Germline loss-of-function mutations in 15 different DNA repair genes are present in 22% of 1412 patients with ovarian, peritoneal or fallopian tube cancers not selected for age at diagnosis or family history. M. I. Harrell1, B. Norquist1, T. Walsh2, M. K. Lee2, S. Casadei2, K. P. Pennington1, K. J. Agnew1, A. Thornton1, M. C. King2, M. J. Birrer3, E. M. Swisher1 1) Division of Gynecologic Oncology, University of Washington, Seattle, WA; 2) Division of Medical Genetics, University of Washington, Seattle, WA; 3) Department of Medicine, Dana Farber Cancer Center, Boston, MA.

   Germline mutations in BRCA1 and BRCA2, and to a lesser extent in the mismatch repair genes, have heretofore been thought to explain nearly all of hereditary ovarian cancer. Recently, more ovarian cancer susceptibility genes have been identified. In order to assess the frequency and spectrum of inherited mutations in many genes, we evaluated women with ovarian, peritoneal, and fallopian tube cancers unselected for age at diagnosis or family history. Participants consisted of two groups: 577 cases prospectively enrolled at diagnosis at UW; and 835 cases enrolled in the GOG218 (n=775) or GOG213 (n=60), randomized phase III treatment trials for advanced stage ovarian cancers. Germline DNA from all 1412 patients was sequenced using BROCA, a targeted capture and massively parallel genomic sequencing approach that detects all mutation classes. Cohort 1 was sequenced with the clinical version of BROCA, while cohort 2 was sequenced with BROCA-HR, an extended version of BROCA modified to include additional genes in the Fanconi anemia-BRCA (FA-BRCA) and related pathways. Only truncating and functionally validated missense mutations were counted. Mutation rates were similar for the two cohorts. Of all patients, 22% (304/1412) carried a damaging mutation in a gene known to be implicated in inherited ovarian or breast cancer. Of these 304 patients, 74% (226/304) carried mutations in BRCA1 (142) or BRCA2 (84) and only 1.6% (5/304) carried mutations in a mismatch repair gene (2 in MSH2, 3 in MSH6, 1 in MLH1). Germline damaging mutations in genes not generally sequenced for inherited ovarian cancer were present in 24% (73/304) of patients: 21 in BRIP1, 13 in CHEK2, 10 in RAD51C, 9 in RAD51D, 7 in PALB2, 4 in TP53, 3 in NBN, 3 in BARD1, 2 in MRE11A, and 1 in XRCC2. In addition, germline damaging mutations were detected in 15 other DNA repair candidate genes, which will be evaluated for loss of heterozygosity in tumors and, for familial cases, for co-segregation with ovarian and related cancers. In conclusion, the distribution of germline mutations in patients with ovarian carcinomas is wider than previously appreciated, and FA-BRCA genes other than BRCA1/2 explain a large number of cases. Effective identification of hereditary risk will require comprehensive testing of many genes for all women with ovarian, peritoneal or tubal carcinoma regardless of age or family history. Cancer gene panels could do so now in a cost effective manner.

You may contact the first author (during and after the meeting) at