Abstract
BACKGROUND: Colorectal cancer (CRC) is among the most common and lethal cancers worldwide. Recent advances in tumor immunotherapy have highlighted the importance of T-cell subsets and the tumor microenvironment (TME) in CRC, both critical for mounting a successful anti-tumor immune response. Thus, there is an urgent need for comprehensive research in these areas to accelerate the development of personalized immunotherapeutic strategies for CRC patients. Therefore, this study aims to explore T-cell heterogeneity, identify characteristic genes, and develop a reliable prognostic model to predict patient outcomes and immunotherapy responses. METHODS: CRC single-cell RNA sequencing (scRNA-seq) data were downloaded from the Gene Expression Omnibus (GEO) database. We identified T cell feature genes and employed Cox and least absolute shrinkage and selection operator (LASSO) regression methods to construct a prognostic model using The Cancer Genome Atlas (TCGA) dataset, while GEO dataset was utilized for validation. Additionally, we analyzed the immune microenvironment, immune checkpoints, and drug sensitivity between the high-risk and low-risk groups. RESULTS: Based on scRNA-seq data, we identified 4,440 T cell marker genes. There were also interactions between T cells and various other cell types. The results of the enrichment analysis revealed that these marker genes were primarily involved in immune-related pathways. After conducting univariate Cox and LASSO analyses, we developed a prognostic model based on the expression levels of three prognostic genes: TIMP1, HMMR, and HIST1H1C. The feasibility of the model was validated using external validation cohort. In addition, the high-risk group exhibited higher immune scores, stromal scores, ESTIMATE scores, and Tumor Immune Dysfunction and Exclusion (TIDE) prediction scores compared to the low-risk group, indicating a poorer response to immunotherapy. CONCLUSIONS: This study discovered three key genes with prognostic relevance in CRC, providing valuable insights into T-cell heterogeneity and the associated prognostic risk model. The results demonstrated that this model accurately predicts patient prognosis and responses to immunotherapy.