Abstract
Background: Prostate cancer (PCa) is a major health concern, and current PSA screening is limited by low specificity and the risk of overdiagnosis. Extracellular vesicle (EV)-derived RNA biomarkers offer a promising non-invasive alternative for early detection. Methods: We utilized a tethered cationic lipoplex nanoparticle (TCLN) biochip for amplification-free EV RNA detection at the single-vesicle level. Eight candidate RNAs (four miRNAs, three mRNAs, and one lncRNA) were profiled in serum and urine samples from PCa patients, benign prostatic hyperplasia (BPH) patients, and healthy controls (HC). Diagnostic and risk stratification performance was evaluated in discovery and validation cohorts, with qRT-PCR used for validation. Results: TCLN reliably detected the candidate RNA biomarkers from PCa cell lines and clinical samples, with strong concordance to qRT-PCR. Serum EV RNAs (miR-141, miR-375, Let-7c) and urine EV RNAs (miR-141, miR-375, PCA3 lncRNA, T1-E2) effectively distinguished PCa patients from controls. Combined EV RNA biomarkers in serum and urine achieved diagnostic area-under-the-curve (AUCs) of 0.824 and 0.741, respectively, surpassing those of prostate-specific antigen (PSA) alone. Serum miR-141, miR-375, and urine miR-141, miR-375, and PCA3 lncRNA, also showed remarkable correlations with PCa Gleason score (GS), tumor stage, and metastatic status. Conclusion: The TCLN biochip enables sensitive, amplification-free detection of EV RNA biomarkers from serum and urine. Key markers such as miR-141, miR-375, and PCA3 showed strong diagnostic and risk stratification value in PCa. This non-invasive approach holds promise for improving early detection and clinical risk assessment.