To evaluate the seminal biomarker signature (sTWEAK, exomiR-221-3p, and exomiR-222-3p) in several prostate cancer (PCa) settings: diagnosis, tumor aggressiveness prognosis (International Society of Urological Pathology [ISUP] grade in radical prostatectomy specimens and biochemical recurrence [BCR]), and identifying biopsy upgrading in confirmatory biopsies during active surveillance (AS).

Semen samples were collected from 136 PCa patients classified by prostatectomy specimen ISUP grades and classified as low-risk (ISUP I-II, n = 85) and high-risk (ISUP ≥ III, n = 51), along with 20 healthy controls, 35 patients with negative biopsies, and 51 low-risk PCa patients under AS. The relative expression levels of exomiR-221-3p and exomiR-222-3p in semen exovesicles were measured. ELISA was used to quantify sTWEAK levels. Statistical analyses, including multivariable binary logistic, Cox multivariate regression analysis, Survival Classification Regression Tree models, and ROC, among others, were used to evaluate the predictive value of the biomarkers.

A model combining sTWEAK and exomiR-221-3p predicted high-risk PCa (AUC = 0.937), showing a linear biomarker-probability relationship but no clear cutoff. A model comprising PSA density and sTWEAK could be used for predicting BCR. sTWEAK alone distinguished PCa cases (AUC = 0.897). In AS patients, sTWEAK levels plus the percentage of positive biopsy cores best predicted upgrading (AUC = 0.824).

Incorporating sTWEAK and exomiR-221-3p, along with clinical parameters, offers a promising diagnostic panel for PCa and serves as a useful predictive tool for AS upgrading and BCR prognosis.