BRIP1 loss-of-function mutations confer high risk for familial ovarian cancer, but not familial breast cancer

Nana Weber-Lassalle, Jan Hauke, Juliane Ramser, Lisa Richters, Eva Groß, Britta Blümcke, Andrea Gehrig, Anne-Karin Kahlert, Clemens R. Müller, Karl Hackmann, Ellen Honisch, Konstantin Weber-Lassalle, Dieter Niederacher, Julika Borde, Holger Thiele, Corinna Ernst, Janine Altmüller, Guido Neidhardt, Peter Nürnberg, Kristina Klaschik, Christopher Schroeder, Konrad Platzer, Alexander E. Volk, Shan Wang-Gohrke, Walter Just, Bernd Auber, Christian Kubisch, Gunnar Schmidt, Judit Horvath, Barbara Wappenschmidt, Christoph Engel, Norbert Arnold, Bernd Dworniczak, Kerstin Rhiem, Alfons Meindl, Rita K. Schmutzler, Eric Hahnen

Germline mutations in the BRIP1 gene have been described as conferring a moderate risk for ovarian cancer (OC), while the role of BRIP1 in breast cancer (BC) pathogenesis remains controversial. To assess the role of deleterious BRIP1 germline mutations in BC/OC predisposition, 6341 well-characterized index patients with BC, 706 index patients with OC, and 2189 geographically matched female controls were screened for loss-of-function (LoF) mutations and potentially damaging missense variants. All index patients met the inclusion criteria of the German Consortium for Hereditary Breast and Ovarian Cancer for germline testing and tested negative for pathogenic BRCA1/2 variants. BRIP1 LoF mutations confer a high OC risk in familial index patients (odds ratio (OR) = 20.97, 95% confidence interval (CI) = 12.02-36.57, P < 0.0001) and in the subgroup of index patients with late-onset OC (OR = 29.91, 95% CI = 14.99-59.66, P < 0.0001). No significant association of BRIP1 LoF mutations with familial BC was observed (OR = 1.81 95% CI = 1.00-3.30, P = 0.0623). In the subgroup of familial BC index patients without a family history of OC there was also no apparent association (OR = 1.42, 95% CI = 0.70-2.90, P = 0.3030). In 1027 familial BC index patients with a family history of OC, the BRIP1 mutation prevalence was significantly higher than that observed in controls (OR = 3.59, 95% CI = 1.43-9.01; P = 0.0168). Based on the negative association between BRIP1 LoF mutations and familial BC in the absence of an OC family history, we conclude that the elevated mutation prevalence in the latter cohort was driven by the occurrence of OC in these families. Compared with controls, predicted damaging rare missense variants were significantly more prevalent in OC (P = 0.0014) but not in BC (P = 0.0693) patients. To avoid ambiguous results, studies aimed at assessing the impact of candidate predisposition gene mutations on BC risk might differentiate between BC index patients with an OC family history and those without. In familial cases, we suggest that BRIP1 is a high-risk gene for late-onset OC but not a BC predisposition gene, though minor effects cannot be excluded.