Performance of a clinically validated urine exosome gene expression test to predict high grade prostate cancer in men with a prior negative biopsy.

Abstract

119 Background: The over-diagnosis of low grade, indolent prostate cancer (PCa) initiates significant patient anxiety regarding treatment options and health risk. To address this clinical need, we developed and subsequently validated a noninvasive, easy to administer, supportive diagnostic test (ExoDx Prostate(IntelliScore), EPI) to guide the initial prostate biopsy decision process for men with PSA 2-10ng/mL. The EPI test is a urine exosome gene expression assay to discriminate high-grade (GS 7 PCa) from low-grade (GS 6) and benign disease, thereby potentially reducing the number of unnecessary biopsies. As many men have had prior negative/ benign biopsies, we sought to understand the performance of EPI in this selected population. Methods: Utilizing men with prior negative biopsies from two clinical trials we evaluated performance of the EPI test with respect to subsequent biopsy outcomes. We used the previously validated and alternative EPI cut-points of 15.6 and 20, respectively, to assess performance with AUC, sensitivity, and NPV. Results: A total of N = 286 patients with a prior negative biopsy; mean age 65 years, mean PSA 6.2 ng/mL, 22% positive family history, 15% African American; 32% positive biopsy rate: 20% GS6 (ISUP1) and 12% > / = GS7 PCa (ISUP≥2). AUC 0.68 vs. AUC 0.59 for standard of care (i.e. PSA, age, race, family history) and AUC 0.51 for PSA alone. By comparison the EPI initial biopsy AUC was 0.71. Using the previously validated EPI cut-points of 15.6 (or alternative 20) yielded NPVs of 92% and 95%, and number of avoided biopsies of 23% and 32%, respectively, and sensitivity of 85.7% for both cut-points. Notably, EPI in the initial biopsy setting had a sensitivity of 92 vs. 87% and an NPV of 91 vs. 90%, for the 15.6 and 20 cut-points, respectively. Conclusions: The EPI test is a noninvasive, first-catch, non-DRE gene expression array that accurately discriminates low-grade and benign biopsy outcomes from high-grade prostate cancer. The test performs well in both the initial and prior negative biopsy patients and has the potential to reduce the overall number of unnecessary biopsies in both settings.