Cancer-specific mortality differences in specimen-confined radical prostatectomy patients according to lymph node invasion

Abstract

Purpose: To test cancer-specific mortality (CSM) differences in specimen-confined (pT2) prostate cancer (PCa) at radical prostatectomy (RP) with lymph node dissection (LND) according to lymph node invasion (LNI). Methods: RP + LND pT2 PCa patients were identified (surveillance, epidemiology, and end results 2010-2015). CSM-FS rates were tested in Kaplan-Meier plots and multivariable Cox-regression (MCR) models. Sensitivity analyses respectively addressing patients with 6 or more lymph nodes analyzed and pT2 pN1 patients were performed. Results: Overall, 32,258 patients with pT2 PCa at RP + LND were identified. Of these, 448 (1.4%) patients harbored LNI. Five-year CSM-free estimates were 99.6% for pN0 vs. 96.4% for pN1 (P < .001). In MCR models, pN1 (HR: 3.4, P < .001) independently predicted higher CSM. In sensitivity analyses addressing patients with 6 or more lymph nodes analyzed (n = 15,437), 328 (2.1%) pN1 patients were identified. In this subgroup, 5-year CSM-free estimates were 99.6% for pN0 vs. 96.3% for pN1 (P < .001) and, in MCR models, pN1 independently predicted higher CSM (HR: 4.4, P < .001). In sensitivity analyses addressing pT2 pN1 patients, 5-year CSM-free estimates were 99.3, 100 and 84.8% for ISUP GG 1-3 vs. 4 vs. 5, respectively (P < .001). Conclusions: In patients with pT2 PCa a small proportion harbor LNI (1.4%-2.1%). In such patients, CSM rate is higher (HR 3.4-4.4, P < .001). This higher CSM risk seems to virtually exclusively apply to ISUP GG5 patients (84.8% 5-year CSM-free rate).