Serum testosterone to prostate specific antigen ratio as predictor of prostate cancer: a diagnostic test evaluation

Abstract

Background: Prostate cancer is suspected by an increase in serum PSA level, and the diagnosis is confirmed with a systematic biopsy. There is no established cutoff value of PSA levels for recommending biopsy and it shows a lack of specificity when PSA levels are in the “gray zone”. The objectives are to study the diagnostic ability and effectiveness of serum testosterone-PSA ratio in predicting prostate cancer and its severity in patients with normal DRE and serum PSA level of 4-20 ng/ml. Methods: Patients more than 40 years of age with normal DRE and elevated serum PSA (4-20 ng/ml) admitted in the Department of Urology, Government medical college Kozhikode will be included. The sTT-PSA ratio will be calculated with sTT and PSA in ng/ml. 12 core prostate biopsy is performed. The HPR results will be depicted as benign and malignant, with malignancy as grade groups (1-5) based on Gleason scoring. Results: In our study cancer detection rate is 20.8%. Mean PSA and testosterone levels are 10.01ng/ml and 4.84 ng/ml in benign group, 15.45 ng/ml and 4.03ng/ml in cancer group. Mean T-PSA ratio are 0.52 and 0.27 in benign and cancer group respectively. Based on ROC curve with a cut off 0.336, T-PSA has the diagnostic ability for Ca prostate with sensitivity of 79.17% and specificity of 89.01%. There is a positive correlation between T-PSA ratio and grade group but it is statistically not significant. Conclusions: The present study suggests that the sTT/PSA ratio might be considered a predictor of the risk of prostate cancer.