Context-dependent interactions between immune checkpoint and epithelial-mesenchymal transition programs shape the tumor microenvironment in esophageal squamous cell carcinoma
TL;DR
Findings delineate context-dependent transcriptional interactions between ICP and EMT pathways in ESCC and highlight the complexity of the tumor immune microenvironment beyond single-parameter biomarkers such as TMB.
Abstract
Immune checkpoint (ICP) and epithelial-mesenchymal transition (EMT) pathways are key biological processes implicated in the progression of esophageal squamous cell carcinoma (ESCC). However, their transcriptional interplay and interactions with the tumor immune microenvironment remain incompletely characterized. Transcriptomic, somatic mutation, and clinical data from TCGA-ESCC (n = 95) were analyzed using a predefined hypothesis-driven gene panel comprising five ICP genes (PDCD1, CD274, CTLA-4, HAVCR2, LAG-3) and two EMT markers (TWIST1, MMP13). Immune cell infiltration was estimated using multiple computational deconvolution algorithms. Tumor mutational burden (TMB), ESTIMATE, and TIDE scores were assessed as exploratory immune-related metrics. Findings were validated in an independent microarray cohort (GSE53624). ICP and EMT genes formed distinct yet partially overlapping expression modules. HAVCR2 demonstrated the strongest and most consistent association with EMT-related signatures across datasets. Elevated expression of most ICP genes and MMP13 was associated with poorer overall survival, whereas TWIST1 exhibited dataset-dependent prognostic behavior. ICP-high tumors were characterized by features of an immunologically active yet regulatory-enriched microenvironment, while EMT-related genes were linked to fibroblast- and stroma-dominant profiles. Advanced-stage tumors displayed concurrent immune activation and immune evasion signals. TMB was low overall and tended to decreased across stages, with no significant prognostic association. These findings delineate context-dependent transcriptional interactions between ICP and EMT pathways in ESCC and highlight the complexity of the tumor immune microenvironment beyond single-parameter biomarkers such as TMB. All associations are exploratory and hypothesis-generating, requiring mechanistic and prospective validation prior to translational application.