Datasets:

yirenc
/

guidelines38K

Modalities:

Formats:

Size:

Tags:

Libraries:

Dataset card Viewer Files Files and versions Community

Dataset Viewer

Auto-converted to Parquet

View in Dataset Viewer

Split (1)

train · 38k rows

clean_text stringlengths 35 1.63M
# QUESTIONS Diagnosis/Staging What benefit to clinical management does positron emission tomography (PET) or positron emission tomography/computed tomography (PET/CT) contribute to the diagnosis or staging of head and neck cancer? What benefit to clinical management does PET or PET/CT contribute to the assessment of treatment response for head and neck cancer? What benefit to clinical management does PET or PET/CT contribute when recurrence of head and neck cancer is suspected but not proven? What benefit to clinical management does PET or PET/CT contribute to restaging at the time of documented recurrence for head and neck cancer? What is the role of PET when a solitary metastasis is identified at the time of recurrence and a metastectomy is being contemplated? # TARGET POPULATION Patients with head and neck cancer are the target population for this recommendation report. # INTENDED PURPOSE This recommendation report is intended to guide the Ontario PET Steering Committee in their decision making concerning indications for the use of PET imaging. This recommendation report may also be useful in informing clinical decision making regarding the appropriate role of PET imaging and in guiding priorities for future PET imaging research. # RECOMMENDATIONS AND KEY EVIDENCE These recommendations are based on an evidentiary foundation consisting of one recent high-quality U.K. Health Technology Assessment (HTA) systematic review (1) that included systematic review and primary study literature for the period from 2000 to August 2005, an update of this systematic review undertaken to retrieve the same level of evidence for the period from August 2005 to June 2008, and a subsequent literature search was conducted to retrieve literature from June 2008 to July 2011. PET is recommended in the M and bilateral nodal staging of all patients with head and neck squamous cell carcinoma where conventional imaging is equivocal, or where treatment may be significantly modified. HTA review 2007 (1): One systematic review of four primary studies and one additional primary study showed PET was sensitive and specific and useful where doubt exists (CT/MRI gave different and less optimal results). PET changed stage and treatment planning. 32) and Guido et al (33) indicated that the addition of PET improved primary tumour delineation and nodal staging and subsequently changed the clinical management of several patients in each study. PET is recommended in all patients after conventional imaging and in addition to, or prior to, diagnostic panendoscopy where the primary site is unknown. HTA review 2007 (1): Two systematic reviews (each with eight primary studies) and two additional primary studies showed that PET can detect primary unknown tumours in patients with cervical lymph node metastases. PET detects 30% of primary tumours, including those missed by conventional imaging. 2005-2008 update: One primary study showed that PET is better than conventional imaging in detecting site of primary tumour (Chen et al. 2008 2011 update: One primary study indicated that patients with cervical metastasis and an unknown primary site after undergoing conventional imaging or clinical examination benefit from PET/CT prior to panendoscopy (Rudmik et al PET is recommended for staging and assessment of recurrence of patients with nasopharyngeal carcinoma if conventional imaging is equivocal. HTA review 2007 (1): This topic was not addressed in the HTA review. # Qualifying Statements This report makes no distinction between studies examining PET and those examining PET/CT. Conventional imaging refers to CT and/or magnetic resonance imaging (MRI) unless otherwise specified. Retrospective design studies were excluded from this review, but several exist favouring the use of PET for head and neck cancer. With respect to primary site (T): - PET appears to be more accurate for the diagnosis of primary tumours, especially in cases where CT/MRI results are equivocal (2008-2011 update: Guido et al Wang et al. o PET can identify the primary site in 30% of cases when undetected by clinical assessment and conventional imaging. - PET can detect some synchronous primaries that may be missed by other modalities. With respect to regional nodes (N): In the clinically N-0 neck, PET does not appear to be better than conventional imaging, because of an unacceptably high false-negative rate. There is little evidence that PET leads to change in patient management (2005-2008 update: Hafidh et al Ng et al Schoder et al Wensing et al Kim et al ; 2008-2011 update: Moeller et al and Kyzas et al Liao et al. There was moderate evidence that PET scanning changed nodal staging status and/or radiation treatment planning. However, in many cases there was no pathologic confirmation of PET versus conventional imaging discrepancy. Exceptions were cases where distant metastatic disease was identified by PET and changed treatment (2005-2008 update: Connell et al. # With respect to distant disease (M): - There is strong evidence that PET imaging is valuable in detecting distant metastatic disease and is better than conventional imaging. The advantage of PET is overwhelming for patients at high risk for distant disease, which includes locally advanced disease and nasopharyngeal carcinoma. The substantial incidence of falsepositive rates of PET may mitigate the advantages for low-risk patients (2008-2011 update: Kim et al Law et al Lonneux et al Martin et al Ng et al Senft et al Yamazaki et al Wang et al. # Qualifying Statements With respect to recurrence and tumour surveillance after treatment, the evidence suggests that sites of disease that are clinically accessible for assessment did not benefit from PET imaging. However, for disease sites that were either not clinically accessible or difficult to examine, PET imaging showed significant advantages over conventional evaluation. - Larynx: moderate evidence that PET is beneficial/better than conventional imaging in detecting recurrent disease. PET also reduced the need for debilitating laryngeal biopsies (2005-2008 update: Gordin et al Brouwer et al. . With respect to the role of PET in assessing status of neck lymphadenopathy following radiation or chemoradiation, moderate evidence suggests that PET-directed management of the neck after therapy, appropriately spares neck dissections in patients with PETnegative residual CT abnormalities (2008-2011 update: Porceddu et al. Copyright This report is copyrighted by Cancer Care Ontario; the report and the illustrations herein may not be reproduced without the express written permission of Cancer Care Ontario. Cancer Care Ontario reserves the right at any time, and at its sole discretion, to change or revoke this authorization.
# GUIDELINE OBJECTIVES To update clinical guidance on the use of multigene profiling assays in individuals with earlystage invasive breast cancer. # TARGET POPULATION Individuals diagnosed with early-stage invasive breast cancer for whom further information is needed for prognosis and treatment decision making. In this guideline, early-stage invasive breast cancer is defined as stage I to III breast cancers that are surgically operable and do not have evidence of inflammatory, locally recurrent or distant metastatic disease with pT1-T3, pN0-N1a based on surgical pathologic staging. # INTENDED USERS This guideline is targeted for clinicians and policy makers involved in the diagnosis and treatment of breast cancer. The purpose of this guideline is to determine the clinical utility of multigene profiling assays (i.e., Oncotype DX, MammaPrint, Prosigna, EndoPredict, and Breast Cancer Index), not to identify which assay is better. No prospective studies have compared these head-to-head. Given that the assays use different scoring systems and classification systems, please refer to Table 1-1 for a summary of each of the assays. Further, this guideline does not cover the utility of multigene profiling assays in helping to guide clinical treatment decisions regarding the use of either neoadjuvant chemotherapy or radiation. # RECOMMENDATIONS, KEY EVIDENCE, AND JUSTIFICATION Recommendation 1 In patients with early-stage estrogen receptor (ER)-positive/human epidermal growth factor 2 (HER2)-negative breast cancer, clinicians should consider using multigene profiling assays (i.e., Oncotype DX, MammaPrint, Prosigna, EndoPredict, and Breast Cancer Index) to help guide the use of systemic therapy. # Qualifying Statements for Recommendation 1 - There is currently insufficient evidence to use multigene profiling assays among patients with either HER2-positive or triple negative breast cancers. - Multigene profiling assays are recommended for use in patients with lymph nodenegative or lymph node-positive (1-3 lymph nodes) disease who are under consideration for adjuvant chemotherapy if the use is supported by other clinical, pathological, or patient-related factors. Clinical and pathological features include patient age, tumour grade, tumour size and nodal status. - One multigene profiling assay should be requested per patient to guide a specific treatment decision. Requesting multiple tests with different multigene profiling assays on an individual tumour specimen to guide a single treatment decision is discouraged. Additional testing may be considered for patients with either repeat metachronous breast cancer diagnoses or synchronous breast cancer diagnoses where tumour specimens display varying morphologies, grade or hormone receptor status. - Multigene profiling assays should be interpreted cautiously in premenopausal patients where a significant benefit from adjuvant chemotherapy may still exist despite a lowrisk score. - It is uncertain whether at least some of the benefit of chemotherapy among premenopausal patients may be due to chemotherapy induced amenorrhea versus the cytotoxic effects of treatment. - The Prosigna, EndoPredict/EPclin, and Breast Cancer Index assays can identify low-risk node-positive patients whose prognostic outcomes are favourable; however, these assays have not demonstrated predictive evidence to support withholding adjuvant chemotherapy among higher risk, node-positive, ER-positive, HER2-negative breast cancer patients. # Recommendation 5 The evidence to support the use of molecular profiling to select the duration of endocrine therapy is evolving. In patients with ER-positive disease, clinicians may consider using a Breast Cancer Index (BCI) (H/I) high assay result to support a decision to extend adjuvant endocrine therapy if the decision is supported by other clinical, pathological, or patientrelated factors. # Qualifying Statements for Recommendation 5 - While a number of studies have demonstrated clinical utility of BCI for extending adjuvant endocrine therapy, the preliminary results of the NSABP B42 trial are negative leading to some uncertainty. Treatment decisions should be based on all available clinical and pathological information for each patient, rather than depending only on multigene profiling tests. - MammaPrint, Oncotype DX, Prosigna, and EndoPredict currently have insufficient evidence to guide extension of adjuvant endocrine therapy; however, these molecular assays may prognosticate a very low rate of disease recurrence that might not justify an extension of endocrine therapy.
# GUIDELINE OBJECTIVES To make recommendations with respect to the use of T2-weighted magnetic resonance imaging (MRI) ± functional sequences in the pre-treatment local staging of patients with newly diagnosed prostate cancer. - MRI refers to T2-weighted MRI. - Functional sequences include dynamic contrast-enhanced imaging (DCE), diffusionweighted imaging (DWI), and proton magnetic resonance spectroscopic imaging (MRS). - In this guideline, the terminology of MRI ± functional sequences is used interchangeably with MRI ± DCE, DWI, and MRS (See Glossary of Terms, Appendix 1). # TARGET POPULATION Men with newly diagnosed biopsy-confirmed prostate cancer who are under consideration for radical treatment. # INTENDED USERS Clinicians who are involved in the staging and treatment of prostate cancer patients. # RECOMMENDATIONS Recommendation 1 Multiparametric MRI (mpMRI) use for pre-treatment local staging of prostate cancer is a reasonable option for assessment of extraprostatic extension (EPE) in intermediate-and highrisk patients being considered for radical therapy if knowledge of EPE will alter management. # Qualifying Statements for Recommendation 1 - mpMRI is the addition of two or more functional sequences to
Evidence-Based Series 4-5 is CURRENT as of November 2022. In June 2016 a literature search and review determined that the recommendations were still valid and not causing harm but all relevant aspects of the topic were not covered by the original guideline. This document was identified for an UPDATE but it is anticipated that key evidence will be published in the next few years and a new guideline will be undertaken at that time.# New evidence found by update searches since completion of the original guideline is consistent with the original recommendations. # Guideline Question For women with cervical cancer in whom radiotherapy is considered appropriate, does the addition of concurrent platinum-based chemotherapy improve survival and quality of life with acceptable toxicity? # Target Population These recommendations apply to women with cervical cancer for whom primary treatment with radiotherapy is being considered: -those with locally advanced cervical cancer, -those with bulky clinical stage IB (>4 cm) cervical cancer, who are treated with radiotherapy, -those with high-risk early-stage cervical cancer (node-positive or margin-positive), who will be treated with radiotherapy following hysterectomy. - Women with cervical cancer for whom treatment with radiotherapy is being considered (described above) should be offered concurrent cisplatin with their course of radiotherapy. There are no direct comparisons of different cisplatin regimens. Based on the review of the available toxicity data from the randomized controlled trials, the Disease Site Group felt that cisplatinum should be given weekly (40 mg/m 2. # Qualifying Statements - Despite this recommendation, other schedules and doses have been used; thus, there is no conclusive evidence that one dose and schedule is better than the other. There is insufficient evidence available to make recommendations on the addition of 5fluorouracil to cisplatin during radiotherapy Entries to MEDLINE (1966through June 2004, EMBASE (1980EMBASE ( through week 25, 2004, CANCERLIT (1975through October 2002, and Cochrane Library (2004, Issue 2) databases and abstracts published in the proceedings of the annual meetings of the American Society of Clinical Oncology from 1999 to 2004 were systematically searched for evidence relevant to this practice guideline report. Evidence was selected and reviewed by members of the Practice Guidelines Initiative's Gynecology Cancer Disease Site Group and methodologists. This practice guideline report has been reviewed and approved by the Gynecology Cancer Disease Site Group, comprised of medical oncologists, radiation oncologists, a pathologist, an oncology nurse and patient representatives. External review by Ontario practitioners is obtained for all practice guideline reports through a mailed survey. Final approval of the practice guideline report is obtained from the Practice Guidelines Coordinating Committee. The Practice Guidelines Initiative has a formal standardized process to ensure the currency of each guideline report. This process consists of the periodic review and evaluation of the scientific literature and, where appropriate, integration of this literature with the original guideline information. # Key Evidence - Eight randomized controlled trials were eligible for the evidence review: six compared cisplatin-based chemotherapy plus radiotherapy to radiotherapy alone (in one of those trials, para-aortic radiotherapy was added to pelvic radiotherapy in the control arm) and two compared cisplatin-based chemotherapy plus radiotherapy to radiotherapy plus hydroxyurea. - The guideline authors pooled survival data from published reports. Pooled survival rates detected a statistically significant effect in favour of cisplatin-based chemotherapy plus radiotherapy compared with radiotherapy alone or with hydroxyurea (relative risk of death, 0.74; 95% confidence interval, 0.64 to 0.86). - The pooled relative risk of death among the six trials that enrolled only women with locally advanced cervical cancer was 0.78 (95% confidence interval, 0.67 to 0.90) in favour of cisplatin-based chemotherapy and radiotherapy. - The pooled relative risk for the two trials in high-risk early-stage disease also demonstrated a significant benefit for the addition of cisplatin-based chemotherapy to radiotherapy (relative risk, 0.56; 95% confidence interval, 0.41 to 0.77). - Rates of serious hematologic, gastrointestinal and genitourinary acute adverse effects are higher with cisplatin-based chemotherapy plus radiotherapy than with radiotherapy alone. For further information about this practice guideline, please contact the authors through the PEBC via: Phone: 905-527-4322 ext. 42822 Fax: 905-526-6775 E-mail: ccopgi@mcmaster.ca The Practice Guidelines Initiative is sponsored by: Cancer Care Ontario & the Ontario Ministry of Health and Long-term Care. Visit for all additional Practice Guidelines Initiative reports.
# GUIDELINE OBJECTIVES To make recommendations with respect to the role of adjuvant systemic chemotherapy in stage II and III colon cancer patients who have undergone complete resection with curative intent. # TARGET POPULATION The target population consists of adult patients with stage II and III colon cancer who have undergone complete resection with curative intent as primary therapy. # INTENDED USERS Intended users of this guidance document are clinicians involved in the delivery of adjuvant systemic chemotherapy for stage II and III colon cancer patients. # Stage II Colon Cancer Recommendation 1 The routine use of adjuvant chemotherapy for all patients with stage II colon cancer is not recommended. However, adjuvant therapy is a reasonable option for the subset of patients with high-risk stage II disease. While there is controversy about which tumour features denote high risk in stage II patients, this subset includes patients with inadequately sampled nodes, T4 lesions, perforation at the site of the tumour, or poorly differentiated histology in the absence of microsatellite instability (MSI) or mismatch repair deficiency (dMMR). # Qualifying Statements for Recommendation 1 - The clinical decision should be based on discussions with the patient about the nature of the evidence supporting treatment, the anticipated morbidity, the presence of high-risk prognostic features on individual prognosis, and patient preferences. - The enrolment of resected stage II patients in clinical trials is encouraged. Additional trials comparing adjuvant therapy with observation are needed and are ethically acceptable in stage II colon cancer. # Recommendation 2 When treated with adjuvant therapy, high-risk stage II patients should receive a fluoropyrimidine. There are insufficient data in support of oxaliplatin providing additional benefit to all high-risk individuals. The 2015 guideline recommendations have been ENDORSED, which means that the recommendations are still current and relevant for decision making. Please see Section 6: Document Assessment and Review for a summary of updated evidence published between 2015 and 2018, and for details on how this guideline was ENDORSED. # Qualifying Statements for Recommendation 2 - It would be reasonable to consider FOLFOX for high-risk patients as part of an informed discussion between patients and their medical oncologists regarding treatment options. # Added to the 2019 Endorsement - Additional evidence is expected that will inform decisions on duration of treatment with oxaliplatin-based treatment in patients with stage II disease. The following data are from a recent abstract (Iveson, ASCO, 2019), and thus should be considered with caution. The IDEA collaboration evaluated 3 vs 6 months of therapy in a randomized, pre-planned, pooled analysis of 4 RCTs focusing on high-risk stage II patients. The decision to use CAPOX or FOLFOX was left to the treating physician. Noninferiority was not met for DFS comparing 3 vs 6 months (HR 1.18, 95% CI 1.05 to 1.31; noninferiority margin was 1.2). Five-year DFS was 80.7% vs 84.0% for 3 and 6 months, respectively. There was a significant reduction in grade 3 to 5 toxicity with 3 months of therapy (irrespective of regimen). See Section 6 for details. Most patients suitable for oxaliplatin-based combination chemotherapy should discuss the differences between CAPOX and FOLFOX with their oncologist and choose a balance between efficacy and toxicity: - The IDEA results suggest that 3 months of CAPOX results in very similar efficacy to 6 months, whereas it appears that 3 months of FOLFOX resulted in slightly lower DFS (but the interaction test for duration and regimen was not statistically significant). - The duration of 5-FU monotherapy was not addressed in IDEA, and should remain 6 months. # Recommendation 3 Adjuvant chemotherapy with a fluoropyrimidine monotherapy regimen following surgery in patients who have MSI/dMMR is not recommended. MSI/dMMR testing should be performed for all stage II patients for whom adjuvant chemotherapy is being considered. In stage II (in the absence of high-risk features) where a patient does not require adjuvant chemotherapy, MSI/dMMR testing is not recommended as it will not influence that decision. # Qualifying Statements for Recommendation 3 - In patients with high-risk stage II colon cancer (e.g., T4) and high MSI/dMMR status (a low risk factor), the choice of treatment is between observation and FOLFOX, but data are lacking to guide this decision. # Stage III Colon Cancer # Qualifying Statements for Recommendation 4 - 5-FU may be given intravenously in combination with LV and oxaliplatin in the regimens known as FOLFOX or FLOX, or capecitabine may be given orally in combination with intravenous oxaliplatin in the regimen known as XELOX. These oxaliplatin-containing regimens have demonstrated superior overall survival when compared with 5-FU plus LV and are the recommended regimens. Oxaliplatin administration is associated with a 12.5% risk of severe neuropathy which is permanent in approximately 1% of patients. This needs to be considered in conjunction with the expected benefits of therapy. - Owing to the toxicity profile of FLOX, it is used less frequently than FOLFOX. - Some patients would not be considered appropriate for oxaliplatin-containing regimens. Examples include patients with underlying neurological conditions or at increased risk of neuropathy, patients at increased risk for infections, and patients likely to poorly tolerate infections as a result of chemotherapy. For these patients the treatment options are: o oral capecitabine which has equivalent efficacy to intravenous bolus 5-FU/LV. Capecitabine results in significantly less diarrhea, stomatitis, neutropenia, nausea/vomiting, and alopecia but significantly more hand-foot syndrome when compared with bolus 5-FU/LV. o 5-FU in combination with LV - Suitable patients should be offered entry into clinical trials testing new adjuvant treatments for resected stage III colon cancer. - Patients have begun their adjuvant treatment within four to nine weeks of surgery in the adjuvant randomized controlled trials of resected colon cancer. # Added to the 2019 Endorsement - The IDEA collaboration evaluated 3 vs 6 months of therapy in a randomized, pre-planned, pooled analysis of 6 individual trials focusing on stage III patients. The treatment choice of CAPOX or FOLFOX was left to the treating physician. Overall, noninferiority was not met for 3 vs 6 months (3-year DFS HR 1.07, 95% CI 1.0 to1.15; noninferiority margin was 1.12). Pre-planned sub-group analysis revealed superiority for 6 months of FOLFOX, whereas 3 months of CAPOX was found to be noninferior to 6 months. 3 months of treatment was associated with lower rates of adverse events independent of chemotherapy regimen (Grothey et al, NEJM, 2018). An unplanned analysis was devised sub-dividing patients into "low" and "high" risk stage III disease, and is the basis for our statements below. See Section 6 for details. - Low-risk stage III (T1-3 N1): 3 months of CAPOX is preferred over FOLFOX. Although the overall trial was negative for the primary endpoint, the shorter duration of treatment strikes a reasonable balance between efficacy and neurotoxicity of oxaliplatin (3 months noninferior to 6 months: HR 1.01, 95% CI 0.90 to 1.12). The pros and cons of 3 vs 6 months should be discussed with patients. Alternatively, 5-FU/capecitabine monotherapy for 6 months' duration remains an option, especially for patients with contraindications to oxaliplatin or preferences for oral chemotherapy. - High-risk stage III (T4 +/-N2): 6 months of oxaliplatin-based chemotherapy (CAPOX or FOLFOX). Although the overall trial was negative for the primary endpoint, the shorter duration of treatment resulted in lower DFS (6 months superior to 3 months: HR 1.12, 95% CI 1.03 to 1.23). The longer duration of therapy is associated with higher rates of neurotoxicity. The pros and cons of CAPOX vs FOLFOX need to be discussed with patients. # Recommendation 5 Although post hoc analyses of studies have not shown a clear benefit of adjuvant fluoropyrimidine plus oxaliplatin regimens in patients older than 70 years of age, it is reasonable to consider FOLFOX for patients older than 70 years as part of an informed discussion between patients and their medical oncologists regarding treatment options.
# Section 1: Recommendations Section 2: Recommendations and Key Evidence Section 3: Guideline Methods Overview Section 4: Systematic Review Section 5: Internal and External Review For information about this document, please contact Alex Sun, the lead author, through the PEBC via: Phone: 905-527-4322 ext. 42822 Fax: 905 526-6775 E-mail: ccopgi@mcmaster.ca For information about the PEBC and the most current version of all reports, please visit the CCO website at or contact the PEBC office at: Phone: 905-527-4322 ext. 42822 Fax: 905 526-6775 E-mail: ccopgi@mcmaster.ca Copyright This report is copyrighted by Cancer Care Ontario; the report and the illustrations herein may not be reproduced without the express written permission of Cancer Care Ontario. Cancer Care Ontario reserves the right at any time, and at its sole discretion, to change or revoke this authorization. # PEBC Report Citation (Vancouver
These guideline recommendations have been ENDORSED, which means that the recommendations are still current and relevant for decision making.# GUIDELINE OBJECTIVE How should patients presenting to family physicians and other primary care providers (PCPs) with signs and/or symptoms of prostate cancer, including incidental prostate specific antigen (PSA) test results, be managed? The following questions are the factors considered in answering the overall question: RESEARCH QUESTIONS 1. What signs, symptoms, and other clinical features that present in primary care are predictive of prostate cancer? 2. What is the diagnostic accuracy of investigations commonly considered for patients presenting with signs and/or symptoms of prostate cancer? 3. What major, known risk factors increase the likelihood of prostate cancer in patients presenting with signs and/or symptoms of prostate cancer? 4. Which factors are associated with delayed referral? Which delay factors can be attributed to patients, and which factors can be attributed to providers or system-related factors? Does a delay in the time to consultation affect patient outcome? # TARGET POPULATION Adult male patients presenting in primary care settings with signs, including incidental PSA results (defined as results not ordered by the attending FP or other primary care provider, or symptoms suggestive of prostate cancer comprise the target population. This guideline does not provide recommendations for screening healthy patients or opportunistic PSA testing. # INTENDED USERS This guideline is targeted to family physicians (FPs), general practitioners (GPs), emergency room physicians, other PCPs (nurse practitioners, registered nurses, and physician assistants), and urologists. For the purposes of this document, we have referred to FPs, GPs, emergency room physicians, and other PCPs as "FPs and other PCPs". The guidelines are also intended for policymakers to help ensure that resources are in place so that target wait times are achieved. They are intended to coincide with the introduction of prostate cancer Diagnostic Assessment Programs (DAPS) in Ontario. DAPs provide a single point of referral, coordination of care using a clinical navigator, fast tracking of diagnostic tests, and a multidisciplinary team approach. They are an Ontario-wide strategic priority designed to improve patient access and outcomes, as outlined in the Ontario Cancer Plan, 2005-2011 and 2011-2014 (1). Added in December 2019: Formal Cancer Care Ontario DAPs no longer exist in Ontario, but many hospitals provide ongoing multidisciplinary team approaches to diagnosing prostate cancer. The following recommendations were adapted from the New Zealand Guidelines Group (NZGG) guideline "Suspected Cancer in Primary Care: Guidelines for Investigation, Referral and Reducing Ethnic Disparities" and the National Institute for Health and Clinical Excellence (NICE 2005), "Referral Guidelines for Suspected Cancer" (2,3).The recommendations below reflect the integration of the NZGG 2009 and NICE 2005 recommendations, an updated systematic review of the research evidence since the NZGG 2009 and the NICE 2005 guidelines, and consensus by the PEBC Prostate Cancer Referral Working Group (see Section 2: Appendix 1 for a list of members) (2,3). The recommended wait times for referral were based on consensus as opposed to strong evidence from well-conducted studies. During the review process for this document in December 2016 when Version 2 of this guideline was ENDORSED, the Expert Panel noted that these wait time targets should be the goal, but may not always be possible. The recommended wait times for referral were based on consensus rather than strong evidence from well-designed studies. These targets are the goal, but may not always be possible. # KEY EVIDENCE AND JUSTIFICATION All recommended wait times were based on consensus of the Working Group. The Canadian Association of Radiation Oncology recommended a wait time from referral to consultation with a radiation oncologist of no longer than 10 working days (6). This was taken into consideration when developing the wait times in this guideline. The primary care literature evidence examining the diagnostic accuracy of tests for prostate cancer was very weak. Two studies suggested that DREs performed by FPs may be useful in identifying patients who should be referred (7,8), and four studies suggested that PSA values were good predictors of prostate cancer with PPVs ranging from 34.3% to 47% (7,(9)(10)(11). The working group chose to endorse the recommendations from NICE 2005 and NZGG 2009 to recommend a DRE and PSA test for all patients with symptoms of metastatic prostate cancer (2,3). NICE 2005 recommended performing a DRE and PSA test for all men with LUTS and NZGG 2009 recommended these tests only for older men with LUTS (2,3). The working group chose to recommend a DRE for all men with LUTS and a PSA test for selected patients with LUTS, following discussion and treatment. The limited evidence from the systematic review suggested that men with LUTS may not be at any higher risk for prostate cancer or have a poorer prognosis than asymptomatic men would be (9,12). The Canadian Urological Association's benign prostatic hyperplasia guideline for men presenting with LUTS recommended a DRE for all men and a PSA test for selected patients (13). The working group chose to be consistent with this guideline. # Recommendation 1. Actions for Patients with Symptoms of Metastatic Prostate Cancer The NZGG 2009 guideline recommendation that patients with symptoms of metastatic prostate cancer should have a DRE and PSA was endorsed (3). An age threshold of 40 years was included at the suggestion of the Expert Panel and due to the few cases of prostate cancer in men under 40 years in Canada (14). The working group did not think it necessary for a man with erectile dysfunction to undergo a DRE and PSA test and therefore excluded it as a symptom of metastatic prostate cancer. This is consistent with the NZGG 2009 guideline but in contrast to the NICE 2005 guideline (2,3). The working group also excluded unexplained hematuria as a symptom of metastatic prostate cancer because although it can be associated with advanced prostate cancer, the Working Group believed the vast majority of men with gross hematuria usually have a different underlying cause such as benign prostate hyperplasia, bladder or renal cancer, stones or infections. The working group believed hematuria requires urologic assessment but is not part of a prostate cancer care algorithm. a-c. The cut-off values of 10 and 20 ng/ml were taken from the D'Amico classification system for categorizing patients at low risk (cT1-cT2a, Gleason 20 ng/ml or Gleason >7) for prostate cancer (15,16). Although this was not developed in the primary care population, the working group chose to include this classification system because it is widely used to classify risk of prostate cancer and using these thresholds provides guidance for family physicians in determining their course of action. # Recommendation 2. Actions for Patients with Lower Urinary Tract Symptoms (LUTS) The recommendation that a man with LUTS should have a DRE and a discussion about PSA testing was consistent with the Canadian Urological Association's guideline for benign prostatic hyperplasia (13). The working group referred to the individual risk assessment developed by the Canadian Partnership Against Cancer as a guide to who should be given a PSA test (4). This document describes the benefits and harms of PSA testing. The working group also endorsed the recommendations to exclude urinary infection before PSA testing and to postpone PSA testing for at least one month after treatment from the NICE 2005 and NZGG 2009 guidelines (2,3). a. This recommendation was endorsed from the NICE 2005 guideline (2). b. The age-based PSA values were endorsed from the NZGG 2009 guidelines (3). c-e. Please refer to a-c in the previous section under Recommendation 1: Actions for Patients with Symptoms of Metastatic Prostate Cancer. f. i. A cut-off risk value of 5% was chosen because in Ontario, Canada, the hospital admission rate for urological complications within 30 days of TRUS-guided biopsy was found to be 4.1% in 2005 (17). The working group decided to use 5% as a cut-off to separate patients into a higher risk category because for these patients the risk of high-grade prostate cancer would be higher than the risk of complications from TRUSguided biopsy. ii-iii. The prostate risk calculator developed at Sunnybrook Hospital, Toronto, Ontario, Canada, showed a net benefit (the relative value of false-positive versus falsenegative results) when a risk of 15% for aggressive prostate cancer was chosen as a threshold to agree to a biopsy (18). Based on the consensus of the working group a conservative cut-off risk value of 20% was chosen. # Recommendation 3. Actions for Patients with Incidental PSA Although this guideline excludes patients in a screening program, the working group thought that FPs and other PCPs need guidance on how to manage patients with incidental PSA test results, a frequently encountered occurrence in practice. Opportunistic screening has been excluded because it is beyond the scope of this guideline. The working group believed that if an incidental PSA test was abnormal, then standard practise would be to perform a DRE. A hard or irregular prostate on DRE may increase the urgency of referral. Cases with enlarged, smooth prostates were excluded because it was beyond the scope of this guideline since it was not considered to be a sign of prostate cancer. Also, although a rising PSA level could be considered a sign of prostate cancer, the working group believed the guideline was sufficiently thorough to include most possible scenarios for prostate cancer using the absolute PSA values. Furthermore, there were no studies examining the factors associated with delayed referral that could directly inform these recommendations. # FUTURE RESEARCH Further studies are required that specifically investigate the diagnostic performance of signs, symptoms, or tests for prostate cancer in the primary care setting. # GLOSSARY Age-based PSA Age-based PSA values (upper limit of normal): 40-50 years: 2.5 ng/ml 50-60 years: 3.5 ng/ml 60-70 years: 4.5 ng/ml 70 years and over: 6.5 ng/ml Note: This is an example of an age-based range cited in the NZGG resource: Testing for prostate cancer: a consultation resource, 2008 (19). Differences in PSA assay can lead to differences in age-based ranges reported by laboratories. Prostate Risk Calculator developed by Nam et al 2011 is available here: (5). The prostate risk calculator includes the free:total PSA ratio, which is the ratio of free PSA, unbound to serum proteins, to total PSA. This ratio is decreased in men with prostate cancer (20). The free:total PSA ratio in some cases may be charged a laboratory fee to the patient. If this ratio is not determined, then a value of 0.1 can be entered into the risk calculator. # Symptoms of metastatic prostate cancer A healthy 70 year old vigorous gentleman, on no medications, who ran marathons yearly in the spring presented to a FP. He lived in Florida in the winter and usually was seen only once yearly in the spring. He came home to Canada earlier than usual as he had urinary retention in Florida, was catheterized but was having tremendous lower back pain. This thin, muscular man had never complained about lower back pain before. On examination, a firm fixed pelvic mass was noted. DRE noted a firm, irregular, fixed, and enlarged prostate. The urologist saw him within two days. A presumptive diagnosis of prostate cancer with bone metastasis was made. The PSA was 20ng/ml. Although diagnosis of prostate cancer was likely, the patient refused a biopsy and further diagnostic tests. His pain was quite severe and he was admitted to a palliative care unit for pain control and died within three weeks. A healthy 72 year old man with some symptoms of urinary retention and urgency presented to a FP. His older brother was diagnosed with prostate cancer at age 76. Urine analysis was negative and DRE found a smooth, normal prostate. The FP and patient discussed having a PSA test but the patient refused and asked to see a urologist to discuss the LUTS and his family history and was seen two months later. After a discussion with the urologist, the patient agreed to have a PSA and the result was 4.9ng/ml. The urologist explained to the patient that the result was within normal limits for his age. The patient elected to be followed with serial PSAs and DREs by his family physician. No treatments were initiated for the patient's symptoms of some urinary retention and urgency which seemed to resolve spontaneously. Since the first visit with the urologist, the PSA has been monitored every three months and has not increased beyond 6.8ng/ml in two years. # Incidental PSA A healthy 49 year old banker had a PSA test as part of a comprehensive medical examination offered through his insurance company. The physical examination was normal but the PSA was elevated for his age. He presented to his family doctor with a PSA of 3.5ng/ml and no other symptoms. The family doctor on DRE found a smooth, normal prostate. The family doctor evaluated the patient's risk for prostate cancer at 10-20% using the Prostate Risk Cancer nomogram and the patient elected to repeat the PSA and DRE in a few months. However, after further consideration at home, the patient called and asked to be referred to a urologist for a consultation. Copyright This report is copyrighted by Cancer Care Ontario; the report and the illustrations herein may not be reproduced without the express written permission of Cancer Care Ontario. Cancer Care Ontario reserves the right at any time, and at its sole discretion, to change or revoke this authorization.
To jointly update the Cancer Care Ontario guideline on brachytherapy for patients with prostate cancer to account for new evidence.An Update Panel conducted a targeted systematic literature review and identified more recent randomized controlled trials comparing dose-escalated external beam radiation therapy (EBRT) with brachytherapy in men with prostate cancer.Five randomized controlled trials provided the evidence for this update.For patients with low-risk prostate cancer who require or choose active treatment, low-dose rate brachytherapy (LDR) alone, EBRT alone, and/or radical prostatectomy (RP) should be offered to eligible patients. For patients with intermediate-risk prostate cancer choosing EBRT with or without androgen-deprivation therapy, brachytherapy boost (LDR or high-dose rate should be offered to eligible patients. For low-intermediate risk prostate cancer (Gleason 7, prostate-specific antigen 10 ng/mL or Gleason 6, prostate-specific antigen, 10 to 20 ng/mL), LDR brachytherapy alone may be offered as monotherapy. For patients with high-risk prostate cancer receiving EBRT and androgen-deprivation therapy, brachytherapy boost (LDR or HDR) should be offered to eligible patients. Iodine-125 and palladium-103 are each reasonable isotope options for patients receiving LDR brachytherapy; no recommendation can be made for or against using cesium-131 or HDR monotherapy. Patients should be encouraged to participate in clinical trials to test novel or targeted approaches to this disease. Additional information is available at www.asco.org/Brachytherapy-guideline and www.asco.org/ guidelineswiki.# INTRODUCTION The goal of this update is to provide oncologists, other health care practitioners, patients, and caregivers with recommendations regarding the use of brachytherapy for patients with prostate cancer that includes the most recent evidence. Prostate cancer is the most commonly diagnosed cancer in men. In 2016, it is estimated that there will be 180,890 new cases, along with an estimated 26,120 deaths. 1 For this reason, there is great interest in finding optimum treatment strategies to reduce the burden of disease in this patient population. The Cancer Care Ontario systematic review 2 and clinical practice guideline 3 on low-dose rate (LDR) brachytherapy for patients with low-or intermediate-risk prostate cancer were both published in 2013, and since then randomized evidence has been made available that might alter the original recommendations. The goal of this joint update is to consider this new evidence and determine if the original recommendations remain valid or if updates are warranted. The scope of this guideline covers brachytherapy boost and monotherapy. Currently, the American Society for Radiation Oncology, ASCO, and the American Urologic Association THE BOTTOM LINE Brachytherapy for Patients With Prostate Cancer: American Society of Clinical Oncology/Cancer Care Ontario Joint Guideline Update Guideline Questions 1. In patients with newly diagnosed prostate cancer, what is the efficacy of brachytherapy alone for clinical outcomes compared with external beam radiation therapy (EBRT) alone or radical prostatectomy (RP) alone? 2. In patients with newly diagnosed prostate cancer, what is the efficacy of brachytherapy combined with EBRT for clinical outcomes compared with brachytherapy alone, EBRT alone, or RP alone? 3. Among the isotopes used for low-dose rate (LDR) brachytherapy (eg, iodine-125 palladium-103 and cesium-131, which isotope maximizes clinical outcomes when used in patients with newly diagnosed prostate cancer? # Target Population Patients with newly diagnosed prostate cancer who require or choose active treatment and are not considering, or are not suitable for, active surveillance. # Target Audience Radiation oncologists, urological surgeons, and other clinicians who provide care for patients defined by the target population. A systematic review of the literature was performed and relevant evidence was evaluated for inclusion into this updated clinical practice guideline using the signals approach. # Updated Recommendations - For patients with low-risk prostate cancer who require or choose active treatment, LDR alone, EBRT alone, or RP should be offered to eligible patients - For patients with intermediate-risk prostate cancer choosing EBRT with or without androgen-deprivation therapy (ADT), brachytherapy boost (LDR or high-dose rate should be offered to eligible patients. For low-intermediate risk prostate cancer (Gleason 7, prostate-specific antigen 10 ng/mL or Gleason 6, prostate-specific antigen, 10 to 20 ng/mL) LDR brachytherapy alone may be offered as monotherapy. For patients with high-risk prostate cancer receiving EBRT and ADT, brachytherapy boost (LDR or HDR) should be offered to eligible patients. - 125 I and 103 Pd are each reasonable isotope options for patients receiving LDR brachytherapy; no recommendation can be made for or against using 131 Cs or HDR monotherapy. - Patients should be encouraged to participate in clinical trials to test novel or targeted approaches to this disease. # Qualifying Statements - Patients should be counseled about all their management options (surgery, EBRT, active surveillance, as applicable) in a balanced, objective manner, preferably from multiple disciplines. - Recommendation for low-risk patients is unchanged from initial guideline, because no new randomized data informing this question have been presented or published since. - Patients ineligible for brachytherapy may include: moderate to severe baseline urinary symptoms, large prostate volume, medically unfit, prior transurethral resection of the prostate, and contraindications to radiation treatment. - ADT may be given in neoadjuvant, concurrent, and/or adjuvant settings at physician discretion. It is noted that neoadjuvant ADT may cytoreduce the prostate volume sufficiently to allow brachytherapy - There may be increased genitourinary toxicity compared with EBRT alone. - Brachytherapy should be performed at a center following strict quality-assurance standards. - It cannot be determined whether there is an overall or cause-specific survival advantage for brachytherapy compared with EBRT alone, because none of the trials were designed or powered to detect a meaningful difference in survival outcomes. (continued on following page) are performing a joint review of hypofractionated radiotherapy (including stereotactic ablative body radiotherapy. 10 Where reported, these classifications are summarized. Androgen-deprivation therapy (ADT) is a standard for patients with high-risk prostate cancer treated with radiotherapy and can be considered for those with intermediate-risk disease. 11 Where reported, use and duration of ADT was also summarized. Evidence was also collected through a systematic review of the medical literature. Publications were included if they were phase III randomized clinical trials of brachytherapy compared with either EBRT or RP in men with prostate cancer. These publications were identified by rerunning the original strategy in MEDLINE, EMBASE, and the Cochrane database of systematic reviews, for the period from the original search in 2011 through to the end of August 2015. A final search for important papers was made in December 2016. Of the 32 publications identified, six 5,6,12-15 publications (addressing five RCTs) met the eligibility criteria and form the evidence base for this update. The Update Committee contributed to the development of the guideline, provided critical review, and finalized the guideline recommendations. All ASCO guidelines are reviewed and approved by the ASCO Clinical Practice Guidelines Committee. # Guideline Disclaimers The Clinical Practice Guidelines and other guidance published herein are provided by the American Society of Clinical Oncology, Inc (ASCO) to assist providers in clinical decision making. The information herein should not be relied upon as being complete or accurate, nor should it be considered as inclusive of all proper treatments or methods of care or as a statement of the standard of care. With the rapid development of scientific knowledge, new evidence may emerge between the time information is developed and when it is published or read. The information is not continually updated and may not reflect the most recent evidence. The information addresses only the topics specifically identified therein and is not applicable to other interventions, diseases, or stages of diseases. This information does not mandate any particular course of medical care. Further, the information is not intended to substitute for the independent professional judgment of the treating provider, as the information does not account for individual variation among patients. Recommendations reflect high, moderate, or low confidence that the recommendation reflects the net effect of a given course of action. The use of words like "must," "must not," "should," and "should not" indicates that a course of action is recommended or not recommended for either most or many patients, but there is latitude for the treating physician to select other courses of action in individual cases. In all cases, the selected course of action should be considered by the treating provider in the context of treating the individual patient. Use of the information is voluntary. ASCO provides this information on an "as is" basis and makes no warranty, express or implied, regarding the information. ASCO specifically disclaims any warranties of merchantability or fitness for a particular use or purpose. ASCO assumes no responsibility for any injury or damage to persons or property arising out of or related to any use of this information or for any errors or omissions. This is the most recent information as of the publication date. For the most recent information, and to submit new evidence, please visit www.asco.org/Brachytherapy-guideline and the ASCO Guidelines Wiki (www.asco.org/guidelineswiki). # THE BOTTOM LINE (CONTINUED) Additional Resources Additional information is available at www.asco.org/Brachytherapy-guideline and www.asco.org/guidelineswiki. Patient information is available at www.cancer.net. ASCO believes that cancer clinical trials are vital to inform medical decisions and improve cancer care, and that all patients should have the opportunity to participate jco.org of a qualified clinician. CCO makes no representations or guarantees of any kind whatsoever regarding the report content or its use or application and disclaims any responsibility for its use or application in any way. # Guideline and Conflicts of Interest The Update Panel (Appendix Table A1, online only) was assembled in accordance with ASCO's Conflict of Interest Management Procedures for Clinical Practice Guidelines ("Procedures," summarized at . asco.org/rwc). Members of the Panel completed ASCO's disclosure form, which requires disclosure of financial and other interests that are relevant to the subject matter of the guideline, including relationships with commercial entities that are reasonably likely to experience direct regulatory or commercial impact as a result of promulgation of the guideline. Categories for disclosure include employment; leadership; stock or other ownership; honoraria, consulting or advisory role; speaker's bureau; research funding; patents, royalties, other intellectual property; expert testimony; travel, accommodations, expenses; and other relationships. In accordance with the Procedures, the majority of the members of the Panel did not disclose any such relationships. Five 5,6,12-15 RCT reports were obtained for this targeted update. All five of these trials were randomized; three were available in fully published form, one 6 was available as both a fully published paper that reported on efficacy outcomes and an abstract 5 that reported on the toxicity outcomes, and one 15 was available in abstract form only. Four 5,12,13,15 of these trials were phase III, and the trial reported by Morton et al 14 was described as being a phase II trial. All efficacy outcomes for the four trials 5,12,13,15 that compared EBRT with brachytherapy are reported in Table 1; adverse effects are reported in Table 2. The earliest trial, the fully published phase III trial reported by Sathya et al 13 in 2005, randomly assigned 104 patients with T2 to 3 nonmetastatic prostate cancer to either EBRT (four-field box radiation to the prostate and seminal vesicles with a 2-cm margin at 66 Gy total dose given in 33 fractions over 6.5 weeks; n 5 53) or brachytherapy boost (iridium implant, 35 Gy over 48 hours) with EBRT (40 Gy, 20 fractions over 4 weeks; n 5 51). Use of concurrent ADT was not reported. Patients were stratified by age, prostate-specific antigen (PSA) levels, Gleason score, tumor stage (T2 v T3), and risk status (intermediate v high). The primary outcome for this trial was biochemical or clinical failure (which was defined as biochemical failure, clinical failure, or death resulting from prostate cancer), and a statistically significant benefit in favor of the EBRT with brachytherapy arm was detected (hazard ratio 0.42; P 5.0024) after a median reported follow-up of 98 months. 16 While the authors report that the treatment effect was greater in the intermediate-risk group compared with the high-risk group, the difference between the HR for biochemical or clinical failure was not significant; however, when adjusted for age, baseline PSA, Gleason score, and tumor stage, the treatment effect (brachytherapy boost v EBRT alone) was more pronounced (HR, 0.31; 95% CI, 0.17 to 0.58; P 5.0002). No differences in the toxicity profile between the two arms were detected. The second trial, the fully published phase III trial first reported by Hoskin et al 12 in 2007, randomly assigned 220 patients with histologically confirmed T1 to 3 prostate cancer, no evidence of metastases, PSA 50, no previous transurethral resection of the prostate, and fitness for general anesthesia to either EBRT (given as either three-field without shaped blocks or threedimensional volumetric planning and conformal three-field plans at 55 Gy given in 20 fractions; n 5 111) or EBRT (35.75 Gy, 13 fractions) with high-dose rate brachytherapy boost (HDR-B; iridium implant delivering a total of 17 Gy in two fractions over 24 hours; n 5 109). Patients were stratified according to tumor stage, PSA levels, and Gleason scores. Neoadjuvant and concurrent ADT (duration not specified) were used in 76% of patients. The primary outcome of interest was biochemical disease-free survival (bDFS), and a statistically significant benefit in favor of the EBRT/ HDR-B was detected (HR, 0.76; P 5.03) after a median follow-up of 30 months. This benefit in favor of the addition of brachytherapy to EBRT was also shown in the comparisons between the tumor stage, PSA levels, and Gleason score groups. No difference was detected in acute toxicity scores between the two arms for grade 2 or higher late bowel or bladder reactions. Health-related quality-of-life scores between the two arms detected a benefit in favor of EBRT/HDR-B (P 5.025) as assessed by the Functional Assessment of Cancer Therapy-Prostate instrument. A second report with a median follow-up of 85 months continued to show superiority of EBRT/HDR-B over EBRT (HR, 0.69; P 5.04). 17 The third trial, a phase III trial, reported in both fully published 6 and abstract form 5 by Morris et al, 5,6 randomly assigned 398 patients with intermediate-and high-risk prostate cancer to either DE-EBRT (whole pelvis EBRT: 46 Gy, 23 fractions followed by conformal EBRT to prostate: 32 Gy, 16 fractions; n 5 200) or LDR-B (whole pelvis EBRT: 46 Gy, 23 fractions followed by an 125 I boost to a minimum dose of 115 Gy to prostate; n 5 198). Patients were stratified by risk category (intermediate v high risk). Twelve months (8 months neoadjuvant, 2 months concurrent, 2 months adjuvant) of ADT was used in all patients. The primary outcome was bDFS as defined by biochemical criteria using the Phoenix (nadir 1 2 ng/mL) threshold. After a median follow-up of 78 months, a statistically significant benefit in favor of EBRT/LDR-B was detected (log-rank P.001). Multivariate analysis confirmed brachytherapy boost (but not risk category) as an independent predictor of bDFS. Assessment of the 5-year cumulative incidence of late grade 3 or higher toxicity detected a significant benefit in grade 3 genitourinary (GU) effects in favor of treatment with DE-EBRT (P.001) but not in grade 4 GU or in grade 3 or higher GI effects. 5 Quality of life was prospectively collected using the Short Form-36 instrument, which assessed physical function, role physical, bodily pain, general health, vitality, social functioning, and emotional and mental health. Additional items to gather data on urinary function, bowel function, and sexual function were added. All items were scored on a scale from 0 to 100. Baselines scores were balanced between treatment arms, but area under the curve differences were detected for bodily pain (P 5.04), general health (P 5.01), sexual function (P 5.02), and urinary function (P 5.006) in favor of treatment with DE-EBRT over LDR-PB. 18 No healthrelated quality-of-life differences were detected for any other domains. The fourth trial, a phase III trial reported in abstract form only by Prestidge et al 15 in 2016, randomly assigned 588 patients with low-intermediate risk prostate cancer (Gleason 6, PSA 10 to 20 ng/mL or Gleason 7, PSA 10 ng/mL) 1:1 to EBRT (45 Gy, 25 fractions mini-pelvis) with LDR boost (110 Gy 125 I or 100 Gy 103 Pd) or LDR alone (145 Gy or 125 Gy, respectively). The primary outcome was progression-free survival (PFS; American Society for Radiation Oncology nadir 1 2 biochemical failure, clinical failure, or death from any cause). After 6.7 years of median follow-up, the independent Data Monitoring Committee recommended releasing the data after the fifth interim analysis. There was no difference in 5-year PFS (85% v 86%; HR, 1.02; futility P.001). There were no differences in acute grade 3 or higher toxicity (8% in each arm) but worse grade 3 or higher late toxicity in the brachytherapy boost arm (12% v 7% overall; 7% v 3% for GU; 3% v 2% for GI; no P values provided). The fifth trial, a phase II trial reported by Morton et al 14 in 2016, randomly assigned 170 patients with low-and intermediaterisk prostate cancer to two HDR monotherapy regimens (ie, no EBRT was used). Patients received 27 Gy, two fractions over 1 week or 19 Gy in a single fraction. Eligible patients also had a prostate volume 60 mL, International Prostate Symptom Score # 18, no previous prostate surgery, and no use of ADT. The primary outcome was grade 2 or higher toxicity. After a median follow-up of 20 months, the only significant difference in acute toxicity detected between the treatment arms was for fatigue in favor of the single fraction (16% v 32%; P 5.029). There were no differences in acute grade 2 or higher GU toxicity, acute grade 2 or higher GI toxicity, or quality of life (measured by Expanded Prostate Index Composite every 6 months). For late toxicity, a difference was detected between treatment arms in favor of the single fraction for grade 2 erectile dysfunction (12% v 29%; P 5.025). No differences in late GI toxicity were reported. bDFS was not reported. There may be higher late GU toxicities associated with LDR-B compared with HDR-B. In the ASCENDE-RT 5 study, there were 20% grade 3 to 4 GU toxicities, half of which were due to urinary strictures requiring dilatation (S. Tyldesley, personal communication, November 2015). In the Hoskin et al study, 6% of patients in the HDR boost arm had urethral stricture at 5 years post-treatment. 17 In a widely used dose fractionation scheme internationally (HDR-B 15 Gy in one fraction followed by EBRT 37.5 Gy, 15 fractions), there was 0.8% grade 3 to 4 GU toxicity (0% strictures). 19 The differences in toxicity between these brachytherapy procedures may be due to relative dose sparing of the membranous urethra, which is associated with lower stricture rates. However, because cross-trial comparisons should only be used for hypothesis generation, we will have to await the results of randomized data. The Canadian Cancer Trials Group launched a national phase III LDR versus HDR monotherapy RCT in 2016 (clinicaltrials.gov NCT pending). # UPDATED RECOMMENDATIONS Please refer to the Bottom Line Box for the updated recommendations and the accompanying Qualifying Statements. †Comprising clinic visits every 6 months until 5 years (yearly thereafter) for prospective collection of patient-and physician-reported adverse effects, complications, and quality of life; PSA and testosterone levels measured every 6 months to assess predefined primary end point of PFS standard nadir 1 2 ng/mL (Phoenix) threshold. # COST CONSIDERATIONS ASCO recognizes that there is often a wide array of choices for treating many cancer types, with often a wide disparity in cost to patients and payers (despite much difference in effectiveness or toxicity). 20 Halpern et al 21 reported that of the radiation modalities used in the treatment of prostate cancer, from the Medicare payer perspective, LDR brachytherapy is the cheapest (compared with SABR, EBRT, or protons). Helou et al 22 showed that in the Canadian health care context, SABR had the higher quality-adjusted life-years and was more cost effective compared with LDR (and both were better than EBRT). Further work is needed to articulate cost, cost-effectiveness, and cost-utility differences between the various prostate cancer treatment approaches. # LIMITATIONS OF THE RESEARCH There are four 5,12,13,15 small-to medium-sized RCTs addressing the question of brachytherapy boost. The largest and most contemporaneously relevant (because the control arm used an external beam dose of 78 Gy in 39 fractions) trial (ASCENDE-RT, Morris et al 5,6 has published the survival outcomes; 6 however, the toxicity data are only available in abstract form. When the toxicity data are published, these guidelines recommendations will have to be revised if the toxicities are significantly higher than was presented to date. There are also insufficient data for comment on a meaningful difference in overall survival, because all trials were powered for PFS only. The guideline panel will re-evaluate the recommendations as new data emerge, especially from the ASCENDE-RT and Radiation Therapy Oncology Group 0232 trials. # AUTHORS' DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST Disclosures provided by the authors are available with this article at jco.org. We thank Maha Hussain, MD, Cynthia Anderson, MD, and the other members of the ASCO Clinical Practice Guidelines Committee as well as the Cancer Care Ontario Program in Evidence-Based Care Report Approval Panel for their thoughtful reviews and insightful comments on this guideline document. We also thank Rodney Breau, MD, MSc, W. Robert Lee, MD, MS, Med, and Ronald C. Chen, MD, MPH, for reviewing an earlier draft of this guideline. Finally, we thank the original authors of the Cancer Care Ontario guideline (George Rodrigues, Xiaomei Yao, Andrew Loblaw, Michael Brundage, and Joseph Chin) for their contribution to this effort. # AUTHOR CONTRIBUTIONS The following represents disclosure information provided by authors of this manuscript. All relationships are considered compensated. Relationships are self-held unless noted. I 5 Immediate Family Member, Inst 5 My Institution. Relationships may not relate to the subject matter of this manuscript. For more information about ASCO's conflict of interest policy, please refer to www.asco.org/rwc or ascopubs.org/jco/site/ifc. # Joseph Chin
In Canada in 2010, an estimated 2600 new cases of ovarian cancer will be diagnosed and, of those cases, 1750 women will die, making ovarian cancer the seventh most prevalent form of cancer in Canadian women and their fifth leading cause of cancer death (1). Women with ovarian cancer typically have subtle, non-specific symptoms such as abdominal pain, bloating, changes in bowel frequency, and urinary and/or pelvic symptoms (2), making early detection difficult. Thus, the majority of ovarian cancer cases are diagnosed at an advanced stage when the cancer has spread outside the pelvis (3). Due to the late diagnosis of this disease, the five-year relative survival ratio for ovarian cancer in Canada is only 40% (1). Unfortunately, due to the low-positive predictive values of potential screening tests (CA-125 and ultrasound), there is currently no screening strategy for ovarian cancer (4). Palpation using a bimanual pelvic examination or by radiological imaging (3) can identify an adnexal mass, which is defined as an enlarged lump near the uterus, usually in the ovary or fallopian tube. Adnexal masses include both benign (ovarian cysts, fibroids, and endometriomas) and malignant tumours. There are numerous methods that have been tested in the preoperative identification of adnexal masses suspicious for malignancy. These methods include CA-125, transvaginal and transabdominal ultrasound, MRI, CT scans and the risk of malignancy index (RMI). However, the most appropriate identification method has yet to be determined, mainly due to poor positive and negative predictive values associated with each test in differentiating a benign from a malignant mass. Once the diagnosis of ovarian cancer is confirmed, the patient must undergo surgical staging or debulking. However, which surgical staging and debulking procedure should be used to improve overall survival, progression-free survival, and quality of life in women with ovarian cancer is also unknown. The purpose of this document is to identify evidence that would inform optimal recommended protocols for the identification and surgical management of adnexal masses suspicious for malignancy. The evidence-based series (EBS) guidelines developed by the CCO PEBC use the methods of the Practice Guidelines Development Cycle (5). For this project, the core methodology used to develop the evidentiary base was an update of two previously published systematic reviews: the Agency for Healthcare Research and Quality (AHRQ) report, 2006 (3) and Australian Cancer Network (ACN) Clinical Practice Guideline, 2004 (6). Evidence was selected and reviewed by five members of the PEBC Gynecology Cancer DSG and one methodologist. The systematic review is a convenient and up-to-date source of the best available evidence on the management of an adnexal mass suspicious for malignancy. The body of evidence in this review is primarily comprised of prospective and retrospective cohort studies. That evidence forms the basis of the recommendations developed by the Gynecology Cancer DSG and published in Section 1. The systematic review and companion recommendations are intended to promote evidence-based practice in Ontario, Canada. The PEBC is supported by the Ontario Ministry of Health and Long-Term Care through Cancer Care Ontario. All work produced by the PEBC is editorially independent from its funding source. # Literature Search Strategy Environmental Scan As a first step, an internet search of Canadian and international health organizations and the National Guidelines Clearinghouse (see Appendix 1 for full list) was conducted for existing guidelines and systematic reviews relevant to our research question. Guidelines were included if they were published since 1999 in English. This initial environmental scan yielded 11 practice guidelines; however, one guideline was excluded because the full guideline was available only in French, and another guideline was excluded because only the National Guidelines Clearinghouse summary was available. One evidence report/technology assessment and one clinical practice guideline identified through this environmental scan were deemed to be the most appropriate to answer the guideline questions. The 2006 AHRQ report (3) addressed the identification of an adnexal mass suspicious for malignancy question. The 2004 ACN Clinical Practice Guideline (6) addressed the surgical management of an adnexal mass suspicious for malignancy question. # Update Literature Search Strategy The literature search from the AHRQ report was updated (Appendix 2) using MEDLINE (OVID: January 2004 through week 3, March 2009). In addition, as an exact search strategy for the Australian Cancer Network report was not available, an update of that literature search (Appendix 2) was approximated using the keywords provided in the report using MEDLINE (OVID: January 2004 through week 3, April 2009). This literature search combined diseasespecific terms ('pelvic mass,' 'adnexal mass,' 'pelvic neoplasms,' 'ovarian cancer,' 'ovarian neoplasm,' 'ovarian carcinoma,' 'epithelial ovarian cancer,' 'borderline ovarian tumours' and 'tumours of low malignant potential') with surgical specific terms ('intraoperative pathological examination,' 'frozen section,' 'debulking surgery,' 'fertility sparing,' 'surgical staging,' 'bilateral salpingo-oophorectomy,' 'total hysterectomy,' 'node or nodal dissection,' 'surgical management,' 'treatment,' 'cytoreduction,' 'secondary cytoreduction,' 'interval cytoreduction,' 'laparotomy,' and laparoscopy') for all study designs. Relevant articles and abstracts were selected and reviewed by two reviewers. The reference lists of included studies along with the personal reference lists of the guideline working group were searched for additional studies. # Study Selection Criteria Articles were eligible for inclusion in this systematic review if they were systematic reviews, meta-analyses, clinical practice guidelines, randomized trials, or comparative cohort studies. Studies indentified in the update of the AHRQ report literature search were included based on the same inclusion criteria put forth in the AHRQ report (3). For studies investigating single modality identification of an adnexal mass, the inclusion criteria were: 1) comparison of the test (e.g., bimanual pelvic exam or ultrasound, to histology or negative surgery 2) greater than 20 patients included in study 3) able to construct a 2-by-2 table, which compares the results of the diagnostic test with the definitive histological diagnosis. For studies investigating the use of multi-modality scoring systems (i.e., RMI), the inclusion criteria were: 1) patients with suspicion of cancer 2) studies with scoring, risk score, combined modality approach 3) assesses predictive value of two or more variables using multivariable model 4) greater than 50 patients included in study. Studies identified in the update of the Australian Cancer Network (6) guideline were based on the following selection criteria: 1) greater than 20 patients included in study 2) patients with an adnexal mass suspicious for early stage (I-II) malignancy, 3) two-armed (or greater) study design with a comparison of surgical procedures/techniques/approaches 4) report on at least one of the following outcomes: optimal surgery, overall survival, progression-free or disease-free survival, reduction in the number of surgeries, morbidity, adverse events, quality of life. # Synthesizing the Evidence A bivariate, random-effects meta-regression model was used to produce summary estimates of sensitivity and specificity and to plot summary receiver operating characteristic (ROC) curves with 95% confidence regions. This model, described in detail elsewhere (7)(8)(9), has several advantages over the standard summary ROC approach. Chief among these is the preservation of the two-dimensional nature of the data and the incorporation of any correlation that might exist between sensitivity and specificity (8). The model assumes that the logit sensitivities and specificities are normally distributed and makes use of the variance estimates to compute study weights (7). Heterogeneity in the current review was assessed visually. Given that between-study heterogeneity is widespread for measures of diagnostic accuracy (10), a random-effects model was used for all pooling. This bivariate, randomeffects model takes into account the difference in precision by which sensitivity and specificity have been measured within and across studies, and it incorporates and estimates the amount of between-study variability (8). Statistical analyses were executed with the statistical software package STATA version 11 (11) using the metandi command. The outcomes of the meta-analyses were plotted as summary ROC curves and can be seen in Figures 2A-D. The Gynecology Cancer DSG decided not to pool the surgical studies, but rather to present the results of each study individually in a descriptive fashion. # Quality Appraisal and Data Extraction The Appraisal of Guidelines Research and Evaluation (AGREE) tool (12) was used to evaluate the quality of identified evidence-based guidelines. While all scoring domains of the AGREE tool were considered in the evaluation of guidelines, the Rigour of Development domain, describing the rigour of systematic methods in identifying and evaluating evidence, was considered to be most relevant in application for this systematic review. Systematic reviews and meta-analyses were assessed for quality using the AMSTAR tool (13). The quality of primary studies included assessments for study design, type of data collection, sampling method, and blinding. # Updated Literature Search Results # Identification of an adnexal mass literature Four meta-analyses (14-17) and 67 primary studies (Table 1), pertaining to the identification of an adnexal mass suspicious for malignancy, met the inclusion criteria and are included in this review. # Surgical procedures literature A total of 1809 articles were identified in the updated search for the most appropriate surgical procedure, of which 16 met the inclusion criteria (85)(86)(87)(88)(89)(90)(91)(92)(93)(94)(95)(96)(97)(98)(99)(100). # Study Design and Quality The ACN evidence-based guideline (6) earned 80.2% for the Rigour of Development domain of the AGREE tool. Since the role of AGREE in the assessment of health technology assessments has not yet been formally evaluated, the AHRQ report was not rated with AGREE but with AMSTAR for systematic reviews. It received an overall quality score of 90%. The meta-analyses by Geomini et al (14), Liu et al (15), Medieros et al (17), and Geomini et al (16) received scores of 73%, 55%, 82%, and 82% respectively. Given that, at the time of inclusion, all patients either have ovarian cancer or they do not, diagnostic accuracy studies are, in principle, cross-sectional in nature (101). The diagnostic tests under study are intended to reduce the clinical uncertainty about their status (101). As such, the terms "cohort-type" or "case-control-type" accuracy studies have been proposed depending on the method of patient recruitment. Using such terminology, two casecontrol type studies and 65 cohort-type accuracy studies were included in this review (Figure 1). Data collection occurred prospectively in 64% of the studies, retrospectively in 33%, and not described in 3%. The sampling method was consecutive in 42% of the studies and not reported in the remaining 58%. Pathologists were blinded to the preoperative index test findings in only 4% of studies and not blinded in 1% of studies, with the remaining 94% of studies not describing any such blinding. In general, the quality of evidence for studies identified for the surgical question was poor. No meta-analyses or systematic reviews were identified. Only one of the 16 eligible surgical studies was an RCT (89), and another was designed as a "historically controlled trial" (95), defined as per the Cochrane Collaboration definition of study designs (102). The remaining 14 eligible studies were all retrospective in nature. As the quality of data collection in a retrospective study is often compromised, it is important to take the inherent limitations of such retrospective review designs into consideration. Furthermore, with the exception of 2 large SEER database reviews, the studies tended to be small and were likely underpowered to detect statistically significant differences in survival outcomes. # Outcomes 1. Identification of an adnexal mass suspicious for malignancy Ultrasonography is the most common diagnostic imaging technique for the noninvasive assessment of adnexal masses and is believed to be both a reliable and reproducible method for preoperative discrimination between malignant and benign pelvic masses (103)(104). # 2D versus 3D In addition to the four studies considered in the AHRQ report that analyzed both 2D and 3D ultrasonography, two new prospective studies were identified through the updated literature search (38,48). Analyzing the 2D ultrasonography data from the six studies using a bivariate random effects model yielded an overall sensitivity of 85.3% and specificity of 87.4% (Table 2, Figure 2A). When the data were pooled from the six 3D ultrasonography studies, the estimates improved to an overall sensitivity of 93.5% and specificity of 91.5% (Table 2, Figure 2A). Morphological scoring systems, based on parameters observed from gray-scale sonography, were developed to overcome limitations with operator subjectivity and tumour variability (3). These scoring systems assign and then sum up the score for established sonographic variables. A predetermined specific cutoff value classifies the mass as either malignant or benign. # Sassone Scoring System The model originally described by Sassone et al ( 105) is based on a weighted sum of the following four morphological features: inner wall structure, wall thickness, septa, and echogenicity. A score of greater than 9 is suggestive of malignancy. The AHRQ report identified and conducted a meta-analysis of 15 studies that explicitly used Sassone's criteria. The meta-analysis by Geomini et al ( 16) included 18 studies with cutoffs that ranged from 4 to 15, although 9 was the most common cutoff as in the original report. While our updated literature search did not identify any additional studies reporting on the Sassone et al model, we elected to conduct a meta-analysis of studies from both the AHRQ report and the Geomini et al meta-analysis that met our inclusion criteria, because neither report alone captured all the available data. A bivariate random effects model of these 22 studies generated a pooled sensitivity of 90.4% and specificity of 76.4%. Limiting our analysis to the 17 studies (42,62,67,71,(105)(106)(107)(108)(109)(110)(111)(112)(113)(114)(115)(116)(117) reporting a cutoff point of 9, yielded an overall sensitivity of 88.6% and specificity of 77.5% (Table 3, Figure 2B). # Lerner Scoring System Lerner et al (118) devised a scoring system based on a modification to the Sassone model. The Lerner model differs from Sassone's in several ways, including weighted point value assignments, fewer point values per variable studied, the deletion of one variable found not to be significant (wall thickness), and the inclusion of a new variable called shadowing. A cutoff of 3 was determined to best differentiate benign from malignant masses. The AHRQ report did not separately evaluate the Lerner model. Geomini et al (16) reported eight studies that evaluated the Lerner model, in addition to the original, and found a pooled sensitivity of 90% and specificity of 63% (Table 3). Our updated literature search identified one additional study by Lee et al (49) where 137 masses were evaluated in 123 women, and a sensitivity of 82.1% and specificity of 69.7% were reported. We do not believe that the addition of this one study would change the overall estimates reported by Geomini et al and, thus, did not rerun the analysis. # DePriest Scoring System The model described by DePriest et al (119) uses the weighted sum of cystic wall structure, volume, and septum structure. The score ranges from 0 to 12, with the most common cutoff set at 5. The AHRQ report identified six studies that made use of the DePriest scoring system. Geomini et al (16) evaluated 10 studies and reported a pooled sensitivity of 91% and specificity of 69% (Table 3). Our literature search identified an additional study (49) not included in either the AHRQ report or the meta-analysis by Geomini et al. In 123 women with 137 masses, Lee and colleagues reported a sensitivity of 92.9% and specificity of 59.6%. We do not believe the addition of this one study to the Geomini et al meta-analysis would have changed the overall estimates and, therefore, did not repeat the analysis. # Ferrazzi Scoring System Ferrazzi et al (109) developed their model to include four morphological features: wall structure, septa, vegetations and echogenicity. The weighted sum can range from 4 to 18, with 9 used as a cutoff to suggest malignancy. The AHRQ report identified three studies explicitly using the Ferrazzi et al criteria and the meta-analysis by Geomini et al (16) pooled seven studies. An updated search of the literature identified 1 additional study reporting on the Ferrazzi et al scoring system (58). Pooling the nine available, qualifying studies generated an overall sensitivity of 85.2% and specificity of 85.9% (Table 3, Figure 2B). # Finkler Scoring System Finkler et al (120) developed a 10-point scoring system, where scores of 7 or more indicate malignancy. The AHRQ report identified three studies that evaluated the Finkler system, as did Geomini et al (16). However, Geomini et al found that the summary estimates varied widely in the ROC curve. Despite not having identified any additional studies, we felt it was worthwhile to combine all available data in overall summary estimates. Our bivariate random-effects analysis of the five studies (120-124) allowed in our model generated an overall sensitivity of 83.5% and specificity of 78.2% (Table 3, Figure 2B). # Other Scoring Systems The meta-analysis by Geomini et al ( 16) also considered a model developed by Alcazar and Jurado that made use of the Sassone score as a variable in a logistic regression. Although the model was evaluated in four studies, there was too much variability in the sensitivity and specificity to pool the summary estimates. The AHRQ report identified 53 studies that evaluate ultrasonography in the assessment of adnexal mass morphology. These publications included unique, modified, or unclear scoring systems that did not fit into the other scoring system categories. While there was significant heterogeneity in the criteria used for diagnosis, the report did go ahead and pool sensitivity and specificity. The resulting summary estimates were 86% for sensitivity and 83% for specificity. # Explicit Scoring Systems The AHRQ report considered scoring systems that included data combined from the following categories: 1) imaging, including ultrasound, CT, and MRI; 2) patient risk factors, such as age and menopausal status; and 3) laboratory data, primarily CA-125. # Risk of Malignancy Index (RMI) The RMI, first published by Jacobs and colleagues (125), is based on scores from ultrasound (U), menopausal status (M), and CA-125 data in the following manner: RMI = U x M x CA-125. A cutoff of 200 was used to differentiate between malignant and benign masses in the original study. The AHRQ report (3) identified 11 studies in addition to the original that assessed the diagnostic performance of RMI. The meta-analysis by Geomini et al (16) identified 16 studies in which the RMI was evaluated but considered only the 13 studies that used the original cutoff of 200. An updated literature search identified two new studies (21,32) that considered the diagnostic accuracy of RMI. When a bivariate random effects model was used to pool the data from the 13 studies, with 15 data sets, employing a cutoff of 200 to be indicative of malignancy, the summary sensitivity and specificity were 79.2% and 91.7%, respectively (Table 4, Figure 2B). When the analysis was extended to the 23 studies considering a cutoff level of 50, the summary estimates were 82.1% for sensitivity and 87.8% for specificity. Several years after the RMI was published, improvements to the model were attempted. Tingulstad et al (126) proposed the RMI2, which gives new weights to the ultrasound and menopause components of the original model. The same cutoff level of 200 was recommended for the RMI2 model. The AHRQ report identified four studies in addition to the Tingulstand et al original, while Geomini et al ( 16) evaluated seven studies and, at a cutoff level of 200, reported pooled estimates for sensitivity and specificity of 79% and 81%, respectively (Table 4). The updated literature search indentified only one additional study evaluating the diagnostic accuracy of RMI2. In 194 women, Bensaid et al ( 21) reported a sensitivity of 92% (95% CI, 65 to 100%) and specificity of 80% (95% CI, 75 to 84%) at a cutoff of 125. Because data were not available at the recommended cutoff of 200, we did not repeat the analysis to include the data from Bensaid et al. In 1999, Tingulstad et al (127) suggested a further refinement to the two previous RMI models, again with suggested new weights given to the ultrasound and menopause components. Two hundred was once again the recommended cutoff level for this third model. Both the AHRQ report and the meta-analysis by Geomini et al identified only one additional study (128) that evaluated the RMI3 and reported that this validation study found a sensitivity of 74% and specificity of 91%. These results were very similar to the original RMI3 report, which found a sensitivity of 71% and specificity of 92%. We did not identify any additional literature evaluating RMI3. # Tailor's Model Tailor et al ( 129) developed a scoring system based on an artificial neural-network analysis that incorporates age, menopausal status, morphological features, and Doppler indices. The AHRQ report found four additional studies that evaluated this model but did not pool the data sets. They did note, however, that the subsequent studies each reported poorer performance than did the original publication. Geomini et al (16) found five publications that evaluated the original Tailor model. At a suggested cutoff of 50%, the pooled sensitivity and specificity of four data sets were 60% and 93%, respectively (Table 4). Our updated search of the literature did not identify any other studies that evaluated the diagnostic performance of the Tailor et al model. (130) in the late 1990s to predict the malignancy of an adnexal mass. Through complex modelling techniques, the ANN models investigate the possible existence of non-linear interactions or correlation between variables. The ANN1 model combines the predictive value of menopausal status, serum CA-125, presence/absence of papillary structures, and colour score, and uses a cutoff of 45%. The ANN2 model includes, in addition to menopausal status, serum CA-125, and papillary structures as in ANN1, the following set of morphological parameters: smoothness of internal walls, unilocularity, presence of ascites, and whether the mass is bilateral. A cutoff of 60% was used. Timmerman et al found the sensitivity and specificity of ANN1 to be 94% and 90%, respectively. ANN2 performed slightly better with a sensitivity of 96% and specificity of 94%. Two separate studies validated the models (77,115) and also reported a better performance with ANN2 over ANN1, although performance in subsequent replications was poorer. Mol et al (115) reported ANN2 to have a sensitivity and specificity of 90% and 46%, respectively, while Van Holsbeke et al ( 77) reported 98% and 34%, respectively. Our updated literature search did not identify any additional studies that validated the Timmerman ANN models. While other artificial neural networks have been developed over the years, external validation of these additional models is scarce, and they will not be discussed further. # Logistic Regression Models (LR) 1 and 2 The development of algorithms through statistical modeling that assess the probability of malignancy can also be used to distinguish malignant from benign masses preoperatively. One such logistic regression model (LR1) (131), developed in the late 1990s by Timmerman et al, included the following variables in the analysis: menopausal status, CA-125 level, presence of ≥1 papillary growth (>3mm in length), and a colour score indicative of tumour vascularity and blood flow. At a cutoff of 25%, the sensitivity of the LR1 was 95.9% and specificity was 87.1%. Two external validation studies (77,132) found lower estimates upon replication, where sensitivity was found to be 62% by Valentin et al and 78.1% by Van Holsbeke et al, and specificity was 79% in both studies. A second logistic regression model (LR2) developed by the Timmerman group incorporates the same variables as ANN2 (i.e., menopausal status, serum CA-125, papillary structures, smoothness of internal walls, unilocularity, presence of ascites, and whether the mass is bilateral). Timmerman et al (130) reported, at a cutoff of 60%, a sensitivity and specificity of 95.9% and 85.5%, respectively. Again, at external validation, estimates were considerably lower (sensitivity 90% and 84% and specificity 86% and 75% in Mol et al (115) and Van Holsbeke et al (77), respectively). A number of other logistic regression prediction models are found in the literature, but many are without external validation and will not be considered further in this document. # Doppler Sonography Regular gray-scale sonography can be enhanced with Doppler measurements, which assess the direction of blood flow and its relative velocity. Colour Doppler imaging and pulsed Doppler spectral analysis enable evaluation of ovarian tumour blood flow, analysis of the distribution of blood vessels, and quantitative measurement of blood-flow velocity waveforms. These parameters increase the sensitivity and specificity of ultrasound evaluation of ovarian tumours (133). # 2D Power Doppler (2D PD) The updated search of the literature identified five qualifying studies that evaluated 2D PD technology in the preoperative discrimination between malignant and benign adnexal masses. The sensitivity in these studies ranged from 49% to 100%, while specificity ranged from 74% to 100% (Table 5). # 3D Power Doppler (3D PD) The evaluation of 3D PD technology in the differential assessment of adnexal masses was considered in four qualifying studies identified through the updated literature search. The sensitivity in these studies ranged from 68% to 100%, while specificity ranged from 40% to 98% (Table 6). # Resistance Index (RI) Resistance Index (RI), the difference between peak systolic and maximum enddiastolic flow velocity divided by peak systolic flow velocity, is one of the most common flow criteria in colour Doppler scanning. The AHRQ report identified 32 articles that evaluated RI, although one study (134) actually considered a morphology index, not RI, and was excluded from further analyses. Our updated literature search identified five new studies (41,48,(51)(52)55) that assessed the diagnostic accuracy of Doppler scanning using the RI parameter, but the data from Marret et al (52) was not included because this same data was previously published and already included. The analysis of the summary estimates for these four new studies along with the existing qualifying literature included 42 data sets and yielded an overall sensitivity of 77.2% and specificity of 89.8% (Table 7, Figure 2C). # Pulsatility Index Pulsatility Index (PI) is defined as the difference between peak systolic and enddiastolic flow velocity, divided by the time-averaged flow velocity. The AHRQ report analyzed the PI from 20 studies, while the updated literature search identified only one additional study (51) looking at PI that met the inclusion criteria. When we analysed the 21 studies with 22 data sets, using a bivariate random effects model, we obtained an overall sensitivity of 80.6% and specificity of 79.9% (Table 7, Figure 2C). # Peak Systolic Velocity The peak or maximum systolic velocity (PSV) is the maximum flow recorded in any visualized artery and, along with RI and PI, it is one of the most common flow criteria. Our analysis of the six studies included in the AHRQ report plus one additional identified paper (55) yielded an overall sensitivity of 80.0% and specificity of 84.2% (Table 7, Figure 2C). The AHRQ report considered Doppler studies that did not measure or calculate waveforms but, rather, looked at the presence of vascularity within the mass or a direct count of the vessels observed. The report identified 10 such studies with sensitivities that ranged from 77% to 100% and specificities that ranged from 30% to 94%. Pooling these studies resulted in an overall sensitivity of 88% and specificity of 78% (Table 7). Our updated literature search did not identify any additional studies investigating Doppler visualization. # Combined Morphology and Doppler 2D Ultrasonography plus Doppler A combination of morphological and vascular imaging was developed to try to improve the differentiation of malignant and benign adnexal masses. The AHRQ report identified nine studies that described such a combined modality. An additional three studies were identified through our updated search of the literature. When a bivariate random effect model was used to analyses all 12 studies, an overall sensitivity of 91.0% (95% CI, 84.8% to 94.8%) and specificity of 91.7% (95% CI, 81.1% to 96.6%) was obtained. (Figure 2C). # 3D Ultrasonography plus Doppler The updated search of the literature identified two qualifying studies that evaluated 3D ultrasonography plus Doppler technology in the differential assessment of adnexal masses (Table 8). # Other Imaging Modalities MRI The AHRQ report identified 15 articles investigating the performance of MRI in the diagnosis of adnexal masses. An updated search of the literature since the AHRQ report was published identified a meta-analysis ( 15) evaluating, among other things, the MRI modality. Liu et al considered 10 studies reporting 13 data. In addition to this meta-analysis, six primary studies, not included in the Liu et al meta-analysis or the AHRQ report, were also identified during the updated literature search. Again, using a bivariate random effects model, we conducted a meta-analysis of the 15 studies from the AHRQ report, the one study exclusive to the Liu et al meta-analysis that qualified for the current analysis, and the six additional primary studies identified. This analysis of 22 studies with 24 data sets yielded an overall sensitivity of 91.9% and a specificity of 88.4% (Table 9, Figure 2A). The AHRQ report described three studies that looked at the performance of CT in diagnosing adnexal masses. Liu et al (15) evaluated a total of four studies, two of which also appeared in the AHRQ report and two that did not meet the AHRQ inclusion criteria. Our updated search identified an additional four studies evaluating CT in the diagnosis of adnexal masses. A bivariate random effects analysis of these four studies, along with the three studies (with four data sets) considered in the AHRQ report yielded an overall sensitivity of 87.2% and specificity of 84.0% (Table 9, Figure 2A). Three studies investigating positron emission tomography (PET) were evaluated in the AHRQ report. Our updated search identified two additional studies, both investigating the accuracy of the combined PET/CT modality. Since results for PET alone were not available, no further PET analyses were undertaken. The AHRQ report found a pooled sensitivity of 67% and specificity of 79% in the diagnosis of adnexal masses with PET technology (Table 9). # Serum Marker CA-125 The AHRQ report identified 66 studies that investigated the use of CA-125 serum marker in the evaluation of an adnexal mass. An updated literature search identified 11 new studies, with 14 data sets, investigating the use of CA-125 in the diagnosis of an adnexal mass that met our inclusion criteria. In keeping with the AHRQ report and the most commonly used threshold of 35 U/mL, we conducted a meta-analysis of the 51 studies, with 52 data sets, that used a threshold of 35 U/mL. Eight of these studies (20,(31)(32)(33)53,57,58,135) were published since the AHRQ report. The addition of these newer studies to the analysis did little to change the AHRQ summary estimates. We calculated an overall sensitivity of 78.7% and a specificity of 77.9% (Table 10, Figure 2D). Since many of the studies reported a threshold other than 35 U/mL, we re-ran the analysis using 50 U/mL as the cutoff point. The analysis included 66 studies with 72 data sets, including 10 studies published since the AHRQ report. This yielded an overall sensitivity of 79.0% and specificity of 78.3%. # Frozen Section While considered an intraoperative assessment rather than a preoperative one, frozensection diagnosis can help to guide further surgical management of ovarian tumours. Accordingly, the accuracy of this technique is of great consequence and, as such, this method of diagnosing a suspicious adnexal mass was deemed valuable for consideration in this report. The updated search of the literature identified two systematic reviews and 15 primary studies (19,(23)(24)(25)35,37,39,43,56,61,65,66,79,81,82) published in or since 2004 that considered the diagnostic accuracy of frozen-section diagnosis and were not included in either systematic review. All 15 studies were retrospective cohort-like in design, with seven reporting the selection of consecutive patients. Only one study reported the blinding of pathologists from the final histopathologic diagnosis when interpreting results of frozen sections. One study specifically reported that pathologists were not blinded, and the remaining 13 studies made no reference to blinding status. Geomini et al considered the accuracy of frozen-section diagnosis in a 2005 systematic review (14). The literature search ranged from 1966 to mid-2003, with 18 primary studies qualifying for inclusion. When borderline tumours were classified as malignant, the sensitivity of frozen-section diagnosis ranged from 65% to 97% and the specificity between 97% and 100%. Classifying borderline tumours as benign resulted in a sensitivity of 71% to 100% and a specificity of 98.3% to 100%. Medeiros and colleagues (17) also conducted a systematic review and meta-analysis on the accuracy of frozen-section analysis that included 14 primary studies. The literature search period in this review ranged from 1984 to the end of 2003. The pooled sensitivity and specificity of frozen-section diagnosis distinguishing between benign and borderline or malignant ovarian tumours was 99% (95% CI, 89 to 99%) and 88% (95% CI, 86 to 90%), respectively. In our analysis of the 15 primary frozen section studies published since 2004, borderline tumours were considered malignant and counted as such in the 2x2 tables. Furthermore, any deferred cases reported were excluded from the 2x2 tables and, consequently, from the analysis. A bivariate random effects analysis of these 15 studies yielded an overall sensitivity of 89.2% (95% CI, 86.3 to 91.5%) and specificity of 97.9% (95% CI, 96.6 to 98.7%) (Table 10, Figure 2D). # ACOG/SGO Referral Guidelines The American College of Obstetricians and Gynecologists (ACOG) and the Society of Gynecologic Oncologists (SGO) jointly published guidelines for the referral of women with pelvic masses that are suspicious for ovarian cancer to gynecologic oncologists (136). The referral guidelines are based on patient age, CA-125 level (>200 U/mL for premenopausal, >35 U/mL for postmenopausal), physical findings, imaging results, and a family history of breast or ovarian cancer in a first-degree relative. The referral guidelines were validated, and their role in distinguishing benign from malignant masses was tested in a multicentre setting with 1035 patients (44). In premenopausal women, the sensitivity and specificity of the referral guidelines in differentiating benign and malignant masses was 70% and 69%, respectively. Sensitivity was # D. CA125 and Frozen Section SROC Curves Sensitivity enhanced in postmenopausal women, where the guideline correctly identified 94% of ovarian cancer patients. Specificity in this group was reported to be 58%. The predictive value of the referral guidelines was further evaluated in a prospective cohort study of 837 consecutive patients (30). In premenopausal women, the sensitivity and specificity were 79.2% and 69.8%, respectively. In postmenopausal women, sensitivity was found to be 93.2% and specificity was 59.9%. In considering early-versus late-staged disease, the referral guidelines performed better in terms of sensitivity in late-staged disease, especially in postmenopausal women where sensitivity reached 98.3%. In premenopausal women, the referral guidelines were 92.3% sensitive in distinguishing malignant from benign cases. # Surgical procedures for an adnexal mass suspicious for malignancy The Australian Cancer Network (ACN) 2004 guideline ( 6) on the management of women with epithelial ovarian cancer made recommendations on, among other things, the most appropriate surgical approach to take in such patients. Eight studies were included in the surgery for pelvic mass section of the guideline and are not discussed further here. An updated search of the literature identified an additional 16 studies (Table 11) published since the 2004 ACN guideline that met our inclusion criteria. A randomized controlled trial (RCT) ( 89) of patients undergoing a systematic aortic and pelvic lymphadenectomy versus those undergoing lymph node sampling reported no statistically significant difference in five-year progression-free (p=0.16) or overall survival (p=0.56). However, this RCT was underpowered to detect an effect of systematic lymphadenectomy on survival. The sample size calculation in this study was undertaken to detect a difference in prevalence of lymph node positivity, the study's primary outcome. The targeted sample size required to detect an effect of lymphadenectomy on survival, the secondary outcome, was deemed unattainable by the researchers. Despite the reduced power to detect a statistical difference in the secondary outcomes, the study reported a trend favouring lymphadenectomy in terms of progression-free (HR, 0.72; 95% CI, 0.46 to 1.14) and overall (HR, 0.85; 95% CI, 0.49 to 1.47) survival. Eight additional studies (85-88,90-93) investigated the survival impact of comprehensive surgical staging in women diagnosed with early-stage ovarian cancer. In two large population-based studies (85)(86), consisting exclusively of over 6600 early-stage epithelial ovarian cancer patients, Chan et al found that surgical staging with lymphadenectomy was associated with improved three-year (p<0.001) (86) and five-year disease-specific survival (p<0.001) (85) compared to staging procedures without lymphadenectomy. Similarly, Oksefjell et al reported a statistically significant improvement in 5-year overall survival rates in patients undergoing a lymphadenectomy versus those that did not (87% vs. 64%, respectively; p=0.02). Survival analyses performed by both Skirnisdottir et al ( 91) and Hornung et al ( 88) also demonstrated a statistically significant benefit in disease-free survival (p=0.004 and p=0.0007, respectively) for patients undergoing a lymphadenectomy versus those that did not. Hornung and colleagues also considered overall survival and reported a statistically significant difference (p=0.0008) in the two patient groups in favour of the lymphadenectomy group. Conversely, in 205 patients with pTI-IIb clear cell carcinoma, Suzuki et al (92) found that patients who underwent systemic lymphadenectomy did not show a significant improvement in disease-free (p=0.353) or overall survival (p=0.645) compared to those that did not. Similarly, Cho et al (87) report no significant difference between groups in progression-free and overall survival rates in patients with stage I mucinous epithelial ovarian tumours undergoing complete staging versus those whose staging was incomplete. In an attempt to determine the benefit of surgically staging ovarian low malignant potential (LMP) tumours, Wong and colleagues (93) retrospectively reviewed the records of 247 patients with tumours of borderline malignancy and found no statistically significant difference in rates of recurrence or mortality between patients surgically staged and those who were unstaged. Six studies (94-99, 137) compared laparoscopy versus laparotomy for the surgical management of women with apparent early ovarian cancer or borderline tumours. Patient sample sizes in these studies ranged from 34 to 360. In the three studies (95-97) that considered patients with early epithelial ovarian cancer (EOC), no statistical difference in survival rates was detected between patients undergoing a laparoscopy versus laparotomy. Similarly, in the management of patients with early borderline ovarian tumours, Romagnolo et al (98), Park et al (99), and Desfeux et al (94) found that a laparoscopic versus laparotomic surgical approach did not appear to influence survival rates, although Desfeux et al acknowledged that the number of events was too small to allow for proper statistical testing. Fertility-preserving treatments are often desirable for women of reproductive age who are diagnosed with borderline ovarian tumours (BOT). Two studies compared the impact of conservative fertility-sparing surgeries versus more extensive surgical approaches. Yinon et al (100) specifically compared rates of recurrence in 40 patients who underwent unilateral salpingo-oophorectomy versus 22 patients who underwent cystectomy only. No statistical difference in recurrence rates was detected (27.5% vs. 22.7%, p=0.8). Similarly, in a larger study of 360 women with BOT, Park et al (99) found no difference in disease-free survival between patients that underwent radical or fertility-sparing surgery (p=0.651). # Adverse Events While the surgical technique did not appear to impact patient survival, there were differences detected in surgical outcomes and complication rates. Cho et al (87) found a statistically significant difference in complication rates, with 12.9% experiencing a complication in the completely staged group versus 1.0% in those incompletely staged (p<0.001). Ghezzi et al (95) reported a statistically significant difference in the rates of minor postoperative complications, with 6.7% of patients in the laparoscopy group experiencing such an event compared to 42.1% of patients in the laparotomy group (p=0.047). Romagnolo et al (98) reported a difference in the cases of tumour rupture or spilling during surgery, with 34.6% ruptures recorded in the laparoscopic group compared to 6.6% in patients undergoing a laparotomy (p<0.0001). Similarly, Lecuru et al (96) found 31% of laparoscopic patients experienced intraoperative tumour rupture versus 16% in the laparotomy group. However, this difference did not reach statistically significance. In patients with borderline tumours, the difference in the occurrence of intraoperative tumour rupture was found not to be statistically associated with the surgical approach according to Desfeux et al. ( 94) The basic diagnostic work-up of patients with a suspicious adnexal mass involves a gynecological exam, ultrasound imaging and testing of serum tumour markers. While this approach is often sufficient in detecting advanced disease, the diagnosis of early-staged ovarian cancer is more challenging. In an attempt to determine the best method for the identification and diagnosis of a suspicious adnexal mass, we systematically reviewed existing guidelines and the literature. The intention and necessity of preoperative diagnosis of an adnexal mass is to both triage the patient and better define the surgical options. Recent Canadian guidelines on the evaluation and referral of ovarian masses (138) recommend that patients with a high-level risk of underlying malignancy be reviewed in consultation with a gynecologic oncologist. Indeed, studies considering the impact of physician's specialty on the survival of patients with early stage ovarian cancer have shown a trend towards improved survival when gynecologic oncologists perform the surgery (139)(140). Along with triaging appropriate patients to subspecialists, the preoperative diagnosis of an adnexal mass provides the anatomic details necessary to inform the surgical options. For patients with clearly benign masses, or likely benign masses with health issues, observation only is the usual management. Such preoperative diagnostic information can also help to distinguish patients who require extensive surgery over those for whom a more conservative surgical approach is adequate. Statistical comparisons between diagnostic techniques could not be performed in this review. Instead, the assessment of various modalities in differentiating benign from malignant masses is based on inspection of the summary data obtained from the meta-analyses conducted. These results suggest that 3D ultrasonography has both a higher sensitivity and specificity when compared to 2D ultrasound. Established morphological scoring systems also performed with respectable sensitivity and specificity, each system with equivalent diagnostic competence. Explicit scoring systems did not appear to perform as well as other diagnostic testing methods. Not only was external validation of the models lacking in many cases, where results were available, but performance was often poorer upon subsequent replication. Assessment of an adnexal mass by colour Doppler technology, using the RI, PI, and PSV indices, was neither as sensitive nor as specific as simple ultrasonography. Furthermore, because of overlap of vascular parameters between malignant and benign masses, a firm diagnosis based on Doppler evaluation alone can be problematic (141). Summary estimates from studies considering combined morphology and Doppler assessment were higher than the estimates for either modality alone. Both sensitivity and specificity of this combined approach were high. Of the three imaging modalities considered, MRI appeared to perform the best, although results were not statistically different from CT as determined by overlapping confidence intervals. PET did not appear to perform as well as either MRI or CT, although only three studies were considered in the analysis. The measurement of the CA-125 tumour marker appears to be less reliable than other available assessment methods; however, results were not stratified by menopausal status. It is widely reported that differences exist in CA-125 levels between premenopausal and postmenopausal women, even with the same histological diagnosis (75). Finally, frozen-section analysis has both a high sensitivity and an especially high specificity in the assessment of adnexal masses. In the treatment of ovarian cancer, the importance of surgical management is universally recognized. Clearly, complete surgical resection is required to improve a patient's prognosis. However, it is not clear how aggressive a surgical approach is necessary in earlystaged ovarian cancer. The included evidence suggests that systematic lymphadenectomy improves survival, as does proper surgical staging. There is an exception to this benefit, however, with tumours of low malignant potential. In such tumours, conservative fertilitypreserving surgical approaches appear to have no adverse survival effect (99). The accuracy and adequacy of surgical staging by laparotomy or laparoscopic approaches appears to be comparable, with neither approach conferring a survival advantage compared to the other. In spite of this, many surgeons are unwilling to perform laparoscopic surgical staging in early-staged ovarian cancer due to the potential risks of intraoperative tumour rupture, port-site metastasis, and dissemination of the tumour (97). Intraoperative tumour rupture was indeed reported to occur more frequently in patients undergoing laparoscopy versus laparotomy in two retrospective cohort studies (96,98). However, unequivocal prospective comparative data supporting the existence of an increased occurrence of intraoperative tumour rupture in ovarian cancer patients managed by laparoscopy is still lacking (95). The evidence included in this systematic review is not without significant limitations. In the identification of an adnexal mass literature, the population of patients often included women with a suspicious mass undergoing surgery. This inclusion of operative patients could escalate the prevalence of malignancy in comparison to what would be expected in a primary care population of women presenting with a suspicious adnexal mass. This has implications for the generalizability of the results, and it can inflate the test's sensitivity. Furthermore, the included studies did not often allow for stratification by menopausal status. Given the lower likelihood of ovarian malignancy in premenopausal patients, the accuracy of a diagnostic test can be reduced in a sample that includes a high proportion of premenopausal patients. In addition to these methodological issues, the quality assessment of the included literature revealed several further shortcomings, especially in study design and reporting. Blinding is a crucial issue in diagnostic accuracy studies, as it is necessary to prevent information bias (142). While the index test always preceded surgery in the included studies, thereby by nature blinding the outcome, the reference test should also be interpreted without knowledge of the index test results (142). The vast majority of studies considered in the identification of a suspicious adnexal mass did not report such blinding. One third of the diagnostic studies had data collection that occurred retrospectively or was not reported. Retrospective study designs are inherently more prone to bias than are prospective studies and can be more difficult to interpret, especially if the sampling did not include consecutive patients. The reporting of consecutive patient sampling occurred in less than half the included studies. Many of the surgical studies were also prone to biases inherent in their retrospective designs. Moreover, the small sample of included patients meant many studies were underpowered to detect statistically significant results. There is an obvious need for improvement in the quality of primary research in this area. Ideally, future research would consist of randomized clinical trials with a non-inferiority design where survival outcomes are considered. Despite these limitations, the best available evidence with respect to the questions posed was collected and included. A rigorous systematic review and meta-analysis, planned a priori, provided an abundant evidentiary base and the context and direction for the development of recommendations. There are numerous methodologies that have been considered in the preoperative identification of adnexal masses suspicious for malignancy. Results suggest that 3D ultrasonography has both a higher sensitivity and specificity when compared to 2D ultrasound. Established morphological scoring systems also performed with respectable sensitivity and specificity, each morphological system with equivalent diagnostic competence. Explicit scoring systems did not perform as well as other diagnostic testing methods. Assessment of an adnexal mass by colour Doppler technology, using the RI, PI, and PSV indices, was neither as sensitive nor as specific as simple ultrasonography. Summary estimates from studies considering combined morphology and Doppler assessment were higher than the estimates for either modality alone. Both sensitivity and specificity of this combined approach were high. Of the three imaging modalities considered, MRI appeared to perform the best, although results were not statistically different from CT. PET did not perform as well as either MRI or CT. The measurement of the CA-125 tumour marker appears to be less reliable than do other available assessment methods. Finally, frozen section analysis has both a high sensitivity and especially high specificity in the assessment of adnexal masses. The evidence suggests that systematic lymphadenectomy and proper surgical staging improve survival. Conservative fertility-preserving surgical approaches are an acceptable option in women with low malignant potential tumours. The accuracy and adequacy of surgical staging by laparotomy or laparoscopic approaches appears to be comparable, with neither approach conferring a survival advantage. # CONFLICT OF INTEREST No conflicts of interest were declared for J Dodge, A Covens, C Lacchetti, L Elit, T Le, M Devries-Aboud, or M Fung Kee Fung. # JOURNAL REFERENCE The following systematic reviews and meta-analysis has been published in Gynecologic Oncology (© 2012 Elsevier Inc.;/): The Gynecology Cancer Disease Site Group would like to thank J Dodge, A Covens, C Lacchetti, L Elit, T Le, M Devries-Aboud, and M Fung Kee Fung for taking the lead in drafting this systematic review with meta-analysis. The PEBC supports a network of disease-specific panels, termed Disease Site Groups (DSGs), as well as other groups or panels called together for a specific topic, all mandated to develop the PEBC products. These panels are comprised of clinicians, other health care providers and decision makers, methodologists, and community representatives from across the province. The PEBC is well known for producing evidence-based guidelines, known as Evidencebased Series (EBS) reports, using the methods of the Practice Guidelines Development Cycle (1,2). The EBS report consists of an evidentiary base (typically a systematic review), an interpretation of and consensus agreement on that evidence by our Groups or Panels, the resulting recommendations, and an external review by Ontario clinicians and other stakeholders in the province for whom the topic is relevant. The PEBC has a formal standardized process to ensure the currency of each document, through the periodic review and evaluation of the scientific literature and, where appropriate, the integration of that literature with the original guideline information. # The Evidence-Based Series Each EBS is comprised of three sections: - Section 1: Guideline Recommendations. Contains the clinical recommendations derived from a systematic review of the clinical and scientific literature and its interpretation by the Group or Panel involved and a formalized external review in Ontario by review participants. # DEVELOPMENT OF THIS EVIDENCE-BASED SERIES Development and Internal Review This EBS was developed by the Gynecology DSG of the CCO PEBC. The series is a convenient and up-to-date source of the best available evidence on management of a suspicious adnexal mass, developed through review of the evidentiary base, evidence synthesis, and input from external review participants in Ontario. # Report Approval Panel Prior to the submission of this EBS draft report for external review, the report was reviewed and approved by the PEBC Report Approval Panel, which consists of two members, including an oncologist, with expertise in clinical and methodology issues. The key issues raised by the Report Approval Panel and the modifications made by the Gynecology DSG are below with the DSG responses bulleted below the comments of the panel members: - If pathology is still the gold standard, as suggested in your last qualifying statement, what is the role of the other diagnostic technologies? o The need to have clarity in the preoperative diagnosis is to 1) triage the patients to a specialist, and 2) better define the surgical options (e.g., conservative vs. radical). We have now added a paragraph to the discussion regarding the intention and necessity of preoperative diagnosis of an adnexal mass. - The authors' first recommendation, and the one for which the most detailed analysis exists, concludes that 3D US, CT and MRI are "all recommended" with considerations of more "local factors" then suggested as determinants of the modality of choice. The authors should reconsider whether they have missed an opportunity to make a more definitive recommendation that accounts for the "equality" in diagnostic efficacy and what can be reasonably assumed about cost, access, harm (e.g., radiation exposure) and patient inconvenience. o We thank the reviewer for his comments. We have taken this opportunity to add further recommendations based on the Working Group's expert consensus opinion that takes into consideration availability, access, and harm. - The authors consider various diagnostic tools separately (e.g., imaging, CA-125). Is there a risk that, in practice, these modalities are used in combination and in doing so, diagnostic properties are changed? Related to this theme, are there important differences in the eligibility of patients included in any analysis of a single modality, in which a second modality criterion was required for inclusion? The bolded text was removed from the recommendations section, and it was further delineated that the recommendations were based on expert opinion. Furthermore, the recommendation on PET was moved to indicate that the role of PET in cervical cancer has not been prospectively evaluated. o In many circumstances, we have evaluated the combination of diagnostic tests such as ultrasound and CA-125 - In contrast to the diagnostic efficacy section, the section that deals with "therapy" (What is the most appropriate surgical procedure for a woman who presents with an adnexal mass suspicious for malignancy?) does not include conventional guideline methodology or reporting. As this question is a largely a therapeutic question, standard methodologies would move down a hierarchy of acceptable guidelines, metaanalysis of randomized controlled trials (RCTs), RCTs, non-randomized comparisons, etc. The authors do not provide background in their Methods section about their approach to this question (including literature search criteria). The Results section largely is divided into positive vs. negative studies as opposed to a systematic presentation of the literature. As a result, the juxtaposition of data and conclusions is associated with ambiguity and comes across as potentially representing "expert opinion" as opposed to an evidenced-based recommendation. The authors should reconsider their approach to this question. o We thank the reviewer for these comments. We did indeed use standard methodologies in the development of this section of the guideline but acknowledge that this was not clear in the reporting. As such, we have explained more explicitly in the Methods section what was done. Furthermore, we rearranged the reporting of the results so that the one RCT is discussed first. - The authors might wish to clarify whether the post-diagnostic therapeutic pathway includes multiple modalities that require systematic review in order to assess linkage. Again, while this comment may reflect lack of content expertise, if these patients should be considered for adjuvant therapy using another modality, does the use of that subsequent modality interact with the choice of the surgical approach? In addressing the surgical approach, and especially if best therapy involves sequential use of multiple modalities, the authors may need to carefully dissect out the prognostic importance of patients being able to undergo a more extensive surgical procedure vs. the therapeutic benefit of that intervention. o There is indeed a link between surgery and adjuvant therapy. However, the Working Group viewed these two interventions separately, both in terms of prognosis and as therapeutic approaches. As such, a discussion on adjuvant therapy in patients with a suspicious adnexal mass was deemed beyond the scope of this document. # External Review by Ontario Clinicians and Other Experts The PEBC external review process is two-pronged and includes a targeted peer review that is intended to obtain direct feedback on the draft report from a small number of specified content experts and a professional consultation that is intended to facilitate dissemination of the final guidance report to Ontario practitioners. Following the review and discussion of Section 1: Recommendations and Section 2: Evidentiary Base of this EBS and the review and approval of the report by the PEBC Report Approval Panel, the Gynecology Cancer DSG circulated Sections 1 and 2 to external review participants for review and feedback. Box 1 summarizes the draft recommendations and supporting evidence developed by the working group. # BOX 1: DRAFT RECOMMENDATIONS (approved for external review 2011-04-08) QUESTIONS - What is the optimal strategy for preoperative identification of the adnexal mass suspicious for ovarian cancer? 2. What is the most appropriate surgical procedure for a woman who presents with an adnexal mass suspicious for ovarian cancer? # TARGET POPULATION The target population of this guideline is adult women presenting with a suspicious adnexal mass, either symptomatic or asymptomatic. # INTENDED USERS This guideline is targeted for clinicians managing the care of women with a suspicious adnexal mass, specifically general gynecologists and gynecological oncologists. # RECOMMENDATIONS Identification of an Adnexal Mass Suspicious for Ovarian Cancer ➢ Sonography, particularly three-dimensional (3D) sonography, magnetic resonance imaging (MRI), and computerized tomography (CT) imaging are each recommended for differentiating malignant from benign ovarian masses. However, the working group offers the following further recommendations, based on their expert consensus opinion and the consideration of availability, access, and harm: - Transvaginal sonography should be the first modality of choice, where technically feasible, in patients with a suspicious, isolated ovarian mass. # - MRI is the most appropriate test to help clarify the malignant potential in patients where ultrasound may be unreliable. - CT is most useful in cases where extra ovarian disease is suspected or needs to be ruled out. # Key Evidence - The diagnostic performance of each diagnostic technology was compared and contrasted based on the summary data on sensitivity and specificity obtained from the meta-analysis. - A meta-analysis of six cohort studies that investigated 3D sonography (1-6) indicated an enhanced sensitivity of 93.5% and specificity of 91.5% with 3D technology (Section 2, Figure 2A). - A meta-analysis of 22 cohort studies with 24 data sets that investigated the effectiveness of MRI in the diagnosis of adnexal masses (7-28) found an overall sensitivity of 91.9% and specificity of 88.4% (Section 2, Figure 2A). - A meta-analysis of seven studies with eight data sets considering CT technology (2,10,12,14,22,29,30) yielded an overall sensitivity of 87.2% and specificity of 84.0% (Section 2, Figure 2A). # ➢ Evaluation of an adnexal mass by Doppler technology alone is not recommended. Doppler technology should be combined with a morphological assessment. # Key Evidence - This recommendation is based on the results of several meta-analyses on Doppler indices but not direct comparisons between them. Rather, the summary data from these meta-analyses were inspected, and reasonable sensitivities and specificities were noted. - A meta-analysis of the resistance index (RI) included 35 cohort studies (2,5,17, with 42 data sets and yielded an overall sensitivity of 77.2% and specificity of 89.8% (Section 2, Figure 2C). - A meta-analysis of 21 cohort studies with 22 data sets that evaluated the Pulsatility Index (PI) found an overall sensitivity of 80.6% and specificity of 79.9% (Section 2, Figure 2C). - A meta-analysis of the peak systolic velocity (PSV) included seven cohort studies (32,33,37,42,50,51,62) and found an overall sensitivity of 80.0% and specificity of 84.2% (Section 2, Figure 2C). # Qualifying Statement - Assessment of an adnexal mass by colour Doppler technology, using the RI, PI and PSV indices, was neither as sensitive nor specific as simple ultrasonography. Furthermore, because of the overlap of vascular parameters between malignant and benign masses, a firm diagnosis based on Doppler evaluation alone can be problematic. ➢ Morphological scoring systems can be used to differentiate benign from malignant adnexal masses. The choice between the scoring systems should be made based on demonstrated accuracy and clinician preference. # Key Evidence - The diagnostic performance of each morphological scoring system was compared and contrasted based on the summary data on sensitivity and specificity obtained from the meta-analysis. - A meta-analysis of 17 cohort studies (32,33,46,52,57,63-74) assessing the Sassone model at a cutoff of 9, found an overall sensitivity of 88.6% and specificity of 77.5% (Section 2, Figure 2B). - A meta-analysis of nine studies (33,35,63,65,70,(75)(76)(77)(78) considering the Ferrazzi scoring system reported an overall sensitivity and specificity of 85.2% and 85.9%, respectively (Section 2, Figure 2B). - A meta-analysis of five studies (36,(79)(80)(81)(82) measuring the performance of the Finkler scoring system found an overall sensitivity of 83.5% and specificity of 78.2% (Section 2, Figure 2B). - A meta-analysis of data from 13 RMI studies (70,(83)(84)(85)(86)(87)(88)(89)(90)(91)(92)(93)(94), with 15 data sets, employing a cutoff of 200 to be indicative of malignancy, reported the summary sensitivity and specificity were 79.2% and 91.7%, respectively (Section 2, Figure 2B). ➢ As a stand-alone modality, serum CA-125 is not recommended for distinguishing between benign and malignant adnexal masses. # Key Evidence - This recommendation is based on a meta-analysis of 49 2D). # Qualifying Statement - CA-125 values are of limited use in premenopausal women and elevated in only 50% of early-stage ovarian cancers. Caution should be used in interpreting values in such patients. ➢ Frozen section for the intraoperative diagnosis of a suspicious adnexal mass is recommended in settings where availability and patient preferences allow. # Key Evidence - This recommendation is based on a meta-analysis of frozen section diagnoses that included 15 cohort studies (7,(132)(133)(134)(135)(136)(137)(138)(139)(140)(141)(142)(143)(144)(145) and yielded an overall sensitivity of 89.2% and specificity of 97.9% (Section 2, Figure 2D). # Surgical Procedures for an Adnexal Mass Suspicious for Malignancy ➢ Comprehensive surgical staging with lymphadenectomy is recommended for the surgical management of patients with early-stage ovarian cancer to improve survival. # Key Evidence # QUALIFYING STATEMENTS The Gynecology Cancer Disease Site Group (DSG) acknowledges that, despite definitions and criteria, it is unrealistic to expect that 100% of ovarian cancers will be identified as suspicious preoperatively. Pathology remains the gold standard. Targeted Peer Review: During the guideline development process, two targeted peer reviewers from Ontario and one from the USA considered to be clinical and/or methodological experts on the topic were identified by the working group. Several weeks prior to completion of the draft report, the nominees were contacted by email and asked to serve as reviewers. Three reviewers agreed, and the draft report and a questionnaire were sent via email for their review. The questionnaire consisted of items evaluating the methods, results, and interpretive summary used to inform the draft recommendations and whether the draft recommendations should be approved as a guideline. Written comments were invited. The questionnaire and draft document were sent out on April 8, 2011. Follow-up reminders were sent at two weeks (email) and at four weeks (telephone call). Professional Consultation: Feedback was obtained through a brief online survey of health care professionals who are the intended users of the guideline. Gynecologists and gynecologic oncologists in the PEBC database were contacted by email to inform them of the survey. Participants were asked to rate the overall quality of the guideline (Section 1) and whether they would use and/or recommend it. Written comments were invited. Participants were contacted by email and directed to the survey website where they were provided with access to the survey, the guideline recommendations (Section 1) and the evidentiary base (Section 2). The notification email was sent on April 13, 2011. The consultation period ended on June 10, 2011. The working group reviewed the results of the survey. Targeted Peer Review: one reviewer out of the three that were invited provided a response. This reviewer's responses are summarized in Table 1. No significant barriers or enablers were reported. # Summary of Written Comments The reviewer advised that references by L. Cohen and A. Fleischer be added to the evidence base. The authors were not able to gather more information from this reviewer regarding exactly which publications had been missed. The authors examined whether references by Cohen and/or Fleischer had been considered at any time during the guideline development process. Cohen (2001) was considered by the AHRQ review and reported in Section 2 under "Other Scoring Systems". An additional paper by Cohen and Fleischer was excluded by the AHRQ. One Fleischer paper was included in the evidence base for the guideline. In the end, no modifications to the evidence base were made on the basis of this reviewer comment. Professional Consultation: Sixty responses were received. Key results of the feedback survey are summarized in Table 2. 3. I would recommend this guideline for use in practice. 0(0) 0(0) 7( 13) 22( 39) 31( 55) # What are the barriers or enablers to the implementation of this guideline report? Practitioners listed barriers to the implementation of this guideline and suggested ways that uptake could be enhanced or enabled. The feedback included the following: Some practitioners mentioned that lack of awareness of the guideline would be a barrier to uptake. They recommended making primary care physicians aware of the guideline, and one reviewer suggested presentation of the guideline at group rounds to make the entire team aware of current literature and recommendations. Many practitioners reported that lack of resources would be a barrier to implementation of the guideline, especially in smaller communities. Specific resources that were noted as lacking included: - Gynecologic oncologists (e.g., for staging using lymphadectomy). - MRI only available in larger communities and delays with MRI booking. - CT access. - Patients may not always have easy access to an ultrasound department and quality of reporting is variable. - Pathologists with experience in gynecologic pathology for frozen section diagnoses. - Regional Minimally Invasive Surgery programs in the community. In order to enable the implementation of the guideline, practitioners also suggested: - A summary table outlining the sensitivities and specificities of the various diagnostic procedures. - Guidance for referral of patients to a gynecologist or gynecologic oncologist for nongynecologist clinicians who identify an adnexal mass. - Educating radiologists about appropriate use of ultrasound. - Boundaries to indicate whether the risk for a specific woman is low or high after being investigated. - A recommendations regarding CA-125 use for post-op women. - A summary table outlining the sensitivities and specificities of the various diagnostic procedures is presented in Section 2 of the report. - Referral criteria are considered to be outside of the scope of this review and guideline. For more information on referral criteria, readers may refer to a guideline published in 2009 by the Society of Obstetricians and Gynaecologists of Canada: Initial evaluation and referral guidelines for management of pelvic/ovarian masses (138). - Cutoffs for the scoring systems included in the report were added to Appendix 2 in Section 1. - CA-125 use post-op is outside of the scope of this guideline. In the event that the reviewer meant that we should clarify the recommendation regarding CA-125 use as a stand-alone modality in postmenopausal women, we amended the CA-125 recommendation so that it no longer refers only to premenopausal women, but to all women, as there is no evidence for its efficacy as a stand-alone modality for any age group. # Summary of Written Comments Twenty of the sixty responders to professional consultation provided additional written comments. The majority indicated that the document was of high quality and would be of use to practitioners. Suggestions for improvements or additions to the document included: - Mention of specific suspicious features may prove useful for the radiologist colleague accessing this guideline. 2. There is clear operative requirement of staging and lymphadectomy, but the role of tertiary referral is not clear. 3. Several comments were made relating the scoring systems described in the report. The feedback generally indicated that there are many practitioners in the province who are not aware of these scoring systems. A direct link from the recommendations to the scoring systems was requested. It was also suggested that the guideline recommend one scoring system that would be the most reliable. Other comments related to scoring systems include: a. Many practitioners are using the Risk of Malignancy Index (RMI) and it should be available as an appendix to this guideline. Also related to RMI: i. Recommendation on the use of RMI II and how the general gynecologist should use it for guidance when considering referral to a tertiary care centre. ii. Reference to RMI cutoff of 200 instead of 400 which is what is commonly used. 4. There was also a request for an appendix for ultrasound features of malignancy and definitions of resistance index (RI) pulsatility index (PI), and Peak Systologic (sic) Velocity (PSV). - As specific suspicious features are described in Section 2 of the report, no changes were made based on this comment. 2. The topic of referral is beyond the scope of this guideline and was therefore not considered for inclusion. As mentioned above, readers may refer to the Society of Obstetricians and Gynaecologists of Canada guideline Initial evaluation and referral guidelines for management of pelvic/ovarian masses (138). 3. Response to comments related to the scoring systems: a. An appendix was added to Section 1 with descriptions of the Sassone, DePriest, Ferrazzi, Lerner, Finkler, and RMI scoring systems. References to the original publications were also added to Section 2 of the report. Because the sensitivity and specificity are comparable for these scoring systems, it is not possible to recommend one of them as most reliable. Therefore, the wording of the recommendation was adjusted to make this clear and to indicate that the choice of scoring system should be based on clinician preference. b. The recommendation was also reworded to state that choice of version of the RMI should be based on clinician preference as they all have comparable sensitivity and specificity. c. The cutoff value of 200 for the RMI was based on the cutoff used in the initial report by Jacobs et al. The sensitivity and specificity values reported here are based on this cutoff. This was also reported in the AHRQ report as the most common cutoff value, therefore, the group decided to not to include reference to a cutoff of 400. d. Other definitions such as RI, PI, and PSV can be found in Section 2 of the document. This EBS report reflects the integration of feedback obtained through the external review process with final approval given by the Gynecology Cancer Disease Site Group, and the Report Approval Panel of the PEBC. Updates of the report will be conducted as new evidence informing the question of interest emerges. In September 2015, this document was assessed in accordance with the PEBC Document Assessment and Review Protocol and was determined to require a review. As part of the review, a PEBC methodologist conducted an updated search of the literature. A clinical expert (TL) reviewed and interpreted the new eligible evidence and proposed the existing recommendations could be endorsed. The Gynecology Cancer Disease Site Group (DSG) endorsed the recommendations found in Section 1 (Clinical Practice Guideline) on September 9, 2016. # DOCUMENT ASSESSMENT AND REVIEW RESULTS # Questions Considered - What is the optimal strategy for preoperative identification of the adnexal mass suspicious for ovarian cancer? - What is the most appropriate surgical procedure for a woman who presents with an adnexal mass suspicious for ovarian cancer # Literature Search and New Evidence The new search (January 2011 to January 2016) yielded 9386 papers. Three hundred and eighty six were retained for full text review. After the full text review, 93 studies were retained. Brief results of these searches are shown in the Document Review Tool. # Impact on Guidelines and Its Recommendations The new data supports existing recommendations. Hence, the Gynecology Cancer DSG ENDORSED the 2011 recommendations on management of a suspicious adnexal mass. A new qualifyling statement was added to the recommendation on ultrasound-based morphological scoring systems. "Ultrasound diagnostic criteria using a set of simple rules to distinguish between benign and malignant masses, and the IOTA (International Ovarian Tumour Analysis) predictive adnexal model had been extensively studied with acceptable sensitivity and specificity. This can serve as potential alternative diagnostic strategy to the RMI score." The following paragraphs are the justification for the inclusion of this qualifying statement.. Inconclusive result can be classified using subjective assessment by an experienced ultrasound operator. This strategy has been recently validated in 4848 patients with adnexal masses against histopathologic diagnosis with a sensitivity of 99.7% and specificity of 33.7%. In addition, the IOTA group also constructed an Assessment of Different Neoplasias in the adneXa (ADNEX) model. This is a risk prediction model to differentiate between benign, borderline tumours, stage I invasive, stage II-IV invasive ovarian cancer and secondary metastatic cancer. The ADNEX model consists of three clinical predictors and six ultrasound predictors. The clinical predictors are age (years), serum CA-125 (U/mL) and type of center to which the patient has been referred for ultrasound examination. Type of center was divided into oncology centers versus other hospitals. The ultrasound predictors are the maximal diameter of the adnexal mass (mm), proportion of solid tissue (%), number of papillary projections (0, 1, 2, 3, > 3), presence of more than 10 cyst locules (yes/no), acoustic shadows (yes/no), and presence of ascites (yes/no). The model performed well in differentiating between benign and malignant masses with the area under the receiver operating characteristic curves (AUC) of the ADNEX model of 0.954 (95% confidence interval 0.947 to 0.961) on the development data and 0.943 (0.934 to 0.952) on the validation data. The sensitivity for diagnosis of malignancy was 96.5% and specificity was 71.3% on the validation data set. 3. What is the optimal strategy for preoperative identification of the adnexal mass suspicious for ovarian cancer? 4. What is the most appropriate surgical procedure for a woman who presents with an adnexal mass suspicious for ovarian cancer? The target population of this guideline is adult women presenting with a suspicious adnexal mass, either symptomatic or asymptomatic. # Study Section Criteria: Articles were eligible for inclusion in this systematic review if they were systematic reviews, meta-analyses, clinical practice guidelines, randomized trials, or comparative cohort studies. Studies identified in the update of the AHRQ report literature search were included based on the same inclusion criteria put forth in the AHRQ report (3). For studies investigating single modality identification of an adnexal mass, the inclusion criteria were: 1) comparison of the test (e.g., bimanual pelvic exam or ultrasound, to histology or negative surgery 2) greater than 20 patients included in study 3) able to construct a 2-by-2 table, which compares the results of the diagnostic test with the definitive histological diagnosis. For studies investigating the use of multi-modality scoring systems (i.e., RMI), the inclusion criteria were: 1) patients with suspicion of cancer 2) studies with scoring, risk score, combined modality approach 3) assesses predictive value of two or more variables using multivariable model 4) greater than 50 patients included in study. Studies identified in the update of the Australian Cancer Network (6) guideline were based on the following selection criteria: 1) greater than 20 patients included in study 2) patients with an adnexal mass suspicious for early stage (I-II) malignancy, 3) two-armed (or greater) study design with a comparison of surgical procedures/techniques/approaches 4) report on at least one of the following outcomes: optimal surgery, overall survival, progression-free or disease-free survival, reduction in the number of surgeries, morbidity, adverse events, quality of life. The SUVmax of malignant tumors was significantly higher than that of benign and borderline lesions (mean, 7.8, 1.7, 2.4; P < 0.05). Among malignant tumors, SUVmax was significantly lower in mucinous adenocarcinomas compared with nonmucinous malignant tumors (mean, 3.3, 8.4; P < 0.05) and lower in clear cell adenocarcinomas compared with other subtypes of nonmucinous malignant tumors (mean, 6.0, 9.4; P < 0.05). The SUVmax cutoff that best differentiated malignant lesions from benign/borderline lesions was 2.4 for mucinous and 4.0 for nonmucinous tumors. These cutoffs correctly classified lesions as malignant or not in 88 The frequency of invasive malignancy was 10% in small tumors, 19% in medium-sized tumors and 40% in large tumors; 11% of the large tumors were borderline tumors vs 3% and 4%, respectively, of the small and medium-sized tumors. The type of benign histology also differed among the three subgroups. For all methods, sensitivity with regard to malignancy was lowest in small tumors (56-84% vs 67-93% in medium-sized tumors and 74-95% in large tumors) while specificity was lowest in large tumors (60-87%vs 83-95% in medium-sized tumors and 83-96% in small tumors. The DOR and the AUC value were highest in medium-sized tumors and the AUC was largest in tumors with a largest diameter of 7-11 cm. # Author Study type Comparison N Results Frozen sections were compared with final paraffin sections 27 cases had a diagnostic discrepancy. # Rakhshan 2009 Not stated Frozen section diagnosis of adnexal masses was compared with permanent section diagnosis as the gold standard. The overall accuracy of frozen section diagnosis was 95.7%. The sensitivity of frozen section diagnosis for benign, borderline and malignant lesions was 99, 60, and 92%, respectively. The tumor size in discrepant cases was larger than the concordant cases, however no association between mucinous histology and inaccurate diagnosis was found. # OUTCOMES DEFINITIONS - ARCHIVED -An archived document is a document that will no longer be tracked or updated but may still be useful for academic or other informational purposes. The document is moved to a separate section of the Web site and each page is watermarked with the phrase "ARCHIVED". - ENDORSED -An endorsed document is a document that the DSG/GDG has reviewed for currency and relevance and determined to be still useful as guidance for clinical decision making. A document may be endorsed because the DSG/GDG feels the current recommendations and evidence are sufficient, or it may be endorsed after a literature search uncovers no evidence that would alter the recommendations in any important way. - DELAY -A Delay means that there is reason to believe new, important evidence will be released within the next year that should be considered before taking further action. - UPDATE -An Update means that the DSG/GDG recognizes that there is new evidence that makes changes to the existing recommendations in the guideline necessary but these changes are more involved and significant than can be accomplished through the Document Assessment and Review process. The DSG/GDG will rewrite the guideline at the earliest opportunity to reflect this new evidence. Until that time, the document will still be available as its existing recommendations are still of some use in clinical decision making.
To determine the risk of cutaneous malignant melanoma (herein referred to as melanoma) associated with use of indoor tanning devices, including impact of age at first use and frequency of use on the relative risk of developing melanoma. # TARGET POPULATION All users of indoor tanning beds are the target population of this guideline. # INTENDED USERS This guideline is intended for use by clinicians, other health care providers, users and potential users of indoor tanning devices in Ontario. # RECOMMENDATIONS, KEY EVIDENCE, AND JUSTIFICATION # RECOMMENDATION 1 Use of indoor tanning devices should be avoided to reduce risk of melanoma. # Summary of Key Evidence for Recommendation 1 A systematic review with meta-analysis (1) based on pooling of 27 cohort and casecontrol studies found a significant association between ever use of indoor tanning devices and increased risk of developing melanoma (relative risk 1.25: 95% confidence interval 1.09-1.43; p<0.05). # Justification for Recommendation 1 There is strong evidence linking the use of indoor tanning devices to an increased risk of developing melanoma. Although the meta-analysis (1) lacked detail on some elements of interest for the included studies, the current systematic review of the literature verified the clinical homogeneity of the pooled studies. The Use of Indoor Tanning Devices Guideline Development Group (GDG) believes that the current evidence informs a strong recommendation. # Qualifying Statements for Recommendation 1 The International Agency for Research on Cancer (IARC) recently declared solar ultraviolet radiation (UVR) from indoor tanning devices a carcinogen (2). Both UVA and UVB have been shown to cause direct DNA damage through production of DNA mutations (UVA at a lower level than UVB), as well as indirect DNA damage via production of reactive oxygen species. Although UVB radiation can initiate the production of vitamin D in the skin, there are no data to support that artificial UVR is superior to oral supplementation with vitamin D to increase serum levels of this vitamin. Given the significant risk of melanoma as a consequence of using tanning devices, the GDG concludes that risks that arise from the use of tanning devices far outweigh any perceived benefit to their use. This systematic review evaluated studies from 2000 to present with the goal of capturing the impact of modern tanning beds, which have been designed to more accurately mimic UVR. However, the identified meta-analysis conducted by Boniol et al (1) included studies published from 1981 through 2012 and evaluated an older generation of tanning beds. It is hypothesized that future studies assessing the impact of modern tanning beds could potentially amplify the effects found in the current review. # RECOMMENDATION 2 All individuals should avoid use of indoor tanning devices, especially those at a younger age. # Summary of Key Evidence for Recommendation 2 A recent and comprehensive systematic review with meta-analysis (1) found an increased risk of melanoma in those who initiated tanning devices use at a younger age (RR, 1.59: 95%CI, 1.36-1.85; p<0.05). Data were pooled from 13 studies, 12 of which adjusted for confounders related to sun exposure and sun sensitivity. # Justification for Recommendation 2 Both the rate of tanning device use in youths, as well as the incidence of melanoma diagnosis in 15 to 34 year olds has been increasing. Moreover, the meta-analysis by Boniol et al (1) demonstrated that the younger a person starts using indoor tanning devices, the higher the risk of developing melanoma in their lifetime. These are concerning statistics, and the GDG concludes that the current evidence informs a strong recommendation. # Qualifying Statements for Recommendation 2 Based on the evidence, the GDG has not set an age cut-off for "younger age." The identified meta-analysis defined young age as under age 35 (1). However, not all the studies included in the analysis defined an age for younger age; in those that did, younger age was defined as anywhere from 18 to 35. In the three included case-control studies that found an increased risk of melanoma with a definitive age cut-off, younger age was defined as less than 25 years (3), less than 35 years (4) and less than 18 years (5). The GDG concludes that these data point to an association between tanning bed use and increased risk of developing melanoma at any younger age of first use: defining a specific age cut-off would only be speculative and would not add to the recommendation. # RECOMMENDATION 3 There is no safe lower limit of exposure to artificial UVR from indoor tanning devices. # Summary of Key Evidence for Recommendation 3 When evaluating the risk associated with frequent use of indoor tanning devices, both number of sessions and length of tanning sessions were considered. The meta-analysis conducted by Boniol et al (1) found a 1.8% increased risk of developing melanoma for each additional session of tanning device use per year (95%CI, 0.0-3.8%; p<0.05). Additionally, when Boniol et al ( 1) conducted an analysis of 14 studies that reported relative risks with frequent tanning bed use, they found a 42% increased risk of developing melanoma with high tanning bed use (RR, 1.42; 95%CI, 1.15-1.74; p<0.05). One additional case-control study (6), which was not included in the Boniol et al meta-analysis (1), similarly found an association between increased risk of melanoma and both the number of sessions and length of sessions (p=0.04). # Justification for Recommendation 3 Based on the association between ever use of indoor tanning devices and increased risk of developing melanoma, plus the greater risk associated with frequent use of indoor tanning devices, the evidence indicates that there is no safe lower limit of exposure to artificial UVR from indoor tanning devices. Updating All PEBC documents are maintained and updated through an annual assessment and subsequent review process. This is described in the PEBC Document Assessment and Review Protocol, available on the CCO website at:
# RECOMMENDATIONS, KEY EVIDENCE, AND JUSTIFICATION RECOMMENDATION 1 DNMT3A mutation testing should be included as a biomarker test in cytogenetically normal AML patients. # Summary of Key Evidence for Recommendation Four (8,10,13,15) of the eight studies (8)(9)(10)(11)(12)(13)(14)(15) included in the systematic review reported a statistically significant difference in Overall Survival (OS) between DNMT3A wild type (DNMT3A-wt) and DNMT3A-mutated (DNMT3A-mut) populations favouring the non-mutated gene; the remaining four that did not provide statistical data did demonstrate a similar trend. The meta-analysis of these data resulted in an overall estimated hazard ratio 1.66 (95%CI, 1.23-2.24; p=0.0010) favouring patients that are DNMT3A wild type. This strongly suggests that the mutational status of DNMT3A has prognostic value. However, the interpretation of the meta-analysis using OS may be limited by the fact that two studies were omitted as complete information was not available. # Justification for Recommendation 1 The available evidence shows DNMT3A mutation status has good prognostic value in this patient population. # Qualifying Statements for Recommendation 1 This recommendation is based on evidence currently available. Despite the heterogeneous nature of the studies included, the likelihood of having a series of large homogenous studies done in this patient population is low due to the nature of the disease and its management. # FUTURE RESEARCH Currently, there are no ongoing trials on DNMT3A mutation status in this patient population. The PEBC is a provincial initiative of Cancer Care Ontario supported by the Ontario Ministry of Health and Long-Term Care. All work produced by the PEBC is editorially independent from the Ontario Ministry of Health and Long-Term Care. Updating All PEBC documents are maintained and updated as described in the PEBC Document Assessment and Review Protocol.
# TARGET POPULATION All adult MM patients considered for treatment that includes blood or marrow transplantation. # RECOMMENDATIONS AND SUPPORTING EVIDENCE Autologous SCT is the recommended treatment option for patients with newly diagnosed MM, as part of the initial treatment plan. # Supporting evidence Evidence included in four Clinical Practice Guidelines (CPGs) (1)(2)(3)(4) suggests that a single transplant with autologous SCT should be offered to all MM patients who are free from severe co-morbidities and who are younger than 65 years of age, following the initial treatment with high-dose chemotherapy. The SCT Steering Committee acknowledged that, while the evidence upon which the CPGs were based almost uniformly excluded patients older than age 65, there was no reason not to offer SCT to patients 65 or older who have good performance status and no co-morbidities that would be a contraindication to transplantation. The SCT Steering Committee also acknowledges that ongoing trials are investigating the value of upfront treatment with combinations of novel agents and the deferring of transplantation to a later date. # Tandem (double) autologous SCT is an option for patients with MM who respond to the first autologous transplant with less than a very good partial response, but not progressive disease. # Supporting evidence Evidence included in two CPGs (1,2) suggests that double autologous SCT should be offered to patients who did not achieve a complete remission after their initial autologous SCT. While the CPGs recommended a second transplant be offered to patients who did not achieve complete remission following their first transplant, the SCT Steering Committee acknowledged that offering a second transplant to patients who achieve less than a very good partial response is reasonable as long as there is no progressive disease. # Allogeneic transplantation is an option for patients with high-risk MM preferably within the context of an investigative study. # Supporting evidence Evidence from three of the CPGs did not support the use of allogeneic SCT from HLA-matched related donors as a primary treatment, but the conclusion is that it may be offered to patients <50 years of age who are not expected to benefit from autologous SCT (e.g., chromosome 13 deletion) within the investigative setting only (1,2,4). # Repeat autologous transplantation is an option for patients with MM who relapse after a long remission (> 2 years) to a single autologous transplant. Supporting evidence Despite a lack of good quality evidence, the SCT Steering Committee's consensus opinion is that patients who relapse after a long remission following a single transplant should be offered a second transplant. # QUALIFYING STATEMENT The patient selection process and the ultimate decision to perform an SCT should take into account not only disease-related characteristics, but also comorbidities and patient preferences. Evidence on the role of SCT in the management of MM is emerging rapidly. This topic is also the subject of Program in Evidence-based Care (PEBC) Evidence-based Series (EBS) 6-6, which will be updated to incorporate new data. EBS 6-6 differs from this report in that it includes only evidence comparing high-dose chemotherapy and SCT in patients with MM, whereas this report includes comparisons of all interventions including SCT such as radiotherapy and other treatment modalities. # FUTURE RESEARCH Future research in this setting should continue to explore novel chemotherapy and supportive therapy options along with SCT. Better management of co-morbidities may allow clinicians to offer SCT to patients currently not eligible for treatment. # IMPLICATIONS FOR POLICY Transplantation for myeloma remains the most frequent indication in Ontario for autologous transplantation. As of this report, and in the foreseeable future, it is highly unlikely that the indication for transplant in such patients will change. With the use of more effective induction regimens, it is possible that more patients will be eligible for transplant with myeloma. # RELATED PROGRAM IN EVIDENCE-BASED CARE REPORTS The PEBC is a provincial initiative of Cancer Care Ontario supported by the Ontario Ministry of Health and Long-Term Care through Cancer Care Ontario. All work produced by the PEBC is editorially independent from its funding source. Copyright This report is copyrighted by Cancer Care Ontario; the report and the illustrations herein may not be reproduced without the express written permission of Cancer Care Ontario. Cancer Care Ontario reserves the right at any time, and at its sole discretion, to change or revoke this authorization.
# GUIDELINE OBJECTIVES # INTENDED USERS This guideline is intended for provincial policy makers, primary care physicians, nurse practitioners, radiologists, respirologists, thoracic surgeons, thoracic oncologists, and any health professionals involved with patients who may be at risk for developing lung cancer. # RECOMMENDATIONS, KEY EVIDENCE, AND JUSTIFICATION Screening High-risk Populations for Lung Cancer: The Working Group is in favour of screening high-risk individuals for lung cancer with LDCT. The primary evidence base for this proposal is the NLST, a large (>50,000 participants) RCT that compared LDCT screening with CXR and showed a 20% decrease in death from lung cancer in high-risk persons (4). The primary benefit associated with LDCT screening is a statistically significant reduction in mortality, both lung cancer specific and all cause. LDCT can identify smaller nodules than can CXR and thus can detect lung cancer at an earlier stage when a cure is more possible. Under current circumstances, most lung cancer patients are diagnosed at an advanced stage, and lung cancer accounts for more than a quarter of all cancer deaths (6). LDCT screening is not without risk. CT scanning, with its acquisition of multiple images, exposes an individual to a greater radiation dose than does CXR and may place patients at increased risk of lung and breast cancer. Based on models from official bodies and commissioned studies of estimates of harm from radiation, Bach et al estimate using the NLST data that one cancer death may be caused by radiation from imaging per 2500 persons screened (5). The serial CT scans required as part of a screening program necessitate judicious and efficient use of the technology with strict rules pertaining to quality control and training. The information obtained from a CT scan of the chest provides more precise visualization of lung nodules leading to a higher rate of detection of lung nodules. Although the majority of these nodules (>90%) will be benign, the detection of these nodules may lead to further imaging and follow-up that can involve invasive diagnostic procedures and possibly to harmful and unnecessary treatment. Completely addressing the clinical and cost implications of this high false-positive rate is critical and remains a challenge. In the interim, the Working Group endorses a strict application of screening to only a high-risk targeted population. In general, the recommendations below reflect the parameters of the NLST (4). Where there are deviations from those parameters, we provide justification. While there are still ongoing trials comparing LDCT with usual care, none are as large (and therefore as statistically powerful) as the NLST, and it is unlikely that another trial the size of the NLST will be undertaken. Some aspects of the ongoing trials may affect the recommendations once their results are known, and we have qualified our recommendations to acknowledge these uncertainties. # RECOMMENDATIONS AND SPECIFIC EVIDENCE Main Recommendation Recommendation 1: Screening for lung cancer with LDCT is recommended in high-risk populations defined as persons 55 to 74 years of age with a minimum smoking history of ≥30 pack-years who currently smoke or have quit within the past 15 years and are disease free at the time of screening. Pack-years = number of cigarette packs smoked per day x the number of years smoked. # Key Evidence - Among the studies in the collaborative review, the age for initiation of screening ranged from 47 to 60 years in the RCTs and from 40 to 60 years in the single-arm studies (5). - The upper age for screening ranged from 69 to 80 years in the RCTs and 73 to 85 years in the single-arm studies (5). The NLST initiated screening in persons ≥55 years of age and stopped at age 75 years (4). - The minimum smoking history in the RCTs ranged from ≥15 to ≥30 pack-years, and in the single-arm studies from ≥10 to ≥20 pack-years (5). The NLST enrolled persons with a smoking history of ≥30 pack-years and former smokers who had quit within the previous 15 years (4). - Seven RCTs reported previous cancer history in the eligibility criteria, stipulating a minimum numbers of years disease free since a previous cancer diagnosis. These ranged from 5 years to an indefinite period with variations for different types of cancers. Among 11 single-arm studies, this criterion was described as a minimum of 5 years since a previous cancer diagnosis, any previous lung cancer, any known pulmonary metastases, and any previous cancer diagnosis (5). In the NLST, exclusion criteria were a previous diagnosis of lung cancer, a previous diagnosis of other cancer within the previous 5 years, chest CT scan within 18 months before enrollment, haemoptysis, or unexplained weight loss >6.8 kg in the preceding year (4). - There is no evidence to support a specific age to initiate screening, a specific age to cease screening, or a specific screening-frequency interval. The highest quality and most compelling evidence is from the NLST. As such, the parameters used in this trial were endorsed by the Working Group as clinically reasonable. Patient acceptability, cost-effectiveness, feasibility, and system capacity may influence whether or not these parameters are reasonable and implementable. - Smoking history is a subjective risk factor, and we acknowledge that it cannot be precisely measured. If smoking is begun in early adulthood (i.e., early 20s) as it commonly is, by age 50 to 55, most people will have exceeded 20 pack-years. Although the NLST enrolled participants with a minimum smoking history of 30 pack-years, several other studies used a threshold of 20 pack-years or less. These studies had lung cancer detection rates similar to those of the NLST. It is anticipated that an increased detection rate would lead to a mortality reduction. The Working Group agreed on a 30 pack-year smoking history threshold to recommend lung cancer screening, aligning with that study entry criterion in the NLST. The panel will update this recommendation when the results of the NELSON trial (which had a 15 pack-year requirement) are published. - It is reasonable to define the screening population by age and smoking history, but there is currently insufficient evidence to include participants based on other risk factors such as family history, passive smoking, occupational exposure, radon exposure, previous cancer, and other diseases. # Qualifying Statements Screening may be a reasonable option in persons with a smoking history of <30 packyears. However, as this risk group was not included in the NLST, an explicit recommendation in favour of screening such persons cannot be made at this time. A current trial (NELSON) includes patients with a minimum smoking history of 15 pack-years and may provide additional data to determine the minimum smoking history appropriate for screening. # Defining a Positive Result on LDCT and Follow-up of a Positive Result # Key Evidence - Most of the studies published since 2008 used multi-detector CT scanners. The voltage ranged from 100 to 140 kVp, with all but one study using 120 to 140 kVp. The current ranged from 20 to 100 mAs, with all but one study not exceeding 60 mAs. The average effective dose was reported in 5 studies and ranged from 0.6 to 1.5 mSv (5). The NLST used multi-detector scanners with a minimum of 4 channels, 120 to 140 kVp, 20 to 30 mAs, and an average effective dose of 1.5 mSv (4). - Among the studies, collimation ranged from 0.75 to 10 mm (5). Collimation in the NLST was ≤2.5 mm (4). - Nodule size found on LDCT warranting further investigation ranged from a minimum size of any diameter to a maximum of >15 mm (5). In the NLST, nodules measuring ≥4 mm received further work-up (4). - Nine studies defined tumour growth. Growth can be determined with calliper measurements of diameter (6 studies) or 3-dimensional volume measurements (4 studies). One RCT and one single-arm study described significant growth as an increase in tumour diameter of ≥1 mm. Three single-arm studies described significant growth as an increase in diameter in at least 1 dimension. Two RCTs described growth as a change in tumour volume of ≥25%. One single-arm study defined growing lesions as those with volume-doubling time between 30 and 400 days, and another used tumour volume and time between high-resolution CT scans to calculate doubling time (5). A definition of growth was not reported in the NLST. - Guidance on the presentation and clinical work-up of a lung cancer diagnosis is detailed in the CCO Lung Cancer Diagnosis Pathway (7). - For screening modality, the parameters listed in the recommendations are derived from the NLST and ongoing studies. - With respect to collimation, newer scanners are able to provide 1-mm collimation with a short breath-hold time, but a large amount of images are produced making scrolling and reading cumbersome. At the current time, the collimation used in the NLST is recommended. - With respect to nodule size warranting further investigation, the recommendation deviates from the parameters of the NLST. In general, the smaller the nodule that defines a positive scan, the larger the number of positive scans, and the larger the number of false-positive results and unnecessary investigations for benign nodules. Based on a 4-mm threshold, 7191 of 26,309 (27.3%) scans in the NLST were positive; 6921 (96%) of the positive results were false positive. A 5-mm threshold will lower the rate of false-positive results, and if nodules between 4 and 5 mm are assessed on an annual scan, it is unlikely a significant finding will be missed. A prospective study of 1035 high-risk individuals found that nodules <5 mm identified by LDCT could be safely monitored at 1-year intervals (8). A retrospective study of two cohorts of patients (n=1000 and n=1897) determined that had no immediate attention been given to nodules between 3 and 5 mm until the first annual repeat screening, immediate further work-up would have been recommended in only 13% of patients rather than the 28% that received diagnostic interventions (9). Raising the threshold for a positive scan from a diameter of 4 mm to a diameter of 5 mm will help lower the false-positive rate without sacrificing the early detection of curable lung cancers. A recent study has suggested that increasing the threshold for a positive scan to 7 or 8 mm may decrease further work-up without delaying diagnosis (10). This will be revisited in future versions of this guideline when more information becomes available. - The recommended follow-up is based on common standard of care actions in the presence of positive findings. Short-term follow-up CT scans are recommended in the event of a positive-screening CT scan to assess the growth of a parenchymal nodule. These CT scans do not need to cover the entire chest; it is sufficient to limit the scan to the location of a nodule (i.e., a slab of a few centimetres covering the location of the nodule). This can substantially decrease the radiation exposure to the patient. # LDCT Screening Interval Recommendation 3: Persons at high risk for lung cancer should commence screening with an initial LDCT scan followed by annual screens for 2 consecutive years, and then once every 2 years after each negative (-ve) scan. a A positive (+ve) test is defined as a solid nodule ≥5 mm or a non-solid nodule (part solid or ground glass) ≥8 mm. b If the nodule appearance dictates a different approach (e.g., bronchoscopy or PET), this can be chosen at the discretion of the reading physician. c Doubling time of between 30 and 400 days. d Lung Cancer Diagnosis Pathway (7). # Key Evidence - LDCT was done on an annual basis in 18 studies; on years 1, 2, and 4 in one study; every 6 months in one study; and after 2 years in one study (5). The NLST conducted LDCT screens annually for 3 years (4). - The MILD trial did not demonstrate a shift to higher stage disease with biennial screening compared with annual screening. Of 49 lung cancers, 20 were detected in the biennial group and 29 in the annual group (11). - The current evidence stems from research studies on lung cancer screening, which by definition have a beginning and an end (e.g., in the case of the NLST, three rounds of screening). This guideline, however, extends this evidence to a screening program, which does not have a defined end. The annual to biennial approach is based on best evidence balancing expected benefit from regular scanning with accumulated harms from costs, radiation, and burden on the health care system. - The current evidence is not sufficient to confirm the benefit of a specific screening interval. The recommendation of annual screening for 3 years is subject to change when longer term trial evidence or further stratification methods become available from the NELSON trial. # Organized Versus Opportunistic Screening The decision to implement an organized, population-based screening program involves many factors, not just the existence of supportive RCT clinical evidence. However, because the benefit of screening to date has only been demonstrated in the context of an organized screening effort (i.e., a randomized clinical trial that compared two types of screening technology), it is the opinion of the Working Group that screening should be conducted in a manner similar to the NLST trial: that is, in an organized fashion. The ASCO guideline also supports screening of high-risk individuals, but only in the settings that can deliver comprehensive care such as that provided to NLST participants. The NLST authors themselves advise restraint in contemplating lung cancer screening recommendations on the basis of the NLST findings claiming the need for rigorous analysis of the cost-effectiveness of LDCT, and the weighing of the reduction in mortality against the harms of positive screening results, overdiagnosis, and cost (4). However, we are aware that these issues would be examined by provincial policy makers before screening policy decisions were made and approved. Because of the potential harms that may arise with LDCT screening done contrary to the recommendations above, a program is required that explicitly describes the target population that will benefit the most, the referral process, the frequency and duration of screening, the locations where screening may take place, the personnel involved in performing and interpreting the scans, and the precise criteria that define a positive scan. The inclusion of smoking-cessation counseling within the screening program is crucial. If elements of data collection and monitoring, quality assurance, and evaluation are built into the screening program from the start, it can be modified while in operation. Opportunistic screening takes the form of CT scans applied to individuals who are asymptomatic, may not qualify for the test, or are referred on an ad hoc basis outside of a programmatic structure. These scans often include contrast, are not done with the low-dose technique, and lack appropriate follow-up of detected lung nodules. This type of screening results in unnecessarily high radiation to the individual, potential side effects from contrast, and invasive procedures for potentially benign lesions. The Working Group believes strongly that screening outside a centre with experience and expertise in identifying the high-risk population, interpreting results and counselling patients, and performing the appropriate diagnostic techniques is ill advised. Such ad hoc screening will lead to an increase in the false-positive rate and in peri-procedure morbidity and mortality, and will threaten to mitigate some or all of the benefits of the screening process. # Next Steps The Lung Cancer Screening Working Group believes that the benefits of screening highrisk populations for lung cancer with LDCT outweigh the harms. The benefits stem from the documented improvement in mortality observed in the NLST showing that LDCT can not only detect small, early-stage lung cancers, but it can also facilitate curing an individual of lung cancer. The harms stem from the investigation itself (radiation exposure) and the sequelae from the false-positive results (detection of lung nodules that ultimately turn out to be benign), and the risk associated with diagnostic evaluation. We address the concern over radiation exposure by recommending a low-dose regimen and by increasing the screening interval to every 2 years after three negative annual scans. We also suggest that the follow-up CT of a suspicious nodule be done as a limited scan to further reduce the radiation exposure. We address the impact of false-positive results by the definition of a positive CT scan: we intentionally deviated from the parameters of the NLST in this instance. In the NLST, the threshold for a positive result was a nodule ≥4 mm in diameter. At baseline, >27% of the screening tests were positive and 96% of those were false-positive results. By increasing the threshold of a positive test to 5 mm, the rate of positive baseline scans can be reduced to <20% while still detecting early-stage, curable lung cancers. We also recommend a follow-up algorithm of CT-detected nodules that is simple and straightforward based on size and growth, and results in an extremely low rate of invasive procedures for benign lesions (12). Lung cancer screening with LDCT is recommended and can be most effectively and safely offered through an organized screening program and administered by specialized centres with multidisciplinary care teams. To determine whether or not a population-based screening program is appropriate for Ontario will require the CCO Prevention and Cancer Control division to investigate the other criteria relevant to the decision-making process. Priorities include: - Safety and effectiveness (long-term) Copyright This report is copyrighted by Cancer Care Ontario; the report and the illustrations herein may not be reproduced without the express written permission of Cancer Care Ontario. Cancer Care Ontario reserves the right at any time, and at its sole discretion, to change or revoke this authorization.
# INTENDED PURPOSE - This recommendation report is primarily intended to guide the Ontario PET Steering Committee in their decision making concerning indications for the use of PET imaging. - This recommendation report may also be useful in informing clinical decision making regarding the appropriate role of PET imaging and in guiding priorities for future PET imaging research. # RECOMMENDATIONS AND KEY EVIDENCE These recommendations are based on an evidentiary foundation consisting of one recent high-quality systematic review from the U.S. Agency for Health Research and Quality (AHRQ) (1) that included primary study literature for the period from 2003 to March 2008. # Diagnosis/Staging PET is not recommended for primary diagnosis of pancreatic cancer. Eleven prospective studies were identified that evaluated the role of PET or PET/CT in the diagnosis of a suspicious pancreatic mass. Sensitivity ranged from 69% to 97%, and specificity ranged from 61% to 97% (Giorgi et al Nishiyama et al Rasmussen et al van Kouwen et al Bang et al Heinrich et al Lemke et al Lytras et al Maemura et al Sperti et al Casneuf et al. Meta-analysis of four prospective studies evaluating the diagnostic performance of PET for the purpose of primary diagnosis (Giorgi et al Nishiyama et al Rasumussen et al van Kouwen et al yielded a pooled positive likelihood ratio (+LR) of 4.28 (95% CI 2.07 to 8.86) and negative likelihood ratio (-LR) of 0.21 (CI 0.12 to 0.40). These LRs had moderate heterogeneity, presenting some difficulties in determining overall accuracy. Meta-analysis of seven prospective studies evaluating the diagnostic performance of PET with the purpose of primary diagnosis and staging (Bang et al Casneuf et al Lemke et al Lytras et al Maemura et al Ruf et al Sperti et al yielded a +LR of 2.77 (CI 1.62 to 4.73) and -LR of 0.19 (CI 0.10 to 0.34). There was considerable heterogeneity, limiting determination of the overall accuracy of PET. Meta-analysis of three studies on PET/CT (Casneuf et al Heinrich et al Lemke et al yielded a homogenous +LR of 2.69 (CI 1.84 to 3.94) and -LR of 0.16 (CI 0.10 to 0.26). These pooled LRs suggest that PET and PET/CT offer small benefit in ruling in and ruling out pancreatic cancer when investigating a suspicious pancreatic mass; therefore, they may be useful in establishing a diagnosis when standard investigations are not confirmatory. Five studies compared PET or PET/CT with CT in the diagnosis of a suspicious pancreatic mass. In two that compared PET, PET/CT, and CT (Lemke et al Casneuf et al, PET/CT had the better diagnostic performance. # Qualifying Statements - The gold standard as well as the clinical goal is biopsy. When biopsy is inconclusive or not possible and the diagnosis remains in doubt, the above evidence supports the use of PET/CT where a positive result would lead to surgical resection for purposes of both diagnosis and treatment. Neuroendocrine tumours of the pancreas are known to be unreliably fluorodeoxyglucose (FDG) avid. PET is recommended for staging if a patient is a candidate for potentially curative surgical resection as determined by conventional staging. In four studies (Bang et al Heinrich et al Nishiyama et al Sperti et al, staging and treatment strategy changed after PET or PET/CT scan in 12% to 69% of cases. In one study (Heinrich et al with 46 patients with pancreatic carcinoma, standard staging followed by PET/CT improved the detection of distant metastases compared with standard staging alone (88% vs 56%, p=0.06 McNemar test). In Nishiyama et al 16 of 42 patients were found to have distant metastases by radiologic evaluation or cytological verification. With the combination of PET and CT, all metastatic sites were detected. Based on the above studies, staging and hence surgical management are impacted in a substantial proportion of patients who are candidates for surgery. # Qualifying Statement - The clinical importance of change in treatment strategy as an outcome, despite a lack of strong evidence, is noted. # Assessment of Treatment Response A recommendation cannot be made for or against the use of PET to guide clinical management based on assessment of treatment response due to insufficient evidence. One study (Bang et al showed that PET was superior to CT in the detection of treatment response after chemoradiation. Of 102 patients evaluated for a suspicious pancreatic mass, 15 with confirmed pancreatic cancer received chemoradiation. CT did not detect any responders while PET detected 5/15 therapy responders. The response after chemoradiation correlated with longer time to progression (TTP) compared with nonresponders (399 vs 233 days). A second study (Maemura et al showed that in 23 patients who received chemoradiation, an SUV <7.0 was correlated with improved survival. The above results are based on two small nonrandomized studies and therefore are not strong enough to make a recommendation for using PET in evaluating treatment response outside of a clinical trial. # Qualifying Statement - A recommendation for PET cannot be made in the setting of incomplete resection due to lack of evidence. Recurrence/Restaging PET is not recommended for clinical management of suspected recurrence, nor for restaging at the time of recurrence, due to insufficient evidence and lack of effective therapeutic options. One study (Ruf et al compared PET with CT in 31 patients who had suspected recurrence based on symptoms or increased CA 19-9 levels. While PET had higher sensitivity than CT for the detection of recurrence overall (96% versus 39%), and for nonhepatic intra-and extraabdominal metastases, CT had a superior sensitivity for the detection of liver metastases (92% vs 42%). However, patient outcomes based on these results were not reported. In a subset of 12 patients in Casneuf et al (12) who were being screened for recurrent pancreatic cancer, the sensitivity, specificity, and accuracy were not different between PET, PET/CT, and CT. In neither study was a reported change in management identified based on scanning modality. # Qualifying Statement - Pancreatic cancer has high overall mortality, and recurrence is uniformly fatal. At this time, there are insufficient treatment options that improve the outlook in patients who recur after surgical resection that would allow PET to contribute to management. PET imaging in recurrent disease should be restricted to clinical trials. # Solitary Metastasis Identified at Time of Recurrence A recommendation cannot be made for or against the use of PET for staging if a solitary metastasis is identified at recurrence as there are no trials that identify the utility of PET scanning in this setting. No studies exist that examine PET in this setting. # Qualifying Statement

End of preview. Expand in Dataset Viewer.

README.md exists but content is empty. Use the Edit dataset card button to edit it.

Downloads last month: 1

Edit dataset card

Size of downloaded dataset files:

Size of the auto-converted Parquet files:

Number of rows: