Article Text

Download PDFPDF

Short Report
The BRCA1 c. 5096G>A p.Arg1699Gln (R1699Q) intermediate risk variant: breast and ovarian cancer risk estimation and recommendations for clinical management from the ENIGMA consortium
  1. Setareh Moghadasi1,
  2. Huong D Meeks2,
  3. Maaike PG Vreeswijk3,
  4. Linda AM Janssen1,
  5. Åke Borg4,
  6. Hans Ehrencrona5,6,
  7. Ylva Paulsson-Karlsson7,
  8. Barbara Wappenschmidt8,9,10,
  9. Christoph Engel11,12,
  10. Andrea Gehrig13,14,15,
  11. Norbert Arnold16,
  12. Thomas Van Overeem Hansen17,18,
  13. Mads Thomassen19,
  14. Uffe Birk Jensen20,
  15. Torben A Kruse19,
  16. Bent Ejlertsen18,21,
  17. Anne-Marie Gerdes18,22,
  18. Inge Søkilde Pedersen23,24,
  19. Sandrine M Caputo25,
  20. Fergus Couch26,
  21. Emily J Hallberg27,
  22. Ans MW van den Ouweland28,
  23. Margriet J Collée28,
  24. Erik Teugels29,
  25. Muriel A Adank30,
  26. Rob B van der Luijt31,32,
  27. Arjen R Mensenkamp33,
  28. Jan C Oosterwijk34,
  29. Marinus J Blok35,
  30. Nicolas Janin36,
  31. Kathleen BM Claes37,
  32. Kathy Tucker38,
  33. Valeria Viassolo39,40,
  34. Amanda Ewart Toland41,
  35. Diana E Eccles42,
  36. Peter Devilee3,
  37. Christie J Van Asperen1,
  38. Amanda B Spurdle43,
  39. David E Goldgar44,
  40. Encarna Gómez García35
  1. 1 Department of Clinical Genetics, Leiden University Medical Centre, Leiden, The Netherlands
  2. 2 Huntsman Cancer Institute, University of Utah, Salt Lake City, Utah, USA
  3. 3 Department of Human Genetics, Leiden University Medical Centre, Leiden, The Netherlands
  4. 4 Division of Oncology, Department of Clinical Sciences, Lund University, Lund, Sweden
  5. 5 Department of Clinical Genetics, Lund University, Lund, Sweden
  6. 6 Department of Clinical Genetics, Laboratory Medicine, Office for Medical Services, Lund University, Lund, Sweden
  7. 7 Department of Immunology, Genetics and Pathology, Uppsala University, Uppsala, Sweden
  8. 8 Centre of Familial Breast and Ovarian Cancer, University Hospital of Cologne, Cologne, Germany
  9. 9 Department of Gynaecology and Obstetrics and Centre for Integrated Oncology (CIO), University Hospital of Cologne, Cologne, Germany
  10. 10 Centre for Molecular Medicine Cologne (CMMC), University Hospital of Cologne, Cologne, Germany
  11. 11 Institute for Medical Informatics, University of Leipzig, Leipzig, Germany
  12. 12 Department of Statistics and Epidemiology, University of Leipzig, Leipzig, Germany
  13. 13 Centre of Familial Breast and Ovarian Cancer, University Würzburg, Würzburg, Germany
  14. 14 Department of Medical Genetics, University Würzburg, Würzburg, Germany
  15. 15 Institute of Human Genetics, University Würzburg, Würzburg, Germany
  16. 16 Department of Gynaecology and Obstetrics, University Hospital of Schleswig-Holstein, Campus Kiel, Christian-Albrechts University Kiel, Kiel, Germany
  17. 17 Center for Genomic Medicine, University of Copenhagen, Copenhagen, DenmarK
  18. 18 Department of Rigshospitalet, University of Copenhagen, Copenhagen, DenmarK
  19. 19 Department of Clinical Genetics, Odense University Hospital, Odense, Denmark
  20. 20 Department of Clinical Genetics, Aarhus University Hospital, Aarhus, Denmark
  21. 21 Department of Oncology, University of Copenhagen, Copenhagen, Denmark
  22. 22 Department of Clinical Genetics, University of Copenhagen, Copenhagen, Denmark
  23. 23 Section of Molecular Diagnostics, Aalborg University Hospital, Aalborg, Denmark
  24. 24 Department of Clinical Biochemistry, Aalborg University Hospital, Aalborg, Denmark
  25. 25 Institut Curie, Service de génétique, Paris, France
  26. 26 Department of Laboratory Medicine and Pathology, Mayo Clinic, Rochester, Minnesota, USA
  27. 27 Department of Health Sciences Research, Mayo Clinic, Rochester, Minnesota, USA
  28. 28 Department of Clinical Genetics, Erasmus Medical Centre, Rotterdam, The Netherlands
  29. 29 Familial Cancer Clinic and Medical Oncology, University Hospital Brussels, Belgium
  30. 30 Department of Clinical Genetics, VU Medical Centre, Amsterdam, The Netherlands
  31. 31 Division of Biomedical Genetics, University Medical Centre Utrecht, Utrecht, The Netherlands
  32. 32 Department of Genetics, University Medical Centre Utrecht, Utrecht, The Netherlands
  33. 33 Department of Human Genetics, Radboudumc Nijmegen, The Netherlands
  34. 34 Department of Genetics, University of Groningen, University Medical Centre, Groningen, The Netherlands
  35. 35 Department of Clinical Genetics, Maastricht University Medical Centre, Maastricht, The Netherlands
  36. 36 Department of Service de Génétique, Cliniques universitaires Saint-Luc, Bruxelles, Belgium
  37. 37 Centre for Medical Genetics, Ghent University Hospital, Belgium
  38. 38 Hereditary Cancer Service, Prince of Wales (and St George Hospitals) Hospital, Randwick, New South Wales, Australia
  39. 39 Department of Oncogenetics and Cancer Prevention Unit, Geneva University Hospitals, Geneva, Switzerland
  40. 40 Division of Oncology, Geneva University Hospitals, Geneva, Switzerland
  41. 41 Department of Cancer Biology and Genetics, The Ohio State University, Columbus, Ohio, USA
  42. 42 Faculty of Medicine, University of Southampton, Southampton, UK
  43. 43 Genetics and Computational Biology Division, QIMR Berghofer Medical Research Institute, Herston, Brisbane, Australia
  44. 44 Huntsman Cancer Institute and Department of Dermatology, University of Utah School of Medicine Salt Lake City, Salt Lake City, Utah, USA
  1. Correspondence to Dr Encarna Gómez García, Department of Clinical Genetics Maastricht University Medical Centre, Maastricht 6202 AZ, The Netherlands; encarna.gomezgarcia{at}mumc.nl

Abstract

Background We previously showed that the BRCA1 variant c.5096G>A p.Arg1699Gln (R1699Q) was associated with an intermediate risk of breast cancer (BC) and ovarian cancer (OC). This study aimed to assess these cancer risks for R1699Q carriers in a larger cohort, including follow-up of previously studied families, to further define cancer risks and to propose adjusted clinical management of female BRCA1*R1699Q carriers.

Methods Data were collected from 129 BRCA1*R1699Q families ascertained internationally by ENIGMA (Evidence-based Network for the Interpretation of Germline Mutant Alleles) consortium members. A modified segregation analysis was used to calculate BC and OC risks. Relative risks were calculated under both monogenic model and major gene plus polygenic model assumptions.

Results In this cohort the cumulative risk of BC and OC by age 70 years was 20% and 6%, respectively. The relative risk for developing cancer was higher when using a model that included the effects of both the R1699Q variant and a residual polygenic component compared with monogenic model (for BC 3.67 vs 2.83, and for OC 6.41 vs 5.83).

Conclusion Our results confirm that BRCA1*R1699Q confers an intermediate risk for BC and OC. Breast surveillance for female carriers based on mammogram annually from age 40 is advised. Bilateral salpingo-oophorectomy should be considered based on family history.

  • BRCA1
  • R1699Q
  • breastcancer
  • ovarian cancer
  • intermediate cancer risk
  • Surveillance

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Introduction

In 2008, the International Agency for Research on Cancer (IARC) proposed a standardised five-tier classification system applicable to sequence-based results in highly penetrant cancer predisposition genes and linked the likelihood of pathogenicity to clinical actions.1 The multifactorial likelihood model (MLM) is commonly used to calculate the probability of pathogenicity2 of individual BRCA1 and BRCA2 variants. It is used in the IARC five-tier classification system to categorise each variant into a specific class. The MLM combines complementary sources of data (ie, physicochemical properties,3 family history,4 cosegregation of the variant with disease in a family5 and co-occurrence of the variant with a pathogenic BRCA1 or BRCA2 variant in trans6) to determine the probability that a given variant has a cancer risk equivalent to known high-risk pathogenic (predominantly truncating) variants.

The BRCA1 variant c.5096G>A p.Arg1699Gln (hereafter termed BRCA1*R1699Q) was initially classified as class 3 (variant of uncertain significance) using the MLM method.1 A subsequent study7 included functional assays to assess pathogenicity, but did not yield conclusive results. Indeed this variant, located in the BRCA1 carboxyl terminal region of the transcriptional transactivation domain, and at the interface of the phosphopeptide binding region, demonstrated ambiguous behaviour in a variety of functional assays, when compared with the pathogenic BRCA1 variant c.5095C>T p.Arg1699Trp (BRCA1*R1699W) at the same residue, wild-type BRCA1 and other known pathogenic missense variants.7 Other models based on family history analysis of BRCA-ness8 or cosegregation within a family5 also gave inconclusive results.

In 2012, members of the ENIGMA consortium (Evidence-based Network for the Interpretation of Germline Mutant Alleles)9 reported on the family histories of 69 families carrying BRCA1*R1699Q.10 Comparison of BRCA1 carrier prediction scores of probands using the BOADICEA risk prediction tool11 showed that BRCA1*R1699Q variant carriers had family histories that were less ‘BRCA1-like’ than BRCA1*R1699W carriers but more ‘BRCA1-like’ than BRCA-X families (families with no detectable BRCA1 or BRCA2 pathogenic mutation). Second, modified segregation analysis was used in a subset of 30 families and showed lower risks of breast cancer (BC) or ovarian cancer (OC) (estimated cumulative risk to age 70: 24%) than BRCA1*R1699W (58%) and the ‘average’ pathogenic BRCA1 truncating variant (68%).10 Due to the relatively small number of families with cosegregation data in that study, age-specific cancer risks could not be established with a high degree of precision.

The aim of the present study was to update the BC and OC risk estimates associated with BRCA1*R1699Q in a larger series that included newly identified families, as well as some of the previously studied families, which had been updated with cosegregation data as a result of cascade screening. Based on these results, we propose recommendations for the clinical management of the carriers and their family members.

Materials and methods

Data collection

All families participating in this study included one or more individuals referred to a cancer family clinic because of a personal history of BC and/or OC, and/or a family history consistent with hereditary BC and/or OC.

Each index case had a confirmed BRCA1*R1699Q variant. ENIGMA members, including those from centres that had contributed pedigrees to the previous study, were asked to provide updated pedigrees (if possible) and additional families segregating BRCA1*R1699Q identified after the close of enrolment of the previous study. Pedigrees and patient-specific data such as ages at diagnoses and genotypes were collected from a total of 129 families from 11 different countries, of which 91 families had at least one additional person genotyped, and were thus informative for estimating BC and OC risks. From these 91 families, 30 had been included in the segregation analysis in our previous study10 (see online supplementary table S1). When ages of diagnosis were missing, we conservatively assumed them to be age 65, and for unaffected women we imputed their age using other pedigree members using the PedPro suite of programs (www.bjfenglab.org, accessed 21 September 2016).

Supplementary Material

Supplementary Table 1

Statistical analysis

Data sets

In order to account for ascertainment bias, the likelihood of the pedigree phenotypes and BRCA1*R1699Q genotypes was calculated conditional on the pedigree phenotypes and the BRCA1*R1699Q genotype of the index case. Cancer risks were estimated using the following data sets:

The primary analysis (hereafter termed main analysis) included all 129 informative pedigrees from both the previous study and the present recruitment. The second analysis (subanalysis 1) was similar to the main analysis, except that for the genotypes and phenotypes from the previous study only information gathered since the previous study is included. In this analysis, the likelihood was conditioned on the genotype of the index case and pedigree phenotypes of the new families and all genotypes and pedigree phenotypes in the previous pedigrees as they were in the previous analysis in 2012. In fact the index patients carrier status and affected status are not used to estimate the hazard/risk ratios on which the cumulative risks are based. The last analysis (subanalysis 2) included only the 60 pedigrees that were recruited for this study.

Data from subanalyses 1 and 2 are shown in the online supplementary materilas.

Cancer risk estimation methods

BC and OC risks were estimated using modified segregation analysis with the MENDEL package of programs.12 For each data set, the analysis was performed under each of the following assumptions:  (1) the relative risk (RR) across age groups was assumed to be constant; and (2) the RR was assumed to be a continuous, piecewise linear function of age, which was constant before age 40 years and after age 60 years and linear between ages 40 and 60 years. For both models, baseline population incidence rates were assumed to be those for the UK 2003–2007 (Cancer Incidence in Five Continents Reports (IARC-WHO; update November 2010).13

For both these analyses we first used a model assuming a single major gene only (the BRCA1*R1699Q variant) and second a model that included the major gene and a polygenic background effect. From the resulting estimates of BC and OC relative risk, age-specific cumulative risk estimates were calculated based on the cumulative incidence Λ(t): F(t)=1 − exp(−Λ(t)), and the corresponding CIs were calculated using a parametric bootstrap with 5000 replications.14

Results

Descriptive characteristics of the cohort

Our cohort included 129 separate families with a total of 4024 family members, from whom 309 women were proven BRCA1*R1699Q carriers and 173 were proven non-carriers. For 91 families, in addition to genotyping data of the proband, at least one additional genotype was available (see online supplementary table S2). Descriptive characteristics of the cohort about BC and OC cancer history and age distribution are listed in the online supplementary table S2.

BC and OC risks

Online supplementary figure S1 and supplementary table S2 show the age distribution for BC and OC for the female carriers. The sharpest increase of BC occurred between ages 40 and 49. For OC this was between ages 50 and 59. The youngest case of BC was diagnosed at age 25, for OC this was age 35.

Supplementary Material

Supplementary Figure 1

Cumulative risks for this variant by age 70 years are estimated to be 20% (95% CI 13% to 32%) for BC and 6% (95% CI 3% to 25%) for OC. The risks are lower than for high-risk BRCA1 truncating variants and higher than for the general population in all the three data sets. Figure 1 shows the corresponding curves for the main analysis. Online supplementary figures S2 and S3 and supplementary tables S3 and S4 show comparable results for all the data sets under both assumptions.

Figure 1

Cumulative risks (%) for breast cancer (left graph) and ovarian cancer (right graph) by age for carriers of BRCA1*R1699Q based on the main analysis (blue line). The corresponding curves or the cumulative risk conferred by average pathogenic BRCA1 variants (red line) and for the general population (green line) are also shown. Cumulative risks are calculated using segregation analysis, major gene model assuming relative risk as a continuous, piecewise linear function of age.

Effect of other genetic factors on cancer risks

In order to study the effect of other (genetic) factors on risk, HRs were calculated based on the ‘major gene only’ model and the ‘major gene and polygenic’ model under both assumptions.

For the main analysis, HRs for BC are higher in the major gene plus polygenic model compared with the major gene only model, both when assuming constant RR across age groups, and when modelled as a continuous piecewise linear function of age. HRs for OC are higher in the major gene plus polygenic model when assuming constant RR. When assuming RR as a continuous, piecewise linear function of age, the HR is higher for the major gene plus polygenic model when the individual is older than 60 years old, suggesting that modifiers might be especially important for the late-onset disease (table 1). Online supplementary table S5 shows the HRs for the subanalyses.

Table 1

Modified segregation analysis results from MENDEL in the main analysis (a) assuming constant relative risk across age groups and (b) assuming relative risk as a continuous, piecewise linear function of age

Discussion

After publication of the study by Spurdle et al 10 in 2012, many cancer clinics started offering cascade screening to relatives of carriers of the BRCA1*R1699Q variant. However, in the absence of robust estimates of cancer risks, it was not clear whether available guidelines for BRCA carriers would also be suitable for female carriers of BRCA1*R1699Q.

The cumulative risks estimated from the main analysis and the two subanalyses were lower than for the average BRCA1 truncating pathogenic variant, yet still substantially higher than the rates in the general population. Cumulative risk by age 70 years was estimated to be 20% (95% CI 13% to 32%) for BC and 6% (95% CI 3% to 25%) for OC.

Our results strongly confirm our previous findings that this variant has reduced penetrance,10 and can thus be termed an intermediate risk variant conferring risks lower than that for the average pathogenic variant in a high-risk cancer predisposition gene. These risk estimates are consistent with those reported for disease-associated variants in so-called ‘moderate risk’ genes, defined as genes in which pathogenic variants have an RR between 2 and 5.15 16

Interestingly, our results show that the estimated HRs are in general slightly higher when the ‘major gene plus polygenic’ model is used compared with the ‘major gene only’ model, which is especially evident in the late-onset disease (>60 years) group. This means that in addition to BRCA1*R1699Q, other genetic and/or environmental factors seem to contribute to the magnitude of the BC and OC risk in carriers. Indeed, recent literature15–17 indicates that single nucleotide polymorphisms are important determinants of personal cancer risk in women carrying a deleterious disease-associated variant especially in moderate risk genes. As those factors are mostly unmeasured or unknown, an indirect estimation of clustering of risk factors can be deduced taking the family history into account. This is particularly relevant to consider when deciding surveillance for healthy relatives who are non-carriers of deleterious variants in the moderate risk genes, or non-carriers of intermediate risk variants in ‘high-risk cancer predisposition genes’ such as BRCA1 or BRCA2.

The relevance of these findings for clinical management of BRCA1*R1699Q carriers and their relatives was considered during the Clinical Working Group meeting at the April 2016 ENIGMA conference, held in Prague, which was attended by 38 members with expertise in laboratory research, statistics and clinical genetics. Recommendations for CHEK2 c.1100delC carriers17 18 and country-specific guidelines including Oncoline (The Netherlands: http://www.oncoline.nl, accessed 21 September 2016), National Institute for Health and Care Excellence (UK: https://www.nice.org.uk, accessed 21 September 2016) and National Comprehensive Cancer Network (USA: https://www.nccn.org, accessed 21 September 2016) were used as a framework to guide discussion. A consensus and majority-based discussion led to the following opinions and recommendations:

Female non-carriers of BRCA1*R1699Q from BRCA1*R1699Q families

Surveillance should depend on (family) history of cancer, for example, on the risk calculated using programs like BOADICEA.11

Female carriers of BRCA1*R1699Q

A cumulative risk of BC (20% (95% CI 13% to 32%)) does not by itself justify preventive mastectomy or breast MRI. Breast surveillance for female carriers based on annual mammogram from age 40 up to 50 years and inclusion in population screening afterwards is advised.

Combining with family history, the BC risk might be estimated to be higher than the risk conferred by the variant alone. If this is the case, the surveillance advice for BRCA1*R1699Q carriers can be ‘overruled’ by the higher family history risk and additional genetic testing can be considered.

The specific genes included will vary across countries dependent on testing practices, which incorporate availability and extent of panel-based testing, eligibility for health insurance or state-based testing, clinical guidelines for ascertainment including number and types of cancer reported in families, etc (ENIGMA, unpublished findings). Genetic testing for variants in other genes using a panel approach for a range of BC/OC susceptibility genes may offer some additional genotype-based information about risk in those cases; however, penetrance estimates for the majority of other genes beyond BRCA1, BRCA2 and PALB2 are imprecise.16 Furthermore, it is still unclear how genetic risks are best combined to produce more accurate, individualised, risk estimates.

The BRCA1*R1699Q variant carriers have lower OC risk (6% (95% CI 3% to 25%)), compared with that for BRCA1 carriers (39% (95% CI 22% to 51%)) and BRCA2 carriers (11% (95% CI 4.1% to 18%)).19 Bilateral salpingo-oophorectomy (BSO) is the standard preventive treatment in the Netherlands for high-risk pathogenic variant carriers, performed at age 35–40 for BRCA1 and 40–45 for BRCA2 (http://www.oncoline.nl). Routine surveillance for OC is not effective and is no longer offered to carriers.20 The magnitude of OC risk for R1699Q carriers suggests that BSO, if performed, may be postponed until age 50. We advise BSO surgery should be offered at age 50, based on the age-related cumulative risks for OC obtained from the study. The cumulative lifetime risk of OC for someone in the general population is approximately 1.5%, but the vast majority of risk occurs after 50 years of age. From our study the cumulative OC risk for BRCA1*R1699Q carriers by age 50 is lower than the cumulative population risk for OC and rises significantly after age 55. Although BSO surgery could be offered at any age after the genetic risk is identified, we base our guidance on a pragmatic balance between cancer prevention and minimum adverse effects from early oestrogen deprivation, achieved if the surgery is timed around the current average age for the menopause in the Western society (52 years).

However, as for BC risk management, and considering the wide CI for the estimated risk of OC, information about cancer history in the family should be taken into account for decision making.

Conclusion

Our analysis of a large cohort of 129 families, using several analytical approaches, confirms that the BRCA1*R1699Q variant is associated with intermediate cancer risks (compared with the average BRCA1 truncating variant). It also provides evidence that cancer risk in carriers is likely to be influenced by other genetic factors. Based on our findings, we propose recommendations for the clinical management of BRCA1*R1699Q carriers and non-carriers. We recommend that follow-up and screening in these families are performed in a research setting in order to enable future assessment of the utility of the proposed surveillance.

Supplementary Material

Supplementary Figure 2

Supplementary Material

Supplementary Figure 3

Acknowledgments

We thank the many families who participated in this study. This work is supported by the efforts of laboratory and clinical staff from many centres around the world. In particular, we would like to acknowledge all members of the ENIGMA consortium who contributed to discussions about clinical management of variant carriers, members of the GC-HBOC (German Consortium of Hereditary Breast and Ovarian Cancer), members of the kConFab (Kathleen Cuningham Foundation Consortium for Research into Familial Breast Cancer),and the members of these consortia: LOB Dutch Belgian Consortium: R Blok, D Bodmer, P Bouwman, K Claes, S Debrake, A Gheldof, J J P Gille, E Gomez Garcia, F Hogervorst, A H van der Hout, L van de kolk, M J  Koudijs, R B  van der Luijt, A Mensenkamp, G Michils, S Moghadasi, P Nederlof, A van den Ouweland, E Roisenberg, K Segers, N van der Stoep, K Storm, S Seneca, E Teugels, C M Tops, M Vogel, M P G Vreeswijk, J T Wijnen, S Willocx; SWE-BRCA, The Swedish BRCA1 and BRCA2 Study Collaborators: Gothenburg, Sahlgrenska University Hospital: Zakaria Einbeigi; Linköping University Hospital: Marie Stenmark-Askmalm; Lund University Hospital: Hans Ehrencrona, Therese Törngren, Anders Kvist, Åke Borg; Stockholm, Karolinska University Hospital: Brita Arver, Annika Lindblom, Emma Tham; Umeå University Hospital: Beatrice Melin; Uppsala University Hospital: Ylva Paulsson-Karlsson; French COVAR group collaborators: M Mathieu-Dramard:  CHU Amiens, Amiens; O Ingster: Centre Paul Papin, Angers; P Gesta: Centre Hospitalier d’Angoulême, Angoulème; H Dreyfus: Institut Sainte-Catherine, Avignon; M A Collonge-Rame, J L Bresson, C Populaire: CHU Besançon, Besançon; M Longy; E Barouk-Simonet, V Bubien; N Sévenet, F Bonnet, N Jones: Institut Bergonié, Bordeaux; I Mortemousque: CH Jacques Cœur, Bourges; S Audebert-Bellanger: CHU de Brest, Brest; Pascaline B; A Hardouin, D Vaur, S Krieger, L Castéra: Centre François Baclesse, Caen; S Ferrer: Centre Hospitalier Hôtel Dieu, Chambery; N Uhrhammer, Y J Bignon: Centre Jean Perrin, Clermont-Ferrand; L Faivre-Olivier, S El Chehadeh; S Lizard: CHU de Dijon, Dijon; O Béra: CHU de Fort de France, Martinique; D Leroux, H Dreyfus, A Béchet: CHU de Grenoble, Grenoble; H Ranjatoelina: CHU Sud Réunion, La Réunion; V Layet: Hôpital Flaubert, Le Havre; C Adenis; J-P Peyrat, F Revillion: Centre Oscar Lambret, Lille; S Lejeune, S Manouvrier-Hanu: CHRU Lille, Lille; L Vénat-Bouvet: CHU Dupuytren, Limoges; C Lasset, V Bonadona; S Mazoyer; O Sininilkova, M Léone, N Boutry-Kryza: Centre Léon Bérard, Lyon; S Giraud: Hospices Civils de Lyon, Lyon; H Sobol, T Noguchi, V Bourdon, A Remenieras, F Eisinger: Institut Paoli-Calmettes, Marseille; H Zattara: CHU La Timone, Marseille; I Coupier; J-M Rey, P-O Harmand: CHU Arnaud de Villeneuve, Montpelier; E Luporsi: Centre Alexis Vautrin, Nancy; M Bronner, J Sokolowska-Gillois, P Jonveaux, C Philippe: CHU Nancy, Nancy; C Delnatte, V Guibert, S Bézieau, E Cauchin: Centre René Gauducheau, Nantes; A Lortholary: Centre Catherine de Sienne, Nantes; V Mari; M Frenay: Centre AntoineLacassagne, Nice; J Chiesa: CHRU Caremeau, Nîmes; E Rouleau, C Lefol, R Lidereau; V Caux-Moncoutier, L Golmard, C Houdayer, B Buecher, M Gauthier-Villars, M Belotti, A Depauw, S Demontety, D Stoppa-Lyonnet: Institut Curie, Paris; F Coulet, F Soubrier, C Colas, A Fajac, M Warcoin: Groupe Hospitalier Pitié-Salpêtrière, Paris; P-L Puig-HEGP, Paris; M Lackmy-Port-Lis: CHU de Pointe à Pitre, Guadeloupe; A Marie Savoye; C Delvincourt, O Beaudoux: Institut Jean Godinot, Reims; Dominique Gaillard; C Poirsier; M Mozelle-Nivoix: CHU de Reims, Reims; C Dugast; C Abadie: Centre Eugène Marquis, Rennes; T Frebourg, J Tinat, A Rossi, I Thevenet: CHU de Rouen, Rouen; F Prieur; M Lebrun: CHU Saint-Etienne, Saint-Etienne; C Noguès, E Fourme, C Sénéchal, A-M Birot, T Kogut-Kubiak: Institut Curie, Saint-Cloud; J-P Fricker, H Nehme-Schuster; D Muller, J Abecassis: Centre Paul Strauss, Strasbourg; C Maugard: Hopital Universitaire de Strasbourg, Strasbourg; L Gladieff; V Feillel; C Toulas:  Institut Claudius Regaud, Toulouse; M Mozelle: CHG Troyes, Troyes; O Caron; M Guillaud-Bataille, B Bressac-De Paillerets: Institute Gustave Roussy, Villejuif. We wish to thank Heather Thorne, Eveline Niedermayr, all the kConFab research nurses and staff, the heads and staff of the Family Cancer Clinics, and the Clinical Follow Up Study (which has received funding from the NHMRC, the National Breast Cancer Foundation, Cancer Australia, and the National Institute of Health (USA)) for their contributions to this resource, and the many families who contribute to kConFab. kConFab is supported by a grant from the National Breast Cancer Foundation, and previously by the National Health and Medical Research Council (NHMRC), the Queensland Cancer Fund, the Cancer Councils of New South Wales, Victoria, Tasmania and South Australia, and the Cancer Foundation of Western Australia.

References

Footnotes

  • SM and HDM contributed equally.

  • Funding The Netherlands Organization for Scientific Research (NWO), research program Mosaic (Grant 017.008.022); Van de Kampfonds from Leiden University Medical Centre (Grant 30.925). ABS is supported by an Australian NHMRC Senior Research Fellowship. SMC was supported by the French National Cancer Institute (INCa).

  • Competing interests EGG has received an honorarium in the past 3 years from AstraZeneca for giving a course and a lecture. HE (or rather, his department with him as primary contact) has received funding from Novartis Oncology (unrestricted grant) and AstraZeneca (invited speaker). KT has received an honorarium for chairing a mainstreaming genetic testing subcommittee and day seminar for AstraZeneca. AET declares to have received an honorarium from American Cancer Society for grant review, NIH NCI PDQ as editorial board, Italian Ministry of Health for grant review. DEE receives an honorarium from AstraZeneca via a contract with the university to provide consultancy advice from time to time (one or two advisory boards each year on average at the moment). DEG has received royalties from patents on the BRCA1 and BRCA2 genes from the University of Utah that are licensed to Myriad Genetics. All the other authors declare to have no conflicts of interest.

  • Ethics approval All authors made a significant contribution to data collection, data interpretation, writing and critical assessment of this study.

    Specifically: SM, EGG, DEG and ABS were responsible for study concept and design. HDM and DEG performed the statistical analyses. SM and EGG wrote the manuscript.

    SM, EGG, LAMJ and MPGV were responsible for data collection. K

    BMC, HE, YPK, BW, CE, AG, NA, TvOH, MT, UBJ, TAK, BE, AMG, ISP, SMC, FC, EJH, AMWvdO, MJC, ET, MAA, RBvdL, ARM, JCO, MJB, NJ, KC, KT, VV, AET, DEE, PD and CJvA provided the family data analysed in this study.

    EGG coordinated the study.

    All authors approved the final manuscript submitted.

  • Ethics approval Not applicable.

  • Provenance and peer review Not commissioned; externally peer reviewed.