Retrospective analysis of oncologic outcomes by germline and somatic genetic status in patients with ovarian cancer

Objectives: To describe results of tumor testing among women with ovarian cancer and to identify associations with oncologic outcomes according to genetic variant status. Methods: Institutional review board (IRB)-approved, retrospective review of tumor genomic results obtained within a single Gynecologic Oncology division. Genomic results, oncologic outcomes and other clinicopathologic data points were abstracted from medical records. Platinum sensitivity and response to PARP inhibitors were compared between patients with: germline BRCA (gBRCA) pathogenic variants (PV), somatic BRCA (sBRCA) PVs and no gBRCA PVs, and loss of heterozygosity (LOH)/genome instability and no gBRCA or sBRCA PVs. These groups are referred to as “gBRCA,” “sBRCA,” and “LOH,” respectively. Results: A total of 261 patients with epithelial ovarian/fallopian tube/primary peritoneal cancer had tumor testing between 2017 and 2020. There were 26 (10%) gBRCA PV carriers [15 BRCA1, 11 BRCA2] identified, 48/261 (18%) tumors with only sBRCA PVs and 13/261 (5%) tumors with LOH and no gBRCA or sBRCA PVs. There were no significant differences in histologies between groups. Of the patients with gBRCA and sBRCA respectively, 85% and 69% were primary platinum sensitive and 8% and 17% were primary platinum resistant, respectively (p=0.40). The average number of platinum-based regimens used before platinum resistance developed was 2.67 for gBRCA (N=6), 2.18 for sBRCA (N=11), 3.2 for LOH (N=5). These groups were not significantly different (p=0.17).23/26 (88%) gBRCA, 32/48 (67%) sBRCA, and 10/13 (77%) LOH patients received PARP inhibitors (PARPi) for treatment. 3 gBRCA and 4 sBRCA patients received PARPi more than once. Median PFS for PARPi use in the frontline was 6 months compared to 5 months in the recurrent setting. There were no significant differences between PFS on PARPi in the upfront or recurrent setting. PFS for PARPi used in the recurrent setting for gBRCA was 8 months (95% CI: 4.6-11.5) and 6.5 months (95% CI: 3.4-9.6) for sBRCA compared to 4 months for LOH (95% CI: 0.74-7.26; p=0.33). Download : Download high-res image (97KB) Download : Download full-size image Conclusions: Twenty-eight percent of ovarian cancer patients had a gBRCA or sBRCA PVs. Platinum sensitivity is similar between patients with gBRCA, sBRCA and LOH groups, however, patients with gBRCA and LOH received slightly more lines of platinum than those with sBRCA. There was a trend towards lower PFS for PARPi used in the recurrent setting in the LOH group compared to gBRCA or sBRCA. These outcomes should be evaluated in a larger population to determine the reliability of LOH assays for predicting functional homologous recombination deficiency and susceptibility to PARPi in the primary or recurrent setting.