Open Access

Robert Benedikt Höh, Christoph Würnschimmel, Rocco Simone Flammia, Benedikt Horlemann, Gabriele Sorce, Francesco Chierigo, Zhe Tian, Fred Saad, Markus Graefen, Michele Gallucci, Alberto Briganti, Carlo Terrone, Shahrokh F. Shariat, Luis Kluth, Philipp Mandel, Felix Chun, Pierre I. Karakiewicz

• Background: To test for differences in cancer-specific mortality (CSM) rates between radical prostatectomy (RP) vs external beam radiotherapy (EBRT) in National Comprehensive Cancer Network (NCCN) high-risk African American patients, as well as Johns Hopkins University (JHU) high-risk and very high-risk patients. Materials and methods: Within the Surveillance, Epidemiology, and End Results database (2010–2016), we identified 4165 NCCN high-risk patients, of whom 1944 (46.7%) and 2221 (53.3%) patients qualified for JHU high-risk or very high-risk definitions. Of all 4165 patients, 1390 (33.5%) were treated with RP versus 2775 (66.6%) with EBRT. Cumulative incidence plots and competing risks regression models addressed CSM before and after 1:1 propensity score matching between RP and EBRT NCCN high-risk patients. Subsequently, analyses were repeated separately in JHU high-risk and very high-risk subgroups. Finally, all analyses were repeated after landmark analyses were applied. Results: In the NCCN high-risk cohort, 5-year CSM rates for RP versus EBRT were 2.4 versus 5.2%, yielding a multivariable hazard ratio of 0.50 (95% confidence interval [CI] 0.30–0.84, p = 0.009) favoring RP. In JHU very high-risk patients 5-year CSM rates for RP versus EBRT were 3.7 versus 8.4%, respectively, yielding a multivariable hazard ratio of 0.51 (95% CI: 0.28–0.95, p = 0.03) favoring RP. Conversely, in JHU high-risk patients, no significant CSM difference was recorded between RP vs EBRT (5-year CSM rates: 1.3 vs 1.3%; multivariable hazard ratio: 0.55, 95% CI: 0.16–1.90, p = 0.3). Observations were confirmed in propensity score-matched and landmark analyses adjusted cohorts. Conclusions: In JHU very high-risk African American patients, RP may hold a CSM advantage over EBRT, but not in JHU high-risk African American patients.