Performance characteristics of ExoDx urinary biomarker: A single-institution analysis.

e17090 Background: Prostate cancer screening relies on prostate-specific antigen, which may lead to unnecessary prostate biopsies. The ExoDx Prostate (EPI) test, a urinary exosome assay, has been developed for the detection of clinically significant (Gleason Grade Group ≥2) prostate cancer (csPCa) to improve specificity. This study analyzes the performance characteristics of EPI compared to/in concert with a multiparametric MRI (mpMRI)-based pathway at a single academic institution. Methods: Patients who underwent EPI testing between 10/2019 to 5/2023 were reviewed for EPI score, demographics, MRI characteristics, and biopsy results. Patients were categorized by EPI score (high risk ≥ 15.6 vs. low risk < 15.6). Outcome variables of biopsy and MRI were compared between EPI risk groups using Chi square and Mann Whitney-U test. Spearman correlation was used to estimate the correlation between EPI score and PI-RADS and Gleason score. Receiver operator curve (ROC) analysis assessed the performance of EPI and PI-RADS score for prediction of csPCa. An optimal cutoff value was determined for EPI using ROC, and the sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were assessed. Statistical comparisons were two-sided with a p-value <0.05 being statistically significant. Results: Of n=173 patients with EPI scores, 28.9% were low risk and 71.1% were high risk. EPI had weak correlation with PI-RADS score on prostate MRI (r = 0.30, p <0.001). An optimal EPI score cutoff for the detection of csPCa of 26.43 was determined (sensitivity 0.720, specificity 0.537, PPV 0.367, NPV 0.800). On ROC analysis, PI-RADS with score of ≥4 had better predictive performance for detecting csPCa with an Area Under the Curve (AUC) of 74.6% (sensitivity: 0.696, specificity: 0.721, PPV: 0.485, NPV: 0.862), versus an AUC of 63.1% with EPI testing. By utilizing a criteria of PI-RADS score ≥4 and EPI ≥ 26.43, we found improved PPV and specificity over either alone (PPV: 0.556 vs 0.485 vs 0.367, respectively) (specificity: 0.873 vs 0.721 vs 0.537, respectively). In addition, we found that with utilization of either PI-RADS score ≥4 or EPI ≥ 26.43, there was improved NPV and sensitivity (NPV: 0.962 vs 0.862 vs 0.800, respectively) (sensitivity: 0.957 vs 0.696 vs 0.720, respectively). Conclusions: Our study supports EPI testing as a way to assess risk for csPCa, ideally in combination with mpMRI, and also prompts reevaluation of 15.6 as an optimal cutoff EPI score.