Alcohol Consumption, Cigarette Smoking, and Risk of Breast Cancer for BRCA1 and BRCA2 Mutation Carriers: Results from The BRCA1 and BRCA2 Cohort Consortium

Carriers of pathogenic variants (mutations) in the BRCA1 and BRCA2 genes are at very high risk of developing breast cancer and ovarian cancer. We recently reported cumulative risks of breast cancer to 80 years of 72% [95% confidence interval (CI), 65%–79%] for BRCA1 mutation carriers and 69% (95% CI, 61%–77%) for BRCA2 mutation carriers based on prospective follow-up of unaffected female mutation carriers (1). However, the associations of lifestyle risk factors on breast cancer risk for BRCA1 and BRCA2 (BRCA1/2) mutation carriers remain uncertain. The Oxford collaborative reanalysis of 53 epidemiologic studies concluded that for women unselected for family history, alcohol consumption was associated with increased breast cancer risk, while there was no association between smoking and breast cancer risk (2). However, some recent studies have found that breast cancer risk may be increased if smoking starts early in life, that is, before menarche or a first full-term pregnancy (FFTP; refs. 3–5). Of the studies that have attempted to identify lifestyle factors that modify breast cancer risk for BRCA mutation carriers, few have examined associations with smoking or alcohol consumption and the results are inconsistent (6–15), possibly due to methodologic limitations and small sample sizes. In view of the very high breast cancer risk for BRCA1/2 mutation carriers, together with the well-known carcinogenic and mutagenic activity of alcohol metabolites (16) and tobacco components (17), and the widespread consumption of alcohol and tobacco, it is important to derive reliable estimates of the associations of alcohol and tobacco consumption with breast cancer risk for BRCA1 and BRCA2 mutation carriers. Moreover, given the role of BRCA1 and BRCA2 in DNA repair, it is plausible that smoking and alcohol consumption could have a disproportionate effect for mutation carriers at least in terms of absolute risk. Furthermore, recent experimental data have shown a haplo-insufficiency for BRCA2 and a replication fork instability in BRCA2 heterozygous cells induced by acetaldehyde, an endogenous product of alcohol catabolism (18).

To provide more reliable estimates of the associations of these lifestyle factors with breast cancer risk for mutation carriers, we analyzed data from the largest available cohort of nearly 10,000 BRCA1 and BRCA2 mutation carriers (1) and compared the results from this prospective analysis with the results from the retrospective analysis from same cohort.

We harmonized risk factor and follow-up data from three prospective cohorts: The International BRCA1/2 Carrier Cohort Study (IBCCS; ref. 19), the Kathleen Cuningham Foundation Consortium for Research Into Familial Breast Cancer (kConFab) Follow-Up Study (20, 21), and the Breast Cancer Family Registry (BCFR; ref. 22). The combined cohort (“The BRCA1 and BRCA2 Cohort Consortium”) included data from 21 centers in Western countries (Supplementary Table S1). The total cohort enrolled 9,845 BRCA1 and BRCA2 mutation carriers ages 18–80 years (after excluding 14 carriers of a mutation in both genes; refs. 19, 23). Sixty-six percent of the study participants were enrolled in one of the five ongoing nationwide studies in the United Kingdom and Ireland [Epidemiological Study of Familial Breast Cancer (EMBRACE)], France [Gene Etude Prospective Sein Ovaire (GENEPSO)], the Netherlands [Hereditary Breast and Ovarian cancer study Netherlands (HEBON)], Australia and New Zealand (kConFab), or Austria [Medical University of Vienna (MUV)]. The other studies were based on regional clinical genetic centers or were population based (three centers of the BCFR).

Women were eligible for this analysis if they were 18–80 years of age and had tested positive for a pathogenic BRCA1 or BRCA2 mutation. The total group with follow-up for a first breast cancer and eligible for retrospective or prospective analyses consisted of 9,845 women (9,232 for retrospective and 3,886 for prospective analysis), including 6,032 BRCA1 and 3,813 BRCA2 mutation carriers (Fig. 1). Women who were unaffected with breast cancer at baseline were excluded from prospective analyses if either: they had ovarian cancer (415 BRCA1 carriers and 142 BRCA2); other cancer (146 BRCA1 and 141 BRCA2); risk-reducing mastectomy (RRM; 298 BRCA1 and 139 BRCA2); or did not have follow-up data (360 BRCA1; 226 BRCA2). Participants provided written informed consent and each study was approved by the relevant ethical committee. Study participants were invited to complete a baseline questionnaire at enrollment and regular follow-up questionnaires. The questionnaires requested detailed information on known or suspected risk factors for breast and ovarian cancer. The primary sources of information on cancer occurrence were: self-report via questionnaire only (six studies, 8% of the study group), self-report with medical record validation (two studies, 37%), medical records (four studies, 18%), and linkage to cancer registries (four studies, 37%), although some studies had a mix of these diagnostic sources. Information on vital status was obtained from municipal or death registries or from contact with family members.

We collected information on ever smoking (defined as at least one pack of cigarettes per month for 1 year), current smoking intensity (average number of cigarettes per day), age started smoking, age stopped smoking, and total duration of smoking (in years) and average number of cigarettes per day during this period. Questionnaires also asked about ever alcohol use (at least one glass per month for 1 year), alcohol use in the last year (i.e., current use), and total years of consumption. In most studies, separate questions were asked about types of alcohol and for each type the amount consumed per week. Some studies asked about alcohol use at age 20 years; women in studies without this information (e.g., BCFR) were treated as missing for variables related to alcohol consumption at age 20 years.

After data harmonization across studies, smoking variables were converted to ever/never smoking, number of cigarettes per day (current or past for ex-smokers) in five categories (0; 1–5, 6–10, 11–20, and >20), years of smoking and estimated number of pack-years in three categories (<1, 1–20, and >20), age started in three categories (≤15, 16–19, and ≥20 years), and timing relative to their FFTP. Alcohol variables were converted to ever/never, and total average number of standard drinks per day at age 20 years and in the year prior to completing the questionnaire.

To assess the association between alcohol and tobacco consumption and the risk of breast cancer, we used Cox proportional hazards regression models. Women were eligible for prospective analyses if they were free of cancer and had no history of RRM at the start of follow-up (enrollment/baseline questionnaire or mutation test, whichever came last); for participants recruited in a research setting, follow-up was considered to begin at enrollment. The primary endpoint was breast cancer [invasive (n = 393) or in situ (n = 33) for the prospective analyses] diagnosed more than 1 month after enrollment. The censoring event was the first of diagnosis of primary breast cancer (invasive or in situ), diagnosis of another cancer, RRM, last questionnaire, last information from external source (e.g., linkage), loss to follow-up, age 80 years, or death. Alcohol and tobacco variables were analyzed as fixed in the models because timing of changes in consumption was too uncertain to generate time-dependent variables. Analyses were adjusted for alcohol consumption (ever vs. never) when tobacco consumption was analyzed and for smoking (ever vs. never) when alcohol consumption was analyzed. Because the consumption of alcoholic beverages and tobacco consumption might interact with other breast cancer risk factors (24, 25), we also performed analyses adjusted on additional potential confounders like, age at menarche (<12, ≥12–<13, ≥13–<14, ≥14–<15, and ≥15 years, age missing, or never had menstrual period), age at FFTP (<30 and ≥30+nulliparous), number of full-term pregnancies (0, 1, and ≥2), body mass index (<18.5, 18.5–24.9, 25–29.9, 30 kg/m²or greater, and missing), oral contraceptive use (ever, never, and missing), bilateral oophorectomy (yes and no), and number of affected relatives with breast cancer (0, 1, ≥2, and unknown). We conducted separate analyses for BRCA1 and BRCA2 mutation carriers. We stratified for birth cohort and study and used robust variance estimation to account for familial clustering.

In addition to the prospective analysis, we conducted full-cohort retrospective analyses, in which follow-up was assumed to start at birth and women were followed until the first of diagnosis of primary breast cancer (invasive or in situ), diagnosis of another cancer, RRM, start of prospective follow-up (baseline questionnaire or mutation test, whichever came last), or age 80 years. Thus, there was no overlap in follow-up period for individual women included in the retrospective and prospective analyses. Because of nonrandom sampling of prevalent cases of breast cancer, all analyses of retrospective data were performed using the weighted regression approach described by Antoniou and colleagues (26). Because changes in habits might occur after a breast cancer diagnosis, the number of glasses per day consumed during the last year and the number of pack years at the date of baseline questionnaire were not included in the retrospective analyses.

To minimize potential survival bias, we also conducted an additional retrospective analysis of a pseudoincident cohort, which was defined as follow-up starting 5 years before enrollment/baseline questionnaire, and thus only included cases diagnosed within the 5 years prior to enrolment.

Analyses were stratified by birth cohort into four groups (≤1950, 1951–1959, 1960–1968, and ≥1969 for retrospective analyses and ≤1957, 1958–1966, 1967–1974, and ≥1975 for prospective analyses). We also assessed associations by birth cohort and study. All statistical analyses were performed using STATA (version 14, StataCorp).

Tables 1 and 2 summarize the descriptive statistics of the two cohorts of BRCA1 and BRCA2 mutation carriers, respectively. Comparison of the retrospective and prospective analysis identified differences in distributions of cigarette consumption: we observed more ex-smokers and never smokers (i.e., noncurrent smokers) in the retrospective analysis than in the prospective analysis for both BRCA1 (84.9% vs. 75.5% for cases and 81.1% vs. 79.0% for unaffected women) and BRCA2 (85.0% vs. 84.1% for cases and 82.4 % vs. 81.3% for unaffected women) mutation carriers.

For BRCA1 mutation carriers, there were no associations between breast cancer risk and the alcohol measures examined, except for reduced risk associated with higher alcohol consumption in the retrospective analysis, with an HR_R of 0.59 (95% CI, 0.43–0.81; P = 0.001) for more than two glasses of alcohol consumed at age 20 years when compared with zero glasses of alcohol (Table 3). However, no associations were observed with alcohol consumption at age 20 years or at baseline in the prospective analyses.

There were no associations with ever smoking, number of cigarettes smoked, pack-years, or age at start smoking, in either the prospective or retrospective analysis. However, among parous women, increased breast cancer risk was associated with more than 5 years of smoking before their FFTP in both prospective and retrospective analyses when compared with parous women who never smoked (HR_R = 1.19; 95% CI, 1.02–1.39 and HR_P = 1.36; 95% CI, 0.99–1.87, respectively). Among nulliparous women, there was no evidence of an association with ever smoking when compared with never smoking (HR_R = 0.97; 95% CI, 0.74–1.27 and HR_P = 1.20; 95% CI, 0.68–2.12). Figure 2 displays the cumulative risks of breast cancer for women who smoked for more than 5 years before their FFTP compared with parous never smokers for BRCA1 mutation carriers.

Both, ever use of smoking and alcohol drinking were associated with increased breast cancer risk when compared with women who neither smoked nor drank alcohol for BRCA2 mutation carriers, but only in the retrospective analysis (ever alcohol consumption in nonsmokers HR_R = 1.32; 95% CI, 1.03,1.70; P = 0.03 and ever smoking in nondrinkers HR_R = 1.37; 95% CI, 1.00–1.89; P = 0.05), ever alcohol and ever smoking (i.e., at least one glass per month for 1 year plus at least one pack of cigarettes per month for 1 year; HR_R = 1.43; 95% CI, 1.13–1.81; P = 0.003; Table 4). Similar to the findings for BRCA1 mutation carriers, in both prospective and retrospective analyses we observed an increased breast cancer risk associated with having smoked more than 5 years before a FFTP (HR_R = 1.25; 95% CI, 1.01–1.55 and HR_P = 1.30; 95% CI, 0.83–2.02, respectively), but the estimates were statistically significant only in the retrospective analysis.

Results from retrospective analyses that used the pseudoincident cohort were consistent with those from the full-cohort retrospective analyses, except for BRCA1 mutation carriers where smoking more than 5 years before a FFTP point estimate slightly lowered and significance disappeared (Supplementary Data, Supplementary Table S2). We observed no significant heterogeneity in the HRs for smoking more than 5 years before a FFTP (Supplementary Figs. S1 and S2), with the exception of heterogeneity by birth cohort for the prospective analysis for BRCA1 mutation carriers which is due to the most recent birth cohort where the HR_p was high, although bounded by a large CI (see Supplementary Fig. S2A). There was no significant heterogeneity by study group with regard to the association we observed for high alcohol consumption in the retrospective analysis for BRCA1 or BRCA2 mutation carriers (P = 0.19 and 0.14, respectively; data not shown).

In our primary analyses, we did not adjust the analyses for other possible confounders because most of the other risk factors for breast cancer are unlikely to be correlated with the primary alcohol and smoking exposures of interest. The multivariable-adjusted results are presented in Supplementary Tables S3 and S4. As expected, multivariable adjustment did not materially change the HR estimates and more importantly, the overall conclusions of the study.

We performed separate analyses for BRCA1 and BRCA2 cohorts based on the hypothesis that the role of the two genes may be different in response to carcinogens from alcohol and tobacco. However, we also performed a pooled analysis which, again, did not change drastically our initial findings nor our conclusions (Supplementary Table S5).

Using data from the largest international cohort of BRCA1 and BRCA2 mutation carriers, we examined associations with alcohol consumption and smoking separately, using both independent retrospective and prospective data. We found no evidence of an overall association between cigarette consumption with breast cancer risk, except for the BRCA2 mutation carriers in the retrospective analysis. However, among parous women, we observed that mutation carriers who smoked more than 5 years before their FFTP had a significantly increased risk of breast cancer. This association was seen in both prospective and retrospective analyses, and was seen for both BRCA1 and BRCA2 mutation carriers, although the confidence limits for BRCA2 mutation carriers were wider. The consistency of these findings for mutations carriers of either gene, as well as similar point estimates between prospective and retrospective analyses support the overall conclusion that this time window prior to breast tissue differentiation from pregnancies may be a particularly sensitive window for environmental carcinogenesis.

Unlike in the general population (2, 27) and in accordance with other studies on BRCA1 and BRCA2 mutation carriers (6, 28), our findings do not support a positive association between alcohol intake and breast cancer risk, although power was somewhat limited to detect the relatively modest association observed in prior studies (3, 27).

Findings from studies that have examined associations between smoking and breast cancer risk for BRCA1 and BRCA2 mutation carriers have been inconsistent. Some reported a null association (12–15), two reported a negative association (9, 10), and two reported a positive association (8, 13), although the latter study showed this association only for BRCA1 mutation carriers with a history of smoking (13). While retrospective studies have the advantage of larger size and full life history of smoking, prospective studies have the advantage that reporting of behaviors is not influenced by disease.

In our study, women with BRCA1 mutations who drank more than two glasses of alcohol per day were at decreased breast cancer risk, but only in the retrospective analyses. This discrepancy in results between the two designs for heavier consumers might be explained by survival bias. While tobacco consumption has been suggested as a poor prognostic factor, particularly for women with a diagnosis of triple-negative and luminal A–like breast tumors (29), the association of alcohol with prognosis is less clear. Regular drinking of 0.5 standard drinks or more per day has been shown to be associated with higher risk of breast cancer recurrence, particularly among postmenopausal women (30). Therefore, if women who are heavy consumers are more likely to die after a diagnosis of breast cancer than nondrinking women with breast cancer, the inclusion of prevalent cases in a retrospective analysis may bias results toward unity or even lead to an artefactual negative association (8).

Major strengths of our study include the large sample size for both retrospective and prospective cohorts with very good follow-up and the largest number of BRCA1 and BRCA2 prospective mutation carrier breast cancer cases studied to date. Potential weaknesses include the fact that information on alcohol intake and tobacco consumption was self-reported with accompanying potential exposure misclassification and the potential for the retrospective analyses to be affected by survival bias due to the inclusion of prevalent cases. However, the prospective part of our study minimized recall and survival biases.

As in the general population (5), we found a consistent association of increased breast cancer risk with cigarette smoking for mutation carriers who smoked for more than 5 years before their FFTP. The period preceding a FFTP has been shown to be a critical period for breast carcinogenesis (31, 32), particularly for women with a mutation in BRCA1 or BRCA2 (33), and potentially even more so for women who accumulated DNA defects during the years before a FFTP because of smoking (Fig. 2).

With the exception of the association with smoking for more than 5 years before a FFTP, no associations were found for most smoking-related variables for either BRCA1 or BRCA2 mutation carriers. Similarly, no association with alcohol consumption was found in the prospective analysis. However, associations with these two lifestyle factors might be complex and need more detailed information on consumption (e.g., quantities and calendar years of starting and stopping) and timing to be able to prospectively investigate them as time-dependent exposures, and extended follow-up might shed further light upon associations of smoking and alcohol with breast cancer risk for BRCA1 and BRCA2 mutation carriers.

In summary, we found no substantial association of breast cancer risk with alcohol consumption or smoking except for women who smoked for more than 5 years before their FFTP. These findings suggest that smoking during the prereproductive years may increase breast cancer risk for mutation carriers, warranting further investigation.

D.G. Evans is a consultant for AstraZeneca. M. Friedlander reports receiving speakers bureau honoraria from AstraZeneca, MSD, Takeda, Lilly, and Novartis, and is a consultant/advisory board member for AbbVie. C.F. Singer reports receiving speakers bureau honoraria from Novartis and Amgen, and is a consultant/advisory board member for Roche. No potential conflicts of interest were disclosed by the other authors.

The content of this manuscript does not necessarily reflect the views or policies of the National Cancer Institute or any of the collaborating centers in the BCFR, nor does mention of trade names, commercial products, or organizations imply endorsement by the U.S. government or the Breast Cancer Family Registry.

Conception and design: M.B. Terry, A.C. Antoniou, K.-A. Phillips, C.F. Singer, H. Olsson, J.L. Hopper, R.L. Milne, D.F. Easton, F.E. Van Leeuwen, M.A. Rookus, N. Andrieu, D.E. Goldgar

Development of methodology: H. Li, A.C. Antoniou, M.B. Daly, J.L. Hopper, D.F. Easton, M.A. Rookus, N. Andrieu, D.E. Goldgar

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): M.B. Terry, A.C. Antoniou, K.-A. Phillips, K. Kast, C. Engel, C. Noguès, D. Stoppa-Lyonnet, C. Lasset, P. Berthet, V. Mari, O. Caron, D. Barrowdale, D. Frost, C. Brewer, D.G. Evans, L. Izatt, L. Side, L. Walker, M. Tischkowitz, M.T. Rogers, M.E. Porteous, H.E.J. Meijers-Heijboer, J.J.P. Gille, M.J. Blok, N. Hoogerbrugge, M.B. Daly, I.L. Andrulis, S.S. Buys, E.M. John, S.-A. McLachlan, M. Friedlander, Y.Y. Tan, A. Osorio, T. Caldes, A. Jakubowska, J. Simard, C.F. Singer, E. Olah, M. Navratilova, L. Foretova, A.-M. Gerdes, B. Arver, H. Olsson, R.K. Schmutzler, J.L. Hopper, R.L. Milne, D.F. Easton, M.A. Rookus

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): H. Li, M.B. Terry, D.G. Evans, S.-A. McLachlan, T. Caldes, C.F. Singer, R.L. Milne, F.E. Van Leeuwen, M.A. Rookus, N. Andrieu, D.E. Goldgar

Writing, review, and/or revision of the manuscript: H. Li, M.B. Terry, K.-A. Phillips, K. Kast, C. Engel, C. Noguès, D. Stoppa-Lyonnet, O. Caron, D. Barrowdale, D. Frost, D.G. Evans, L. Izatt, L. Walker, M. Tischkowitz, M.T. Rogers, H.E.J. Meijers-Heijboer, M.J. Blok, N. Hoogerbrugge, M.B. Daly, I.L. Andrulis, S.S. Buys, E.M. John, S.-A. McLachlan, M. Friedlander, Y.Y. Tan, A. Osorio, A. Jakubowska, J. Simard, C.F. Singer, E. Olah, L. Foretova, A.-M. Gerdes, H. Olsson, R.K. Schmutzler, J.L. Hopper, R.L. Milne, D.F. Easton, F.E. Van Leeuwen, M.A. Rookus, N. Andrieu, D.E. Goldgar

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): T.M. Mooij, C. Engel, D. Barrowdale, D. Frost, D.G. Evans, C.F. Singer, M.-J. Roos-Blom, J.L. Hopper

Study supervision: K.-A. Phillips, J.L. Hopper, N. Andrieu, D.E. Goldgar

Other (coordinator of the GENEPSO study): C. Noguès

The Breast Cancer Family Registry (BCFR), and authors Mary Beth Terry, Irene Andrulis, David Goldgar, Saundra Buys, Esther John, and Mary Daly, were supported by grant UM1CA164920 from the NCI. This work was partially supported by the Spanish Ministry of Economy and Competitiveness (MINECO) SAF2014-57680-R and the Spanish Research Network on Rare diseases (CIBERER). This work was supported by the Canadian Institutes of Health Research for the “CIHR Team in Familial Risks of Breast Cancer” program (grant no. CRN-87521) and the Ministry of Economic Development, Innovation and Export Trade (grant no. PSR-SIIRI-701). The DKFZ study was supported by the German Cancer Research Center (DKFZ). EMBRACE and D.F. Easton are supported by Cancer Research UK grants C1287/A10118 and C1287/A11990. D. Gareth Evans is supported by an NIHR grant to the Biomedical Research Centre, Manchester (IS-BRC-1215-20007). The Investigators at The Institute of Cancer Research and The Royal Marsden NHS Foundation Trust are supported by an NIHR grant to the Biomedical Research Centre at The Institute of Cancer Research and The Royal Marsden NHS Foundation Trust. A.C. Antoniou is funded by Cancer Research UK grants C12292/A20861 and C12292/A11174. The German Consortium of Hereditary Breast and Ovarian Cancer (GC-HBOC) is supported by the German Cancer Aid (grant no. 110837, to R.K. Schmutzler). This work was supported by LIFE – Leipzig Research Center for Civilization Diseases, Universität Leipzig. LIFE is funded by means of the European Union, by the European Regional Development Fund (ERDF), and by means of the Free State of Saxony within the framework of the excellence initiative. The national French cohort, GENEPSO, was supported by a grant from the Fondation de France and the Ligue Nationale Contre le Cancer and is being supported by a grant from INCa as part of the European program ERA-NET on Translational Cancer Research (TRANSCAN-JTC2012, no. 2014-008). Hospital Clinico San Carlos (HCSC) was supported by a grant RD12/0036/0006 and 15/00059 from ISCIII (Spain) and partially supported by European Regional Development FEDER funds. The HEBON study is supported by the Dutch Cancer Society grants NKI1998-1854, NKI2004-3088, NKI2007-3756, the Netherlands Organisation of Scientific Research grant NWO 91109024, the Pink Ribbon grants 110005 and 2014-187.WO76, the BBMRI grant NWO 184.021.007/CP46, and the Transcan grant JTC 2012 Cancer 12-054. The International Hereditary Cancer Centre (IHCC) was supported by grant PBZ_KBN_122/P05/2004 and The National Centre for Research and Development (NCBR) within the framework of the international ERA-NET TRANSAN JTC 2012 application no. Cancer 12-054 (contract no. ERA-NET-TRANSCAN/07/2014). This work was also supported by grants to kConFab and the kConFab Follow-Up Study from Cancer Australia (809195, 1100868), the Australian National Breast Cancer Foundation (IF 17), the National Health and Medical Research Council (454508, 288704, 145684), the NIH (1RO1CA159868), the Queensland Cancer Fund, the Cancer Councils of New South Wales, Victoria, Tasmania and South Australia, and the Cancer Foundation of Western Australia. K.-A. Phillips is an Australian National Breast Cancer Foundation fellow. L. Foretova was supported by MH CZ - DRO (MMCI, 00209805) and by MEYS - NPS I - LO1413 (to L. Foretova, M. Navratilova). E. Olah and The Hungarian Breast and Ovarian Cancer Study were supported by Hungarian Research Grants KTIA-OTKA CK-80745, NKFI OTKA K-112228, and the Norwegian EEA Financial Mechanism HU0115/NA/2008-3/ÖP-9. H. Olsson and Lund-BRCA collaborators are supported by the Swedish Cancer Society, Lund Hospital Funds, and European Research Council Advanced Grant ERC-2011-294576. Stockholm-BRCA collaborators are supported by the Swedish Cancer Society. BCFR wishes to thank members and participants in the Breast Cancer Family Registry from the New York, Northern California, Ontario, Philadelphia, and Utah sites for their contributions to the study. BCFR-Australia wish to acknowledge Maggie Angelakos, Judi Maskiell, Gillian Dite, and Helen Tsimiklis. CNIO thanks Alicia Barroso, Rosario Alonso, and Guillermo Pita for their assistance. We acknowledge the GENEPSO Centers (the Coordinating Center: Institut Paoli-Calmettes, Marseille, France: Catherine Noguès, Lilian Laborde, Pauline Pontois, Emanuelle Breysse, Margot Berline) and the Collaborating Centers (Institut Curie, Paris, France: Dominique Stoppa-Lyonnet, Marion Gauthier-Villars, Bruno Buecher, Chrystelle Colas; Institut Gustave Roussy, Villejuif, France: Olivier Caron; Hôpital René Huguenin/Institut Curie, Saint Cloud: Catherine Noguès, Emmanuelle Mouret-Fourme, Claire Saule, Chrystelle Colas; Centre Paul Strauss, Strasbourg, France: Jean-Pierre Fricker; Centre Léon Bérard, Lyon, France: Christine Lasset, Valérie Bonadona, Sophie Dussard; Centre François Baclesse, Caen, France: Pascaline Berthet; Hôpital d'Enfants CHU Dijon – Centre Georges François Leclerc, Dijon, France: Laurence Faivre; Centre Alexis Vautrin, Vandoeuvre-les-Nancy, France: Elisabeth Luporsi; Centre Antoine Lacassagne, Nice, France: Véronique Mari; Institut Claudius Regaud, Toulouse, France: Laurence Gladieff; Réseau Oncogénétique Poitou Charente, Niort: Paul Gesta, Stéphanie Chieze-Valéro; Institut Paoli-Calmettes, Marseille, France: Catherine Noguès, Jessica Moretta Hagay Sobol, François Eisinger, Cornel Popovici; Institut Bergonié, Bordeaux, France: Michel Longy; Centre Eugène Marquis, Rennes: Louise Grivelli; GH Pitié Salpétrière, Paris, France: Chrystelle Colas, Florent Soubrier, Patrick Benusiglio; CHU Arnaud de Villeneuve, Montpellier, France: Isabelle Coupier, Pascal Pujol, Carole Corsini; Centres Paul Papin/ICO, Angers: Marie-Emmanuelle Morin-Meschin; Clinique Catherine de Sienne, Nantes, France: Alain Lortholary; Centre Oscar Lambret, Lille, France: Claude Adenis, Audrey Maillez; Institut Jean Godinot, Reims, France: Tan Dat Nguyen; Centre René Gauducheau, Nantes, France: Capucine Delnatte, Caroline Abadie; Centre Henri Becquerel, Rouen, France: Julie Tinat, Isabelle Tennevet; Hôpital Civil, Strasbourg, France: Christine Maugard; Centre Jean Perrin, Clermont-Ferrand, France: Yves-Jean Bignon, Mathilde Gay Bellile; Polyclinique Courlancy, Reims, France: Clotilde Penet; Clinique Sainte Catherine, Avignon, France: Hélène Dreyfus; Hôpital Saint-Louis, Paris, France: Odile Cohen-Haguenauer; CHRU Dupuytren, Limoges, France: Brigitte Gilbert, Laurence Venat-Bouvet; CHU de Grenoble: Dominique Leroux, Clémentine Legrand, Hélène Dreyfus; Hôpital de la Timone, Marseille, France: Hélène Zattara-Cannoni; Hôpital Jacques Monod, le Havre, France: Valérie Layet, Elodie Lacaze; CHU Chambéry, Chambéry, France: Sandra Fert-Ferrer; Hôpital Clarac, Fort de France: Odile Bera; CHU la Milétrie, Poitiers, France: Brigitte Gilbert-Dussardier, David Tougeron, Stéphanie Chieze-Valéro; Hôpital Saint Louis, la Rochelle, France: Hakima Lallaoui; CH Pellegrin, Bordeaux, France: Julie Tinat). HCSC acknowledges Alicia Tosar and Paula Diaque for their technical assistance. The Hereditary Breast and Ovarian Cancer Research Group Netherlands (HEBON) consists of the following Collaborating Centers: Netherlands Cancer Institute (coordinating center), Amsterdam, the Netherlands: M.A. Rookus, F.B.L. Hogervorst, F.E. van Leeuwen, M.A. Adank, M.K. Schmidt, N.S. Russell, J.L. de Lange, R. Wijnands, D.J. Jenner; Erasmus Medical Center, Rotterdam, the Netherlands: J.M. Collée, A.M.W. van den Ouweland, M.J. Hooning, C. Seynaeve, C.H.M. van Deurzen, I.M. Obdeijn; Leiden University Medical Center, Leiden, the Netherlands: C.J. van Asperen, J.T. Wijnen, R.A.E.M. Tollenaar, P. Devilee, T.C.T.E.F. van Cronenburg; Radboud University Nijmegen Medical Center, Nijmegen, the Netherlands: C.M. Kets, A.R. Mensenkamp; University Medical Center Utrecht, Utrecht, the Netherlands: M.G.E.M. Ausems, R.B. van der Luijt, C.C. van der Pol; Amsterdam Medical Center, Amsterdam, the Netherlands: C.M. Aalfs, H.E.J. Meijers-Heijboer, T.A.M. van Os; VU University Medical Center, Amsterdam, the Netherlands: K. van Engelen, J.J.P. Gille, Q. Waisfisz; Maastricht University Medical Center, Maastricht, the Netherlands: E.B. Gómez-Garcia, M.J. Blok; University of Groningen, Groningen, the Netherlands: J.C. Oosterwijk, A.H. van der Hout, M.J. Mourits, G.H. de Bock; The Netherlands Comprehensive Cancer Organisation (IKNL): S. Siesling, J.Verloop; the nationwide network and registry of histo- and cytopathology in the Netherlands (PALGA): L.I.H. Overbeek. HEBON thanks the study participants and the registration teams of IKNL and PALGA for part of the data collection. INHERIT would like to thank Dr. Martine Dumont for sample management and skillful assistance. We thank Stephanie Nesci, Sandra Picken, Lucy Stanhope, Sarah O'Connor, Heather Thorne, Eveline Niedermayr, all the kConFab research nurses and staff, the heads and staff of the Australian and New Zealand Family Cancer Clinics, and the many families who contribute to kConFab for their contributions to this resource. MMCI, Brno, Czech Republic - many thanks to Dita Hanouskova, research nurse, Jitka Berkovcova, research technician, for data collection and management. We wish to thank the Hungarian Breast and Ovarian Cancer Study Group members (Maria Balogh, Janos Papp, Matrai Zoltan, Judit Franko Department of Molecular Genetics, National Institute of Oncology, Budapest, Hungary) and the clinicians and patients for their contributions to this study. Swedish scientists participating as SWE-BRCA collaborators are from Lund University and University Hospital (Lund, Sweden): Håkan Olsson, Carolina Ellberg; and from Stockholm and Karolinska University Hospital (Stockholm, Sweden): Brita Arver.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.