Breast and ovarian cancer penetrance estimates derived from germline multiple-gene sequencing results in women

Purpose Multiple-gene, next-generation sequencing panels are increasingly used to assess hereditary cancer risks of patients with diverse personal and family cancer histories. The magnitude of breast and ovarian cancer risk associated with many clinically tested genes, and independent of family cancer history, remains to be quantified. Methods We queried a commercial laboratory database of 95,561 women tested clinically for hereditary cancer risk with a 25-gene (APC, ATM, BARD1, BMPR1A, BRCA1, BRCA2, BRIP1, CDH1, CDK4, CHEK2, MLH2, MSH2, MSH6, MUTYH, NBN, P14ARF, P16, PALB2, PMS2, PTEN, RAD51C, RAD51D, SMAD4, STK11, and TP53) next-generation sequencing panel. Multivariable logistic regression models accounting for family history were used to examine the association between pathogenic mutations and breast or ovarian cancer. As a confirmatory approach, a matched case-control analysis was conducted, defining cases as patients with breast or ovarian cancer and controls as women without cancer. Results One or more pathogenic mutations were detected in 6,775 (7%) of 95,561 women. Eight genes (ATM, BARD1, BRCA1, BRCA2, CHEK2, PALB2, PTEN, and TP53) were associated with breast cancer, with odds ratios (ORs) ranging from two-fold (ATM: OR, 1.74; 95% CI, 1.46 to 2.07) to six-fold (BRCA1: OR, 5.91; 95% CI, 5.25 to 6.67). Eleven genes (ATM, BRCA1, BRCA2, BRIP1, MLH1, MSH2, MSH6, NBN, STK11, RAD51C, and RAD51D) were associated with ovarian cancer, withORranging from two-fold (ATM:OR, 1.69;95%CI, 1.19 to 2.40) to 40-fold (STK11: OR, 41.9; 95% CI, 5.55 to 315). Multivariable models and matched case-control analyses yielded similar results. ConclusionAmongnearly 100,000 clinically tested women,7%carried a pathogenic mutation in one or more cancer-associated genes. Associated breast and ovarian cancer risks ranged from two- to 40-fold after controlling for family history. These results may inform cancer risk counseling.