Neuro-Immuno-Stromal Context in Colorectal Cancer: An Enteric Glial Cell-Driven Prognostic Model via Machine Learning Predicts Survival, Recurrence, and Therapy Response.

Background: Enteric glial cells (EGCs) have been implicated in colorectal cancer (CRC) progression. This study aimed to develop and validate a prognostic model integrating EGC- and CRC-associated gene expression to predict patient survival, recurrence, metastasis, and therapy response.

Methods: Bulk and single-cell RNA sequencing data were analyzed, and a machine learning-based model was constructed using the RSF random forest algorithm. The model's prognostic value was evaluated through survival analysis, pathway enrichment, immune profiling, and therapy response predictions.

Results: The model effectively stratified patients into high- and low-risk groups, with high-risk patients exhibiting significantly worse overall survival (OS) and an increased likelihood of recurrence and metastasis. Gene Set Enrichment Analysis (GSEA) identified key pathways associated with tumor progression, immune regulation, and microenvironmental interactions. The model was significantly correlated with immune cell infiltration and chemokine signaling. High-risk patients exhibited reduced immune therapy efficacy and distinct drug sensitivity profiles, suggesting its potential to guide personalized treatment strategies.

Conclusion: This model serves as a valuable tool for CRC prognosis and treatment stratification, with potential clinical applications pending further validation.