Search

Simple Summary: Active surveillance is the preferred treatment strategy for low-risk prostate cancer and includes regular monitoring by control biopsies, which bear the risk of various side effects. In case of risk reclassification, therapy is switched to radical treatment. Currently used clinical parameters only possess a limited capability to indicate risk reclassification. Molecular markers identified via liquid biopsies, such as urine, could facilitate the detection of aggressive disease since they provide a more global assessment of prostate cancer than tissue biopsies. Moreover, an improved predictability could reduce the number of control biopsies needed during active surveillance. In this study, we identified a set of molecular markers from the urine of men on active surveillance that could predict the outcome of control biopsies. The combination of these molecular markers with clinical parameters resulted in further improved predictability of risk reclassification and, thus, has the potential to refine the monitoring strategies in active surveillance. Serum prostate-specific antigen (PSA), its derivatives, and magnetic resonance tomography (MRI) lack sufficient specificity and sensitivity for the prediction of risk reclassification of prostate cancer (PCa) patients on active surveillance (AS). We investigated selected transcripts in urinary extracellular vesicles (uEV) from PCa patients on AS to predict PCa risk reclassification (defined by ISUP 1 with PSA > 10 ng/mL or ISUP 2-5 with any PSA level) in control biopsy. Before the control biopsy, urine samples were prospectively collected from 72 patients, of whom 43% were reclassified during AS. Following RNA isolation from uEV, multiplexed reverse transcription, and pre-amplification, 29 PCa-associated transcripts were quantified by quantitative PCR. The predictive ability of the transcripts to indicate PCa risk reclassification was assessed by receiver operating characteristic (ROC) curve analyses via calculation of the area under the curve (AUC) and was then compared to clinical parameters followed by multivariate regression analysis. ROC curve analyses revealed a predictive potential for AMACR, HPN, MALAT1, PCA3, and PCAT29 (AUC = 0.614–0.655, p < 0.1). PSA, PSA density, PSA velocity, and MRI maxPI-RADS showed AUC values of 0.681–0.747 (p < 0.05), with accuracies for indicating a PCa risk reclassification of 64–68%. A model including AMACR, MALAT1, PCAT29, PSA density, and MRI maxPI-RADS resulted in an AUC of 0.867 (p < 0.001) with a sensitivity, specificity, and accuracy of 87%, 83%, and 85%, respectively, thus surpassing the predictive power of the individual markers. These findings highlight the potential of uEV transcripts in combination with clinical parameters as monitoring markers during the AS of PCa. [ABSTRACT FROM AUTHOR]