作者
C. Loveday,A. Garrett,P. Law,S. Hanks,E. Poyastro-Pearson,J.W. Adlard,J. Barwell,J. Berg,A.F. Brady,C. Brewer,C. Chapman,J. Cook,R. Davidson,A. Donaldson,F. Douglas,L. Greenhalgh,A. Henderson,L. Izatt,A. Kumar,F. Lalloo,Z. Miedzybrodzka,P.J. Morrison,J. Paterson,M. Porteous,M.T. Rogers,L. Walker,D. Eccles,D.G. Evans,K. Snape,H. Hanson,R.S. Houlston,C. Turnbull
摘要
Abstract
Background
Breast cancer has a significant heritable basis, of which approximately 60% remains unexplained. Testing for BRCA1/BRCA2 offers useful discrimination of breast cancer risk within families, and identification of additional breast cancer susceptibility genes could offer clinical utility. Patients and methods
We included 2,135 invasive breast cancer cases recruited via the BOCS study, a retrospective UK study of familial breast cancer. Eligibility criteria: female, BRCA-negative, white European ethnicity, and one of: i) breast cancer family history, ii) bilateral disease, iii) young age of onset (<30 years), iv) concomitant ovarian cancer. We undertook exome sequencing of cases and performed gene-level burden testing of rare damaging variants against those from 51,377 ethnicity-matched population controls from gnomAD. Results
159/2135 (7.4%) cases had a qualifying variant in an established breast cancer susceptibility gene, with minimal evidence of signal in other cancer susceptibility genes. Known breast cancer susceptibility genes PALB2, CHEK2 and ATM were the only genes to retain statistical significance after correcting for multiple testing. Due to the enrichment of hereditary cases in the series, we had good power (>80%) to detect a gene of BRCA1-like risk (odds ratio=10.6) down to a population minor allele frequency of 4.6 x 10-5 (1 in 10,799, less than one tenth that of BRCA1)and of PALB2-like risk (odds ratio=5.0) down to a population minor allele frequency of 2.8 x 10-4 (1 in 1,779, less than half that of PALB2). Power was lower for identification of novel moderate penetrance genes (odds ratio=2-3) like CHEK2 and ATM. Conclusions
This is the largest case-control whole-exome analysis of enriched breast cancer published to date. Whilst additional breast cancer susceptibility genes likely exist, those of high penetrance are likely to be of very low mutational frequency. Contention exists regarding the clinical utility of such genes.