Associations of common variants at 1p11.2 and 14q24.1 (RAD51L1) with breast cancer risk and heterogeneity by tumor subtype: findings from the Breast Cancer Association Consortium

Research output: Contribution to journalJournal articleResearchpeer-review

Jonine D Figueroa, Montserrat Garcia-Closas, Manjeet Humphreys, Radka Platte, John L Hopper, Melissa C Southey, Carmel Apicella, Fleur Hammet, Marjanka K Schmidt, Annegien Broeks, Rob A E M Tollenaar, Laura J Van't Veer, Peter A Fasching, Matthias W Beckmann, Arif B Ekici, Reiner Strick, Julian Peto, Isabel dos Santos Silva, Olivia Fletcher, Nichola Johnson & 31 others Elinor Sawyer, Ian Tomlinson, Michael Kerin, Barbara Burwinkel, Federik Marme, Andreas Schneeweiss, Christof Sohn, Stig Bojesen, Henrik Flyger, Børge G Nordestgaard, Javier Benítez, Roger L Milne, Jose Ignacio Arias, M Pilar Zamora, Hermann Brenner, Heiko Müller, Volker Arndt, Nazneen Rahman, Clare Turnbull, Sheila Seal, Anthony Renwick, Hiltrud Brauch, Christina Justenhoven, Thomas Brüning, Jenny Chang-Claude, Rebecca Hein, Shan Wang-Gohrke, Thilo Dörk, Peter Schürmann, Michael Bremer, GENICA Network

A genome-wide association study (GWAS) identified single-nucleotide polymorphisms (SNPs) at 1p11.2 and 14q24.1 (RAD51L1) as breast cancer susceptibility loci. The initial GWAS suggested stronger effects for both loci for estrogen receptor (ER)-positive tumors. Using data from the Breast Cancer Association Consortium (BCAC), we sought to determine whether risks differ by ER, progesterone receptor (PR), human epidermal growth factor receptor 2 (HER2), grade, node status, tumor size, and ductal or lobular morphology. We genotyped rs11249433 at 1p.11.2, and two highly correlated SNPs rs999737 and rs10483813 (r(2)= 0.98) at 14q24.1 (RAD51L1), for up to 46 036 invasive breast cancer cases and 46 930 controls from 39 studies. Analyses by tumor characteristics focused on subjects reporting to be white women of European ancestry and were based on 25 458 cases, of which 87% had ER data. The SNP at 1p11.2 showed significantly stronger associations with ER-positive tumors [per-allele odds ratio (OR) for ER-positive tumors was 1.13, 95% CI = 1.10-1.16 and, for ER-negative tumors, OR was 1.03, 95% CI = 0.98-1.07, case-only P-heterogeneity = 7.6 × 10(-5)]. The association with ER-positive tumors was stronger for tumors of lower grade (case-only P= 6.7 × 10(-3)) and lobular histology (case-only P= 0.01). SNPs at 14q24.1 were associated with risk for most tumor subtypes evaluated, including triple-negative breast cancers, which has not been described previously. Our results underscore the need for large pooling efforts with tumor pathology data to help refine risk estimates for SNP associations with susceptibility to different subtypes of breast cancer.
Original languageEnglish
JournalHuman Molecular Genetics
Volume20
Issue number23
Pages (from-to)4693-706
Number of pages14
ISSN0964-6906
DOIs
Publication statusPublished - 2011

ID: 40184024