Epidemiological studies related to androgens and prostate cancer (PC) have focused on serum determination of testosterone, androstenedione (A4), and DHEA, with inconsistent results. Herein, we hypothesized that differences in androgen biosynthetic and metabolic pathways, rather than differences in specific androgen concentrations, are associated with prostatic carcinogenesis. Therefore, spot urine samples from 111 incident PC cases with Gleason score at diagnosis and 227 healthy population controls, were analyzed. Urinary androgen concentrations (nanograms/milligrams of creatinine) were determined by ultra-performance liquid chromatography-tandem mass spectrometry (UPLC-MS). Using a factor analysis, we identified three androgen urinary excretion patterns. In a subsample, we evaluated a modification effect of the androgen receptor (AR) CAG polymorphism. Pattern I, characterized by A4 and testosterone hydroxylated metabolites (11β-OHT; 2β-OHT; 15β-OHT; 2α-OHT; 6β-OHT), was associated with high PC odds among carriers of AR gene (CAG)>19 repeats (OR: 3.67 95% CI: 1.23-11.0; P for interaction= 0.009). Conversely, higher testosterone excretion (pattern III), was marginally associated with lower (OR: 0.35 95% CI: 0.12-1.00, P for trend= 0.08) poorly differentiated PC (Gleason ≥8). No clear association was observed with pattern II (DHEA; 16α and 16β-OHT). Our results were consistent with the previous evidence which suggests that the C11-oxy backdoor pathway is important for prostatic carcinogenesis. Androgen urine excretion analysis could be useful for PC diagnosis, treatment, and prognosis; however, further studies with a larger number of samples and the urinary determination of 11-ketoandrogens are necessary.
Keywords: 11-oxyandrogens; androgen receptor; prostate cancer; urinary androgen excretion.