More than 25% of breast cancer families with wild-type results from commercial genetic testing of BRCA1 and BRCA2 are resolved by BROCA sequencing of all known breast cancer genes. T. Walsh1, S. Casadei1, M. K. Lee1, A. M. Thornton1, G. Bernier1, C. Spurrell1, J. Mandell1, T. Lajus2, E. Swisher1, M.-C. King1 1) Medical Genetics, University of Washington, Seattle, WA; 2) Coordenadora de Pesquisa Translacional Liga Contra o Cāncer, Brazil.

   A challenge to the present practice of genetic testing for inherited risk of breast cancer is how to explain the illness in breast cancer patients with severe family history, but negative (wild-type) results from commercial testing for BRCA1 and BRCA2. To address this problem, we applied BROCA, a targeted capture sequencing approach, to identify all single base pair substitutions, insertion-deletions, and copy number variants in all known breast cancer genes. Our subjects were all living affected persons in 800 families with at least three cases of breast and/or ovarian cancer. Probands of all families had received wild-type results of commercial genetic testing of BRCA1 and BRCA2. Of the 800 families, 206 (26%) were resolved by BROCA. The 206 resolved families harbored 166 different damaging germline mutations in 21 different genes. Of the resolved families, 39% (80/206) carried mutations in BRCA1 or BRCA2 that were not detected by commercial sequencing of the proband, either because the patient did not have comprehensive commercial large rearrangement testing in addition to commercial Sanger sequencing; or the proband was wild-type for BRCA1 and BRCA2, but other cases in the family carried a BRCA1 or BRCA2 mutation; or the family carried a BRCA1 or BRCA2 mutation not reported by commercial testing (e.g. mutation of a splice enhancer, a small damaging in-frame indel, or a missense with good experimental evidence of damaging effect.) Of the resolved families, 37% (77/206) carried mutations in other breast cancer genes whose role in inherited predisposition is well characterized: 39 in CHEK2, 28 in PALB2, and 10 in TP53. Finally, of the resolved families, 20% (41/206) carried mutations in breast cancer genes that have been published but whose role in inherited predisposition has been less fully characterized: 15 in ATM, 4 in BARD1, 5 in BRIP1, 1 in CDH1, 2 in ABRAXAS, 2 in NBN, 3 in RAD51C, 5 in RAD51D, 1 in STK11, 1 in XRCC2. Our results indicate that families severely affected by breast cancer but with wild-type results from commercial genetic testing are well served by complete genomic sequencing of all known breast cancer genes. As such comprehensive testing becomes more widespread, it will be important to determine more precisely the risks associated with damaging mutations in each of these genes so as to incorporate them most effectively into clinical care.

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