Prospective validation of a urine-based proteomics test for predicting clinically significant prostate cancer in biopsy-naïve patients.

BACKGROUND AND

objectiveProstate cancer (PCa) remains a major health burden, while existing diagnostic approaches are limited by overdiagnosis that leads to unnecessary biopsies, and by underdetection of clinically significant disease (csPCa). This study aims to prospectively validate a urine-based 19-biomarker model (19-BM) using capillary electrophoresis-mass spectrometry (CE-MS) for predicting csPCa in patients at risk.

methodsOne hundred sixty-one biopsy-naïve patients at risk for csPCa underwent pre-biopsy CE-MS analysis in urine, multiparametric magnetic resonance imaging (mpMRI), and biopsy at two major hospitals. The 19-BM performance was assessed in comparison with Prostate-Specific Antigen (PSA), the European Randomised Study of Screening for Prostate Cancer (ERSPC), and mpMRI. Performance metrics included Area Under the ROC Curve (AUC), sensitivity, specificity, negative, and positive predictive values. One hundred patients receiving radical prostatectomy were evaluated to investigate the correlation of the 19-BM scores with the prostatectomy pathology. KEY FINDINGS AND LIMITATIONS: The 19-BM scores achieved an AUC of 0.79, outperforming PSA (AUC 0.54, p < 0.0001), ERSPC (AUC 0.63, p = 0.0108), PSAD (AUC 0.61, p = 0.0021) and mpMRI (AUC 0.65, p = 0.0069). In patients with low Prostate Imaging Reporting and Data System scores (PI-RADS), 19-BM showed a sensitivity of 75% and specificity of 90%. Prospective validation of a previous nomogram combining 19-BM with mpMRI yielded an AUC of 0.82 (with specificity of 90%, sensitivity 64%).

conclusionsAfter prospective validation, the 19-BM demonstrates robust performance in predicting csPCa. The urine-based 19BM can serve as a valuable complement to current diagnostic methods, potentially reducing the need for unnecessary invasive procedures, also when combined with mpMRI.