The RECOURSE Study: Long-term Oncologic Outcomes Associated With Robotically Assisted Minimally Invasive Procedures for Endometrial, Cervical, Colorectal, Lung, or Prostate Cancer: A Systematic Review and Meta-analysis

Abstract

Objective: To assess long-term outcomes with robotic versus laparoscopic/thoracoscopic and open surgery for colorectal, urologic, endometrial, cervical, and thoracic cancers.

Background: Minimally invasive surgery provides perioperative benefits and similar oncological outcomes compared with open surgery. Recent robotic surgery data have questioned long-term benefits.

Methods: A systematic review and meta-analysis of cancer outcomes based on surgical approach was conducted based on Preferred Reporting Items for Systematic Reviews and Meta-analyses guidelines using Pubmed, Scopus, and Embase. Hazard ratios for recurrence, disease-free survival (DFS), and overall survival (OS) were extracted/estimated using a hierarchical decision tree and pooled in RevMan 5.4 using inverse-variance fixed-effect (heterogeneity nonsignificant) or random effect models.

Results: Of 31,204 references, 199 were included (7 randomized, 23 database, 15 prospective, 154 retrospective studies)-157,876 robotic, 68,007 laparoscopic/thoracoscopic, and 234,649 open cases. Cervical cancer: OS and DFS were similar between robotic and laparoscopic [1.01 (0.56, 1.80), P =0.98] or open [1.18 (0.99, 1.41), P =0.06] surgery; 2 papers reported less recurrence with open surgery [2.30 (1.32, 4.01), P =0.003]. Endometrial cancer: the only significant result favored robotic over open surgery [OS; 0.77 (0.71, 0.83), P <0.001]. Lobectomy: DFS favored robotic over thoracoscopic surgery [0.74 (0.59, 0.93), P =0.009]; OS favored robotic over open surgery [0.93 (0.87, 1.00), P =0.04]. Prostatectomy: recurrence was less with robotic versus laparoscopic surgery [0.77 (0.68, 0.87), P <0.0001]; OS favored robotic over open surgery [0.78 (0.72, 0.85), P <0.0001]. Low-anterior resection: OS significantly favored robotic over laparoscopic [0.76 (0.63, 0.91), P =0.004] and open surgery [0.83 (0.74, 0.93), P =0.001].

Conclusions: Long-term outcomes were similar for robotic versus laparoscopic/thoracoscopic and open surgery, with no safety signal or indication requiring further research (PROSPERO Reg#CRD42021240519).

Figure 1.. Cervical Cancer Forest Plots.

Forest plots showing hazard ratios for individual papers as well as pooled outcomes for cervical cancer for A) incidence of recurrence for RAS vs OPEN, B) disease-free survival (DFS) for RAS vs LAP, and C) for RAS vs OPEN, D) overall survival (OS) for RAS vs LAP, and E) for RAS vs OPEN. RAS=robotic-assisted surgery, LAP=laparoscopy, OPEN=open surgery, CI=confidence interval, BMI=body mass index, KM=Kaplan-Meier, yr=year, mo=month, HR=hazard ratio, FU=follow-up, sig=significant, diff=different, PFS=progression-free survival, calc=calculated, DOD=dead of disease, DOO=dead of other, AWD=alive with disease.

Figure 2.. Endometrial Cancer Forest Plots.

Forest plots showing hazard ratios for individual papers as well as pooled outcomes for endometrial cancer for A) disease-free survival (DFS) for RAS vs. LAP, and B) for RAS vs. OPEN, C) overall survival for RAS vs. LAP, and D) for RAS vs. OPEN. RAS=robotic-assisted surgery, LAP=laparoscopy, OPEN=open surgery, CI=confidence interval, BMI=body mass index, KM=Kaplan-Meier, yr=year, mo=month, HR=hazard ratio, FU=follow-up, sig=significant, diff=different, PFS=progression-free survival, calc=calculated, DOD=dead of disease, DOO=dead of other, AWD=alive with disease.

Figure 3.. Lobectomy for Lung Cancer Forest Plots.

Forest plots showing hazard ratios for individual papers as well as pooled outcomes for lobectomy for lung cancer for A) disease-free survival (DFS) for RAS vs. VATS, and B) for RAS vs. OPEN, C) overall survival (OS) for RAS vs. VATS, and D) for RAS vs. OPEN. RAS=robotic-assisted surgery, VATS=video-assisted thoracoscopic surgery, OPEN=open surgery, CI=confidence interval, KM=Kaplan-Meier, ref=reference group, sig=significant, diff=different, clin=clinical, pt=patient.

Figure 4.. Prostatectomy Forest Plots.

Forest plots showing hazard ratios for individual papers as well as pooled outcomes for prostate cancer for A) Biochemical-recurrence (BCR) for RAS vs. LAP (BCR for RAS vs. OPEN is in Supplementary Figure 7), B) BCR-Free Survival (BCRFS) for RAS vs. LAP and for C) RAS vs. OPEN, D) Overall Survival (OS) for RAS vs. LAP and for E) RAS vs. OPEN. RAS=robotic-assisted surgery, LAP=laparoscopy, OPEN=open surgery, CI=confidence interval, KM=Kaplan-Meier, GS=gleason score, NS=not significant, yr=year, mo=month, HR=hazard ratio, ref=reference group, sig=significant, calc=calculated.

Figure 5.. Forest Plots for Low-Anterior Resection for Rectal Cancer.

Forest plots of hazard ratios for low-anterior resection for rectal cancer. A) local recurrence RAS vs. LAP, and B) for RAS vs. OPEN, C) Disease-Free Survival (DFS) for RAS vs. LAP, and D) for RAS vs. OPEN, E) Overall Survival (OS) for RAS vs. LAP, and F) for RAS vs. OPEN.