5072 Background: Taxane-based chemotherapy (taxane) and poly(ADP-ribose) polymerase inhibitors (PARPi) are established life-prolonging therapies for patients with homologous recombination repair (HRR)–altered metastatic castration-resistant prostate cancer (mCRPC), particularly tumors with BRCA1/2 alterations. However, data informing the optimal sequencing of PARPi and taxane chemotherapy in this patient population is limited, especially because subsequent treatments may not be captured in reports of clinical trials. Real-world evidence may help describe treatment sequencing patterns and outcomes. Methods: We conducted a retrospective cohort study using the nationwide Veterans Affairs (VA) healthcare system. Patients with mCRPC and HRR alterations treated with either PARPi or taxanes for mCRPC were included. HRR mutation status was ascertained from genomic results captured within the VA database and included different platform/assays. Baseline characteristics were assessed at mCRPC treatment initiation. The primary endpoint was overall survival (OS) from the initiation of PARPi or taxane. Secondary analyses included OS comparisons by initial treatment class and OS from initiation of second-line therapy among patients who received both PARPi and taxane chemotherapy. Survival was estimated using Kaplan–Meier methods and compared using Cox proportional hazards models with inverse probability of treatment weighting (IPTW) to adjust for baseline differences. Results: Between Jan 2014 - May 2025, a total of 418 patients (26.8% non-Hispanic Black, median age 74.5 years), with HRR-altered mCRPC (31.3% BRCA2, 26.8% ATM, 21.1% CDK12, 13.4% CHEK2) were included in the final analysis. Metastatic involvement included bone metastases in 80.9% of patients, and visceral metastases in 11.2%. Median PSA was 30.2 ng/mL. Prior to mCRPC, 58.6% of patients had received ARPIs and 14.1% had received taxane chemotherapy. With a median follow-up of 14.1 months, patients received PARPi (N=211) or taxane (N=207), and 30.1% received both treatments for mCRPC. OS was similar between patients initially treated with PARPi versus taxane for mCRPC (HR 1.09, 95% CI 0.84-1.41). Among patients who received both therapies, OS did not differ based on treatment sequence (PARPi→taxane vs taxane→PARPi; HR 0.90, 95% CI 0.41-1.85). Conclusions: In this real-world VA cohort of HRR-altered mCRPC, overall survival was similar whether PARP inhibitors or taxanes were used as initial therapy and did not differ by subsequent treatment sequence. Treatment crossover was observed in routine clinical practice.
Corrigan et al. (2026) studied this question.