Androgens have been implicated in prostate tumorigenesis, but prospective studies have overall reported no association between circulating levels of androgens and risk of prostate cancer. However, some recent studies have shown that a high level of testosterone increase the risk of non-aggressive tumors but is associated with a decreased risk of aggressive tumors.
We prospectively measured plasma levels of total testosterone, androstanediol glucuronide (A-diol-g) and sex hormone binding globuline (SHBG) and calculated estimated levels of free testosterone, in a nested case-control study of 392 cases and 392 matched controls.
None of the studied hormones were significantly associated with prostate cancer risk in the full study group or in subgroups according to tumor aggressiveness. Odds ratios in the full study group, for top versus bottom quartile, was for total testosterone 1.25 (95% CI = 0.79–2.00; Ptrend = 0.51); free testosterone, 1.31 (95% CI = 0.82–2.07; Ptrend = 0.35); A-diol-g, 0.88 (95% CI = 0.59–1.33; Ptrend = 0.77); and for SHBG, 1.01 (95% CI = 0.64–1.58; Ptrend = 0.94).
We found no significant associations between androgen levels and risk of prostate cancer in this population-based, non-screened cohort. Prostate 67: 1230–1237, 2007. © 2007 Wiley-Liss, Inc.