Prognostic value of testosterone castration levels following androgen deprivation and high-dose radiotherapy in localized prostate cancer: Results from a phase III trial.

BACKGROUND/OBJECTIVE The optimal prognostic value of testosterone following androgen deprivation therapy (ADT) is controversial. We studied the effect of serum testosterone levels on clinical outcome in localized prostate cancer (PCa) treated with ADT and high-dose radiotherapy (HRT). PATIENTS AND METHODS The DART01/05 trial randomized 355 men with intermediate and high-risk PCa to 4 months of ADT plus HRT (STADT, N = 178) or the same treatment followed by 24 months of ADT (LTADT, N = 177). This study included patients treated with LTADT who had at least 3 determinations of testosterone during ADT (N = 154). Patients were stratified into 3 subgroups by testosterone level: minimum <20 ng/dL; median 20-49 ng/dL; and maximum ≥50 ng/dL. Kaplan-Meyer and Cox regression analysis were used for overall survival (OS) and Fine & Gray regression model for metastasis free survival (MFS), biochemical disease-free survival (bDFS) and time to TT recovery. RESULTS There were no statistically significant differences in 10-year bDFS, MFS, or OS between the <20 ng/mL and 20-49 ng/dL subgroups. Multivariate analysis showed that a median testosterone ≥50 ng/dL was significantly associated with a decrease in bDFS (HR: 6.58, 95%CI 1.28-33.76, p = 0.03). Time to testosterone recovery after ADT did not correlate with bDFS, MFS, or OS and was not significantly associated with any of the testosterone subgroups. CONCLUSIONS Our results do not support the concept that additional serum testosterone suppression below 20 ng/dL is associated with better outcomes than 20-49 ng/dL. Time to testosterone recovery after ADT and HRT did not impact clinical failure.