Résumé
ABSTRACT Objectives To evaluate the predictive value of TRIMprob test to detect prostate cancer (PCa) in patients referred to prostate biopsy (PB). Material and Methods Patients with PSA <10ng/mL and rectal exam without findings suggestive of prostate cancer were selected for TRIMprob evaluation. Exam was performed by a single operator through transperineal approach. Patients admitted for the study were submitted to TRIMprob and multiparametric magnetic resonance (mpMRI) and posteriorly to PB. Results In total, 77 patients were included. TRIMprob showed evidences of PCa in 25 (32.5%) and was negative in 52 patients (67.5%). The rate of detection of prostate cancer at biopsy was higher in patients with positive TRIMprob (16/25; 64.0%) than in patients with negative TRIMprob (11/52; 21.1%; p<0.001). Sensitivity, specificity, positive predictive value, negative predictive value and accuracy of TRIMprob were respectively 61.5%, 82.0%, 64.0%, 80.3% and 74.0%. ROC curve showed the following areas under the curve values for TRIMprob, mpMRI and combination of TRIMprob + mpMRI: 0.706; 0.662 and 0.741 respectively. At combined analysis, when both TRIMprob and mpMRI were negative for prostate cancer, accuracy was 96.3% or only 1 in 27 PB was positive (3.7%). Conclusions Trimprob had similar predictive value for PCa in patients submitted to PB as mpMRI. Combined TRIMprob and mpMRI showed higher accuracy than when performed singly.