PD34-09 EFFICACY OF PHYSICIAN'S CHOICE OF ENZALUTAMIDE OR ABIRATERONE IN THE CONTROL ARM OF PROFOUND

Abstract

INTRODUCTION AND OBJECTIVE: PROfound (NCT02987543) was the first positive Phase III PARP inhibitor trial demonstrating that men with metastatic castration-resistant prostate cancer (mCRPC) with alterations in homologous recombination repair (HRR) genes whose disease had progressed on next-generation hormonal agent (NHA) had significantly prolonged radiographic progression-free survival (rPFS) with olaparib (ola) versus physician's choice of abiraterone (abi) or enzalutamide (enza). A significant improvement in overall survival (OS) with olaparib was demonstrated in Cohort A (BRCA1, BRCA2 or ATM). Considerate of a potential difference in the efficacy of NHA sequencing, we report an exploratory analysis of the efficacy of ola versus either abi or enza. METHODS: Patients (pts) had mCRPC with alterations in ≥1 of 15 genes with a direct or indirect role in HRR and had disease progression on prior NHA. Pts were randomized 2:1 to ola or control of abi or enza. At radiographic disease progression, pts could cross over to ola if they met the study criteria; OS is reported without adjustment for crossover. RESULTS: Prior to enrollment, 40% of pts had received abi only, 41% had received enza only and 19% had received both during their treatment. In Cohort A, median rPFS was 7.39 months (m) for pts in the ola arm versus 3.52 and 3.55 m for those receiving abi or enza, respectively (Table). Of the pts who received abi or enza, 66% crossed over to receive ola at disease progression. In Cohort A, OS was 19.09 m in the ola arm, and 14.51 and 14.87 m for pts receiving abi and enza, respectively. The efficacy of abi and of enza is consistent with previous reports of the control arm in PROfound. CONCLUSIONS: These exploratory analyses highlight the benefit of ola versus abi or enza, supporting the statistically significant benefit in the prespecified analysis for ola versus control for rPFS and OS in Cohort A and for rPFS in the overall study population. The benefit of sequential use of NHA appears to be limited, with no differentiation in outcomes observed between the pts receiving abi or enza. Importantly, improvement in OS was observed despite substantial crossover to ola and suggests a benefit for giving ola earlier in the treatment sequence for men with mCRPC with alterations in HRR genes and progression on prior NHA.