Efficacy of olaparib (ola) plus abiraterone (abi) versus placebo (pbo) plus abi in patients (pts) with metastatic castration-resistant prostate cancer (mCRPC) with a germline or somatic BRCA mutation in the PROpel trial.

219 Background: PROpel (NCT03732820) met its primary endpoint showing statistically significant investigator-assessed (INV) radiographic progression-free survival (rPFS) benefit with ola + abi vs pbo + abi in first-line mCRPC in pts enrolled irrespective of homologous recombination repair gene mutation (HRRm) status (intention-to-treat [ITT] hazard ratio [HR] 0.66, 95% CI 0.54–0.81; P <0.001). At final prespecified analysis (ITT), median overall survival (OS) with ola + abi vs pbo + abi was 42.1 vs 34.7 months (HR 0.81, 0.67–1.00; P =0.054). In pts assigned to HRRm subgroups using aggregated tumor tissue and circulating tumor DNA (ctDNA) test results ( post hoc analyses), the greatest benefit for ola + abi vs pbo + abi was in pts with BRCAm (rPFS HR 0.23, 0.12–0.43; OS HR 0.29, 0.14–0.56). Concordance based on HRRm status between tumor tissue and ctDNA testing was also high (80% +ve, 87% -ve predictive agreement). We report post hoc efficacy analyses in pts with BRCAm of germline (g) or somatic (s) origin. Methods: PROpel was a double-blind Phase III trial. Pts were randomized 1:1 to ola (300 mg twice daily [bid]) or pbo, and abi (1000 mg once daily) + prednisone/prednisolone (5 mg bid) until disease progression, unacceptable toxicity, or withdrawal of consent. Tumor BRCAm status was determined using aggregated results from tumor tissue (FoundationOne CDx) and ctDNA (FoundationOne Liquid CDx) tests, and g/s status by blood test (Myriad MyRisk). Results: Of the 85 BRCAm pts, 77 were evaluable for g/s status by blood test. Of these, 32% (n=25) had a BRCAm of g origin and 68% (n=52) of s origin. HRs for pts with g and s BRCAm for rPFS (INV 0.13 and 0.19, BICR 0.15 and 0.17) and OS (0.23 and 0.26) all favored ola + abi vs pbo + abi (Table). Conclusions: Ola + abi showed clinical benefit vs pbo + abi for rPFS and OS in pts with g or s BRCAm, supporting earlier findings in the overall BRCAm population of PROpel. Also, as g testing alone does not detect s mutations, these results highlight the importance of robust biomarker testing (which is currently underutilized in real-world practice), including tumor tissue or ctDNA testing, to inform treatment options. Clinical trial information: NCT03732820 . Germline BRCAm, (n=25) Somatic BRCAm, (n=52) Ola + Abi(n=15) Pbo + Abi (n=10) Ola + Abi (n=27) Pbo + Abi (n=25) rPFS* (INV, primary endpoint) Events, n (%) 5 (33.3) 10 (100) 7 (25.9) 17 (68.0) Median rPFS NR 6.9 NR 11.1 HR 0.13 (0.04–0.38) 0.19 (0.07–0.45) rPFS*(BICR, sensitivity analysis) Events 5 (33.3) 10 (100) 6 (22.2) 18 (72.0) Median rPFS NR 5.9 NR 9.1 HR 0.15 (0.05–0.45) 0.17 (0.06–0.41) OS † (key secondary endpoint) Events, n (%) 5 (33.3) 9 (90.0) 6 (22.2) 15 (60.0) Median OS NR 18.4 NR 27.5 HR 0.23 (0.07–0.67) 0.26 (0.09–0.65) *Primary analysis, DCO 30 July 2021; † Final prespecified analysis, DCO 12 Oct 2022. BICR, blinded independent central review; DCO, data cutoff; NR, not reached.