This is the first randomized surgical study in bladder cancer randomizing patients with pT2 or greater disease and without neoadjuvant chemotherapy to either a limited lymph node dissection or an extended lymph node dissection. The limited node dissection consisted of external iliac, internal iliac, and obturator lymph nodes while the extended lymph node dissection included these areas in addition to the deep obturator, presacral, interaortocaval, paracaval, and periaortic lymph nodes up to the inferior mesenteric artery. Patients with organ-confined disease and negative nodes were observed whereas patients with non-organ-confined disease and/or positive lymph nodes were offered adjuvant chemotherapy. The primary endpoint was 5-year recurrence-free survival (RFS). Secondary endpoints include complication rate, influence of adjuvant chemotherapy, disease-specific survival, overall survival, and local recurrences and distant metastases.
The proximal extent of a lymph node dissection (LND) at the time of radical cystectomy has always been an issue of controversy. Some reports suggest that a dissection to the common iliac bifurcation or to the ureter crossing over the iliac vessels is adequate whereas others suggest that more proximal dissection to the inferior mesenteric artery may affect survival. Furthermore, removal of the presacral lymph nodes and the deep obturator lymph nodes are not routinely done by all surgeons.
Ultimately 437 patients were randomized but 64 patients dropped out due to inclusion/exclusion violations leaving 190 patients in the limited LND group and 183 in the extended LND group by an intention to treat analysis. Median follow-up was about 33 months. The lymph node positive rate was 27.4% for the limited LND group and 20.2% for the extended LND group. Overall, the positive node rate was 23.9% among patients and approximately 14% of patients received adjuvant chemotherapy. The median lymph node removal was 19 for limited LND and 31 for extended LND, p < 0.01. The 5-year RFS was 61.5% for the limited LND and 67% for the extended LND, p = 0.34. Secondary endpoints were also not statistically significant including 5-year disease-specific survival, 5-year OS, and complication rate. On post-hoc analysis of organ-confined tumors (pT2), there was a survival benefit in 5-year RFS (62.5% vs. 85% , p = 0.01) and 5-year OS (59.1% vs. 79.8% , p = 0.04) favoring the extended LND group. In 89 patients with positive lymph nodes, adjuvant chemotherapy improved median RFS (35.9 months vs. 8.8 months, p = 0.0002).