Logo image
Back
Characterization of the tumor microenvironment in endometrial cancer
Abstract   Peer reviewed

Characterization of the tumor microenvironment in endometrial cancer

Alexandra Morell, Robert Wenham, Talicia Savage, Casey Cosgrove, Marisa Moroney, Marilyn Huang, Eugenia Girda, Laura Holman, Jessica Parker, Craig Shriver, …
Gynecologic oncology, Vol.208(Supplement), pp.S414-S415
05/2026
DOI: 10.1016/j.ygyno.2026.01.711

View Online

Abstract

Objectives To describe the tumor microenvironment in endometrial malignancies and determine if differences in immune cell composition have prognostic significance. Methods The Oncology Research Information Exchange Network (ORIEN) collaborative was used to identify a cohort of patients with endometrial malignancies. CIBERSORT was used to estimate the immune cell composition (22 different immune cell types) of the entire cohort. Immune cell composition was compared against various demographic and clinicopathologic variables using Wilcoxon tests or Kruskal-Wallis tests. Kaplan-Meier curves were used to assess overall survival (OS) based on total immune cell infiltration. Univariate Cox regression analyses were performed to determine whether the abundance of individual immune cell types was correlated with patient survival. Consensus clustering was used to identify distinct immune subtype groupings, which were then used to stratify patients and evaluate associations with OS. Results A total of 1191 endometrial cancer patients were in the initial cohort, of which 987 had analyzable data. Fig. 1 demonstrates the immune cell composition of the entire endometrial cancer cohort. The most prevalent immune cells were CD4 resting memory T cells (RMTC) (14.4%), plasma cells (11.5%), M2 macrophages (11.1%), and activated mast cells (10.2%). M2 macrophages (p = 0.003) and CD4 RMTCs (p = 0.003) significantly differed by histology. M2 macrophages (p = 0.038) and CD4 RMTC (p = 0.038) also significantly differed by race. CD8 T cells (p = 0.032), CD4 RMTC (p = 0.002), resting dendritic cells (p = 0.032), and M1 macrophages (p = 0.000) significantly differed by grade. Appalachian patients had a significantly higher proportion of CD4 RMTC compared with non-Appalachian patients (p = 0.037). Immune cell infiltration (high vs low) was not associated with OS (p = 0.057). On univariate Cox proportional hazards analysis, activated dendritic cells (p = 0.002) and plasma cells (p = 0.002) were associated with improved overall survival. However, after correction for multiple testing across all immune cell types, only activated dendritic cells remained significantly associated with improved survival (p.adj = 0.040). 3 distinct immune subtype groupings were identified on consensus clustering; however, no difference in OS was seen between clusters (p = 0.860). Conclusions Immune cell composition differed by histology, race, grade, and Appalachian status, mainly with regard to M2 macrophages and CD4 RMTC. However, M2 macrophages and CD4 RMTC did not have prognostic significance. The only immune cell with prognostic significance was activated dendritic cells, which have also been associated with improved survival in other cancer types. Future directions include assessing the difference in immune cell composition by molecular subtype and investigating if certain tumor microenvironment immune profiles are associated with response to immunotherapy in endometrial cancer.

Details

Metrics

1 Record Views
Logo image