Survival of microsatellite-stable endometrioid endometrial cancer patients after minimally invasive surgery: An analysis of the Cancer Genome Atlas data

Objective. This study aims to investigate the survival impact of minimally invasive surgery on endometrial cancer (EC) patients with different histology and microsatellite status. Methods. This is a retrospective study based on the Cancer Genome Atlas (TCGA) data. 519 eligible EC patients were divided into four subgroups according to histology and microsatellite status. Kaplan-Meier survival analyses were conducted in all patients and four subgroups to compare the survival outcome after two surgeries (open vs. minimally invasive). Propensity score matching and propensity score covariate adjustment models were used to control confounders. To establish survival prediction models for EC patients, multivariate stepwise Cox regressions were conducted. Results. Among the eligible patients, 318 (61.3%) received open surgery and 201 (38.7%) received minimally invasive surgery. Overall survival was similar between the two groups (p = 0.33), but the latter showed significantly shorter recurrence-free survival (RFS) (p = 0.005). Subgroup analyses revealed the survival influence of surgical approach was only significant in microsatellite-stable (MSS) endometrioid EC patients. These results were verified by Kaplan-Meier survival analyses after propensity score matching and propensity score covariate adjustment models. Finally, the survival influence of multiple clinicopathological factors was analyzed. After stepwise Cox regressions, minimally invasive surgery was found to be independent risk factor for shorter RFS of all patients (hazard ratio [HR] = 2.038, 95% confidence interval [CI] 1.111-3.741, p = 0.02) and MSS patients (HR = 2.449, 95% CI 1.064-5.639, p = 0.04). Conclusions. Minimally invasive surgery is associated with more rapid recurrence in MSS endometrioid EC patients, thus indicating the necessity of microsatellite testing for guiding EC surgery. (C) 2020 Published by Elsevier Inc.