Implementing PALB2 gene testing in breast and ovarian cancer patients in UK. N. Rahman1,2, E. Ruark1, S. Seal1, A. Renwick1, E. Ramsay1, S. Powell1, M. Warren-Perry1, H. Hanson1, C. Lord3, C. Turnbull1 1) Division of Genetics & Epidemiology, The Institute of Cancer Research, London, United Kingdom; 2) Cancer Genetics Unit, Royal Marsden NHS Foundation Trust, London, United Kingdom; 3) Breast Cancer Research Division, The Institute of Cancer Research, London, United Kingdom.

   In 2007 we reported that PALB2 mutations confer moderately increased risks of breast cancer. The cancer association was confirmed by many other groups, but the risks conferred have been variable, with some studies suggesting much higher risks, potentially warranting interventions similar to those offered to BRCA1/2 mutation carriers. Together with the advent of affordable gene panel testing this has led to PALB2 testing entering the clinical arena. To explore clinical utility of PALB2 testing in UK we performed PALB2 mutation analysis in 5982 samples detecting mutations in 58/3906 cancer cases vs 1/2076 population controls (P= 4x10-10).Via segregation analysis of the full pedigree information from 3200 families ascertained on the basis of personal +/- family history of breast and ovarian cancer through Genetic clinics, and 2076 controls we found a 2.1 fold increased risk of breast cancer (95% CI=1.7-2.6, P=3.35x10-12), consistent with our previous estimate, and a 2.6 fold increased risk of ovarian cancer (95% CI=1.5-4.7, P=0.001). We also confirmed that PALB2 defects generate sensitization to PARP inhibitors and show that this is as extensive as that caused by gene silencing of BRCA1 or BRCA2. These data confirm PALB2 mutations predispose to breast cancer and provide the first robust evidence of an association with ovarian cancer. Testing for PALB2 mutations in cancer patients that are having BRCA mutation analysis anyway is justifiable (if it incurs no additional cost), as it may aid treatment decisions, for example eligibility to PARP inhibitor trials. However, caution is required in using PALB2 mutations in a predictive capacity in unaffected relatives. Overall, our large analysis suggest the cancer risks in UK families are moderate and not of the level to warrant major interventions such as mastectomy, unless additional factors present (e.g. a strong family history). Larger prospective studies will be invaluable in clarifying cancer risks further.