The Phase 3 TALAPRO-2 trial met its primary endpoint, showing improved radiographic progression-free survival (rPFS) for TALA + ENZA vs placebo (PBO) + ENZA as 1L treatment in pts with mCRPC unselected for homologous recombination repair (HRR) gene alterations (all-comers; cohort 1). Here we report final OS data, a descriptive update of rPFS, and extended safety follow-up in cohort 1.
Methods
In cohort 1, pts were randomized 1:1 to ENZA 160 mg + either TALA 0.5 mg (0.35 mg if moderate renal impairment) or PBO once daily and stratified by prior abiraterone or docetaxel (yes/no) for castration-sensitive PC and HRR gene alteration status. Key eligibility criteria included asymptomatic or mildly symptomatic mCRPC, ECOG PS ≤1, ongoing androgen deprivation therapy, and no prior life-prolonging therapy for CRPC. The primary endpoint was rPFS by blinded independent central review. OS was an alpha-protected key secondary endpoint. For statistical significance at the final OS analysis, the stratified log-rank 2-sided P value needed to be ≤0.022 using a group sequential design with O’Brien-Fleming spending function.
Results
Overall, 805 pts were randomized, 402 to TALA+ENZA and 403 to PBO+ENZA. At data cutoff (Sept 3, 2024), 211 pts (52%) in the TALA + ENZA arm and 243 pts (60%) in the PBO + ENZA arm had died; median follow-up was 52.5 and 53.0 months, respectively. Hazard ratio (HR) for OS with TALA + ENZA vs PBO + ENZA was 0.796 (95% CI, 0.661–0.958; 2-sided P=0.0155); median OS (95% CI), 45.8 months (39.4–50.8) vs 37.0 months (34.1–40.4 months), respectively. In prespecified subgroup analyses, OS favored TALA + ENZA vs PBO + ENZA in pts who were HRR-deficient (n=169; HR, 0.549; 95% CI, 0.364–0.826; P=0.0035) or HRR–non-deficient/unknown (n=636; HR, 0.878; 95% CI, 0.713–1.080; P=0.218). In exploratory analyses of pts with results available for both circulating tumor DNA and tumor tissue, OS favored TALA + ENZA vs PBO + ENZA in pts without BRCA1/2 alterations (n=439; HR, 0.749; 95% CI, 0.582–0.963; P=0.024) and in pts without HRR alterations (n=314; HR, 0.782; 95% CI, 0.582–1.050; P=0.101). Consistent with the primary analysis, updated rPFS data favored TALA + ENZA vs PBO + ENZA (HR, 0.667; 95% CI, 0.551–0.807; P<0.0001); median rPFS, 33.1 vs 19.5 months, respectively. Consistent with primary results, the most common grade ≥3 TEAEs with TALA + ENZA were anemia (49%) and neutropenia (19%). TEAEs were generally manageable; 86 pts (22%) discontinued TALA due to TEAEs.
Conclusions
TALA + ENZA demonstrated a statistically significant and clinically meaningful improvement in OS vs standard-of-care ENZA as 1L treatment in pts with mCRPC unselected for HRR gene alterations. rPFS continued to favor TALA + ENZA. No new safety signals were identified with extended follow-up. Clinical trial information: NCT03395197.