Does baseline total testosterone improve the yielding of prostate cancer screening?

Abstract

Previous studies suggest an association between total testosterone (tT) and prostate cancer, but results are conflicting and it is not clear if accounting for the tT levels improves the yielding of patient selection for prostatic biopsy. We evaluated the potential for tT levels and the tT/total PSA (tPSA) ratio to be used as diagnostic tools for prostate cancer and its relation with cancer aggressiveness. We measured tT, tPSA and free PSA (fPSA) in fasting blood samples of 1570 subjects consecutively referred for prostate biopsy due to abnormal digital rectal examination and/or elevated tPSA levels. These values were compared between groups defined according to the pathological results of the biopsy. No significant difference was observed in tT levels when comparing cases with prostate cancer, high grade prostate intraepithelial neoplasia, pathological prostatitis, benign prostatic hyperplasia or no alteration (median: 4.26 versus 4.44 versus 4.31 versus 4.16 pg/mL, respectively; p = 0.643). The tT/tPSA ratio had a better area under the curve than tT alone (0.62, 95% CI, 0.59-0.65 versus 0.50, 95% CI, 0.47-0.53), but worse than the f/tPSA ratio (0.70, 95% CI, 0.67-0.73). In multivariate analysis, using the median of distribution as cut-off no significant association was observed between tT or tT/tPSA and prostate cancer (OR = 1.06, 95% CI, 0.84-1.33; OR = 0.94, 95% CI, 0.70-1.27, respectively). The tT levels were not significantly different across Gleason score groups (p = 0.553). In patients suspected of having prostate cancer the tT levels are not useful to improve the yielding of patient selection for prostatic biopsy or to predict cancer aggressiveness. © 2011 Elsevier Ltd. All rights reserved.