Presentation Authors: Karim Touijer*, Daniel Sjoberg, Nicole Benfante, James Eastham, Vincent Laudone, Ehdaie Behfar, Peter Scardino, Andrew Vickers, New York, NY
Introduction: Pelvic lymph node dissection (PLND) in prostate cancer remains controversial . Although it represents the most reliable procedure for lymph node staging, a therapeutic benefit remains unproven and a consensus has not been reached as to what anatomical extent the PLND should be performed.
Methods: Between October 2011 and March 2017, 1573 men undergoing radical prostatectomy at Memorial Sloan Kettering Cancer Center consented to the trial. After 73 patients were randomized, the study design changed to cluster randomization with crossover. Each surgeon was randomized to use a limited (external iliac nodal packet) vs. extended (external iliac, obturator fossa and hypogastric nodal packets) PLND template for a 3-month period. At the end of the 3-month period, surgeons were once again randomized. The primary endpoint was time to biochemical recurrence defined as a PSA of >0.1 ng/mL followed by a confirmatory PSA rise. Cox regression was used to compare groups, using the sandwich estimator to allow for correlation within surgeon.
Results: A total of 1480 patients were registered: 757 randomized to the extended and 723 to the limited template. The median number of nodes retrieved was 12 (IQR 8, 17) and 14 (IQR 10, 20) The rate of positive nodes was 11.2% vs. 13.6% (risk difference 2.4%, 95% C.I. -1.0%, 5.8%, p=0.2) for the limited and extended PLND, respectively. There were no grade IV or V complications. Rates for grade II and III complications were almost identical (12% in limited vs. 11% in extended). There was no significant difference in biochemical recurrence between groups (hazard ratio 0.99; 95% CI 0.83, 1.17; p=0.9).
Conclusions: The extended pelvic lymph node dissection did not improve freedom from biochemical recurrence over a limited node dissection in men with clinically localized prostate cancer. However, there was only a small difference in nodal count and the rate of positive nodes between the two templates.