Urinary microRNA-based signature improves accuracy of detection of clinically relevant prostate cancer within the prostate-specific antigen grey zone.

At present, prostate-specific antigen (PSA) is used as a clinical biomarker for prostate cancer (PCa) diagnosis; however, a large number of patients with benign prostate hyperplasia (BPH) with PSA levels in the 'gray area' (4-10 ng/ml) are currently subjected to unnecessary biopsy due to overdiagnosis. Certain microRNAs (miRs) have been proven to be useful biomarkers, several of which are detectable in bodily fluids. The present study identified and validated a urinary miR‑based signature to enhance the specificity of PCa diagnosis and to reduce the number of patients with benign conditions undergoing biopsy. Seventy‑three urine samples from Mexican patients with diagnosis of PCa with a Gleason score ≥7 and 70 patients diagnosed with BPH were collected after digital rectal examination (DRE) of the prostate. miR expression profiles were determined using TaqMan Low Density Array experiments, and normalized Ct values for the miRs were compared between PCa and BPH groups. Receiver operating characteristic (ROC) curve analysis was performed to evaluate whether miR detection in urine is suitable for distinguishing patients with PCa from those with BPH. The identified miR‑100/200b signature was significantly correlated with PCa. Using a multivariable logistic regression approach, a base model including the clinical variables age, prostate‑specific antigen (PSA), the percentage of free PSA and DRE was generated, and a second base model additionally contained the miR‑100/200b signature. ROC analysis demonstrated that the combined model significantly outperformed the capacity of PSA (P<0.001) and the base model (P=0.01) to discriminate between PCa and BPH patients. In terms of evaluation of the sub‑group of patients in the gray zone of PSA levels, the performance of the combined model for predicting PCa cases was significantly superior to PSA level determination (P<0.001) and the base model (P=0.009). In addition, decision curve analysis demonstrated that the use of the combined model increased the clinical benefit for patients and produced a substantial reduction in unnecessary biopsies across a range of reasonable threshold probabilities (10‑50%). Detection of the urinary miR signature identified in the present study as part of clinical diagnostic procedures will enhance the accuracy of PCa diagnosis and provide a clinical benefit for patients with BPH by sparing them from undergoing invasive biopsy. To the best of our knowledge, the present study was the first to describe the profiling of urinary miR100 and miR-200b levels for the clinical diagnosis of PCa.