The Prognostic Value of a Tumor Microenvironment-Based Immune Cell Infiltration Score Model in Colon Cancer

The current study aimed to construct a prognostic predictive model based on tumor microenvironment. CIBERSORT and ESTIMATE algorithms were used to reveal the immune cell infiltration (ICI) landscape of colon cancer. Patients were classified into three clusters by ConsensusClusterPlus algorithm. ICI scores of each patient were determined by principal component analysis. Patients were divided into high and low ICI score groups. Survival, gene expression, and somatic mutation of the two groups were compared. We found that patients with no lymph node invasion, no metastasis, T1-2 disease, and stage I-II had higher ICI scores. Calcium signaling pathway, leukocyte transendothelial migration pathway, MAPK signaling pathway, TGF beta pathway, and Wnt signaling pathway were enriched in the high ICI score group. Immune-checkpoint and immune-activity associated genes were decreased in high ICI score patients. Patients in the high ICI score group had better survival. Prognostic value of ICI score was independent of tumor mutational burden (TMB). The ICI score model constructed in the current study may serve as an independent prognostic biomarker in colon cancer.

Automated summary

In this study, a prognostic predictive model based on tumor microenvironment was constructed, revealing a correlation between immune cell infiltration landscape and survival in colon cancer patients. The high ICI score group showed enrichment in specific pathways and better survival, with the prognostic value of ICI score being independent of tumor mutational burden.