Germline Mutations in BRCA1 and BRCA2 in Breast-Ovarian Families From a Breast Cancer Risk Evaluation Clinic

Abstract

PURPOSE: Data from the Breast Cancer Linkage Consortium suggest that the proportion of familial breast and ovarian cancers linked to BRCA1 or BRCA2 may be as high as 98% depending on the characteristics of the families, suggesting that mutations in BRCA1 or BRCA2 may entirely account for hereditary breast and ovarian cancer families. We sought to determine what proportion of families with both breast and ovarian cancers seen in a breast cancer risk evaluation clinic are accounted for by coding region germline mutations in BRCA1 and BRCA2 as compared to a linkage study group. We also evaluated what clinical parameters were predictive of mutation status. PATIENTS AND METHODS: Affected women from 100 families with at least one case of breast cancer and at least one case of ovarian cancer in the same lineage were screened for germline mutations in the entire coding regions of BRCA1 and BRCA2 by conformation-sensitive gel electrophoresis, apolymerase chain reaction–based heteroduplex analysis, or direct sequencing. RESULTS: Unequivocal deleterious mutations were found in 55% (55 of 100) of the families studied. Mutations in BRCA1 and BRCA2 accounted for 80% and 20% of the mutations overall, respectively. Using multivariate analysis, the strongest predictors of detecting a mutation in BRCA1 or BRCA2 in this study group were the presence of a single family member with both breast and ovarian cancer (P < .0009; odds ratio [OR], 5.68; 95% confidence interval [CI], 2.04 to 15.76) and a young average age at breast cancer diagnosis in the family (P < .0016; OR, 1.69; 95% CI, 1.23 to 2.38). CONCLUSION: These results suggest that at least half of breast/ovarian families evaluated in a high-risk cancer evaluation clinic may have germline mutations in BRCA1 or BRCA2. Whether the remaining families have mutations in noncoding regions in BRCA1, mutations in other, as-yet-unidentified, low-penetrance susceptibility genes, or represent chance clustering remains to be determined.