1039 - Baseline serum testosterone - does it influence androgen deprivation therapy outcomes in hormone naïve advanced prostate cancer patients?

Abstract

Purpose To study baseline serum testosterone's prognostic value in hormone naive advanced prostate cancer patients receiving continuous androgen deprivation therapy. Materials and methods The study population undergoing continuous androgen deprivation therapy (agonist or antagonist) with 1-year follow-up was pooled for post-hoc analysis from two large prospective, randomized, parallel-arm phase 3b trials (NCT00295750-Global; NCT00928434-USA). Survival end-points were evaluated for baseline serum testosterone effect as a continuous variable, and compared for low ( > 250 ng/dL) groups based on the saturation model, using Kaplan Meier survival estimates, log rank test, and Cox proportional hazards regression models incorporating established clinically important baseline factors. Results Limitations: On intention-to-treat analysis, 138 (16.5%) of 838 eligible men had baseline serum testosterone > 20 ng/ml categories (38% each). The lowest baseline serum testosterone quartile cut-off value was 282 ng/dL (n=206). Multi-variable analysis showed a significant baseline serum testosterone effect for all survival end points. For the saturation model low cut-off Conclusions Lower baseline serum testosterone was significantly associated with worse study survival end-points in hormone naive advanced prostate cancer patients undergoing continuous medical castration. Future well-designed studies should compare continuous androgen deprivation therapy (the current gold standard) with newer alternatives, to optimize individualized management in these men.