Accurate and noninvasive prostate cancer detection using plasma-derived extracellular vesicle RNA

Prostate cancer (PCa) is the most commonly diagnosed noncutaneous malignancy in men and a leading cause of cancer-related death worldwide. Its clinical heterogeneity, ranging from indolent to aggressive disease, presents major diagnostic and therapeutic challenges. However, prostate-specific antigen (PSA), the current standard biomarker, lacks sufficient sensitivity and specificity, resulting in both overdiagnosis and missed cases. Therefore, more accurate noninvasive biomarkers are urgently needed. In this study, we isolated plasma-derived extracellular vesicles (EVs) using a wheat germ agglutinin (WGA)-conjugated magnetic bead method and identified differentially expressed transcripts between PCa patient and healthy control (HC) EVs. Candidate biomarkers were selected through weighted gene co-expression network analysis (WGCNA) and Random Forest modeling, followed by validation using RT-qPCR in independent cohorts including PCa, benign prostatic hyperplasia (BPH), and HC samples. We identified three EV-derived RNAs (NM_024955, NR_047469, and NR_002564), which were significantly dysregulated in PCa compared with both HC and BPH. Combined analysis demonstrated high diagnostic performance, outperforming PSA and notably NM_024955 expression showed a strong correlation, indicating potential relevance to disease severity. These findings suggest that lectin-based EV isolation coupled with RNA profiling provides a robust and scalable platform for PCa diagnosis, offering promising noninvasive biomarkers to complement or improve current PSA-based screening.