Background: The accuracy of the extracellular vesicle-fingerprint score (EV-FPS) test to predict clinically significant prostate cancer (PCa; Gleason grade (GG) ≥ 3) from indolent disease (GG ≤ 2) and avoid unnecessary prostate biopsies was determined at the point of prostate biopsy decision. Methods: Clinical data, health information, and blood samples were collected from a prospective validation cohort of 415 men, without prior PCa diagnosis, referred to urology clinics for prostate biopsy or transurethral prostate surgery (June 2014-Dec 2016). The patient’s EV-FPS risk score was calculated by combining machine learning model-analyzed microflow cytometry data from EV biomarkers with logistic regression-analyzed patient-centric clinical features. The plasma-derived EV biomarkers were prostate-specific membrane antigen, polysialic acid and ghrelin-growth hormone receptor. The patient clinical features were; age, ethnicity, PCa family history, PSA levels, abnormal digital rectal examination (DRE) and prior negative prostate biopsy. Together, the biomarkers and clinical features provided specificity for clinically significant PCa. Results: The EV-FPS test identified clinically significant PCa patients with high accuracy (0.81 area under curve) at 95% sensitivity and 97% negative predictive value. Using a 7.85% probability cut-off after test validation; 95% of the patients with GG ≥ 3 would have been found before biopsy, 35% biopsies would have been avoided and diagnosis of GG ≥ 3 PCa would have been missed in only 5% of the cohort. Conclusions: This minimally invasive EV-FPS test accurately predicted clinically significant PCa in men with high EV-FPS risk scores, high PSA level and/or abnormal DRE. Therefore, men with low EV-FPS risk scores could potentially avoid unnecessary prostate biopsies. Clinical care cut-offs to calculate the number of biopsies that could have been avoided, and the percentage of GG ≥ 1 to GG ≥ 3 PCa that could have had a delayed diagnosis.