European Journal of Cancer 106 (2019) 54e60 Available online at www.sciencedirect.com ScienceDirect journal homepage: www.ejcancer.com Position Paper Genetic counselling and testing of susceptibility genes for therapeutic decision-making in breast cancerdan European consensus statement and expert recommendations Christian F. Singer a,*,1, Judith Balmaña b, Nicole Bürki c, Suzette Delaloge d, Maria Elisabetta Filieri e, Anna-Marie Gerdes f, Eli Marie Grindedal g, Sileni Han h, Oskar Johansson i, Bella Kaufman j, Mateja Krajc k, Niklas Loman l, Edith Olah m, Shani Paluch-Shimon n, Natalija Dedic Plavetic o, Kamil Pohlodek p, Kerstin Rhiem q, Manuel Teixeira r, D. Gareth Evans s a Medical University of Vienna, Department of Obstetrics and Gynecology, Vienna, Austria b Medical Oncology Department, Hospital Vall d’Hebron, Vall d’Hebron, Institute of Oncology, Universitat Autònoma de Barcelona, Barcelona, Spain c Department of Gynecology and Obstetrics, University Hospital Basel (UHB), Spitalstrasse 21, 4031, Basel, Switzerland d Department of Cancer Medicine, Gustave Roussy, 114 Rue Edouard Vaillant, 94800 Villejuif, France e Department of Medical Oncology, University Hospital of Modena, Via del Pozzo, Modena, Italy f Department of Clinical Genetics, Rigshospitalet, Copenhagen, Denmark g Department of Medical Genetics, Oslo University Hospital, Oslo, Norway h Department of Gynecology and Obstetrics, UZ Leuven, Leuven, Belgium i Landspitali-the National University Hospital of Iceland, Reykjavik 101, Iceland j Breast Oncology Institute, Sheba Medical Center, Tel Hashomer, Israel k Institute of Oncology Ljubljana, Slovenia, Zaloska 2, 1000 Ljubljana, Slovenia l Department of Clinical Sciences, Division of Oncology and Pathology, Lund University Hospital, 221 85 Lund, Sweden m Department of Molecular Genetics, National Institute of Oncology, Budapest, Hungary n Shaare Zedek Medical Centre, Jerusalem, Israel o Department of Oncology, Division of Medical Oncology, University Hospital Centre Zagreb, University of Zagreb, School of Medicine, Zagreb, Croatia p Second Department of Gynecology and Obstetrics, Comenius University of Bratislava, Faculty of Medicine, 82606 Bratislava, Slovakia q Center for Familial Breast and Ovarian Cancer, Center for Integrated Oncology, Medical Faculty, University Hospital of Cologne, D-50931 Cologne, Germany r Department of Genetics, Portuguese Oncology Institute, Porto, Portugal * Corresponding author: Department of OB/GYN and Comprehensive Cancer Center, Medical University of Vienna Waehringer Guertel 18-20, 1090 Vienna, Austria. Fax: þ43 (0)1 40400 23230. E-mail address: christian.singer@meduniwien.ac.at (C.F. Singer). 1 www.meduniwien.ac.at https://doi.org/10.1016/j.ejca.2018.10.007 0959-8049/ª 2018 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http:// creativecommons.org/licenses/by-nc-nd/4.0/). C.F. Singer et al. / European Journal of Cancer 106 (2019) 54e60 55 s Department of Genomic Medicine, Division of Evolution and Genomic Science, University of Manchester, MAHSC, St Mary’s Hospital, Manchester M13 9WL, United Kingdom Received 21 August 2018; received in revised form 9 October 2018; accepted 10 October 2018 Available online 22 November 2018 KEYWORDS Abstract An international panel of experts representing 17 European countries and Israel BRCA; convened to discuss current needs and future developments in BRCA testing and counselling BRCA1; and to issue consensus recommendations. The experts agreed that, with the increasing avail- BRCA2; ability of high-throughput testing platforms and the registration of poly-ADP-ribose-polymer- Metastatic breast ase inhibitors, the need for genetic counselling and testing will rapidly increase in the near cancer; future. Consequently, the already existing shortage of genetic counsellors is expected to Genetic testing; worsen and to compromise the quality of care particularly in individuals and families with sus- Genetic counselling; pected or proven hereditary breast or ovarian cancer. Increasing educational efforts within the Hereditary breast breast cancer caregiver community may alleviate this limitation by enabling all involved spe- cancer cialities to perform genetic counselling. In the therapeutic setting, for patients with a clinical suspicion of genetic susceptibility and if the results may have an immediate impact on the ther- apeutic strategy, the majority voted that BRCA1/2 testing should be performed after histolog- ical diagnosis of breast cancer, regardless of oestrogen receptor and human epidermal growth factor receptor 2 (HER2) status. Experts also agreed that, in the predictive and therapeutic setting, genetic testing should be limited to individuals with a personal or family history sug- gestive of a BRCA1/2 pathogenic variant and should also include high-risk actionable genes beyond BRCA1/2. Of high-risk actionable genes, all pathological variants (i.e. class IV and V) should be reported; class III variants of unknown significance, should be reported provided that the current lack of clinical utility of the variant is expressly stated. Genetic counselling should always address the possibility that already tested individuals might be re-contacted in case new information on a particular variant results in a re-classification. ª 2018 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). 1. Background reducing surgery but increasingly also has therapeutic implications for women who have already developed As per the most recent data from the International breast or ovarian cancer: BRCA mutation carriers with Agency for Research on Cancer (IARC), the lifetime newly diagnosed early breast cancer often chose bilateral risk for a woman in most parts of Europe and Israel mastectomy over unilateral lumpectomy [4]. In addition, varies between approximately 8% and 12% [1]. Women women with newly diagnosed breast cancer who un- with one or more affected family members face a further dergo rapid early genetic testing have also been increase in their lifetime risk, but the cumulative risk is demonstrated to prefer mastectomy over lumpec- highest in women who carry a BRCA1 or BRCA2 tomy [5]. Two recent clinical trials with advanced breast germline mutation: from a prospective study, BRCA1 cancer have convincingly demonstrated that the poly- carriers will develop breast cancer by age 80 years with a ADP-ribose-polymerase (PARP)-inhibitors olaparib probability of 72%, and BRCA2 carriers, with a prob- and talazoparib prolong progression-free survival of ability of 69% [2]. Mutation carriers who have been BRCA1 and BRCA2 mutation carriers beyond what diagnosed with breast cancer have a 26% (BRCA2) to conventional chemotherapy regimens are able to achieve 40% (BRCA1) chance of developing contralateral breast in this setting [6e13]. A large adjuvant trial investigating cancer in the following 20 years. Women with a BRCA the efficacy of olaparib in BRCA1/2 germline mutation variant also carry a substantial risk for developing carriers with early breast cancer (OLYMPIA) is ongoing ovarian cancer, which ranges from 17% in BRCA2 to [14]. 44% in BRCA1 carriers, and which is considerably With the availability of Next Generation Sequencing higher than the 2% lifetime risk of the general (NGS), it is now possible to offer mutational analysis population [2,3]. for BRCA1 and BRCA2 and several other moderate- to Knowledge of the individual mutation status thus not high-risk penetrance genes that elevate the lifetime risk only allows to assess individual cancer risks and to of developing breast cancer as part of a diagnostic consider intensified early detection strategies or risk routine with a short turnaround time and diminishing 56 C.F. Singer et al. / European Journal of Cancer 106 (2019) 54e60 costs. The increasing role of and demand for genetic circulated in an iterative open email process until testing for therapeutic purposes has infrastructural, consensus was reached. The results reported below are legislative and financial consequences. In the light of explained based on the votes and discussions of the these recent developments and the arising clinical and panel. ethical questions, a panel of experts in BRCA testing For the purpose of the consensus meeting, the term representing 17 European countries and Israel convened ‘predictive testing’ was defined as genetic testing of to assemble information on the current status of BRCA healthy individuals with a personal/familial (hereditary) and other gene testing across Europe and Israel and to breast and/or ovarian cancer (HBOC) background, formulate consensus recommendations for BRCA irrespective of whether an identified pathogenic variant testing in the metastatic breast cancer context. has been described in their family or not, and testing of individuals with a history of breast or ovarian cancer in 2. Methods whom the test result has no direct therapeutic conse- quence. The aim of predictive testing is to provide a Nineteen experts from the following countries genetic cancer risk estimation. ‘Therapeutic testing’ was participated in the consensus process: Austria, Belgium, defined as genetic testing in which the result has a direct Germany, Croatia, Denmark, France, Hungary, Ice- or indirect implication on cancer treatment. The term land, Israel, Italy, Norway, Portugal, Slovakia, ‘pathogenic variant’ was used for class IV and V genetic Slovenia, Spain, Sweden, Switzerland and the UK. The variations [16]. medical specialities represented are as follows: clinical or medical geneticists (32%), oncologists (37%), gynaecol- 3. Recommendations ogists (26%) and surgeons (5%). Forty-three percent worked in an academic setting (for instance non-uni- 3.1. Genetic counselling versityebased national research centres or university- associated public health hospitals or national cancer center with genetic research), 30% at a university, 13% in Given the fact that the identification of pathogenic non-profit organisations, 4% in the private setting and variants in BRCA1 or BRCA2 and other additional 9% in other types of work settings. genes has clinical implications, a growing demand for A consensus method was applied [15]. Twenty-seven counselling and testing is expected in the near future. questions relevant to genetic counselling and testing in While the experts agreed that the availability of high- affected and non-affected women with a familial breast throughput sequencing technologies such as NGS or ovarian cancer background were sent out to panel across Europe is adequate to cover the increased de- members before the meeting and were open for discus- mand for testing in the years to come, the majority of sion and subsequent voting during the expert panel experts (60%) also stated that it will not be possible to meeting. Conflict of interest statements were collected in address the need for predictive and therapeutic coun- writing at the meeting, and panelists with a conflict of selling within the next five years [Q18], if restricted to interest in one or more questions were asked to abstain clinical geneticists exclusively (as currently organised in from voting at the respective question. Questions could most countries). be modified during the voting session if >50% of pan- elists agreed. Most questions were not modified; how- 3.1.1. Counselling in the predictive setting ever, some questions were rendered more precisely Within the next 5 years, the workload for predictive following discussion. The meeting was recorded, and all counselling was expected to increase by 50% by 39% of modifications from the pre-defined questions were experts, and to increase by 100% by another 33% of documented in writing. Voting results were translated experts [Q19]. All experts agreed that waiting times for into panel recommendations conveying the strength of counselling in the predictive setting should not exceed 2 panel support for each recommendation based on the months, but given the limitations in infrastructure for following rules: An ‘agreement/consensus’ was defined appropriate counselling and aftercare, a range of 2e6 as an agreement among >75% of panelists, and a ‘ma- months would be acceptable [Q10]. To decrease waiting jority’ was defined as an agreement among 50%e75% of times in non-affected members from HBOC families, the panelists. The proportion of panelists abstaining from majority of experts (71%) agreed that genetic counselling vote was recorded for each question. in the predictive setting shoulddif possible within na- We present here the results of all panel discussions tional legislationdnot be restricted to clinical geneticists regarding the 27 questions submitted to the panel. In the or genetic counsellors [Q16]. It was also noted that Appendix, the questions asked and the votes obtained predictive counselling of unaffected individuals from for each question are summarised; the individual ques- HBOC families usually involves a broader perspective tion numbers are cross-referenced between this sum- than therapeutic counselling, and usually covers indi- mary and the Appendix (e.g. [Q1] in the summary refers vidual risks for other malignancies, descendant’s risk to to question 1 in the appendix). The written report was inherit mutations, preventive options, lifestyle issues and C.F. Singer et al. / European Journal of Cancer 106 (2019) 54e60 57 so on. And, therefore, professional training, particularly The experts concurred with 94% agreement that of non-geneticists, needs to account for these aspects. In predictive genetic testing in healthy individuals with a addition, predictive counselling often necessitates pro- familial background of breast cancer should not be fessional psycho-oncological support which is best given restricted to BRCA1/2 [Q5]. A set of high-risk actionable by involving trained psycho-oncologists. genes with evidence of clinical impact was defined as a minimum required panel of high risk actionable genes 3.1.2. Counselling in the therapeutic setting [Q7]. This basic gene panel was set to include BRCA1, With regards to genetic counselling in a therapeutic BRCA2, TP53 (particularly if the patient has an early setting, all experts agreed that the workload will increase disease onset or family history suggestive of Li- in the years to come: the majority of experts expected the Fraumeni syndrome) and PALB2. It was, however, workload to double (44%) or increase by more than also agreed that different populations may require 100% (33%) within the next 5 years [Q20]. Because different gene sets. Gene panels discussed at the meeting counselling of already affected individuals is often pri- included BRCA1, BRCA2, TP53, PTEN, PALB2, STK- marily focused on direct therapeutic implications and 11, ATM, CHEK2, CDH-1, MLH1, MSH2, MSH6, recurrence risks, all experts agreed thatdwithin the BRIP1 and RAD51D/C. Different panelists felt a justi- respective national legal frameworkdall adequately fication for a selection of some of these genes, but there trained professions, including clinical geneticists, genetic was an agreement between the experts that the previ- counsellors (physicians and non-physicians), oncologists, ously mentioned four genes represented the absolute surgeons, gynaecologists and trained nurses, should be minimum of genes that should be tested. authorised to perform genetic counselling for affected Experts also agreed (83%) that, given the low cost women [Q15]. In patients with metastatic breast cancer, and high-throughput that can now be achieved by NGS, experts unanimously agreed that the waiting time be- analysing a number of the most common BRCA1/2 tween the indication for genetic counselling and the variant loci (i.e. “hotspot testing”), rather than availability of the result of the genetic analysis should not completely sequencing the whole genes is not acceptable exceed 3 weeks to avoid significant delays in treatment, [Q14]. A possible exception would only be founder because the test results often have immediate therapeutic mutations that represent more than 99% of pathogenic implications for this group of patients [Q11]. In addition, variants identified in the gene of interest in a specific in patients with newly diagnosed early breast cancer, in geographical region or setting. whom surgical decision-making depends on the result of Overall, the panel expressed that, in the setting of genetic testing, preferential testing should be attempted limited resources, the goal of genetic counselling and to prevent delays in curative surgery. testing strategies should rather focus on providing BRCA1/2 testing to a larger number of potential car- 3.1.3. Educational needs riers, than on investigating large gene panels in a smaller The experts concluded that, to enable easily accessible number of individuals. and nationwide genetic counselling, and to guarantee short waiting times particularly in therapeutic counsel- 4.2. Genetic testing in the therapeutic setting ling, educational efforts are urgently needed. Initiatives suggested by the experts included online tutorials, The experts agreed (89%) that currently BRCA1/2 testing webinars or training courses in genetic counselling that is not indicated solely on the basis of a diagnosis of breast could be offered to interested physicians or breast care cancer, because the prevalence of BRCA1/2 germline nurses. Because gene panel testing now also allows mutations is low in the absence of a suggestive family detection of pathogenic variants in genes beyond history, or young age at onset [Q1], although it was shown BRCA1 and BRCA2 that may have substantial clinical that, for example, in Norway more than 60% of identified implications for other malignancies, most major guide- mutation carriers did not have a suggestive family history lines also recommend a multidisciplinary and multilevel [19]. The diagnosis of TNBC, the presence of a family counselling approach [6e9,17,18]. history, young age at onset and the presence of a clinical setting in which the detection of a BRCA1/2 germline 4. Genetic testing mutation would qualify a patient for PARPi treatment are indications for BRCA1/2 testing [Q21]. In these cases, the 4.1. Genetic testing in the predictive setting majority of experts (69%) recommend that testing should be offered after the histological proof of breast cancer and During the panel discussion, questions arose around the should not be restricted to the advanced cancer setting true definition of ‘predictive’ in the light that very often [Q22]. Already today, the presence of a suggestive family testing starts with an affected family member and rela- history and young age at onset are relative indications for tives are tested as a second step. The panel agreed on the expedited BRCA1/2 testing in many countries, if an early definition set forth in the methods section of this breast cancer patient considers bilateral mastectomy in publication. case a germline mutation is detected. 58 C.F. Singer et al. / European Journal of Cancer 106 (2019) 54e60 This implies that with the approval of PARPi in counselling pathway [Q36], which should also involve advanced disease, genetic testing may increasingly informed consent before gene testing. become an integral part of the routine workup in both early and advanced breast cancer. 4.3. Genetic testing in breast cancer tissue When asked to whom genetic testing should be offered in the metastatic setting, the majority of experts The majority of experts (74%; two experts abstained) (60%) voted that BRCA1/2 testing should be offered to agreed that in routine clinical practice genetic testing of all patients, while 20% of experts voted that BRCA1/2 tumour tissue for the detection of somatic BRCA1/2 and testing should be limited to patient with a familial/per- other breast cancereassociated genes should not be part sonal history suggestive of a BRCA1/2 mutation [Q27j. of a diagnostic algorithm in metastatic breast cancer The majority of experts consequently also voted that [spontaneous question arising from discussion]. The BRCA1/2 testing should be offered to patients with presence of somatic gene alterations does not currently human epidermal growth factor receptor 2 (HER2) have a therapeutic consequence and does not absolve positive (59%) [Q26] and HR positive (57%) [Q28] dis- from germline testing. It should, however, be noted that ease, even if family history is not suggestive. This the prevalence of somatic mutations in BRCA1/2 is recommendation was given in light of the recently likely to be higher than previously thought. In a recently published Assessment of the Efficacy and Safety of published article including 273 Swedish breast cancer Olaparib Monotherapy Versus Physicians Choice patients, the likelihood of a BRCA1/2 mutation being Chemotherapy in the Treatment of Metastatic Breast somatic was w1/3, and germline, 2/3 [23]. It was there- Cancer Patients With Germline BRCA1/2 Mutations. fore remarked that, while available evidence does not (OlympiAD) [11] and A Study Evaluating Talazoparib currently support routine BRCA1/2 or panel testing in (BMN 673), a PARP Inhibitor, in Advanced and/or tumour tissue, it is nevertheless important for research Metastatic Breast Cancer Patients With BRCA Muta- purposes and may have clinical consequences in the tion (EMBRACA Study) (EMBRACA) [13] trial re- future. It will therefore be particularly important to sults, and explicitly acknowledged the fact that the determine the prevalence of somatic genetic variants of likelihood for a pathogenic variant in BRCA1/2 in a BRCA1/2 and other breast cancer related genes in met- breast cancer patient, who has no family history, is astatic tissue, and to further improve the quality of ge- around 2% [20]. It was, however, also discussed that netic testing in tissue biopsies. clinical data addressing the efficacy of PARPi in HER2 positive metastatic breast cancer are not available. Therefore, the panelists argued that it is debatable 5. Reporting whether countries with limited health-care resources should implement a more restrictive testing strategy. 5.1. Reporting of mutations and VUS in BRCA1 and Seventy-four percent of experts were in favour of also BRCA2 testing additional breast cancereassociated genes [Q2], even if the efficacy of PARPi has not been clinically The majority of experts (74%) agreed that variants of validated in the non-BRCA1/2 setting, and although few unknown significance (VUS) should be reported in genes are known, in particular PALB2 [21] and BRIP1 BRCA1/2 [Q6]. However, experts felt that, unless a [22], in which a functional alteration might still have variant is classified as pathogenic (i.e. class IV and V potential surgical consequences such as bilateral mas- variants), it should not be used for medical decisions or tectomy or risk reducing bilateral salpingo-oophorec- for predictive testing in relatives at risk. In this context, tomy. However, the therapeutic benefit of risk-reducing it was cautioned that with increasing medical knowl- surgery in mutation carriers with advanced breast cancer edge, the currently used five-tier classification of a is questionable. particular variant may change over time, and genetic Despite the unanimous support for generalised counselling should therefore always address the possi- testing, it was also cautioned that as long as the test bility that already tested individuals might be recon- results particularly in low or intermediate penetrance tacted in case new information on a particular variant genes are not clinically actionable, the mere knowledge results in a re-classification. of ‘having a mutation’ might result in false treatment The experts, unanimously believe that an interna- expectations in affected individuals. The experts felt that tional initiative is preferable over isolated local research, particularly pretest counselling of breast (and ovarian) to study and to subsequently reclassify class III variants cancer patients should convey the limited clinical con- into either class IV/V or class I/II variants, and that sequences particularly in women with non-BRCA1/2 laboratories should be mandated to routinely reevaluate mutations. Pretest counselling should also include im- individual BRCA1/2 sequence variants prospectively to plications of the genetic result for relatives, particularly reclassify, if new evidence becomes available. This in case high penetrance gene mutations are detected. All should already be addressed in a forward-looking experts were in favour of a structured oncological statement in the initial laboratory report. C.F. Singer et al. / European Journal of Cancer 106 (2019) 54e60 59 5.2. Reporting of germline mutations and VUS in gene AZ had no role in the organization of the meeting, in the panels invitation of panel members, and in the selection of questions. Participants were solely selected because of Experts agreed (79%) that in principle, VUS should not scientific expertise in the BRCA field and regional dis- only be reported if in BRCA1/2 but also in other gene tribution, thereby representing European countries and panel genes [Q9]. It was, however, also pointed out by Israel. several experts that VUS in high risk genes would have higher clinical importance than VUS in lower risk genes. Disclosures Consequently, in a setting where large gene panels are used, which also comprise genes which are associated Judith Balmaña, Nicole Bürki, Maria Elisabetta Filieri, with a small or ill-defined increase in cancer risks, 16% Eli Marie Grindedal, Sileni Han, Oskar Johansson, of experts would report VUS only in genes with a high Bella Kaufman, Mateja Krajc, Edith Olah, Kamil relative risk. Forty-seven percent of panelists suggested Pohlodek, Christian F. Singer, and Manuel Teixeira VUS should only be reported in genes with moderate have nothing to declare. Suzette Delaloge received lec- and high relative risk, whereas 37% opted for reporting ture fees from and is a member of the advisory board of all genetic findings, even if the VUS-associated risk is AstraZeneca and Pfizer. D. Gareth Evans has received low [Q8]. For most experts, such a prioritisation was travel grants from AstraZeneca and Amgen. Anna- deemed to be important given the fact that VUS in low Marie Gerdes was a member of the advisory board of and moderate penetrance genes are probably very AstraZeneca in 2016. Niklas Loman received lecture fees common in the general population as well, and their from AstraZeneca and Novartis and has participated in detection in individuals from HBOC families would an advisory board of Amgen. Shani Paluch-Shimon has therefore not have clinical consequences. This is a participated an advisory board of AstraZeneca. Natalija particularly important aspect in the predictive testing of Dedic Plavetic received lecture fees from AstraZeneca. young women in whom the knowledge of a genetic Kerstin Rhiem received lecture fees from and, in 2012, variationdwhich they might perceive as ‘potentially was a member of the advisory board of AstraZeneca. dangerous’dcould result in a significant psychological burden and an ethical dilemma in their life planning. Conflict of interest statement 6. Conclusion None declared. The clinical efficacy of BRCA1/2-specific targeted ther- Acknowledgements apies and the growing demand for predictive testing in women with a HBOC background have resulted in a Margit Hemetsberger, hemetsberger medical services, profound increase in the demand for genetic counselling Vienna, Austria, helped with the writing of this manu- and testing. It is expected that the current restriction of script. Philipp Pappenscheller, Vienna Medical Univer- counselling to clinical geneticists in several countries will sity, helped with the evaluation of the questionnaires lead to a shortage in counselling slots and to a prolon- and voting results and the organisation of the meeting. gation of already long waiting times. Alternative options D Gareth Evans is supported by the NIHR Manchester to provide genetic counselling are needed, and education Biomedical Research Centre. of oncological caregivers (both medical and non- medical) is an appropriate strategy to overcome the Appendix A. Supplementary data current shortages. Genetic testing in both, the predictive and therapeutic setting, should not be limited to BRCA1 Supplementary data to this article can be found online and BRCA2 but should also include genes beyond at https://doi.org/10.1016/j.ejca.2018.10.007. BRCA1/2. VUS in general pose a considerable challenge because of the limited actionability and potential psy- References chological consequences in carriers and their families. Somatic testing in tumour tissue is not presently rec- [1] Bray F, Colombet M, Mery L, Piñeros M, Znaor A, Zanetti R, ommended in breast cancer patients, but ongoing et al. Cancer incidence in five continents, vol. XI (electronic research might challenge this recommendation in the version). In: Cancer IAfRo; 2017. Lyon. [2] Kuchenbaecker KB, Hopper JL, Barnes DR, Phillips KA, future. Mooij TM, Roos-Blom MJ, et al. Risks of breast, ovarian, and contralateral breast cancer for BRCA1 and BRCA2 mutation Role of the funding source carriers. J Am Med Assoc 2017;317(23):2402e16. [3] Chen S, Parmigiani G. Meta-analysis of BRCA1 and BRCA2 penetrance. J Clin Oncol 2007;25(11):1329e33. Travel expenses for (some) participants were spon- [4] Chiba A, Hoskin TL, Hallberg EJ, Cogswell JA, Heins CN, sored by an unrestricted grant from AstraZeneca (AZ). Couch FJ, et al. Impact that timing of genetic mutation diagnosis 60 C.F. Singer et al. / European Journal of Cancer 106 (2019) 54e60 has on surgical decision making and outcome for breast cancer (BC) and a germline BRCA1/2 mutation. Ann BRCA1/BRCA2 mutation carriers with breast cancer. Ann Surg Oncol 2017;28(suppl 5):v43e67. Oncol 2016;23(10):3232e8. [15] Fink A, Kosecoff J, Chassin M, Brook RH. Consensus methods: [5] Cortesi L, Razzaboni E, Toss A, De Matteis E, Marchi I, characteristics and guidelines for use. Am J Public Health 1984; Medici V, et al. A rapid genetic counselling and testing in newly 74(9):979e83. diagnosed breast cancer is associated with high rate of risk- [16] Plon SE, Eccles DM, Easton D, Foulkes WD, Genuardi M, reducing mastectomy in BRCA1/2-positive Italian women. Ann Greenblatt MS, et al. Sequence variant classification and report- Oncol 2014;25(1):57e63. ing: recommendations for improving the interpretation of cancer [6] Cardoso F, Costa A, Senkus E, Aapro M, Andre F, Barrios CH, susceptibility genetic test results. Hum Mutat 2008;29(11): et al. 3rd ESO-ESMO international consensus guidelines for 1282e91. advanced breast cancer (ABC 3). Ann Oncol 2017;28(12):3111. [17] National Institute for Health and Care Excellence. In: Familial [7] Cardoso F, Costa A, Senkus E, Aapro M, Andre F, Barrios CH, breast cancer: classification, care and managing breast cancer and et al. 3rd ESO-ESMO international consensus guidelines for related risks in people with a family history of breast cancer; 2017. advanced breast cancer (ABC 3). Ann Oncol 2017;28(1):16e33. [18] National Comprehensive Cancer Network. Version 1.2018. In: [8] Cardoso F, Costa A, Senkus E, Aapro M, Andre F, Barrios CH, NCCN clinical practice guidelines in oncology (NCCN Guide- et al. Corrigendum to "3rd ESO-ESMO international consensus lines): genetic/familial high-risk assessment: breast and ovarian; guidelines for advanced breast cancer (ABC 3)" [Breast 31 2017. (February 2017) 244-259]. Breast 2017;32:269e70. [19] Grindedal EM, Heramb C, Karsrud I, Ariansen SL, Maehle L, [9] Cardoso F, Costa A, Senkus E, Aapro M, Andre F, Barrios CH, Undlien DE, et al. Current guidelines for BRCA testing of breast et al. 3rd ESO-ESMO international consensus guidelines for cancer patients are insufficient to detect all mutation carriers. Advanced Breast Cancer (ABC 3). Breast 2017;31:244e59. BMC Cancer 2017;17(1):438. [10] Metcalfe K, Gershman S, Ghadirian P, Lynch HT, Snyder C, [20] Nilsson MP, Torngren T, Henriksson K, Kristoffersson U, Tung N, et al. Contralateral mastectomy and survival after breast Kvist A, Silfverberg B, et al. BRCAsearch: written pre-test in- cancer in carriers of BRCA1 and BRCA2 mutations: retrospective formation and BRCA1/2 germline mutation testing in unselected analysis. BMJ 2014;348:g226. patients with newly diagnosed breast cancer. Breast Cancer Res [11] Robson M, Im SA, Senkus E, Xu B, Domchek SM, Masuda N, Treat 2018;168(1):117e26. et al. Olaparib for metastatic breast cancer in patients with a [21] Antoniou AC, Casadei S, Heikkinen T, Barrowdale D, Pylkas K, germline BRCA mutation. N Engl J Med 2017;377(6):523e33. Roberts J, et al. Breast-cancer risk in families with mutations in [12] Tutt A, Tovey H, Cheang MCU, Kernaghan S, Kilburn L, PALB2. N Engl J Med 2014;371(6):497e506. Gazinska P, et al. Carboplatin in BRCA1/2-mutated and triple- [22] Ramus SJ, Song H, Dicks E, Tyrer JP, Rosenthal AN, negative breast cancer BRCAness subgroups: the TNT Trial. Intermaggio MP, et al. Germline mutations in the BRIP1, Nat Med 2018;24(5):628e37. BARD1, PALB2, and NBN genes in women with ovarian cancer. [13] Litton JK, Rugo HS, Ettl J, Hurvitz SA, Goncalves A, Lee KH, J Natl Cancer Inst 2015;107(11). et al. Talazoparib in patients with advanced breast cancer and a [23] Winter C, Nilsson MP, Olsson E, George AM, Chen Y, Kvist A, germline BRCA mutation. N Engl J Med 2018;379(8):753e63. et al. Targeted sequencing of BRCA1 and BRCA2 across a large [14] Tutt A, Kaufman B, Garber J, Gelber R, McFadden E, Goessl C, unselected breast cancer cohort suggests that one-third of muta- et al. 216TiP - OlympiA: a randomized Phase III trial of olaparib tions are somatic. Ann Oncol 2016;27(8):1532e8. as adjuvant therapy in patients with high-risk HER2-negative