Identification of Immune Subtypes of Esophageal Adenocarcinoma to Predict Prognosis and Immunotherapy Response

Abstract

A low response rate limits the application of immune checkpoint inhibitors (ICIs) in the treatment of esophageal adenocarcinoma (EAC), which requires the precise characterization of heterogeneous tumor microenvironments. This study aimed to identify the molecular features and tumor microenvironment compositions of EAC to facilitate patient stratification and provide novel strategies to improve clinical outcomes. Here, we performed consensus molecular subtyping with nonnegative matrix factorization (NMF) using EAC data from the Cancer Genome Atlas (TCGA) and identified two distinct subtypes with significant prognostic differences and differences in tumor microenvironments. The findings were further validated in independent EAC cohorts and potential response to ICI therapy was estimated using Tumor Immune Dysfunction and Exclusion (TIDE) and SubMap methods. Our findings suggest that EAC patients of subtype 2 with low levels of cancer-associated fibroblasts, tumor associated macrophages (TAMs), and MDSCs in the tumor microenvironment may benefit from PD-1 blockade therapy, while patients of subtype 1 are more responsive to chemotherapy or combination therapy. These findings might improve our understanding of immunotherapy efficacy and be useful in the development of new strategies to better guide immunotherapy and targeted therapy in the treatment of EAC.