Efficacy of third-line and later (3L+) therapies post poly (ADP-ribose) polymerase inhibitor (PARPi) exposure in recurrent platinum-sensitive ovarian cancer (PSOC): A pooled clinical trial database analysis.

5579 Background: Patients (pts) with PSOC often experience reduced efficacy and tolerability with each successive treatment (tx). While PARPi therapies have demonstrated clinical benefit in frontline and maintenance settings, most pts eventually experience progression of disease (PD) with limited tx options. Data establishing standard of care for PSOC was published prior to the PARPi era. This study evaluated the efficacy of tx in PSOC subsequent to PARPi exposure. Methods: Pooled pt-level data from <5 multinational clinical trials (CT) involving PARPi tx were sourced from the Medidata Clinical Cloud and included pts with PSOC diagnosis, ≥2 prior lines of platinum-based chemotherapy (PBC), most recent platinum-free interval (PFI) ≥6 months (mo), prior PARPi tx, initiation of tx subsequent to PARPi (defined as the index tx), and ECOG Performance Status (PS) ≤1 prior to the index tx. Index date was defined as the initiation of index tx. Outcomes included overall response rate (ORR), progression-free survival (PFS), and overall survival (OS). PFS was defined as the time from the index date to PD, death or start of new tx; pts with no PFS event were censored at the end of follow-up. PFS and OS were estimated using the Kaplan-Meier method and compared across subgroups with the log-rank test. Exploratory subgroup analyses of PFS and OS were conducted per factors identified in multivariable Cox models. Results: Among 130 pts (≥65 years, 36.9%; White, 87.7%; FIGO Stage III/IV, 89.2%; ECOG PS 0/1, 59.2%/40.8%; ≥3 prior lines of PBC, 23.8% [range, 2-5]; PFI ≥12 mo, 61.5%), the median duration of PARPi use was 13.24 mo (IQR, 9.22), and 96.9% experienced PD ≤106 days from PARPi discontinuation. Index tx included PBC (74.6%) and non-PBC (25.4%); 61.5% received combination tx. Median duration of index tx was 4.01 mo (95% CI, 3.71-4.93). ORR on index tx was 16.9% (95% CI, 10.9-24.5). Median PFS was 6.11 mo (95% CI, 5.06-7.39), with longer PFS in pts with PFI ≥12 mo vs 6 to <12 mo (7.39 mo [95% CI, 6.08-8.77] vs 4.44 mo [3.25-6.14]; P =0.002) and in pts with combination therapy vs monotherapy index tx (7.39 mo [95% CI, 6.21-8.61] vs 3.71 mo [95% CI, 3.12-5.75]; P =0.026). Median OS was 19.35 mo (95% CI, 17.77-22.08), with longer OS in pts with PFI ≥12 mo vs 6 to <12 mo (23.03 mo [95% CI, 19.29-33.81] vs 15.08 mo [95% CI, 11.89-19.68]; P <0.0001). PFS and OS were similar in PBC vs non-PBC as index tx. Conclusions: Median PFS with 3L+ tx for PSOC following PARPi exposure was 6.11 mo, establishing an efficacy benchmark for tx subsequent to PARPi exposure in this unique pt population and highlighting the need for more effective tx. Due to the lack of regular per-protocol imaging assessments after the start of non-trial tx, PFS values may be overestimated. However, pooled CT data with long-term follow-up provide valuable insights not available from other sources.