MP45-04: SelectMDx Urine Test Diagnose Men with High-Grade Prostate Cancer

Whitney Stanton, Aurora, CO, E David Crawford*, La Jolla, CA, Paul Arangua, Joshua Romero, Francisco G. La Rosa, Adrie van Bokhoven, M Scott Lucia, Paul Maroni, Aurora, CO, Wendy Poage, Centennial, CO, Jack Schalken, Nijmegen, Netherlands, Fernando Kim, Denver, CO, Jack Groskopf, Irvine, CA, Priya Werahera, Aurora, CO

Introduction: Some men are candidates for watchful-waiting (WW) or active surveillance (AS) following TRUS biopsy findings of low-grade prostate cancer (PCa) or benign disease. However, there is an unmet clinical need to reliably rule out high-grade PCa (HGPCa) in these men. The SelectMDx™ test measures urinary mRNA levels of the homeobox C6 (HOXC6) and distal-less homeobox 1 (DLX1) biomarkers and has been clinically validated for detection of HGPCa upon subsequent TRUS biopsy. We investigated the utility of the SelectMDx test to identify HGPCa in a cohort of men who underwent transperineal mapping biopsy (TPMB) after TRUS biopsy by comparing the SelectMDx results to histopathological findings of TPMB.

Methods: In this retrospective study, 105 patients opted for TPMB to confirm the histopathological findings of their initial TRUS biopsies. Post-DRE, first-void urine specimens were collected from each patient prior to TPMB. SelectMDx testing was performed at MDxHealth, blinded with respect to TPMB outcome. Histopathology of each TPMB was independently read by a genitourinary pathologist. HGPCa was defined as ISUP Grade Group (GG) =2. SelectMDx performance characteristics for detection of HGPCa were determined by comparison to TPMB histopathology data.

Results: 29 patients on 5ARI and 8 patients failed SelectMDx test were excluded. The remaining 68 patients were included with median age of 63 (IQR 57-68) years and median PSA 5.1 (3.0-7.6) ng/mL. TRUS biopsy diagnosed 10 (15%) with benign disease, 37 (54%) patients with GG1, and 21 (31%) with =GG2. TPMB identified 12 (18%) patients with benign pathology, 26 (38%) with GG1 PCa, and 30 (44%) with HGPCa (23 GG2, 3 GG3 and 4 GG4). For detection of HGPCa, the SelectMDx test yielded sensitivity of 97% (95% C.I. 83-100%), specificity 37% (22-54%), positive predictive value 55% and negative predictive value (NPV) 93%. The single HGPCa identified in a SelectMDx negative patient was GG2. In a logistic regression analysis including age and PSA, SelectMDx was the only significant predictor of HGPCa at TPMB (ß = 1.09 (p=0.0037), odds ratio 2.99, AUC = 0.738).

Conclusions: The SelectMDx urine test demonstrated high sensitivity and NPV for discriminating between patients with HGPCa vs. patients with GG1 or benign disease at biopsy. These results support the use of SelectMDx to help identify men for WW or AS following TRUS biopsy. Study results also support the use of non-invasive tests to help identify men with HGPCa who may benefit from prostate biopsy, thereby reducing over-detection of indolent disease.

Source of Funding: Bingham Foundation, Schramm Foundation, Prostate Cancer Biorepository, and MDx Health