Characterization of BRCA Deficiency in Ovarian Cancer

Simple Summary Ovarian cancer (OC) is a highly lethal malignancy. Major improvements in treatment are expected from the identification of molecular features that may predict outcome or be used as therapeutic targets. Among genetic defects relevant for OC are those of BRCA1 and BRCA2 genes. Indeed, at least 20% of OC patients carry inherited or acquired BRCA1/2 pathogenic variants, the identification of which is important for treatment and prevention. A comprehensive study of 30 OC patients revealed that 7 (23%) had BRCA alterations (6 inherited and 1 acquired) detectable by usual clinical testing, while another 5 patients (17%) showed epigenetic silencing of BRCA1 in the tumor, which would have escaped standard sequencing analysis, and one had an inherited variant in another gene: RAD51C, involved in the same DNA repair mechanism as BRCA1 and BRCA2. Patients with BRCA deficit showed greater genomic instability, but better survival, than those with no evidence of BRCA deficit. BRCA testing is recommended in all Ovarian Cancer (OC) patients, but the optimal approach is debated. The landscape of BRCA alterations was explored in 30 consecutive OC patients: 6 (20.0%) carried germline pathogenic variants, 1 (3.3%) a somatic mutation of BRCA2, 2 (6.7%) unclassified germline variants in BRCA1, and 5 (16.7%) hypermethylation of the BRCA1 promoter. Overall, 12 patients (40.0%) showed BRCA deficit (BD), due to inactivation of both alleles of either BRCA1 or BRCA2, while 18 (60.0%) had undetected/unclear BRCA deficit (BU). Regarding sequence changes, analysis performed on Formalin-Fixed-Paraffin-Embedded tissue through a validated diagnostic protocol showed 100% accuracy, compared with 96.3% for Snap-Frozen tissue and 77.8% for the pre-diagnostic Formalin-Fixed-Paraffin-Embedded protocol. BD tumors, compared to BU, showed a significantly higher rate of small genomic rearrangements. After a median follow-up of 60.3 months, the mean PFS was 54.9 +/- 27.2 months in BD patients and 34.6 +/- 26.7 months in BU patients (p = 0.055). The analysis of other cancer genes in BU patients identified a carrier of a pathogenic germline variant in RAD51C. Thus, BRCA sequencing alone may miss tumors potentially responsive to specific treatments (due to BRCA1 promoter methylation or mutations in other genes) while unvalidated FFPE approaches may yield false-positive results.