UC Irvine

Importance: Prostate cancer remains as one of the most common malignancies in men in the United States. Testosterone Replacement Therapy (TRT) is known to improve functional and sexual outcomes in hypogonadal men. TRT historically has been contraindicated for the management of hypogonadism in men with prostate cancer (PC). In 2020, our team demonstrated in a cohort of 850 men that TRT reduced biochemical recurrence (BCR) by 53% (median follow-up 3.5 yrs.). The present study extends the median follow-up time up to 5.84 years [4.18; 7.24].

Objective: This study mainly aims to investigate the utility of TRT to reduce BCR.

Methods: A retrospective review of prospectively collected data between December 2009 and June 2018 undergoing robotic radical prostatectomy by a single surgeon was conducted. 152 patients who had TRT post-RP with a cFT (below 25%) were frequency-matched to 420 patients not receiving TRT with respect to pathological stage (p-stage) and Gleason Grade Group (GGG). Furthermore, the TRT group was propensity-score matched to 304 patients. Propensity scores were calculated using a multivariable logistic model that incorporated the following variables: age, BMI, Gleason group, PSA, SHIM, prostate size, pathological stage, total testosterone (TT), sex hormone-binding globulin (SHBG), and free testosterone (FT). Biochemical recurrence (BCR) was defined as two consecutive prostate specific antigen (PSA) blood tests >0.2 ng/mL. A cox regression model and Kaplan-Meier (KM) graphs were used to analyze rate and time to BCR. We also used our cox regression model of the time to BCR with truncated time of 10.32 to calculate restricted mean survival time (RMST) for the TRT vs no-TRT groups.

Results: With propensity-score matching, there were no baseline differences in demographics observed except in TT (p-value= <0.001), cFT (p-value <0.001), and follow-up time (p-value = 0.004). In the cox regression multivariate analysis of the propensity- scored matched control, patients with a higher calculated free testosterone were less likely to have a BCR (p-value= 0.013). In the same analysis, patients with a higher GGG, p-stage, and preoperative PSA (p-value= <0.001) were more likely to have a BCR. Patients receiving TRT had 47% reduction in BCR rates (p-value = 0.011) after a median follow-up time of 5.84 [4.18;7.24] years. Furthermore, our 10-year cox regression model for the time to BCR showed RMST difference between the TRT and no-TRT groups: in patients who did recur, patients on TRT had an increased latency of 0.44 years.

Conclusion and Relevance: With the long-term follow-up, TRT was observed to have a continuing protective effect in preventing and delaying BCR in patients post-RARP when compared to patients who did not receive TRT.