Rapid Noninvasive Classification of Prostatic Disease Using Paper Spray Ionization Mass Spectrometry (PSI-MS)-Based Nontargeted Metabolomics

Paper spray ionization mass spectrometry (PSI-MS)-based nontargeted metabolomics was employed to rapidly classify prostate health conditions using urine liquid biopsies. Urine samples from healthy volunteers and patients with benign prostatic hyperplasia (BPH), prostatitis, and prostate cancer (PC) were analyzed to identify unique metabolic signatures. After filtering, 1050 distinct metabolic features (70-800 Da) were detected across all samples. Multivariate analysis including partial least-squares discriminant analysis (PLS-DA) and hierarchical clustering revealed distinct metabolic profiles among different prostate health conditions, with PC samples forming separate clusters from benign conditions and healthy controls. Machine learning algorithms such as random forest analysis identified top discriminatory metabolic features, while significance analysis of microarrays (SAM) confirmed meaningful biological changes above analytical noise. Receiver Operating Characteristic (ROC) analysis demonstrated high predictive accuracy: BPH (AUC = 0.90), prostatitis (AUC = 0.75), and PC (AUC = 0.98). Condition-specific metabolic vulnerabilities were identified, including stress-response metabolites in BPH, infection-related compounds in prostatitis, and extensive metabolic reprogramming in PC. The study identified 37 significant features with varying degrees of significance among conditions. These findings demonstrate PSI-MS-based metabolomics as a promising rapid, noninvasive diagnostic tool for prostate health assessment, providing molecular insights into disease-specific metabolic alterations suitable for clinical applications.