878P - Quantification of genetic variants as marker of Brca-like phenotype in ovarian cancer

Abstract

PARP inhibitors have repeatedly been demonstrated to be active in BRCA mutant ovarian cancers. However, tumors with a BRCA-like phenotype could also benefit from such treatments. Thus, defining molecularly this subtype of ovarian neoplasias has become a priority. We aimed to study the potential role of the quantification of genetic variants in a limited set of genes in this regard. We designed a retrospective multicenter study in four collaborating institutions in Spain. Adult patients diagnosed with epithelial ovarian cancer, stage IC or superior, from 2008 to date were eligible. Clinical data (regarding demographics and treatment outcome) were extracted from medical records by an external data monitor. Whole exome sequencing was performed in extreme cases (best and worst responders) while a panel of five genes highly involved in homologous recombination (BRCA 1 and 2, ATM, CHEK2, RAD51C) was studied through Next Generation Sequencing in middle cases, not previously tested for BRCA mutations. We present the quantitative results of these last patients. In total 220 patients have been included so far. Discovery cohort accomplish for the first 90 cases. Median age was 61 (range 37-87) and tumor stage was I in 10 cases (9%), II in 5 (6%), III in 59 (66%), IV in 13 (14%) and not available in 3 (5%). Tumor histologies were papillary serous 73 (81%), mucinous 1 (1%), endometroid 5 (6%), clear cell 7 (13%), adenocarcinoma 2 (2%), and non-epithelial in 2 (2%). Up to 37 cases have been already analyzed. Median number of genetic variants was 26 (range 12-41). Kaplan Meier test showed a platinum-free interval of 14.3 months (95% Confidence Interval [CI] 10.2-18.5) vs 48.3 (95% CI 0-101.2) for patients with a number of variants below or above the median, respectively. Though exploratory, our results point towards an association between the number of genetic variants in selected genes and a better outcome in advanced ovarian cancer. Mature data of the whole cohort will be presented at the meeting.