Somatic versus germline BRCA mutations screening in ovarian cancer.

Abstract

e13100 Background: ovarian cancer (OC) with germinal or somatic BRCA mutations responds better to platinum and to PARP-inhibitors. There is great enthusiasm about BRCA somatic screening. Our aim was to analyse the correlation between BRCA somatic and germline mutational profile. Methods: a cohort of 23 pts was obtained by cross-linking OC pts from the South Portuguese Cancer Registry, between 2009-2014 and BRCA mutation carriers identified in our Clinic. Medical records were reviewed: demographic and clinico-pathologic data obtained. Germinal screening: pts were pre-screened for the BRCA2 Portuguese founder mutation (PFM), analysed for BRCA point mutations (different screening methodologies were used in diagnostic timeline: initially CSGE, then CSCE and finally NGS) and for large rearrangements by MLPA. Somatic screening: DNA was extracted from 5 sections of 10µm of FFPE tissue and analysed for BRCA1/2 genes by NGS. Results: clinico-pathological features of the 23 pts revealed mainly high-grade serous histology (96%). Mean age at diagnosis was 54 years old (33-76); BRCA2 were older than BRCA1 carriers (62 vs 51). Most pts presented at advanced stage (70% stages III-IV; 30% stage I-II). Seventeen were BRCA1 carriers and 6 were BRCA2 (5 of those PFM, a large insertion of an Alu element). Somatic correlation: 8 pts (5 BRCA1, 3PFM) were already analysed and 100% correlation was observed for all point mutations. One additional BRCA2 somatic mutation was detected (with a variant allele frequency of 53% whereas 2 others were < 7%); interestingly exclusive somatic mutations were only observed in PFM carriers (known not to be NGS detectable). The remaining 15 pts are under analysis. Conclusions: it was expected that the PFM and other large rearrangements would not be detected with NGS. A specific somatic screening for the PFM may be possible but other rearrangements are found by MLPA in our population (10% of all BRCA mutations). Preliminary data adds to the evidence that NGS OC somatic screening will identify all germinal point mutations and an indeterminate number of additional pts with exclusive somatic mutations. An ideal correlation needs integration of differenttechniques that may increase complexity, time and cost.