AUA 2019: A Genomic Classifier Shows Improved Prediction of Oncologic Outcomes in African-American Men Treated with Radical Prostatectomy

Chicago, IL ( Adjuvant radiotherapy has been proven, in three randomized controlled trials, to improve biochemical recurrence for men with high risk features following radical prostatectomy. However, it is not widely used, in part due to concerns regarding overtreatment and the toxicity of radiation. While randomized data comparing adjuvant and salvage radiotherapy strategies is currently maturing, there remains a clinical need for tools to better risk stratify patients following radical prostatectomy. One such tool is genomic testing: one genomic test which may be useful in this context is the Decipher Genomic Classifier1. While this genomic classifier has been approved and adopted, relatively little is known about how it performs among African American men. In a podium presentation at the American Urologic Association Annual Meetin, Ms. Howard, on behalf of Dr. Freedland and colleagues, presented the results of their study comparing the prognostic ability of Decipher GC and the CAPRA-S clinical risk score among African American and non-African American men who underwent radical prostatectomy at the Durham Veterans Affairs Medical Center.

The authors identified 557 patients who underwent radical prostatectomy at the Veterans Affairs Medical Center Durham between 1989 and 2013 and were deemed at clinical high risk for recurrence on the basis of stage pT3a, positive surgical margins, seminal vesicle invasion (stage pT3b), or receipt of post-operative radiation therapy. The authors compared the performance of the Decipher GS with the CAPRA-S clinical risk score using cox proportional hazards models and survival c-index for assessing the risk of metastasis and prostate cancer-specific mortality.

Among the 557 included patients, 306 (55%) were African American. 10.4% of patients were classified as low-risk on the basis of the CAPRA-S clinical risk score while 50.4% of patients were similarly classified on the basis of the Decipher GC. Over a median follow-up of 9 years, 40 patients developed metastasis and 18 patients died of prostate cancer. Following multivariable adjustment, both the CAPRA-S clinical risk score (HR 1.27, 95% CI 1.01-1.58) and Decipher GC (HR 1.30, 95% CI 1.01-1.69) were significant predictors of metastasis among non-African American men. However, only the Decipher GC was a significant predictor of metastasis among African American men (HR 1.70, 95% CI 1.31-2.20). Assessing PCSM, Decipher GC, but not CAPRA-S clinical risk score, was a significant predictor of prostate cancer mortality among both non-African American men (HR 1.54, 9%% CI 1.01-2.53) and African American men (HR 1.65, 95% CI 1.19-2.42). The survival c-index for the Decipher GC was higher among African American men (metastasis: 0.84, 95% CI 0.76-0.90; PCSM: 0.82, 95% CI 0.61-0.93) than among non-African American men (metastasis: 0.70, 95% CI 0.63-0.80; PCSM: 0.73, 95% CI 0.63-0.84).

These results confirm numerous previous studies which have demonstrated the validity of the Decipher GC for predicting metastasis and PCSM among non-African American men following radical prostatectomy. They further suggest that, although the Decipher GC is relatively poorly studied among African American men, it may actually perform better among these men than among non-African American men.

Presented by: Lauren Howard, Biostatistician at Duke University Medical Center, Durham, North Carolina
Co-authors: Stephen Freedland, Marguerite du Plessis, Jingbin Zhang, Amanda De Hoedt, Elai Davicioni

Written by: Christopher J.D. Wallis, Urology Resident, University of Toronto, @WallisCJD on Twitter at the American Urological Association's 2019 Annual Meeting (AUA 2019), May 3 – 6, 2019 in Chicago, Illinois

  1. Lobo JM, Trifiletti DM, Sturz VN, et al. Cost-effectiveness of the Decipher Genomic Classifier to Guide Individualized Decisions for Early Radiation Therapy After Prostatectomy for Prostate Cancer. Clin Genitourin Cancer 2017; 15(3): e299-e309.