Real-world use and outcomes of darolutamide (DARO), enzalutamide (ENZA), and apalutamide (APA) for nonmetastatic castration-resistant prostate cancer (nmCRPC): Race subgroup analysis.

158 Background: Comparative real-world (RW) data on androgen receptor inhibitor (ARI; DARO, ENZA, APA) use in nmCRPC are limited. DEAR-EXT (NCT06013475) compared the RW utilization, tolerability, and outcomes of all 3 ARIs. Here we present a race subgroup analysis, based on data from the Black and White patient (pt) cohorts. Methods: DEAR-EXT is a retrospective chart review cohort study of men aged ≥18 years in the Precision Point Specialty Analytics network of US urology practices who initiated ARI treatment for nmCRPC for the first time (index date 8/2019–3/2023). Pts were categorized into White and Black cohorts based on their self-reported race. Key endpoints were time to and reasons for initial ARI treatment discontinuation, and metastasis-free survival (defined as time from index date until progression to mCRPC or death). Descriptive statistics and time-to-event analyses were conducted using unadjusted KM estimates and Cox proportional hazards models adjusted for baseline characteristics (age, race, insurance coverage, index year, PSA, time from nmCRPC diagnosis, and Gleason score). Results: A total of 282 Black and 923 White pts were included, of whom 41% received DARO, 44% ENZA, and 15% APA, in both race cohorts. Baseline characteristics were well balanced across treatment arms within each race cohort, except Gleason score and index year. Duration of follow-up was consistent across treatment subgroups (25.5–28.6 months). In both cohorts, the proportion of pts who discontinued their initial ARI treatment was lower for DARO vs ENZA or APA (Black: 33%, 45%, 49%; White: 38%, 53%, 50%, respectively). Similarly, in both race cohorts, progression to mCRPC or death was less frequent for DARO than ENZA or APA (Black: 26%, 35%, 42%; White: 34%, 49%, 45%, respectively). In adjusted Cox regression models, the risk of discontinuation and the risk of mCRPC progression or death in the DARO group were consistently lower in Black and White pts compared with ENZA and APA (range of hazard ratios [HRs] 0.49–0.79), while the differences between ENZA and APA were much less pronounced (HRs 0.80–1.16; Table). The most common reasons for ARI discontinuation were adverse events and disease progression/death, both occurring less frequently in the DARO group. Conclusions: DEAR-EXT real-world data reinforce DARO as an effective, well-tolerated treatment for nmCRPC, demonstrating potentially greater clinical benefits vs ENZA and APA among Black and White pts. Clinical trial information: NCT06013475 . Adjusted HR (95% Cl) DARO vs ENZA DARO vs APA ENZA vs APA Black pts Risk of ARI discontinuation 0.62 (0.40–0.97) 0.49 (0.28–0.88) 0.80 (0.47–1.37) Risk of mCRPC or death 0.79 (0.47–1.31) 0.67 (0.35–1.28) 0.85 (0.47–1.55) White pts Risk of ARI discontinuation 0.68 (0.54–0.85) 0.72 (0.53–0.96) 1.05 (0.80–1.38) Risk of mCRPC or death 0.60 (0.47–0.75) 0.69 (0.51–0.94) 1.16 (0.87–1.55)