Association of Long-Term Dynamics in Circulating Testosterone with Serum PSA in Prostate Cancer-Free Men with Initial-PSA < 4ng/mL.

Abstract

We previously reported that an accelerated decline in circulating testosterone level is associated with a higher risk of prostate cancer (PCa). This study is to examine whether testosterone change rate is related to serum prostate-specific antigen (PSA) concentration among PCa-free men. Longitudinal data were derived from electronic medical records at a tertiary hospital in the Southeastern USA. PCa-free men with initial-PSA < 4ng/mL and ≥ 2 testosterone measurements were included (n = 632). Three PSA measures (peak, the most recent, and average PSA) during the study period (from first testosterone measurement to the most recent hospital visit) were examined using multivariable-adjusted geometric means and were compared across quintiles of testosterone change rate (ng/dL/month) and current testosterone level (cross-sectional). Mean (standard deviation, SD) age at baseline was 59.3 (10.5) years; mean study period was 93.0 (55.3) months. After adjusting for covariates including baseline testosterone, the three PSA measures all significantly increased across quintile of testosterone change rate from increase to decline (peak PSA: quint 1 = 1.09, quint 5 = 1.41; the most recent PSA: quint 1 = 0.85, quint 5 = 1.00; average PSA: quint 1 = 0.89, quint 5 = 1.02; all Ptrend < 0.001). But current testosterone level was not associated with PSA levels. Stratified analyses indicated men with higher adiposity (body mass index > 24.1kg/m2) or lower baseline testosterone (≤ 296ng/dL) were more sensitive to testosterone change in regard to PSA. Among PCa-free men, accelerated testosterone decline might correlate with higher serum PSA concentration. It will help to elucidate the mechanisms relating aging-accompanying testosterone dynamics to prostate carcinogenesis.