RESUMEN
OBJECTIVE: To evaluate preoperative total testosterone (TT) as a predictor of positive surgical margins (PSM) in prostate cancer (PCA). PATIENTS AND METHODS: During the period from November 2014 to July 2017, preoperative TT was measured in 476 PCA patients undergoing only radical prostatectomy (RP) and including all risk classes. Surgical margins were stated negative, focal positive (single and less than 1 mL), and multifocal positive (more than 1). The risk of TT and clinical factors associated with the risk of PSM (focal or multifocal versus negative) was evaluated by the multinomial logistic regression model. RESULTS: Overall, PSM were detected in 149 cases (31.3%), which included 99 patients with focal cancer invasion (20.8%) and 50 subjects with multifocal cancer invasion (10.5%). In univariate analysis, PSM associated with higher median levels of TT and prostate-specific antigen than controls. Multifocal PSM associated with higher rates of high-risk PCA (42%) than focal (22.2%) or control cases (18.3%). In multivariate analysis, TT was the only independent factor positively associated with the risk of focal PSM when compared to controls (OR 1.002; p = 0.035). TT (OR 1.003; p = 0.002) and high-risk PCA (OR 1.002; p = 0.047) were independent factors, which positively associated with the risk of multifocal PSM when compared to controls. Risk models were computed. CONCLUSIONS: In a large and contemporary cohort of patients elected to primary RP, TT was an -independent positive factor associated with the risk of focal and multifocal PSM. TT associated with aggressive PCA biology.