Long-Term Mortality in Patients with Positive Lymph Nodes at the Time of Radical Prostatectomy

Abstract

Background: The aim of this study was to determine prognostic factors and to provide long-term mortality data in patients with positive lymph nodes at the time of radical prostatectomy in a sample with long-term follow-up. Methods: A total of 527 patients with complete data sets treated in the years 1992–2014 were studied. The median follow-up was 7.2 years. The median number of removed lymph nodes was 15. Age, year of surgery, Gleason score, local tumor stage, prostate-specific antigen level, lymph node density, lymph node count and the number of positive lymph nodes were included in multivariable competing risk analyses with prostate cancer mortality as endpoint. Results: After 20 years, 28% of patients (95% CI 20–36%) died from non-prostate cancer (competing) causes, whereas 29% (95% CI 23–36%) died from prostate cancer. Only lymph node density (stratified by the median of 11.1%; hazard ratio [HR] 1.66, 95% CI 1.04–2.64, p = 0.0340) and Gleason score (8–10 vs. <8: HR 5.97, 95% CI 3.18–11.23, p < 0.0001) were independent predictors of prostate cancer mortality. Patients with a Gleason score <8 and a lymph node density < median had a 20-year prostate cancer mortality of only 5% (95% CI 0–10%), whereas this rate in patients with Gleason score 8–10 and a lymph node density ≥ median was 44% (95% CI 32–56%), p < 0.0001. Conclusions: Mortality in patients with positive lymph nodes was determined by tumor aggressiveness and the relative extent of spread; neither the year of surgery nor the number of removed lymph nodes was associated with outcome. Patients with a lymph node density of <11.1% and a Gleason score <8 had an excellent long-term outcome.