamide 100 mg/kg, Busulfan 16 mg/kg N.a Mtx + CsA Chronic (extensive): gut, eye, and lung Hemorrhagic cystitis, Herpes oesophagitis 12 14 y MUD BM 10/10 (11/12) TNC: 3.5 × 108/kg -/ - MAC*: Fludarabine 160 mg/m2, Treosulfan 42 g/m2,Thiotepa 8 mg/kg, ATG Grafalon 3 × 10 mg/kg day + 84: > 99% Mtx + CsA No Impetigo day + 40 13 11 y MMFD (father) PBSC, TCRab + /CD19 + depletion in vitro Haploidentical 10.3 + / +  MAC*: Fludarabine 160 mg/m2, Thiotepa 10 mg/kg, Melphalan 140 mg/m2, ATG Grafalon 3 × 10 mg/kg  > 99% MMF (until day + 28) No None Abbreviations: BM, bone marrow; CMV, cytomegalovirus; CsA, cyclosporine A; Css, concentration at steady-state; Cya, cyclosporine A; GVHD, graft versus host disease; i.v., intravenous; MAC, myeloablative conditioning; MMF, mycophenolate mofetil; MMFD, mismatched family donor; MRD, matched related donor; Mtx, methotrexate: MUD, matched unrelated donor; N.a., not available; PBSC, peripheral blood stem cells; RIC, reduced-intensity conditioning; TDM, therapeutic drug monitoring; Tx, transplantation *According to HSCT recommendations by the EWOG-MDS study group Clinical Outcome The clinical outcome of all 14 patients is presented in Table 1. Two adult patients (18%) died after allo-HSCT. Patient 5 had persistent thrombocytopenia and died of a cerebral hemorrhage 8 months post transplantation. Patient 11 developed respiratory failure due to cGVHD in the lungs and received a bilateral pulmonary transplant 5 years post allo-HSCT. However, he developed chronic pulmonary rejection and died 2 years after lung transplantation and 7 years after allo-HSCT. The mean follow-up of the nine patients still alive after allo-HSCT is 26 months (range 3–78 months). The incidence of aGvHD and cGVHD among the eleven transplanted patients was 25% and 33%, respectively, all occurring in patients > 18 years of age (Table 4). None of the pediatric patients had experienced aGVHD or cGVHD, serious infectious complications, or any serious or unexpected transplant-related acute or late toxicity. Their transplantation courses were uneventful and did not principally differ from MDS patients without germline disease-causing GATA2 variants. One patient is listed for allo-HSCT (Patient 14) and two patients are followed regularly in the out-patient clinic (Patients 3 and 8). Discussion This retrospective study describes clinical features and outcome of 14 patients from ten families diagnosed with GATA2 deficiency in Norway. The main findings were as follows: (i) We found a diverse clinical phenotype dominated by cytopenia (13/14), myeloid neoplasia (MDS/AML) (10/14), warts (8/14), and hearing loss (7/14). (ii) We observed two novel clinical features multiple aneurysms of small vessels (n = 1) and early graying (n = 1) that could be associated with GATA2 deficiency. (iii) The majority of patients (11/14) had already undergone allo-HSCT at the time of our analysis, illustrating the need for allo-HSCT in a large proportion of patients with GATA2 deficiency in Norway, and most likely in other countries. (iv) Genetic testing should be offered to first-degree relatives, particularly children, to identify individuals with GATA2 deficiency that need close surveillance. Genetic testing performed as soon as a clinical suspicion is raised, increases the likelihood of an early and correct molecular diagnosis. In Norway, exome-based panels including GATA2 are offered as a routine diagnostic laboratory service for constitutional hematological disorders and PID [21]. However, this approach will not detect all GATA2 variants. Germline variants located in the intronic transcriptional enhancer elements, the cis-acting E-box/GATA and ETS motifs within intron 5 (NM_001145661.1), may cause GATA2 deficiency[34]. Supplementary Sanger sequencing of the enhancer element sequences in intron 5[34] was only performed in one of the laboratories (see Supplementary Methods). Despite supplementary copy number variant calling from exome data, with chromosomal microarray and MLPA only in selected patients (Supplementary Methods), some structural variants may go undetected. In addition, synonymous disease-causing GATA2 variants resulting in selective loss of mutated RNA were recently reported [35]. Disease-causing intronic variants, small intragenic variants, such as structural variants and synonymous variants, may also have escaped detection and hence, individuals with GATA2 deficiency may have been overlooked. With the recent introduction of targeted NGS panels in the work-up of myeloid neoplasms searching for somatic GATA2 variants, unexpected germline variants can be identified, which may reveal the underlying constitutional cause of the myeloid disease and increase the prevalence of known GATA2 deficiency. Improving the NGS panels targeting both germline and somatic variants by deeper coverage of the whole genes including important non-exonic regions, better algorithms for detection of structural variants, and attention to rare synonymous variants may enhance the identification of GATA2 deficiency. In this case series, we found two clinical features that have not yet been described in GATA2 deficiency, namely small vessel aneurysm and early graying. Multiple small vessel aneurysms found in Patient 10 may be secondary to vasculitis, or could also represent a novel vascular feature associated with GATA2 deficiency. In fact, it has been suggested that alteration in GATA2 expression may be of importance for vascular integrity[36]. The observation of premature graying may be a coincidence. In the absence of telomere biology disorders as in the present patients, one may speculate if this feature may reflect a GATA2-linked autoimmune phenomenon which has gone undetected. One of the aims of this retrospective study was to describe the clinical characteristics, GATA2 variants, and other molecular variants representing risk factors for clonal evolution that could aid us in the difficult decision regarding: “Who and when to transplant”? The high proportion of patients (79%) that had already undergone allo-HSCT in our cohort was somewhat surprising. Donadieu et al. have published the largest cohort of patients with GATA2 deficiency, and found that only 28 patients (35%) of 79 patients had undergone allo-HSCT. However, this low percentage of allo-HSCT did not correspond with the severity of the disease in this cohort. At the age of 40, the authors reported a mortality rate of 35% and a hematological malignancy rate of 80%[14]. The high proportion of allo-HSCT in our study may have been influenced by increased awareness of negative prospective clinical markers of GATA2 deficiency. Based on the findings from our study and the high morbidity and mortality rate reported by Donadieu et al., it is clear that these patients need to be monitored closely. Ideally, allo-HSCT should be performed before they develop malignancies (both solid tumors and hematological malignancies)[37] or severe/recurrent infections causing organ failure. In our opinion, a history of disseminated viral infection, aggressive HPV infection (particular with dysplasia), or myeloid clonal disease is clear indication to consider allo-HSCT[14]. First-degree relatives with a severe outcome of the disease may further strengthen the indication for an early allo-HSCT in symptomatic patients with GATA2 deficiency. Overall, the decision to perform an allo-HSCT requires careful weighing of potential gain (restore immune function; diminish the risk of hematological malignancies) versus possible transplant complications, including GVHD and transplant-related mortality. This is particularly challenging given the lack of genotype–phenotype correlation. Keeping in mind that the observation time is short for some of the patients in our study, the survival rate after allo-HSCT was 82% (9/11). In patients with GATA2 deficiency, previous publications have reported 86% survival 2 years after HSCT (n = 22)[16], 73% and 62% survival 1 and 5 years after HSCT, respectively (n = 28)[14], 72%, 65%, and 54% survival 1, 2, and 4 years after HSCT, respectively (n = 21)[9], and 57% 3, 5 years after HSCT (n = 14)[38]. These cohorts are, however, not necessarily comparable in terms of severity of disease and conditioning regimen. Two children were diagnosed with GATA2 deficiency after family screening. The hematological surveillance of one of these children led to detection of hematological abnormalities consistent with MDS and, in the end, a timely allo-HSCT. We therefore recommend genetic testing of children of affected adults and hematological surveillance of individuals with known pathogenic germline GATA2 variants. This includes annual BM investigations with morphological and cytogenetic evaluations, and testing with NGS myeloid panel to screen for somatic occurring molecular drivers of malignancies. Monosomy 7 and trisomy 8 have been reported by others to be the major cytogenetic aberrations in hematopoietic cells of patients with GATA2 deficiency and MDS[15]. Advanced MDS disease and monosomy 7 have been related to worse outcome, especially for pediatric patients with GATA2 germline disease[20]. We found monosomy 7 only in ¼, but trisomy 8 in half of the karyotyped patients, which is in line with a previous published study by McReynolds et al. [39]. Two patients in our GATA2 cohort had monosomy 7 and trisomy 8, both were children. Some of the other somatic variants in our cohort occurred in genes previously reported to be mutated in GATA2 deficiency with MDS, such as ASXL1[40], STAG2, SEPTBP1, and RUNX1[41, 42]. Interestingly, one adult patient with GATA2 deficiency and MDS-related AML had a der(1;7) in the leukemic clone, a translocation that has recently been shown to be enriched in pediatric MDS patients with germline GATA2 mutations[43]. However, our number of patients are too small to determine possible genotype–phenotype correlations related to clonal disease progression. Previous reports have suggested that plasma levels of FLT3LG can be used as predictor of hematological disease in GATA2 deficiency, and used in clinical monitoring post-HSCT [25]. Unfortunately, we lack serum or plasma samples taken before and after HSCT in our cohort, and the role of FLT3LG as a disease progression marker could be explored in future studies. Our main conclusion of this study is that the majority of patients with symptomatic GATA2 deficiency will need allo-HSCT, and close surveillance of these patients is important to find the “optimal window” for allo-HSCT. We advocate a more offensive approach to allo-HSCT than previously described. Supplementary Information Below is the link to the electronic supplementary material.Supplementary file1 (DOCX 55 KB) Acknowledgements Our gratitude goes to the Section for Cancer Cytogenetics who karyotyped 9 of the 14 patients reported here (Table 2). Three of the reported families have been followed up at the Department of Medical Genetics, University Hospital of North Norway. We want to thank our colleagues, MD Gry Hoem, MD Hilde Yttervik, MD Specialist in medical genetics and pediatrics Marie Falkenberg Smeland, and MD PhD Øyvind Holsbø Hald at the Department of Medical Genetics, University Hospital of North Norway, Tromsø, Norway, for their clinical work and genetic counselling. The three pediatric patients are enrolled in the registry of the European Working Group of MDS in Childhood (EWOG-MDS; ClinicalTrials.gov Identifier: NCT00662090). Author Contribution SFJ, JB, AEM, PA, AS-P, TGD, and IN wrote the paper. The genetic analyses were performed by AS-P, MAK, HS, ØH, SS, and EL. SFJ, JB, AEM, EG, YF, CA, AB, IH, TF, BF, AS-P, TGD, and IN collected clinical data. SS did the bone marrow analyses. All authors critically revised the manuscript for important intellectual content and approved the final version of the manuscript. Funding Open access funding provided by University of Oslo (incl Oslo University Hospital). Data Availability Upon request. Code Availability Not applicable. Declarations Ethics Approval Five of the patients had previously consented to be part of a genetic PID research project approved by the regional ethical committee (REC. 2014/1270–1), three patients were diagnosed with MDS in childhood and consented to be registered into the EWOG-MDS-2006 study (2015/1651/REC Nord), and all adult patients who underwent allo-HSCT had consented to publication of data (REC 11909). Studies were performed according to the Declaration of Helsinki. Consent to Participate All adult living patients signed a written informed consent for publication. For children < 18 years, consent was given by their parents. Consent for Publication Obtained. Conflict of Interest The authors declare no competing interests. Abbreviations aGVHD Acute graft versus host disease allo-HSCT Allogeneic hematopoietic stem cell transplantation AML Acute myelogenous leukemia ARDS Acute respiratory distress syndrome BM Bone marrow cGVHD Chronic graft versus host disease GVHD Graft versus host disease HPV Human papilloma virus MDS Myelodysplastic syndrome MRD Matched related donor MUD Matched unrelated donors NGS Next-generation sequencing PID Primary immunodeficiency VAF Variant allele frequency WES Whole exome sequencing ZNF1 Zink finger 1 ZNF2 Zink finger 2 Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Asbjørg Stray-Pedersen, Tobias Gedde-Dahl and Ingvild Nordøy contributed equally

BUSULFAN, CYCLOSPORINE, FLUDARABINE PHOSPHATE, METHOTREXATE

DrugsGivenReaction

CC BY

34893945

2021-12-10

Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Osteoporosis'.

A Nationwide Study of GATA2 Deficiency in Norway-the Majority of Patients Have Undergone Allo-HSCT. OBJECTIVE GATA2 deficiency is a rare primary immunodeficiency that has become increasingly recognized due to improved molecular diagnostics and clinical awareness. The only cure for GATA2 deficiency is allogeneic hematopoietic stem cell transplantation (allo-HSCT). The inconsistency of genotype-phenotype correlations makes the decision regarding "who and when" to transplant challenging. Despite considerable morbidity and mortality, the reported proportion of patients with GATA2 deficiency that has undergone allo-HSCT is low (~ 35%). The purpose of this study was to explore if detailed clinical, genetic, and bone marrow characteristics could predict end-point outcome, i.e., death and allo-HSCT. METHODS All medical genetics departments in Norway were contacted to identify GATA2 deficient individuals. Clinical information, genetic variants, treatment, and outcome were subsequently retrieved from the patients' medical records. RESULTS Between 2013 and 2020, we identified 10 index cases or probands, four additional symptomatic patients, and no asymptomatic patients with germline GATA2 variants. These patients had a diverse clinical phenotype dominated by cytopenia (13/14), myeloid neoplasia (10/14), warts (8/14), and hearing loss (7/14). No valid genotype-phenotype correlations were found in our data set, and the phenotypes varied also within families. We found that 11/14 patients (79%), with known GATA2 deficiency, had already undergone allo-HSCT. In addition, one patient is awaiting allo-HSCT. The indications to perform allo-HSCT were myeloid neoplasia, disseminated viral infection, severe obliterating bronchiolitis, and/or HPV-associated in situ carcinoma. Two patients died, 8 months and 7 years after allo-HSCT, respectively. CONCLUSIONS Our main conclusion is that the majority of patients with symptomatic GATA2 deficiency will need allo-HSCT, and a close surveillance of these patients is important to find the "optimal window" for allo-HSCT. We advocate a more offensive approach to allo-HSCT than previously described. pmcIntroduction GATA2 deficiency is a rare primary immunodeficiency (PID), first described in 2011[1–3] that has become gradually more recognized due to improved molecular diagnostics and increased clinical awareness. GATA2, as a “master” transcription factor, plays a critical role in hematopoietic development[4]. Through cooperative processes that include other transcription factors, it controls the transition from hemogenic endothelium to hematopoietic stem cells and is required for survival and self-renewal of these cells[5]. GATA2 is also important for other tissue-forming stem cells, e.g., in the inner ear[6]. The heterozygous variants causing GATA2 deficiency are located both in coding, non-coding and enhancer regions[7]. The disease-causing loss-of-function variants can be localized across the gene. These variants can lead to defective DNA-binding capacity of the transcription factor and may cause disease through haploinsufficiency of the functional protein[5, 8]. Missense variants within the zink finger 2 (ZNF2) domain are the most frequent germline disease-causing GATA2 variants [9]. It has been estimated that approximately 1/3 of the patients have an autosomal dominant inherited disease-causing variant[10], whereas the remaining have a de novo GATA2 variant[7]. Of note, somatic variants in GATA2 are known to be drivers of myeloid neoplasia in adults. Such variants are diverse, may cause gain-of-function effects, and be located across the whole gene. This includes missense variants in the zink finger 1 (ZNF1) domain, which has not been observed in constitutional GATA2 deficiency[8]. Typically, GATA2 deficiency becomes clinically apparent in late childhood to early adulthood. The phenotype is heterogeneous, without any clear genotype–phenotype correlation, and with an incomplete clinical penetrance[11]. Symptoms may include recurrent or severe infections, warts, cytopenia (including monocytopenia), lymphedema, alveolar proteinosis, and malignant myeloid disease[9]. Infectious complications in GATA2 deficiency are likely due to deficiency of monocytes, NK cells, and B-lymphocytes as well as defective innate immune responses, including impaired type I interferon production[12]. This leads to both increased susceptibility to viral infections (e.g., human papilloma virus [HPV, warts] and herpes virus infections), non-tuberculous mycobacteria, and to more common bacterial respiratory infections. Hearing loss is a common clinical feature of GATA2 deficiency and is related to the critical role of GATA2 in vestibular morphogenesis of semicircular ducts and generation of the perilymphatic space around the inner ear’s semicircular canals[6, 13]. A substantial proportion of patients develop immunodeficiency, myelodysplastic syndrome (MDS), or acute myelogenous leukemia (AML) as initial manifestation[9, 14]. GATA2 deficiency is considered the most common hereditary predisposition to pediatric MDS[15]. The only cure for GATA2 deficiency is allogeneic hematopoietic stem cell transplantation (allo-HSCT) and results are encouraging[16–20]. However, the main challenge is deciding who and when to transplant due to the complexity and inconsistency of phenotype-genotype correlation in GATA2 deficiency[9]. To further elucidate this important issue, we present detailed clinical and molecular characteristics, treatment, and outcome of 14 Norwegian patients with germline GATA2 variants diagnosed between 2013 and 2020. The main aim of our study was to explore if detailed clinical, genetic, and bone marrow (BM) characteristics could predict end-point outcome such as death and allo-HSCT in patients with GATA2 deficiency. Methods Identification of Patients and Clinical Characteristics The first aim of this study was to obtain a complete overview of all patients with known GATA2 deficiency in Norway. For this purpose, a network of clinical immunologists, hematologists, pediatricians, and geneticists at Oslo University Hospital (OUH) collected clinical and laboratory data on patients with GATA2 deficiency at their institution. In addition, all medical genetics departments in Norway were contacted to identify any additional GATA2 deficient individuals. Clinical information, genetic variants, treatment, and outcome were subsequently retrieved from the patients’ medical records. Patients were enrolled into the study at OUH where most of the data was obtained, while supplemental data from Patient 3 was collected at the University Hospital of North Norway, Tromsø. Informed Consent Five of the patients had previously consented to be part of a genetic PID research project approved by the regional ethical committee (REC. 2014/1270–1), three patients were diagnosed with MDS in childhood and consented to be registered into the EWOG-MDS-2006 study (2015/1651/REC Nord), and all adult patients who underwent allo-HSCT had consented to publication of data (REC 11,909). In addition, due to the detailed clinical information published herein, all adult living patients signed an additional written informed consent for publication of their data and was given the opportunity to review the manuscript prior to publication. For children < 18 years, consent was given by their parents. This is in line with the recommendation given by the Ethical Constitutional board at OUH. Genetic Analyses Whole exome sequencing (WES) with in silico filtering for genes causing primary immunodeficiency disorders was performed in the probands and affected relatives as part of a routine laboratory service (Patients 2, 3, 8, 9, 10, 12, and 14) or on a research basis (Patients 4, 5, 6, and 7) as previously described (Supplemental methods)[21]. Patient 1 had severe cytopenia (Table 1), and the first attempt to extract DNA from peripheral blood was not successful. A skin biopsy was therefore performed to extract DNA from fibroblasts. In parallel, peripheral blood (from puncture of the fingertip) was applied directly to a Guthrie filter card, and by using multiple filter card punches, enough DNA was extracted to run next-generation sequencing (NGS) with an amplicon-based targeted panel for constitutional variants in PID genes (Supplemental methods). By using this rapid amplicon-based method, the molecular result was available within 3 working days[22]. DNA later extracted from fibroblasts confirmed the GATA2 variant by Sanger sequencing. Also, for Patient 13, who had advanced MDS with pancytopenia, the NGS results were available within 3 working days, with parental testing performed in parallel to evaluate as fast as possible the availability of a healthy unaffected matched related donor (MRD).Table 1 Clinical characteristics and outcome in patients with GATA2 deficiency Patient no Family Sex Current age Age at onset of symptoms/age at genetic diagnosis Infections Hearing loss Hematologic abnormalities Autoimmunity/immune dysregulation Miscellaneous HSCT, age Outcome Viral Bacterial 1 A (father of P2 and P3) M 44y 5y/41y HPV: warts HSV: disseminated disease Ear infections as a child Yes Hypoplastic BM: cytopenia, trilinear hypoplasia No No 41y Alive 2 A (son of P1) M 16y 7y/14y HPV: warts No No MDS-EB-1 No No 16y Alive 3 A (daughter of P1) F 13y 8y/9y HPV: warts No No No No No ND Alive 4a B (monozygotic twin to P5) F 45y 21y/38y HPV: warts, carcinoma in situ EBV: prolonged viremia Recurrent respiratory inf Suppurative skin infection after BCG * Yes No Progressive obliterating bronchiolitis, lupus-like syndrome Miscarriage 39y Alive 5a B (monozygotic twin to P4) F † (39y) 24y/38y HPV: warts, carcinoma VZV and EBV: prolonged viremia Recurrent respiratory inf Suppurative skin inf. after BCG* Yes MDS-MLD (hypoplastic) Progressive obliterating bronchiolitis, lupus-like syndrome DVT × 2 Squamous cell carcinoma in the cervix, rectum, and anus 39y Deceased 8 m post-HSCTb 6a C M 31y 11y/26y No Recurrent respiratory inf Yes MDS-MLD (hypoplastic) - Fever of unknown origin, recurrent pneumothorax 29y Alive 7a D F 23y 6y/17y HPV: warts Recurrent respiratory inf No MDS-MLD (hypoplastic) Interstitial lung disease Lymphedema, acne, rosacea, rash, fatigue 22y Alive 8 E F 56y 0y/53y No No No Hypoplastic BM No Lymphedema, premature graying ND Alive 9 F F 24y 15y/23y No No No AML with MDS-related changes Erythema nodosum DVT, PE, juvenile myoclonic epilepsy, epicanthic fold 23y Alive 10 G (sibling to P11) F 32y 6y/31y HPV: warts, cervix dysplasia Recurrent respiratory inf Yes MDS-MLD No Aneurysm of small vessels, hidradenitis suppurative, liver lesions: focal nodular hyperplasia 32y Alive 11 G (sibling to P10) M † (34y) 22y/PM No Recurrent skin and respiratory inf No MDS-MLD No Acne, rosacea, necrotizing fasciitis, pilonidal cysts, skin infections, ulcerations 27y Deceased 7y post-HSCTc 12 H F 19y 14y/14y No No Yesd MDS-RCC (hypoplastic) BPD/Asthma Born premature (week 26 + 5), BPD 14y Alive 13 I M 13y 9y/11 y HPV: warts No No MDS-EB1 Asthma Chronic skin abscesses, congenital ptosis 11y Alive 14 J F 31 23y/31y No No Yes MDS-SLD (hypoplastic) No Born prematurely (week 25), cerebral palsy, congenital hip dysplasia Planned Alive Abbreviations: AML, acute myeloid leukemia; BCG, bacille Calmette Geurin; BM, bone marrow; BPD, bronchopulmonary dysplasia, CT, computer tomography, HPV, human papilloma virus; HSCT, hematopoietic stem cell transplantation; Inf., infection; m, months; MDS, myelodysplastic syndrome; MDS-EB1, MDS with excess of blasts type 1; MDS-MLD, MDS with multilineage dysplasia; MDS-SLD, MDS with single lineage dysplasia; MDS-RCC, MDS subtype refractory cytopenia of childhood; ND, not done, PM, post mortem; VTE, venous thromboembolism; y, years, †; deceased aThese patients have previously been published in Stray-Pedersen, Sorte et al. 2016 (Patient 4 was 84.1, Patient 5 was 84.4, Patient 6 was 88.1, and Patient 7 was 86.1)[21] bThe patient was doing well after HSCT, but died unexpectedly of a cerebral hemorrhage cThe patient underwent lung transplantation for chronic lung GVHD 58 months after HSCT, and died of chronic lung rejection 26 months after bilateral lung transplantation dThe patient has reduced hearing, but this was confirmed after HSCT. Her hearing loss may be due to the disease-causing GATA2 variant, but may also be secondary to complications of HSCT therapy, e.g., aminoglycosides The molecular diagnosis in Patient 11 was confirmed post mortem using a BM sample collected prior to allo-HSCT (Table 2). Methods for testing for somatic occurring sequence variants on DNA extracted from whole blood or BM, and methods for testing chromosomal aberration on BM cells are described in Supplemental methods.Table 2 Constitutional and acquired genetic findings in patients with GATA2 deficiency Patient no Hematological abnormalities Constitutional heterozygous variants in GATA2, NM_001145661.1, predicted protein effect, domain, occurrence, novelty, and reference Somatic variants, predicted protein effect, VAF in BM/blood (prior to HSCT) Karyotype in BM (closest to HSCT)d  + 8  − 7 1 Hypoplastic bone marrow c.1062_1064del, p.Thr358del, in-frame exon 6, ZNF2, likely de novo, novel variant Unknown 46,XY[25/25] No No 2 MDS-EB1 c.1062_1064del, p.Thr358del, in-frame exon 6, ZNF2, paternal inherited, novel variant Acquired germline donor variant in GATA2 Post-HSCTb: c.1215G > T, p.(Lys405Asn) missense exon 7, outside and distal to ZNF2 domain, VAF: 49,5% BM NM_001145661.1 (GATA2): c.1168_1170del, p.(Lys390del), in-frame exon 7, in ZNF2, VAF: 40.2% BM NM_006758.2(U2AF1): c.470A>G, p.(Gln157Arg) VAF: 44,0% BM 46,XY,-7 + 8[20/20] Yes Yes 3 No c.1062_1064del, p.Thr358del, in-frame exon 6, ZNF2, paternal inherited, novel variant Unknown Unknown N.a N.a 4 a No c.1143 + 5G > A, p.Asn381fs*20, splice defect intron 6, ZNF2, de novo, novel, reported by usc)[21] Unknown Unknown N.a N.a 5 a MDS-MLD (hypoplastic) c.1143 + 5G > A, p.Asn381fs*20, splice defect intron 6, ZNF2, de novo, novel, reported by usc[21] None 46,XX[18/18] No No 6a MDS-MLD (hypoplastic) c.1078 T > A, p.Trp360Arg, missense exon 6, ZNF2, de novo, variant previously reported by others[23] Unknown 46,XY[25/25], but FISH MYC(8q24): + 8 in 14/303 metaphases Yes No 7a MDS-MLD (hypoplastic) c.1061C > T, p.Thr354Met, missense exon 6, ZNF2, de novo, but a recurrent GATA2 variant[21, 24, 25] NM_001042749.2(STAG2): c.2534-2A > G, predicted splice variant with loss of acceptor site, Chr.X, VAF: 11.7% blood 47,XX, + 8[4/10]/46,XX[6/10] Yes No 8 Hypoplastic bone marrow c.1017 + 1G > T, loss of donor splice site, splice defect intron 5, ZNF1, both parents deceased and not tested, novel variant Unknown 46,XX[25/25] No No 9 AML with MDS-related changes c.163C > T, p.Gln55*, nonsense exon 3, TAD domain, likely de novo (see pedigree), novel variant VAF:48.7% in BM, 49,4% in buccal swap NM_001754.4(RUNX1): c.593A > G,p.(Asp198Gly) VAF:15% BM NM_156039.3(CSF3R): c.2326C > T, p.(Gln776*) VAF: 12.5% BM NM_032458.2(PHF6): c.309C > G, p.(Tyr103Ter) VAF:12.0% BM NM_033632.3(FBXW7): c.1513C > T, p.(Arg505Cys), VAF:11.7% BM 46,XX, der(1;7)(q10;p10), + 1[11/20]/46,XX [9/20] No No 10 MDS-MLD c.1084C > T, p.Arg362*, nonsense exon 6, ZNF2, likely inherited, variant previously reported by others[15, 26–28] NM_001123385.1(BCOR): c.529_530del, p.(Ser177ProfsTer8), VAF: 23.0% BM 49∼50,XX, + 6, + 8, + 21? + 21[cp7/8]/46,XX[1/8] Yes No 11 MDS-MLD c.1084C > T, p.Arg362*, nonsense exon 6, ZNF2, likely inherited, variant previously reported by others[15, 26–28] NM_015338.5(ASXL1): c.2324 T > G, p.(Leu775Ter) VAF: 20.5% BM NM_001042749.1(STAG2): c.2990 T > A, p.(Leu997Ter) VAF: 9.6% BM 47∼48,XY, + 8[10/15],der(16)t(1;16)(q21;q24[10/15], + der(16)t(1;16)[1/15], + 21[6][cp11/15]/46,XY[3/15] Trisomy 8, evolving to unbalanced 1;16 translocation and later Trisomy 21 Yes No 12 MDS-RCC c.1098_1100delGGA, p.Asp367del, in-frame exon 6, ZNF2, de novo, novel variant None 46,XX,-7, + 8[15/20] Yes Yes 13 MDS-EB1 c.1021_1024insGCCG, p.Ala342Glyfs*43, frameshift exon 6, ZNF1 de novo, variant previously reported[29] NM_015338.5(ASXL1): c.1854dupA, p.(Ala619SerfsTer16), VAF:17.0%, BM NM_015559.2 (SETBP1): c.2612 T > C, p.(Ile871Thr), VAF: 16.3%, BM 45,XY,-7[12/12] No Yes 14 MDS-SLD (hypoplastic) c.1114G > A, p.(Ala372Thr), missense exon 6, ZNF2, variant previously reported [14] NM_001042749.1(STAG2): c.707del; p.(Asn236IlefsTer20) VAF: 5.1%, BM 46,XX[25/25] No No Abbreviations: AML, acute myeloid leukemia; BM, bone marrow; Chr, chromosome; HSCT, hematopoietic stem cell transplantation; MDS, myelodysplastic syndrome; MDS-EB1, MDS with excess blasts type 1; MDS-MLD, MDS with multilineage dysplasia; MDS-RCC, MDS subtype refractory cytopenia of childhood; MDS-SLD, MDS with single lineage dysplasia; N.a., Not applicable; TAD, N-terminal transactivation domain, ZNF2, Zinc finger 2 domain in GATA2 protein; VAF, variant allele frequency aThese patients have previously been published in Stray-Pedersen, Sorte et al. (2016) (Patient 4 was 84.1, Patient 5 was 84.4, Patient 6 was 88.1, and Patient 7 was 86.1)[21] bThis disease-related GATA2 variant was detected in a routine BM at day + 28 post-HSCT; it turned out to be donor-derived (from a MUD) cWES identified a potential splicing variant in GATA2 (c.1143 + 5G > A) in Patient 4. The variant was predicted (Alamut®) to inactivate the donor site of GATA2 exon 5. PCR amplification of GATA2 exon 4 to 7 on cDNA showed that most transcripts were normally spliced resulting in a main product of ~ 400 bps, as seen in the normal control. A slightly longer PCR product including 64 bps of intron 6 sequence via a cryptic donor site in intron 6 (NM_001145661.1), was observed in the sample from Patient 4, but not in the control (see Supplementary information). Sanger sequencing identified the GATA2 splicing variant in the proband’s deceased monozygotic twin (Patient 5). Details described in Supplemental Figure E6 in Stray-Pedersen, Sorte et al. (2016)[21] dNomenclature according to ISCN (The International System for Human Cytogenetic Nomenclature) 2020 guidelines Results Characteristics of Patients Between 2013 and 2020, ten index cases, or probands, and four additional symptomatic patients with germline GATA2 variants were identified (9 females, 5 male, Table 1). Five adult patients were diagnosed by infectious disease specialists (Patients 1, 4, 5, 10, and 11) where infections (mostly HPV infection/warts and recurrent bacterial airway infections) were prominent symptoms. Four additional adult patients were identified by hematologists (Patients 6, 8, 9, and 14), where three were referred with pancytopenia and one patient had AML (also with pancytopenia). Three patients were diagnosed by pediatricians, two patients with MDS (Patients 12 and 13) and one patient with extensive warts and NK-/B-cell deficiency (Patient 7). Additionally, two patients with GATA2 deficiency were identified after family screening (Patients 2 and 3). We did not detect any asymptomatic individuals with GATA2 deficiency in this study. The mean age for debut of symptoms, that we considered related to GATA2 deficiency, was 12 years (range 0–24 years, Supplemental Table S1). The median time from these symptoms to a diagnosis of GATA2 deficiency was 11 years, range 0–53 years (Supplemental Table S1). Retrospectively, hearing loss, warts, and skin manifestations were the most common early symptoms, which in some patients became apparent many years before the genetic diagnosis of GATA2 deficiency was made (Supplemental Table S1). A summary of the patients’ clinical characteristics is given in Table 1. Viral infections such as HPV-associated warts were common, affecting eight patients. In addition, two patients had disseminated BCG infections (after vaccination), and one patient had a life-threatening disseminated HSV infection (originating from genitalia and disseminating to CNS and liver). Two patients experienced prolonged EBV and/or Varicella zoster viremia. Furthermore, six patients had recurrent bacterial airway infections. Interestingly, one patient had early graying (Patient 8), with normal telomere length, and one patient had multiple aneurysms of small vessels (coronary arteries, axillary arteries, and an iliac artery; Patient 10), which both represent clinical characteristics not previously described in GATA2 deficiency. In Patient 10, Varicella zoster infection was excluded as a cause of vasculitis with negative VZV PCR in blood. In addition, two patients had obliterating bronchiolitis (Patients 4 and 5), which has been reported in only one previous patient with GATA2 deficiency [30]. Affected cell lineages and immunoglobulin levels prior to allo-HSCT are listed in Table 3. As expected, the majority of patients had decreased levels of monocytes (11/14) and one patient had increased levels of monocytes (Patient 12). In addition, decreased levels of B cells (10/11) and NK cells (9/11) were common findings (three patients did not have NK- and B cells measured before allo-HSCT).Table 3 Immunoglobulin levels and affected cell lineage in peripheral blood prior to HSCT Patient no Affected cell lineage (normal range) Immunoglobulins Time before HSCT (months) CD19 + , cells × 106/L (100–500) NK cells × 106 /L (100–400) CD3 +  T- cells × 106/L (800–2400) CD4 +  T- cells × 106/L (500–1400) CD8 +  T-cells × 106/L (200–2000) Monocytes  × 109/L (0.2–0.8) Neutrophils  × 109/L (1.5–7.3) IgG g/L (6.9–15.7) IgG2 g/L (1.69–15.7) 1 0 0 118 43 64 0.0 0.2 5.9 ND 5 2 10 10 550 200 300 0.3 2.9 9.0 ND 4.5 3 160 150 1930 780 950 0.2 2.4 8.4 ND NA 4  < 10 15 349 185 145 0.0 3.4 9.5 1,24 17 5 6 1 259 134 66 0.0 6.0 13.0 0.79 221 6 2 0 809 495 326 0.1 0.5 44.32 2.3 4 7 18 19 913 545 341 0.0 2.8 14.0 2.60 3 83 40 2 1335 546 761 0.1 1.9 9.7 ND NA 9 ND ND ND ND ND 0.0 4.2 16.2 - 1 10 70 206 247 134 108 0.1 1.7 18.5 0.79 6 11 ND ND ND ND ND 0.0 0.9 9.5 2.11 10 12 ND ND ND ND ND 1.7 2.6 8.1 ND 1 134 28 13 1073 633 417 0.2 0.3 12.8 ND 0.5 14 8 11 676 323 341 0.1 1.2 11.9 1.14 NA Abnormal values are given in bold 1On Prednisolone 20 mg a day when these samples were taken 2Hypergammaglobulinemia on IVIG due to IgG2 deficiency 3The values are from the time at diagnosis of GATA2 deficiency 3 years ago 4The reference values for Patient 13 who was 12 years old at the time of HSCT are CD19 200–600, NK 70–1200, CD3 800–3500, cd4 400–1200, cd8 200–1200 (given in cells × 106/L) and for IgG the normal reference value was 6.1–14.9 g/L ND, not done; NA, not applicable Germline GATA2 Variants and Somatic Variants in Other Genes Ten different GATA2 pathogenic, or likely pathogenic, variants were found in 14 patients (Table 2). All identified constitutional GATA2 variants, except one, were located in the ZNF2 domain, corresponding to or in close proximity to exons 5 and 6 (Table 2). Three nonsense variants (p.Ala342Glyfs*43, p.Arg362*, p.Asn381fs*20), one + 1 splicing variant, three missense variants (p.Thr354Met, p.Trp360Arg, and p.Ala372Thr), all previously reported to be disease-causing[14, 23–25], and two novel in-frame deletions (p.Thr358del, p.Asp367del) were found. The variant located in exon 3, outside the ZNF2 domain, was a nonsense variant (c.163C > T, p.Gln55*). It was initially identified by the NGS myeloid panel with variant allele frequency (VAF) 49% in the DNA from the patient’s BM and later verified to be germline with VAF 49% in a buccal DNA sample (Patient 9, Table 2). Patient 2 had a paternal inherited in-frame deletion, c.1062_1064del (p.Thr358del), in the ZNF2 domain, and a somatic in-frame deletion, c.1168_1170del (p.Lys390del), with a fairly high VAF, 40.2% in BM. As expected, these two in-frame deletions were no longer detectable after allo-HSCT. Surprisingly, in the first post-transplant BM sample at day + 28, we detected another acquired GATA2 variant, c.1215G > T (p.Lys405Asn) with VAF 49.5%. This missense mutation variant, affecting an amino acid located C-terminal to the ZNF2 domain, is a variant of unknown clinical significance. It is most likely a rare benign variant, which in retrospect was confirmed to be constitutional in the unrelated BM donor (Table 2). It is evaluated to ACMG category 3 minus, since altogether 5 heterozygote individuals with the same amino acid change p.Lys405Asn are reported in GnomAD (v.2.1.1)[31]. As far as we know, missense variants located outside the ZNF2 domain rarely represent constitutional susceptibility to development of myelodysplasia. One exception is the p.Ser447Arg located C-terminal of the ZNF2 domain[32], while no other missense variants outside the ZNF2 domain are currently defined as pathogenic or likely pathogenic in ClinVar (www.clinvar.com) as of year 2021. Karyotype abnormalities and somatic variants in other genes observed in the 14 patients are presented in Table 2. Trisomy 8 (n = 6), monosomy 7 (n = 3), STAG2 variants (n = 3), ASXL1 variants (n = 2), a combination of somatic variants in RUNX1/CSF3R/PHF6/FBXW7 (n = 1), and variants in the following MDS genes[33] were observed once in separate individuals: BCOR, SETBP1, U2AF1, and somatic GATA2. One adult GATA2 deficient patient who developed AML had an unbalanced translocation der(1;7) in the leukemic clone. Since GATA2 deficiency is considered the most common hereditary predisposition to pediatric MDS, we estimated the proportion of pediatric patients diagnosed with MDS in Norway that had a germline GATA2 variant in the same time period (2013–2020). We found that three out of 14 pediatric patients diagnosed with MDS (21%) had GATA2 deficiency. Of note, these are 14 pediatric MDS patients and not the same cohort of 14 GATA2 deficient patients described above (except three overlapping pediatric patients, Patients 2, 12, and 13, with MDS). Two of the 3 pediatric GATA2 deficient patients with MDS had both monosomy 7 and trisomy 8 in their bone marrow cells. Patients 4, 5, 8, 9, 10, and 11 from Family B, E, F, and G had frameshift or other definitive loss-of-function variants, while Patients 1, 2, 3, 6, 7, 12, 13, and 14 from Family A, C, D, H, I, and J had in-frame deletions or missense variants. No specific genotype–phenotype correlations were found in our data set, i.e., regarding debut of symptoms, type and distribution of infections, age of transition to MDS/AML, somatic occurring variants in blood and BM. The severity of the clinical presentations also varied within families. Families and Predictive Genetic Testing The pedigrees of the 10 families are presented in detail in Fig. 1. Patient 1 (Family A) had three apparently healthy children, when he was diagnosed with GATA2 deficiency. After genetic testing of first-degree relatives, we found that two of his children (Patients 2 and 3) had inherited the GATA2 variant. For Patient 2, initial clinical work-up revealed only mild cytopenia and warts. However, within 2 years of follow-up, he developed pancytopenia and transfusion dependency and was diagnosed with MDS-EB1. His sister, Patient 3, has warts as her only clinical manifestation, but will be followed up regularly for development of cytopenia/MDS.Fig. 1 Pedigrees of the ten families, including 14 patients, with known GATA2 deficiency. Solid symbols denote affected status. Individuals marked in gray are deceased and not tested for GATA2 deficiency but are suspected to carry the disease-causing variant. In family G, the mother of Patients 10 and 11 died at age 30 of acute respiratory distress syndrome, 27 years ago. She also had lymphedema since birth. In light of their mother’s medical history, the GATA2 variant is probably maternally inherited. The father is alive and healthy. In family J, the mother of Patient 14 had a combined B and T cell defect, warts, myelodysplastic syndrome, lymphedema, and recurrent respiratory tract infections. She died of vulval cancer at the age of 38. The maternal grandfather of Patient 14 died of acute leukemia at the age of 33. WT, wild-type In Family G, two siblings had the same germline GATA2 variant (Patients 10 and 11). Their mother died 27 years ago, at the age of 30, of acute respiratory distress syndrome (ARDS), of unknown etiology. She also had lymphedema since birth. In light of their mother’s medical history with lymphedema and ARDS, which could be secondary to complications related GATA2 deficiency, the GATA2 variant is probably maternally inherited. Their father is alive and healthy. The deceased mother of Patient 14 (Family J) had a combined B- and T cell defect, warts, MDS, lymphedema, and recurrent respiratory tract infections. At the age of 38 (years), she died of metastatic vulval cancer. The maternal grandfather of Patient 14 died of acute leukemia at the age of 33. Considering the family history, it is very likely that Patient 14 had inherited her germline GATA2 variant from her maternal grandfather via her mother. Both individuals died before GATA2 deficiency was acknowledged as a cause of PID. Her mother’s siblings are now offered genetic counselling/testing for GATA2 deficiency. Allo-HSCT in Patients with GATA2 Deficiency Twelve of 14 (86%) patients with GATA2 deficiency were found to have a clinical indication, cytogenetic findings, and/or molecular findings warranting to proceed to allo-HSCT. As of today, 11 patients have undergone allo-HSCT, whereas one is recently accepted for allo-HSCT (Patient 14). Clinical features that lead to the decision to perform allo-HSCT were previous life-threatening disseminated HSV infection (Patient 1), severe obliterating bronchiolitis and in situ carcinoma (Patients 4 and 5), MDS with cytogenetic abnormalities (monosomy 7) and/or excess of blasts with high likelihood of progression to leukemic transformation (Patients 2, 6, 7, 9, 12, 13, and 14), MDS and warts with high-grade dysplasia (Patient 10), and symptoms of severe immunodeficiency and MDS (Patient 11). Details on the allo-HSCT procedure, including conditioning, donor selection, stem cell source, donor/recipient cytomegalovirus status, donor chimerism, graft versus host disease (GVHD) prophylaxis, and the occurrence of GVHD, are presented in Table 4.Table 4 HSCT details for eleven patients with GATA2 deficiency Patient no Age at HSCT Donor Stem cell source HLA Match CD34 + , × 106/kg CMV status d/r Conditioning regimen & In vivo T-cell depletion Chimerism %Day + 28 GVHD prophylaxis GVHD Complications 1 41 y MUD PBSC 10/10 (11/12) 7.8 -/ +  RIC: Fludarabine 150 mg/ m2, Busulfan 8 mg/kg, ATG Thymoglobulin 4 mg/kg 99% Mtx + CsA No Hemorrhagic cystitis (BK-virus) 2 16 y MUD PBSC 10/10 (10/12) 9.7 -/ - MAC*: Busulfan for 4 days (TDM; Css 825 ng/ml), Cyclophosphamide 120 mg/kg, Melphalan 140 mg/m2, ATG Grafalon 3 × 10 mg/kg  > 99% Mtx + CsA No E. coli sepsis; BK-virus cystitis; mucositis (grade 3) 4 39 y MUD PBSC 10/10 (12/12) 10.6 + / +  RIC: Fludarabine 150 mg/ m2, Busulfan 8 mg/kg. ATG Thymoglobulin 4 mg/kg  > 99% Mtx + CsA Chronic: skin and gut CMV reactivation 5 39 y MUD PBSC 10/10 (11/12) 5.2 +/ - RIC: Fludarabine 90 mg/m2, 2 Gy TBI 99% Mtx + CsA No Enterococcus faecalis sepsis (2 months post-HSCT), prolonged cytopenia, died of intracerebral hemorrhage 8 months post-HSCT 6 29 y MRD PBSC HLA-id sibling 5.4 + / +  RIC: Fludarabine 150 mg/m2, Busulfan 8 mg/kg 98% Mtx + CsA Chronic: liver, oral mucosa and genitalia Cytopenia at day + 33, osteoporosis, compression fractures 7 22 y MUD PBSC 10/10 6.9 -/ - MAC: Fludarabine 150 mg/m2, Treosulfan 42 g/m2, ATG Thymoglobulin 4 mg/kg  > 99% Mtx + CsA Chronic (limited): skin Moraxella nonliquefaciens sepsis on day 0. E. coli sepsis day + 14. Oral mucositis grade IV. PTLD 7 weeks post-HSCT 9 23 y MUD PBSC 10/10 (11/12) 4.3 +/ - MAC: Fludarabine 160 mg/m2, Busulfan 12,8 mg/kg (i.v.), ATG Thymoglobulin 4 mg/kg  > 99% Mtx + CsA Acute GvHD grade I: skin None 10 32 y MUD PBSC 10/10 (11/12) 5.9 -/ - MAC: Fludarabine 150 mg/m2, Treosulfan 42 g/m2, ATG Thymoglobulin 4 mg/kg  > 99% Mtx + CsA Acute GvHD: serositis None 11 27 y MUD PBSC 10/10 (10/12) 10.2 -/ - MAC: Cyclophosphamide 100 mg/kg, Busulfan 16 mg/kg N.a Mtx + CsA Chronic (extensive): gut, eye, and lung Hemorrhagic cystitis, Herpes oesophagitis 12 14 y MUD BM 10/10 (11/12) TNC: 3.5 × 108/kg -/ - MAC*: Fludarabine 160 mg/m2, Treosulfan 42 g/m2,Thiotepa 8 mg/kg, ATG Grafalon 3 × 10 mg/kg day + 84: > 99% Mtx + CsA No Impetigo day + 40 13 11 y MMFD (father) PBSC, TCRab + /CD19 + depletion in vitro Haploidentical 10.3 + / +  MAC*: Fludarabine 160 mg/m2, Thiotepa 10 mg/kg, Melphalan 140 mg/m2, ATG Grafalon 3 × 10 mg/kg  > 99% MMF (until day + 28) No None Abbreviations: BM, bone marrow; CMV, cytomegalovirus; CsA, cyclosporine A; Css, concentration at steady-state; Cya, cyclosporine A; GVHD, graft versus host disease; i.v., intravenous; MAC, myeloablative conditioning; MMF, mycophenolate mofetil; MMFD, mismatched family donor; MRD, matched related donor; Mtx, methotrexate: MUD, matched unrelated donor; N.a., not available; PBSC, peripheral blood stem cells; RIC, reduced-intensity conditioning; TDM, therapeutic drug monitoring; Tx, transplantation *According to HSCT recommendations by the EWOG-MDS study group Clinical Outcome The clinical outcome of all 14 patients is presented in Table 1. Two adult patients (18%) died after allo-HSCT. Patient 5 had persistent thrombocytopenia and died of a cerebral hemorrhage 8 months post transplantation. Patient 11 developed respiratory failure due to cGVHD in the lungs and received a bilateral pulmonary transplant 5 years post allo-HSCT. However, he developed chronic pulmonary rejection and died 2 years after lung transplantation and 7 years after allo-HSCT. The mean follow-up of the nine patients still alive after allo-HSCT is 26 months (range 3–78 months). The incidence of aGvHD and cGVHD among the eleven transplanted patients was 25% and 33%, respectively, all occurring in patients > 18 years of age (Table 4). None of the pediatric patients had experienced aGVHD or cGVHD, serious infectious complications, or any serious or unexpected transplant-related acute or late toxicity. Their transplantation courses were uneventful and did not principally differ from MDS patients without germline disease-causing GATA2 variants. One patient is listed for allo-HSCT (Patient 14) and two patients are followed regularly in the out-patient clinic (Patients 3 and 8). Discussion This retrospective study describes clinical features and outcome of 14 patients from ten families diagnosed with GATA2 deficiency in Norway. The main findings were as follows: (i) We found a diverse clinical phenotype dominated by cytopenia (13/14), myeloid neoplasia (MDS/AML) (10/14), warts (8/14), and hearing loss (7/14). (ii) We observed two novel clinical features multiple aneurysms of small vessels (n = 1) and early graying (n = 1) that could be associated with GATA2 deficiency. (iii) The majority of patients (11/14) had already undergone allo-HSCT at the time of our analysis, illustrating the need for allo-HSCT in a large proportion of patients with GATA2 deficiency in Norway, and most likely in other countries. (iv) Genetic testing should be offered to first-degree relatives, particularly children, to identify individuals with GATA2 deficiency that need close surveillance. Genetic testing performed as soon as a clinical suspicion is raised, increases the likelihood of an early and correct molecular diagnosis. In Norway, exome-based panels including GATA2 are offered as a routine diagnostic laboratory service for constitutional hematological disorders and PID [21]. However, this approach will not detect all GATA2 variants. Germline variants located in the intronic transcriptional enhancer elements, the cis-acting E-box/GATA and ETS motifs within intron 5 (NM_001145661.1), may cause GATA2 deficiency[34]. Supplementary Sanger sequencing of the enhancer element sequences in intron 5[34] was only performed in one of the laboratories (see Supplementary Methods). Despite supplementary copy number variant calling from exome data, with chromosomal microarray and MLPA only in selected patients (Supplementary Methods), some structural variants may go undetected. In addition, synonymous disease-causing GATA2 variants resulting in selective loss of mutated RNA were recently reported [35]. Disease-causing intronic variants, small intragenic variants, such as structural variants and synonymous variants, may also have escaped detection and hence, individuals with GATA2 deficiency may have been overlooked. With the recent introduction of targeted NGS panels in the work-up of myeloid neoplasms searching for somatic GATA2 variants, unexpected germline variants can be identified, which may reveal the underlying constitutional cause of the myeloid disease and increase the prevalence of known GATA2 deficiency. Improving the NGS panels targeting both germline and somatic variants by deeper coverage of the whole genes including important non-exonic regions, better algorithms for detection of structural variants, and attention to rare synonymous variants may enhance the identification of GATA2 deficiency. In this case series, we found two clinical features that have not yet been described in GATA2 deficiency, namely small vessel aneurysm and early graying. Multiple small vessel aneurysms found in Patient 10 may be secondary to vasculitis, or could also represent a novel vascular feature associated with GATA2 deficiency. In fact, it has been suggested that alteration in GATA2 expression may be of importance for vascular integrity[36]. The observation of premature graying may be a coincidence. In the absence of telomere biology disorders as in the present patients, one may speculate if this feature may reflect a GATA2-linked autoimmune phenomenon which has gone undetected. One of the aims of this retrospective study was to describe the clinical characteristics, GATA2 variants, and other molecular variants representing risk factors for clonal evolution that could aid us in the difficult decision regarding: “Who and when to transplant”? The high proportion of patients (79%) that had already undergone allo-HSCT in our cohort was somewhat surprising. Donadieu et al. have published the largest cohort of patients with GATA2 deficiency, and found that only 28 patients (35%) of 79 patients had undergone allo-HSCT. However, this low percentage of allo-HSCT did not correspond with the severity of the disease in this cohort. At the age of 40, the authors reported a mortality rate of 35% and a hematological malignancy rate of 80%[14]. The high proportion of allo-HSCT in our study may have been influenced by increased awareness of negative prospective clinical markers of GATA2 deficiency. Based on the findings from our study and the high morbidity and mortality rate reported by Donadieu et al., it is clear that these patients need to be monitored closely. Ideally, allo-HSCT should be performed before they develop malignancies (both solid tumors and hematological malignancies)[37] or severe/recurrent infections causing organ failure. In our opinion, a history of disseminated viral infection, aggressive HPV infection (particular with dysplasia), or myeloid clonal disease is clear indication to consider allo-HSCT[14]. First-degree relatives with a severe outcome of the disease may further strengthen the indication for an early allo-HSCT in symptomatic patients with GATA2 deficiency. Overall, the decision to perform an allo-HSCT requires careful weighing of potential gain (restore immune function; diminish the risk of hematological malignancies) versus possible transplant complications, including GVHD and transplant-related mortality. This is particularly challenging given the lack of genotype–phenotype correlation. Keeping in mind that the observation time is short for some of the patients in our study, the survival rate after allo-HSCT was 82% (9/11). In patients with GATA2 deficiency, previous publications have reported 86% survival 2 years after HSCT (n = 22)[16], 73% and 62% survival 1 and 5 years after HSCT, respectively (n = 28)[14], 72%, 65%, and 54% survival 1, 2, and 4 years after HSCT, respectively (n = 21)[9], and 57% 3, 5 years after HSCT (n = 14)[38]. These cohorts are, however, not necessarily comparable in terms of severity of disease and conditioning regimen. Two children were diagnosed with GATA2 deficiency after family screening. The hematological surveillance of one of these children led to detection of hematological abnormalities consistent with MDS and, in the end, a timely allo-HSCT. We therefore recommend genetic testing of children of affected adults and hematological surveillance of individuals with known pathogenic germline GATA2 variants. This includes annual BM investigations with morphological and cytogenetic evaluations, and testing with NGS myeloid panel to screen for somatic occurring molecular drivers of malignancies. Monosomy 7 and trisomy 8 have been reported by others to be the major cytogenetic aberrations in hematopoietic cells of patients with GATA2 deficiency and MDS[15]. Advanced MDS disease and monosomy 7 have been related to worse outcome, especially for pediatric patients with GATA2 germline disease[20]. We found monosomy 7 only in ¼, but trisomy 8 in half of the karyotyped patients, which is in line with a previous published study by McReynolds et al. [39]. Two patients in our GATA2 cohort had monosomy 7 and trisomy 8, both were children. Some of the other somatic variants in our cohort occurred in genes previously reported to be mutated in GATA2 deficiency with MDS, such as ASXL1[40], STAG2, SEPTBP1, and RUNX1[41, 42]. Interestingly, one adult patient with GATA2 deficiency and MDS-related AML had a der(1;7) in the leukemic clone, a translocation that has recently been shown to be enriched in pediatric MDS patients with germline GATA2 mutations[43]. However, our number of patients are too small to determine possible genotype–phenotype correlations related to clonal disease progression. Previous reports have suggested that plasma levels of FLT3LG can be used as predictor of hematological disease in GATA2 deficiency, and used in clinical monitoring post-HSCT [25]. Unfortunately, we lack serum or plasma samples taken before and after HSCT in our cohort, and the role of FLT3LG as a disease progression marker could be explored in future studies. Our main conclusion of this study is that the majority of patients with symptomatic GATA2 deficiency will need allo-HSCT, and close surveillance of these patients is important to find the “optimal window” for allo-HSCT. We advocate a more offensive approach to allo-HSCT than previously described. Supplementary Information Below is the link to the electronic supplementary material.Supplementary file1 (DOCX 55 KB) Acknowledgements Our gratitude goes to the Section for Cancer Cytogenetics who karyotyped 9 of the 14 patients reported here (Table 2). Three of the reported families have been followed up at the Department of Medical Genetics, University Hospital of North Norway. We want to thank our colleagues, MD Gry Hoem, MD Hilde Yttervik, MD Specialist in medical genetics and pediatrics Marie Falkenberg Smeland, and MD PhD Øyvind Holsbø Hald at the Department of Medical Genetics, University Hospital of North Norway, Tromsø, Norway, for their clinical work and genetic counselling. The three pediatric patients are enrolled in the registry of the European Working Group of MDS in Childhood (EWOG-MDS; ClinicalTrials.gov Identifier: NCT00662090). Author Contribution SFJ, JB, AEM, PA, AS-P, TGD, and IN wrote the paper. The genetic analyses were performed by AS-P, MAK, HS, ØH, SS, and EL. SFJ, JB, AEM, EG, YF, CA, AB, IH, TF, BF, AS-P, TGD, and IN collected clinical data. SS did the bone marrow analyses. All authors critically revised the manuscript for important intellectual content and approved the final version of the manuscript. Funding Open access funding provided by University of Oslo (incl Oslo University Hospital). Data Availability Upon request. Code Availability Not applicable. Declarations Ethics Approval Five of the patients had previously consented to be part of a genetic PID research project approved by the regional ethical committee (REC. 2014/1270–1), three patients were diagnosed with MDS in childhood and consented to be registered into the EWOG-MDS-2006 study (2015/1651/REC Nord), and all adult patients who underwent allo-HSCT had consented to publication of data (REC 11909). Studies were performed according to the Declaration of Helsinki. Consent to Participate All adult living patients signed a written informed consent for publication. For children < 18 years, consent was given by their parents. Consent for Publication Obtained. Conflict of Interest The authors declare no competing interests. Abbreviations aGVHD Acute graft versus host disease allo-HSCT Allogeneic hematopoietic stem cell transplantation AML Acute myelogenous leukemia ARDS Acute respiratory distress syndrome BM Bone marrow cGVHD Chronic graft versus host disease GVHD Graft versus host disease HPV Human papilloma virus MDS Myelodysplastic syndrome MRD Matched related donor MUD Matched unrelated donors NGS Next-generation sequencing PID Primary immunodeficiency VAF Variant allele frequency WES Whole exome sequencing ZNF1 Zink finger 1 ZNF2 Zink finger 2 Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Asbjørg Stray-Pedersen, Tobias Gedde-Dahl and Ingvild Nordøy contributed equally

BUSULFAN, CYCLOSPORINE, FLUDARABINE PHOSPHATE, METHOTREXATE

DrugsGivenReaction

CC BY

34893945

2021-12-10

What was the dosage of drug 'BUSULFAN'?

A Nationwide Study of GATA2 Deficiency in Norway-the Majority of Patients Have Undergone Allo-HSCT. OBJECTIVE GATA2 deficiency is a rare primary immunodeficiency that has become increasingly recognized due to improved molecular diagnostics and clinical awareness. The only cure for GATA2 deficiency is allogeneic hematopoietic stem cell transplantation (allo-HSCT). The inconsistency of genotype-phenotype correlations makes the decision regarding "who and when" to transplant challenging. Despite considerable morbidity and mortality, the reported proportion of patients with GATA2 deficiency that has undergone allo-HSCT is low (~ 35%). The purpose of this study was to explore if detailed clinical, genetic, and bone marrow characteristics could predict end-point outcome, i.e., death and allo-HSCT. METHODS All medical genetics departments in Norway were contacted to identify GATA2 deficient individuals. Clinical information, genetic variants, treatment, and outcome were subsequently retrieved from the patients' medical records. RESULTS Between 2013 and 2020, we identified 10 index cases or probands, four additional symptomatic patients, and no asymptomatic patients with germline GATA2 variants. These patients had a diverse clinical phenotype dominated by cytopenia (13/14), myeloid neoplasia (10/14), warts (8/14), and hearing loss (7/14). No valid genotype-phenotype correlations were found in our data set, and the phenotypes varied also within families. We found that 11/14 patients (79%), with known GATA2 deficiency, had already undergone allo-HSCT. In addition, one patient is awaiting allo-HSCT. The indications to perform allo-HSCT were myeloid neoplasia, disseminated viral infection, severe obliterating bronchiolitis, and/or HPV-associated in situ carcinoma. Two patients died, 8 months and 7 years after allo-HSCT, respectively. CONCLUSIONS Our main conclusion is that the majority of patients with symptomatic GATA2 deficiency will need allo-HSCT, and a close surveillance of these patients is important to find the "optimal window" for allo-HSCT. We advocate a more offensive approach to allo-HSCT than previously described. pmcIntroduction GATA2 deficiency is a rare primary immunodeficiency (PID), first described in 2011[1–3] that has become gradually more recognized due to improved molecular diagnostics and increased clinical awareness. GATA2, as a “master” transcription factor, plays a critical role in hematopoietic development[4]. Through cooperative processes that include other transcription factors, it controls the transition from hemogenic endothelium to hematopoietic stem cells and is required for survival and self-renewal of these cells[5]. GATA2 is also important for other tissue-forming stem cells, e.g., in the inner ear[6]. The heterozygous variants causing GATA2 deficiency are located both in coding, non-coding and enhancer regions[7]. The disease-causing loss-of-function variants can be localized across the gene. These variants can lead to defective DNA-binding capacity of the transcription factor and may cause disease through haploinsufficiency of the functional protein[5, 8]. Missense variants within the zink finger 2 (ZNF2) domain are the most frequent germline disease-causing GATA2 variants [9]. It has been estimated that approximately 1/3 of the patients have an autosomal dominant inherited disease-causing variant[10], whereas the remaining have a de novo GATA2 variant[7]. Of note, somatic variants in GATA2 are known to be drivers of myeloid neoplasia in adults. Such variants are diverse, may cause gain-of-function effects, and be located across the whole gene. This includes missense variants in the zink finger 1 (ZNF1) domain, which has not been observed in constitutional GATA2 deficiency[8]. Typically, GATA2 deficiency becomes clinically apparent in late childhood to early adulthood. The phenotype is heterogeneous, without any clear genotype–phenotype correlation, and with an incomplete clinical penetrance[11]. Symptoms may include recurrent or severe infections, warts, cytopenia (including monocytopenia), lymphedema, alveolar proteinosis, and malignant myeloid disease[9]. Infectious complications in GATA2 deficiency are likely due to deficiency of monocytes, NK cells, and B-lymphocytes as well as defective innate immune responses, including impaired type I interferon production[12]. This leads to both increased susceptibility to viral infections (e.g., human papilloma virus [HPV, warts] and herpes virus infections), non-tuberculous mycobacteria, and to more common bacterial respiratory infections. Hearing loss is a common clinical feature of GATA2 deficiency and is related to the critical role of GATA2 in vestibular morphogenesis of semicircular ducts and generation of the perilymphatic space around the inner ear’s semicircular canals[6, 13]. A substantial proportion of patients develop immunodeficiency, myelodysplastic syndrome (MDS), or acute myelogenous leukemia (AML) as initial manifestation[9, 14]. GATA2 deficiency is considered the most common hereditary predisposition to pediatric MDS[15]. The only cure for GATA2 deficiency is allogeneic hematopoietic stem cell transplantation (allo-HSCT) and results are encouraging[16–20]. However, the main challenge is deciding who and when to transplant due to the complexity and inconsistency of phenotype-genotype correlation in GATA2 deficiency[9]. To further elucidate this important issue, we present detailed clinical and molecular characteristics, treatment, and outcome of 14 Norwegian patients with germline GATA2 variants diagnosed between 2013 and 2020. The main aim of our study was to explore if detailed clinical, genetic, and bone marrow (BM) characteristics could predict end-point outcome such as death and allo-HSCT in patients with GATA2 deficiency. Methods Identification of Patients and Clinical Characteristics The first aim of this study was to obtain a complete overview of all patients with known GATA2 deficiency in Norway. For this purpose, a network of clinical immunologists, hematologists, pediatricians, and geneticists at Oslo University Hospital (OUH) collected clinical and laboratory data on patients with GATA2 deficiency at their institution. In addition, all medical genetics departments in Norway were contacted to identify any additional GATA2 deficient individuals. Clinical information, genetic variants, treatment, and outcome were subsequently retrieved from the patients’ medical records. Patients were enrolled into the study at OUH where most of the data was obtained, while supplemental data from Patient 3 was collected at the University Hospital of North Norway, Tromsø. Informed Consent Five of the patients had previously consented to be part of a genetic PID research project approved by the regional ethical committee (REC. 2014/1270–1), three patients were diagnosed with MDS in childhood and consented to be registered into the EWOG-MDS-2006 study (2015/1651/REC Nord), and all adult patients who underwent allo-HSCT had consented to publication of data (REC 11,909). In addition, due to the detailed clinical information published herein, all adult living patients signed an additional written informed consent for publication of their data and was given the opportunity to review the manuscript prior to publication. For children < 18 years, consent was given by their parents. This is in line with the recommendation given by the Ethical Constitutional board at OUH. Genetic Analyses Whole exome sequencing (WES) with in silico filtering for genes causing primary immunodeficiency disorders was performed in the probands and affected relatives as part of a routine laboratory service (Patients 2, 3, 8, 9, 10, 12, and 14) or on a research basis (Patients 4, 5, 6, and 7) as previously described (Supplemental methods)[21]. Patient 1 had severe cytopenia (Table 1), and the first attempt to extract DNA from peripheral blood was not successful. A skin biopsy was therefore performed to extract DNA from fibroblasts. In parallel, peripheral blood (from puncture of the fingertip) was applied directly to a Guthrie filter card, and by using multiple filter card punches, enough DNA was extracted to run next-generation sequencing (NGS) with an amplicon-based targeted panel for constitutional variants in PID genes (Supplemental methods). By using this rapid amplicon-based method, the molecular result was available within 3 working days[22]. DNA later extracted from fibroblasts confirmed the GATA2 variant by Sanger sequencing. Also, for Patient 13, who had advanced MDS with pancytopenia, the NGS results were available within 3 working days, with parental testing performed in parallel to evaluate as fast as possible the availability of a healthy unaffected matched related donor (MRD).Table 1 Clinical characteristics and outcome in patients with GATA2 deficiency Patient no Family Sex Current age Age at onset of symptoms/age at genetic diagnosis Infections Hearing loss Hematologic abnormalities Autoimmunity/immune dysregulation Miscellaneous HSCT, age Outcome Viral Bacterial 1 A (father of P2 and P3) M 44y 5y/41y HPV: warts HSV: disseminated disease Ear infections as a child Yes Hypoplastic BM: cytopenia, trilinear hypoplasia No No 41y Alive 2 A (son of P1) M 16y 7y/14y HPV: warts No No MDS-EB-1 No No 16y Alive 3 A (daughter of P1) F 13y 8y/9y HPV: warts No No No No No ND Alive 4a B (monozygotic twin to P5) F 45y 21y/38y HPV: warts, carcinoma in situ EBV: prolonged viremia Recurrent respiratory inf Suppurative skin infection after BCG * Yes No Progressive obliterating bronchiolitis, lupus-like syndrome Miscarriage 39y Alive 5a B (monozygotic twin to P4) F † (39y) 24y/38y HPV: warts, carcinoma VZV and EBV: prolonged viremia Recurrent respiratory inf Suppurative skin inf. after BCG* Yes MDS-MLD (hypoplastic) Progressive obliterating bronchiolitis, lupus-like syndrome DVT × 2 Squamous cell carcinoma in the cervix, rectum, and anus 39y Deceased 8 m post-HSCTb 6a C M 31y 11y/26y No Recurrent respiratory inf Yes MDS-MLD (hypoplastic) - Fever of unknown origin, recurrent pneumothorax 29y Alive 7a D F 23y 6y/17y HPV: warts Recurrent respiratory inf No MDS-MLD (hypoplastic) Interstitial lung disease Lymphedema, acne, rosacea, rash, fatigue 22y Alive 8 E F 56y 0y/53y No No No Hypoplastic BM No Lymphedema, premature graying ND Alive 9 F F 24y 15y/23y No No No AML with MDS-related changes Erythema nodosum DVT, PE, juvenile myoclonic epilepsy, epicanthic fold 23y Alive 10 G (sibling to P11) F 32y 6y/31y HPV: warts, cervix dysplasia Recurrent respiratory inf Yes MDS-MLD No Aneurysm of small vessels, hidradenitis suppurative, liver lesions: focal nodular hyperplasia 32y Alive 11 G (sibling to P10) M † (34y) 22y/PM No Recurrent skin and respiratory inf No MDS-MLD No Acne, rosacea, necrotizing fasciitis, pilonidal cysts, skin infections, ulcerations 27y Deceased 7y post-HSCTc 12 H F 19y 14y/14y No No Yesd MDS-RCC (hypoplastic) BPD/Asthma Born premature (week 26 + 5), BPD 14y Alive 13 I M 13y 9y/11 y HPV: warts No No MDS-EB1 Asthma Chronic skin abscesses, congenital ptosis 11y Alive 14 J F 31 23y/31y No No Yes MDS-SLD (hypoplastic) No Born prematurely (week 25), cerebral palsy, congenital hip dysplasia Planned Alive Abbreviations: AML, acute myeloid leukemia; BCG, bacille Calmette Geurin; BM, bone marrow; BPD, bronchopulmonary dysplasia, CT, computer tomography, HPV, human papilloma virus; HSCT, hematopoietic stem cell transplantation; Inf., infection; m, months; MDS, myelodysplastic syndrome; MDS-EB1, MDS with excess of blasts type 1; MDS-MLD, MDS with multilineage dysplasia; MDS-SLD, MDS with single lineage dysplasia; MDS-RCC, MDS subtype refractory cytopenia of childhood; ND, not done, PM, post mortem; VTE, venous thromboembolism; y, years, †; deceased aThese patients have previously been published in Stray-Pedersen, Sorte et al. 2016 (Patient 4 was 84.1, Patient 5 was 84.4, Patient 6 was 88.1, and Patient 7 was 86.1)[21] bThe patient was doing well after HSCT, but died unexpectedly of a cerebral hemorrhage cThe patient underwent lung transplantation for chronic lung GVHD 58 months after HSCT, and died of chronic lung rejection 26 months after bilateral lung transplantation dThe patient has reduced hearing, but this was confirmed after HSCT. Her hearing loss may be due to the disease-causing GATA2 variant, but may also be secondary to complications of HSCT therapy, e.g., aminoglycosides The molecular diagnosis in Patient 11 was confirmed post mortem using a BM sample collected prior to allo-HSCT (Table 2). Methods for testing for somatic occurring sequence variants on DNA extracted from whole blood or BM, and methods for testing chromosomal aberration on BM cells are described in Supplemental methods.Table 2 Constitutional and acquired genetic findings in patients with GATA2 deficiency Patient no Hematological abnormalities Constitutional heterozygous variants in GATA2, NM_001145661.1, predicted protein effect, domain, occurrence, novelty, and reference Somatic variants, predicted protein effect, VAF in BM/blood (prior to HSCT) Karyotype in BM (closest to HSCT)d  + 8  − 7 1 Hypoplastic bone marrow c.1062_1064del, p.Thr358del, in-frame exon 6, ZNF2, likely de novo, novel variant Unknown 46,XY[25/25] No No 2 MDS-EB1 c.1062_1064del, p.Thr358del, in-frame exon 6, ZNF2, paternal inherited, novel variant Acquired germline donor variant in GATA2 Post-HSCTb: c.1215G > T, p.(Lys405Asn) missense exon 7, outside and distal to ZNF2 domain, VAF: 49,5% BM NM_001145661.1 (GATA2): c.1168_1170del, p.(Lys390del), in-frame exon 7, in ZNF2, VAF: 40.2% BM NM_006758.2(U2AF1): c.470A>G, p.(Gln157Arg) VAF: 44,0% BM 46,XY,-7 + 8[20/20] Yes Yes 3 No c.1062_1064del, p.Thr358del, in-frame exon 6, ZNF2, paternal inherited, novel variant Unknown Unknown N.a N.a 4 a No c.1143 + 5G > A, p.Asn381fs*20, splice defect intron 6, ZNF2, de novo, novel, reported by usc)[21] Unknown Unknown N.a N.a 5 a MDS-MLD (hypoplastic) c.1143 + 5G > A, p.Asn381fs*20, splice defect intron 6, ZNF2, de novo, novel, reported by usc[21] None 46,XX[18/18] No No 6a MDS-MLD (hypoplastic) c.1078 T > A, p.Trp360Arg, missense exon 6, ZNF2, de novo, variant previously reported by others[23] Unknown 46,XY[25/25], but FISH MYC(8q24): + 8 in 14/303 metaphases Yes No 7a MDS-MLD (hypoplastic) c.1061C > T, p.Thr354Met, missense exon 6, ZNF2, de novo, but a recurrent GATA2 variant[21, 24, 25] NM_001042749.2(STAG2): c.2534-2A > G, predicted splice variant with loss of acceptor site, Chr.X, VAF: 11.7% blood 47,XX, + 8[4/10]/46,XX[6/10] Yes No 8 Hypoplastic bone marrow c.1017 + 1G > T, loss of donor splice site, splice defect intron 5, ZNF1, both parents deceased and not tested, novel variant Unknown 46,XX[25/25] No No 9 AML with MDS-related changes c.163C > T, p.Gln55*, nonsense exon 3, TAD domain, likely de novo (see pedigree), novel variant VAF:48.7% in BM, 49,4% in buccal swap NM_001754.4(RUNX1): c.593A > G,p.(Asp198Gly) VAF:15% BM NM_156039.3(CSF3R): c.2326C > T, p.(Gln776*) VAF: 12.5% BM NM_032458.2(PHF6): c.309C > G, p.(Tyr103Ter) VAF:12.0% BM NM_033632.3(FBXW7): c.1513C > T, p.(Arg505Cys), VAF:11.7% BM 46,XX, der(1;7)(q10;p10), + 1[11/20]/46,XX [9/20] No No 10 MDS-MLD c.1084C > T, p.Arg362*, nonsense exon 6, ZNF2, likely inherited, variant previously reported by others[15, 26–28] NM_001123385.1(BCOR): c.529_530del, p.(Ser177ProfsTer8), VAF: 23.0% BM 49∼50,XX, + 6, + 8, + 21? + 21[cp7/8]/46,XX[1/8] Yes No 11 MDS-MLD c.1084C > T, p.Arg362*, nonsense exon 6, ZNF2, likely inherited, variant previously reported by others[15, 26–28] NM_015338.5(ASXL1): c.2324 T > G, p.(Leu775Ter) VAF: 20.5% BM NM_001042749.1(STAG2): c.2990 T > A, p.(Leu997Ter) VAF: 9.6% BM 47∼48,XY, + 8[10/15],der(16)t(1;16)(q21;q24[10/15], + der(16)t(1;16)[1/15], + 21[6][cp11/15]/46,XY[3/15] Trisomy 8, evolving to unbalanced 1;16 translocation and later Trisomy 21 Yes No 12 MDS-RCC c.1098_1100delGGA, p.Asp367del, in-frame exon 6, ZNF2, de novo, novel variant None 46,XX,-7, + 8[15/20] Yes Yes 13 MDS-EB1 c.1021_1024insGCCG, p.Ala342Glyfs*43, frameshift exon 6, ZNF1 de novo, variant previously reported[29] NM_015338.5(ASXL1): c.1854dupA, p.(Ala619SerfsTer16), VAF:17.0%, BM NM_015559.2 (SETBP1): c.2612 T > C, p.(Ile871Thr), VAF: 16.3%, BM 45,XY,-7[12/12] No Yes 14 MDS-SLD (hypoplastic) c.1114G > A, p.(Ala372Thr), missense exon 6, ZNF2, variant previously reported [14] NM_001042749.1(STAG2): c.707del; p.(Asn236IlefsTer20) VAF: 5.1%, BM 46,XX[25/25] No No Abbreviations: AML, acute myeloid leukemia; BM, bone marrow; Chr, chromosome; HSCT, hematopoietic stem cell transplantation; MDS, myelodysplastic syndrome; MDS-EB1, MDS with excess blasts type 1; MDS-MLD, MDS with multilineage dysplasia; MDS-RCC, MDS subtype refractory cytopenia of childhood; MDS-SLD, MDS with single lineage dysplasia; N.a., Not applicable; TAD, N-terminal transactivation domain, ZNF2, Zinc finger 2 domain in GATA2 protein; VAF, variant allele frequency aThese patients have previously been published in Stray-Pedersen, Sorte et al. (2016) (Patient 4 was 84.1, Patient 5 was 84.4, Patient 6 was 88.1, and Patient 7 was 86.1)[21] bThis disease-related GATA2 variant was detected in a routine BM at day + 28 post-HSCT; it turned out to be donor-derived (from a MUD) cWES identified a potential splicing variant in GATA2 (c.1143 + 5G > A) in Patient 4. The variant was predicted (Alamut®) to inactivate the donor site of GATA2 exon 5. PCR amplification of GATA2 exon 4 to 7 on cDNA showed that most transcripts were normally spliced resulting in a main product of ~ 400 bps, as seen in the normal control. A slightly longer PCR product including 64 bps of intron 6 sequence via a cryptic donor site in intron 6 (NM_001145661.1), was observed in the sample from Patient 4, but not in the control (see Supplementary information). Sanger sequencing identified the GATA2 splicing variant in the proband’s deceased monozygotic twin (Patient 5). Details described in Supplemental Figure E6 in Stray-Pedersen, Sorte et al. (2016)[21] dNomenclature according to ISCN (The International System for Human Cytogenetic Nomenclature) 2020 guidelines Results Characteristics of Patients Between 2013 and 2020, ten index cases, or probands, and four additional symptomatic patients with germline GATA2 variants were identified (9 females, 5 male, Table 1). Five adult patients were diagnosed by infectious disease specialists (Patients 1, 4, 5, 10, and 11) where infections (mostly HPV infection/warts and recurrent bacterial airway infections) were prominent symptoms. Four additional adult patients were identified by hematologists (Patients 6, 8, 9, and 14), where three were referred with pancytopenia and one patient had AML (also with pancytopenia). Three patients were diagnosed by pediatricians, two patients with MDS (Patients 12 and 13) and one patient with extensive warts and NK-/B-cell deficiency (Patient 7). Additionally, two patients with GATA2 deficiency were identified after family screening (Patients 2 and 3). We did not detect any asymptomatic individuals with GATA2 deficiency in this study. The mean age for debut of symptoms, that we considered related to GATA2 deficiency, was 12 years (range 0–24 years, Supplemental Table S1). The median time from these symptoms to a diagnosis of GATA2 deficiency was 11 years, range 0–53 years (Supplemental Table S1). Retrospectively, hearing loss, warts, and skin manifestations were the most common early symptoms, which in some patients became apparent many years before the genetic diagnosis of GATA2 deficiency was made (Supplemental Table S1). A summary of the patients’ clinical characteristics is given in Table 1. Viral infections such as HPV-associated warts were common, affecting eight patients. In addition, two patients had disseminated BCG infections (after vaccination), and one patient had a life-threatening disseminated HSV infection (originating from genitalia and disseminating to CNS and liver). Two patients experienced prolonged EBV and/or Varicella zoster viremia. Furthermore, six patients had recurrent bacterial airway infections. Interestingly, one patient had early graying (Patient 8), with normal telomere length, and one patient had multiple aneurysms of small vessels (coronary arteries, axillary arteries, and an iliac artery; Patient 10), which both represent clinical characteristics not previously described in GATA2 deficiency. In Patient 10, Varicella zoster infection was excluded as a cause of vasculitis with negative VZV PCR in blood. In addition, two patients had obliterating bronchiolitis (Patients 4 and 5), which has been reported in only one previous patient with GATA2 deficiency [30]. Affected cell lineages and immunoglobulin levels prior to allo-HSCT are listed in Table 3. As expected, the majority of patients had decreased levels of monocytes (11/14) and one patient had increased levels of monocytes (Patient 12). In addition, decreased levels of B cells (10/11) and NK cells (9/11) were common findings (three patients did not have NK- and B cells measured before allo-HSCT).Table 3 Immunoglobulin levels and affected cell lineage in peripheral blood prior to HSCT Patient no Affected cell lineage (normal range) Immunoglobulins Time before HSCT (months) CD19 + , cells × 106/L (100–500) NK cells × 106 /L (100–400) CD3 +  T- cells × 106/L (800–2400) CD4 +  T- cells × 106/L (500–1400) CD8 +  T-cells × 106/L (200–2000) Monocytes  × 109/L (0.2–0.8) Neutrophils  × 109/L (1.5–7.3) IgG g/L (6.9–15.7) IgG2 g/L (1.69–15.7) 1 0 0 118 43 64 0.0 0.2 5.9 ND 5 2 10 10 550 200 300 0.3 2.9 9.0 ND 4.5 3 160 150 1930 780 950 0.2 2.4 8.4 ND NA 4  < 10 15 349 185 145 0.0 3.4 9.5 1,24 17 5 6 1 259 134 66 0.0 6.0 13.0 0.79 221 6 2 0 809 495 326 0.1 0.5 44.32 2.3 4 7 18 19 913 545 341 0.0 2.8 14.0 2.60 3 83 40 2 1335 546 761 0.1 1.9 9.7 ND NA 9 ND ND ND ND ND 0.0 4.2 16.2 - 1 10 70 206 247 134 108 0.1 1.7 18.5 0.79 6 11 ND ND ND ND ND 0.0 0.9 9.5 2.11 10 12 ND ND ND ND ND 1.7 2.6 8.1 ND 1 134 28 13 1073 633 417 0.2 0.3 12.8 ND 0.5 14 8 11 676 323 341 0.1 1.2 11.9 1.14 NA Abnormal values are given in bold 1On Prednisolone 20 mg a day when these samples were taken 2Hypergammaglobulinemia on IVIG due to IgG2 deficiency 3The values are from the time at diagnosis of GATA2 deficiency 3 years ago 4The reference values for Patient 13 who was 12 years old at the time of HSCT are CD19 200–600, NK 70–1200, CD3 800–3500, cd4 400–1200, cd8 200–1200 (given in cells × 106/L) and for IgG the normal reference value was 6.1–14.9 g/L ND, not done; NA, not applicable Germline GATA2 Variants and Somatic Variants in Other Genes Ten different GATA2 pathogenic, or likely pathogenic, variants were found in 14 patients (Table 2). All identified constitutional GATA2 variants, except one, were located in the ZNF2 domain, corresponding to or in close proximity to exons 5 and 6 (Table 2). Three nonsense variants (p.Ala342Glyfs*43, p.Arg362*, p.Asn381fs*20), one + 1 splicing variant, three missense variants (p.Thr354Met, p.Trp360Arg, and p.Ala372Thr), all previously reported to be disease-causing[14, 23–25], and two novel in-frame deletions (p.Thr358del, p.Asp367del) were found. The variant located in exon 3, outside the ZNF2 domain, was a nonsense variant (c.163C > T, p.Gln55*). It was initially identified by the NGS myeloid panel with variant allele frequency (VAF) 49% in the DNA from the patient’s BM and later verified to be germline with VAF 49% in a buccal DNA sample (Patient 9, Table 2). Patient 2 had a paternal inherited in-frame deletion, c.1062_1064del (p.Thr358del), in the ZNF2 domain, and a somatic in-frame deletion, c.1168_1170del (p.Lys390del), with a fairly high VAF, 40.2% in BM. As expected, these two in-frame deletions were no longer detectable after allo-HSCT. Surprisingly, in the first post-transplant BM sample at day + 28, we detected another acquired GATA2 variant, c.1215G > T (p.Lys405Asn) with VAF 49.5%. This missense mutation variant, affecting an amino acid located C-terminal to the ZNF2 domain, is a variant of unknown clinical significance. It is most likely a rare benign variant, which in retrospect was confirmed to be constitutional in the unrelated BM donor (Table 2). It is evaluated to ACMG category 3 minus, since altogether 5 heterozygote individuals with the same amino acid change p.Lys405Asn are reported in GnomAD (v.2.1.1)[31]. As far as we know, missense variants located outside the ZNF2 domain rarely represent constitutional susceptibility to development of myelodysplasia. One exception is the p.Ser447Arg located C-terminal of the ZNF2 domain[32], while no other missense variants outside the ZNF2 domain are currently defined as pathogenic or likely pathogenic in ClinVar (www.clinvar.com) as of year 2021. Karyotype abnormalities and somatic variants in other genes observed in the 14 patients are presented in Table 2. Trisomy 8 (n = 6), monosomy 7 (n = 3), STAG2 variants (n = 3), ASXL1 variants (n = 2), a combination of somatic variants in RUNX1/CSF3R/PHF6/FBXW7 (n = 1), and variants in the following MDS genes[33] were observed once in separate individuals: BCOR, SETBP1, U2AF1, and somatic GATA2. One adult GATA2 deficient patient who developed AML had an unbalanced translocation der(1;7) in the leukemic clone. Since GATA2 deficiency is considered the most common hereditary predisposition to pediatric MDS, we estimated the proportion of pediatric patients diagnosed with MDS in Norway that had a germline GATA2 variant in the same time period (2013–2020). We found that three out of 14 pediatric patients diagnosed with MDS (21%) had GATA2 deficiency. Of note, these are 14 pediatric MDS patients and not the same cohort of 14 GATA2 deficient patients described above (except three overlapping pediatric patients, Patients 2, 12, and 13, with MDS). Two of the 3 pediatric GATA2 deficient patients with MDS had both monosomy 7 and trisomy 8 in their bone marrow cells. Patients 4, 5, 8, 9, 10, and 11 from Family B, E, F, and G had frameshift or other definitive loss-of-function variants, while Patients 1, 2, 3, 6, 7, 12, 13, and 14 from Family A, C, D, H, I, and J had in-frame deletions or missense variants. No specific genotype–phenotype correlations were found in our data set, i.e., regarding debut of symptoms, type and distribution of infections, age of transition to MDS/AML, somatic occurring variants in blood and BM. The severity of the clinical presentations also varied within families. Families and Predictive Genetic Testing The pedigrees of the 10 families are presented in detail in Fig. 1. Patient 1 (Family A) had three apparently healthy children, when he was diagnosed with GATA2 deficiency. After genetic testing of first-degree relatives, we found that two of his children (Patients 2 and 3) had inherited the GATA2 variant. For Patient 2, initial clinical work-up revealed only mild cytopenia and warts. However, within 2 years of follow-up, he developed pancytopenia and transfusion dependency and was diagnosed with MDS-EB1. His sister, Patient 3, has warts as her only clinical manifestation, but will be followed up regularly for development of cytopenia/MDS.Fig. 1 Pedigrees of the ten families, including 14 patients, with known GATA2 deficiency. Solid symbols denote affected status. Individuals marked in gray are deceased and not tested for GATA2 deficiency but are suspected to carry the disease-causing variant. In family G, the mother of Patients 10 and 11 died at age 30 of acute respiratory distress syndrome, 27 years ago. She also had lymphedema since birth. In light of their mother’s medical history, the GATA2 variant is probably maternally inherited. The father is alive and healthy. In family J, the mother of Patient 14 had a combined B and T cell defect, warts, myelodysplastic syndrome, lymphedema, and recurrent respiratory tract infections. She died of vulval cancer at the age of 38. The maternal grandfather of Patient 14 died of acute leukemia at the age of 33. WT, wild-type In Family G, two siblings had the same germline GATA2 variant (Patients 10 and 11). Their mother died 27 years ago, at the age of 30, of acute respiratory distress syndrome (ARDS), of unknown etiology. She also had lymphedema since birth. In light of their mother’s medical history with lymphedema and ARDS, which could be secondary to complications related GATA2 deficiency, the GATA2 variant is probably maternally inherited. Their father is alive and healthy. The deceased mother of Patient 14 (Family J) had a combined B- and T cell defect, warts, MDS, lymphedema, and recurrent respiratory tract infections. At the age of 38 (years), she died of metastatic vulval cancer. The maternal grandfather of Patient 14 died of acute leukemia at the age of 33. Considering the family history, it is very likely that Patient 14 had inherited her germline GATA2 variant from her maternal grandfather via her mother. Both individuals died before GATA2 deficiency was acknowledged as a cause of PID. Her mother’s siblings are now offered genetic counselling/testing for GATA2 deficiency. Allo-HSCT in Patients with GATA2 Deficiency Twelve of 14 (86%) patients with GATA2 deficiency were found to have a clinical indication, cytogenetic findings, and/or molecular findings warranting to proceed to allo-HSCT. As of today, 11 patients have undergone allo-HSCT, whereas one is recently accepted for allo-HSCT (Patient 14). Clinical features that lead to the decision to perform allo-HSCT were previous life-threatening disseminated HSV infection (Patient 1), severe obliterating bronchiolitis and in situ carcinoma (Patients 4 and 5), MDS with cytogenetic abnormalities (monosomy 7) and/or excess of blasts with high likelihood of progression to leukemic transformation (Patients 2, 6, 7, 9, 12, 13, and 14), MDS and warts with high-grade dysplasia (Patient 10), and symptoms of severe immunodeficiency and MDS (Patient 11). Details on the allo-HSCT procedure, including conditioning, donor selection, stem cell source, donor/recipient cytomegalovirus status, donor chimerism, graft versus host disease (GVHD) prophylaxis, and the occurrence of GVHD, are presented in Table 4.Table 4 HSCT details for eleven patients with GATA2 deficiency Patient no Age at HSCT Donor Stem cell source HLA Match CD34 + , × 106/kg CMV status d/r Conditioning regimen & In vivo T-cell depletion Chimerism %Day + 28 GVHD prophylaxis GVHD Complications 1 41 y MUD PBSC 10/10 (11/12) 7.8 -/ +  RIC: Fludarabine 150 mg/ m2, Busulfan 8 mg/kg, ATG Thymoglobulin 4 mg/kg 99% Mtx + CsA No Hemorrhagic cystitis (BK-virus) 2 16 y MUD PBSC 10/10 (10/12) 9.7 -/ - MAC*: Busulfan for 4 days (TDM; Css 825 ng/ml), Cyclophosphamide 120 mg/kg, Melphalan 140 mg/m2, ATG Grafalon 3 × 10 mg/kg  > 99% Mtx + CsA No E. coli sepsis; BK-virus cystitis; mucositis (grade 3) 4 39 y MUD PBSC 10/10 (12/12) 10.6 + / +  RIC: Fludarabine 150 mg/ m2, Busulfan 8 mg/kg. ATG Thymoglobulin 4 mg/kg  > 99% Mtx + CsA Chronic: skin and gut CMV reactivation 5 39 y MUD PBSC 10/10 (11/12) 5.2 +/ - RIC: Fludarabine 90 mg/m2, 2 Gy TBI 99% Mtx + CsA No Enterococcus faecalis sepsis (2 months post-HSCT), prolonged cytopenia, died of intracerebral hemorrhage 8 months post-HSCT 6 29 y MRD PBSC HLA-id sibling 5.4 + / +  RIC: Fludarabine 150 mg/m2, Busulfan 8 mg/kg 98% Mtx + CsA Chronic: liver, oral mucosa and genitalia Cytopenia at day + 33, osteoporosis, compression fractures 7 22 y MUD PBSC 10/10 6.9 -/ - MAC: Fludarabine 150 mg/m2, Treosulfan 42 g/m2, ATG Thymoglobulin 4 mg/kg  > 99% Mtx + CsA Chronic (limited): skin Moraxella nonliquefaciens sepsis on day 0. E. coli sepsis day + 14. Oral mucositis grade IV. PTLD 7 weeks post-HSCT 9 23 y MUD PBSC 10/10 (11/12) 4.3 +/ - MAC: Fludarabine 160 mg/m2, Busulfan 12,8 mg/kg (i.v.), ATG Thymoglobulin 4 mg/kg  > 99% Mtx + CsA Acute GvHD grade I: skin None 10 32 y MUD PBSC 10/10 (11/12) 5.9 -/ - MAC: Fludarabine 150 mg/m2, Treosulfan 42 g/m2, ATG Thymoglobulin 4 mg/kg  > 99% Mtx + CsA Acute GvHD: serositis None 11 27 y MUD PBSC 10/10 (10/12) 10.2 -/ - MAC: Cyclophosphamide 100 mg/kg, Busulfan 16 mg/kg N.a Mtx + CsA Chronic (extensive): gut, eye, and lung Hemorrhagic cystitis, Herpes oesophagitis 12 14 y MUD BM 10/10 (11/12) TNC: 3.5 × 108/kg -/ - MAC*: Fludarabine 160 mg/m2, Treosulfan 42 g/m2,Thiotepa 8 mg/kg, ATG Grafalon 3 × 10 mg/kg day + 84: > 99% Mtx + CsA No Impetigo day + 40 13 11 y MMFD (father) PBSC, TCRab + /CD19 + depletion in vitro Haploidentical 10.3 + / +  MAC*: Fludarabine 160 mg/m2, Thiotepa 10 mg/kg, Melphalan 140 mg/m2, ATG Grafalon 3 × 10 mg/kg  > 99% MMF (until day + 28) No None Abbreviations: BM, bone marrow; CMV, cytomegalovirus; CsA, cyclosporine A; Css, concentration at steady-state; Cya, cyclosporine A; GVHD, graft versus host disease; i.v., intravenous; MAC, myeloablative conditioning; MMF, mycophenolate mofetil; MMFD, mismatched family donor; MRD, matched related donor; Mtx, methotrexate: MUD, matched unrelated donor; N.a., not available; PBSC, peripheral blood stem cells; RIC, reduced-intensity conditioning; TDM, therapeutic drug monitoring; Tx, transplantation *According to HSCT recommendations by the EWOG-MDS study group Clinical Outcome The clinical outcome of all 14 patients is presented in Table 1. Two adult patients (18%) died after allo-HSCT. Patient 5 had persistent thrombocytopenia and died of a cerebral hemorrhage 8 months post transplantation. Patient 11 developed respiratory failure due to cGVHD in the lungs and received a bilateral pulmonary transplant 5 years post allo-HSCT. However, he developed chronic pulmonary rejection and died 2 years after lung transplantation and 7 years after allo-HSCT. The mean follow-up of the nine patients still alive after allo-HSCT is 26 months (range 3–78 months). The incidence of aGvHD and cGVHD among the eleven transplanted patients was 25% and 33%, respectively, all occurring in patients > 18 years of age (Table 4). None of the pediatric patients had experienced aGVHD or cGVHD, serious infectious complications, or any serious or unexpected transplant-related acute or late toxicity. Their transplantation courses were uneventful and did not principally differ from MDS patients without germline disease-causing GATA2 variants. One patient is listed for allo-HSCT (Patient 14) and two patients are followed regularly in the out-patient clinic (Patients 3 and 8). Discussion This retrospective study describes clinical features and outcome of 14 patients from ten families diagnosed with GATA2 deficiency in Norway. The main findings were as follows: (i) We found a diverse clinical phenotype dominated by cytopenia (13/14), myeloid neoplasia (MDS/AML) (10/14), warts (8/14), and hearing loss (7/14). (ii) We observed two novel clinical features multiple aneurysms of small vessels (n = 1) and early graying (n = 1) that could be associated with GATA2 deficiency. (iii) The majority of patients (11/14) had already undergone allo-HSCT at the time of our analysis, illustrating the need for allo-HSCT in a large proportion of patients with GATA2 deficiency in Norway, and most likely in other countries. (iv) Genetic testing should be offered to first-degree relatives, particularly children, to identify individuals with GATA2 deficiency that need close surveillance. Genetic testing performed as soon as a clinical suspicion is raised, increases the likelihood of an early and correct molecular diagnosis. In Norway, exome-based panels including GATA2 are offered as a routine diagnostic laboratory service for constitutional hematological disorders and PID [21]. However, this approach will not detect all GATA2 variants. Germline variants located in the intronic transcriptional enhancer elements, the cis-acting E-box/GATA and ETS motifs within intron 5 (NM_001145661.1), may cause GATA2 deficiency[34]. Supplementary Sanger sequencing of the enhancer element sequences in intron 5[34] was only performed in one of the laboratories (see Supplementary Methods). Despite supplementary copy number variant calling from exome data, with chromosomal microarray and MLPA only in selected patients (Supplementary Methods), some structural variants may go undetected. In addition, synonymous disease-causing GATA2 variants resulting in selective loss of mutated RNA were recently reported [35]. Disease-causing intronic variants, small intragenic variants, such as structural variants and synonymous variants, may also have escaped detection and hence, individuals with GATA2 deficiency may have been overlooked. With the recent introduction of targeted NGS panels in the work-up of myeloid neoplasms searching for somatic GATA2 variants, unexpected germline variants can be identified, which may reveal the underlying constitutional cause of the myeloid disease and increase the prevalence of known GATA2 deficiency. Improving the NGS panels targeting both germline and somatic variants by deeper coverage of the whole genes including important non-exonic regions, better algorithms for detection of structural variants, and attention to rare synonymous variants may enhance the identification of GATA2 deficiency. In this case series, we found two clinical features that have not yet been described in GATA2 deficiency, namely small vessel aneurysm and early graying. Multiple small vessel aneurysms found in Patient 10 may be secondary to vasculitis, or could also represent a novel vascular feature associated with GATA2 deficiency. In fact, it has been suggested that alteration in GATA2 expression may be of importance for vascular integrity[36]. The observation of premature graying may be a coincidence. In the absence of telomere biology disorders as in the present patients, one may speculate if this feature may reflect a GATA2-linked autoimmune phenomenon which has gone undetected. One of the aims of this retrospective study was to describe the clinical characteristics, GATA2 variants, and other molecular variants representing risk factors for clonal evolution that could aid us in the difficult decision regarding: “Who and when to transplant”? The high proportion of patients (79%) that had already undergone allo-HSCT in our cohort was somewhat surprising. Donadieu et al. have published the largest cohort of patients with GATA2 deficiency, and found that only 28 patients (35%) of 79 patients had undergone allo-HSCT. However, this low percentage of allo-HSCT did not correspond with the severity of the disease in this cohort. At the age of 40, the authors reported a mortality rate of 35% and a hematological malignancy rate of 80%[14]. The high proportion of allo-HSCT in our study may have been influenced by increased awareness of negative prospective clinical markers of GATA2 deficiency. Based on the findings from our study and the high morbidity and mortality rate reported by Donadieu et al., it is clear that these patients need to be monitored closely. Ideally, allo-HSCT should be performed before they develop malignancies (both solid tumors and hematological malignancies)[37] or severe/recurrent infections causing organ failure. In our opinion, a history of disseminated viral infection, aggressive HPV infection (particular with dysplasia), or myeloid clonal disease is clear indication to consider allo-HSCT[14]. First-degree relatives with a severe outcome of the disease may further strengthen the indication for an early allo-HSCT in symptomatic patients with GATA2 deficiency. Overall, the decision to perform an allo-HSCT requires careful weighing of potential gain (restore immune function; diminish the risk of hematological malignancies) versus possible transplant complications, including GVHD and transplant-related mortality. This is particularly challenging given the lack of genotype–phenotype correlation. Keeping in mind that the observation time is short for some of the patients in our study, the survival rate after allo-HSCT was 82% (9/11). In patients with GATA2 deficiency, previous publications have reported 86% survival 2 years after HSCT (n = 22)[16], 73% and 62% survival 1 and 5 years after HSCT, respectively (n = 28)[14], 72%, 65%, and 54% survival 1, 2, and 4 years after HSCT, respectively (n = 21)[9], and 57% 3, 5 years after HSCT (n = 14)[38]. These cohorts are, however, not necessarily comparable in terms of severity of disease and conditioning regimen. Two children were diagnosed with GATA2 deficiency after family screening. The hematological surveillance of one of these children led to detection of hematological abnormalities consistent with MDS and, in the end, a timely allo-HSCT. We therefore recommend genetic testing of children of affected adults and hematological surveillance of individuals with known pathogenic germline GATA2 variants. This includes annual BM investigations with morphological and cytogenetic evaluations, and testing with NGS myeloid panel to screen for somatic occurring molecular drivers of malignancies. Monosomy 7 and trisomy 8 have been reported by others to be the major cytogenetic aberrations in hematopoietic cells of patients with GATA2 deficiency and MDS[15]. Advanced MDS disease and monosomy 7 have been related to worse outcome, especially for pediatric patients with GATA2 germline disease[20]. We found monosomy 7 only in ¼, but trisomy 8 in half of the karyotyped patients, which is in line with a previous published study by McReynolds et al. [39]. Two patients in our GATA2 cohort had monosomy 7 and trisomy 8, both were children. Some of the other somatic variants in our cohort occurred in genes previously reported to be mutated in GATA2 deficiency with MDS, such as ASXL1[40], STAG2, SEPTBP1, and RUNX1[41, 42]. Interestingly, one adult patient with GATA2 deficiency and MDS-related AML had a der(1;7) in the leukemic clone, a translocation that has recently been shown to be enriched in pediatric MDS patients with germline GATA2 mutations[43]. However, our number of patients are too small to determine possible genotype–phenotype correlations related to clonal disease progression. Previous reports have suggested that plasma levels of FLT3LG can be used as predictor of hematological disease in GATA2 deficiency, and used in clinical monitoring post-HSCT [25]. Unfortunately, we lack serum or plasma samples taken before and after HSCT in our cohort, and the role of FLT3LG as a disease progression marker could be explored in future studies. Our main conclusion of this study is that the majority of patients with symptomatic GATA2 deficiency will need allo-HSCT, and close surveillance of these patients is important to find the “optimal window” for allo-HSCT. We advocate a more offensive approach to allo-HSCT than previously described. Supplementary Information Below is the link to the electronic supplementary material.Supplementary file1 (DOCX 55 KB) Acknowledgements Our gratitude goes to the Section for Cancer Cytogenetics who karyotyped 9 of the 14 patients reported here (Table 2). Three of the reported families have been followed up at the Department of Medical Genetics, University Hospital of North Norway. We want to thank our colleagues, MD Gry Hoem, MD Hilde Yttervik, MD Specialist in medical genetics and pediatrics Marie Falkenberg Smeland, and MD PhD Øyvind Holsbø Hald at the Department of Medical Genetics, University Hospital of North Norway, Tromsø, Norway, for their clinical work and genetic counselling. The three pediatric patients are enrolled in the registry of the European Working Group of MDS in Childhood (EWOG-MDS; ClinicalTrials.gov Identifier: NCT00662090). Author Contribution SFJ, JB, AEM, PA, AS-P, TGD, and IN wrote the paper. The genetic analyses were performed by AS-P, MAK, HS, ØH, SS, and EL. SFJ, JB, AEM, EG, YF, CA, AB, IH, TF, BF, AS-P, TGD, and IN collected clinical data. SS did the bone marrow analyses. All authors critically revised the manuscript for important intellectual content and approved the final version of the manuscript. Funding Open access funding provided by University of Oslo (incl Oslo University Hospital). Data Availability Upon request. Code Availability Not applicable. Declarations Ethics Approval Five of the patients had previously consented to be part of a genetic PID research project approved by the regional ethical committee (REC. 2014/1270–1), three patients were diagnosed with MDS in childhood and consented to be registered into the EWOG-MDS-2006 study (2015/1651/REC Nord), and all adult patients who underwent allo-HSCT had consented to publication of data (REC 11909). Studies were performed according to the Declaration of Helsinki. Consent to Participate All adult living patients signed a written informed consent for publication. For children < 18 years, consent was given by their parents. Consent for Publication Obtained. Conflict of Interest The authors declare no competing interests. Abbreviations aGVHD Acute graft versus host disease allo-HSCT Allogeneic hematopoietic stem cell transplantation AML Acute myelogenous leukemia ARDS Acute respiratory distress syndrome BM Bone marrow cGVHD Chronic graft versus host disease GVHD Graft versus host disease HPV Human papilloma virus MDS Myelodysplastic syndrome MRD Matched related donor MUD Matched unrelated donors NGS Next-generation sequencing PID Primary immunodeficiency VAF Variant allele frequency WES Whole exome sequencing ZNF1 Zink finger 1 ZNF2 Zink finger 2 Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Asbjørg Stray-Pedersen, Tobias Gedde-Dahl and Ingvild Nordøy contributed equally

8 MG/KG

DrugDosageText

CC BY

34893945

2021-12-10

What was the dosage of drug 'FLUDARABINE PHOSPHATE'?

A Nationwide Study of GATA2 Deficiency in Norway-the Majority of Patients Have Undergone Allo-HSCT. OBJECTIVE GATA2 deficiency is a rare primary immunodeficiency that has become increasingly recognized due to improved molecular diagnostics and clinical awareness. The only cure for GATA2 deficiency is allogeneic hematopoietic stem cell transplantation (allo-HSCT). The inconsistency of genotype-phenotype correlations makes the decision regarding "who and when" to transplant challenging. Despite considerable morbidity and mortality, the reported proportion of patients with GATA2 deficiency that has undergone allo-HSCT is low (~ 35%). The purpose of this study was to explore if detailed clinical, genetic, and bone marrow characteristics could predict end-point outcome, i.e., death and allo-HSCT. METHODS All medical genetics departments in Norway were contacted to identify GATA2 deficient individuals. Clinical information, genetic variants, treatment, and outcome were subsequently retrieved from the patients' medical records. RESULTS Between 2013 and 2020, we identified 10 index cases or probands, four additional symptomatic patients, and no asymptomatic patients with germline GATA2 variants. These patients had a diverse clinical phenotype dominated by cytopenia (13/14), myeloid neoplasia (10/14), warts (8/14), and hearing loss (7/14). No valid genotype-phenotype correlations were found in our data set, and the phenotypes varied also within families. We found that 11/14 patients (79%), with known GATA2 deficiency, had already undergone allo-HSCT. In addition, one patient is awaiting allo-HSCT. The indications to perform allo-HSCT were myeloid neoplasia, disseminated viral infection, severe obliterating bronchiolitis, and/or HPV-associated in situ carcinoma. Two patients died, 8 months and 7 years after allo-HSCT, respectively. CONCLUSIONS Our main conclusion is that the majority of patients with symptomatic GATA2 deficiency will need allo-HSCT, and a close surveillance of these patients is important to find the "optimal window" for allo-HSCT. We advocate a more offensive approach to allo-HSCT than previously described. pmcIntroduction GATA2 deficiency is a rare primary immunodeficiency (PID), first described in 2011[1–3] that has become gradually more recognized due to improved molecular diagnostics and increased clinical awareness. GATA2, as a “master” transcription factor, plays a critical role in hematopoietic development[4]. Through cooperative processes that include other transcription factors, it controls the transition from hemogenic endothelium to hematopoietic stem cells and is required for survival and self-renewal of these cells[5]. GATA2 is also important for other tissue-forming stem cells, e.g., in the inner ear[6]. The heterozygous variants causing GATA2 deficiency are located both in coding, non-coding and enhancer regions[7]. The disease-causing loss-of-function variants can be localized across the gene. These variants can lead to defective DNA-binding capacity of the transcription factor and may cause disease through haploinsufficiency of the functional protein[5, 8]. Missense variants within the zink finger 2 (ZNF2) domain are the most frequent germline disease-causing GATA2 variants [9]. It has been estimated that approximately 1/3 of the patients have an autosomal dominant inherited disease-causing variant[10], whereas the remaining have a de novo GATA2 variant[7]. Of note, somatic variants in GATA2 are known to be drivers of myeloid neoplasia in adults. Such variants are diverse, may cause gain-of-function effects, and be located across the whole gene. This includes missense variants in the zink finger 1 (ZNF1) domain, which has not been observed in constitutional GATA2 deficiency[8]. Typically, GATA2 deficiency becomes clinically apparent in late childhood to early adulthood. The phenotype is heterogeneous, without any clear genotype–phenotype correlation, and with an incomplete clinical penetrance[11]. Symptoms may include recurrent or severe infections, warts, cytopenia (including monocytopenia), lymphedema, alveolar proteinosis, and malignant myeloid disease[9]. Infectious complications in GATA2 deficiency are likely due to deficiency of monocytes, NK cells, and B-lymphocytes as well as defective innate immune responses, including impaired type I interferon production[12]. This leads to both increased susceptibility to viral infections (e.g., human papilloma virus [HPV, warts] and herpes virus infections), non-tuberculous mycobacteria, and to more common bacterial respiratory infections. Hearing loss is a common clinical feature of GATA2 deficiency and is related to the critical role of GATA2 in vestibular morphogenesis of semicircular ducts and generation of the perilymphatic space around the inner ear’s semicircular canals[6, 13]. A substantial proportion of patients develop immunodeficiency, myelodysplastic syndrome (MDS), or acute myelogenous leukemia (AML) as initial manifestation[9, 14]. GATA2 deficiency is considered the most common hereditary predisposition to pediatric MDS[15]. The only cure for GATA2 deficiency is allogeneic hematopoietic stem cell transplantation (allo-HSCT) and results are encouraging[16–20]. However, the main challenge is deciding who and when to transplant due to the complexity and inconsistency of phenotype-genotype correlation in GATA2 deficiency[9]. To further elucidate this important issue, we present detailed clinical and molecular characteristics, treatment, and outcome of 14 Norwegian patients with germline GATA2 variants diagnosed between 2013 and 2020. The main aim of our study was to explore if detailed clinical, genetic, and bone marrow (BM) characteristics could predict end-point outcome such as death and allo-HSCT in patients with GATA2 deficiency. Methods Identification of Patients and Clinical Characteristics The first aim of this study was to obtain a complete overview of all patients with known GATA2 deficiency in Norway. For this purpose, a network of clinical immunologists, hematologists, pediatricians, and geneticists at Oslo University Hospital (OUH) collected clinical and laboratory data on patients with GATA2 deficiency at their institution. In addition, all medical genetics departments in Norway were contacted to identify any additional GATA2 deficient individuals. Clinical information, genetic variants, treatment, and outcome were subsequently retrieved from the patients’ medical records. Patients were enrolled into the study at OUH where most of the data was obtained, while supplemental data from Patient 3 was collected at the University Hospital of North Norway, Tromsø. Informed Consent Five of the patients had previously consented to be part of a genetic PID research project approved by the regional ethical committee (REC. 2014/1270–1), three patients were diagnosed with MDS in childhood and consented to be registered into the EWOG-MDS-2006 study (2015/1651/REC Nord), and all adult patients who underwent allo-HSCT had consented to publication of data (REC 11,909). In addition, due to the detailed clinical information published herein, all adult living patients signed an additional written informed consent for publication of their data and was given the opportunity to review the manuscript prior to publication. For children < 18 years, consent was given by their parents. This is in line with the recommendation given by the Ethical Constitutional board at OUH. Genetic Analyses Whole exome sequencing (WES) with in silico filtering for genes causing primary immunodeficiency disorders was performed in the probands and affected relatives as part of a routine laboratory service (Patients 2, 3, 8, 9, 10, 12, and 14) or on a research basis (Patients 4, 5, 6, and 7) as previously described (Supplemental methods)[21]. Patient 1 had severe cytopenia (Table 1), and the first attempt to extract DNA from peripheral blood was not successful. A skin biopsy was therefore performed to extract DNA from fibroblasts. In parallel, peripheral blood (from puncture of the fingertip) was applied directly to a Guthrie filter card, and by using multiple filter card punches, enough DNA was extracted to run next-generation sequencing (NGS) with an amplicon-based targeted panel for constitutional variants in PID genes (Supplemental methods). By using this rapid amplicon-based method, the molecular result was available within 3 working days[22]. DNA later extracted from fibroblasts confirmed the GATA2 variant by Sanger sequencing. Also, for Patient 13, who had advanced MDS with pancytopenia, the NGS results were available within 3 working days, with parental testing performed in parallel to evaluate as fast as possible the availability of a healthy unaffected matched related donor (MRD).Table 1 Clinical characteristics and outcome in patients with GATA2 deficiency Patient no Family Sex Current age Age at onset of symptoms/age at genetic diagnosis Infections Hearing loss Hematologic abnormalities Autoimmunity/immune dysregulation Miscellaneous HSCT, age Outcome Viral Bacterial 1 A (father of P2 and P3) M 44y 5y/41y HPV: warts HSV: disseminated disease Ear infections as a child Yes Hypoplastic BM: cytopenia, trilinear hypoplasia No No 41y Alive 2 A (son of P1) M 16y 7y/14y HPV: warts No No MDS-EB-1 No No 16y Alive 3 A (daughter of P1) F 13y 8y/9y HPV: warts No No No No No ND Alive 4a B (monozygotic twin to P5) F 45y 21y/38y HPV: warts, carcinoma in situ EBV: prolonged viremia Recurrent respiratory inf Suppurative skin infection after BCG * Yes No Progressive obliterating bronchiolitis, lupus-like syndrome Miscarriage 39y Alive 5a B (monozygotic twin to P4) F † (39y) 24y/38y HPV: warts, carcinoma VZV and EBV: prolonged viremia Recurrent respiratory inf Suppurative skin inf. after BCG* Yes MDS-MLD (hypoplastic) Progressive obliterating bronchiolitis, lupus-like syndrome DVT × 2 Squamous cell carcinoma in the cervix, rectum, and anus 39y Deceased 8 m post-HSCTb 6a C M 31y 11y/26y No Recurrent respiratory inf Yes MDS-MLD (hypoplastic) - Fever of unknown origin, recurrent pneumothorax 29y Alive 7a D F 23y 6y/17y HPV: warts Recurrent respiratory inf No MDS-MLD (hypoplastic) Interstitial lung disease Lymphedema, acne, rosacea, rash, fatigue 22y Alive 8 E F 56y 0y/53y No No No Hypoplastic BM No Lymphedema, premature graying ND Alive 9 F F 24y 15y/23y No No No AML with MDS-related changes Erythema nodosum DVT, PE, juvenile myoclonic epilepsy, epicanthic fold 23y Alive 10 G (sibling to P11) F 32y 6y/31y HPV: warts, cervix dysplasia Recurrent respiratory inf Yes MDS-MLD No Aneurysm of small vessels, hidradenitis suppurative, liver lesions: focal nodular hyperplasia 32y Alive 11 G (sibling to P10) M † (34y) 22y/PM No Recurrent skin and respiratory inf No MDS-MLD No Acne, rosacea, necrotizing fasciitis, pilonidal cysts, skin infections, ulcerations 27y Deceased 7y post-HSCTc 12 H F 19y 14y/14y No No Yesd MDS-RCC (hypoplastic) BPD/Asthma Born premature (week 26 + 5), BPD 14y Alive 13 I M 13y 9y/11 y HPV: warts No No MDS-EB1 Asthma Chronic skin abscesses, congenital ptosis 11y Alive 14 J F 31 23y/31y No No Yes MDS-SLD (hypoplastic) No Born prematurely (week 25), cerebral palsy, congenital hip dysplasia Planned Alive Abbreviations: AML, acute myeloid leukemia; BCG, bacille Calmette Geurin; BM, bone marrow; BPD, bronchopulmonary dysplasia, CT, computer tomography, HPV, human papilloma virus; HSCT, hematopoietic stem cell transplantation; Inf., infection; m, months; MDS, myelodysplastic syndrome; MDS-EB1, MDS with excess of blasts type 1; MDS-MLD, MDS with multilineage dysplasia; MDS-SLD, MDS with single lineage dysplasia; MDS-RCC, MDS subtype refractory cytopenia of childhood; ND, not done, PM, post mortem; VTE, venous thromboembolism; y, years, †; deceased aThese patients have previously been published in Stray-Pedersen, Sorte et al. 2016 (Patient 4 was 84.1, Patient 5 was 84.4, Patient 6 was 88.1, and Patient 7 was 86.1)[21] bThe patient was doing well after HSCT, but died unexpectedly of a cerebral hemorrhage cThe patient underwent lung transplantation for chronic lung GVHD 58 months after HSCT, and died of chronic lung rejection 26 months after bilateral lung transplantation dThe patient has reduced hearing, but this was confirmed after HSCT. Her hearing loss may be due to the disease-causing GATA2 variant, but may also be secondary to complications of HSCT therapy, e.g., aminoglycosides The molecular diagnosis in Patient 11 was confirmed post mortem using a BM sample collected prior to allo-HSCT (Table 2). Methods for testing for somatic occurring sequence variants on DNA extracted from whole blood or BM, and methods for testing chromosomal aberration on BM cells are described in Supplemental methods.Table 2 Constitutional and acquired genetic findings in patients with GATA2 deficiency Patient no Hematological abnormalities Constitutional heterozygous variants in GATA2, NM_001145661.1, predicted protein effect, domain, occurrence, novelty, and reference Somatic variants, predicted protein effect, VAF in BM/blood (prior to HSCT) Karyotype in BM (closest to HSCT)d  + 8  − 7 1 Hypoplastic bone marrow c.1062_1064del, p.Thr358del, in-frame exon 6, ZNF2, likely de novo, novel variant Unknown 46,XY[25/25] No No 2 MDS-EB1 c.1062_1064del, p.Thr358del, in-frame exon 6, ZNF2, paternal inherited, novel variant Acquired germline donor variant in GATA2 Post-HSCTb: c.1215G > T, p.(Lys405Asn) missense exon 7, outside and distal to ZNF2 domain, VAF: 49,5% BM NM_001145661.1 (GATA2): c.1168_1170del, p.(Lys390del), in-frame exon 7, in ZNF2, VAF: 40.2% BM NM_006758.2(U2AF1): c.470A>G, p.(Gln157Arg) VAF: 44,0% BM 46,XY,-7 + 8[20/20] Yes Yes 3 No c.1062_1064del, p.Thr358del, in-frame exon 6, ZNF2, paternal inherited, novel variant Unknown Unknown N.a N.a 4 a No c.1143 + 5G > A, p.Asn381fs*20, splice defect intron 6, ZNF2, de novo, novel, reported by usc)[21] Unknown Unknown N.a N.a 5 a MDS-MLD (hypoplastic) c.1143 + 5G > A, p.Asn381fs*20, splice defect intron 6, ZNF2, de novo, novel, reported by usc[21] None 46,XX[18/18] No No 6a MDS-MLD (hypoplastic) c.1078 T > A, p.Trp360Arg, missense exon 6, ZNF2, de novo, variant previously reported by others[23] Unknown 46,XY[25/25], but FISH MYC(8q24): + 8 in 14/303 metaphases Yes No 7a MDS-MLD (hypoplastic) c.1061C > T, p.Thr354Met, missense exon 6, ZNF2, de novo, but a recurrent GATA2 variant[21, 24, 25] NM_001042749.2(STAG2): c.2534-2A > G, predicted splice variant with loss of acceptor site, Chr.X, VAF: 11.7% blood 47,XX, + 8[4/10]/46,XX[6/10] Yes No 8 Hypoplastic bone marrow c.1017 + 1G > T, loss of donor splice site, splice defect intron 5, ZNF1, both parents deceased and not tested, novel variant Unknown 46,XX[25/25] No No 9 AML with MDS-related changes c.163C > T, p.Gln55*, nonsense exon 3, TAD domain, likely de novo (see pedigree), novel variant VAF:48.7% in BM, 49,4% in buccal swap NM_001754.4(RUNX1): c.593A > G,p.(Asp198Gly) VAF:15% BM NM_156039.3(CSF3R): c.2326C > T, p.(Gln776*) VAF: 12.5% BM NM_032458.2(PHF6): c.309C > G, p.(Tyr103Ter) VAF:12.0% BM NM_033632.3(FBXW7): c.1513C > T, p.(Arg505Cys), VAF:11.7% BM 46,XX, der(1;7)(q10;p10), + 1[11/20]/46,XX [9/20] No No 10 MDS-MLD c.1084C > T, p.Arg362*, nonsense exon 6, ZNF2, likely inherited, variant previously reported by others[15, 26–28] NM_001123385.1(BCOR): c.529_530del, p.(Ser177ProfsTer8), VAF: 23.0% BM 49∼50,XX, + 6, + 8, + 21? + 21[cp7/8]/46,XX[1/8] Yes No 11 MDS-MLD c.1084C > T, p.Arg362*, nonsense exon 6, ZNF2, likely inherited, variant previously reported by others[15, 26–28] NM_015338.5(ASXL1): c.2324 T > G, p.(Leu775Ter) VAF: 20.5% BM NM_001042749.1(STAG2): c.2990 T > A, p.(Leu997Ter) VAF: 9.6% BM 47∼48,XY, + 8[10/15],der(16)t(1;16)(q21;q24[10/15], + der(16)t(1;16)[1/15], + 21[6][cp11/15]/46,XY[3/15] Trisomy 8, evolving to unbalanced 1;16 translocation and later Trisomy 21 Yes No 12 MDS-RCC c.1098_1100delGGA, p.Asp367del, in-frame exon 6, ZNF2, de novo, novel variant None 46,XX,-7, + 8[15/20] Yes Yes 13 MDS-EB1 c.1021_1024insGCCG, p.Ala342Glyfs*43, frameshift exon 6, ZNF1 de novo, variant previously reported[29] NM_015338.5(ASXL1): c.1854dupA, p.(Ala619SerfsTer16), VAF:17.0%, BM NM_015559.2 (SETBP1): c.2612 T > C, p.(Ile871Thr), VAF: 16.3%, BM 45,XY,-7[12/12] No Yes 14 MDS-SLD (hypoplastic) c.1114G > A, p.(Ala372Thr), missense exon 6, ZNF2, variant previously reported [14] NM_001042749.1(STAG2): c.707del; p.(Asn236IlefsTer20) VAF: 5.1%, BM 46,XX[25/25] No No Abbreviations: AML, acute myeloid leukemia; BM, bone marrow; Chr, chromosome; HSCT, hematopoietic stem cell transplantation; MDS, myelodysplastic syndrome; MDS-EB1, MDS with excess blasts type 1; MDS-MLD, MDS with multilineage dysplasia; MDS-RCC, MDS subtype refractory cytopenia of childhood; MDS-SLD, MDS with single lineage dysplasia; N.a., Not applicable; TAD, N-terminal transactivation domain, ZNF2, Zinc finger 2 domain in GATA2 protein; VAF, variant allele frequency aThese patients have previously been published in Stray-Pedersen, Sorte et al. (2016) (Patient 4 was 84.1, Patient 5 was 84.4, Patient 6 was 88.1, and Patient 7 was 86.1)[21] bThis disease-related GATA2 variant was detected in a routine BM at day + 28 post-HSCT; it turned out to be donor-derived (from a MUD) cWES identified a potential splicing variant in GATA2 (c.1143 + 5G > A) in Patient 4. The variant was predicted (Alamut®) to inactivate the donor site of GATA2 exon 5. PCR amplification of GATA2 exon 4 to 7 on cDNA showed that most transcripts were normally spliced resulting in a main product of ~ 400 bps, as seen in the normal control. A slightly longer PCR product including 64 bps of intron 6 sequence via a cryptic donor site in intron 6 (NM_001145661.1), was observed in the sample from Patient 4, but not in the control (see Supplementary information). Sanger sequencing identified the GATA2 splicing variant in the proband’s deceased monozygotic twin (Patient 5). Details described in Supplemental Figure E6 in Stray-Pedersen, Sorte et al. (2016)[21] dNomenclature according to ISCN (The International System for Human Cytogenetic Nomenclature) 2020 guidelines Results Characteristics of Patients Between 2013 and 2020, ten index cases, or probands, and four additional symptomatic patients with germline GATA2 variants were identified (9 females, 5 male, Table 1). Five adult patients were diagnosed by infectious disease specialists (Patients 1, 4, 5, 10, and 11) where infections (mostly HPV infection/warts and recurrent bacterial airway infections) were prominent symptoms. Four additional adult patients were identified by hematologists (Patients 6, 8, 9, and 14), where three were referred with pancytopenia and one patient had AML (also with pancytopenia). Three patients were diagnosed by pediatricians, two patients with MDS (Patients 12 and 13) and one patient with extensive warts and NK-/B-cell deficiency (Patient 7). Additionally, two patients with GATA2 deficiency were identified after family screening (Patients 2 and 3). We did not detect any asymptomatic individuals with GATA2 deficiency in this study. The mean age for debut of symptoms, that we considered related to GATA2 deficiency, was 12 years (range 0–24 years, Supplemental Table S1). The median time from these symptoms to a diagnosis of GATA2 deficiency was 11 years, range 0–53 years (Supplemental Table S1). Retrospectively, hearing loss, warts, and skin manifestations were the most common early symptoms, which in some patients became apparent many years before the genetic diagnosis of GATA2 deficiency was made (Supplemental Table S1). A summary of the patients’ clinical characteristics is given in Table 1. Viral infections such as HPV-associated warts were common, affecting eight patients. In addition, two patients had disseminated BCG infections (after vaccination), and one patient had a life-threatening disseminated HSV infection (originating from genitalia and disseminating to CNS and liver). Two patients experienced prolonged EBV and/or Varicella zoster viremia. Furthermore, six patients had recurrent bacterial airway infections. Interestingly, one patient had early graying (Patient 8), with normal telomere length, and one patient had multiple aneurysms of small vessels (coronary arteries, axillary arteries, and an iliac artery; Patient 10), which both represent clinical characteristics not previously described in GATA2 deficiency. In Patient 10, Varicella zoster infection was excluded as a cause of vasculitis with negative VZV PCR in blood. In addition, two patients had obliterating bronchiolitis (Patients 4 and 5), which has been reported in only one previous patient with GATA2 deficiency [30]. Affected cell lineages and immunoglobulin levels prior to allo-HSCT are listed in Table 3. As expected, the majority of patients had decreased levels of monocytes (11/14) and one patient had increased levels of monocytes (Patient 12). In addition, decreased levels of B cells (10/11) and NK cells (9/11) were common findings (three patients did not have NK- and B cells measured before allo-HSCT).Table 3 Immunoglobulin levels and affected cell lineage in peripheral blood prior to HSCT Patient no Affected cell lineage (normal range) Immunoglobulins Time before HSCT (months) CD19 + , cells × 106/L (100–500) NK cells × 106 /L (100–400) CD3 +  T- cells × 106/L (800–2400) CD4 +  T- cells × 106/L (500–1400) CD8 +  T-cells × 106/L (200–2000) Monocytes  × 109/L (0.2–0.8) Neutrophils  × 109/L (1.5–7.3) IgG g/L (6.9–15.7) IgG2 g/L (1.69–15.7) 1 0 0 118 43 64 0.0 0.2 5.9 ND 5 2 10 10 550 200 300 0.3 2.9 9.0 ND 4.5 3 160 150 1930 780 950 0.2 2.4 8.4 ND NA 4  < 10 15 349 185 145 0.0 3.4 9.5 1,24 17 5 6 1 259 134 66 0.0 6.0 13.0 0.79 221 6 2 0 809 495 326 0.1 0.5 44.32 2.3 4 7 18 19 913 545 341 0.0 2.8 14.0 2.60 3 83 40 2 1335 546 761 0.1 1.9 9.7 ND NA 9 ND ND ND ND ND 0.0 4.2 16.2 - 1 10 70 206 247 134 108 0.1 1.7 18.5 0.79 6 11 ND ND ND ND ND 0.0 0.9 9.5 2.11 10 12 ND ND ND ND ND 1.7 2.6 8.1 ND 1 134 28 13 1073 633 417 0.2 0.3 12.8 ND 0.5 14 8 11 676 323 341 0.1 1.2 11.9 1.14 NA Abnormal values are given in bold 1On Prednisolone 20 mg a day when these samples were taken 2Hypergammaglobulinemia on IVIG due to IgG2 deficiency 3The values are from the time at diagnosis of GATA2 deficiency 3 years ago 4The reference values for Patient 13 who was 12 years old at the time of HSCT are CD19 200–600, NK 70–1200, CD3 800–3500, cd4 400–1200, cd8 200–1200 (given in cells × 106/L) and for IgG the normal reference value was 6.1–14.9 g/L ND, not done; NA, not applicable Germline GATA2 Variants and Somatic Variants in Other Genes Ten different GATA2 pathogenic, or likely pathogenic, variants were found in 14 patients (Table 2). All identified constitutional GATA2 variants, except one, were located in the ZNF2 domain, corresponding to or in close proximity to exons 5 and 6 (Table 2). Three nonsense variants (p.Ala342Glyfs*43, p.Arg362*, p.Asn381fs*20), one + 1 splicing variant, three missense variants (p.Thr354Met, p.Trp360Arg, and p.Ala372Thr), all previously reported to be disease-causing[14, 23–25], and two novel in-frame deletions (p.Thr358del, p.Asp367del) were found. The variant located in exon 3, outside the ZNF2 domain, was a nonsense variant (c.163C > T, p.Gln55*). It was initially identified by the NGS myeloid panel with variant allele frequency (VAF) 49% in the DNA from the patient’s BM and later verified to be germline with VAF 49% in a buccal DNA sample (Patient 9, Table 2). Patient 2 had a paternal inherited in-frame deletion, c.1062_1064del (p.Thr358del), in the ZNF2 domain, and a somatic in-frame deletion, c.1168_1170del (p.Lys390del), with a fairly high VAF, 40.2% in BM. As expected, these two in-frame deletions were no longer detectable after allo-HSCT. Surprisingly, in the first post-transplant BM sample at day + 28, we detected another acquired GATA2 variant, c.1215G > T (p.Lys405Asn) with VAF 49.5%. This missense mutation variant, affecting an amino acid located C-terminal to the ZNF2 domain, is a variant of unknown clinical significance. It is most likely a rare benign variant, which in retrospect was confirmed to be constitutional in the unrelated BM donor (Table 2). It is evaluated to ACMG category 3 minus, since altogether 5 heterozygote individuals with the same amino acid change p.Lys405Asn are reported in GnomAD (v.2.1.1)[31]. As far as we know, missense variants located outside the ZNF2 domain rarely represent constitutional susceptibility to development of myelodysplasia. One exception is the p.Ser447Arg located C-terminal of the ZNF2 domain[32], while no other missense variants outside the ZNF2 domain are currently defined as pathogenic or likely pathogenic in ClinVar (www.clinvar.com) as of year 2021. Karyotype abnormalities and somatic variants in other genes observed in the 14 patients are presented in Table 2. Trisomy 8 (n = 6), monosomy 7 (n = 3), STAG2 variants (n = 3), ASXL1 variants (n = 2), a combination of somatic variants in RUNX1/CSF3R/PHF6/FBXW7 (n = 1), and variants in the following MDS genes[33] were observed once in separate individuals: BCOR, SETBP1, U2AF1, and somatic GATA2. One adult GATA2 deficient patient who developed AML had an unbalanced translocation der(1;7) in the leukemic clone. Since GATA2 deficiency is considered the most common hereditary predisposition to pediatric MDS, we estimated the proportion of pediatric patients diagnosed with MDS in Norway that had a germline GATA2 variant in the same time period (2013–2020). We found that three out of 14 pediatric patients diagnosed with MDS (21%) had GATA2 deficiency. Of note, these are 14 pediatric MDS patients and not the same cohort of 14 GATA2 deficient patients described above (except three overlapping pediatric patients, Patients 2, 12, and 13, with MDS). Two of the 3 pediatric GATA2 deficient patients with MDS had both monosomy 7 and trisomy 8 in their bone marrow cells. Patients 4, 5, 8, 9, 10, and 11 from Family B, E, F, and G had frameshift or other definitive loss-of-function variants, while Patients 1, 2, 3, 6, 7, 12, 13, and 14 from Family A, C, D, H, I, and J had in-frame deletions or missense variants. No specific genotype–phenotype correlations were found in our data set, i.e., regarding debut of symptoms, type and distribution of infections, age of transition to MDS/AML, somatic occurring variants in blood and BM. The severity of the clinical presentations also varied within families. Families and Predictive Genetic Testing The pedigrees of the 10 families are presented in detail in Fig. 1. Patient 1 (Family A) had three apparently healthy children, when he was diagnosed with GATA2 deficiency. After genetic testing of first-degree relatives, we found that two of his children (Patients 2 and 3) had inherited the GATA2 variant. For Patient 2, initial clinical work-up revealed only mild cytopenia and warts. However, within 2 years of follow-up, he developed pancytopenia and transfusion dependency and was diagnosed with MDS-EB1. His sister, Patient 3, has warts as her only clinical manifestation, but will be followed up regularly for development of cytopenia/MDS.Fig. 1 Pedigrees of the ten families, including 14 patients, with known GATA2 deficiency. Solid symbols denote affected status. Individuals marked in gray are deceased and not tested for GATA2 deficiency but are suspected to carry the disease-causing variant. In family G, the mother of Patients 10 and 11 died at age 30 of acute respiratory distress syndrome, 27 years ago. She also had lymphedema since birth. In light of their mother’s medical history, the GATA2 variant is probably maternally inherited. The father is alive and healthy. In family J, the mother of Patient 14 had a combined B and T cell defect, warts, myelodysplastic syndrome, lymphedema, and recurrent respiratory tract infections. She died of vulval cancer at the age of 38. The maternal grandfather of Patient 14 died of acute leukemia at the age of 33. WT, wild-type In Family G, two siblings had the same germline GATA2 variant (Patients 10 and 11). Their mother died 27 years ago, at the age of 30, of acute respiratory distress syndrome (ARDS), of unknown etiology. She also had lymphedema since birth. In light of their mother’s medical history with lymphedema and ARDS, which could be secondary to complications related GATA2 deficiency, the GATA2 variant is probably maternally inherited. Their father is alive and healthy. The deceased mother of Patient 14 (Family J) had a combined B- and T cell defect, warts, MDS, lymphedema, and recurrent respiratory tract infections. At the age of 38 (years), she died of metastatic vulval cancer. The maternal grandfather of Patient 14 died of acute leukemia at the age of 33. Considering the family history, it is very likely that Patient 14 had inherited her germline GATA2 variant from her maternal grandfather via her mother. Both individuals died before GATA2 deficiency was acknowledged as a cause of PID. Her mother’s siblings are now offered genetic counselling/testing for GATA2 deficiency. Allo-HSCT in Patients with GATA2 Deficiency Twelve of 14 (86%) patients with GATA2 deficiency were found to have a clinical indication, cytogenetic findings, and/or molecular findings warranting to proceed to allo-HSCT. As of today, 11 patients have undergone allo-HSCT, whereas one is recently accepted for allo-HSCT (Patient 14). Clinical features that lead to the decision to perform allo-HSCT were previous life-threatening disseminated HSV infection (Patient 1), severe obliterating bronchiolitis and in situ carcinoma (Patients 4 and 5), MDS with cytogenetic abnormalities (monosomy 7) and/or excess of blasts with high likelihood of progression to leukemic transformation (Patients 2, 6, 7, 9, 12, 13, and 14), MDS and warts with high-grade dysplasia (Patient 10), and symptoms of severe immunodeficiency and MDS (Patient 11). Details on the allo-HSCT procedure, including conditioning, donor selection, stem cell source, donor/recipient cytomegalovirus status, donor chimerism, graft versus host disease (GVHD) prophylaxis, and the occurrence of GVHD, are presented in Table 4.Table 4 HSCT details for eleven patients with GATA2 deficiency Patient no Age at HSCT Donor Stem cell source HLA Match CD34 + , × 106/kg CMV status d/r Conditioning regimen & In vivo T-cell depletion Chimerism %Day + 28 GVHD prophylaxis GVHD Complications 1 41 y MUD PBSC 10/10 (11/12) 7.8 -/ +  RIC: Fludarabine 150 mg/ m2, Busulfan 8 mg/kg, ATG Thymoglobulin 4 mg/kg 99% Mtx + CsA No Hemorrhagic cystitis (BK-virus) 2 16 y MUD PBSC 10/10 (10/12) 9.7 -/ - MAC*: Busulfan for 4 days (TDM; Css 825 ng/ml), Cyclophosphamide 120 mg/kg, Melphalan 140 mg/m2, ATG Grafalon 3 × 10 mg/kg  > 99% Mtx + CsA No E. coli sepsis; BK-virus cystitis; mucositis (grade 3) 4 39 y MUD PBSC 10/10 (12/12) 10.6 + / +  RIC: Fludarabine 150 mg/ m2, Busulfan 8 mg/kg. ATG Thymoglobulin 4 mg/kg  > 99% Mtx + CsA Chronic: skin and gut CMV reactivation 5 39 y MUD PBSC 10/10 (11/12) 5.2 +/ - RIC: Fludarabine 90 mg/m2, 2 Gy TBI 99% Mtx + CsA No Enterococcus faecalis sepsis (2 months post-HSCT), prolonged cytopenia, died of intracerebral hemorrhage 8 months post-HSCT 6 29 y MRD PBSC HLA-id sibling 5.4 + / +  RIC: Fludarabine 150 mg/m2, Busulfan 8 mg/kg 98% Mtx + CsA Chronic: liver, oral mucosa and genitalia Cytopenia at day + 33, osteoporosis, compression fractures 7 22 y MUD PBSC 10/10 6.9 -/ - MAC: Fludarabine 150 mg/m2, Treosulfan 42 g/m2, ATG Thymoglobulin 4 mg/kg  > 99% Mtx + CsA Chronic (limited): skin Moraxella nonliquefaciens sepsis on day 0. E. coli sepsis day + 14. Oral mucositis grade IV. PTLD 7 weeks post-HSCT 9 23 y MUD PBSC 10/10 (11/12) 4.3 +/ - MAC: Fludarabine 160 mg/m2, Busulfan 12,8 mg/kg (i.v.), ATG Thymoglobulin 4 mg/kg  > 99% Mtx + CsA Acute GvHD grade I: skin None 10 32 y MUD PBSC 10/10 (11/12) 5.9 -/ - MAC: Fludarabine 150 mg/m2, Treosulfan 42 g/m2, ATG Thymoglobulin 4 mg/kg  > 99% Mtx + CsA Acute GvHD: serositis None 11 27 y MUD PBSC 10/10 (10/12) 10.2 -/ - MAC: Cyclophosphamide 100 mg/kg, Busulfan 16 mg/kg N.a Mtx + CsA Chronic (extensive): gut, eye, and lung Hemorrhagic cystitis, Herpes oesophagitis 12 14 y MUD BM 10/10 (11/12) TNC: 3.5 × 108/kg -/ - MAC*: Fludarabine 160 mg/m2, Treosulfan 42 g/m2,Thiotepa 8 mg/kg, ATG Grafalon 3 × 10 mg/kg day + 84: > 99% Mtx + CsA No Impetigo day + 40 13 11 y MMFD (father) PBSC, TCRab + /CD19 + depletion in vitro Haploidentical 10.3 + / +  MAC*: Fludarabine 160 mg/m2, Thiotepa 10 mg/kg, Melphalan 140 mg/m2, ATG Grafalon 3 × 10 mg/kg  > 99% MMF (until day + 28) No None Abbreviations: BM, bone marrow; CMV, cytomegalovirus; CsA, cyclosporine A; Css, concentration at steady-state; Cya, cyclosporine A; GVHD, graft versus host disease; i.v., intravenous; MAC, myeloablative conditioning; MMF, mycophenolate mofetil; MMFD, mismatched family donor; MRD, matched related donor; Mtx, methotrexate: MUD, matched unrelated donor; N.a., not available; PBSC, peripheral blood stem cells; RIC, reduced-intensity conditioning; TDM, therapeutic drug monitoring; Tx, transplantation *According to HSCT recommendations by the EWOG-MDS study group Clinical Outcome The clinical outcome of all 14 patients is presented in Table 1. Two adult patients (18%) died after allo-HSCT. Patient 5 had persistent thrombocytopenia and died of a cerebral hemorrhage 8 months post transplantation. Patient 11 developed respiratory failure due to cGVHD in the lungs and received a bilateral pulmonary transplant 5 years post allo-HSCT. However, he developed chronic pulmonary rejection and died 2 years after lung transplantation and 7 years after allo-HSCT. The mean follow-up of the nine patients still alive after allo-HSCT is 26 months (range 3–78 months). The incidence of aGvHD and cGVHD among the eleven transplanted patients was 25% and 33%, respectively, all occurring in patients > 18 years of age (Table 4). None of the pediatric patients had experienced aGVHD or cGVHD, serious infectious complications, or any serious or unexpected transplant-related acute or late toxicity. Their transplantation courses were uneventful and did not principally differ from MDS patients without germline disease-causing GATA2 variants. One patient is listed for allo-HSCT (Patient 14) and two patients are followed regularly in the out-patient clinic (Patients 3 and 8). Discussion This retrospective study describes clinical features and outcome of 14 patients from ten families diagnosed with GATA2 deficiency in Norway. The main findings were as follows: (i) We found a diverse clinical phenotype dominated by cytopenia (13/14), myeloid neoplasia (MDS/AML) (10/14), warts (8/14), and hearing loss (7/14). (ii) We observed two novel clinical features multiple aneurysms of small vessels (n = 1) and early graying (n = 1) that could be associated with GATA2 deficiency. (iii) The majority of patients (11/14) had already undergone allo-HSCT at the time of our analysis, illustrating the need for allo-HSCT in a large proportion of patients with GATA2 deficiency in Norway, and most likely in other countries. (iv) Genetic testing should be offered to first-degree relatives, particularly children, to identify individuals with GATA2 deficiency that need close surveillance. Genetic testing performed as soon as a clinical suspicion is raised, increases the likelihood of an early and correct molecular diagnosis. In Norway, exome-based panels including GATA2 are offered as a routine diagnostic laboratory service for constitutional hematological disorders and PID [21]. However, this approach will not detect all GATA2 variants. Germline variants located in the intronic transcriptional enhancer elements, the cis-acting E-box/GATA and ETS motifs within intron 5 (NM_001145661.1), may cause GATA2 deficiency[34]. Supplementary Sanger sequencing of the enhancer element sequences in intron 5[34] was only performed in one of the laboratories (see Supplementary Methods). Despite supplementary copy number variant calling from exome data, with chromosomal microarray and MLPA only in selected patients (Supplementary Methods), some structural variants may go undetected. In addition, synonymous disease-causing GATA2 variants resulting in selective loss of mutated RNA were recently reported [35]. Disease-causing intronic variants, small intragenic variants, such as structural variants and synonymous variants, may also have escaped detection and hence, individuals with GATA2 deficiency may have been overlooked. With the recent introduction of targeted NGS panels in the work-up of myeloid neoplasms searching for somatic GATA2 variants, unexpected germline variants can be identified, which may reveal the underlying constitutional cause of the myeloid disease and increase the prevalence of known GATA2 deficiency. Improving the NGS panels targeting both germline and somatic variants by deeper coverage of the whole genes including important non-exonic regions, better algorithms for detection of structural variants, and attention to rare synonymous variants may enhance the identification of GATA2 deficiency. In this case series, we found two clinical features that have not yet been described in GATA2 deficiency, namely small vessel aneurysm and early graying. Multiple small vessel aneurysms found in Patient 10 may be secondary to vasculitis, or could also represent a novel vascular feature associated with GATA2 deficiency. In fact, it has been suggested that alteration in GATA2 expression may be of importance for vascular integrity[36]. The observation of premature graying may be a coincidence. In the absence of telomere biology disorders as in the present patients, one may speculate if this feature may reflect a GATA2-linked autoimmune phenomenon which has gone undetected. One of the aims of this retrospective study was to describe the clinical characteristics, GATA2 variants, and other molecular variants representing risk factors for clonal evolution that could aid us in the difficult decision regarding: “Who and when to transplant”? The high proportion of patients (79%) that had already undergone allo-HSCT in our cohort was somewhat surprising. Donadieu et al. have published the largest cohort of patients with GATA2 deficiency, and found that only 28 patients (35%) of 79 patients had undergone allo-HSCT. However, this low percentage of allo-HSCT did not correspond with the severity of the disease in this cohort. At the age of 40, the authors reported a mortality rate of 35% and a hematological malignancy rate of 80%[14]. The high proportion of allo-HSCT in our study may have been influenced by increased awareness of negative prospective clinical markers of GATA2 deficiency. Based on the findings from our study and the high morbidity and mortality rate reported by Donadieu et al., it is clear that these patients need to be monitored closely. Ideally, allo-HSCT should be performed before they develop malignancies (both solid tumors and hematological malignancies)[37] or severe/recurrent infections causing organ failure. In our opinion, a history of disseminated viral infection, aggressive HPV infection (particular with dysplasia), or myeloid clonal disease is clear indication to consider allo-HSCT[14]. First-degree relatives with a severe outcome of the disease may further strengthen the indication for an early allo-HSCT in symptomatic patients with GATA2 deficiency. Overall, the decision to perform an allo-HSCT requires careful weighing of potential gain (restore immune function; diminish the risk of hematological malignancies) versus possible transplant complications, including GVHD and transplant-related mortality. This is particularly challenging given the lack of genotype–phenotype correlation. Keeping in mind that the observation time is short for some of the patients in our study, the survival rate after allo-HSCT was 82% (9/11). In patients with GATA2 deficiency, previous publications have reported 86% survival 2 years after HSCT (n = 22)[16], 73% and 62% survival 1 and 5 years after HSCT, respectively (n = 28)[14], 72%, 65%, and 54% survival 1, 2, and 4 years after HSCT, respectively (n = 21)[9], and 57% 3, 5 years after HSCT (n = 14)[38]. These cohorts are, however, not necessarily comparable in terms of severity of disease and conditioning regimen. Two children were diagnosed with GATA2 deficiency after family screening. The hematological surveillance of one of these children led to detection of hematological abnormalities consistent with MDS and, in the end, a timely allo-HSCT. We therefore recommend genetic testing of children of affected adults and hematological surveillance of individuals with known pathogenic germline GATA2 variants. This includes annual BM investigations with morphological and cytogenetic evaluations, and testing with NGS myeloid panel to screen for somatic occurring molecular drivers of malignancies. Monosomy 7 and trisomy 8 have been reported by others to be the major cytogenetic aberrations in hematopoietic cells of patients with GATA2 deficiency and MDS[15]. Advanced MDS disease and monosomy 7 have been related to worse outcome, especially for pediatric patients with GATA2 germline disease[20]. We found monosomy 7 only in ¼, but trisomy 8 in half of the karyotyped patients, which is in line with a previous published study by McReynolds et al. [39]. Two patients in our GATA2 cohort had monosomy 7 and trisomy 8, both were children. Some of the other somatic variants in our cohort occurred in genes previously reported to be mutated in GATA2 deficiency with MDS, such as ASXL1[40], STAG2, SEPTBP1, and RUNX1[41, 42]. Interestingly, one adult patient with GATA2 deficiency and MDS-related AML had a der(1;7) in the leukemic clone, a translocation that has recently been shown to be enriched in pediatric MDS patients with germline GATA2 mutations[43]. However, our number of patients are too small to determine possible genotype–phenotype correlations related to clonal disease progression. Previous reports have suggested that plasma levels of FLT3LG can be used as predictor of hematological disease in GATA2 deficiency, and used in clinical monitoring post-HSCT [25]. Unfortunately, we lack serum or plasma samples taken before and after HSCT in our cohort, and the role of FLT3LG as a disease progression marker could be explored in future studies. Our main conclusion of this study is that the majority of patients with symptomatic GATA2 deficiency will need allo-HSCT, and close surveillance of these patients is important to find the “optimal window” for allo-HSCT. We advocate a more offensive approach to allo-HSCT than previously described. Supplementary Information Below is the link to the electronic supplementary material.Supplementary file1 (DOCX 55 KB) Acknowledgements Our gratitude goes to the Section for Cancer Cytogenetics who karyotyped 9 of the 14 patients reported here (Table 2). Three of the reported families have been followed up at the Department of Medical Genetics, University Hospital of North Norway. We want to thank our colleagues, MD Gry Hoem, MD Hilde Yttervik, MD Specialist in medical genetics and pediatrics Marie Falkenberg Smeland, and MD PhD Øyvind Holsbø Hald at the Department of Medical Genetics, University Hospital of North Norway, Tromsø, Norway, for their clinical work and genetic counselling. The three pediatric patients are enrolled in the registry of the European Working Group of MDS in Childhood (EWOG-MDS; ClinicalTrials.gov Identifier: NCT00662090). Author Contribution SFJ, JB, AEM, PA, AS-P, TGD, and IN wrote the paper. The genetic analyses were performed by AS-P, MAK, HS, ØH, SS, and EL. SFJ, JB, AEM, EG, YF, CA, AB, IH, TF, BF, AS-P, TGD, and IN collected clinical data. SS did the bone marrow analyses. All authors critically revised the manuscript for important intellectual content and approved the final version of the manuscript. Funding Open access funding provided by University of Oslo (incl Oslo University Hospital). Data Availability Upon request. Code Availability Not applicable. Declarations Ethics Approval Five of the patients had previously consented to be part of a genetic PID research project approved by the regional ethical committee (REC. 2014/1270–1), three patients were diagnosed with MDS in childhood and consented to be registered into the EWOG-MDS-2006 study (2015/1651/REC Nord), and all adult patients who underwent allo-HSCT had consented to publication of data (REC 11909). Studies were performed according to the Declaration of Helsinki. Consent to Participate All adult living patients signed a written informed consent for publication. For children < 18 years, consent was given by their parents. Consent for Publication Obtained. Conflict of Interest The authors declare no competing interests. Abbreviations aGVHD Acute graft versus host disease allo-HSCT Allogeneic hematopoietic stem cell transplantation AML Acute myelogenous leukemia ARDS Acute respiratory distress syndrome BM Bone marrow cGVHD Chronic graft versus host disease GVHD Graft versus host disease HPV Human papilloma virus MDS Myelodysplastic syndrome MRD Matched related donor MUD Matched unrelated donors NGS Next-generation sequencing PID Primary immunodeficiency VAF Variant allele frequency WES Whole exome sequencing ZNF1 Zink finger 1 ZNF2 Zink finger 2 Publisher's Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Asbjørg Stray-Pedersen, Tobias Gedde-Dahl and Ingvild Nordøy contributed equally

150 MG/M2

DrugDosageText

CC BY

34893945

2021-12-10

Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Anaemia'.

The first reported case of trastuzumab induced interstitial lung disease associated with anti-neutrophil cytoplasmic antibody vasculitis - A case report and a prospective cohort study on the usefulness of neutrophil derived biomarkers in monitoring vasculitis disease activity during follow-up. Targeted therapies against human epidermal growth factor receptor 2 (HER2) are associated with increased interstitial lung disease (ILD). Trastuzumab, lapatinib, pertuzumab, and trastuzumab emtansine have markedly extended HER2 breast cancer survival but current knowledge on how these HER2-targeted agents induce interstitial lung disease is still poorly defined due to limited cases in the literature. Physicians mostly managed this complication by dose interruption, dose de-escalation, or discontinuation with success. In 2019, the FDA had granted accelerated approval on trastuzumab deruxtecan (T-Dxd) in HER2 breast cancer in the late line setting. Severe ILD incidence rate was over ten percent and led to fatal outcomes in 2.2% of patients in the T-Dxd trial. Searching for biomarkers to detect ILD incidence before it becomes clinically fulminant or for treatment response monitoring is of high clinical value. A Case of life-threatening trastuzumab-induced ILD was encountered in our facility. The ILD was confirmed to be antineutrophil cytoplasmic antibody (ANCA) pulmonary capillaritis. The biomarker of neutrophil extracellular traps (NETs), serum MPO-DNA complex, showed a good correlation with the clinical severity. Soon after B cell depleting agent rituximab usage, the serum MPO-DNA outperformed ANCA autoantibody and maintained its correlation with clinical severity. In addition to the trastuzumab-induced ILD case, a prospective cohort in our facility also confirmed the usefulness of MPO-DNA in monitoring vasculitis activity. We postulated that upfront testing with biomarkers of vasculitis during HER2 targeted treatment with high ILD incidence may be beneficial in the future. pmc1 Introduction The HER-2 gene is a poor prognostic factor in breast cancer and is amplified in 20–25% of patients [1]. Trastuzumab is a humanized monoclonal antibody that targets the extracellular domain of the human epidermal growth factor receptor 2 (HER2). Since its first approval in 1998, subsequently developed HER2 agents such as lapatinib, pertuzumab, and trastuzumab emtansine have markedly extended the HER2 breast cancer patient survival [[2], [3], [4]]. Drug-induced interstitial lung disease (DIILD) accounts for 3–5% of interstitial lung disease. Cancer therapy is the leading cause of DIILD and accounts for 23–51% of the cases [5]. Although the incidence of trastuzumab-related ILD in the registration trials was low at 0.5%, the mortality rate with treatment-related ILD was around 20% [[6], [7], [8]]. In 2019 the novel HER-2 antibody-drug conjugate trastuzumab deruxtecan (T-Dxd) had recently drawn attention for an increased ILD incidence while being approved to treat late-stage HER-2 positive metastatic breast cancer. T-Dxd therapy achieved a response rate up to 60%, but at the cost of ILD up to 13.6%, and 2.2% treatment-related death [9]. Therefore an unmet medical need for developing effective methods in predicting or monitoring HER-2 targeted agents induced ILD is required. 1.1 Case history A 59-year-old female was diagnosed with invasive ductal carcinoma after undergoing a core needle biopsy for a right breast lump. She received a simple mastectomy, and the surgical pathology showed stage IIA, pT2N0, HER-2/neu 3+, weak positive estrogen receptor (2%), and was negative for progesterone receptor. Adjuvant therapy comprising 12 weeks of weekly paclitaxel plus 52 weeks of tri-weekly trastuzumab was initiated three weeks after the surgery. However, at week 16, the treatment course was interrupted due to an ER visit for blood-tinged sputum over five days. A chest X-ray showed bilateral centrilobular pulmonary infiltration (Fig. 1A). The complete blood count showed neutrophilic leukocytosis (WBC 11.01 k/μL, neutrophil 74.2%, lymphocyte 17.6%, eosinophil 2.5%), anemia and thrombocytopenia (hemoglobin 9.8 g/dL, platelet 126 k/μL). The coagulation profile, as well as liver and kidney biochemistry, were within normal limits. She was admitted for empirical community-acquired pneumonia treatment with ceftriaxone and azithromycin. However, on the third day, the patient's chest X-ray showed rapid infiltration progression (Fig. 1B), and she had a fever that spiked to 39 °C with hypoxia requiring oxygen mask support. A chest CT with contrast showed diffuse peribronchovascular consolidation (Fig. 1C–D). The culture and serology on multiple pneumonia pathogens, including S. pneumonia, Legionella, Chlamydia, Influenza, Aspergillus, Cryptococcus, and CMV virus, showed negative results. Cultures for blood and sputum were also negative. Non-infectious pneumonitis was suspected, and a connective tissue disease survey was conducted. However, hypoxic respiratory failure progressed, and the patient received endotracheal intubation and was transferred to the intensive care unit. The patient underwent a bronchoscopy examination that showed diffuse mucosal bleeding of the entire airway (Fig. 2A and B). Bronchial wash cytology showed neutrophil predominance without malignant cells. Bronchial wash specimens showed no microorganisms, and the cultures remained negative. However, the connective tissue disease survey revealed antineutrophil cytoplasmic antibody (ANCA) positivity both via an indirect immunofluorescence method (IMMCO Diagnostics) and captured enzyme-linked immunoassays (EliA). The survey showed positive neutrophil cytoplasmic staining (cANCA, cytoplasmic ANCA) and an anti-PR3 titer of 40.28 IU/mL (normal <2.0). The antinuclear antibody, anti-dsDNA, rheumatoid factor, C3, C4, anti-GBM, and anti-SSA/SSB were negative.Fig. 1 A-D. (A) Chest radiograph (CXR) at initial presentation revealed multi-lobar consolidation with a centi-hilar pattern. (B) CXR on the third day showed rapid bilateral pulmonary infiltrate progression (C–D) Computed tomography (CT) on the third day showed a diffuse, multi-lobar consolidation with a peri-bronchovascular pattern. Fig. 1 Fig. 2 A-B Bronchoscopy examination of the tracheal (A) and main carina (B) revealed fresh blood in the whole airway without active vascular bleeding. Fig. 2 A lung biopsy via video-assisted thoracoscopic surgery (VATS) was performed for tissue diagnosis. Microscopically, the resected lung tissue showed marked interstitial fibrosis and residual airspaces of varying size (Fig. 3A). The remaining airspaces revealed extensive hemorrhage, as demonstrated by the presence of erythrocytes and histiocytes mixed with fibrinous exudate in the alveolar spaces. The presence of necrotic neutrophils in the alveolar septa and airspaces were indicative of acute capillaritis (Fig. 3B). A diagnosis of acute capillaritis associated with diffuse alveolar damage of the organizing stage was made (Fig. 3C). Treatment with intravenous methylprednisolone (1 mg/kg/day) was administered. The fever resolved, and the chest X-ray showed improvement on the seventh day. The patient was extubated after 14 days of ventilator support.Fig. 3 A Lung, left upper lobe, wedge resection, low power field (100X), interstitial fibrosis with collagen deposits (C) and alveolar airspace obliteration (A). B High power field showing alveolar hemorrhage (A), and necrotic neutrophils (P) in a capillary wall (V) and alveoli. C Comparison of the patient sample to a sample from a healthy individual shows marked airspace obliteration due to hemorrhage and interstitial thickening. Fig. 3 Trastuzumab-related ANCA capillaritis was diagnosed after reviewing the patient's history. She had no pulmonary symptoms or chest X-ray abnormality in previous health check-ups. The Naranjo algorithm, an adverse drug reaction (ADE) assessment tool having a scale of 1–10 [10], was applied with trastuzumab scoring 7 out of 10 (Table 1). Probable trastuzumab-related ADE was suspected. Considering this ADE as a life-threatening event, the trastuzumab schedule was permanently discontinued and replaced with hormonal adjuvant therapy. She also received rheumatologist follow up since this event at the outpatient clinic.Table 1 Probable Trastuzumab related adverse drug reaction, with Naranjo score of 7 according to the patients clinical history. Table 1No Question Yes No Do Not Know I Are there previous conclusive reports on this reaction? +1 0 0 II Did the adverse event appear after the suspected drug was administered? +2 −1 0 III Did the adverse reaction improve when the drug was discontinued or a specific antagonist was administered? +1 0 0 VI Did the adverse reaction reappear when the drug was administered? +2 −1 0 V Are there alternative causes (other than the drug) that could on their own have caused the reaction? −1 +2 0 VI Did the reaction reappear when a placebo was given? −1 +1 0 VII Was the drug detected in the blood (or other fluids) in concentration known to be toxic? +1 0 0 VIII Was the reaction more severe when the dose was increased or less severe when the dose was decreased? +1 0 0 IX Did the patient have a similar reaction to the same or similar drugs in any previous exposure? +1 0 0 X Was the adverse event confirmed by any objective evidence? +1 0 0 Naranjo score for estimating the probability of adverse drug reactions; 0, doubtful ADR; 1–4 possible ADR; 5–8, probable ADR; ≥9 definite ADR. 1.2 Utilization of different vasculitis markers in the trastuzumab ILD case Systemic methylprednisolone had induced rapid remission of the chest X-ray infiltration and inflammatory marker (e.g. C-reactive protein, erythrocyte sedimentation rate, D-dimer) improvement during the initial induction period (Table 2). However, the patient still manifested with residual vasculitis disease activity of leg weakness, numbness, and also persistent microscopic hematuria and proteinuria. Occasional expectoration of blood clot content was also noted despite no evident change in the chest X-ray appearance. The measurement of Birmingham Vasculitis Activity Score (BVAS) [11], a tool of vasculitis activity measurement consisted of 9 organs, and a score of 0–33 for persistent symptoms, and 0 to 63 for new or worse symptoms, still scored 10–20 for persistent symptoms in the patient. The patient also encountered a rapid increase in the anti-PR3 titer eight months after discontinuation of trastuzumab (Table 2). Recent evidence has supported neutrophil as the dominant infiltrate within vasculitis lesions [[12], [13], [14]]. And the discovery of neutrophil extracellular traps (NETs), a component of cell-free DNA, histone, proteinase 3 (PR3), and myeloperoxidase (MPO) released by ANCA-stimulated neutrophil, could induce vasculitis damage such as thrombus formation and endothelial damage. The patient was enrolled in a prospective vasculitis cohort in our facility for testing on vasculitis neutrophil-derived NETs and had also started rituximab (500 mg D1/D15, every 6 months) therapy for better control for the DIILD.Table 2 Trend of inflammatory markers since interstitial lung disease event onset; arrowheads indicate rituximab treatment. MPO-DNA showed a good correlation with Birmingham Vasculitis Activity Score (BVAS), especially in the late phase (r = 0.996, p = 0.004) after Rituximab treatment. This AAV patient had a protracted disease course that involved BVAS score rebound despite the PR3 autoantibody titer remaining below the upper limit. The results indicate that the MPO-DNA complex, which is a novel biomarker of neutrophil activation, could better monitor vasculitis activity. Table 2 2 Prospective cohort on neutrophil derived biomarkers in vasculitis disease activity 2.1 Materials and methods From 2017 to 2021, a prospective cohort study addressing neutrophil related biomarkers in the evaluation of patients with vasculitis was initiated in our facility (Suppl. Fig. S1). The aim was to explore the relationship between levels of cell-free MPO-DNA, a biomarker for NETosis and the clinical activity of systemic vasculitis. Including the Case presented in the manuscript, a total of eight patients with vasculitis and 17 healthy controls (a ratio of 1:2) were enrolled, and serial serum testing was obtained (Suppl. Fig. S2). The study was approved by the hospital Institutional Review Board in 2017 (NTUH: 201612147RINA). Informed consents were obtained from all patients and healthy donors. 2.2 Measurement of MPO-DNA and ANCA We tested each vasculitis individual for the cell-free MPO-DNA via a “sandwich” ELISA with an anti-MPO polyclonal antibody (GeneTex, GTX22088, Irvine, Ca, USA). ELISA microplates were coated with the MPO monoclonal antibody overnight to capture MPO-associated DNAs. Anti-ds DNA-specific monoclonal antibodies (Abcam, Cambridge, ab27156, UK; 1:2000) were added to bind MPO-associated DNA, followed by binding of a horseradish peroxidase-conjugated anti-mouse IgG antibody (Jackson ImmunoResearch, 115-035-003) for detection. A peroxidase substrate (3,3′,5,5′-Tetramethylbenzidine) was added to react with the conjugated peroxidase to yield a blue product. The reaction was halted by adding 2 N H2SO4, and the absorbance of the final product was measured at 450 nm and was transformed to an arbitrary unit (a.u.) according to the previous study [15]. An MPO-DNA assay standard curve is provided (Suppl. Fig. S3). ANCA assessment was performed using anti-proteinase (PR3) and anti-myeloperoxidase(MPO) specific immunoassays (EliA; Thermo Fisher Scientific, Waltham, MA, USA). The assay was performed according to the manufacturer's instructions. An antibody concentration was considered positive if: MPO >5.0 IU/mL and PR3 >3.0 IU/mL (normal range were provided by the manufacturer). 2.3 Statistical analysis We utilized Stata statistical software (V. 14.0, Stata Corporation, College Station, TX) for the analysis. The MPO-DNA cut-off point between healthy controls and vasculitis patients was obtained according to the maximum Youden index (Sensitivity + Specificity - 1) to capture the performance of this dichotomous diagnostic test [16]. Statistical significance of the difference between two sets of continuous variables was analyzed using a Mann-Whitney U test. A p-value less than 0.05 was defined as statistically significant. 3 Results 3.1 MPO DNA performance in the prospective vasculitis cohort The MPO-DNA values were significantly higher in the systemic vasculitis group (n = 8; 0.092 ± 0.071 arbitrary unit (a.u.)) relative to healthy donors (n = 17; 0.015 ± 0.018 a.u., p = 0.01) (Table 3). The Youden J index representing the maximum potential effectiveness of the upper-lower limit of the MPO-DNA cut-off value was 0.047 (a.u.), the red dashed line (Table 3). An MPO-DNA cut-off level at ≤ 0.047 a.u. was established to differentiate between healthy control and vasculitis patients. We also tested MPO-DNA serum level, erythrocyte sedimentation rate, C-reactive protein correlation with their Birmingham Vasculitis Activity Score. The MPO-DNA showed a moderate positive correlation [ r (n = 12) = 0.596, p = 0.025]; but the C-reactive protein [ r(n = 10) = 0.402, p = 0.196] and the erythrocyte sedimentation rate [ r(n = 9) = 0.119, p = 0.727] showed no correlation with clinical vasculitis severity. These results suggest that MPO-DNA levels may also be useful in monitoring vasculitis activity compared to standard inflammation biomarkers (Table 4).Table 3 MPO-DNA values were significantly higher in the systemic vasculitis group [n = 8; 0.092 ± 0.071 arbitrary units (A.U.)] relative to healthy donors [ n = 17; 0.015 ± 0.018 A U., p = 0.01 ]. Youden J index representing the maximum potential effectiveness of the upper-lower limit of the MPO-DNA cut-off value, which was set at 0.047 (A.U.), as indicated by the red dashed line. Table 3 Table 4 The MPO-DNA serum levels, erythrocyte sedimentation rates, and C-reactive protein correlation with the Birmingham Vasculitis Activity Score for the eight vasculitis patients, the MPO-DNA showed a moderate positive correlation but the C-reactive protein and erythrocyte sedimentation rate did not. Table 4 3.2 MPO DNA performance in the trastuzumab ILD case During follow-up of the Case presenting trastuzumab-induced ILD, we found that despite standard vasculitis anti-PR3 is capable of showing good correlation during initial glucocorticoid therapy with the Birmingham vasculitis score (BVAS). The correlation was lost after salvage treatment with the B-cell targeting agent rituximab was initiated (Table 2). MPO-DNA levels exhibited a better correlation with the clinical vasculitis activity BVAS score during the rituximab treatment period. 4 Discussion Interstitial lung disease (ILD) induced by HER2-targeted agents is a well-known adverse drug reaction, but the mechanism is ill-defined and considered low in incidence [17]. HER-2 targeted agents such as trastuzumab, trastuzumab emtansine had been reported to induce cutaneous vasculitis in the previous literature [[18], [19], [20]], but pulmonary vasculitis due to HER2 targeted agents had not been reported yet. A recent review of 9886 patients investigating anti-HER2 therapies for HER2 breast cancer had reported the overall incidence of ILD was 2.4%. The incidence of grade 1–2, grade 3–4, and grade 5 events were 66.7%, 23.0%, and 0.2% respectively. The agents leading to the highest ILD incidence was trastuzumab combined with everolimus or paclitaxel. The incidence of ILD-related deaths was highest among patients receiving trastuzumab deruxtecan (T-Dxd), with an incidence of around 2% [17]. Recently the oncology society has drawn attention to this adverse event due to the recent accelerated approval of T-Dxd [9,21]. The novel antibody-drug conjugate had shown a 60% response rate in third or later line metastatic HER-2 positive breast cancer but at the cost of treatment-related ILD up to 13% and 2% treatment-related death. This high rate of side effects will inhibit its potential to be utilized as the frontline therapy or even be placed in the curative adjuvant or neoadjuvant setting. A recently published article by Kumagai et al. had reported successful induction of T-Dxd interstitial lung disease in a cynomolgus monkey model. Receiving T-Dxd in a monkey model developed interstitial lung disease, whereas receiving Dxd does not. Although most ILD lesions were found within the alveoli, the HER2 expression in lungs was limited to the bronchial level [22]. Vasculitis due to a pathway via ANCA autoantibodies, an indirect mechanism by the immune system, could explain why the injury was not at the location where T-Dxd was uptaken. Moreover, a two-step mechanism of pathogenesis explains how DIILD (drug-induced ILD) manifests later after anti-HER2 therapies. A literature review was performed to summarize previously reported cases of trastuzumab-related ILD. A total of 8 cases were identified before our Case [[23], [24], [25], [26], [27], [28]]. Two of the reported cases had available pathology that showed organizing pneumonia or diffuse alveolar damage. No traceable serum marker was reported in previously reported patients. All previously reported cases that we identified had late onset with a median time of onset two months after the first trastuzumab exposure (Table 5). The pattern of CT infiltration largely comprised ground-glass opacity and patch consolidations, with only one case exhibiting nodular lesions. Most patients recovered following prednisolone-based monotherapy, and only one patient died. Long-term follow up was not conducted in these cases. Our case also had a protracted course of vasculitis up to two years. We found that both ANCA auto-antibody testing and NET related biomarker, MPO-DNA should be tested concomitantly and especially when B cell depleting agents (such as rituximab use in our case) inhibit the autoantibody production, but vasculitis-related neutrophil activity is still high. Reliance on ANCA autoantibodies alone to monitor vasculitis activity may not be sufficient.Table 5 Case comparison of previously reported trastuzumab-related interstitial lung disease [[12], [13], [14], [15], [16], [17]]. Table 5Case Age Gender Previous CRT (setting) Tmab exposure CT pattern Pathology Treatment Recovery Naranjo Score 1 49 F [23] AC, RT, D (adjuvant) 3 months Subpleural consolidation Organizing pneumonia n/aa Yes 6 2 56 F [25] FAC, D (salvage) 4 months Consolidation and pleural effusion BAL: eosinophil 18% Prednisolone (40 mg/d) Yes 8 3 51 F [26] P (neoadjuvant) 10 weeks Ground glass opacity TBLB: Intra-alveolar hemorrhage, interstitial inflammation Prednisolone (40 mg/d) Yes 6 4 63 F [24] D (neoadjuvant) 5 weeks Peripheral consolidation, GGO Diffuse alveolar damage Prednisonec No (mortality) 7 5 67 F [35] DC (adjuvant) 4 months GGO n/a Prednisone Cyclophosphamideb Yes 8 6 68 F [28] EC (adjuvant) 3 months Patch infiltration, GGO n/a Semi-pulse steroidc Yes 6 7 62 F [28] DC (adjuvant) 3 months Patchy consolidation, poorly defined nodules n/a Prednisolonec Yes 7 8 71 F [36] Salvage Tmab followed by T-DM1 Tmab 12 wks T-DM1 6 wks Diffuse interstitial infiltrates n/a Methylprednisolone (1 mg/kg/d) Yes 5 9 59 F P (adjuvant) 3 months Patchy consolidation ANCA capillaritis, acute interstitial pneumonia Prednisolone (40 mg/d) Yes 7 CRT, chemoradiotherapy; Tmab, trastuzumab; T-DM1, trastuzumab emtansine; GGO, ground glass opacity; A, doxorubicin; C, cyclophosphamide; RT, radiotherapy; F, 5-fluorouracil; P, paclitaxel; D, docetaxel; BAL, bronchoalveolar lavage; TBLB, transbronchial lung biopsy. a Resolved spontaneously after trastuzumab discontinuation. b Cyclophosphamide was used upon trastuzumab rechallenge. c The dose used was not reported. Systemic glucocorticoids combined with either rituximab or cyclophosphamide are recommended for induction therapy for life-threatening AAV [29]. Autoreactive B cell depletion by rituximab can effectively reduce disease activity and decrease pathogenic ANCA concentrations in AAV patients. Although adverse events after rituximab include hepatitis B reactivation [30] and progressive multifocal leukoencephalopathy [31], the safety profile of rituximab is still considered to be favorable. Further salvage therapy or investigational agents included belimumab [32], a monoclonal antibody against soluble B cell-activating factor (BAFF) that can induce B cell apoptosis and showed clinical activity as an add-on therapy. The use of therapeutic plasma exchange is also beneficial for carefully selected patients experiencing severe diffuse alveolar hemorrhage or a serum creatinine level of ≥500 mmol/L [33]. There are still two crucial pieces of information that cannot be determined from our Case. First, neither serum levels of anti-PR3/anti-MPO level prior to trastuzumab exposure were available for this case. A prospective examination of vasculitis markers may be considered in the future to determine whether ANCA has clinical utility in predicting side effects during HER2 targeted therapy, especially antibody or antibody drug-conjugates. Secondly, the majority of drug-induced AAV cases present with anti-MPO positivity [34], but our case presented with anti-PR3 positivity without any upper airway symptoms or granuloma formation. Whether this difference is related to the medication administered or due to individual patient differences cannot be explained at present. In conclusion, this Case is the first clinical evidence of HER-2 targeted therapy induces pulmonary ILD via ANCA autoantibodies. Early suspicion with testing of ANCA level can be life-saving when a biopsy cannot be immediately obtained. Pre-treatment ANCA level testing may also be considered in the future when treating patients with HER-2 agents with higher risk profiles of ILD. In addition, serum MPO-DNA, could be a biomarker with promising clinical implication, as the results from this case and a prospective vasculitis cohort in our facility indicated. Authors’ contribution C.H.C., Y.M.H., and Y.M.K., wrote the original draft writing with input from all authors; Y.M.K., C.J.J., designed, developed and conducted the experiments; Y.M.K., L.C., are the primary physicians responsible for the patient medical treatment; Y.L.C., reviews the surgical pathology and issues the report.; S.C.H., Y.M.H., C.H.L., contributed to the final version of the manuscript; All authors actively participated in the discussion and suggestion for the manuscript. Funding This article was subsidized for English editing by 10.13039/501100006477 National Taiwan University under the Excellence Improvement Program for Doctoral Students (grant number 108-2926-I-002-002-MY4), sponsored by the 10.13039/501100004663 Ministry of Science and Technology, Taiwan . This work was supported by Ministry of Science and Technology (MOST 108-2314-B-002-097-MY3, 107-2314-B-002-253-, 106-2314-B-002-100-), 10.13039/501100005762 National Taiwan University Hospital (NTUH 110-M4972) and Taiwan Rheumatology Association Research Fund (TRARF-2018). Ethics approval This study was approved by the hospital Institutional Review Board (NTUH: 201612147RINA). The informed consent was obtained from the patients. Availability of data and material The author(s) confirm that the data supporting the findings of this study are available within the article. Code availability The author(s) confirm that the data supporting the findings of this study are available within the article. Declaration of competing interest The author(s) had declared that no conflicts of interest exist. Appendix A Supplementary data The following is the Supplementary data to this article:Multimedia component 1 Multimedia component 1 Appendix A Supplementary data to this article can be found online at https://doi.org/10.1016/j.breast.2021.11.016.

PACLITAXEL, TRASTUZUMAB

DrugsGivenReaction

CC BY-NC-ND

34894465

2021-11-29

Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Antineutrophil cytoplasmic antibody'.

The first reported case of trastuzumab induced interstitial lung disease associated with anti-neutrophil cytoplasmic antibody vasculitis - A case report and a prospective cohort study on the usefulness of neutrophil derived biomarkers in monitoring vasculitis disease activity during follow-up. Targeted therapies against human epidermal growth factor receptor 2 (HER2) are associated with increased interstitial lung disease (ILD). Trastuzumab, lapatinib, pertuzumab, and trastuzumab emtansine have markedly extended HER2 breast cancer survival but current knowledge on how these HER2-targeted agents induce interstitial lung disease is still poorly defined due to limited cases in the literature. Physicians mostly managed this complication by dose interruption, dose de-escalation, or discontinuation with success. In 2019, the FDA had granted accelerated approval on trastuzumab deruxtecan (T-Dxd) in HER2 breast cancer in the late line setting. Severe ILD incidence rate was over ten percent and led to fatal outcomes in 2.2% of patients in the T-Dxd trial. Searching for biomarkers to detect ILD incidence before it becomes clinically fulminant or for treatment response monitoring is of high clinical value. A Case of life-threatening trastuzumab-induced ILD was encountered in our facility. The ILD was confirmed to be antineutrophil cytoplasmic antibody (ANCA) pulmonary capillaritis. The biomarker of neutrophil extracellular traps (NETs), serum MPO-DNA complex, showed a good correlation with the clinical severity. Soon after B cell depleting agent rituximab usage, the serum MPO-DNA outperformed ANCA autoantibody and maintained its correlation with clinical severity. In addition to the trastuzumab-induced ILD case, a prospective cohort in our facility also confirmed the usefulness of MPO-DNA in monitoring vasculitis activity. We postulated that upfront testing with biomarkers of vasculitis during HER2 targeted treatment with high ILD incidence may be beneficial in the future. pmc1 Introduction The HER-2 gene is a poor prognostic factor in breast cancer and is amplified in 20–25% of patients [1]. Trastuzumab is a humanized monoclonal antibody that targets the extracellular domain of the human epidermal growth factor receptor 2 (HER2). Since its first approval in 1998, subsequently developed HER2 agents such as lapatinib, pertuzumab, and trastuzumab emtansine have markedly extended the HER2 breast cancer patient survival [[2], [3], [4]]. Drug-induced interstitial lung disease (DIILD) accounts for 3–5% of interstitial lung disease. Cancer therapy is the leading cause of DIILD and accounts for 23–51% of the cases [5]. Although the incidence of trastuzumab-related ILD in the registration trials was low at 0.5%, the mortality rate with treatment-related ILD was around 20% [[6], [7], [8]]. In 2019 the novel HER-2 antibody-drug conjugate trastuzumab deruxtecan (T-Dxd) had recently drawn attention for an increased ILD incidence while being approved to treat late-stage HER-2 positive metastatic breast cancer. T-Dxd therapy achieved a response rate up to 60%, but at the cost of ILD up to 13.6%, and 2.2% treatment-related death [9]. Therefore an unmet medical need for developing effective methods in predicting or monitoring HER-2 targeted agents induced ILD is required. 1.1 Case history A 59-year-old female was diagnosed with invasive ductal carcinoma after undergoing a core needle biopsy for a right breast lump. She received a simple mastectomy, and the surgical pathology showed stage IIA, pT2N0, HER-2/neu 3+, weak positive estrogen receptor (2%), and was negative for progesterone receptor. Adjuvant therapy comprising 12 weeks of weekly paclitaxel plus 52 weeks of tri-weekly trastuzumab was initiated three weeks after the surgery. However, at week 16, the treatment course was interrupted due to an ER visit for blood-tinged sputum over five days. A chest X-ray showed bilateral centrilobular pulmonary infiltration (Fig. 1A). The complete blood count showed neutrophilic leukocytosis (WBC 11.01 k/μL, neutrophil 74.2%, lymphocyte 17.6%, eosinophil 2.5%), anemia and thrombocytopenia (hemoglobin 9.8 g/dL, platelet 126 k/μL). The coagulation profile, as well as liver and kidney biochemistry, were within normal limits. She was admitted for empirical community-acquired pneumonia treatment with ceftriaxone and azithromycin. However, on the third day, the patient's chest X-ray showed rapid infiltration progression (Fig. 1B), and she had a fever that spiked to 39 °C with hypoxia requiring oxygen mask support. A chest CT with contrast showed diffuse peribronchovascular consolidation (Fig. 1C–D). The culture and serology on multiple pneumonia pathogens, including S. pneumonia, Legionella, Chlamydia, Influenza, Aspergillus, Cryptococcus, and CMV virus, showed negative results. Cultures for blood and sputum were also negative. Non-infectious pneumonitis was suspected, and a connective tissue disease survey was conducted. However, hypoxic respiratory failure progressed, and the patient received endotracheal intubation and was transferred to the intensive care unit. The patient underwent a bronchoscopy examination that showed diffuse mucosal bleeding of the entire airway (Fig. 2A and B). Bronchial wash cytology showed neutrophil predominance without malignant cells. Bronchial wash specimens showed no microorganisms, and the cultures remained negative. However, the connective tissue disease survey revealed antineutrophil cytoplasmic antibody (ANCA) positivity both via an indirect immunofluorescence method (IMMCO Diagnostics) and captured enzyme-linked immunoassays (EliA). The survey showed positive neutrophil cytoplasmic staining (cANCA, cytoplasmic ANCA) and an anti-PR3 titer of 40.28 IU/mL (normal <2.0). The antinuclear antibody, anti-dsDNA, rheumatoid factor, C3, C4, anti-GBM, and anti-SSA/SSB were negative.Fig. 1 A-D. (A) Chest radiograph (CXR) at initial presentation revealed multi-lobar consolidation with a centi-hilar pattern. (B) CXR on the third day showed rapid bilateral pulmonary infiltrate progression (C–D) Computed tomography (CT) on the third day showed a diffuse, multi-lobar consolidation with a peri-bronchovascular pattern. Fig. 1 Fig. 2 A-B Bronchoscopy examination of the tracheal (A) and main carina (B) revealed fresh blood in the whole airway without active vascular bleeding. Fig. 2 A lung biopsy via video-assisted thoracoscopic surgery (VATS) was performed for tissue diagnosis. Microscopically, the resected lung tissue showed marked interstitial fibrosis and residual airspaces of varying size (Fig. 3A). The remaining airspaces revealed extensive hemorrhage, as demonstrated by the presence of erythrocytes and histiocytes mixed with fibrinous exudate in the alveolar spaces. The presence of necrotic neutrophils in the alveolar septa and airspaces were indicative of acute capillaritis (Fig. 3B). A diagnosis of acute capillaritis associated with diffuse alveolar damage of the organizing stage was made (Fig. 3C). Treatment with intravenous methylprednisolone (1 mg/kg/day) was administered. The fever resolved, and the chest X-ray showed improvement on the seventh day. The patient was extubated after 14 days of ventilator support.Fig. 3 A Lung, left upper lobe, wedge resection, low power field (100X), interstitial fibrosis with collagen deposits (C) and alveolar airspace obliteration (A). B High power field showing alveolar hemorrhage (A), and necrotic neutrophils (P) in a capillary wall (V) and alveoli. C Comparison of the patient sample to a sample from a healthy individual shows marked airspace obliteration due to hemorrhage and interstitial thickening. Fig. 3 Trastuzumab-related ANCA capillaritis was diagnosed after reviewing the patient's history. She had no pulmonary symptoms or chest X-ray abnormality in previous health check-ups. The Naranjo algorithm, an adverse drug reaction (ADE) assessment tool having a scale of 1–10 [10], was applied with trastuzumab scoring 7 out of 10 (Table 1). Probable trastuzumab-related ADE was suspected. Considering this ADE as a life-threatening event, the trastuzumab schedule was permanently discontinued and replaced with hormonal adjuvant therapy. She also received rheumatologist follow up since this event at the outpatient clinic.Table 1 Probable Trastuzumab related adverse drug reaction, with Naranjo score of 7 according to the patients clinical history. Table 1No Question Yes No Do Not Know I Are there previous conclusive reports on this reaction? +1 0 0 II Did the adverse event appear after the suspected drug was administered? +2 −1 0 III Did the adverse reaction improve when the drug was discontinued or a specific antagonist was administered? +1 0 0 VI Did the adverse reaction reappear when the drug was administered? +2 −1 0 V Are there alternative causes (other than the drug) that could on their own have caused the reaction? −1 +2 0 VI Did the reaction reappear when a placebo was given? −1 +1 0 VII Was the drug detected in the blood (or other fluids) in concentration known to be toxic? +1 0 0 VIII Was the reaction more severe when the dose was increased or less severe when the dose was decreased? +1 0 0 IX Did the patient have a similar reaction to the same or similar drugs in any previous exposure? +1 0 0 X Was the adverse event confirmed by any objective evidence? +1 0 0 Naranjo score for estimating the probability of adverse drug reactions; 0, doubtful ADR; 1–4 possible ADR; 5–8, probable ADR; ≥9 definite ADR. 1.2 Utilization of different vasculitis markers in the trastuzumab ILD case Systemic methylprednisolone had induced rapid remission of the chest X-ray infiltration and inflammatory marker (e.g. C-reactive protein, erythrocyte sedimentation rate, D-dimer) improvement during the initial induction period (Table 2). However, the patient still manifested with residual vasculitis disease activity of leg weakness, numbness, and also persistent microscopic hematuria and proteinuria. Occasional expectoration of blood clot content was also noted despite no evident change in the chest X-ray appearance. The measurement of Birmingham Vasculitis Activity Score (BVAS) [11], a tool of vasculitis activity measurement consisted of 9 organs, and a score of 0–33 for persistent symptoms, and 0 to 63 for new or worse symptoms, still scored 10–20 for persistent symptoms in the patient. The patient also encountered a rapid increase in the anti-PR3 titer eight months after discontinuation of trastuzumab (Table 2). Recent evidence has supported neutrophil as the dominant infiltrate within vasculitis lesions [[12], [13], [14]]. And the discovery of neutrophil extracellular traps (NETs), a component of cell-free DNA, histone, proteinase 3 (PR3), and myeloperoxidase (MPO) released by ANCA-stimulated neutrophil, could induce vasculitis damage such as thrombus formation and endothelial damage. The patient was enrolled in a prospective vasculitis cohort in our facility for testing on vasculitis neutrophil-derived NETs and had also started rituximab (500 mg D1/D15, every 6 months) therapy for better control for the DIILD.Table 2 Trend of inflammatory markers since interstitial lung disease event onset; arrowheads indicate rituximab treatment. MPO-DNA showed a good correlation with Birmingham Vasculitis Activity Score (BVAS), especially in the late phase (r = 0.996, p = 0.004) after Rituximab treatment. This AAV patient had a protracted disease course that involved BVAS score rebound despite the PR3 autoantibody titer remaining below the upper limit. The results indicate that the MPO-DNA complex, which is a novel biomarker of neutrophil activation, could better monitor vasculitis activity. Table 2 2 Prospective cohort on neutrophil derived biomarkers in vasculitis disease activity 2.1 Materials and methods From 2017 to 2021, a prospective cohort study addressing neutrophil related biomarkers in the evaluation of patients with vasculitis was initiated in our facility (Suppl. Fig. S1). The aim was to explore the relationship between levels of cell-free MPO-DNA, a biomarker for NETosis and the clinical activity of systemic vasculitis. Including the Case presented in the manuscript, a total of eight patients with vasculitis and 17 healthy controls (a ratio of 1:2) were enrolled, and serial serum testing was obtained (Suppl. Fig. S2). The study was approved by the hospital Institutional Review Board in 2017 (NTUH: 201612147RINA). Informed consents were obtained from all patients and healthy donors. 2.2 Measurement of MPO-DNA and ANCA We tested each vasculitis individual for the cell-free MPO-DNA via a “sandwich” ELISA with an anti-MPO polyclonal antibody (GeneTex, GTX22088, Irvine, Ca, USA). ELISA microplates were coated with the MPO monoclonal antibody overnight to capture MPO-associated DNAs. Anti-ds DNA-specific monoclonal antibodies (Abcam, Cambridge, ab27156, UK; 1:2000) were added to bind MPO-associated DNA, followed by binding of a horseradish peroxidase-conjugated anti-mouse IgG antibody (Jackson ImmunoResearch, 115-035-003) for detection. A peroxidase substrate (3,3′,5,5′-Tetramethylbenzidine) was added to react with the conjugated peroxidase to yield a blue product. The reaction was halted by adding 2 N H2SO4, and the absorbance of the final product was measured at 450 nm and was transformed to an arbitrary unit (a.u.) according to the previous study [15]. An MPO-DNA assay standard curve is provided (Suppl. Fig. S3). ANCA assessment was performed using anti-proteinase (PR3) and anti-myeloperoxidase(MPO) specific immunoassays (EliA; Thermo Fisher Scientific, Waltham, MA, USA). The assay was performed according to the manufacturer's instructions. An antibody concentration was considered positive if: MPO >5.0 IU/mL and PR3 >3.0 IU/mL (normal range were provided by the manufacturer). 2.3 Statistical analysis We utilized Stata statistical software (V. 14.0, Stata Corporation, College Station, TX) for the analysis. The MPO-DNA cut-off point between healthy controls and vasculitis patients was obtained according to the maximum Youden index (Sensitivity + Specificity - 1) to capture the performance of this dichotomous diagnostic test [16]. Statistical significance of the difference between two sets of continuous variables was analyzed using a Mann-Whitney U test. A p-value less than 0.05 was defined as statistically significant. 3 Results 3.1 MPO DNA performance in the prospective vasculitis cohort The MPO-DNA values were significantly higher in the systemic vasculitis group (n = 8; 0.092 ± 0.071 arbitrary unit (a.u.)) relative to healthy donors (n = 17; 0.015 ± 0.018 a.u., p = 0.01) (Table 3). The Youden J index representing the maximum potential effectiveness of the upper-lower limit of the MPO-DNA cut-off value was 0.047 (a.u.), the red dashed line (Table 3). An MPO-DNA cut-off level at ≤ 0.047 a.u. was established to differentiate between healthy control and vasculitis patients. We also tested MPO-DNA serum level, erythrocyte sedimentation rate, C-reactive protein correlation with their Birmingham Vasculitis Activity Score. The MPO-DNA showed a moderate positive correlation [ r (n = 12) = 0.596, p = 0.025]; but the C-reactive protein [ r(n = 10) = 0.402, p = 0.196] and the erythrocyte sedimentation rate [ r(n = 9) = 0.119, p = 0.727] showed no correlation with clinical vasculitis severity. These results suggest that MPO-DNA levels may also be useful in monitoring vasculitis activity compared to standard inflammation biomarkers (Table 4).Table 3 MPO-DNA values were significantly higher in the systemic vasculitis group [n = 8; 0.092 ± 0.071 arbitrary units (A.U.)] relative to healthy donors [ n = 17; 0.015 ± 0.018 A U., p = 0.01 ]. Youden J index representing the maximum potential effectiveness of the upper-lower limit of the MPO-DNA cut-off value, which was set at 0.047 (A.U.), as indicated by the red dashed line. Table 3 Table 4 The MPO-DNA serum levels, erythrocyte sedimentation rates, and C-reactive protein correlation with the Birmingham Vasculitis Activity Score for the eight vasculitis patients, the MPO-DNA showed a moderate positive correlation but the C-reactive protein and erythrocyte sedimentation rate did not. Table 4 3.2 MPO DNA performance in the trastuzumab ILD case During follow-up of the Case presenting trastuzumab-induced ILD, we found that despite standard vasculitis anti-PR3 is capable of showing good correlation during initial glucocorticoid therapy with the Birmingham vasculitis score (BVAS). The correlation was lost after salvage treatment with the B-cell targeting agent rituximab was initiated (Table 2). MPO-DNA levels exhibited a better correlation with the clinical vasculitis activity BVAS score during the rituximab treatment period. 4 Discussion Interstitial lung disease (ILD) induced by HER2-targeted agents is a well-known adverse drug reaction, but the mechanism is ill-defined and considered low in incidence [17]. HER-2 targeted agents such as trastuzumab, trastuzumab emtansine had been reported to induce cutaneous vasculitis in the previous literature [[18], [19], [20]], but pulmonary vasculitis due to HER2 targeted agents had not been reported yet. A recent review of 9886 patients investigating anti-HER2 therapies for HER2 breast cancer had reported the overall incidence of ILD was 2.4%. The incidence of grade 1–2, grade 3–4, and grade 5 events were 66.7%, 23.0%, and 0.2% respectively. The agents leading to the highest ILD incidence was trastuzumab combined with everolimus or paclitaxel. The incidence of ILD-related deaths was highest among patients receiving trastuzumab deruxtecan (T-Dxd), with an incidence of around 2% [17]. Recently the oncology society has drawn attention to this adverse event due to the recent accelerated approval of T-Dxd [9,21]. The novel antibody-drug conjugate had shown a 60% response rate in third or later line metastatic HER-2 positive breast cancer but at the cost of treatment-related ILD up to 13% and 2% treatment-related death. This high rate of side effects will inhibit its potential to be utilized as the frontline therapy or even be placed in the curative adjuvant or neoadjuvant setting. A recently published article by Kumagai et al. had reported successful induction of T-Dxd interstitial lung disease in a cynomolgus monkey model. Receiving T-Dxd in a monkey model developed interstitial lung disease, whereas receiving Dxd does not. Although most ILD lesions were found within the alveoli, the HER2 expression in lungs was limited to the bronchial level [22]. Vasculitis due to a pathway via ANCA autoantibodies, an indirect mechanism by the immune system, could explain why the injury was not at the location where T-Dxd was uptaken. Moreover, a two-step mechanism of pathogenesis explains how DIILD (drug-induced ILD) manifests later after anti-HER2 therapies. A literature review was performed to summarize previously reported cases of trastuzumab-related ILD. A total of 8 cases were identified before our Case [[23], [24], [25], [26], [27], [28]]. Two of the reported cases had available pathology that showed organizing pneumonia or diffuse alveolar damage. No traceable serum marker was reported in previously reported patients. All previously reported cases that we identified had late onset with a median time of onset two months after the first trastuzumab exposure (Table 5). The pattern of CT infiltration largely comprised ground-glass opacity and patch consolidations, with only one case exhibiting nodular lesions. Most patients recovered following prednisolone-based monotherapy, and only one patient died. Long-term follow up was not conducted in these cases. Our case also had a protracted course of vasculitis up to two years. We found that both ANCA auto-antibody testing and NET related biomarker, MPO-DNA should be tested concomitantly and especially when B cell depleting agents (such as rituximab use in our case) inhibit the autoantibody production, but vasculitis-related neutrophil activity is still high. Reliance on ANCA autoantibodies alone to monitor vasculitis activity may not be sufficient.Table 5 Case comparison of previously reported trastuzumab-related interstitial lung disease [[12], [13], [14], [15], [16], [17]]. Table 5Case Age Gender Previous CRT (setting) Tmab exposure CT pattern Pathology Treatment Recovery Naranjo Score 1 49 F [23] AC, RT, D (adjuvant) 3 months Subpleural consolidation Organizing pneumonia n/aa Yes 6 2 56 F [25] FAC, D (salvage) 4 months Consolidation and pleural effusion BAL: eosinophil 18% Prednisolone (40 mg/d) Yes 8 3 51 F [26] P (neoadjuvant) 10 weeks Ground glass opacity TBLB: Intra-alveolar hemorrhage, interstitial inflammation Prednisolone (40 mg/d) Yes 6 4 63 F [24] D (neoadjuvant) 5 weeks Peripheral consolidation, GGO Diffuse alveolar damage Prednisonec No (mortality) 7 5 67 F [35] DC (adjuvant) 4 months GGO n/a Prednisone Cyclophosphamideb Yes 8 6 68 F [28] EC (adjuvant) 3 months Patch infiltration, GGO n/a Semi-pulse steroidc Yes 6 7 62 F [28] DC (adjuvant) 3 months Patchy consolidation, poorly defined nodules n/a Prednisolonec Yes 7 8 71 F [36] Salvage Tmab followed by T-DM1 Tmab 12 wks T-DM1 6 wks Diffuse interstitial infiltrates n/a Methylprednisolone (1 mg/kg/d) Yes 5 9 59 F P (adjuvant) 3 months Patchy consolidation ANCA capillaritis, acute interstitial pneumonia Prednisolone (40 mg/d) Yes 7 CRT, chemoradiotherapy; Tmab, trastuzumab; T-DM1, trastuzumab emtansine; GGO, ground glass opacity; A, doxorubicin; C, cyclophosphamide; RT, radiotherapy; F, 5-fluorouracil; P, paclitaxel; D, docetaxel; BAL, bronchoalveolar lavage; TBLB, transbronchial lung biopsy. a Resolved spontaneously after trastuzumab discontinuation. b Cyclophosphamide was used upon trastuzumab rechallenge. c The dose used was not reported. Systemic glucocorticoids combined with either rituximab or cyclophosphamide are recommended for induction therapy for life-threatening AAV [29]. Autoreactive B cell depletion by rituximab can effectively reduce disease activity and decrease pathogenic ANCA concentrations in AAV patients. Although adverse events after rituximab include hepatitis B reactivation [30] and progressive multifocal leukoencephalopathy [31], the safety profile of rituximab is still considered to be favorable. Further salvage therapy or investigational agents included belimumab [32], a monoclonal antibody against soluble B cell-activating factor (BAFF) that can induce B cell apoptosis and showed clinical activity as an add-on therapy. The use of therapeutic plasma exchange is also beneficial for carefully selected patients experiencing severe diffuse alveolar hemorrhage or a serum creatinine level of ≥500 mmol/L [33]. There are still two crucial pieces of information that cannot be determined from our Case. First, neither serum levels of anti-PR3/anti-MPO level prior to trastuzumab exposure were available for this case. A prospective examination of vasculitis markers may be considered in the future to determine whether ANCA has clinical utility in predicting side effects during HER2 targeted therapy, especially antibody or antibody drug-conjugates. Secondly, the majority of drug-induced AAV cases present with anti-MPO positivity [34], but our case presented with anti-PR3 positivity without any upper airway symptoms or granuloma formation. Whether this difference is related to the medication administered or due to individual patient differences cannot be explained at present. In conclusion, this Case is the first clinical evidence of HER-2 targeted therapy induces pulmonary ILD via ANCA autoantibodies. Early suspicion with testing of ANCA level can be life-saving when a biopsy cannot be immediately obtained. Pre-treatment ANCA level testing may also be considered in the future when treating patients with HER-2 agents with higher risk profiles of ILD. In addition, serum MPO-DNA, could be a biomarker with promising clinical implication, as the results from this case and a prospective vasculitis cohort in our facility indicated. Authors’ contribution C.H.C., Y.M.H., and Y.M.K., wrote the original draft writing with input from all authors; Y.M.K., C.J.J., designed, developed and conducted the experiments; Y.M.K., L.C., are the primary physicians responsible for the patient medical treatment; Y.L.C., reviews the surgical pathology and issues the report.; S.C.H., Y.M.H., C.H.L., contributed to the final version of the manuscript; All authors actively participated in the discussion and suggestion for the manuscript. Funding This article was subsidized for English editing by 10.13039/501100006477 National Taiwan University under the Excellence Improvement Program for Doctoral Students (grant number 108-2926-I-002-002-MY4), sponsored by the 10.13039/501100004663 Ministry of Science and Technology, Taiwan . This work was supported by Ministry of Science and Technology (MOST 108-2314-B-002-097-MY3, 107-2314-B-002-253-, 106-2314-B-002-100-), 10.13039/501100005762 National Taiwan University Hospital (NTUH 110-M4972) and Taiwan Rheumatology Association Research Fund (TRARF-2018). Ethics approval This study was approved by the hospital Institutional Review Board (NTUH: 201612147RINA). The informed consent was obtained from the patients. Availability of data and material The author(s) confirm that the data supporting the findings of this study are available within the article. Code availability The author(s) confirm that the data supporting the findings of this study are available within the article. Declaration of competing interest The author(s) had declared that no conflicts of interest exist. Appendix A Supplementary data The following is the Supplementary data to this article:Multimedia component 1 Multimedia component 1 Appendix A Supplementary data to this article can be found online at https://doi.org/10.1016/j.breast.2021.11.016.

TRASTUZUMAB

DrugsGivenReaction

CC BY-NC-ND

34894465

2021-11-29

Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Interstitial lung disease'.

The first reported case of trastuzumab induced interstitial lung disease associated with anti-neutrophil cytoplasmic antibody vasculitis - A case report and a prospective cohort study on the usefulness of neutrophil derived biomarkers in monitoring vasculitis disease activity during follow-up. Targeted therapies against human epidermal growth factor receptor 2 (HER2) are associated with increased interstitial lung disease (ILD). Trastuzumab, lapatinib, pertuzumab, and trastuzumab emtansine have markedly extended HER2 breast cancer survival but current knowledge on how these HER2-targeted agents induce interstitial lung disease is still poorly defined due to limited cases in the literature. Physicians mostly managed this complication by dose interruption, dose de-escalation, or discontinuation with success. In 2019, the FDA had granted accelerated approval on trastuzumab deruxtecan (T-Dxd) in HER2 breast cancer in the late line setting. Severe ILD incidence rate was over ten percent and led to fatal outcomes in 2.2% of patients in the T-Dxd trial. Searching for biomarkers to detect ILD incidence before it becomes clinically fulminant or for treatment response monitoring is of high clinical value. A Case of life-threatening trastuzumab-induced ILD was encountered in our facility. The ILD was confirmed to be antineutrophil cytoplasmic antibody (ANCA) pulmonary capillaritis. The biomarker of neutrophil extracellular traps (NETs), serum MPO-DNA complex, showed a good correlation with the clinical severity. Soon after B cell depleting agent rituximab usage, the serum MPO-DNA outperformed ANCA autoantibody and maintained its correlation with clinical severity. In addition to the trastuzumab-induced ILD case, a prospective cohort in our facility also confirmed the usefulness of MPO-DNA in monitoring vasculitis activity. We postulated that upfront testing with biomarkers of vasculitis during HER2 targeted treatment with high ILD incidence may be beneficial in the future. pmc1 Introduction The HER-2 gene is a poor prognostic factor in breast cancer and is amplified in 20–25% of patients [1]. Trastuzumab is a humanized monoclonal antibody that targets the extracellular domain of the human epidermal growth factor receptor 2 (HER2). Since its first approval in 1998, subsequently developed HER2 agents such as lapatinib, pertuzumab, and trastuzumab emtansine have markedly extended the HER2 breast cancer patient survival [[2], [3], [4]]. Drug-induced interstitial lung disease (DIILD) accounts for 3–5% of interstitial lung disease. Cancer therapy is the leading cause of DIILD and accounts for 23–51% of the cases [5]. Although the incidence of trastuzumab-related ILD in the registration trials was low at 0.5%, the mortality rate with treatment-related ILD was around 20% [[6], [7], [8]]. In 2019 the novel HER-2 antibody-drug conjugate trastuzumab deruxtecan (T-Dxd) had recently drawn attention for an increased ILD incidence while being approved to treat late-stage HER-2 positive metastatic breast cancer. T-Dxd therapy achieved a response rate up to 60%, but at the cost of ILD up to 13.6%, and 2.2% treatment-related death [9]. Therefore an unmet medical need for developing effective methods in predicting or monitoring HER-2 targeted agents induced ILD is required. 1.1 Case history A 59-year-old female was diagnosed with invasive ductal carcinoma after undergoing a core needle biopsy for a right breast lump. She received a simple mastectomy, and the surgical pathology showed stage IIA, pT2N0, HER-2/neu 3+, weak positive estrogen receptor (2%), and was negative for progesterone receptor. Adjuvant therapy comprising 12 weeks of weekly paclitaxel plus 52 weeks of tri-weekly trastuzumab was initiated three weeks after the surgery. However, at week 16, the treatment course was interrupted due to an ER visit for blood-tinged sputum over five days. A chest X-ray showed bilateral centrilobular pulmonary infiltration (Fig. 1A). The complete blood count showed neutrophilic leukocytosis (WBC 11.01 k/μL, neutrophil 74.2%, lymphocyte 17.6%, eosinophil 2.5%), anemia and thrombocytopenia (hemoglobin 9.8 g/dL, platelet 126 k/μL). The coagulation profile, as well as liver and kidney biochemistry, were within normal limits. She was admitted for empirical community-acquired pneumonia treatment with ceftriaxone and azithromycin. However, on the third day, the patient's chest X-ray showed rapid infiltration progression (Fig. 1B), and she had a fever that spiked to 39 °C with hypoxia requiring oxygen mask support. A chest CT with contrast showed diffuse peribronchovascular consolidation (Fig. 1C–D). The culture and serology on multiple pneumonia pathogens, including S. pneumonia, Legionella, Chlamydia, Influenza, Aspergillus, Cryptococcus, and CMV virus, showed negative results. Cultures for blood and sputum were also negative. Non-infectious pneumonitis was suspected, and a connective tissue disease survey was conducted. However, hypoxic respiratory failure progressed, and the patient received endotracheal intubation and was transferred to the intensive care unit. The patient underwent a bronchoscopy examination that showed diffuse mucosal bleeding of the entire airway (Fig. 2A and B). Bronchial wash cytology showed neutrophil predominance without malignant cells. Bronchial wash specimens showed no microorganisms, and the cultures remained negative. However, the connective tissue disease survey revealed antineutrophil cytoplasmic antibody (ANCA) positivity both via an indirect immunofluorescence method (IMMCO Diagnostics) and captured enzyme-linked immunoassays (EliA). The survey showed positive neutrophil cytoplasmic staining (cANCA, cytoplasmic ANCA) and an anti-PR3 titer of 40.28 IU/mL (normal <2.0). The antinuclear antibody, anti-dsDNA, rheumatoid factor, C3, C4, anti-GBM, and anti-SSA/SSB were negative.Fig. 1 A-D. (A) Chest radiograph (CXR) at initial presentation revealed multi-lobar consolidation with a centi-hilar pattern. (B) CXR on the third day showed rapid bilateral pulmonary infiltrate progression (C–D) Computed tomography (CT) on the third day showed a diffuse, multi-lobar consolidation with a peri-bronchovascular pattern. Fig. 1 Fig. 2 A-B Bronchoscopy examination of the tracheal (A) and main carina (B) revealed fresh blood in the whole airway without active vascular bleeding. Fig. 2 A lung biopsy via video-assisted thoracoscopic surgery (VATS) was performed for tissue diagnosis. Microscopically, the resected lung tissue showed marked interstitial fibrosis and residual airspaces of varying size (Fig. 3A). The remaining airspaces revealed extensive hemorrhage, as demonstrated by the presence of erythrocytes and histiocytes mixed with fibrinous exudate in the alveolar spaces. The presence of necrotic neutrophils in the alveolar septa and airspaces were indicative of acute capillaritis (Fig. 3B). A diagnosis of acute capillaritis associated with diffuse alveolar damage of the organizing stage was made (Fig. 3C). Treatment with intravenous methylprednisolone (1 mg/kg/day) was administered. The fever resolved, and the chest X-ray showed improvement on the seventh day. The patient was extubated after 14 days of ventilator support.Fig. 3 A Lung, left upper lobe, wedge resection, low power field (100X), interstitial fibrosis with collagen deposits (C) and alveolar airspace obliteration (A). B High power field showing alveolar hemorrhage (A), and necrotic neutrophils (P) in a capillary wall (V) and alveoli. C Comparison of the patient sample to a sample from a healthy individual shows marked airspace obliteration due to hemorrhage and interstitial thickening. Fig. 3 Trastuzumab-related ANCA capillaritis was diagnosed after reviewing the patient's history. She had no pulmonary symptoms or chest X-ray abnormality in previous health check-ups. The Naranjo algorithm, an adverse drug reaction (ADE) assessment tool having a scale of 1–10 [10], was applied with trastuzumab scoring 7 out of 10 (Table 1). Probable trastuzumab-related ADE was suspected. Considering this ADE as a life-threatening event, the trastuzumab schedule was permanently discontinued and replaced with hormonal adjuvant therapy. She also received rheumatologist follow up since this event at the outpatient clinic.Table 1 Probable Trastuzumab related adverse drug reaction, with Naranjo score of 7 according to the patients clinical history. Table 1No Question Yes No Do Not Know I Are there previous conclusive reports on this reaction? +1 0 0 II Did the adverse event appear after the suspected drug was administered? +2 −1 0 III Did the adverse reaction improve when the drug was discontinued or a specific antagonist was administered? +1 0 0 VI Did the adverse reaction reappear when the drug was administered? +2 −1 0 V Are there alternative causes (other than the drug) that could on their own have caused the reaction? −1 +2 0 VI Did the reaction reappear when a placebo was given? −1 +1 0 VII Was the drug detected in the blood (or other fluids) in concentration known to be toxic? +1 0 0 VIII Was the reaction more severe when the dose was increased or less severe when the dose was decreased? +1 0 0 IX Did the patient have a similar reaction to the same or similar drugs in any previous exposure? +1 0 0 X Was the adverse event confirmed by any objective evidence? +1 0 0 Naranjo score for estimating the probability of adverse drug reactions; 0, doubtful ADR; 1–4 possible ADR; 5–8, probable ADR; ≥9 definite ADR. 1.2 Utilization of different vasculitis markers in the trastuzumab ILD case Systemic methylprednisolone had induced rapid remission of the chest X-ray infiltration and inflammatory marker (e.g. C-reactive protein, erythrocyte sedimentation rate, D-dimer) improvement during the initial induction period (Table 2). However, the patient still manifested with residual vasculitis disease activity of leg weakness, numbness, and also persistent microscopic hematuria and proteinuria. Occasional expectoration of blood clot content was also noted despite no evident change in the chest X-ray appearance. The measurement of Birmingham Vasculitis Activity Score (BVAS) [11], a tool of vasculitis activity measurement consisted of 9 organs, and a score of 0–33 for persistent symptoms, and 0 to 63 for new or worse symptoms, still scored 10–20 for persistent symptoms in the patient. The patient also encountered a rapid increase in the anti-PR3 titer eight months after discontinuation of trastuzumab (Table 2). Recent evidence has supported neutrophil as the dominant infiltrate within vasculitis lesions [[12], [13], [14]]. And the discovery of neutrophil extracellular traps (NETs), a component of cell-free DNA, histone, proteinase 3 (PR3), and myeloperoxidase (MPO) released by ANCA-stimulated neutrophil, could induce vasculitis damage such as thrombus formation and endothelial damage. The patient was enrolled in a prospective vasculitis cohort in our facility for testing on vasculitis neutrophil-derived NETs and had also started rituximab (500 mg D1/D15, every 6 months) therapy for better control for the DIILD.Table 2 Trend of inflammatory markers since interstitial lung disease event onset; arrowheads indicate rituximab treatment. MPO-DNA showed a good correlation with Birmingham Vasculitis Activity Score (BVAS), especially in the late phase (r = 0.996, p = 0.004) after Rituximab treatment. This AAV patient had a protracted disease course that involved BVAS score rebound despite the PR3 autoantibody titer remaining below the upper limit. The results indicate that the MPO-DNA complex, which is a novel biomarker of neutrophil activation, could better monitor vasculitis activity. Table 2 2 Prospective cohort on neutrophil derived biomarkers in vasculitis disease activity 2.1 Materials and methods From 2017 to 2021, a prospective cohort study addressing neutrophil related biomarkers in the evaluation of patients with vasculitis was initiated in our facility (Suppl. Fig. S1). The aim was to explore the relationship between levels of cell-free MPO-DNA, a biomarker for NETosis and the clinical activity of systemic vasculitis. Including the Case presented in the manuscript, a total of eight patients with vasculitis and 17 healthy controls (a ratio of 1:2) were enrolled, and serial serum testing was obtained (Suppl. Fig. S2). The study was approved by the hospital Institutional Review Board in 2017 (NTUH: 201612147RINA). Informed consents were obtained from all patients and healthy donors. 2.2 Measurement of MPO-DNA and ANCA We tested each vasculitis individual for the cell-free MPO-DNA via a “sandwich” ELISA with an anti-MPO polyclonal antibody (GeneTex, GTX22088, Irvine, Ca, USA). ELISA microplates were coated with the MPO monoclonal antibody overnight to capture MPO-associated DNAs. Anti-ds DNA-specific monoclonal antibodies (Abcam, Cambridge, ab27156, UK; 1:2000) were added to bind MPO-associated DNA, followed by binding of a horseradish peroxidase-conjugated anti-mouse IgG antibody (Jackson ImmunoResearch, 115-035-003) for detection. A peroxidase substrate (3,3′,5,5′-Tetramethylbenzidine) was added to react with the conjugated peroxidase to yield a blue product. The reaction was halted by adding 2 N H2SO4, and the absorbance of the final product was measured at 450 nm and was transformed to an arbitrary unit (a.u.) according to the previous study [15]. An MPO-DNA assay standard curve is provided (Suppl. Fig. S3). ANCA assessment was performed using anti-proteinase (PR3) and anti-myeloperoxidase(MPO) specific immunoassays (EliA; Thermo Fisher Scientific, Waltham, MA, USA). The assay was performed according to the manufacturer's instructions. An antibody concentration was considered positive if: MPO >5.0 IU/mL and PR3 >3.0 IU/mL (normal range were provided by the manufacturer). 2.3 Statistical analysis We utilized Stata statistical software (V. 14.0, Stata Corporation, College Station, TX) for the analysis. The MPO-DNA cut-off point between healthy controls and vasculitis patients was obtained according to the maximum Youden index (Sensitivity + Specificity - 1) to capture the performance of this dichotomous diagnostic test [16]. Statistical significance of the difference between two sets of continuous variables was analyzed using a Mann-Whitney U test. A p-value less than 0.05 was defined as statistically significant. 3 Results 3.1 MPO DNA performance in the prospective vasculitis cohort The MPO-DNA values were significantly higher in the systemic vasculitis group (n = 8; 0.092 ± 0.071 arbitrary unit (a.u.)) relative to healthy donors (n = 17; 0.015 ± 0.018 a.u., p = 0.01) (Table 3). The Youden J index representing the maximum potential effectiveness of the upper-lower limit of the MPO-DNA cut-off value was 0.047 (a.u.), the red dashed line (Table 3). An MPO-DNA cut-off level at ≤ 0.047 a.u. was established to differentiate between healthy control and vasculitis patients. We also tested MPO-DNA serum level, erythrocyte sedimentation rate, C-reactive protein correlation with their Birmingham Vasculitis Activity Score. The MPO-DNA showed a moderate positive correlation [ r (n = 12) = 0.596, p = 0.025]; but the C-reactive protein [ r(n = 10) = 0.402, p = 0.196] and the erythrocyte sedimentation rate [ r(n = 9) = 0.119, p = 0.727] showed no correlation with clinical vasculitis severity. These results suggest that MPO-DNA levels may also be useful in monitoring vasculitis activity compared to standard inflammation biomarkers (Table 4).Table 3 MPO-DNA values were significantly higher in the systemic vasculitis group [n = 8; 0.092 ± 0.071 arbitrary units (A.U.)] relative to healthy donors [ n = 17; 0.015 ± 0.018 A U., p = 0.01 ]. Youden J index representing the maximum potential effectiveness of the upper-lower limit of the MPO-DNA cut-off value, which was set at 0.047 (A.U.), as indicated by the red dashed line. Table 3 Table 4 The MPO-DNA serum levels, erythrocyte sedimentation rates, and C-reactive protein correlation with the Birmingham Vasculitis Activity Score for the eight vasculitis patients, the MPO-DNA showed a moderate positive correlation but the C-reactive protein and erythrocyte sedimentation rate did not. Table 4 3.2 MPO DNA performance in the trastuzumab ILD case During follow-up of the Case presenting trastuzumab-induced ILD, we found that despite standard vasculitis anti-PR3 is capable of showing good correlation during initial glucocorticoid therapy with the Birmingham vasculitis score (BVAS). The correlation was lost after salvage treatment with the B-cell targeting agent rituximab was initiated (Table 2). MPO-DNA levels exhibited a better correlation with the clinical vasculitis activity BVAS score during the rituximab treatment period. 4 Discussion Interstitial lung disease (ILD) induced by HER2-targeted agents is a well-known adverse drug reaction, but the mechanism is ill-defined and considered low in incidence [17]. HER-2 targeted agents such as trastuzumab, trastuzumab emtansine had been reported to induce cutaneous vasculitis in the previous literature [[18], [19], [20]], but pulmonary vasculitis due to HER2 targeted agents had not been reported yet. A recent review of 9886 patients investigating anti-HER2 therapies for HER2 breast cancer had reported the overall incidence of ILD was 2.4%. The incidence of grade 1–2, grade 3–4, and grade 5 events were 66.7%, 23.0%, and 0.2% respectively. The agents leading to the highest ILD incidence was trastuzumab combined with everolimus or paclitaxel. The incidence of ILD-related deaths was highest among patients receiving trastuzumab deruxtecan (T-Dxd), with an incidence of around 2% [17]. Recently the oncology society has drawn attention to this adverse event due to the recent accelerated approval of T-Dxd [9,21]. The novel antibody-drug conjugate had shown a 60% response rate in third or later line metastatic HER-2 positive breast cancer but at the cost of treatment-related ILD up to 13% and 2% treatment-related death. This high rate of side effects will inhibit its potential to be utilized as the frontline therapy or even be placed in the curative adjuvant or neoadjuvant setting. A recently published article by Kumagai et al. had reported successful induction of T-Dxd interstitial lung disease in a cynomolgus monkey model. Receiving T-Dxd in a monkey model developed interstitial lung disease, whereas receiving Dxd does not. Although most ILD lesions were found within the alveoli, the HER2 expression in lungs was limited to the bronchial level [22]. Vasculitis due to a pathway via ANCA autoantibodies, an indirect mechanism by the immune system, could explain why the injury was not at the location where T-Dxd was uptaken. Moreover, a two-step mechanism of pathogenesis explains how DIILD (drug-induced ILD) manifests later after anti-HER2 therapies. A literature review was performed to summarize previously reported cases of trastuzumab-related ILD. A total of 8 cases were identified before our Case [[23], [24], [25], [26], [27], [28]]. Two of the reported cases had available pathology that showed organizing pneumonia or diffuse alveolar damage. No traceable serum marker was reported in previously reported patients. All previously reported cases that we identified had late onset with a median time of onset two months after the first trastuzumab exposure (Table 5). The pattern of CT infiltration largely comprised ground-glass opacity and patch consolidations, with only one case exhibiting nodular lesions. Most patients recovered following prednisolone-based monotherapy, and only one patient died. Long-term follow up was not conducted in these cases. Our case also had a protracted course of vasculitis up to two years. We found that both ANCA auto-antibody testing and NET related biomarker, MPO-DNA should be tested concomitantly and especially when B cell depleting agents (such as rituximab use in our case) inhibit the autoantibody production, but vasculitis-related neutrophil activity is still high. Reliance on ANCA autoantibodies alone to monitor vasculitis activity may not be sufficient.Table 5 Case comparison of previously reported trastuzumab-related interstitial lung disease [[12], [13], [14], [15], [16], [17]]. Table 5Case Age Gender Previous CRT (setting) Tmab exposure CT pattern Pathology Treatment Recovery Naranjo Score 1 49 F [23] AC, RT, D (adjuvant) 3 months Subpleural consolidation Organizing pneumonia n/aa Yes 6 2 56 F [25] FAC, D (salvage) 4 months Consolidation and pleural effusion BAL: eosinophil 18% Prednisolone (40 mg/d) Yes 8 3 51 F [26] P (neoadjuvant) 10 weeks Ground glass opacity TBLB: Intra-alveolar hemorrhage, interstitial inflammation Prednisolone (40 mg/d) Yes 6 4 63 F [24] D (neoadjuvant) 5 weeks Peripheral consolidation, GGO Diffuse alveolar damage Prednisonec No (mortality) 7 5 67 F [35] DC (adjuvant) 4 months GGO n/a Prednisone Cyclophosphamideb Yes 8 6 68 F [28] EC (adjuvant) 3 months Patch infiltration, GGO n/a Semi-pulse steroidc Yes 6 7 62 F [28] DC (adjuvant) 3 months Patchy consolidation, poorly defined nodules n/a Prednisolonec Yes 7 8 71 F [36] Salvage Tmab followed by T-DM1 Tmab 12 wks T-DM1 6 wks Diffuse interstitial infiltrates n/a Methylprednisolone (1 mg/kg/d) Yes 5 9 59 F P (adjuvant) 3 months Patchy consolidation ANCA capillaritis, acute interstitial pneumonia Prednisolone (40 mg/d) Yes 7 CRT, chemoradiotherapy; Tmab, trastuzumab; T-DM1, trastuzumab emtansine; GGO, ground glass opacity; A, doxorubicin; C, cyclophosphamide; RT, radiotherapy; F, 5-fluorouracil; P, paclitaxel; D, docetaxel; BAL, bronchoalveolar lavage; TBLB, transbronchial lung biopsy. a Resolved spontaneously after trastuzumab discontinuation. b Cyclophosphamide was used upon trastuzumab rechallenge. c The dose used was not reported. Systemic glucocorticoids combined with either rituximab or cyclophosphamide are recommended for induction therapy for life-threatening AAV [29]. Autoreactive B cell depletion by rituximab can effectively reduce disease activity and decrease pathogenic ANCA concentrations in AAV patients. Although adverse events after rituximab include hepatitis B reactivation [30] and progressive multifocal leukoencephalopathy [31], the safety profile of rituximab is still considered to be favorable. Further salvage therapy or investigational agents included belimumab [32], a monoclonal antibody against soluble B cell-activating factor (BAFF) that can induce B cell apoptosis and showed clinical activity as an add-on therapy. The use of therapeutic plasma exchange is also beneficial for carefully selected patients experiencing severe diffuse alveolar hemorrhage or a serum creatinine level of ≥500 mmol/L [33]. There are still two crucial pieces of information that cannot be determined from our Case. First, neither serum levels of anti-PR3/anti-MPO level prior to trastuzumab exposure were available for this case. A prospective examination of vasculitis markers may be considered in the future to determine whether ANCA has clinical utility in predicting side effects during HER2 targeted therapy, especially antibody or antibody drug-conjugates. Secondly, the majority of drug-induced AAV cases present with anti-MPO positivity [34], but our case presented with anti-PR3 positivity without any upper airway symptoms or granuloma formation. Whether this difference is related to the medication administered or due to individual patient differences cannot be explained at present. In conclusion, this Case is the first clinical evidence of HER-2 targeted therapy induces pulmonary ILD via ANCA autoantibodies. Early suspicion with testing of ANCA level can be life-saving when a biopsy cannot be immediately obtained. Pre-treatment ANCA level testing may also be considered in the future when treating patients with HER-2 agents with higher risk profiles of ILD. In addition, serum MPO-DNA, could be a biomarker with promising clinical implication, as the results from this case and a prospective vasculitis cohort in our facility indicated. Authors’ contribution C.H.C., Y.M.H., and Y.M.K., wrote the original draft writing with input from all authors; Y.M.K., C.J.J., designed, developed and conducted the experiments; Y.M.K., L.C., are the primary physicians responsible for the patient medical treatment; Y.L.C., reviews the surgical pathology and issues the report.; S.C.H., Y.M.H., C.H.L., contributed to the final version of the manuscript; All authors actively participated in the discussion and suggestion for the manuscript. Funding This article was subsidized for English editing by 10.13039/501100006477 National Taiwan University under the Excellence Improvement Program for Doctoral Students (grant number 108-2926-I-002-002-MY4), sponsored by the 10.13039/501100004663 Ministry of Science and Technology, Taiwan . This work was supported by Ministry of Science and Technology (MOST 108-2314-B-002-097-MY3, 107-2314-B-002-253-, 106-2314-B-002-100-), 10.13039/501100005762 National Taiwan University Hospital (NTUH 110-M4972) and Taiwan Rheumatology Association Research Fund (TRARF-2018). Ethics approval This study was approved by the hospital Institutional Review Board (NTUH: 201612147RINA). The informed consent was obtained from the patients. Availability of data and material The author(s) confirm that the data supporting the findings of this study are available within the article. Code availability The author(s) confirm that the data supporting the findings of this study are available within the article. Declaration of competing interest The author(s) had declared that no conflicts of interest exist. Appendix A Supplementary data The following is the Supplementary data to this article:Multimedia component 1 Multimedia component 1 Appendix A Supplementary data to this article can be found online at https://doi.org/10.1016/j.breast.2021.11.016.

PACLITAXEL, TRASTUZUMAB

DrugsGivenReaction

CC BY-NC-ND

34894465

2021-11-29

Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Mucosal haemorrhage'.

The first reported case of trastuzumab induced interstitial lung disease associated with anti-neutrophil cytoplasmic antibody vasculitis - A case report and a prospective cohort study on the usefulness of neutrophil derived biomarkers in monitoring vasculitis disease activity during follow-up. Targeted therapies against human epidermal growth factor receptor 2 (HER2) are associated with increased interstitial lung disease (ILD). Trastuzumab, lapatinib, pertuzumab, and trastuzumab emtansine have markedly extended HER2 breast cancer survival but current knowledge on how these HER2-targeted agents induce interstitial lung disease is still poorly defined due to limited cases in the literature. Physicians mostly managed this complication by dose interruption, dose de-escalation, or discontinuation with success. In 2019, the FDA had granted accelerated approval on trastuzumab deruxtecan (T-Dxd) in HER2 breast cancer in the late line setting. Severe ILD incidence rate was over ten percent and led to fatal outcomes in 2.2% of patients in the T-Dxd trial. Searching for biomarkers to detect ILD incidence before it becomes clinically fulminant or for treatment response monitoring is of high clinical value. A Case of life-threatening trastuzumab-induced ILD was encountered in our facility. The ILD was confirmed to be antineutrophil cytoplasmic antibody (ANCA) pulmonary capillaritis. The biomarker of neutrophil extracellular traps (NETs), serum MPO-DNA complex, showed a good correlation with the clinical severity. Soon after B cell depleting agent rituximab usage, the serum MPO-DNA outperformed ANCA autoantibody and maintained its correlation with clinical severity. In addition to the trastuzumab-induced ILD case, a prospective cohort in our facility also confirmed the usefulness of MPO-DNA in monitoring vasculitis activity. We postulated that upfront testing with biomarkers of vasculitis during HER2 targeted treatment with high ILD incidence may be beneficial in the future. pmc1 Introduction The HER-2 gene is a poor prognostic factor in breast cancer and is amplified in 20–25% of patients [1]. Trastuzumab is a humanized monoclonal antibody that targets the extracellular domain of the human epidermal growth factor receptor 2 (HER2). Since its first approval in 1998, subsequently developed HER2 agents such as lapatinib, pertuzumab, and trastuzumab emtansine have markedly extended the HER2 breast cancer patient survival [[2], [3], [4]]. Drug-induced interstitial lung disease (DIILD) accounts for 3–5% of interstitial lung disease. Cancer therapy is the leading cause of DIILD and accounts for 23–51% of the cases [5]. Although the incidence of trastuzumab-related ILD in the registration trials was low at 0.5%, the mortality rate with treatment-related ILD was around 20% [[6], [7], [8]]. In 2019 the novel HER-2 antibody-drug conjugate trastuzumab deruxtecan (T-Dxd) had recently drawn attention for an increased ILD incidence while being approved to treat late-stage HER-2 positive metastatic breast cancer. T-Dxd therapy achieved a response rate up to 60%, but at the cost of ILD up to 13.6%, and 2.2% treatment-related death [9]. Therefore an unmet medical need for developing effective methods in predicting or monitoring HER-2 targeted agents induced ILD is required. 1.1 Case history A 59-year-old female was diagnosed with invasive ductal carcinoma after undergoing a core needle biopsy for a right breast lump. She received a simple mastectomy, and the surgical pathology showed stage IIA, pT2N0, HER-2/neu 3+, weak positive estrogen receptor (2%), and was negative for progesterone receptor. Adjuvant therapy comprising 12 weeks of weekly paclitaxel plus 52 weeks of tri-weekly trastuzumab was initiated three weeks after the surgery. However, at week 16, the treatment course was interrupted due to an ER visit for blood-tinged sputum over five days. A chest X-ray showed bilateral centrilobular pulmonary infiltration (Fig. 1A). The complete blood count showed neutrophilic leukocytosis (WBC 11.01 k/μL, neutrophil 74.2%, lymphocyte 17.6%, eosinophil 2.5%), anemia and thrombocytopenia (hemoglobin 9.8 g/dL, platelet 126 k/μL). The coagulation profile, as well as liver and kidney biochemistry, were within normal limits. She was admitted for empirical community-acquired pneumonia treatment with ceftriaxone and azithromycin. However, on the third day, the patient's chest X-ray showed rapid infiltration progression (Fig. 1B), and she had a fever that spiked to 39 °C with hypoxia requiring oxygen mask support. A chest CT with contrast showed diffuse peribronchovascular consolidation (Fig. 1C–D). The culture and serology on multiple pneumonia pathogens, including S. pneumonia, Legionella, Chlamydia, Influenza, Aspergillus, Cryptococcus, and CMV virus, showed negative results. Cultures for blood and sputum were also negative. Non-infectious pneumonitis was suspected, and a connective tissue disease survey was conducted. However, hypoxic respiratory failure progressed, and the patient received endotracheal intubation and was transferred to the intensive care unit. The patient underwent a bronchoscopy examination that showed diffuse mucosal bleeding of the entire airway (Fig. 2A and B). Bronchial wash cytology showed neutrophil predominance without malignant cells. Bronchial wash specimens showed no microorganisms, and the cultures remained negative. However, the connective tissue disease survey revealed antineutrophil cytoplasmic antibody (ANCA) positivity both via an indirect immunofluorescence method (IMMCO Diagnostics) and captured enzyme-linked immunoassays (EliA). The survey showed positive neutrophil cytoplasmic staining (cANCA, cytoplasmic ANCA) and an anti-PR3 titer of 40.28 IU/mL (normal <2.0). The antinuclear antibody, anti-dsDNA, rheumatoid factor, C3, C4, anti-GBM, and anti-SSA/SSB were negative.Fig. 1 A-D. (A) Chest radiograph (CXR) at initial presentation revealed multi-lobar consolidation with a centi-hilar pattern. (B) CXR on the third day showed rapid bilateral pulmonary infiltrate progression (C–D) Computed tomography (CT) on the third day showed a diffuse, multi-lobar consolidation with a peri-bronchovascular pattern. Fig. 1 Fig. 2 A-B Bronchoscopy examination of the tracheal (A) and main carina (B) revealed fresh blood in the whole airway without active vascular bleeding. Fig. 2 A lung biopsy via video-assisted thoracoscopic surgery (VATS) was performed for tissue diagnosis. Microscopically, the resected lung tissue showed marked interstitial fibrosis and residual airspaces of varying size (Fig. 3A). The remaining airspaces revealed extensive hemorrhage, as demonstrated by the presence of erythrocytes and histiocytes mixed with fibrinous exudate in the alveolar spaces. The presence of necrotic neutrophils in the alveolar septa and airspaces were indicative of acute capillaritis (Fig. 3B). A diagnosis of acute capillaritis associated with diffuse alveolar damage of the organizing stage was made (Fig. 3C). Treatment with intravenous methylprednisolone (1 mg/kg/day) was administered. The fever resolved, and the chest X-ray showed improvement on the seventh day. The patient was extubated after 14 days of ventilator support.Fig. 3 A Lung, left upper lobe, wedge resection, low power field (100X), interstitial fibrosis with collagen deposits (C) and alveolar airspace obliteration (A). B High power field showing alveolar hemorrhage (A), and necrotic neutrophils (P) in a capillary wall (V) and alveoli. C Comparison of the patient sample to a sample from a healthy individual shows marked airspace obliteration due to hemorrhage and interstitial thickening. Fig. 3 Trastuzumab-related ANCA capillaritis was diagnosed after reviewing the patient's history. She had no pulmonary symptoms or chest X-ray abnormality in previous health check-ups. The Naranjo algorithm, an adverse drug reaction (ADE) assessment tool having a scale of 1–10 [10], was applied with trastuzumab scoring 7 out of 10 (Table 1). Probable trastuzumab-related ADE was suspected. Considering this ADE as a life-threatening event, the trastuzumab schedule was permanently discontinued and replaced with hormonal adjuvant therapy. She also received rheumatologist follow up since this event at the outpatient clinic.Table 1 Probable Trastuzumab related adverse drug reaction, with Naranjo score of 7 according to the patients clinical history. Table 1No Question Yes No Do Not Know I Are there previous conclusive reports on this reaction? +1 0 0 II Did the adverse event appear after the suspected drug was administered? +2 −1 0 III Did the adverse reaction improve when the drug was discontinued or a specific antagonist was administered? +1 0 0 VI Did the adverse reaction reappear when the drug was administered? +2 −1 0 V Are there alternative causes (other than the drug) that could on their own have caused the reaction? −1 +2 0 VI Did the reaction reappear when a placebo was given? −1 +1 0 VII Was the drug detected in the blood (or other fluids) in concentration known to be toxic? +1 0 0 VIII Was the reaction more severe when the dose was increased or less severe when the dose was decreased? +1 0 0 IX Did the patient have a similar reaction to the same or similar drugs in any previous exposure? +1 0 0 X Was the adverse event confirmed by any objective evidence? +1 0 0 Naranjo score for estimating the probability of adverse drug reactions; 0, doubtful ADR; 1–4 possible ADR; 5–8, probable ADR; ≥9 definite ADR. 1.2 Utilization of different vasculitis markers in the trastuzumab ILD case Systemic methylprednisolone had induced rapid remission of the chest X-ray infiltration and inflammatory marker (e.g. C-reactive protein, erythrocyte sedimentation rate, D-dimer) improvement during the initial induction period (Table 2). However, the patient still manifested with residual vasculitis disease activity of leg weakness, numbness, and also persistent microscopic hematuria and proteinuria. Occasional expectoration of blood clot content was also noted despite no evident change in the chest X-ray appearance. The measurement of Birmingham Vasculitis Activity Score (BVAS) [11], a tool of vasculitis activity measurement consisted of 9 organs, and a score of 0–33 for persistent symptoms, and 0 to 63 for new or worse symptoms, still scored 10–20 for persistent symptoms in the patient. The patient also encountered a rapid increase in the anti-PR3 titer eight months after discontinuation of trastuzumab (Table 2). Recent evidence has supported neutrophil as the dominant infiltrate within vasculitis lesions [[12], [13], [14]]. And the discovery of neutrophil extracellular traps (NETs), a component of cell-free DNA, histone, proteinase 3 (PR3), and myeloperoxidase (MPO) released by ANCA-stimulated neutrophil, could induce vasculitis damage such as thrombus formation and endothelial damage. The patient was enrolled in a prospective vasculitis cohort in our facility for testing on vasculitis neutrophil-derived NETs and had also started rituximab (500 mg D1/D15, every 6 months) therapy for better control for the DIILD.Table 2 Trend of inflammatory markers since interstitial lung disease event onset; arrowheads indicate rituximab treatment. MPO-DNA showed a good correlation with Birmingham Vasculitis Activity Score (BVAS), especially in the late phase (r = 0.996, p = 0.004) after Rituximab treatment. This AAV patient had a protracted disease course that involved BVAS score rebound despite the PR3 autoantibody titer remaining below the upper limit. The results indicate that the MPO-DNA complex, which is a novel biomarker of neutrophil activation, could better monitor vasculitis activity. Table 2 2 Prospective cohort on neutrophil derived biomarkers in vasculitis disease activity 2.1 Materials and methods From 2017 to 2021, a prospective cohort study addressing neutrophil related biomarkers in the evaluation of patients with vasculitis was initiated in our facility (Suppl. Fig. S1). The aim was to explore the relationship between levels of cell-free MPO-DNA, a biomarker for NETosis and the clinical activity of systemic vasculitis. Including the Case presented in the manuscript, a total of eight patients with vasculitis and 17 healthy controls (a ratio of 1:2) were enrolled, and serial serum testing was obtained (Suppl. Fig. S2). The study was approved by the hospital Institutional Review Board in 2017 (NTUH: 201612147RINA). Informed consents were obtained from all patients and healthy donors. 2.2 Measurement of MPO-DNA and ANCA We tested each vasculitis individual for the cell-free MPO-DNA via a “sandwich” ELISA with an anti-MPO polyclonal antibody (GeneTex, GTX22088, Irvine, Ca, USA). ELISA microplates were coated with the MPO monoclonal antibody overnight to capture MPO-associated DNAs. Anti-ds DNA-specific monoclonal antibodies (Abcam, Cambridge, ab27156, UK; 1:2000) were added to bind MPO-associated DNA, followed by binding of a horseradish peroxidase-conjugated anti-mouse IgG antibody (Jackson ImmunoResearch, 115-035-003) for detection. A peroxidase substrate (3,3′,5,5′-Tetramethylbenzidine) was added to react with the conjugated peroxidase to yield a blue product. The reaction was halted by adding 2 N H2SO4, and the absorbance of the final product was measured at 450 nm and was transformed to an arbitrary unit (a.u.) according to the previous study [15]. An MPO-DNA assay standard curve is provided (Suppl. Fig. S3). ANCA assessment was performed using anti-proteinase (PR3) and anti-myeloperoxidase(MPO) specific immunoassays (EliA; Thermo Fisher Scientific, Waltham, MA, USA). The assay was performed according to the manufacturer's instructions. An antibody concentration was considered positive if: MPO >5.0 IU/mL and PR3 >3.0 IU/mL (normal range were provided by the manufacturer). 2.3 Statistical analysis We utilized Stata statistical software (V. 14.0, Stata Corporation, College Station, TX) for the analysis. The MPO-DNA cut-off point between healthy controls and vasculitis patients was obtained according to the maximum Youden index (Sensitivity + Specificity - 1) to capture the performance of this dichotomous diagnostic test [16]. Statistical significance of the difference between two sets of continuous variables was analyzed using a Mann-Whitney U test. A p-value less than 0.05 was defined as statistically significant. 3 Results 3.1 MPO DNA performance in the prospective vasculitis cohort The MPO-DNA values were significantly higher in the systemic vasculitis group (n = 8; 0.092 ± 0.071 arbitrary unit (a.u.)) relative to healthy donors (n = 17; 0.015 ± 0.018 a.u., p = 0.01) (Table 3). The Youden J index representing the maximum potential effectiveness of the upper-lower limit of the MPO-DNA cut-off value was 0.047 (a.u.), the red dashed line (Table 3). An MPO-DNA cut-off level at ≤ 0.047 a.u. was established to differentiate between healthy control and vasculitis patients. We also tested MPO-DNA serum level, erythrocyte sedimentation rate, C-reactive protein correlation with their Birmingham Vasculitis Activity Score. The MPO-DNA showed a moderate positive correlation [ r (n = 12) = 0.596, p = 0.025]; but the C-reactive protein [ r(n = 10) = 0.402, p = 0.196] and the erythrocyte sedimentation rate [ r(n = 9) = 0.119, p = 0.727] showed no correlation with clinical vasculitis severity. These results suggest that MPO-DNA levels may also be useful in monitoring vasculitis activity compared to standard inflammation biomarkers (Table 4).Table 3 MPO-DNA values were significantly higher in the systemic vasculitis group [n = 8; 0.092 ± 0.071 arbitrary units (A.U.)] relative to healthy donors [ n = 17; 0.015 ± 0.018 A U., p = 0.01 ]. Youden J index representing the maximum potential effectiveness of the upper-lower limit of the MPO-DNA cut-off value, which was set at 0.047 (A.U.), as indicated by the red dashed line. Table 3 Table 4 The MPO-DNA serum levels, erythrocyte sedimentation rates, and C-reactive protein correlation with the Birmingham Vasculitis Activity Score for the eight vasculitis patients, the MPO-DNA showed a moderate positive correlation but the C-reactive protein and erythrocyte sedimentation rate did not. Table 4 3.2 MPO DNA performance in the trastuzumab ILD case During follow-up of the Case presenting trastuzumab-induced ILD, we found that despite standard vasculitis anti-PR3 is capable of showing good correlation during initial glucocorticoid therapy with the Birmingham vasculitis score (BVAS). The correlation was lost after salvage treatment with the B-cell targeting agent rituximab was initiated (Table 2). MPO-DNA levels exhibited a better correlation with the clinical vasculitis activity BVAS score during the rituximab treatment period. 4 Discussion Interstitial lung disease (ILD) induced by HER2-targeted agents is a well-known adverse drug reaction, but the mechanism is ill-defined and considered low in incidence [17]. HER-2 targeted agents such as trastuzumab, trastuzumab emtansine had been reported to induce cutaneous vasculitis in the previous literature [[18], [19], [20]], but pulmonary vasculitis due to HER2 targeted agents had not been reported yet. A recent review of 9886 patients investigating anti-HER2 therapies for HER2 breast cancer had reported the overall incidence of ILD was 2.4%. The incidence of grade 1–2, grade 3–4, and grade 5 events were 66.7%, 23.0%, and 0.2% respectively. The agents leading to the highest ILD incidence was trastuzumab combined with everolimus or paclitaxel. The incidence of ILD-related deaths was highest among patients receiving trastuzumab deruxtecan (T-Dxd), with an incidence of around 2% [17]. Recently the oncology society has drawn attention to this adverse event due to the recent accelerated approval of T-Dxd [9,21]. The novel antibody-drug conjugate had shown a 60% response rate in third or later line metastatic HER-2 positive breast cancer but at the cost of treatment-related ILD up to 13% and 2% treatment-related death. This high rate of side effects will inhibit its potential to be utilized as the frontline therapy or even be placed in the curative adjuvant or neoadjuvant setting. A recently published article by Kumagai et al. had reported successful induction of T-Dxd interstitial lung disease in a cynomolgus monkey model. Receiving T-Dxd in a monkey model developed interstitial lung disease, whereas receiving Dxd does not. Although most ILD lesions were found within the alveoli, the HER2 expression in lungs was limited to the bronchial level [22]. Vasculitis due to a pathway via ANCA autoantibodies, an indirect mechanism by the immune system, could explain why the injury was not at the location where T-Dxd was uptaken. Moreover, a two-step mechanism of pathogenesis explains how DIILD (drug-induced ILD) manifests later after anti-HER2 therapies. A literature review was performed to summarize previously reported cases of trastuzumab-related ILD. A total of 8 cases were identified before our Case [[23], [24], [25], [26], [27], [28]]. Two of the reported cases had available pathology that showed organizing pneumonia or diffuse alveolar damage. No traceable serum marker was reported in previously reported patients. All previously reported cases that we identified had late onset with a median time of onset two months after the first trastuzumab exposure (Table 5). The pattern of CT infiltration largely comprised ground-glass opacity and patch consolidations, with only one case exhibiting nodular lesions. Most patients recovered following prednisolone-based monotherapy, and only one patient died. Long-term follow up was not conducted in these cases. Our case also had a protracted course of vasculitis up to two years. We found that both ANCA auto-antibody testing and NET related biomarker, MPO-DNA should be tested concomitantly and especially when B cell depleting agents (such as rituximab use in our case) inhibit the autoantibody production, but vasculitis-related neutrophil activity is still high. Reliance on ANCA autoantibodies alone to monitor vasculitis activity may not be sufficient.Table 5 Case comparison of previously reported trastuzumab-related interstitial lung disease [[12], [13], [14], [15], [16], [17]]. Table 5Case Age Gender Previous CRT (setting) Tmab exposure CT pattern Pathology Treatment Recovery Naranjo Score 1 49 F [23] AC, RT, D (adjuvant) 3 months Subpleural consolidation Organizing pneumonia n/aa Yes 6 2 56 F [25] FAC, D (salvage) 4 months Consolidation and pleural effusion BAL: eosinophil 18% Prednisolone (40 mg/d) Yes 8 3 51 F [26] P (neoadjuvant) 10 weeks Ground glass opacity TBLB: Intra-alveolar hemorrhage, interstitial inflammation Prednisolone (40 mg/d) Yes 6 4 63 F [24] D (neoadjuvant) 5 weeks Peripheral consolidation, GGO Diffuse alveolar damage Prednisonec No (mortality) 7 5 67 F [35] DC (adjuvant) 4 months GGO n/a Prednisone Cyclophosphamideb Yes 8 6 68 F [28] EC (adjuvant) 3 months Patch infiltration, GGO n/a Semi-pulse steroidc Yes 6 7 62 F [28] DC (adjuvant) 3 months Patchy consolidation, poorly defined nodules n/a Prednisolonec Yes 7 8 71 F [36] Salvage Tmab followed by T-DM1 Tmab 12 wks T-DM1 6 wks Diffuse interstitial infiltrates n/a Methylprednisolone (1 mg/kg/d) Yes 5 9 59 F P (adjuvant) 3 months Patchy consolidation ANCA capillaritis, acute interstitial pneumonia Prednisolone (40 mg/d) Yes 7 CRT, chemoradiotherapy; Tmab, trastuzumab; T-DM1, trastuzumab emtansine; GGO, ground glass opacity; A, doxorubicin; C, cyclophosphamide; RT, radiotherapy; F, 5-fluorouracil; P, paclitaxel; D, docetaxel; BAL, bronchoalveolar lavage; TBLB, transbronchial lung biopsy. a Resolved spontaneously after trastuzumab discontinuation. b Cyclophosphamide was used upon trastuzumab rechallenge. c The dose used was not reported. Systemic glucocorticoids combined with either rituximab or cyclophosphamide are recommended for induction therapy for life-threatening AAV [29]. Autoreactive B cell depletion by rituximab can effectively reduce disease activity and decrease pathogenic ANCA concentrations in AAV patients. Although adverse events after rituximab include hepatitis B reactivation [30] and progressive multifocal leukoencephalopathy [31], the safety profile of rituximab is still considered to be favorable. Further salvage therapy or investigational agents included belimumab [32], a monoclonal antibody against soluble B cell-activating factor (BAFF) that can induce B cell apoptosis and showed clinical activity as an add-on therapy. The use of therapeutic plasma exchange is also beneficial for carefully selected patients experiencing severe diffuse alveolar hemorrhage or a serum creatinine level of ≥500 mmol/L [33]. There are still two crucial pieces of information that cannot be determined from our Case. First, neither serum levels of anti-PR3/anti-MPO level prior to trastuzumab exposure were available for this case. A prospective examination of vasculitis markers may be considered in the future to determine whether ANCA has clinical utility in predicting side effects during HER2 targeted therapy, especially antibody or antibody drug-conjugates. Secondly, the majority of drug-induced AAV cases present with anti-MPO positivity [34], but our case presented with anti-PR3 positivity without any upper airway symptoms or granuloma formation. Whether this difference is related to the medication administered or due to individual patient differences cannot be explained at present. In conclusion, this Case is the first clinical evidence of HER-2 targeted therapy induces pulmonary ILD via ANCA autoantibodies. Early suspicion with testing of ANCA level can be life-saving when a biopsy cannot be immediately obtained. Pre-treatment ANCA level testing may also be considered in the future when treating patients with HER-2 agents with higher risk profiles of ILD. In addition, serum MPO-DNA, could be a biomarker with promising clinical implication, as the results from this case and a prospective vasculitis cohort in our facility indicated. Authors’ contribution C.H.C., Y.M.H., and Y.M.K., wrote the original draft writing with input from all authors; Y.M.K., C.J.J., designed, developed and conducted the experiments; Y.M.K., L.C., are the primary physicians responsible for the patient medical treatment; Y.L.C., reviews the surgical pathology and issues the report.; S.C.H., Y.M.H., C.H.L., contributed to the final version of the manuscript; All authors actively participated in the discussion and suggestion for the manuscript. Funding This article was subsidized for English editing by 10.13039/501100006477 National Taiwan University under the Excellence Improvement Program for Doctoral Students (grant number 108-2926-I-002-002-MY4), sponsored by the 10.13039/501100004663 Ministry of Science and Technology, Taiwan . This work was supported by Ministry of Science and Technology (MOST 108-2314-B-002-097-MY3, 107-2314-B-002-253-, 106-2314-B-002-100-), 10.13039/501100005762 National Taiwan University Hospital (NTUH 110-M4972) and Taiwan Rheumatology Association Research Fund (TRARF-2018). Ethics approval This study was approved by the hospital Institutional Review Board (NTUH: 201612147RINA). The informed consent was obtained from the patients. Availability of data and material The author(s) confirm that the data supporting the findings of this study are available within the article. Code availability The author(s) confirm that the data supporting the findings of this study are available within the article. Declaration of competing interest The author(s) had declared that no conflicts of interest exist. Appendix A Supplementary data The following is the Supplementary data to this article:Multimedia component 1 Multimedia component 1 Appendix A Supplementary data to this article can be found online at https://doi.org/10.1016/j.breast.2021.11.016.

PACLITAXEL, TRASTUZUMAB

DrugsGivenReaction

CC BY-NC-ND

34894465

2021-11-29

Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Pulmonary vasculitis'.

The first reported case of trastuzumab induced interstitial lung disease associated with anti-neutrophil cytoplasmic antibody vasculitis - A case report and a prospective cohort study on the usefulness of neutrophil derived biomarkers in monitoring vasculitis disease activity during follow-up. Targeted therapies against human epidermal growth factor receptor 2 (HER2) are associated with increased interstitial lung disease (ILD). Trastuzumab, lapatinib, pertuzumab, and trastuzumab emtansine have markedly extended HER2 breast cancer survival but current knowledge on how these HER2-targeted agents induce interstitial lung disease is still poorly defined due to limited cases in the literature. Physicians mostly managed this complication by dose interruption, dose de-escalation, or discontinuation with success. In 2019, the FDA had granted accelerated approval on trastuzumab deruxtecan (T-Dxd) in HER2 breast cancer in the late line setting. Severe ILD incidence rate was over ten percent and led to fatal outcomes in 2.2% of patients in the T-Dxd trial. Searching for biomarkers to detect ILD incidence before it becomes clinically fulminant or for treatment response monitoring is of high clinical value. A Case of life-threatening trastuzumab-induced ILD was encountered in our facility. The ILD was confirmed to be antineutrophil cytoplasmic antibody (ANCA) pulmonary capillaritis. The biomarker of neutrophil extracellular traps (NETs), serum MPO-DNA complex, showed a good correlation with the clinical severity. Soon after B cell depleting agent rituximab usage, the serum MPO-DNA outperformed ANCA autoantibody and maintained its correlation with clinical severity. In addition to the trastuzumab-induced ILD case, a prospective cohort in our facility also confirmed the usefulness of MPO-DNA in monitoring vasculitis activity. We postulated that upfront testing with biomarkers of vasculitis during HER2 targeted treatment with high ILD incidence may be beneficial in the future. pmc1 Introduction The HER-2 gene is a poor prognostic factor in breast cancer and is amplified in 20–25% of patients [1]. Trastuzumab is a humanized monoclonal antibody that targets the extracellular domain of the human epidermal growth factor receptor 2 (HER2). Since its first approval in 1998, subsequently developed HER2 agents such as lapatinib, pertuzumab, and trastuzumab emtansine have markedly extended the HER2 breast cancer patient survival [[2], [3], [4]]. Drug-induced interstitial lung disease (DIILD) accounts for 3–5% of interstitial lung disease. Cancer therapy is the leading cause of DIILD and accounts for 23–51% of the cases [5]. Although the incidence of trastuzumab-related ILD in the registration trials was low at 0.5%, the mortality rate with treatment-related ILD was around 20% [[6], [7], [8]]. In 2019 the novel HER-2 antibody-drug conjugate trastuzumab deruxtecan (T-Dxd) had recently drawn attention for an increased ILD incidence while being approved to treat late-stage HER-2 positive metastatic breast cancer. T-Dxd therapy achieved a response rate up to 60%, but at the cost of ILD up to 13.6%, and 2.2% treatment-related death [9]. Therefore an unmet medical need for developing effective methods in predicting or monitoring HER-2 targeted agents induced ILD is required. 1.1 Case history A 59-year-old female was diagnosed with invasive ductal carcinoma after undergoing a core needle biopsy for a right breast lump. She received a simple mastectomy, and the surgical pathology showed stage IIA, pT2N0, HER-2/neu 3+, weak positive estrogen receptor (2%), and was negative for progesterone receptor. Adjuvant therapy comprising 12 weeks of weekly paclitaxel plus 52 weeks of tri-weekly trastuzumab was initiated three weeks after the surgery. However, at week 16, the treatment course was interrupted due to an ER visit for blood-tinged sputum over five days. A chest X-ray showed bilateral centrilobular pulmonary infiltration (Fig. 1A). The complete blood count showed neutrophilic leukocytosis (WBC 11.01 k/μL, neutrophil 74.2%, lymphocyte 17.6%, eosinophil 2.5%), anemia and thrombocytopenia (hemoglobin 9.8 g/dL, platelet 126 k/μL). The coagulation profile, as well as liver and kidney biochemistry, were within normal limits. She was admitted for empirical community-acquired pneumonia treatment with ceftriaxone and azithromycin. However, on the third day, the patient's chest X-ray showed rapid infiltration progression (Fig. 1B), and she had a fever that spiked to 39 °C with hypoxia requiring oxygen mask support. A chest CT with contrast showed diffuse peribronchovascular consolidation (Fig. 1C–D). The culture and serology on multiple pneumonia pathogens, including S. pneumonia, Legionella, Chlamydia, Influenza, Aspergillus, Cryptococcus, and CMV virus, showed negative results. Cultures for blood and sputum were also negative. Non-infectious pneumonitis was suspected, and a connective tissue disease survey was conducted. However, hypoxic respiratory failure progressed, and the patient received endotracheal intubation and was transferred to the intensive care unit. The patient underwent a bronchoscopy examination that showed diffuse mucosal bleeding of the entire airway (Fig. 2A and B). Bronchial wash cytology showed neutrophil predominance without malignant cells. Bronchial wash specimens showed no microorganisms, and the cultures remained negative. However, the connective tissue disease survey revealed antineutrophil cytoplasmic antibody (ANCA) positivity both via an indirect immunofluorescence method (IMMCO Diagnostics) and captured enzyme-linked immunoassays (EliA). The survey showed positive neutrophil cytoplasmic staining (cANCA, cytoplasmic ANCA) and an anti-PR3 titer of 40.28 IU/mL (normal <2.0). The antinuclear antibody, anti-dsDNA, rheumatoid factor, C3, C4, anti-GBM, and anti-SSA/SSB were negative.Fig. 1 A-D. (A) Chest radiograph (CXR) at initial presentation revealed multi-lobar consolidation with a centi-hilar pattern. (B) CXR on the third day showed rapid bilateral pulmonary infiltrate progression (C–D) Computed tomography (CT) on the third day showed a diffuse, multi-lobar consolidation with a peri-bronchovascular pattern. Fig. 1 Fig. 2 A-B Bronchoscopy examination of the tracheal (A) and main carina (B) revealed fresh blood in the whole airway without active vascular bleeding. Fig. 2 A lung biopsy via video-assisted thoracoscopic surgery (VATS) was performed for tissue diagnosis. Microscopically, the resected lung tissue showed marked interstitial fibrosis and residual airspaces of varying size (Fig. 3A). The remaining airspaces revealed extensive hemorrhage, as demonstrated by the presence of erythrocytes and histiocytes mixed with fibrinous exudate in the alveolar spaces. The presence of necrotic neutrophils in the alveolar septa and airspaces were indicative of acute capillaritis (Fig. 3B). A diagnosis of acute capillaritis associated with diffuse alveolar damage of the organizing stage was made (Fig. 3C). Treatment with intravenous methylprednisolone (1 mg/kg/day) was administered. The fever resolved, and the chest X-ray showed improvement on the seventh day. The patient was extubated after 14 days of ventilator support.Fig. 3 A Lung, left upper lobe, wedge resection, low power field (100X), interstitial fibrosis with collagen deposits (C) and alveolar airspace obliteration (A). B High power field showing alveolar hemorrhage (A), and necrotic neutrophils (P) in a capillary wall (V) and alveoli. C Comparison of the patient sample to a sample from a healthy individual shows marked airspace obliteration due to hemorrhage and interstitial thickening. Fig. 3 Trastuzumab-related ANCA capillaritis was diagnosed after reviewing the patient's history. She had no pulmonary symptoms or chest X-ray abnormality in previous health check-ups. The Naranjo algorithm, an adverse drug reaction (ADE) assessment tool having a scale of 1–10 [10], was applied with trastuzumab scoring 7 out of 10 (Table 1). Probable trastuzumab-related ADE was suspected. Considering this ADE as a life-threatening event, the trastuzumab schedule was permanently discontinued and replaced with hormonal adjuvant therapy. She also received rheumatologist follow up since this event at the outpatient clinic.Table 1 Probable Trastuzumab related adverse drug reaction, with Naranjo score of 7 according to the patients clinical history. Table 1No Question Yes No Do Not Know I Are there previous conclusive reports on this reaction? +1 0 0 II Did the adverse event appear after the suspected drug was administered? +2 −1 0 III Did the adverse reaction improve when the drug was discontinued or a specific antagonist was administered? +1 0 0 VI Did the adverse reaction reappear when the drug was administered? +2 −1 0 V Are there alternative causes (other than the drug) that could on their own have caused the reaction? −1 +2 0 VI Did the reaction reappear when a placebo was given? −1 +1 0 VII Was the drug detected in the blood (or other fluids) in concentration known to be toxic? +1 0 0 VIII Was the reaction more severe when the dose was increased or less severe when the dose was decreased? +1 0 0 IX Did the patient have a similar reaction to the same or similar drugs in any previous exposure? +1 0 0 X Was the adverse event confirmed by any objective evidence? +1 0 0 Naranjo score for estimating the probability of adverse drug reactions; 0, doubtful ADR; 1–4 possible ADR; 5–8, probable ADR; ≥9 definite ADR. 1.2 Utilization of different vasculitis markers in the trastuzumab ILD case Systemic methylprednisolone had induced rapid remission of the chest X-ray infiltration and inflammatory marker (e.g. C-reactive protein, erythrocyte sedimentation rate, D-dimer) improvement during the initial induction period (Table 2). However, the patient still manifested with residual vasculitis disease activity of leg weakness, numbness, and also persistent microscopic hematuria and proteinuria. Occasional expectoration of blood clot content was also noted despite no evident change in the chest X-ray appearance. The measurement of Birmingham Vasculitis Activity Score (BVAS) [11], a tool of vasculitis activity measurement consisted of 9 organs, and a score of 0–33 for persistent symptoms, and 0 to 63 for new or worse symptoms, still scored 10–20 for persistent symptoms in the patient. The patient also encountered a rapid increase in the anti-PR3 titer eight months after discontinuation of trastuzumab (Table 2). Recent evidence has supported neutrophil as the dominant infiltrate within vasculitis lesions [[12], [13], [14]]. And the discovery of neutrophil extracellular traps (NETs), a component of cell-free DNA, histone, proteinase 3 (PR3), and myeloperoxidase (MPO) released by ANCA-stimulated neutrophil, could induce vasculitis damage such as thrombus formation and endothelial damage. The patient was enrolled in a prospective vasculitis cohort in our facility for testing on vasculitis neutrophil-derived NETs and had also started rituximab (500 mg D1/D15, every 6 months) therapy for better control for the DIILD.Table 2 Trend of inflammatory markers since interstitial lung disease event onset; arrowheads indicate rituximab treatment. MPO-DNA showed a good correlation with Birmingham Vasculitis Activity Score (BVAS), especially in the late phase (r = 0.996, p = 0.004) after Rituximab treatment. This AAV patient had a protracted disease course that involved BVAS score rebound despite the PR3 autoantibody titer remaining below the upper limit. The results indicate that the MPO-DNA complex, which is a novel biomarker of neutrophil activation, could better monitor vasculitis activity. Table 2 2 Prospective cohort on neutrophil derived biomarkers in vasculitis disease activity 2.1 Materials and methods From 2017 to 2021, a prospective cohort study addressing neutrophil related biomarkers in the evaluation of patients with vasculitis was initiated in our facility (Suppl. Fig. S1). The aim was to explore the relationship between levels of cell-free MPO-DNA, a biomarker for NETosis and the clinical activity of systemic vasculitis. Including the Case presented in the manuscript, a total of eight patients with vasculitis and 17 healthy controls (a ratio of 1:2) were enrolled, and serial serum testing was obtained (Suppl. Fig. S2). The study was approved by the hospital Institutional Review Board in 2017 (NTUH: 201612147RINA). Informed consents were obtained from all patients and healthy donors. 2.2 Measurement of MPO-DNA and ANCA We tested each vasculitis individual for the cell-free MPO-DNA via a “sandwich” ELISA with an anti-MPO polyclonal antibody (GeneTex, GTX22088, Irvine, Ca, USA). ELISA microplates were coated with the MPO monoclonal antibody overnight to capture MPO-associated DNAs. Anti-ds DNA-specific monoclonal antibodies (Abcam, Cambridge, ab27156, UK; 1:2000) were added to bind MPO-associated DNA, followed by binding of a horseradish peroxidase-conjugated anti-mouse IgG antibody (Jackson ImmunoResearch, 115-035-003) for detection. A peroxidase substrate (3,3′,5,5′-Tetramethylbenzidine) was added to react with the conjugated peroxidase to yield a blue product. The reaction was halted by adding 2 N H2SO4, and the absorbance of the final product was measured at 450 nm and was transformed to an arbitrary unit (a.u.) according to the previous study [15]. An MPO-DNA assay standard curve is provided (Suppl. Fig. S3). ANCA assessment was performed using anti-proteinase (PR3) and anti-myeloperoxidase(MPO) specific immunoassays (EliA; Thermo Fisher Scientific, Waltham, MA, USA). The assay was performed according to the manufacturer's instructions. An antibody concentration was considered positive if: MPO >5.0 IU/mL and PR3 >3.0 IU/mL (normal range were provided by the manufacturer). 2.3 Statistical analysis We utilized Stata statistical software (V. 14.0, Stata Corporation, College Station, TX) for the analysis. The MPO-DNA cut-off point between healthy controls and vasculitis patients was obtained according to the maximum Youden index (Sensitivity + Specificity - 1) to capture the performance of this dichotomous diagnostic test [16]. Statistical significance of the difference between two sets of continuous variables was analyzed using a Mann-Whitney U test. A p-value less than 0.05 was defined as statistically significant. 3 Results 3.1 MPO DNA performance in the prospective vasculitis cohort The MPO-DNA values were significantly higher in the systemic vasculitis group (n = 8; 0.092 ± 0.071 arbitrary unit (a.u.)) relative to healthy donors (n = 17; 0.015 ± 0.018 a.u., p = 0.01) (Table 3). The Youden J index representing the maximum potential effectiveness of the upper-lower limit of the MPO-DNA cut-off value was 0.047 (a.u.), the red dashed line (Table 3). An MPO-DNA cut-off level at ≤ 0.047 a.u. was established to differentiate between healthy control and vasculitis patients. We also tested MPO-DNA serum level, erythrocyte sedimentation rate, C-reactive protein correlation with their Birmingham Vasculitis Activity Score. The MPO-DNA showed a moderate positive correlation [ r (n = 12) = 0.596, p = 0.025]; but the C-reactive protein [ r(n = 10) = 0.402, p = 0.196] and the erythrocyte sedimentation rate [ r(n = 9) = 0.119, p = 0.727] showed no correlation with clinical vasculitis severity. These results suggest that MPO-DNA levels may also be useful in monitoring vasculitis activity compared to standard inflammation biomarkers (Table 4).Table 3 MPO-DNA values were significantly higher in the systemic vasculitis group [n = 8; 0.092 ± 0.071 arbitrary units (A.U.)] relative to healthy donors [ n = 17; 0.015 ± 0.018 A U., p = 0.01 ]. Youden J index representing the maximum potential effectiveness of the upper-lower limit of the MPO-DNA cut-off value, which was set at 0.047 (A.U.), as indicated by the red dashed line. Table 3 Table 4 The MPO-DNA serum levels, erythrocyte sedimentation rates, and C-reactive protein correlation with the Birmingham Vasculitis Activity Score for the eight vasculitis patients, the MPO-DNA showed a moderate positive correlation but the C-reactive protein and erythrocyte sedimentation rate did not. Table 4 3.2 MPO DNA performance in the trastuzumab ILD case During follow-up of the Case presenting trastuzumab-induced ILD, we found that despite standard vasculitis anti-PR3 is capable of showing good correlation during initial glucocorticoid therapy with the Birmingham vasculitis score (BVAS). The correlation was lost after salvage treatment with the B-cell targeting agent rituximab was initiated (Table 2). MPO-DNA levels exhibited a better correlation with the clinical vasculitis activity BVAS score during the rituximab treatment period. 4 Discussion Interstitial lung disease (ILD) induced by HER2-targeted agents is a well-known adverse drug reaction, but the mechanism is ill-defined and considered low in incidence [17]. HER-2 targeted agents such as trastuzumab, trastuzumab emtansine had been reported to induce cutaneous vasculitis in the previous literature [[18], [19], [20]], but pulmonary vasculitis due to HER2 targeted agents had not been reported yet. A recent review of 9886 patients investigating anti-HER2 therapies for HER2 breast cancer had reported the overall incidence of ILD was 2.4%. The incidence of grade 1–2, grade 3–4, and grade 5 events were 66.7%, 23.0%, and 0.2% respectively. The agents leading to the highest ILD incidence was trastuzumab combined with everolimus or paclitaxel. The incidence of ILD-related deaths was highest among patients receiving trastuzumab deruxtecan (T-Dxd), with an incidence of around 2% [17]. Recently the oncology society has drawn attention to this adverse event due to the recent accelerated approval of T-Dxd [9,21]. The novel antibody-drug conjugate had shown a 60% response rate in third or later line metastatic HER-2 positive breast cancer but at the cost of treatment-related ILD up to 13% and 2% treatment-related death. This high rate of side effects will inhibit its potential to be utilized as the frontline therapy or even be placed in the curative adjuvant or neoadjuvant setting. A recently published article by Kumagai et al. had reported successful induction of T-Dxd interstitial lung disease in a cynomolgus monkey model. Receiving T-Dxd in a monkey model developed interstitial lung disease, whereas receiving Dxd does not. Although most ILD lesions were found within the alveoli, the HER2 expression in lungs was limited to the bronchial level [22]. Vasculitis due to a pathway via ANCA autoantibodies, an indirect mechanism by the immune system, could explain why the injury was not at the location where T-Dxd was uptaken. Moreover, a two-step mechanism of pathogenesis explains how DIILD (drug-induced ILD) manifests later after anti-HER2 therapies. A literature review was performed to summarize previously reported cases of trastuzumab-related ILD. A total of 8 cases were identified before our Case [[23], [24], [25], [26], [27], [28]]. Two of the reported cases had available pathology that showed organizing pneumonia or diffuse alveolar damage. No traceable serum marker was reported in previously reported patients. All previously reported cases that we identified had late onset with a median time of onset two months after the first trastuzumab exposure (Table 5). The pattern of CT infiltration largely comprised ground-glass opacity and patch consolidations, with only one case exhibiting nodular lesions. Most patients recovered following prednisolone-based monotherapy, and only one patient died. Long-term follow up was not conducted in these cases. Our case also had a protracted course of vasculitis up to two years. We found that both ANCA auto-antibody testing and NET related biomarker, MPO-DNA should be tested concomitantly and especially when B cell depleting agents (such as rituximab use in our case) inhibit the autoantibody production, but vasculitis-related neutrophil activity is still high. Reliance on ANCA autoantibodies alone to monitor vasculitis activity may not be sufficient.Table 5 Case comparison of previously reported trastuzumab-related interstitial lung disease [[12], [13], [14], [15], [16], [17]]. Table 5Case Age Gender Previous CRT (setting) Tmab exposure CT pattern Pathology Treatment Recovery Naranjo Score 1 49 F [23] AC, RT, D (adjuvant) 3 months Subpleural consolidation Organizing pneumonia n/aa Yes 6 2 56 F [25] FAC, D (salvage) 4 months Consolidation and pleural effusion BAL: eosinophil 18% Prednisolone (40 mg/d) Yes 8 3 51 F [26] P (neoadjuvant) 10 weeks Ground glass opacity TBLB: Intra-alveolar hemorrhage, interstitial inflammation Prednisolone (40 mg/d) Yes 6 4 63 F [24] D (neoadjuvant) 5 weeks Peripheral consolidation, GGO Diffuse alveolar damage Prednisonec No (mortality) 7 5 67 F [35] DC (adjuvant) 4 months GGO n/a Prednisone Cyclophosphamideb Yes 8 6 68 F [28] EC (adjuvant) 3 months Patch infiltration, GGO n/a Semi-pulse steroidc Yes 6 7 62 F [28] DC (adjuvant) 3 months Patchy consolidation, poorly defined nodules n/a Prednisolonec Yes 7 8 71 F [36] Salvage Tmab followed by T-DM1 Tmab 12 wks T-DM1 6 wks Diffuse interstitial infiltrates n/a Methylprednisolone (1 mg/kg/d) Yes 5 9 59 F P (adjuvant) 3 months Patchy consolidation ANCA capillaritis, acute interstitial pneumonia Prednisolone (40 mg/d) Yes 7 CRT, chemoradiotherapy; Tmab, trastuzumab; T-DM1, trastuzumab emtansine; GGO, ground glass opacity; A, doxorubicin; C, cyclophosphamide; RT, radiotherapy; F, 5-fluorouracil; P, paclitaxel; D, docetaxel; BAL, bronchoalveolar lavage; TBLB, transbronchial lung biopsy. a Resolved spontaneously after trastuzumab discontinuation. b Cyclophosphamide was used upon trastuzumab rechallenge. c The dose used was not reported. Systemic glucocorticoids combined with either rituximab or cyclophosphamide are recommended for induction therapy for life-threatening AAV [29]. Autoreactive B cell depletion by rituximab can effectively reduce disease activity and decrease pathogenic ANCA concentrations in AAV patients. Although adverse events after rituximab include hepatitis B reactivation [30] and progressive multifocal leukoencephalopathy [31], the safety profile of rituximab is still considered to be favorable. Further salvage therapy or investigational agents included belimumab [32], a monoclonal antibody against soluble B cell-activating factor (BAFF) that can induce B cell apoptosis and showed clinical activity as an add-on therapy. The use of therapeutic plasma exchange is also beneficial for carefully selected patients experiencing severe diffuse alveolar hemorrhage or a serum creatinine level of ≥500 mmol/L [33]. There are still two crucial pieces of information that cannot be determined from our Case. First, neither serum levels of anti-PR3/anti-MPO level prior to trastuzumab exposure were available for this case. A prospective examination of vasculitis markers may be considered in the future to determine whether ANCA has clinical utility in predicting side effects during HER2 targeted therapy, especially antibody or antibody drug-conjugates. Secondly, the majority of drug-induced AAV cases present with anti-MPO positivity [34], but our case presented with anti-PR3 positivity without any upper airway symptoms or granuloma formation. Whether this difference is related to the medication administered or due to individual patient differences cannot be explained at present. In conclusion, this Case is the first clinical evidence of HER-2 targeted therapy induces pulmonary ILD via ANCA autoantibodies. Early suspicion with testing of ANCA level can be life-saving when a biopsy cannot be immediately obtained. Pre-treatment ANCA level testing may also be considered in the future when treating patients with HER-2 agents with higher risk profiles of ILD. In addition, serum MPO-DNA, could be a biomarker with promising clinical implication, as the results from this case and a prospective vasculitis cohort in our facility indicated. Authors’ contribution C.H.C., Y.M.H., and Y.M.K., wrote the original draft writing with input from all authors; Y.M.K., C.J.J., designed, developed and conducted the experiments; Y.M.K., L.C., are the primary physicians responsible for the patient medical treatment; Y.L.C., reviews the surgical pathology and issues the report.; S.C.H., Y.M.H., C.H.L., contributed to the final version of the manuscript; All authors actively participated in the discussion and suggestion for the manuscript. Funding This article was subsidized for English editing by 10.13039/501100006477 National Taiwan University under the Excellence Improvement Program for Doctoral Students (grant number 108-2926-I-002-002-MY4), sponsored by the 10.13039/501100004663 Ministry of Science and Technology, Taiwan . This work was supported by Ministry of Science and Technology (MOST 108-2314-B-002-097-MY3, 107-2314-B-002-253-, 106-2314-B-002-100-), 10.13039/501100005762 National Taiwan University Hospital (NTUH 110-M4972) and Taiwan Rheumatology Association Research Fund (TRARF-2018). Ethics approval This study was approved by the hospital Institutional Review Board (NTUH: 201612147RINA). The informed consent was obtained from the patients. Availability of data and material The author(s) confirm that the data supporting the findings of this study are available within the article. Code availability The author(s) confirm that the data supporting the findings of this study are available within the article. Declaration of competing interest The author(s) had declared that no conflicts of interest exist. Appendix A Supplementary data The following is the Supplementary data to this article:Multimedia component 1 Multimedia component 1 Appendix A Supplementary data to this article can be found online at https://doi.org/10.1016/j.breast.2021.11.016.

TRASTUZUMAB

DrugsGivenReaction

CC BY-NC-ND

34894465

2021-11-29

Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Respiratory failure'.

The first reported case of trastuzumab induced interstitial lung disease associated with anti-neutrophil cytoplasmic antibody vasculitis - A case report and a prospective cohort study on the usefulness of neutrophil derived biomarkers in monitoring vasculitis disease activity during follow-up. Targeted therapies against human epidermal growth factor receptor 2 (HER2) are associated with increased interstitial lung disease (ILD). Trastuzumab, lapatinib, pertuzumab, and trastuzumab emtansine have markedly extended HER2 breast cancer survival but current knowledge on how these HER2-targeted agents induce interstitial lung disease is still poorly defined due to limited cases in the literature. Physicians mostly managed this complication by dose interruption, dose de-escalation, or discontinuation with success. In 2019, the FDA had granted accelerated approval on trastuzumab deruxtecan (T-Dxd) in HER2 breast cancer in the late line setting. Severe ILD incidence rate was over ten percent and led to fatal outcomes in 2.2% of patients in the T-Dxd trial. Searching for biomarkers to detect ILD incidence before it becomes clinically fulminant or for treatment response monitoring is of high clinical value. A Case of life-threatening trastuzumab-induced ILD was encountered in our facility. The ILD was confirmed to be antineutrophil cytoplasmic antibody (ANCA) pulmonary capillaritis. The biomarker of neutrophil extracellular traps (NETs), serum MPO-DNA complex, showed a good correlation with the clinical severity. Soon after B cell depleting agent rituximab usage, the serum MPO-DNA outperformed ANCA autoantibody and maintained its correlation with clinical severity. In addition to the trastuzumab-induced ILD case, a prospective cohort in our facility also confirmed the usefulness of MPO-DNA in monitoring vasculitis activity. We postulated that upfront testing with biomarkers of vasculitis during HER2 targeted treatment with high ILD incidence may be beneficial in the future. pmc1 Introduction The HER-2 gene is a poor prognostic factor in breast cancer and is amplified in 20–25% of patients [1]. Trastuzumab is a humanized monoclonal antibody that targets the extracellular domain of the human epidermal growth factor receptor 2 (HER2). Since its first approval in 1998, subsequently developed HER2 agents such as lapatinib, pertuzumab, and trastuzumab emtansine have markedly extended the HER2 breast cancer patient survival [[2], [3], [4]]. Drug-induced interstitial lung disease (DIILD) accounts for 3–5% of interstitial lung disease. Cancer therapy is the leading cause of DIILD and accounts for 23–51% of the cases [5]. Although the incidence of trastuzumab-related ILD in the registration trials was low at 0.5%, the mortality rate with treatment-related ILD was around 20% [[6], [7], [8]]. In 2019 the novel HER-2 antibody-drug conjugate trastuzumab deruxtecan (T-Dxd) had recently drawn attention for an increased ILD incidence while being approved to treat late-stage HER-2 positive metastatic breast cancer. T-Dxd therapy achieved a response rate up to 60%, but at the cost of ILD up to 13.6%, and 2.2% treatment-related death [9]. Therefore an unmet medical need for developing effective methods in predicting or monitoring HER-2 targeted agents induced ILD is required. 1.1 Case history A 59-year-old female was diagnosed with invasive ductal carcinoma after undergoing a core needle biopsy for a right breast lump. She received a simple mastectomy, and the surgical pathology showed stage IIA, pT2N0, HER-2/neu 3+, weak positive estrogen receptor (2%), and was negative for progesterone receptor. Adjuvant therapy comprising 12 weeks of weekly paclitaxel plus 52 weeks of tri-weekly trastuzumab was initiated three weeks after the surgery. However, at week 16, the treatment course was interrupted due to an ER visit for blood-tinged sputum over five days. A chest X-ray showed bilateral centrilobular pulmonary infiltration (Fig. 1A). The complete blood count showed neutrophilic leukocytosis (WBC 11.01 k/μL, neutrophil 74.2%, lymphocyte 17.6%, eosinophil 2.5%), anemia and thrombocytopenia (hemoglobin 9.8 g/dL, platelet 126 k/μL). The coagulation profile, as well as liver and kidney biochemistry, were within normal limits. She was admitted for empirical community-acquired pneumonia treatment with ceftriaxone and azithromycin. However, on the third day, the patient's chest X-ray showed rapid infiltration progression (Fig. 1B), and she had a fever that spiked to 39 °C with hypoxia requiring oxygen mask support. A chest CT with contrast showed diffuse peribronchovascular consolidation (Fig. 1C–D). The culture and serology on multiple pneumonia pathogens, including S. pneumonia, Legionella, Chlamydia, Influenza, Aspergillus, Cryptococcus, and CMV virus, showed negative results. Cultures for blood and sputum were also negative. Non-infectious pneumonitis was suspected, and a connective tissue disease survey was conducted. However, hypoxic respiratory failure progressed, and the patient received endotracheal intubation and was transferred to the intensive care unit. The patient underwent a bronchoscopy examination that showed diffuse mucosal bleeding of the entire airway (Fig. 2A and B). Bronchial wash cytology showed neutrophil predominance without malignant cells. Bronchial wash specimens showed no microorganisms, and the cultures remained negative. However, the connective tissue disease survey revealed antineutrophil cytoplasmic antibody (ANCA) positivity both via an indirect immunofluorescence method (IMMCO Diagnostics) and captured enzyme-linked immunoassays (EliA). The survey showed positive neutrophil cytoplasmic staining (cANCA, cytoplasmic ANCA) and an anti-PR3 titer of 40.28 IU/mL (normal <2.0). The antinuclear antibody, anti-dsDNA, rheumatoid factor, C3, C4, anti-GBM, and anti-SSA/SSB were negative.Fig. 1 A-D. (A) Chest radiograph (CXR) at initial presentation revealed multi-lobar consolidation with a centi-hilar pattern. (B) CXR on the third day showed rapid bilateral pulmonary infiltrate progression (C–D) Computed tomography (CT) on the third day showed a diffuse, multi-lobar consolidation with a peri-bronchovascular pattern. Fig. 1 Fig. 2 A-B Bronchoscopy examination of the tracheal (A) and main carina (B) revealed fresh blood in the whole airway without active vascular bleeding. Fig. 2 A lung biopsy via video-assisted thoracoscopic surgery (VATS) was performed for tissue diagnosis. Microscopically, the resected lung tissue showed marked interstitial fibrosis and residual airspaces of varying size (Fig. 3A). The remaining airspaces revealed extensive hemorrhage, as demonstrated by the presence of erythrocytes and histiocytes mixed with fibrinous exudate in the alveolar spaces. The presence of necrotic neutrophils in the alveolar septa and airspaces were indicative of acute capillaritis (Fig. 3B). A diagnosis of acute capillaritis associated with diffuse alveolar damage of the organizing stage was made (Fig. 3C). Treatment with intravenous methylprednisolone (1 mg/kg/day) was administered. The fever resolved, and the chest X-ray showed improvement on the seventh day. The patient was extubated after 14 days of ventilator support.Fig. 3 A Lung, left upper lobe, wedge resection, low power field (100X), interstitial fibrosis with collagen deposits (C) and alveolar airspace obliteration (A). B High power field showing alveolar hemorrhage (A), and necrotic neutrophils (P) in a capillary wall (V) and alveoli. C Comparison of the patient sample to a sample from a healthy individual shows marked airspace obliteration due to hemorrhage and interstitial thickening. Fig. 3 Trastuzumab-related ANCA capillaritis was diagnosed after reviewing the patient's history. She had no pulmonary symptoms or chest X-ray abnormality in previous health check-ups. The Naranjo algorithm, an adverse drug reaction (ADE) assessment tool having a scale of 1–10 [10], was applied with trastuzumab scoring 7 out of 10 (Table 1). Probable trastuzumab-related ADE was suspected. Considering this ADE as a life-threatening event, the trastuzumab schedule was permanently discontinued and replaced with hormonal adjuvant therapy. She also received rheumatologist follow up since this event at the outpatient clinic.Table 1 Probable Trastuzumab related adverse drug reaction, with Naranjo score of 7 according to the patients clinical history. Table 1No Question Yes No Do Not Know I Are there previous conclusive reports on this reaction? +1 0 0 II Did the adverse event appear after the suspected drug was administered? +2 −1 0 III Did the adverse reaction improve when the drug was discontinued or a specific antagonist was administered? +1 0 0 VI Did the adverse reaction reappear when the drug was administered? +2 −1 0 V Are there alternative causes (other than the drug) that could on their own have caused the reaction? −1 +2 0 VI Did the reaction reappear when a placebo was given? −1 +1 0 VII Was the drug detected in the blood (or other fluids) in concentration known to be toxic? +1 0 0 VIII Was the reaction more severe when the dose was increased or less severe when the dose was decreased? +1 0 0 IX Did the patient have a similar reaction to the same or similar drugs in any previous exposure? +1 0 0 X Was the adverse event confirmed by any objective evidence? +1 0 0 Naranjo score for estimating the probability of adverse drug reactions; 0, doubtful ADR; 1–4 possible ADR; 5–8, probable ADR; ≥9 definite ADR. 1.2 Utilization of different vasculitis markers in the trastuzumab ILD case Systemic methylprednisolone had induced rapid remission of the chest X-ray infiltration and inflammatory marker (e.g. C-reactive protein, erythrocyte sedimentation rate, D-dimer) improvement during the initial induction period (Table 2). However, the patient still manifested with residual vasculitis disease activity of leg weakness, numbness, and also persistent microscopic hematuria and proteinuria. Occasional expectoration of blood clot content was also noted despite no evident change in the chest X-ray appearance. The measurement of Birmingham Vasculitis Activity Score (BVAS) [11], a tool of vasculitis activity measurement consisted of 9 organs, and a score of 0–33 for persistent symptoms, and 0 to 63 for new or worse symptoms, still scored 10–20 for persistent symptoms in the patient. The patient also encountered a rapid increase in the anti-PR3 titer eight months after discontinuation of trastuzumab (Table 2). Recent evidence has supported neutrophil as the dominant infiltrate within vasculitis lesions [[12], [13], [14]]. And the discovery of neutrophil extracellular traps (NETs), a component of cell-free DNA, histone, proteinase 3 (PR3), and myeloperoxidase (MPO) released by ANCA-stimulated neutrophil, could induce vasculitis damage such as thrombus formation and endothelial damage. The patient was enrolled in a prospective vasculitis cohort in our facility for testing on vasculitis neutrophil-derived NETs and had also started rituximab (500 mg D1/D15, every 6 months) therapy for better control for the DIILD.Table 2 Trend of inflammatory markers since interstitial lung disease event onset; arrowheads indicate rituximab treatment. MPO-DNA showed a good correlation with Birmingham Vasculitis Activity Score (BVAS), especially in the late phase (r = 0.996, p = 0.004) after Rituximab treatment. This AAV patient had a protracted disease course that involved BVAS score rebound despite the PR3 autoantibody titer remaining below the upper limit. The results indicate that the MPO-DNA complex, which is a novel biomarker of neutrophil activation, could better monitor vasculitis activity. Table 2 2 Prospective cohort on neutrophil derived biomarkers in vasculitis disease activity 2.1 Materials and methods From 2017 to 2021, a prospective cohort study addressing neutrophil related biomarkers in the evaluation of patients with vasculitis was initiated in our facility (Suppl. Fig. S1). The aim was to explore the relationship between levels of cell-free MPO-DNA, a biomarker for NETosis and the clinical activity of systemic vasculitis. Including the Case presented in the manuscript, a total of eight patients with vasculitis and 17 healthy controls (a ratio of 1:2) were enrolled, and serial serum testing was obtained (Suppl. Fig. S2). The study was approved by the hospital Institutional Review Board in 2017 (NTUH: 201612147RINA). Informed consents were obtained from all patients and healthy donors. 2.2 Measurement of MPO-DNA and ANCA We tested each vasculitis individual for the cell-free MPO-DNA via a “sandwich” ELISA with an anti-MPO polyclonal antibody (GeneTex, GTX22088, Irvine, Ca, USA). ELISA microplates were coated with the MPO monoclonal antibody overnight to capture MPO-associated DNAs. Anti-ds DNA-specific monoclonal antibodies (Abcam, Cambridge, ab27156, UK; 1:2000) were added to bind MPO-associated DNA, followed by binding of a horseradish peroxidase-conjugated anti-mouse IgG antibody (Jackson ImmunoResearch, 115-035-003) for detection. A peroxidase substrate (3,3′,5,5′-Tetramethylbenzidine) was added to react with the conjugated peroxidase to yield a blue product. The reaction was halted by adding 2 N H2SO4, and the absorbance of the final product was measured at 450 nm and was transformed to an arbitrary unit (a.u.) according to the previous study [15]. An MPO-DNA assay standard curve is provided (Suppl. Fig. S3). ANCA assessment was performed using anti-proteinase (PR3) and anti-myeloperoxidase(MPO) specific immunoassays (EliA; Thermo Fisher Scientific, Waltham, MA, USA). The assay was performed according to the manufacturer's instructions. An antibody concentration was considered positive if: MPO >5.0 IU/mL and PR3 >3.0 IU/mL (normal range were provided by the manufacturer). 2.3 Statistical analysis We utilized Stata statistical software (V. 14.0, Stata Corporation, College Station, TX) for the analysis. The MPO-DNA cut-off point between healthy controls and vasculitis patients was obtained according to the maximum Youden index (Sensitivity + Specificity - 1) to capture the performance of this dichotomous diagnostic test [16]. Statistical significance of the difference between two sets of continuous variables was analyzed using a Mann-Whitney U test. A p-value less than 0.05 was defined as statistically significant. 3 Results 3.1 MPO DNA performance in the prospective vasculitis cohort The MPO-DNA values were significantly higher in the systemic vasculitis group (n = 8; 0.092 ± 0.071 arbitrary unit (a.u.)) relative to healthy donors (n = 17; 0.015 ± 0.018 a.u., p = 0.01) (Table 3). The Youden J index representing the maximum potential effectiveness of the upper-lower limit of the MPO-DNA cut-off value was 0.047 (a.u.), the red dashed line (Table 3). An MPO-DNA cut-off level at ≤ 0.047 a.u. was established to differentiate between healthy control and vasculitis patients. We also tested MPO-DNA serum level, erythrocyte sedimentation rate, C-reactive protein correlation with their Birmingham Vasculitis Activity Score. The MPO-DNA showed a moderate positive correlation [ r (n = 12) = 0.596, p = 0.025]; but the C-reactive protein [ r(n = 10) = 0.402, p = 0.196] and the erythrocyte sedimentation rate [ r(n = 9) = 0.119, p = 0.727] showed no correlation with clinical vasculitis severity. These results suggest that MPO-DNA levels may also be useful in monitoring vasculitis activity compared to standard inflammation biomarkers (Table 4).Table 3 MPO-DNA values were significantly higher in the systemic vasculitis group [n = 8; 0.092 ± 0.071 arbitrary units (A.U.)] relative to healthy donors [ n = 17; 0.015 ± 0.018 A U., p = 0.01 ]. Youden J index representing the maximum potential effectiveness of the upper-lower limit of the MPO-DNA cut-off value, which was set at 0.047 (A.U.), as indicated by the red dashed line. Table 3 Table 4 The MPO-DNA serum levels, erythrocyte sedimentation rates, and C-reactive protein correlation with the Birmingham Vasculitis Activity Score for the eight vasculitis patients, the MPO-DNA showed a moderate positive correlation but the C-reactive protein and erythrocyte sedimentation rate did not. Table 4 3.2 MPO DNA performance in the trastuzumab ILD case During follow-up of the Case presenting trastuzumab-induced ILD, we found that despite standard vasculitis anti-PR3 is capable of showing good correlation during initial glucocorticoid therapy with the Birmingham vasculitis score (BVAS). The correlation was lost after salvage treatment with the B-cell targeting agent rituximab was initiated (Table 2). MPO-DNA levels exhibited a better correlation with the clinical vasculitis activity BVAS score during the rituximab treatment period. 4 Discussion Interstitial lung disease (ILD) induced by HER2-targeted agents is a well-known adverse drug reaction, but the mechanism is ill-defined and considered low in incidence [17]. HER-2 targeted agents such as trastuzumab, trastuzumab emtansine had been reported to induce cutaneous vasculitis in the previous literature [[18], [19], [20]], but pulmonary vasculitis due to HER2 targeted agents had not been reported yet. A recent review of 9886 patients investigating anti-HER2 therapies for HER2 breast cancer had reported the overall incidence of ILD was 2.4%. The incidence of grade 1–2, grade 3–4, and grade 5 events were 66.7%, 23.0%, and 0.2% respectively. The agents leading to the highest ILD incidence was trastuzumab combined with everolimus or paclitaxel. The incidence of ILD-related deaths was highest among patients receiving trastuzumab deruxtecan (T-Dxd), with an incidence of around 2% [17]. Recently the oncology society has drawn attention to this adverse event due to the recent accelerated approval of T-Dxd [9,21]. The novel antibody-drug conjugate had shown a 60% response rate in third or later line metastatic HER-2 positive breast cancer but at the cost of treatment-related ILD up to 13% and 2% treatment-related death. This high rate of side effects will inhibit its potential to be utilized as the frontline therapy or even be placed in the curative adjuvant or neoadjuvant setting. A recently published article by Kumagai et al. had reported successful induction of T-Dxd interstitial lung disease in a cynomolgus monkey model. Receiving T-Dxd in a monkey model developed interstitial lung disease, whereas receiving Dxd does not. Although most ILD lesions were found within the alveoli, the HER2 expression in lungs was limited to the bronchial level [22]. Vasculitis due to a pathway via ANCA autoantibodies, an indirect mechanism by the immune system, could explain why the injury was not at the location where T-Dxd was uptaken. Moreover, a two-step mechanism of pathogenesis explains how DIILD (drug-induced ILD) manifests later after anti-HER2 therapies. A literature review was performed to summarize previously reported cases of trastuzumab-related ILD. A total of 8 cases were identified before our Case [[23], [24], [25], [26], [27], [28]]. Two of the reported cases had available pathology that showed organizing pneumonia or diffuse alveolar damage. No traceable serum marker was reported in previously reported patients. All previously reported cases that we identified had late onset with a median time of onset two months after the first trastuzumab exposure (Table 5). The pattern of CT infiltration largely comprised ground-glass opacity and patch consolidations, with only one case exhibiting nodular lesions. Most patients recovered following prednisolone-based monotherapy, and only one patient died. Long-term follow up was not conducted in these cases. Our case also had a protracted course of vasculitis up to two years. We found that both ANCA auto-antibody testing and NET related biomarker, MPO-DNA should be tested concomitantly and especially when B cell depleting agents (such as rituximab use in our case) inhibit the autoantibody production, but vasculitis-related neutrophil activity is still high. Reliance on ANCA autoantibodies alone to monitor vasculitis activity may not be sufficient.Table 5 Case comparison of previously reported trastuzumab-related interstitial lung disease [[12], [13], [14], [15], [16], [17]]. Table 5Case Age Gender Previous CRT (setting) Tmab exposure CT pattern Pathology Treatment Recovery Naranjo Score 1 49 F [23] AC, RT, D (adjuvant) 3 months Subpleural consolidation Organizing pneumonia n/aa Yes 6 2 56 F [25] FAC, D (salvage) 4 months Consolidation and pleural effusion BAL: eosinophil 18% Prednisolone (40 mg/d) Yes 8 3 51 F [26] P (neoadjuvant) 10 weeks Ground glass opacity TBLB: Intra-alveolar hemorrhage, interstitial inflammation Prednisolone (40 mg/d) Yes 6 4 63 F [24] D (neoadjuvant) 5 weeks Peripheral consolidation, GGO Diffuse alveolar damage Prednisonec No (mortality) 7 5 67 F [35] DC (adjuvant) 4 months GGO n/a Prednisone Cyclophosphamideb Yes 8 6 68 F [28] EC (adjuvant) 3 months Patch infiltration, GGO n/a Semi-pulse steroidc Yes 6 7 62 F [28] DC (adjuvant) 3 months Patchy consolidation, poorly defined nodules n/a Prednisolonec Yes 7 8 71 F [36] Salvage Tmab followed by T-DM1 Tmab 12 wks T-DM1 6 wks Diffuse interstitial infiltrates n/a Methylprednisolone (1 mg/kg/d) Yes 5 9 59 F P (adjuvant) 3 months Patchy consolidation ANCA capillaritis, acute interstitial pneumonia Prednisolone (40 mg/d) Yes 7 CRT, chemoradiotherapy; Tmab, trastuzumab; T-DM1, trastuzumab emtansine; GGO, ground glass opacity; A, doxorubicin; C, cyclophosphamide; RT, radiotherapy; F, 5-fluorouracil; P, paclitaxel; D, docetaxel; BAL, bronchoalveolar lavage; TBLB, transbronchial lung biopsy. a Resolved spontaneously after trastuzumab discontinuation. b Cyclophosphamide was used upon trastuzumab rechallenge. c The dose used was not reported. Systemic glucocorticoids combined with either rituximab or cyclophosphamide are recommended for induction therapy for life-threatening AAV [29]. Autoreactive B cell depletion by rituximab can effectively reduce disease activity and decrease pathogenic ANCA concentrations in AAV patients. Although adverse events after rituximab include hepatitis B reactivation [30] and progressive multifocal leukoencephalopathy [31], the safety profile of rituximab is still considered to be favorable. Further salvage therapy or investigational agents included belimumab [32], a monoclonal antibody against soluble B cell-activating factor (BAFF) that can induce B cell apoptosis and showed clinical activity as an add-on therapy. The use of therapeutic plasma exchange is also beneficial for carefully selected patients experiencing severe diffuse alveolar hemorrhage or a serum creatinine level of ≥500 mmol/L [33]. There are still two crucial pieces of information that cannot be determined from our Case. First, neither serum levels of anti-PR3/anti-MPO level prior to trastuzumab exposure were available for this case. A prospective examination of vasculitis markers may be considered in the future to determine whether ANCA has clinical utility in predicting side effects during HER2 targeted therapy, especially antibody or antibody drug-conjugates. Secondly, the majority of drug-induced AAV cases present with anti-MPO positivity [34], but our case presented with anti-PR3 positivity without any upper airway symptoms or granuloma formation. Whether this difference is related to the medication administered or due to individual patient differences cannot be explained at present. In conclusion, this Case is the first clinical evidence of HER-2 targeted therapy induces pulmonary ILD via ANCA autoantibodies. Early suspicion with testing of ANCA level can be life-saving when a biopsy cannot be immediately obtained. Pre-treatment ANCA level testing may also be considered in the future when treating patients with HER-2 agents with higher risk profiles of ILD. In addition, serum MPO-DNA, could be a biomarker with promising clinical implication, as the results from this case and a prospective vasculitis cohort in our facility indicated. Authors’ contribution C.H.C., Y.M.H., and Y.M.K., wrote the original draft writing with input from all authors; Y.M.K., C.J.J., designed, developed and conducted the experiments; Y.M.K., L.C., are the primary physicians responsible for the patient medical treatment; Y.L.C., reviews the surgical pathology and issues the report.; S.C.H., Y.M.H., C.H.L., contributed to the final version of the manuscript; All authors actively participated in the discussion and suggestion for the manuscript. Funding This article was subsidized for English editing by 10.13039/501100006477 National Taiwan University under the Excellence Improvement Program for Doctoral Students (grant number 108-2926-I-002-002-MY4), sponsored by the 10.13039/501100004663 Ministry of Science and Technology, Taiwan . This work was supported by Ministry of Science and Technology (MOST 108-2314-B-002-097-MY3, 107-2314-B-002-253-, 106-2314-B-002-100-), 10.13039/501100005762 National Taiwan University Hospital (NTUH 110-M4972) and Taiwan Rheumatology Association Research Fund (TRARF-2018). Ethics approval This study was approved by the hospital Institutional Review Board (NTUH: 201612147RINA). The informed consent was obtained from the patients. Availability of data and material The author(s) confirm that the data supporting the findings of this study are available within the article. Code availability The author(s) confirm that the data supporting the findings of this study are available within the article. Declaration of competing interest The author(s) had declared that no conflicts of interest exist. Appendix A Supplementary data The following is the Supplementary data to this article:Multimedia component 1 Multimedia component 1 Appendix A Supplementary data to this article can be found online at https://doi.org/10.1016/j.breast.2021.11.016.

PACLITAXEL, TRASTUZUMAB

DrugsGivenReaction

CC BY-NC-ND

34894465

2021-11-29

What was the dosage of drug 'PACLITAXEL'?

The first reported case of trastuzumab induced interstitial lung disease associated with anti-neutrophil cytoplasmic antibody vasculitis - A case report and a prospective cohort study on the usefulness of neutrophil derived biomarkers in monitoring vasculitis disease activity during follow-up. Targeted therapies against human epidermal growth factor receptor 2 (HER2) are associated with increased interstitial lung disease (ILD). Trastuzumab, lapatinib, pertuzumab, and trastuzumab emtansine have markedly extended HER2 breast cancer survival but current knowledge on how these HER2-targeted agents induce interstitial lung disease is still poorly defined due to limited cases in the literature. Physicians mostly managed this complication by dose interruption, dose de-escalation, or discontinuation with success. In 2019, the FDA had granted accelerated approval on trastuzumab deruxtecan (T-Dxd) in HER2 breast cancer in the late line setting. Severe ILD incidence rate was over ten percent and led to fatal outcomes in 2.2% of patients in the T-Dxd trial. Searching for biomarkers to detect ILD incidence before it becomes clinically fulminant or for treatment response monitoring is of high clinical value. A Case of life-threatening trastuzumab-induced ILD was encountered in our facility. The ILD was confirmed to be antineutrophil cytoplasmic antibody (ANCA) pulmonary capillaritis. The biomarker of neutrophil extracellular traps (NETs), serum MPO-DNA complex, showed a good correlation with the clinical severity. Soon after B cell depleting agent rituximab usage, the serum MPO-DNA outperformed ANCA autoantibody and maintained its correlation with clinical severity. In addition to the trastuzumab-induced ILD case, a prospective cohort in our facility also confirmed the usefulness of MPO-DNA in monitoring vasculitis activity. We postulated that upfront testing with biomarkers of vasculitis during HER2 targeted treatment with high ILD incidence may be beneficial in the future. pmc1 Introduction The HER-2 gene is a poor prognostic factor in breast cancer and is amplified in 20–25% of patients [1]. Trastuzumab is a humanized monoclonal antibody that targets the extracellular domain of the human epidermal growth factor receptor 2 (HER2). Since its first approval in 1998, subsequently developed HER2 agents such as lapatinib, pertuzumab, and trastuzumab emtansine have markedly extended the HER2 breast cancer patient survival [[2], [3], [4]]. Drug-induced interstitial lung disease (DIILD) accounts for 3–5% of interstitial lung disease. Cancer therapy is the leading cause of DIILD and accounts for 23–51% of the cases [5]. Although the incidence of trastuzumab-related ILD in the registration trials was low at 0.5%, the mortality rate with treatment-related ILD was around 20% [[6], [7], [8]]. In 2019 the novel HER-2 antibody-drug conjugate trastuzumab deruxtecan (T-Dxd) had recently drawn attention for an increased ILD incidence while being approved to treat late-stage HER-2 positive metastatic breast cancer. T-Dxd therapy achieved a response rate up to 60%, but at the cost of ILD up to 13.6%, and 2.2% treatment-related death [9]. Therefore an unmet medical need for developing effective methods in predicting or monitoring HER-2 targeted agents induced ILD is required. 1.1 Case history A 59-year-old female was diagnosed with invasive ductal carcinoma after undergoing a core needle biopsy for a right breast lump. She received a simple mastectomy, and the surgical pathology showed stage IIA, pT2N0, HER-2/neu 3+, weak positive estrogen receptor (2%), and was negative for progesterone receptor. Adjuvant therapy comprising 12 weeks of weekly paclitaxel plus 52 weeks of tri-weekly trastuzumab was initiated three weeks after the surgery. However, at week 16, the treatment course was interrupted due to an ER visit for blood-tinged sputum over five days. A chest X-ray showed bilateral centrilobular pulmonary infiltration (Fig. 1A). The complete blood count showed neutrophilic leukocytosis (WBC 11.01 k/μL, neutrophil 74.2%, lymphocyte 17.6%, eosinophil 2.5%), anemia and thrombocytopenia (hemoglobin 9.8 g/dL, platelet 126 k/μL). The coagulation profile, as well as liver and kidney biochemistry, were within normal limits. She was admitted for empirical community-acquired pneumonia treatment with ceftriaxone and azithromycin. However, on the third day, the patient's chest X-ray showed rapid infiltration progression (Fig. 1B), and she had a fever that spiked to 39 °C with hypoxia requiring oxygen mask support. A chest CT with contrast showed diffuse peribronchovascular consolidation (Fig. 1C–D). The culture and serology on multiple pneumonia pathogens, including S. pneumonia, Legionella, Chlamydia, Influenza, Aspergillus, Cryptococcus, and CMV virus, showed negative results. Cultures for blood and sputum were also negative. Non-infectious pneumonitis was suspected, and a connective tissue disease survey was conducted. However, hypoxic respiratory failure progressed, and the patient received endotracheal intubation and was transferred to the intensive care unit. The patient underwent a bronchoscopy examination that showed diffuse mucosal bleeding of the entire airway (Fig. 2A and B). Bronchial wash cytology showed neutrophil predominance without malignant cells. Bronchial wash specimens showed no microorganisms, and the cultures remained negative. However, the connective tissue disease survey revealed antineutrophil cytoplasmic antibody (ANCA) positivity both via an indirect immunofluorescence method (IMMCO Diagnostics) and captured enzyme-linked immunoassays (EliA). The survey showed positive neutrophil cytoplasmic staining (cANCA, cytoplasmic ANCA) and an anti-PR3 titer of 40.28 IU/mL (normal <2.0). The antinuclear antibody, anti-dsDNA, rheumatoid factor, C3, C4, anti-GBM, and anti-SSA/SSB were negative.Fig. 1 A-D. (A) Chest radiograph (CXR) at initial presentation revealed multi-lobar consolidation with a centi-hilar pattern. (B) CXR on the third day showed rapid bilateral pulmonary infiltrate progression (C–D) Computed tomography (CT) on the third day showed a diffuse, multi-lobar consolidation with a peri-bronchovascular pattern. Fig. 1 Fig. 2 A-B Bronchoscopy examination of the tracheal (A) and main carina (B) revealed fresh blood in the whole airway without active vascular bleeding. Fig. 2 A lung biopsy via video-assisted thoracoscopic surgery (VATS) was performed for tissue diagnosis. Microscopically, the resected lung tissue showed marked interstitial fibrosis and residual airspaces of varying size (Fig. 3A). The remaining airspaces revealed extensive hemorrhage, as demonstrated by the presence of erythrocytes and histiocytes mixed with fibrinous exudate in the alveolar spaces. The presence of necrotic neutrophils in the alveolar septa and airspaces were indicative of acute capillaritis (Fig. 3B). A diagnosis of acute capillaritis associated with diffuse alveolar damage of the organizing stage was made (Fig. 3C). Treatment with intravenous methylprednisolone (1 mg/kg/day) was administered. The fever resolved, and the chest X-ray showed improvement on the seventh day. The patient was extubated after 14 days of ventilator support.Fig. 3 A Lung, left upper lobe, wedge resection, low power field (100X), interstitial fibrosis with collagen deposits (C) and alveolar airspace obliteration (A). B High power field showing alveolar hemorrhage (A), and necrotic neutrophils (P) in a capillary wall (V) and alveoli. C Comparison of the patient sample to a sample from a healthy individual shows marked airspace obliteration due to hemorrhage and interstitial thickening. Fig. 3 Trastuzumab-related ANCA capillaritis was diagnosed after reviewing the patient's history. She had no pulmonary symptoms or chest X-ray abnormality in previous health check-ups. The Naranjo algorithm, an adverse drug reaction (ADE) assessment tool having a scale of 1–10 [10], was applied with trastuzumab scoring 7 out of 10 (Table 1). Probable trastuzumab-related ADE was suspected. Considering this ADE as a life-threatening event, the trastuzumab schedule was permanently discontinued and replaced with hormonal adjuvant therapy. She also received rheumatologist follow up since this event at the outpatient clinic.Table 1 Probable Trastuzumab related adverse drug reaction, with Naranjo score of 7 according to the patients clinical history. Table 1No Question Yes No Do Not Know I Are there previous conclusive reports on this reaction? +1 0 0 II Did the adverse event appear after the suspected drug was administered? +2 −1 0 III Did the adverse reaction improve when the drug was discontinued or a specific antagonist was administered? +1 0 0 VI Did the adverse reaction reappear when the drug was administered? +2 −1 0 V Are there alternative causes (other than the drug) that could on their own have caused the reaction? −1 +2 0 VI Did the reaction reappear when a placebo was given? −1 +1 0 VII Was the drug detected in the blood (or other fluids) in concentration known to be toxic? +1 0 0 VIII Was the reaction more severe when the dose was increased or less severe when the dose was decreased? +1 0 0 IX Did the patient have a similar reaction to the same or similar drugs in any previous exposure? +1 0 0 X Was the adverse event confirmed by any objective evidence? +1 0 0 Naranjo score for estimating the probability of adverse drug reactions; 0, doubtful ADR; 1–4 possible ADR; 5–8, probable ADR; ≥9 definite ADR. 1.2 Utilization of different vasculitis markers in the trastuzumab ILD case Systemic methylprednisolone had induced rapid remission of the chest X-ray infiltration and inflammatory marker (e.g. C-reactive protein, erythrocyte sedimentation rate, D-dimer) improvement during the initial induction period (Table 2). However, the patient still manifested with residual vasculitis disease activity of leg weakness, numbness, and also persistent microscopic hematuria and proteinuria. Occasional expectoration of blood clot content was also noted despite no evident change in the chest X-ray appearance. The measurement of Birmingham Vasculitis Activity Score (BVAS) [11], a tool of vasculitis activity measurement consisted of 9 organs, and a score of 0–33 for persistent symptoms, and 0 to 63 for new or worse symptoms, still scored 10–20 for persistent symptoms in the patient. The patient also encountered a rapid increase in the anti-PR3 titer eight months after discontinuation of trastuzumab (Table 2). Recent evidence has supported neutrophil as the dominant infiltrate within vasculitis lesions [[12], [13], [14]]. And the discovery of neutrophil extracellular traps (NETs), a component of cell-free DNA, histone, proteinase 3 (PR3), and myeloperoxidase (MPO) released by ANCA-stimulated neutrophil, could induce vasculitis damage such as thrombus formation and endothelial damage. The patient was enrolled in a prospective vasculitis cohort in our facility for testing on vasculitis neutrophil-derived NETs and had also started rituximab (500 mg D1/D15, every 6 months) therapy for better control for the DIILD.Table 2 Trend of inflammatory markers since interstitial lung disease event onset; arrowheads indicate rituximab treatment. MPO-DNA showed a good correlation with Birmingham Vasculitis Activity Score (BVAS), especially in the late phase (r = 0.996, p = 0.004) after Rituximab treatment. This AAV patient had a protracted disease course that involved BVAS score rebound despite the PR3 autoantibody titer remaining below the upper limit. The results indicate that the MPO-DNA complex, which is a novel biomarker of neutrophil activation, could better monitor vasculitis activity. Table 2 2 Prospective cohort on neutrophil derived biomarkers in vasculitis disease activity 2.1 Materials and methods From 2017 to 2021, a prospective cohort study addressing neutrophil related biomarkers in the evaluation of patients with vasculitis was initiated in our facility (Suppl. Fig. S1). The aim was to explore the relationship between levels of cell-free MPO-DNA, a biomarker for NETosis and the clinical activity of systemic vasculitis. Including the Case presented in the manuscript, a total of eight patients with vasculitis and 17 healthy controls (a ratio of 1:2) were enrolled, and serial serum testing was obtained (Suppl. Fig. S2). The study was approved by the hospital Institutional Review Board in 2017 (NTUH: 201612147RINA). Informed consents were obtained from all patients and healthy donors. 2.2 Measurement of MPO-DNA and ANCA We tested each vasculitis individual for the cell-free MPO-DNA via a “sandwich” ELISA with an anti-MPO polyclonal antibody (GeneTex, GTX22088, Irvine, Ca, USA). ELISA microplates were coated with the MPO monoclonal antibody overnight to capture MPO-associated DNAs. Anti-ds DNA-specific monoclonal antibodies (Abcam, Cambridge, ab27156, UK; 1:2000) were added to bind MPO-associated DNA, followed by binding of a horseradish peroxidase-conjugated anti-mouse IgG antibody (Jackson ImmunoResearch, 115-035-003) for detection. A peroxidase substrate (3,3′,5,5′-Tetramethylbenzidine) was added to react with the conjugated peroxidase to yield a blue product. The reaction was halted by adding 2 N H2SO4, and the absorbance of the final product was measured at 450 nm and was transformed to an arbitrary unit (a.u.) according to the previous study [15]. An MPO-DNA assay standard curve is provided (Suppl. Fig. S3). ANCA assessment was performed using anti-proteinase (PR3) and anti-myeloperoxidase(MPO) specific immunoassays (EliA; Thermo Fisher Scientific, Waltham, MA, USA). The assay was performed according to the manufacturer's instructions. An antibody concentration was considered positive if: MPO >5.0 IU/mL and PR3 >3.0 IU/mL (normal range were provided by the manufacturer). 2.3 Statistical analysis We utilized Stata statistical software (V. 14.0, Stata Corporation, College Station, TX) for the analysis. The MPO-DNA cut-off point between healthy controls and vasculitis patients was obtained according to the maximum Youden index (Sensitivity + Specificity - 1) to capture the performance of this dichotomous diagnostic test [16]. Statistical significance of the difference between two sets of continuous variables was analyzed using a Mann-Whitney U test. A p-value less than 0.05 was defined as statistically significant. 3 Results 3.1 MPO DNA performance in the prospective vasculitis cohort The MPO-DNA values were significantly higher in the systemic vasculitis group (n = 8; 0.092 ± 0.071 arbitrary unit (a.u.)) relative to healthy donors (n = 17; 0.015 ± 0.018 a.u., p = 0.01) (Table 3). The Youden J index representing the maximum potential effectiveness of the upper-lower limit of the MPO-DNA cut-off value was 0.047 (a.u.), the red dashed line (Table 3). An MPO-DNA cut-off level at ≤ 0.047 a.u. was established to differentiate between healthy control and vasculitis patients. We also tested MPO-DNA serum level, erythrocyte sedimentation rate, C-reactive protein correlation with their Birmingham Vasculitis Activity Score. The MPO-DNA showed a moderate positive correlation [ r (n = 12) = 0.596, p = 0.025]; but the C-reactive protein [ r(n = 10) = 0.402, p = 0.196] and the erythrocyte sedimentation rate [ r(n = 9) = 0.119, p = 0.727] showed no correlation with clinical vasculitis severity. These results suggest that MPO-DNA levels may also be useful in monitoring vasculitis activity compared to standard inflammation biomarkers (Table 4).Table 3 MPO-DNA values were significantly higher in the systemic vasculitis group [n = 8; 0.092 ± 0.071 arbitrary units (A.U.)] relative to healthy donors [ n = 17; 0.015 ± 0.018 A U., p = 0.01 ]. Youden J index representing the maximum potential effectiveness of the upper-lower limit of the MPO-DNA cut-off value, which was set at 0.047 (A.U.), as indicated by the red dashed line. Table 3 Table 4 The MPO-DNA serum levels, erythrocyte sedimentation rates, and C-reactive protein correlation with the Birmingham Vasculitis Activity Score for the eight vasculitis patients, the MPO-DNA showed a moderate positive correlation but the C-reactive protein and erythrocyte sedimentation rate did not. Table 4 3.2 MPO DNA performance in the trastuzumab ILD case During follow-up of the Case presenting trastuzumab-induced ILD, we found that despite standard vasculitis anti-PR3 is capable of showing good correlation during initial glucocorticoid therapy with the Birmingham vasculitis score (BVAS). The correlation was lost after salvage treatment with the B-cell targeting agent rituximab was initiated (Table 2). MPO-DNA levels exhibited a better correlation with the clinical vasculitis activity BVAS score during the rituximab treatment period. 4 Discussion Interstitial lung disease (ILD) induced by HER2-targeted agents is a well-known adverse drug reaction, but the mechanism is ill-defined and considered low in incidence [17]. HER-2 targeted agents such as trastuzumab, trastuzumab emtansine had been reported to induce cutaneous vasculitis in the previous literature [[18], [19], [20]], but pulmonary vasculitis due to HER2 targeted agents had not been reported yet. A recent review of 9886 patients investigating anti-HER2 therapies for HER2 breast cancer had reported the overall incidence of ILD was 2.4%. The incidence of grade 1–2, grade 3–4, and grade 5 events were 66.7%, 23.0%, and 0.2% respectively. The agents leading to the highest ILD incidence was trastuzumab combined with everolimus or paclitaxel. The incidence of ILD-related deaths was highest among patients receiving trastuzumab deruxtecan (T-Dxd), with an incidence of around 2% [17]. Recently the oncology society has drawn attention to this adverse event due to the recent accelerated approval of T-Dxd [9,21]. The novel antibody-drug conjugate had shown a 60% response rate in third or later line metastatic HER-2 positive breast cancer but at the cost of treatment-related ILD up to 13% and 2% treatment-related death. This high rate of side effects will inhibit its potential to be utilized as the frontline therapy or even be placed in the curative adjuvant or neoadjuvant setting. A recently published article by Kumagai et al. had reported successful induction of T-Dxd interstitial lung disease in a cynomolgus monkey model. Receiving T-Dxd in a monkey model developed interstitial lung disease, whereas receiving Dxd does not. Although most ILD lesions were found within the alveoli, the HER2 expression in lungs was limited to the bronchial level [22]. Vasculitis due to a pathway via ANCA autoantibodies, an indirect mechanism by the immune system, could explain why the injury was not at the location where T-Dxd was uptaken. Moreover, a two-step mechanism of pathogenesis explains how DIILD (drug-induced ILD) manifests later after anti-HER2 therapies. A literature review was performed to summarize previously reported cases of trastuzumab-related ILD. A total of 8 cases were identified before our Case [[23], [24], [25], [26], [27], [28]]. Two of the reported cases had available pathology that showed organizing pneumonia or diffuse alveolar damage. No traceable serum marker was reported in previously reported patients. All previously reported cases that we identified had late onset with a median time of onset two months after the first trastuzumab exposure (Table 5). The pattern of CT infiltration largely comprised ground-glass opacity and patch consolidations, with only one case exhibiting nodular lesions. Most patients recovered following prednisolone-based monotherapy, and only one patient died. Long-term follow up was not conducted in these cases. Our case also had a protracted course of vasculitis up to two years. We found that both ANCA auto-antibody testing and NET related biomarker, MPO-DNA should be tested concomitantly and especially when B cell depleting agents (such as rituximab use in our case) inhibit the autoantibody production, but vasculitis-related neutrophil activity is still high. Reliance on ANCA autoantibodies alone to monitor vasculitis activity may not be sufficient.Table 5 Case comparison of previously reported trastuzumab-related interstitial lung disease [[12], [13], [14], [15], [16], [17]]. Table 5Case Age Gender Previous CRT (setting) Tmab exposure CT pattern Pathology Treatment Recovery Naranjo Score 1 49 F [23] AC, RT, D (adjuvant) 3 months Subpleural consolidation Organizing pneumonia n/aa Yes 6 2 56 F [25] FAC, D (salvage) 4 months Consolidation and pleural effusion BAL: eosinophil 18% Prednisolone (40 mg/d) Yes 8 3 51 F [26] P (neoadjuvant) 10 weeks Ground glass opacity TBLB: Intra-alveolar hemorrhage, interstitial inflammation Prednisolone (40 mg/d) Yes 6 4 63 F [24] D (neoadjuvant) 5 weeks Peripheral consolidation, GGO Diffuse alveolar damage Prednisonec No (mortality) 7 5 67 F [35] DC (adjuvant) 4 months GGO n/a Prednisone Cyclophosphamideb Yes 8 6 68 F [28] EC (adjuvant) 3 months Patch infiltration, GGO n/a Semi-pulse steroidc Yes 6 7 62 F [28] DC (adjuvant) 3 months Patchy consolidation, poorly defined nodules n/a Prednisolonec Yes 7 8 71 F [36] Salvage Tmab followed by T-DM1 Tmab 12 wks T-DM1 6 wks Diffuse interstitial infiltrates n/a Methylprednisolone (1 mg/kg/d) Yes 5 9 59 F P (adjuvant) 3 months Patchy consolidation ANCA capillaritis, acute interstitial pneumonia Prednisolone (40 mg/d) Yes 7 CRT, chemoradiotherapy; Tmab, trastuzumab; T-DM1, trastuzumab emtansine; GGO, ground glass opacity; A, doxorubicin; C, cyclophosphamide; RT, radiotherapy; F, 5-fluorouracil; P, paclitaxel; D, docetaxel; BAL, bronchoalveolar lavage; TBLB, transbronchial lung biopsy. a Resolved spontaneously after trastuzumab discontinuation. b Cyclophosphamide was used upon trastuzumab rechallenge. c The dose used was not reported. Systemic glucocorticoids combined with either rituximab or cyclophosphamide are recommended for induction therapy for life-threatening AAV [29]. Autoreactive B cell depletion by rituximab can effectively reduce disease activity and decrease pathogenic ANCA concentrations in AAV patients. Although adverse events after rituximab include hepatitis B reactivation [30] and progressive multifocal leukoencephalopathy [31], the safety profile of rituximab is still considered to be favorable. Further salvage therapy or investigational agents included belimumab [32], a monoclonal antibody against soluble B cell-activating factor (BAFF) that can induce B cell apoptosis and showed clinical activity as an add-on therapy. The use of therapeutic plasma exchange is also beneficial for carefully selected patients experiencing severe diffuse alveolar hemorrhage or a serum creatinine level of ≥500 mmol/L [33]. There are still two crucial pieces of information that cannot be determined from our Case. First, neither serum levels of anti-PR3/anti-MPO level prior to trastuzumab exposure were available for this case. A prospective examination of vasculitis markers may be considered in the future to determine whether ANCA has clinical utility in predicting side effects during HER2 targeted therapy, especially antibody or antibody drug-conjugates. Secondly, the majority of drug-induced AAV cases present with anti-MPO positivity [34], but our case presented with anti-PR3 positivity without any upper airway symptoms or granuloma formation. Whether this difference is related to the medication administered or due to individual patient differences cannot be explained at present. In conclusion, this Case is the first clinical evidence of HER-2 targeted therapy induces pulmonary ILD via ANCA autoantibodies. Early suspicion with testing of ANCA level can be life-saving when a biopsy cannot be immediately obtained. Pre-treatment ANCA level testing may also be considered in the future when treating patients with HER-2 agents with higher risk profiles of ILD. In addition, serum MPO-DNA, could be a biomarker with promising clinical implication, as the results from this case and a prospective vasculitis cohort in our facility indicated. Authors’ contribution C.H.C., Y.M.H., and Y.M.K., wrote the original draft writing with input from all authors; Y.M.K., C.J.J., designed, developed and conducted the experiments; Y.M.K., L.C., are the primary physicians responsible for the patient medical treatment; Y.L.C., reviews the surgical pathology and issues the report.; S.C.H., Y.M.H., C.H.L., contributed to the final version of the manuscript; All authors actively participated in the discussion and suggestion for the manuscript. Funding This article was subsidized for English editing by 10.13039/501100006477 National Taiwan University under the Excellence Improvement Program for Doctoral Students (grant number 108-2926-I-002-002-MY4), sponsored by the 10.13039/501100004663 Ministry of Science and Technology, Taiwan . This work was supported by Ministry of Science and Technology (MOST 108-2314-B-002-097-MY3, 107-2314-B-002-253-, 106-2314-B-002-100-), 10.13039/501100005762 National Taiwan University Hospital (NTUH 110-M4972) and Taiwan Rheumatology Association Research Fund (TRARF-2018). Ethics approval This study was approved by the hospital Institutional Review Board (NTUH: 201612147RINA). The informed consent was obtained from the patients. Availability of data and material The author(s) confirm that the data supporting the findings of this study are available within the article. Code availability The author(s) confirm that the data supporting the findings of this study are available within the article. Declaration of competing interest The author(s) had declared that no conflicts of interest exist. Appendix A Supplementary data The following is the Supplementary data to this article:Multimedia component 1 Multimedia component 1 Appendix A Supplementary data to this article can be found online at https://doi.org/10.1016/j.breast.2021.11.016.

QW (SCHEDULED FOR 12 WEEKS OF PACLITAXEL WEEKLY)

DrugDosageText

CC BY-NC-ND

34894465

2021-11-29

What was the outcome of reaction 'Interstitial lung disease'?

The first reported case of trastuzumab induced interstitial lung disease associated with anti-neutrophil cytoplasmic antibody vasculitis - A case report and a prospective cohort study on the usefulness of neutrophil derived biomarkers in monitoring vasculitis disease activity during follow-up. Targeted therapies against human epidermal growth factor receptor 2 (HER2) are associated with increased interstitial lung disease (ILD). Trastuzumab, lapatinib, pertuzumab, and trastuzumab emtansine have markedly extended HER2 breast cancer survival but current knowledge on how these HER2-targeted agents induce interstitial lung disease is still poorly defined due to limited cases in the literature. Physicians mostly managed this complication by dose interruption, dose de-escalation, or discontinuation with success. In 2019, the FDA had granted accelerated approval on trastuzumab deruxtecan (T-Dxd) in HER2 breast cancer in the late line setting. Severe ILD incidence rate was over ten percent and led to fatal outcomes in 2.2% of patients in the T-Dxd trial. Searching for biomarkers to detect ILD incidence before it becomes clinically fulminant or for treatment response monitoring is of high clinical value. A Case of life-threatening trastuzumab-induced ILD was encountered in our facility. The ILD was confirmed to be antineutrophil cytoplasmic antibody (ANCA) pulmonary capillaritis. The biomarker of neutrophil extracellular traps (NETs), serum MPO-DNA complex, showed a good correlation with the clinical severity. Soon after B cell depleting agent rituximab usage, the serum MPO-DNA outperformed ANCA autoantibody and maintained its correlation with clinical severity. In addition to the trastuzumab-induced ILD case, a prospective cohort in our facility also confirmed the usefulness of MPO-DNA in monitoring vasculitis activity. We postulated that upfront testing with biomarkers of vasculitis during HER2 targeted treatment with high ILD incidence may be beneficial in the future. pmc1 Introduction The HER-2 gene is a poor prognostic factor in breast cancer and is amplified in 20–25% of patients [1]. Trastuzumab is a humanized monoclonal antibody that targets the extracellular domain of the human epidermal growth factor receptor 2 (HER2). Since its first approval in 1998, subsequently developed HER2 agents such as lapatinib, pertuzumab, and trastuzumab emtansine have markedly extended the HER2 breast cancer patient survival [[2], [3], [4]]. Drug-induced interstitial lung disease (DIILD) accounts for 3–5% of interstitial lung disease. Cancer therapy is the leading cause of DIILD and accounts for 23–51% of the cases [5]. Although the incidence of trastuzumab-related ILD in the registration trials was low at 0.5%, the mortality rate with treatment-related ILD was around 20% [[6], [7], [8]]. In 2019 the novel HER-2 antibody-drug conjugate trastuzumab deruxtecan (T-Dxd) had recently drawn attention for an increased ILD incidence while being approved to treat late-stage HER-2 positive metastatic breast cancer. T-Dxd therapy achieved a response rate up to 60%, but at the cost of ILD up to 13.6%, and 2.2% treatment-related death [9]. Therefore an unmet medical need for developing effective methods in predicting or monitoring HER-2 targeted agents induced ILD is required. 1.1 Case history A 59-year-old female was diagnosed with invasive ductal carcinoma after undergoing a core needle biopsy for a right breast lump. She received a simple mastectomy, and the surgical pathology showed stage IIA, pT2N0, HER-2/neu 3+, weak positive estrogen receptor (2%), and was negative for progesterone receptor. Adjuvant therapy comprising 12 weeks of weekly paclitaxel plus 52 weeks of tri-weekly trastuzumab was initiated three weeks after the surgery. However, at week 16, the treatment course was interrupted due to an ER visit for blood-tinged sputum over five days. A chest X-ray showed bilateral centrilobular pulmonary infiltration (Fig. 1A). The complete blood count showed neutrophilic leukocytosis (WBC 11.01 k/μL, neutrophil 74.2%, lymphocyte 17.6%, eosinophil 2.5%), anemia and thrombocytopenia (hemoglobin 9.8 g/dL, platelet 126 k/μL). The coagulation profile, as well as liver and kidney biochemistry, were within normal limits. She was admitted for empirical community-acquired pneumonia treatment with ceftriaxone and azithromycin. However, on the third day, the patient's chest X-ray showed rapid infiltration progression (Fig. 1B), and she had a fever that spiked to 39 °C with hypoxia requiring oxygen mask support. A chest CT with contrast showed diffuse peribronchovascular consolidation (Fig. 1C–D). The culture and serology on multiple pneumonia pathogens, including S. pneumonia, Legionella, Chlamydia, Influenza, Aspergillus, Cryptococcus, and CMV virus, showed negative results. Cultures for blood and sputum were also negative. Non-infectious pneumonitis was suspected, and a connective tissue disease survey was conducted. However, hypoxic respiratory failure progressed, and the patient received endotracheal intubation and was transferred to the intensive care unit. The patient underwent a bronchoscopy examination that showed diffuse mucosal bleeding of the entire airway (Fig. 2A and B). Bronchial wash cytology showed neutrophil predominance without malignant cells. Bronchial wash specimens showed no microorganisms, and the cultures remained negative. However, the connective tissue disease survey revealed antineutrophil cytoplasmic antibody (ANCA) positivity both via an indirect immunofluorescence method (IMMCO Diagnostics) and captured enzyme-linked immunoassays (EliA). The survey showed positive neutrophil cytoplasmic staining (cANCA, cytoplasmic ANCA) and an anti-PR3 titer of 40.28 IU/mL (normal <2.0). The antinuclear antibody, anti-dsDNA, rheumatoid factor, C3, C4, anti-GBM, and anti-SSA/SSB were negative.Fig. 1 A-D. (A) Chest radiograph (CXR) at initial presentation revealed multi-lobar consolidation with a centi-hilar pattern. (B) CXR on the third day showed rapid bilateral pulmonary infiltrate progression (C–D) Computed tomography (CT) on the third day showed a diffuse, multi-lobar consolidation with a peri-bronchovascular pattern. Fig. 1 Fig. 2 A-B Bronchoscopy examination of the tracheal (A) and main carina (B) revealed fresh blood in the whole airway without active vascular bleeding. Fig. 2 A lung biopsy via video-assisted thoracoscopic surgery (VATS) was performed for tissue diagnosis. Microscopically, the resected lung tissue showed marked interstitial fibrosis and residual airspaces of varying size (Fig. 3A). The remaining airspaces revealed extensive hemorrhage, as demonstrated by the presence of erythrocytes and histiocytes mixed with fibrinous exudate in the alveolar spaces. The presence of necrotic neutrophils in the alveolar septa and airspaces were indicative of acute capillaritis (Fig. 3B). A diagnosis of acute capillaritis associated with diffuse alveolar damage of the organizing stage was made (Fig. 3C). Treatment with intravenous methylprednisolone (1 mg/kg/day) was administered. The fever resolved, and the chest X-ray showed improvement on the seventh day. The patient was extubated after 14 days of ventilator support.Fig. 3 A Lung, left upper lobe, wedge resection, low power field (100X), interstitial fibrosis with collagen deposits (C) and alveolar airspace obliteration (A). B High power field showing alveolar hemorrhage (A), and necrotic neutrophils (P) in a capillary wall (V) and alveoli. C Comparison of the patient sample to a sample from a healthy individual shows marked airspace obliteration due to hemorrhage and interstitial thickening. Fig. 3 Trastuzumab-related ANCA capillaritis was diagnosed after reviewing the patient's history. She had no pulmonary symptoms or chest X-ray abnormality in previous health check-ups. The Naranjo algorithm, an adverse drug reaction (ADE) assessment tool having a scale of 1–10 [10], was applied with trastuzumab scoring 7 out of 10 (Table 1). Probable trastuzumab-related ADE was suspected. Considering this ADE as a life-threatening event, the trastuzumab schedule was permanently discontinued and replaced with hormonal adjuvant therapy. She also received rheumatologist follow up since this event at the outpatient clinic.Table 1 Probable Trastuzumab related adverse drug reaction, with Naranjo score of 7 according to the patients clinical history. Table 1No Question Yes No Do Not Know I Are there previous conclusive reports on this reaction? +1 0 0 II Did the adverse event appear after the suspected drug was administered? +2 −1 0 III Did the adverse reaction improve when the drug was discontinued or a specific antagonist was administered? +1 0 0 VI Did the adverse reaction reappear when the drug was administered? +2 −1 0 V Are there alternative causes (other than the drug) that could on their own have caused the reaction? −1 +2 0 VI Did the reaction reappear when a placebo was given? −1 +1 0 VII Was the drug detected in the blood (or other fluids) in concentration known to be toxic? +1 0 0 VIII Was the reaction more severe when the dose was increased or less severe when the dose was decreased? +1 0 0 IX Did the patient have a similar reaction to the same or similar drugs in any previous exposure? +1 0 0 X Was the adverse event confirmed by any objective evidence? +1 0 0 Naranjo score for estimating the probability of adverse drug reactions; 0, doubtful ADR; 1–4 possible ADR; 5–8, probable ADR; ≥9 definite ADR. 1.2 Utilization of different vasculitis markers in the trastuzumab ILD case Systemic methylprednisolone had induced rapid remission of the chest X-ray infiltration and inflammatory marker (e.g. C-reactive protein, erythrocyte sedimentation rate, D-dimer) improvement during the initial induction period (Table 2). However, the patient still manifested with residual vasculitis disease activity of leg weakness, numbness, and also persistent microscopic hematuria and proteinuria. Occasional expectoration of blood clot content was also noted despite no evident change in the chest X-ray appearance. The measurement of Birmingham Vasculitis Activity Score (BVAS) [11], a tool of vasculitis activity measurement consisted of 9 organs, and a score of 0–33 for persistent symptoms, and 0 to 63 for new or worse symptoms, still scored 10–20 for persistent symptoms in the patient. The patient also encountered a rapid increase in the anti-PR3 titer eight months after discontinuation of trastuzumab (Table 2). Recent evidence has supported neutrophil as the dominant infiltrate within vasculitis lesions [[12], [13], [14]]. And the discovery of neutrophil extracellular traps (NETs), a component of cell-free DNA, histone, proteinase 3 (PR3), and myeloperoxidase (MPO) released by ANCA-stimulated neutrophil, could induce vasculitis damage such as thrombus formation and endothelial damage. The patient was enrolled in a prospective vasculitis cohort in our facility for testing on vasculitis neutrophil-derived NETs and had also started rituximab (500 mg D1/D15, every 6 months) therapy for better control for the DIILD.Table 2 Trend of inflammatory markers since interstitial lung disease event onset; arrowheads indicate rituximab treatment. MPO-DNA showed a good correlation with Birmingham Vasculitis Activity Score (BVAS), especially in the late phase (r = 0.996, p = 0.004) after Rituximab treatment. This AAV patient had a protracted disease course that involved BVAS score rebound despite the PR3 autoantibody titer remaining below the upper limit. The results indicate that the MPO-DNA complex, which is a novel biomarker of neutrophil activation, could better monitor vasculitis activity. Table 2 2 Prospective cohort on neutrophil derived biomarkers in vasculitis disease activity 2.1 Materials and methods From 2017 to 2021, a prospective cohort study addressing neutrophil related biomarkers in the evaluation of patients with vasculitis was initiated in our facility (Suppl. Fig. S1). The aim was to explore the relationship between levels of cell-free MPO-DNA, a biomarker for NETosis and the clinical activity of systemic vasculitis. Including the Case presented in the manuscript, a total of eight patients with vasculitis and 17 healthy controls (a ratio of 1:2) were enrolled, and serial serum testing was obtained (Suppl. Fig. S2). The study was approved by the hospital Institutional Review Board in 2017 (NTUH: 201612147RINA). Informed consents were obtained from all patients and healthy donors. 2.2 Measurement of MPO-DNA and ANCA We tested each vasculitis individual for the cell-free MPO-DNA via a “sandwich” ELISA with an anti-MPO polyclonal antibody (GeneTex, GTX22088, Irvine, Ca, USA). ELISA microplates were coated with the MPO monoclonal antibody overnight to capture MPO-associated DNAs. Anti-ds DNA-specific monoclonal antibodies (Abcam, Cambridge, ab27156, UK; 1:2000) were added to bind MPO-associated DNA, followed by binding of a horseradish peroxidase-conjugated anti-mouse IgG antibody (Jackson ImmunoResearch, 115-035-003) for detection. A peroxidase substrate (3,3′,5,5′-Tetramethylbenzidine) was added to react with the conjugated peroxidase to yield a blue product. The reaction was halted by adding 2 N H2SO4, and the absorbance of the final product was measured at 450 nm and was transformed to an arbitrary unit (a.u.) according to the previous study [15]. An MPO-DNA assay standard curve is provided (Suppl. Fig. S3). ANCA assessment was performed using anti-proteinase (PR3) and anti-myeloperoxidase(MPO) specific immunoassays (EliA; Thermo Fisher Scientific, Waltham, MA, USA). The assay was performed according to the manufacturer's instructions. An antibody concentration was considered positive if: MPO >5.0 IU/mL and PR3 >3.0 IU/mL (normal range were provided by the manufacturer). 2.3 Statistical analysis We utilized Stata statistical software (V. 14.0, Stata Corporation, College Station, TX) for the analysis. The MPO-DNA cut-off point between healthy controls and vasculitis patients was obtained according to the maximum Youden index (Sensitivity + Specificity - 1) to capture the performance of this dichotomous diagnostic test [16]. Statistical significance of the difference between two sets of continuous variables was analyzed using a Mann-Whitney U test. A p-value less than 0.05 was defined as statistically significant. 3 Results 3.1 MPO DNA performance in the prospective vasculitis cohort The MPO-DNA values were significantly higher in the systemic vasculitis group (n = 8; 0.092 ± 0.071 arbitrary unit (a.u.)) relative to healthy donors (n = 17; 0.015 ± 0.018 a.u., p = 0.01) (Table 3). The Youden J index representing the maximum potential effectiveness of the upper-lower limit of the MPO-DNA cut-off value was 0.047 (a.u.), the red dashed line (Table 3). An MPO-DNA cut-off level at ≤ 0.047 a.u. was established to differentiate between healthy control and vasculitis patients. We also tested MPO-DNA serum level, erythrocyte sedimentation rate, C-reactive protein correlation with their Birmingham Vasculitis Activity Score. The MPO-DNA showed a moderate positive correlation [ r (n = 12) = 0.596, p = 0.025]; but the C-reactive protein [ r(n = 10) = 0.402, p = 0.196] and the erythrocyte sedimentation rate [ r(n = 9) = 0.119, p = 0.727] showed no correlation with clinical vasculitis severity. These results suggest that MPO-DNA levels may also be useful in monitoring vasculitis activity compared to standard inflammation biomarkers (Table 4).Table 3 MPO-DNA values were significantly higher in the systemic vasculitis group [n = 8; 0.092 ± 0.071 arbitrary units (A.U.)] relative to healthy donors [ n = 17; 0.015 ± 0.018 A U., p = 0.01 ]. Youden J index representing the maximum potential effectiveness of the upper-lower limit of the MPO-DNA cut-off value, which was set at 0.047 (A.U.), as indicated by the red dashed line. Table 3 Table 4 The MPO-DNA serum levels, erythrocyte sedimentation rates, and C-reactive protein correlation with the Birmingham Vasculitis Activity Score for the eight vasculitis patients, the MPO-DNA showed a moderate positive correlation but the C-reactive protein and erythrocyte sedimentation rate did not. Table 4 3.2 MPO DNA performance in the trastuzumab ILD case During follow-up of the Case presenting trastuzumab-induced ILD, we found that despite standard vasculitis anti-PR3 is capable of showing good correlation during initial glucocorticoid therapy with the Birmingham vasculitis score (BVAS). The correlation was lost after salvage treatment with the B-cell targeting agent rituximab was initiated (Table 2). MPO-DNA levels exhibited a better correlation with the clinical vasculitis activity BVAS score during the rituximab treatment period. 4 Discussion Interstitial lung disease (ILD) induced by HER2-targeted agents is a well-known adverse drug reaction, but the mechanism is ill-defined and considered low in incidence [17]. HER-2 targeted agents such as trastuzumab, trastuzumab emtansine had been reported to induce cutaneous vasculitis in the previous literature [[18], [19], [20]], but pulmonary vasculitis due to HER2 targeted agents had not been reported yet. A recent review of 9886 patients investigating anti-HER2 therapies for HER2 breast cancer had reported the overall incidence of ILD was 2.4%. The incidence of grade 1–2, grade 3–4, and grade 5 events were 66.7%, 23.0%, and 0.2% respectively. The agents leading to the highest ILD incidence was trastuzumab combined with everolimus or paclitaxel. The incidence of ILD-related deaths was highest among patients receiving trastuzumab deruxtecan (T-Dxd), with an incidence of around 2% [17]. Recently the oncology society has drawn attention to this adverse event due to the recent accelerated approval of T-Dxd [9,21]. The novel antibody-drug conjugate had shown a 60% response rate in third or later line metastatic HER-2 positive breast cancer but at the cost of treatment-related ILD up to 13% and 2% treatment-related death. This high rate of side effects will inhibit its potential to be utilized as the frontline therapy or even be placed in the curative adjuvant or neoadjuvant setting. A recently published article by Kumagai et al. had reported successful induction of T-Dxd interstitial lung disease in a cynomolgus monkey model. Receiving T-Dxd in a monkey model developed interstitial lung disease, whereas receiving Dxd does not. Although most ILD lesions were found within the alveoli, the HER2 expression in lungs was limited to the bronchial level [22]. Vasculitis due to a pathway via ANCA autoantibodies, an indirect mechanism by the immune system, could explain why the injury was not at the location where T-Dxd was uptaken. Moreover, a two-step mechanism of pathogenesis explains how DIILD (drug-induced ILD) manifests later after anti-HER2 therapies. A literature review was performed to summarize previously reported cases of trastuzumab-related ILD. A total of 8 cases were identified before our Case [[23], [24], [25], [26], [27], [28]]. Two of the reported cases had available pathology that showed organizing pneumonia or diffuse alveolar damage. No traceable serum marker was reported in previously reported patients. All previously reported cases that we identified had late onset with a median time of onset two months after the first trastuzumab exposure (Table 5). The pattern of CT infiltration largely comprised ground-glass opacity and patch consolidations, with only one case exhibiting nodular lesions. Most patients recovered following prednisolone-based monotherapy, and only one patient died. Long-term follow up was not conducted in these cases. Our case also had a protracted course of vasculitis up to two years. We found that both ANCA auto-antibody testing and NET related biomarker, MPO-DNA should be tested concomitantly and especially when B cell depleting agents (such as rituximab use in our case) inhibit the autoantibody production, but vasculitis-related neutrophil activity is still high. Reliance on ANCA autoantibodies alone to monitor vasculitis activity may not be sufficient.Table 5 Case comparison of previously reported trastuzumab-related interstitial lung disease [[12], [13], [14], [15], [16], [17]]. Table 5Case Age Gender Previous CRT (setting) Tmab exposure CT pattern Pathology Treatment Recovery Naranjo Score 1 49 F [23] AC, RT, D (adjuvant) 3 months Subpleural consolidation Organizing pneumonia n/aa Yes 6 2 56 F [25] FAC, D (salvage) 4 months Consolidation and pleural effusion BAL: eosinophil 18% Prednisolone (40 mg/d) Yes 8 3 51 F [26] P (neoadjuvant) 10 weeks Ground glass opacity TBLB: Intra-alveolar hemorrhage, interstitial inflammation Prednisolone (40 mg/d) Yes 6 4 63 F [24] D (neoadjuvant) 5 weeks Peripheral consolidation, GGO Diffuse alveolar damage Prednisonec No (mortality) 7 5 67 F [35] DC (adjuvant) 4 months GGO n/a Prednisone Cyclophosphamideb Yes 8 6 68 F [28] EC (adjuvant) 3 months Patch infiltration, GGO n/a Semi-pulse steroidc Yes 6 7 62 F [28] DC (adjuvant) 3 months Patchy consolidation, poorly defined nodules n/a Prednisolonec Yes 7 8 71 F [36] Salvage Tmab followed by T-DM1 Tmab 12 wks T-DM1 6 wks Diffuse interstitial infiltrates n/a Methylprednisolone (1 mg/kg/d) Yes 5 9 59 F P (adjuvant) 3 months Patchy consolidation ANCA capillaritis, acute interstitial pneumonia Prednisolone (40 mg/d) Yes 7 CRT, chemoradiotherapy; Tmab, trastuzumab; T-DM1, trastuzumab emtansine; GGO, ground glass opacity; A, doxorubicin; C, cyclophosphamide; RT, radiotherapy; F, 5-fluorouracil; P, paclitaxel; D, docetaxel; BAL, bronchoalveolar lavage; TBLB, transbronchial lung biopsy. a Resolved spontaneously after trastuzumab discontinuation. b Cyclophosphamide was used upon trastuzumab rechallenge. c The dose used was not reported. Systemic glucocorticoids combined with either rituximab or cyclophosphamide are recommended for induction therapy for life-threatening AAV [29]. Autoreactive B cell depletion by rituximab can effectively reduce disease activity and decrease pathogenic ANCA concentrations in AAV patients. Although adverse events after rituximab include hepatitis B reactivation [30] and progressive multifocal leukoencephalopathy [31], the safety profile of rituximab is still considered to be favorable. Further salvage therapy or investigational agents included belimumab [32], a monoclonal antibody against soluble B cell-activating factor (BAFF) that can induce B cell apoptosis and showed clinical activity as an add-on therapy. The use of therapeutic plasma exchange is also beneficial for carefully selected patients experiencing severe diffuse alveolar hemorrhage or a serum creatinine level of ≥500 mmol/L [33]. There are still two crucial pieces of information that cannot be determined from our Case. First, neither serum levels of anti-PR3/anti-MPO level prior to trastuzumab exposure were available for this case. A prospective examination of vasculitis markers may be considered in the future to determine whether ANCA has clinical utility in predicting side effects during HER2 targeted therapy, especially antibody or antibody drug-conjugates. Secondly, the majority of drug-induced AAV cases present with anti-MPO positivity [34], but our case presented with anti-PR3 positivity without any upper airway symptoms or granuloma formation. Whether this difference is related to the medication administered or due to individual patient differences cannot be explained at present. In conclusion, this Case is the first clinical evidence of HER-2 targeted therapy induces pulmonary ILD via ANCA autoantibodies. Early suspicion with testing of ANCA level can be life-saving when a biopsy cannot be immediately obtained. Pre-treatment ANCA level testing may also be considered in the future when treating patients with HER-2 agents with higher risk profiles of ILD. In addition, serum MPO-DNA, could be a biomarker with promising clinical implication, as the results from this case and a prospective vasculitis cohort in our facility indicated. Authors’ contribution C.H.C., Y.M.H., and Y.M.K., wrote the original draft writing with input from all authors; Y.M.K., C.J.J., designed, developed and conducted the experiments; Y.M.K., L.C., are the primary physicians responsible for the patient medical treatment; Y.L.C., reviews the surgical pathology and issues the report.; S.C.H., Y.M.H., C.H.L., contributed to the final version of the manuscript; All authors actively participated in the discussion and suggestion for the manuscript. Funding This article was subsidized for English editing by 10.13039/501100006477 National Taiwan University under the Excellence Improvement Program for Doctoral Students (grant number 108-2926-I-002-002-MY4), sponsored by the 10.13039/501100004663 Ministry of Science and Technology, Taiwan . This work was supported by Ministry of Science and Technology (MOST 108-2314-B-002-097-MY3, 107-2314-B-002-253-, 106-2314-B-002-100-), 10.13039/501100005762 National Taiwan University Hospital (NTUH 110-M4972) and Taiwan Rheumatology Association Research Fund (TRARF-2018). Ethics approval This study was approved by the hospital Institutional Review Board (NTUH: 201612147RINA). The informed consent was obtained from the patients. Availability of data and material The author(s) confirm that the data supporting the findings of this study are available within the article. Code availability The author(s) confirm that the data supporting the findings of this study are available within the article. Declaration of competing interest The author(s) had declared that no conflicts of interest exist. Appendix A Supplementary data The following is the Supplementary data to this article:Multimedia component 1 Multimedia component 1 Appendix A Supplementary data to this article can be found online at https://doi.org/10.1016/j.breast.2021.11.016.

Recovering

ReactionOutcome

CC BY-NC-ND

34894465

2021-11-29

What was the outcome of reaction 'Respiratory failure'?

The first reported case of trastuzumab induced interstitial lung disease associated with anti-neutrophil cytoplasmic antibody vasculitis - A case report and a prospective cohort study on the usefulness of neutrophil derived biomarkers in monitoring vasculitis disease activity during follow-up. Targeted therapies against human epidermal growth factor receptor 2 (HER2) are associated with increased interstitial lung disease (ILD). Trastuzumab, lapatinib, pertuzumab, and trastuzumab emtansine have markedly extended HER2 breast cancer survival but current knowledge on how these HER2-targeted agents induce interstitial lung disease is still poorly defined due to limited cases in the literature. Physicians mostly managed this complication by dose interruption, dose de-escalation, or discontinuation with success. In 2019, the FDA had granted accelerated approval on trastuzumab deruxtecan (T-Dxd) in HER2 breast cancer in the late line setting. Severe ILD incidence rate was over ten percent and led to fatal outcomes in 2.2% of patients in the T-Dxd trial. Searching for biomarkers to detect ILD incidence before it becomes clinically fulminant or for treatment response monitoring is of high clinical value. A Case of life-threatening trastuzumab-induced ILD was encountered in our facility. The ILD was confirmed to be antineutrophil cytoplasmic antibody (ANCA) pulmonary capillaritis. The biomarker of neutrophil extracellular traps (NETs), serum MPO-DNA complex, showed a good correlation with the clinical severity. Soon after B cell depleting agent rituximab usage, the serum MPO-DNA outperformed ANCA autoantibody and maintained its correlation with clinical severity. In addition to the trastuzumab-induced ILD case, a prospective cohort in our facility also confirmed the usefulness of MPO-DNA in monitoring vasculitis activity. We postulated that upfront testing with biomarkers of vasculitis during HER2 targeted treatment with high ILD incidence may be beneficial in the future. pmc1 Introduction The HER-2 gene is a poor prognostic factor in breast cancer and is amplified in 20–25% of patients [1]. Trastuzumab is a humanized monoclonal antibody that targets the extracellular domain of the human epidermal growth factor receptor 2 (HER2). Since its first approval in 1998, subsequently developed HER2 agents such as lapatinib, pertuzumab, and trastuzumab emtansine have markedly extended the HER2 breast cancer patient survival [[2], [3], [4]]. Drug-induced interstitial lung disease (DIILD) accounts for 3–5% of interstitial lung disease. Cancer therapy is the leading cause of DIILD and accounts for 23–51% of the cases [5]. Although the incidence of trastuzumab-related ILD in the registration trials was low at 0.5%, the mortality rate with treatment-related ILD was around 20% [[6], [7], [8]]. In 2019 the novel HER-2 antibody-drug conjugate trastuzumab deruxtecan (T-Dxd) had recently drawn attention for an increased ILD incidence while being approved to treat late-stage HER-2 positive metastatic breast cancer. T-Dxd therapy achieved a response rate up to 60%, but at the cost of ILD up to 13.6%, and 2.2% treatment-related death [9]. Therefore an unmet medical need for developing effective methods in predicting or monitoring HER-2 targeted agents induced ILD is required. 1.1 Case history A 59-year-old female was diagnosed with invasive ductal carcinoma after undergoing a core needle biopsy for a right breast lump. She received a simple mastectomy, and the surgical pathology showed stage IIA, pT2N0, HER-2/neu 3+, weak positive estrogen receptor (2%), and was negative for progesterone receptor. Adjuvant therapy comprising 12 weeks of weekly paclitaxel plus 52 weeks of tri-weekly trastuzumab was initiated three weeks after the surgery. However, at week 16, the treatment course was interrupted due to an ER visit for blood-tinged sputum over five days. A chest X-ray showed bilateral centrilobular pulmonary infiltration (Fig. 1A). The complete blood count showed neutrophilic leukocytosis (WBC 11.01 k/μL, neutrophil 74.2%, lymphocyte 17.6%, eosinophil 2.5%), anemia and thrombocytopenia (hemoglobin 9.8 g/dL, platelet 126 k/μL). The coagulation profile, as well as liver and kidney biochemistry, were within normal limits. She was admitted for empirical community-acquired pneumonia treatment with ceftriaxone and azithromycin. However, on the third day, the patient's chest X-ray showed rapid infiltration progression (Fig. 1B), and she had a fever that spiked to 39 °C with hypoxia requiring oxygen mask support. A chest CT with contrast showed diffuse peribronchovascular consolidation (Fig. 1C–D). The culture and serology on multiple pneumonia pathogens, including S. pneumonia, Legionella, Chlamydia, Influenza, Aspergillus, Cryptococcus, and CMV virus, showed negative results. Cultures for blood and sputum were also negative. Non-infectious pneumonitis was suspected, and a connective tissue disease survey was conducted. However, hypoxic respiratory failure progressed, and the patient received endotracheal intubation and was transferred to the intensive care unit. The patient underwent a bronchoscopy examination that showed diffuse mucosal bleeding of the entire airway (Fig. 2A and B). Bronchial wash cytology showed neutrophil predominance without malignant cells. Bronchial wash specimens showed no microorganisms, and the cultures remained negative. However, the connective tissue disease survey revealed antineutrophil cytoplasmic antibody (ANCA) positivity both via an indirect immunofluorescence method (IMMCO Diagnostics) and captured enzyme-linked immunoassays (EliA). The survey showed positive neutrophil cytoplasmic staining (cANCA, cytoplasmic ANCA) and an anti-PR3 titer of 40.28 IU/mL (normal <2.0). The antinuclear antibody, anti-dsDNA, rheumatoid factor, C3, C4, anti-GBM, and anti-SSA/SSB were negative.Fig. 1 A-D. (A) Chest radiograph (CXR) at initial presentation revealed multi-lobar consolidation with a centi-hilar pattern. (B) CXR on the third day showed rapid bilateral pulmonary infiltrate progression (C–D) Computed tomography (CT) on the third day showed a diffuse, multi-lobar consolidation with a peri-bronchovascular pattern. Fig. 1 Fig. 2 A-B Bronchoscopy examination of the tracheal (A) and main carina (B) revealed fresh blood in the whole airway without active vascular bleeding. Fig. 2 A lung biopsy via video-assisted thoracoscopic surgery (VATS) was performed for tissue diagnosis. Microscopically, the resected lung tissue showed marked interstitial fibrosis and residual airspaces of varying size (Fig. 3A). The remaining airspaces revealed extensive hemorrhage, as demonstrated by the presence of erythrocytes and histiocytes mixed with fibrinous exudate in the alveolar spaces. The presence of necrotic neutrophils in the alveolar septa and airspaces were indicative of acute capillaritis (Fig. 3B). A diagnosis of acute capillaritis associated with diffuse alveolar damage of the organizing stage was made (Fig. 3C). Treatment with intravenous methylprednisolone (1 mg/kg/day) was administered. The fever resolved, and the chest X-ray showed improvement on the seventh day. The patient was extubated after 14 days of ventilator support.Fig. 3 A Lung, left upper lobe, wedge resection, low power field (100X), interstitial fibrosis with collagen deposits (C) and alveolar airspace obliteration (A). B High power field showing alveolar hemorrhage (A), and necrotic neutrophils (P) in a capillary wall (V) and alveoli. C Comparison of the patient sample to a sample from a healthy individual shows marked airspace obliteration due to hemorrhage and interstitial thickening. Fig. 3 Trastuzumab-related ANCA capillaritis was diagnosed after reviewing the patient's history. She had no pulmonary symptoms or chest X-ray abnormality in previous health check-ups. The Naranjo algorithm, an adverse drug reaction (ADE) assessment tool having a scale of 1–10 [10], was applied with trastuzumab scoring 7 out of 10 (Table 1). Probable trastuzumab-related ADE was suspected. Considering this ADE as a life-threatening event, the trastuzumab schedule was permanently discontinued and replaced with hormonal adjuvant therapy. She also received rheumatologist follow up since this event at the outpatient clinic.Table 1 Probable Trastuzumab related adverse drug reaction, with Naranjo score of 7 according to the patients clinical history. Table 1No Question Yes No Do Not Know I Are there previous conclusive reports on this reaction? +1 0 0 II Did the adverse event appear after the suspected drug was administered? +2 −1 0 III Did the adverse reaction improve when the drug was discontinued or a specific antagonist was administered? +1 0 0 VI Did the adverse reaction reappear when the drug was administered? +2 −1 0 V Are there alternative causes (other than the drug) that could on their own have caused the reaction? −1 +2 0 VI Did the reaction reappear when a placebo was given? −1 +1 0 VII Was the drug detected in the blood (or other fluids) in concentration known to be toxic? +1 0 0 VIII Was the reaction more severe when the dose was increased or less severe when the dose was decreased? +1 0 0 IX Did the patient have a similar reaction to the same or similar drugs in any previous exposure? +1 0 0 X Was the adverse event confirmed by any objective evidence? +1 0 0 Naranjo score for estimating the probability of adverse drug reactions; 0, doubtful ADR; 1–4 possible ADR; 5–8, probable ADR; ≥9 definite ADR. 1.2 Utilization of different vasculitis markers in the trastuzumab ILD case Systemic methylprednisolone had induced rapid remission of the chest X-ray infiltration and inflammatory marker (e.g. C-reactive protein, erythrocyte sedimentation rate, D-dimer) improvement during the initial induction period (Table 2). However, the patient still manifested with residual vasculitis disease activity of leg weakness, numbness, and also persistent microscopic hematuria and proteinuria. Occasional expectoration of blood clot content was also noted despite no evident change in the chest X-ray appearance. The measurement of Birmingham Vasculitis Activity Score (BVAS) [11], a tool of vasculitis activity measurement consisted of 9 organs, and a score of 0–33 for persistent symptoms, and 0 to 63 for new or worse symptoms, still scored 10–20 for persistent symptoms in the patient. The patient also encountered a rapid increase in the anti-PR3 titer eight months after discontinuation of trastuzumab (Table 2). Recent evidence has supported neutrophil as the dominant infiltrate within vasculitis lesions [[12], [13], [14]]. And the discovery of neutrophil extracellular traps (NETs), a component of cell-free DNA, histone, proteinase 3 (PR3), and myeloperoxidase (MPO) released by ANCA-stimulated neutrophil, could induce vasculitis damage such as thrombus formation and endothelial damage. The patient was enrolled in a prospective vasculitis cohort in our facility for testing on vasculitis neutrophil-derived NETs and had also started rituximab (500 mg D1/D15, every 6 months) therapy for better control for the DIILD.Table 2 Trend of inflammatory markers since interstitial lung disease event onset; arrowheads indicate rituximab treatment. MPO-DNA showed a good correlation with Birmingham Vasculitis Activity Score (BVAS), especially in the late phase (r = 0.996, p = 0.004) after Rituximab treatment. This AAV patient had a protracted disease course that involved BVAS score rebound despite the PR3 autoantibody titer remaining below the upper limit. The results indicate that the MPO-DNA complex, which is a novel biomarker of neutrophil activation, could better monitor vasculitis activity. Table 2 2 Prospective cohort on neutrophil derived biomarkers in vasculitis disease activity 2.1 Materials and methods From 2017 to 2021, a prospective cohort study addressing neutrophil related biomarkers in the evaluation of patients with vasculitis was initiated in our facility (Suppl. Fig. S1). The aim was to explore the relationship between levels of cell-free MPO-DNA, a biomarker for NETosis and the clinical activity of systemic vasculitis. Including the Case presented in the manuscript, a total of eight patients with vasculitis and 17 healthy controls (a ratio of 1:2) were enrolled, and serial serum testing was obtained (Suppl. Fig. S2). The study was approved by the hospital Institutional Review Board in 2017 (NTUH: 201612147RINA). Informed consents were obtained from all patients and healthy donors. 2.2 Measurement of MPO-DNA and ANCA We tested each vasculitis individual for the cell-free MPO-DNA via a “sandwich” ELISA with an anti-MPO polyclonal antibody (GeneTex, GTX22088, Irvine, Ca, USA). ELISA microplates were coated with the MPO monoclonal antibody overnight to capture MPO-associated DNAs. Anti-ds DNA-specific monoclonal antibodies (Abcam, Cambridge, ab27156, UK; 1:2000) were added to bind MPO-associated DNA, followed by binding of a horseradish peroxidase-conjugated anti-mouse IgG antibody (Jackson ImmunoResearch, 115-035-003) for detection. A peroxidase substrate (3,3′,5,5′-Tetramethylbenzidine) was added to react with the conjugated peroxidase to yield a blue product. The reaction was halted by adding 2 N H2SO4, and the absorbance of the final product was measured at 450 nm and was transformed to an arbitrary unit (a.u.) according to the previous study [15]. An MPO-DNA assay standard curve is provided (Suppl. Fig. S3). ANCA assessment was performed using anti-proteinase (PR3) and anti-myeloperoxidase(MPO) specific immunoassays (EliA; Thermo Fisher Scientific, Waltham, MA, USA). The assay was performed according to the manufacturer's instructions. An antibody concentration was considered positive if: MPO >5.0 IU/mL and PR3 >3.0 IU/mL (normal range were provided by the manufacturer). 2.3 Statistical analysis We utilized Stata statistical software (V. 14.0, Stata Corporation, College Station, TX) for the analysis. The MPO-DNA cut-off point between healthy controls and vasculitis patients was obtained according to the maximum Youden index (Sensitivity + Specificity - 1) to capture the performance of this dichotomous diagnostic test [16]. Statistical significance of the difference between two sets of continuous variables was analyzed using a Mann-Whitney U test. A p-value less than 0.05 was defined as statistically significant. 3 Results 3.1 MPO DNA performance in the prospective vasculitis cohort The MPO-DNA values were significantly higher in the systemic vasculitis group (n = 8; 0.092 ± 0.071 arbitrary unit (a.u.)) relative to healthy donors (n = 17; 0.015 ± 0.018 a.u., p = 0.01) (Table 3). The Youden J index representing the maximum potential effectiveness of the upper-lower limit of the MPO-DNA cut-off value was 0.047 (a.u.), the red dashed line (Table 3). An MPO-DNA cut-off level at ≤ 0.047 a.u. was established to differentiate between healthy control and vasculitis patients. We also tested MPO-DNA serum level, erythrocyte sedimentation rate, C-reactive protein correlation with their Birmingham Vasculitis Activity Score. The MPO-DNA showed a moderate positive correlation [ r (n = 12) = 0.596, p = 0.025]; but the C-reactive protein [ r(n = 10) = 0.402, p = 0.196] and the erythrocyte sedimentation rate [ r(n = 9) = 0.119, p = 0.727] showed no correlation with clinical vasculitis severity. These results suggest that MPO-DNA levels may also be useful in monitoring vasculitis activity compared to standard inflammation biomarkers (Table 4).Table 3 MPO-DNA values were significantly higher in the systemic vasculitis group [n = 8; 0.092 ± 0.071 arbitrary units (A.U.)] relative to healthy donors [ n = 17; 0.015 ± 0.018 A U., p = 0.01 ]. Youden J index representing the maximum potential effectiveness of the upper-lower limit of the MPO-DNA cut-off value, which was set at 0.047 (A.U.), as indicated by the red dashed line. Table 3 Table 4 The MPO-DNA serum levels, erythrocyte sedimentation rates, and C-reactive protein correlation with the Birmingham Vasculitis Activity Score for the eight vasculitis patients, the MPO-DNA showed a moderate positive correlation but the C-reactive protein and erythrocyte sedimentation rate did not. Table 4 3.2 MPO DNA performance in the trastuzumab ILD case During follow-up of the Case presenting trastuzumab-induced ILD, we found that despite standard vasculitis anti-PR3 is capable of showing good correlation during initial glucocorticoid therapy with the Birmingham vasculitis score (BVAS). The correlation was lost after salvage treatment with the B-cell targeting agent rituximab was initiated (Table 2). MPO-DNA levels exhibited a better correlation with the clinical vasculitis activity BVAS score during the rituximab treatment period. 4 Discussion Interstitial lung disease (ILD) induced by HER2-targeted agents is a well-known adverse drug reaction, but the mechanism is ill-defined and considered low in incidence [17]. HER-2 targeted agents such as trastuzumab, trastuzumab emtansine had been reported to induce cutaneous vasculitis in the previous literature [[18], [19], [20]], but pulmonary vasculitis due to HER2 targeted agents had not been reported yet. A recent review of 9886 patients investigating anti-HER2 therapies for HER2 breast cancer had reported the overall incidence of ILD was 2.4%. The incidence of grade 1–2, grade 3–4, and grade 5 events were 66.7%, 23.0%, and 0.2% respectively. The agents leading to the highest ILD incidence was trastuzumab combined with everolimus or paclitaxel. The incidence of ILD-related deaths was highest among patients receiving trastuzumab deruxtecan (T-Dxd), with an incidence of around 2% [17]. Recently the oncology society has drawn attention to this adverse event due to the recent accelerated approval of T-Dxd [9,21]. The novel antibody-drug conjugate had shown a 60% response rate in third or later line metastatic HER-2 positive breast cancer but at the cost of treatment-related ILD up to 13% and 2% treatment-related death. This high rate of side effects will inhibit its potential to be utilized as the frontline therapy or even be placed in the curative adjuvant or neoadjuvant setting. A recently published article by Kumagai et al. had reported successful induction of T-Dxd interstitial lung disease in a cynomolgus monkey model. Receiving T-Dxd in a monkey model developed interstitial lung disease, whereas receiving Dxd does not. Although most ILD lesions were found within the alveoli, the HER2 expression in lungs was limited to the bronchial level [22]. Vasculitis due to a pathway via ANCA autoantibodies, an indirect mechanism by the immune system, could explain why the injury was not at the location where T-Dxd was uptaken. Moreover, a two-step mechanism of pathogenesis explains how DIILD (drug-induced ILD) manifests later after anti-HER2 therapies. A literature review was performed to summarize previously reported cases of trastuzumab-related ILD. A total of 8 cases were identified before our Case [[23], [24], [25], [26], [27], [28]]. Two of the reported cases had available pathology that showed organizing pneumonia or diffuse alveolar damage. No traceable serum marker was reported in previously reported patients. All previously reported cases that we identified had late onset with a median time of onset two months after the first trastuzumab exposure (Table 5). The pattern of CT infiltration largely comprised ground-glass opacity and patch consolidations, with only one case exhibiting nodular lesions. Most patients recovered following prednisolone-based monotherapy, and only one patient died. Long-term follow up was not conducted in these cases. Our case also had a protracted course of vasculitis up to two years. We found that both ANCA auto-antibody testing and NET related biomarker, MPO-DNA should be tested concomitantly and especially when B cell depleting agents (such as rituximab use in our case) inhibit the autoantibody production, but vasculitis-related neutrophil activity is still high. Reliance on ANCA autoantibodies alone to monitor vasculitis activity may not be sufficient.Table 5 Case comparison of previously reported trastuzumab-related interstitial lung disease [[12], [13], [14], [15], [16], [17]]. Table 5Case Age Gender Previous CRT (setting) Tmab exposure CT pattern Pathology Treatment Recovery Naranjo Score 1 49 F [23] AC, RT, D (adjuvant) 3 months Subpleural consolidation Organizing pneumonia n/aa Yes 6 2 56 F [25] FAC, D (salvage) 4 months Consolidation and pleural effusion BAL: eosinophil 18% Prednisolone (40 mg/d) Yes 8 3 51 F [26] P (neoadjuvant) 10 weeks Ground glass opacity TBLB: Intra-alveolar hemorrhage, interstitial inflammation Prednisolone (40 mg/d) Yes 6 4 63 F [24] D (neoadjuvant) 5 weeks Peripheral consolidation, GGO Diffuse alveolar damage Prednisonec No (mortality) 7 5 67 F [35] DC (adjuvant) 4 months GGO n/a Prednisone Cyclophosphamideb Yes 8 6 68 F [28] EC (adjuvant) 3 months Patch infiltration, GGO n/a Semi-pulse steroidc Yes 6 7 62 F [28] DC (adjuvant) 3 months Patchy consolidation, poorly defined nodules n/a Prednisolonec Yes 7 8 71 F [36] Salvage Tmab followed by T-DM1 Tmab 12 wks T-DM1 6 wks Diffuse interstitial infiltrates n/a Methylprednisolone (1 mg/kg/d) Yes 5 9 59 F P (adjuvant) 3 months Patchy consolidation ANCA capillaritis, acute interstitial pneumonia Prednisolone (40 mg/d) Yes 7 CRT, chemoradiotherapy; Tmab, trastuzumab; T-DM1, trastuzumab emtansine; GGO, ground glass opacity; A, doxorubicin; C, cyclophosphamide; RT, radiotherapy; F, 5-fluorouracil; P, paclitaxel; D, docetaxel; BAL, bronchoalveolar lavage; TBLB, transbronchial lung biopsy. a Resolved spontaneously after trastuzumab discontinuation. b Cyclophosphamide was used upon trastuzumab rechallenge. c The dose used was not reported. Systemic glucocorticoids combined with either rituximab or cyclophosphamide are recommended for induction therapy for life-threatening AAV [29]. Autoreactive B cell depletion by rituximab can effectively reduce disease activity and decrease pathogenic ANCA concentrations in AAV patients. Although adverse events after rituximab include hepatitis B reactivation [30] and progressive multifocal leukoencephalopathy [31], the safety profile of rituximab is still considered to be favorable. Further salvage therapy or investigational agents included belimumab [32], a monoclonal antibody against soluble B cell-activating factor (BAFF) that can induce B cell apoptosis and showed clinical activity as an add-on therapy. The use of therapeutic plasma exchange is also beneficial for carefully selected patients experiencing severe diffuse alveolar hemorrhage or a serum creatinine level of ≥500 mmol/L [33]. There are still two crucial pieces of information that cannot be determined from our Case. First, neither serum levels of anti-PR3/anti-MPO level prior to trastuzumab exposure were available for this case. A prospective examination of vasculitis markers may be considered in the future to determine whether ANCA has clinical utility in predicting side effects during HER2 targeted therapy, especially antibody or antibody drug-conjugates. Secondly, the majority of drug-induced AAV cases present with anti-MPO positivity [34], but our case presented with anti-PR3 positivity without any upper airway symptoms or granuloma formation. Whether this difference is related to the medication administered or due to individual patient differences cannot be explained at present. In conclusion, this Case is the first clinical evidence of HER-2 targeted therapy induces pulmonary ILD via ANCA autoantibodies. Early suspicion with testing of ANCA level can be life-saving when a biopsy cannot be immediately obtained. Pre-treatment ANCA level testing may also be considered in the future when treating patients with HER-2 agents with higher risk profiles of ILD. In addition, serum MPO-DNA, could be a biomarker with promising clinical implication, as the results from this case and a prospective vasculitis cohort in our facility indicated. Authors’ contribution C.H.C., Y.M.H., and Y.M.K., wrote the original draft writing with input from all authors; Y.M.K., C.J.J., designed, developed and conducted the experiments; Y.M.K., L.C., are the primary physicians responsible for the patient medical treatment; Y.L.C., reviews the surgical pathology and issues the report.; S.C.H., Y.M.H., C.H.L., contributed to the final version of the manuscript; All authors actively participated in the discussion and suggestion for the manuscript. Funding This article was subsidized for English editing by 10.13039/501100006477 National Taiwan University under the Excellence Improvement Program for Doctoral Students (grant number 108-2926-I-002-002-MY4), sponsored by the 10.13039/501100004663 Ministry of Science and Technology, Taiwan . This work was supported by Ministry of Science and Technology (MOST 108-2314-B-002-097-MY3, 107-2314-B-002-253-, 106-2314-B-002-100-), 10.13039/501100005762 National Taiwan University Hospital (NTUH 110-M4972) and Taiwan Rheumatology Association Research Fund (TRARF-2018). Ethics approval This study was approved by the hospital Institutional Review Board (NTUH: 201612147RINA). The informed consent was obtained from the patients. Availability of data and material The author(s) confirm that the data supporting the findings of this study are available within the article. Code availability The author(s) confirm that the data supporting the findings of this study are available within the article. Declaration of competing interest The author(s) had declared that no conflicts of interest exist. Appendix A Supplementary data The following is the Supplementary data to this article:Multimedia component 1 Multimedia component 1 Appendix A Supplementary data to this article can be found online at https://doi.org/10.1016/j.breast.2021.11.016.

Recovering

ReactionOutcome

CC BY-NC-ND

34894465

2021-11-29