Decision Analysis of Pelvic Lymph Node Dissection During Radical Prostatectomy

There is controversy about the decision of whether to perform a pelvic lymph node dissection (PLND) during radical prostatectomy for prostate cancer. While a recent randomized trial reported a reduced risk of metastasis for extended compared with limited PLND, some guidelines do not recommend PLND, at least partly on the basis that it raises the risk of complications such as lymphocele. We conducted a decision analysis of PLND. Our aim was to put varying numerical estimates on benefit, harm, and uncertainty to determine whether, and under what conditions, PLND would do more good than harm.

Our approach was to start first with a simple decision tree for PLND vs no PLND during radical prostatectomy and then determine whether added complexity would be of benefit. We started by using inputs that were unfavorable to PLND-for instance, using an extremely high outlying rate of lymphocele and having no difference in metastasis rates beyond 10 years-aiming to vary these in sensitivity analyses.

Despite starting with unfavorable inputs for PLND, the expected utility of PLND was higher than that for no PLND across a broad range of scenarios, including giving a low subjective probability that PLND was of benefit, high risk of PLND complications, and PLND's reduction in locoregional metastases being considered irrelevant. PLND was also favored in patients with prostate-specific membrane antigen positron emission tomography-negative disease, a finding driven by the imperfect sensitivity of prostate-specific membrane antigen positron emission tomography. The key limitation is that the findings do not apply to patients who have only a trivial risk of metastasis, such as patients with grade group 1 disease.

PLND should be the standard of care for patients undergoing radical prostatectomy for grade group 2 or higher disease.