Robotic vs. laparoscopic resection for rectal cancer with neoadjuvant chemoradiotherapy: perioperative outcomes and survival.

Abstract

Background: Although the safety and effectiveness of robot-assisted surgery for rectal cancer have been demonstrated, its peri-operative safety and long-term survival benefit in patients who have received neoadjuvant chemoradiotherapy (NCRT) remain unclear.

Method: A retrospective cohort of 463 consecutive rectal-cancer patients who underwent either robot-assisted anterior resection or laparoscopic anterior resection after NCRT at Sun Yat-sen University Cancer Center from June 2016 to August 2023 was analyzed. Propensity-score matching (PSM) was applied to balance baseline variables that could affect surgical outcomes and survival. Peri-operative parameters, complication rates, and pathological findings were compared, and 5-year overall survival (OS) and disease-free survival (DFS) were calculated.

Result: A total of 121 patients undergoing robotic-assisted surgery and 342 patients undergoing laparoscopic surgery were included in the study. Operative time was longer in the robot-assisted group both before and after matching. Before matching, the robot-assisted group showed a lower rate of ileostomy (65.3% vs 76.6%, P = 0.021), an advantage that disappeared after PSM (65.3% vs 75.2%, P = 0.063). In the matched cohort, the 5-year DFS was 89.8% for robot-assisted versus 82.3% for laparoscopic (P = 0.130), whereas the 5-year OS was higher in the robot-assisted group than in the laparoscopic group (96.8% vs 86.8%; P = 0.019). In the primary preoperative covariate-adjusted Cox model after PSM, the association between robot-assisted surgery and OS was attenuated and did not reach conventional statistical significance (HR = 0.231, 95% CI 0.052-1.015; P = 0.052).

Conclusions: Robot-assisted and laparoscopic surgery showed comparable perioperative outcomes and DFS after NCRT for rectal cancer. Although an apparent OS difference was observed in the matched cohort, this finding was attenuated in the primary preoperative covariate-adjusted Cox model and should be interpreted cautiously given the potential for residual confounding.