Long-term oncological outcomes of pelvic lymph node dissection during radical prostatectomy

Abstract

Purpose: Lymph node dissection during radical prostatectomy (RP) remains under debate due to an unclear impact on oncological outcomes. Here, we investigate long-term oncological events after RP +/- pelvic lymph node dissection (PLND).

Methods: All patients who had an RP from 2006 to 2021 (N=15,515) in the Danish Prostate Registry were included. Cumulative incidence of biochemical failure (BF), subsequent treatment, and prostate cancer-specific death by the Aalen Johansen estimator from the time of RP were compared for men without lymph nodes removed (pNx), without positive lymph nodes (pN0), and with lymph node invasion (pN1). Area under the curve (AUC) of the receiver operating characteristic was used to determine the added value of PLND for outcome discrimination. Results: The 15-year incidences of BF were pNx = 22% (95% confidence interval [95CI]: 20-23), pN0 = 37% (95CI: 35-39), and pN1 = 70% (95CI: 65-76). Fifteen-year incidences of subsequent treatment were 19% (95CI: 18-20), 31% (95CI: 29-33), and 84% (95CI: 79-89) for men with pNx, pN0, and pN1. Fifteen-year risk of prostate cancer-specific death was 2.1% (95CI: 1.6-2.6), 5.5% (95CI: 4.6-6.5), and 25% (95CI: 18-32) for men with pNx, pN0, and pN1. The AUC for the 10-year BF increased from 74% (95CI: 72-77) to 76% (95CI: 73-78) with the addition of pN0/pN1 in the model. The main limitation was the inability to differentiate the number of malignant nodes removed.

Conclusion: Men with positive lymph nodes after PLND faced a high risk of BF, subsequent treatment, and prostate cancer-specific death. As the addition of PLND did not substantially improve the discriminative ability, the added benefit of PLND is likely limited. The side effects of PLND should be considered in relation to its possible advantages.