Key PointsQuestion
How can BRCA1 and BRCA2 gene testing in patients with cancer be increased?
Findings
In this quality improvement study of 1184 individuals, 5 cancer-based criteria with a 10% mutation detection rate were used by the cancer team to approve genetic testing for patients with cancer.
Meaning
Mainstreaming genetic testing using simple, cancer-based criteria may provide an efficient way to implement consistent, transparent, equitable, cost-effective, patient-centered genetic testing.
Importance
Increasing BRCA1 and BRCA2 (collectively termed herein as BRCA) gene testing is required to improve cancer management and prevent BRCA-related cancers.
Objective
To evaluate mainstream genetic testing using cancer-based criteria in patients with cancer.
Design, Setting, and Participants
A quality improvement study and cost-effectiveness analysis of different BRCA testing selection criteria and access procedures to evaluate feasibility, acceptability, and mutation detection performance was conducted at the Royal Marsden National Health Service Foundation Trust as part of the Mainstreaming Cancer Genetics (MCG) Programme. Participants included 1184 patients with cancer who were undergoing genetic testing between September 1, 2013, and February 28, 2017.
Main Outcomes and Measures
Mutation rates, quality-adjusted life-years (QALYs), and incremental cost-effectiveness ratios were the primary outcomes.
Results
Of the 1184 patients (1158 women [97.8%]) meeting simple cancer-based criteria, 117 had a BRCA mutation (9.9%). The mutation rate was similar in retrospective United Kingdom (10.2% [235 of 2294]) and prospective Malaysian (9.7% [103 of 1061]) breast cancer studies. If traditional family history criteria had been used, more than 50% of the mutation-positive individuals would have been missed. Of the 117 mutation-positive individuals, 115 people (98.3%) attended their genetics appointment and cascade to relatives is underway in all appropriate families (85 of 85). Combining with the equivalent ovarian cancer study provides 5 simple cancer-based criteria for BRCA testing with a 10% mutation rate: (1) ovarian cancer; (2) breast cancer diagnosed when patients are 45 years or younger; (3) 2 primary breast cancers, both diagnosed when patients are 60 years or younger; (4) triple-negative breast cancer; and (5) male breast cancer. A sixth criterion—breast cancer plus a parent, sibling, or child with any of the other criteria—can be added to address family history. Criteria 1 through 5 are considered the MCG criteria, and criteria 1 through 6 are considered the MCGplus criteria. Testing using MCG or MCGplus criteria is cost-effective with cost-effectiveness ratios of $1330 per discounted QALYs and $1225 per discounted QALYs, respectively, and appears to lead to cancer and mortality reductions (MCG: 804 cancers, 161 deaths; MCGplus: 1020 cancers, 204 deaths per year over 50 years). Use of MCG or MCGplus criteria might allow detection of all BRCA mutations in patients with breast cancer in the United Kingdom through testing one-third of patients. Feedback questionnaires from 259 patients and 23 cancer team members (12 oncologists, 8 surgeons, and 3 nurse specialists) showed acceptability of the process with 100% of patients pleased they had genetic testing and 100% of cancer team members confident to approve patients for genetic testing. Use of MCGplus criteria also appeared to be time and resource efficient, requiring 95% fewer genetic consultations than the traditional process.
Conclusions and Relevance
This study suggests that mainstream testing using simple, cancer-based criteria might be able to efficiently deliver consistent, cost-effective, patient-centered BRCA testing.
Germline mutations in BRCA1 or BRCA2 (collectively termed herein as BRCA) contribute to approximately 12% to 15% of ovarian cancer and approximately 3% to 5% of breast cancer in most populations.1-3 Identifying BRCA mutations can deliver substantial benefits in cancer management and cancer prevention.1,4,5 Patients with cancer with BRCA mutations are at increased risk of a second cancer, and many choose to have bilateral mastectomy and/or salpingo-oophorectomy to mitigate this risk.1,4 Furthermore, specific therapeutic options for cancers driven by BRCA mutations are increasingly available.1,5 Identification of a BRCA mutation also provides opportunities for relatives to access improved cancer risk information, screening, and risk-reducing interventions, and it is a cost-effective cancer prevention strategy.4,6-11
When the BRCA genes were discovered, the time and cost of testing precluded routine integration into the cancer diagnostic pathway. Instead, family history (FH) of cancer was used to identify individuals most likely to have a BRCA mutation, with many countries using a 10% chance of a BRCA mutation as the test access threshold.12-14 However, implementation of BRCA testing has been impeded by the complexity of FH criteria and the limited FH information typically available in clinical medicine.15-17 These impediments have led to inappropriate and inconsistent testing.18-22 Furthermore, many patients with cancer with BRCA mutations are not tested because they do not fulfill FH criteria.23,24
These problems have led to systematic underuse and inappropriate use of BRCA testing over the past 2 decades, with consequent lost opportunities for improved cancer management and cancer prevention.14-26 A recent US study estimated that only 20% of eligible individuals are being offered testing, with more than a million eligible individuals not having testing between 2000 and 2010.27 It is further estimated that only 30% of patients with breast cancer and 10% of unaffected individuals with BRCA mutations in the United States have been identified.28 These challenges and outcomes have been similar in many other countries.14-26
To improve BRCA testing, the Mainstreaming Cancer Genetics (MCG) Programme has been developing simplified eligibility criteria and testing access processes.29 Ovarian cancer was addressed first, simplifying eligibility to all women with epithelial ovarian cancer, as the BRCA mutation rate is more than 10% within this group. A mainstream test access model was validated in which patients with ovarian cancer were directly approved for BRCA testing by their cancer team, with patients who were BRCA mutation-positive rather than all patients having an appointment for genetics consultation. The mainstream model has proved to be popular, efficient, and cost-effective and more than 1000 patients with ovarian cancer have had BRCA testing through the mainstream access model in the Royal Marsden National Health Service Foundation Trust.7 Many other centers have adopted similar processes, with comparable positive results.30-34 Herein we report on the development and evaluation of simple, cancer-based eligibility criteria for use in mainstream genetic testing in patients with breast cancer.
The MCG Programme was a quality improvement program aimed at making BRCA genetic testing routine for appropriate patients with cancer (eAppendix 1 in the Supplement). There were 2 subprojects: the first was led by the gynecology-oncology team and has been published29 and the second was the equivalent breast team project, MCG-breast. We used structured literature review, international guidelines, and real-world evidence to determine 5 simple, cancer-based criteria estimated to give an approximate 10% mutation detection rate in patients with breast cancer (eFigure 1 in the Supplement). The Royal Marsden/Institute of Cancer Research/National Institute for Health Research Specialist Biomedical Research Centre Review Board approved the study. Informed written consent was obtained from all participants; they did not receive financial compensation. This study followed the Standards for Quality Improvement Reporting Excellence (SQUIRE) reporting guideline for quality improvement studies.
The MCG-breast study ran for 3.5 years, between September 1, 2013, and February 28, 2017. A total of 1184 patients with breast cancer received genetic testing through MCG-breast during this period. Nine breast cancer predisposition genes (BRCA1 [OMIM 113705], BRCA2 [OMIM 600185], PALB2 [OMIM 610355], TP53 [OMIM 191170], PTEN [OMIM 601728], STK11 [OMIM 602216], CDH1 [OMIM 192090], ATM [OMIM 607585], and CHEK2 [OMIM 604373]) were tested using the TruSight Cancer Panel (IIlumina).
While MCG-breast was running, 182 patients with breast cancer underwent testing through the genetics department because, although the proband did not fulfill an MCG-breast criterion, their FH made them eligible as they had a Manchester Scoring System (MSS) score of 15 or higher (eFigure 2 in the Supplement). These individuals are called the FH series.
We also used data from 2294 patients from the Breast and Ovarian Cancer Susceptibility (BOCS) study, a retrospective UK study of familial breast cancer, and 2575 patients from the Malaysia-Breast Cancer genetics Study (BCGS),35 a prospective Malaysian study of patients with cancer unselected for age or FH, to evaluate the mutation detection rates of the MCG-breast criteria. We used these patient populations to compare MCG-breast criteria with 3 widely used systems for determining BRCA testing eligibility: the MSS1,36 Breast and Ovarian Analysis of Disease Incidence and Cancer Estimation Algorithm (BOADICEA),37 and the National Comprehensive Cancer Network (NCCN) Guidelines.38
Patient and Cancer Team Acceptability
We evaluated patient and cancer team acceptability of the mainstream process using similar questionnaires to the equivalent ovarian cancer programme.29 We initiated the patient feedback once 250 had been consented by the cancer team, sending questionnaires to 259 participants, of whom 129 replied (113 BRCA mutation-negative and 16 BRCA mutation-positive). Twenty-three members of the cancer team completed the cancer team feedback questionnaire, including 12 oncologists, 8 surgeons and 3 nurse specialists. The questionnaire and results are detailed in eTable 2 and eTable 3 in the Supplement.
Cost-effectiveness Analyses
We performed a cost-effectiveness analysis using the model in Eccleston et al7 that was adapted for breast cancer. A patient-level simulation with a lifetime time horizon was constructed in Microsoft Excel (Microsoft Corp). In the model, an index population is compared for testing and no testing scenarios, patients are offered BRCA testing using mainstream testing, and cascade testing to unaffected family members is included. The analysis was conducted according to the Consolidated Health Economic Evaluation Reporting Standards (CHEERS) guideline.
The principal outcome was the cost per quality-adjusted life-year for each individual and aggregated to provide an incremental cost-effectiveness ratio. Other outcomes included total and disaggregated costs and clinical events, such as the number of new cancer cases prevented and the number of lives saved over a 50-year horizon. We also performed a probabilistic sensitivity analysis and constructed a multiple cost-effectiveness acceptability curve. Additional details on the methods are presented in eAppendix 2 in the Supplement.
A total of 1184 individuals with breast cancer (1158 women [97.8%] and 26 men [2.2%]), underwent BRCA testing because they fulfilled 1 or more of the eligibility criteria. Of these, 117 patients had a BRCA mutation, giving an overall mutation rate of 9.9% (Table 1; eTable 1 in the Supplement). The mutation rate of each criterion exceeded 10%. Most individuals met only 1 criterion (926 [78.2%]), with 246 (20.8%) meeting 2 criteria and 12 (0.1%) meeting 3 criteria, with mutation rates of 7.7% (71 of 926), 16.3% (40 of 246), and 50.0% (6 of 12), respectively (Table 1).
Of 117 mutation-positive individuals, 115 patients (98.3%) attended their genetics appointment. Of the 2 missing individuals, 1 declined her appointment and chemotherapy and subsequently died, and the other initially declined her appointment but then recontacted the genetics department. Cascade of information to relatives is underway in 85 families, was not relevant in 16 families, and is of unknown status in 16 families outside our catchment area. We surveyed 259 patients to evaluate acceptability of the process, of whom 129 replied (eTable 2 in the Supplement). Acceptability was high, with 100% (128 of 128) pleased that they had genetic testing and 96.1% (124 of 129) happy that testing was done by their cancer team.
The cancer team acceptability of the process was also high. All 23 team members reported feeling confident to approve patients for BRCA testing during the cancer clinic consultation and believed that the process worked well (eTable 3 in the Supplement).
Evaluation of Testing Criteria in Other Data Sets
To further evaluate the testing eligibility criteria, we used data from 2294 participants in the BOCS Study for whom information to determine eligibility for criteria 1 through 3 was available. Of these, 235 patients (10.2%) had a BRCA mutation and each criterion also met the 10% threshold (Table 1).
To evaluate the criteria in a different population, we used data from the Malaysia-BCGS study, a prospective study of 2575 unselected women with breast cancer .35 Of these, 1061 patients (41.2%) were eligible for testing, of whom 103 women (9.7%) had a BRCA mutation (Table 1). Male breast cancer was not included in the study and the presence of exon deletion/duplication mutations was not tested. Consistent with the MCG-breast and BOCS data, the mutation detection rate in each category was approximately 10% (Table 1).
Comparison With Other Eligibility Criteria
We evaluated which of the 117 mutation-positive probands would have been eligible for testing using the MSS1,36 BOADICEA37 or the NCCN criteria.38 Fifty-five probands (47.0%) had an MSS1 score of 15 or greater and 37 probands (31.6%) had a 10% or greater mutation probability by BOADICEA. One hundred six patients (90.6%) met the NCCN criteria, of whom 103 patients (88.0%) were eligible through their personal cancer history (eTable 1 in the Supplement).
We combined the results of the MCG-breast and the equivalent gynecology-oncology study to develop universal testing criteria that could be used to determine eligibility of any patient with breast or ovarian cancer. We call these the MCG criteria (Figure). The MCG-breast criterion of a woman with breast and ovarian cancer became redundant because any woman with epithelial ovarian cancer has a greater than 10% chance of having a BRCA mutation.29 Thus, the MCG criteria are (1) ovarian cancer, (2) breast cancer diagnosed when patients are 45 years or younger, (3) 2 primary breast cancers diagnosed when patients are 60 years or younger, (4) triple-negative breast cancer, and (5) breast cancer in men (Figure).
Comparison With FH-Based Testing
During the study period, an additional 182 patients with breast cancer were tested through the genetics department. These patients did not fulfill MCG-breast criteria, but their FH made them eligible for testing because their MSS score was 15 or higher (eFigure 2 in the Supplement). Ten patients (5.5%) in this FH series had a BRCA mutation (eTable 4 in the Supplement). All 10 patients had a relative eligible for testing using the MCG criteria, and in 9 families an eligible relative was diagnosed with cancer before the proband (eTable 4 in the Supplement). A total of 151 of 182 patients (83.0%) in the FH series had an eligible relative and 86 individuals (47.3%) had an eligible first-degree relative.
The MCG-breast and FH series samples also underwent analysis of 7 additional breast cancer predisposition genes: 5 rare, higher penetrance predisposition genes (PALB2, TP53, CDH1, PTEN, and STK11) and 2 more common, lower penetrance predisposition genes (ATM and CHEK2).39 One individual (0.05%) in the FH series had a mutation in a higher penetrance non-BRCA predisposition gene compared with 24 patients (2.0%) in the MCG-breast group. In total, 9 patients (4.9%) in the FH series and 51 patients (4.3%) in the MCG-breast group had mutations in non-BRCA cancer predisposition genes (eTable 5 in the Supplement).
We next considered all 1366 individuals tested during the study period through the MCG-breast or FH series, of whom 127 patients (9.3%) had a BRCA mutation. The BRCA mutation detection rate using the different eligibility criteria are given in Table 2. The MCG and NCCN criteria performed best, with 92.1% (n = 117) and 91.3% (n = 116) mutation detection rates, respectively.
We also considered whether it were possible to extend the simple cancer-based criteria to encompass the missed mutations detected through the FH series. Six of 10 FH-series mutations (60.0%) were in patients with breast cancer with first-degree relatives who would have been eligible for testing. We therefore added a sixth category, which is breast cancer and a parent, sibling, or child (ie, a first-degree relative) with any of the other criteria. Criteria 1 through 5 are considered the MCG criteria and 1 through 6 are considered the MCGplus criteria (Figure). Applying the MCGplus criteria to the 1366 (MCG-breast plus FH series) individuals leads to an extra 86 people being eligible for testing, compared with the MCG criteria, and detection of an extra 6 mutations. Overall, MCGplus criteria detected 123 of 127 mutations (96.7%) and retained an overall mutation rate of 9.7% (123 of 1270) (Table 2).
We performed a cost-effectiveness analysis using the ovarian cancer cost-effectiveness model previously described, adapted for breast cancer.7 We compared implementation of BRCA testing using the mainstream process and the MCG or MCGplus criteria with no testing. Both the MCG and MCGplus criteria were cost-effective with cost-effectiveness ratios of $1330 per discounted QALYs (Table 3) and $1225 per discounted QALYs (Table 4), respectively. Multiple cost-effectiveness acceptability curves showed that BRCA testing with MCG or MCGplus criteria had a greater than 99% probability of being cost-effective at a willingness-to-pay threshold of $26 184 per quality-adjusted life-year. With use of the MCG criteria, the model estimates that 804 cancers and 161 deaths would be prevented per year of testing over the subsequent 50 years. With use of the MCGplus criteria, 1020 cancers and 204 deaths are estimated to be prevented per year over 50 years.
Genetic Testing and Consult Requirements
Determining who is eligible for cancer predisposition gene testing based on personal history rather than FH allows one to better estimate testing and resource requirements. To exemplify this, we considered the UK outcome of implementing the mainstream testing model with the MCG or MCGplus criteria. Approximately 55 000 breast cancer cases and 7000 ovarian cancer cases are diagnosed in the United Kingdom each year. We estimate that approximately 20 000 individuals would be eligible for testing using the MCG criteria and approximately 25 000 would be eligible using the MCGplus criteria (eTable 6 in the Supplement). These numbers would lead to the identification of 2000 (MCG) or 2500 (MCGplus) BRCA mutations each year, given the mutation detection rate of approximately 10%. Approximately 12% of ovarian cancer and 3% of breast cancer diagnoses occur in BRCA mutation carriers, which is 2500 per year (840 ovarian cancers and 1650 breast cancers).2,3 Hence, using the MCGplus criteria theoretically allows one to identify all of the BRCA mutations in patients with breast cancer through testing of only a third of the patients. Using the mainstream access model also leads to substantial reduction in testing time and genetic consultation requirements (eFigure 3 in the Supplement). In the traditional access model, everyone has input from the genetics department before and after testing; thus, 50 000 consultations would be required if the MCGplus criteria were used. In the mainstream model, the discussions before and after the test are undertaken by the cancer team during existing appointments. The 2500 BRCA mutation-positive individuals additionally have a genetics department appointment—a 95% reduction in genetic consultation requirements. The posttest cancer and genetic management for the BRCA mutation-positive individuals and their relatives is the same in both models (eFigure 3 in the Supplement). The mainstream access model also results in approximately 85% (4 vs 25 weeks in the United Kingdom) reduction in time to test result, compared with the traditional access model (eFigure 3 in the Supplement).
In this study of 1184 individuals, we estimate that simple, cancer-based criteria may identify patients with breast cancer eligible for genetic testing at an approximate 10% BRCA mutation rate threshold. Our results are supported by data from 2294 patients from a retrospective UK study of familial breast cancer and 2575 patients from a prospective Malaysian study of unselected breast cancer.
Integrating these results with our equivalent study in ovarian cancer leads to 5 simple MCG criteria that can be used to determine which patients with breast or ovarian cancer should be offered genetic testing. The simplicity of the criteria makes it possible to readily and consistently determine who is eligible for testing. In turn, a mainstream testing model may be used whereby the cancer team directly performs testing instead of referring patients to the genetics department. Patients and cancer team members expressed satisfaction in this process, and it also appears to be time- and cost-efficient. In particular, mainstream testing reduces genetic consultation requirements by approximately 95%, as only the 10% mutation-positive individuals have genetic input after the test, compared with the traditional model in which all patients have genetic input before and after testing. It is recognized that, in most countries, a shortage of genetic counselors precludes offering pretest and posttest counseling to all eligible patients with cancer in a timely fashion.40,41 This limitation has led to ad hoc, nonvalidated testing without genetic counseling.41 With the cost of testing decreasing and the therapeutic implications of testing increasing, it is inevitable that testing without prior genetic counselor input will expand. Furthermore, tumor (somatic) genetic testing, which is initiated by the cancer team, is rapidly expanding and it is likely that germline and tumor genetic testing will converge over the next 5 years. We therefore believe it is preferable to explicitly institute germline genetic testing by the cancer team in close collaboration with the genetics department, as we have done herein. This cooperation ensures that the cancer team will give appropriate information before testing and that patients will receive genetic input when required. It is reassuring that we, and others, have found that patients with cancer are supportive of this testing model.29-34
The cancer-based criteria performed better than the FH-based MSS and BOADICEA, both of which would have missed more than 50% of the mutations. This finding is consistent with other studies showing that FH is a suboptimal selection method for identifying patients with cancer who have BRCA mutations.23,24,29 The mutation rate of patients in the FH series who met FH-based criteria but did not meet MCG-breast criteria was only 5.5%. This low level is to be expected; it is inevitable that the mutation rate of FH-based criteria will be reduced if one excludes patients eligible through their cancer status. Historically, such evaluations have taken the reverse approach; they have evaluated the mutation rate in patients with cancer not eligible by FH-based criteria.12 Given that cancer-based criteria are easier and faster to use, we believe prioritizing cancer-based criteria and adding FH-based criteria, if required, is the preferable approach. Moreover, 151 of 182 patients (83.0%) in the FH series had a relative eligible for testing by the MCG criteria, including all of the mutation-positive individuals. This finding suggests that systematic deployment of the MCG criteria would, in time, detect almost all of the mutations currently identified by FH-based criteria as well as many mutations that FH-based criteria miss. Cancer-based criteria also make it easier to estimate and evaluate testing service requirements and performance than when using FH-based criteria, as we demonstrate for the United Kingdom. Use of the cancer-based criteria should facilitate the improvement in BRCA testing delivery that many studies have shown is required.14-26
For countries without a legacy of FH-based testing or those in which FH discussions can deter testing, we believe the 5 MCG criteria offer a simple, useful way to deliver genetic testing to patients with breast and ovarian cancer. For countries that have used FH-based selection criteria for many years, this change may be too radical. To address this issue, we developed MCGplus criteria, which retain the simplicity and mutation rate of the MCG criteria and add a sixth criterion of a patient with breast cancer who has a parent, sibling, or child meeting the MCG criteria. In due course, this sixth criterion should become obsolete because the eligible relative will have been tested during their cancer diagnosis. In our center we have been using the MCGplus criteria in genetics and oncology since November 2017, and this has been well received (eFigure 4 in the Supplement).
In recent years the NCCN has expanded personal cancer-based criteria, which has led to improved mutation detection. The NCCN criteria performed similarly to the MCG criteria but are more complex, as both cancer-based and FH-based criteria are now included. This complexity makes use of the NCCN guidelines time consuming, impeding widespread adherence.16,19,21,35 There are strong similarities between the MCG and NCCN cancer-based criteria. A key difference is that all patients with triple-negative breast cancer are eligible by MCG criteria, but only those younger than 60 years are eligible by the NCCN. Ten of 44 mutation-positive women (22.7%) with triple-negative breast cancer in MCG-breast were diagnosed when they were older than 60 years. Moreover, the mutation rate in triple-negative breast cancer at any age in the MCG-breast, BOCS, and Malaysia-BCGS data was greater than 10%, and this finding is consistent with other studies.12,42,43 It is noteworthy that 106 of 116 mutations (91.4%) detected by the NCCN criteria were in patients eligible for testing due solely to their cancer history, although many also fulfilled FH criteria. This finding suggests that there may be redundancy in the NCCN cancer-based and FH-based criteria that could be streamlined.
Using MCG or MCGplus criteria is cost-effective and appears to result in meaningful cancer and mortality reductions. BRCA testing using many different criteria and testing processes has been shown to be cost-effective.4,6-11 The combined economic benefit and increasing clinical utility of testing appear to provide a strong argument for relaxing the mutation threshold and expanding testing to more patients with breast cancer. Changing the age threshold for breast cancer to 50 years or younger or 60 years or younger and removing MCG criterion 2 would be the logical expansions and simplifications of the MCG criteria. Ultimately, offering testing to all patients with breast cancer will likely occur. However, currently, most centers have limited resources they must use for testing at a specified mutation detection threshold.
The study has limitations. The cost-effectiveness analysis we performed compared testing with no testing. When we initiated the study, no germline genetic testing in oncology was performed in the United Kingdom, so testing vs no testing was considered the most appropriate comparison. Furthermore, a major aspiration of the MCG Programme was to provide data and resources that could be used by the many centers and countries that do not yet offer BRCA testing. However, for some centers, for example, those with surplus genetic resources that can be used in oncology, a more appropriate comparison would be between the MCG criteria and FH-based criteria.
Concurrent testing of other cancer predisposition genes in addition to the BRCA genes is being increasingly performed.44 We suggest that the mainstream testing process can be readily adapted for multigene testing; in our study, it led to an extra 51 (40% increase) posttest genetic consultations for patients with mutations in non-BRCA genes. However, we did not evaluate the performance of MCG criteria for non-BRCA genes. We believe that MCG criteria are likely to perform at least as well as other criteria because more eligible patients receive testing using the MCG criteria and mainstream testing process.45,46 The MCG criteria also do not address BRCA testing in patients with cancers other than breast or ovarian. To our knowledge, there is no evidence that personal history of cancers other than breast or ovarian would meet a 10% mutation detection threshold.
It is well known that underuse and inappropriate use of genetic testing in patients with cancer is impeding cancer management and prevention.14-26 Herein, we presented a streamlined genetic testing process based on short, simple eligibility criteria that can be used by the cancer team. We suggest that mainstreaming testing using these criteria may be an efficient way to deploy limited resources in a consistent, transparent, equitable, cost-effective, patient-centered manner.
Accepted for Publication: April 2, 2019.
Published: May 24, 2019. doi:10.1001/jamanetworkopen.2019.4428
Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2019 Kemp Z et al. JAMA Network Open.
Corresponding Author: Nazneen Rahman, MD, PhD, Division of Genetics and Epidemiology, Institute of Cancer Research, 15 Cotswold Rd, Sutton, Surrey SM2 5NG, United Kingdom (nazneen.rahman@icr.ac.uk).
Author Contributions: Drs Turnbull and Rahman had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
Concept and design: Kemp, Turnbull, George, Rahman.
Acquisition, analysis, or interpretation of data: All authors.
Drafting of the manuscript: Yost, Seal, Strydom, Rahman.
Critical revision of the manuscript for important intellectual content: Kemp, Turnbull, Mahamdallie, Poyastro-Pearson, Warren-Perry, Eccleston, Tan, Teo, Turner, George, Rahman.
Statistical analysis: Turnbull, Yost, Eccleston, Tan, Rahman.
Obtained funding: Kemp, Teo, Rahman.
Administrative, technical, or material support: Turnbull, Seal, Mahamdallie, Poyastro-Pearson, Warren-Perry, Turner, Strydom, George, Rahman.
Supervision: Kemp, Teo, George, Rahman.
Conflict of Interest Disclosures: Dr Kemp reported honoraria from Lilly for an educational talk and from AstraZeneca for educational talks, but the compensation was not personally received. Mr Eccleston reported personal fees from the Institute of Cancer Research during the conduct of the study. Dr Teo reported grants from AstraZeneca outside the submitted work. Dr Turner reported advisory board honoraria from AstraZeneca, Bristol-Myers Squibb, Lilly, Merck Sharpe and Dohme, Novartis, Pfizer, Roche/Genentech, Tesaro, and Bicycle Therapeutics; and research funding from AstraZeneca, BioRad, Pfizer, Roche/Genentech, Clovis, and Guardant Health. Dr George reported grants from Wellcome Trust during the conduct of the study and honoraria from AstraZeneca, Roche, and Tesaro outside the submitted work, but the compensation was not personally received. Dr Rahman reported grants from Royal Marsden/Institute of Cancer Research/National Institute for Health Research Specialist Biomedical Research Centre for Cancer and grants from Wellcome during the conduct of the study, personal fees from AstraZeneca, and personal fees from Genomics plc outside the submitted work. No other disclosures were reported.
Funding/Support: This study was supported by the Royal Marsden/Institute of Cancer Research National Insitute for Health Research Specialist Biomedical Research Centre for Cancer and Wellcome grant 098518.
Role of the Funder/Sponsor: The funding organizations had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Additional Contributions: We thank all the participating clinical, technical and administrative staff who facilitated this quality improvement program, and the many people who participated in discussions and provided invaluable advice about the program. They received no compensation for their contributions other than their salaries. We thank the many physicians and genetic counselors who recruited patients to the Breast and Ovarian Cancer Susceptibility (BOCS) study. Jessie Bull, BSc (Institute of Cancer Research), provided administrative oversight of BOCS recruitment with financial compensation through the study grant. Yoon Sook-Yee, PhD (Cancer Research Malaysia), Douglas Easton, PhD (Center for Cancer Epidemiology, University of Cambridge), and Alison Dunning, PhD (Center for Cancer Epidemiology, University of Cambridge), performed patient recruitment and genetic testing of patients in the Malaysian Breast Cancer Genetic Study, which was funded by Wellcome (203477/Z/16/Z). They received no financial compensation for their contributions to the present study.
2.Lang
GT, Shi
JX, Hu
X,
et al. The spectrum of BRCA mutations and characteristics of BRCA-associated breast cancers in China: screening of 2,991 patients and 1,043 controls by next-generation sequencing.
Int J Cancer. 2017;141(1):129-142. doi:
10.1002/ijc.30692PubMedGoogle ScholarCrossref 3.Li
J, Wen
WX, Eklund
M,
et al. Prevalence of BRCA1 and BRCA2 pathogenic variants in a large, unselected breast cancer cohort.
Int J Cancer. 2019;144(5):1195-1204. doi:
10.1002/ijc.31841Google ScholarCrossref 8.Tuffaha
HW, Mitchell
A, Ward
RL,
et al. Cost-effectiveness analysis of germ-line BRCA testing in women with breast cancer and cascade testing in family members of mutation carriers.
Genet Med. 2018;20(9):985-994. doi:
10.1038/gim.2017.231PubMedGoogle ScholarCrossref 9.Petelin
L, Trainer
AH, Mitchell
G, Liew
D, James
PA. Cost-effectiveness and comparative effectiveness of cancer risk management strategies in
BRCA1/2 mutation carriers: a systematic review.
Genet Med. 2018;20(10):1145-1156. doi:
10.1038/gim.2017.255PubMedGoogle ScholarCrossref 10.Lim
KK, Yoon
SY, Mohd Taib
NA,
et al. Is BRCA mutation testing cost effective for early stage breast cancer patients compared to routine clinical surveillance? the case of an upper middle-income country in Asia.
Appl Health Econ Health Policy. 2018;16(3):395-406. doi:
10.1007/s40258-018-0384-8PubMedGoogle ScholarCrossref 11.Norum
J, Grindedal
EM, Heramb
C,
et al. BRCA mutation carrier detection: a model-based cost-effectiveness analysis comparing the traditional family history approach and the testing of all patients with breast cancer.
ESMO Open. 2018;3(3):e000328. doi:
10.1136/esmoopen-2018-000328PubMedGoogle ScholarCrossref 12.Engel
C, Rhiem
K, Hahnen
E,
et al; German Consortium for Hereditary Breast and Ovarian Cancer (GC-HBOC). Prevalence of pathogenic
BRCA1/2 germline mutations among 802 women with unilateral triple-negative breast cancer without family cancer history.
BMC Cancer. 2018;18(1):265. doi:
10.1186/s12885-018-4029-yPubMedGoogle ScholarCrossref 13.National Collaborating Centre for Cancer. Classification and Care of People at Risk of Familial Breast Cancer and Management of Breast Cancer and Related Risks in People With a Family History of Breast Cancer. Cardiff, UK: National Collaborating Centre for Cancer; 2013.
14.van den Broek
AJ, de Ruiter
K, van ’t Veer
LJ,
et al. Evaluation of the Dutch
BRCA 1/2 clinical genetic center referral criteria in an unselected early breast cancer population.
Eur J Hum Genet. 2015;23(5):588-595. doi:
10.1038/ejhg.2014.161PubMedGoogle ScholarCrossref 15.Bellcross
CA, Leadbetter
S, Alford
SH, Peipins
LA. Prevalence and healthcare actions of women in a large health system with a family history meeting the 2005 USPSTF recommendation for BRCA genetic counseling referral.
Cancer Epidemiol Biomarkers Prev. 2013;22(4):728-735. doi:
10.1158/1055-9965.EPI-12-1280PubMedGoogle ScholarCrossref 16.Bellcross
CA, Peipins
LA, McCarty
FA,
et al. Characteristics associated with genetic counseling referral and
BRCA1/2 testing among women in a large integrated health system.
Genet Med. 2015;17(1):43-50. doi:
10.1038/gim.2014.68PubMedGoogle ScholarCrossref 23.Alemar
B, Gregório
C, Herzog
J,
et al.
BRCA1 and
BRCA2 mutational profile and prevalence in hereditary breast and ovarian cancer (HBOC) probands from Southern Brazil: are international testing criteria appropriate for this specific population?
PLoS One. 2017;12(11):e0187630. doi:
10.1371/journal.pone.0187630PubMedGoogle ScholarCrossref 26.Pujol
P, Lyonnet
DS, Frebourg
T,
et al. Lack of referral for genetic counseling and testing in
BRCA1/2 and Lynch syndromes: a nationwide study based on 240,134 consultations and 134,652 genetic tests.
Breast Cancer Res Treat. 2013;141(1):135-144. doi:
10.1007/s10549-013-2669-9PubMedGoogle ScholarCrossref 30.Høberg-Vetti
H, Bjorvatn
C, Fiane
BE,
et al.
BRCA1/2 testing in newly diagnosed breast and ovarian cancer patients without prior genetic counselling: the DNA-BONus study.
Eur J Hum Genet. 2016;24(6):881-888. doi:
10.1038/ejhg.2015.196PubMedGoogle ScholarCrossref 32.Nilsson
MP, Nilsson
ED, Borg
A, Brandberg
Y, Silfverberg
B, Loman
N. High patient satisfaction with a simplified
BRCA1/2 testing procedure: long-term results of a prospective study.
Breast Cancer Res Treat. 2019;173(2):313-318. doi:
10.1007/s10549-018-5000-yPubMedGoogle ScholarCrossref 33.Nilsson
MP, Törngren
T, Henriksson
K,
et al. BRCAsearch: written pre-test information and
BRCA1/2 germline mutation testing in unselected patients with newly diagnosed breast cancer.
Breast Cancer Res Treat. 2018;168(1):117-126. doi:
10.1007/s10549-017-4584-yPubMedGoogle ScholarCrossref 37.Lee
AJ, Cunningham
AP, Kuchenbaecker
KB, Mavaddat
N, Easton
DF, Antoniou
AC; Consortium of Investigators of Modifiers of
BRCA1/2; Breast Cancer Association Consortium. BOADICEA breast cancer risk prediction model: updates to cancer incidences, tumour pathology and web interface.
Br J Cancer. 2014;110(2):535-545. doi:
10.1038/bjc.2013.730PubMedGoogle ScholarCrossref 39.Taylor
A, Brady
AF, Frayling
IM,
et al; UK Cancer Genetics Group (UK-CGG). Consensus for genes to be included on cancer panel tests offered by UK genetics services: guidelines of the UK Cancer Genetics Group.
J Med Genet. 2018;55(6):372-377.
PubMedGoogle Scholar 42.Couch
FJ, Hart
SN, Sharma
P,
et al. Inherited mutations in 17 breast cancer susceptibility genes among a large triple-negative breast cancer cohort unselected for family history of breast cancer.
J Clin Oncol. 2015;33(4):304-311. doi:
10.1200/JCO.2014.57.1414PubMedGoogle ScholarCrossref