335 - A Validation Study on the Impact of Decipher® Testing on Treatment Recommendations in African American and Non-African American Men with Prostate Cancer (VANDAAM STUDY)

K. Yamoah¹, P. Trivedi², S. Awasthi³, G. D. Grass¹, J. F. Torres-Roca¹, P. A. S. Johnstone², J. Dhillon², J. Park², E. Davicioni⁴, A. Hakansson⁵, Y. Liu⁶, A. Fink², E. Katende², A. DeRenzis⁷, R. Smith⁷, M. Poch², R. Li², B. Manley², D. C. Fernandez⁸, K. Gage⁹, G. Lu-Yao¹⁰, Y. Kim¹¹, E. Katsoulakis¹², A. Leone¹³, N. Vapiwala¹⁴, C. Deville Jr¹⁵, T. R. Rebbeck¹⁶, A. P. Dicker¹⁷, W. K. Kelly¹⁸, R. Burri¹³, C. E. Ercole¹⁹, and J. Pow-Sang²; ¹H. Lee Moffitt Cancer Center and Research Institute, Department of Radiation Oncology, Tampa, FL, ²H. Lee Moffitt Cancer Center & Research Institute, Tampa, FL, ³H. Lee Moffitt Cancer Center and Research Institute, Department of Cancer Epidemiology, Tampa, FL, ⁴Veracyte Inc., San Diego, CA, ⁵Veracyte Inc.,, South San Francisco, CA, ⁶Veracyte, San Diego, CA, ⁷Moffitt Cancer Center, Tampa, FL, ⁸Department of Radiation Oncology, H. Lee Moffitt Cancer Center and Research Institute, Tampa, FL, ⁹H. Lee Moffitt Cancer Center and Research Institute, Tampa, FL, ¹⁰Sidney Kimmel Cancer Center at Jefferson, Philadelphia, PA, ¹¹H. Lee Moffitt Cancer Center and Research Institute, Department of Bioinformatics and Biostatistics, Tampa, FL, ¹²James A. Haley Veterans Hospital, Tampa, FL, ¹³Bay Pines VA Healthcare System, Tampa, FL, ¹⁴Department of Radiation Oncology, University of Pennsylvania, Philadelphia, PA, ¹⁵Department of Radiation Oncology and Molecular Radiation Sciences, Johns Hopkins University School of Medicine, Baltimore, MD, ¹⁶Dana Farber Cancer Institute, Boston, MA, ¹⁷Department of Radiation Oncology, Sidney Kimmel Medical College at Thomas Jefferson University, Philadelphia, PA, ¹⁸Thomas Jefferson University, Philadelphia, PA, ¹⁹James A. Haley Veterans Hospital, Tampa, FL

Purpose/Objective(s): In the era of precision oncology, genomic risk classifiers (GC), such as the Decipher score, have enhanced the personalization of care for patients with localized prostate cancer (PCa). Nonetheless, few prospective studies have prioritized the recruitment of African American men (AAM) to provide Level I evidence to support the generalizability of these GCs in AAM who endures the disproportionate burden of PCa. Here, we report the first prospective validation of the Decipher GC in predicting biochemical recurrence (BCR) in AAM; (VANDAAM; - NCT02723734). Materials/

Methods: Between 2018 and 2021, this multicenter prospective validation study recruited 243 patients (125-AAM, 118 non-African American -NAAM) with low- or intermediate-risk PCa and received treatment for their disease. Patients were recruited on a 1:1 (AAM:NAAM) basis and matched on the following clinical factors: age at diagnosis, Gleason score, stage, and percentage of positive cores from a baseline prostate biopsy. Patients who elected active surveillance were ineligible for participation. Decipher GC testing was ordered for all patients using their biopsy and/or radical prostatectomy (RP) tumor tissue. The primary outcome was to determine whether GC can predict 2-year BCR rates, a surrogate for disease aggressiveness, following standard treatment options. The secondary outcome evaluated the concordance between biopsy- and RP- derived GC risk scores for treatment recommendations.

Results: The final analytical cohort included 207 evaluable cases (104 AAM, 103 NAAM) with both genomic and clinical outcome data. Decipher GC scores were stratified into three risk categories: low (<0.45), intermediate (= 0.45 – < 0.6) and high (>0.6). High-risk GC was associated with a 10-fold increased risk of BCR as compared to low-risk counterparts (HR = 10.37, 95% CI, 1.94-55.35, p=0.006). There was no significant interaction in predictive performance of GC across either race or treatment (RP vs. RT) groups. In the surgical cohort (n=74), biopsy and RP-derived GC scores exhibited a 77% concordance rate, defined as no reclassification in GC based risk categories. Among cases with discordant GC scores (23%), anterior-located tumors that were classified as low-risk on biopsy GC had a 44-fold increased likelihood of being reclassified as high-risk after surgery (OR = 44, 95% CI, 2.19-882.65, p = 0.01).

Conclusion: Prior to VANDAAM, GC had not been formally validated in a prospective clinical study of its prognostic ability for AAM. At the 2-year endpoint, VANDAAM results indicate that GC equally predicted early BCR in both AAM and NAAM. Therefore, GC could be integrated into clinical practice to improve risk stratification to aid in appropriate treatment recommendations for AAM patients with early stage PCa.