Effectiveness of PARP inhibitor maintenance therapy (mPARPi) in advanced ovarian cancer (OC) by BRCA1/2 and HRD signature in real-world practice.

5583 Background: mPARPi following surgery and platinum chemotherapy (PCT) is a standard of care treatment for newly diagnosed advanced OC. Clinical trials have demonstrated benefit of mPARPi for patients (pts) with and without BRCA1/2 mutations ( BRCA+/-). The degree of benefit of mPARPi for pts without homologous recombination deficiency (HRD) biomarkers detected remains in question. This study aimed to compare the effectiveness of mPARPi in real world practice defined by biomarker status ( BRCA+/- mutation and a novel HRD signature [HRDsig]). Methods: This study included pts with advanced OC who received 1 st -line PCT with real-world progression-free survival (rwPFS) of at least 10 months after treatment initiation and received either mPARPi (without bevacizumab) or no maintenance therapy (nm). Pt data was obtained by the US-based de-identified Flatiron Health-Foundation Medicine real-world OC clinico-genomic database, originating from ~280 US cancer clinics (~800 sites of care) between 01/2015 and 09/2022. rwPFS and real-world overall survival (rwOS) were compared between pts +/- biomarkers by Cox models, adjusted for propensity scores accounting for disease stage at diagnosis, ECOG, age, and BRCA status (for the HRDsig analysis). HRDsig+ (Foundation Medicine) status was determined using a pre-specified cutoff. Results: Of 604 included pts, 128 pts received mPARPi (25.8% BRCA+ and 48.4% HRDsig+) and 476 received nm (14.3% BRCA+ and 35.5% HRDsig+). BRCA+ pts receiving mPARPi vs. nm had more favorable rwPFS (HR 0.44, 95% CI 0.24-0.79, p = 0.006), as did BRCA- pts (HR 0.7, 95% CI 0.51-0.95, p = 0.021). More favorable rwOS was not observed for either group (p = 0.7112, p = 0.2066 respectively). HRDsig+ pts receiving mPARPi vs. nm had more favorable rwPFS (HR 0.31, 95% CI 0.20-0.48, p < 0.001) and rwOS (HR 0.35, 95% CI 0.12-1.05, p = 0.061), while no differences were observed for those HRDsig- (rwPFS HR 0.99 0.71-1.40, p = 0.976 / rwOS HR 0.86, 95% CI 0.5-1.47, p = 0.575). A treatment interaction was observed for HRDsig+ vs. HRDsig- (rwPFS p < 0.001 / rwOS p = 0.042) but not for BRCA+ vs. BRCA- (rwPFS p = 0.213 / rwOS p = 0.596). Among BRCA- pts, those HRDsig+ receiving mPARPi vs nm had favorable rwPFS (HR 0.27, 95% CI 0.15-0.49, p < 0.001) and rwOS (HR 0.40, 95% CI 0.08-1.98, p = 0.216), while no difference was observed for those HRDsig- (rwPFS HR 1.02, 95% CI 0.72-1.44, p = 0.923 / rwOS HR 0.86, 95% CI 0.49-1.49, p = 0.590), with a treatment interaction observed for rwPFS (p < 0.001), but not for rwOS (p = 0.120). Conclusions: In this analysis of real-world practice, we observed improved outcomes among biomarker positive pts treated with mPARPi. HRDsig+ pts had improved outcomes, even among those who were BRCA-, while HRDsig- pts showed no enrichment for benefit with mPARPi, suggesting that a novel HRDsig might be able to predict benefit from mPARPi regardless of BRCA status.