Survival Advantage of Laparoscopic Versus Open Resection For Colorectal Liver Metastases A Meta-analysis of Individual Patient Data From Randomized Trials and Propensity-score Matched Studies

Objective: To perform an individual participant data meta-analysis using randomized trials and propensity-score matched (PSM) studies which compared laparoscopic versus open hepatectomy for patients with colorectal liver metastases (CLM). Background: Randomized trials and PSM studies constitute the highest level of evidence in addressing the long-term oncologic efficacy of laparoscopic versus open resection for CLM. However, individual studies are limited by the reporting of overall survival in ways not amenable to traditional methods of meta-analysis, and violation of the proportional hazards assumption. Methods: Survival information of individual patients was reconstructed from the published Kaplan-Meier curves with the aid of a computer vision program. Frequentist and Bayesian survival models (taking into account random-effects and nonproportional hazards) were fitted to compare overall survival of patients who underwent laparoscopic versus open surgery. To handle long plateaus in the tails of survival curves, we also exploited cure models to estimate the fraction of patients effectively cured of disease. Results: Individual patient data from 2 randomized trials and 13 PSM studies involving 3148 participants were reconstructed. Laparoscopic resection was associated with a lower hazard rate of death (stratified hazard ratio = 0.853, 95% confidence interval: 0.754-0.965,P= 0.0114), and there was evidence of time-varying effects (P= 0.0324) in which the magnitude of hazard ratios increased over time. The fractions of long-term cancer survivors were estimated to be 47.4% and 18.0% in the laparoscopy and open surgery groups, respectively. At 10-year follow-up, the restricted mean survival time was 8.6 months (or 12.1%) longer in the laparoscopy arm (P< 0.0001). In a subgroup analysis, elderly patients (>= 65 years old) treated with laparoscopy experienced longer 3-year average life expectancy (+6.2%,P= 0.018), and those who live past the 5-year milestone (46.1%) seem to be cured of disease. Conclusions: This patient-level meta-analysis of high-quality studies demonstrated an unexpected survival benefit in favor of laparoscopic over open resection for CLM in the long-term. From a conservative viewpoint, these results can be interpreted to indicate that laparoscopy is at least not inferior to the standard open approach.