Testosterone replacement therapy is not associated with increased prostate cancer incidence, prostate cancer-specific, or cardiovascular disease-specific mortality in Finnish men

Background: Concerns have been expressed over the safety of testosterone replacement therapy (TRT) in men with late-onset hypogonadism (LOH). Previous studies have shown controversial results regarding the association of TRT with the risk of cardiovascular events or prostate cancer (PCa) incidence, aggressiveness, and mortality. This study explores the overall risk of PCa and risk by tumor grade and stage, as well as mortality from PCa and cardiovascular disease (CVD), among men treated with TRT compared to men without LOH and TRT use. Materials and methods: The study included 78,615 men of age 55–67 years at baseline from the Finnish Randomized Study of Screening for Prostate Cancer (FinRSPC). Follow-up started at randomization and ended at death, emigration, or a common closing date January 1st, 2017. Cox proportional hazards regression model with time-dependent variables and adjustment for age, trial arm, use of other medications, and Charlson comorbidity index was used. Comprehensive information on TRT purchases during 1995–2015 was obtained from the Finnish National Prescription Database. PCa cases were identified from the Finnish Cancer Registry and causes of death obtained from Statistics Finland. Results: Over the course of 18 years of follow-up, 2919 men were on TRT, and 285 PCa cases were diagnosed among them. TRT users did not exhibit a higher incidence or mortality rate of PCa compared to non-users. On the contrary, men using TRT had lower PCa mortality than non-users (HR = 0.52; 95% CI 0.3–0.91). Additionally, TRT users had slightly lower CVD and all-cause mortality compared to non-users (HR = 0.87; 95% CI 0.75–1.01 and HR = 0.93; 95% CI 0.87–1.0, respectively). No time- or dose-dependency of TRT use was evident in any of the analyses. Conclusion: Men using TRT were not associated to increased risk for PCa and did not experience increased PCa- or CVD-specific mortality compared to non-users. Further studies considering blood testosterone levels are warranted.