Nadir Prostate-specific Antigen as an Independent Predictor of Survival Outcomes: A Post Hoc Analysis of the PROSPER Randomized Clinical Trial.

Abstract

This post hoc analysis of PROSPER evaluated the relationship between depth of PSA decline and clinical outcomes in enzalutamide-treated men with nonmetastatic castration-resistant prostate cancer. PROSPER was an international, randomized, double-blind, placebo-controlled, phase 3 trial that demonstrated significantly improved metastasis-free survival and overall survival with androgen deprivation therapy plus enzalutamide vs placebo. A total of 905 enzalutamide-treated men were included in this post hoc analysis. Metastasis-free survival (primary endpoint) and overall survival (secondary endpoint) were evaluated for 4 mutually exclusive subgroups defined by PSA decline: <50% (reference); ≥50% to <90%; ≥90%, nadir ≥0.2 ng/mL; and ≥90%, nadir <0.2 ng/mL. Medians and 95% confidence intervals were determined using a 12-month landmark analysis; hazard ratios and P values were based on an unstratified Cox proportional analysis model. In enzalutamide-treated men, PSA declines of <50%, ≥50% to <90%, ≥90% with nadir ≥0.2 ng/mL, and ≥90% with nadir <0.2 ng/mL were associated with median metastasis-free survival in months (95% confidence intervals) of 22.1 (14.8-not reached), 34.2 (29.4-not reached), 36.6 (33.4-not reached), and not reached, respectively, and overall survival in months (95% confidence intervals) of 40.8 (31.7-44.9), 54.4 (49.0-67.0), 64.3 (63.4-not reached), and not reached, respectively. There was a statistically significant correlation between greater depth of PSA decline and improved clinical outcomes, suggesting a previously underappreciated relationship between changes in PSA levels and clinical outcomes in nonmetastatic castration-resistant prostate cancer.