Testosterone replacement therapy in men with prostate cancer: a time-varying analysis.

INTRODUCTION The use of testosterone replacement therapy (TRT) in men with prostate cancer is controversial given concerns of androgen-related cancer progression. Although emerging evidence suggests that TRT may be safe in this setting, no study has investigated dose-related effects. AIM We used time-varying analysis to determine whether increasing TRT exposure is associated with worse outcomes. METHODS Using linked Surveillance, Epidemiology, and End Results-Medicare data, we identified 149,354 men diagnosed with prostate cancer from 1991 to 2007. Subjects treated with TRT were stratified by duration of treatment. Weighted propensity score methods were used to adjust for differences between groups. A Cox proportional hazards model was constructed to assess the effect of injectable TRT exposure on outcomes. MAIN OUTCOME MEASURE Overall mortality (OM), prostate cancer-specific mortality (PCSM), and use of salvage androgen deprivation therapy (ADT). RESULTS Men treated with TRT, regardless of duration, did not experience higher OM or PCSM (all hazard ratio [HR] <1.0, all P ≤ 0.002). We found no difference in use of salvage ADT in the ≤ 30-day and 31-60 day groups compared with no-TRT (HR 1.23 and 1.05, P=0.06 and 0.81, respectively), whereas it was lower for men on long-term TRT (HR 0.70, P=0.04). CONCLUSIONS TRT following prostate cancer diagnosis and treatment does not increase mortality or the use of salvage ADT. Using time-varying analysis, we demonstrate that longer duration of TRT is not associated with adverse mortality or greater need for ADT.