Abstract
Immunogenic cell death (ICD) plays a critical role in the host’s antitumor immune response, and its dysregulation is linked to poor prognosis in colorectal cancer (CRC). Identifying ICD-related genes and developing prognostic models can enhance the prediction of patient outcomes and facilitate personalized treatments. Utilizing data from CRC patients in the Cancer Genome Atlas, we employed weighted gene co-expression network analysis, LASSO regression analysis, and multivariate Cox regression screening to identify immunogenic cell death-related prognostic genes (IPGs) in CRC patients. An IPGs scoring system was constructed, allowing for the stratification of patients into high-risk and low-risk groups based on the best cutoff ICD-RiskScore, with the Gene Expression Omnibus dataset serving as a validation cohort. We then examined the correlations between the high-risk and low-risk groups concerning clinicopathological features, immune infiltration, immunotherapy responses, and tumor mutational burden. Additionally, we analyzed the cellular landscape and cell-cell interactions in colorectal cancer at a single-cell resolution. A prognostic model comprising 12 immune-related prognostic genes (IPGs)—CAV2, NOXA1, TGFB2, WRNIP1, CCL22, ICAM2, APOBEC3F, COTL1, CTSD, IDO1, NFATC1, and RUNX3—was established, demonstrating a high area under the curve value in survival analysis, which indicates robust predictive performance. In comparison to the low-risk group, the high-risk group exhibited lower survival rates, higher mutation rates of genes, and was associated with more advanced tumor stages. Single-sample gene set enrichment analysis revealed that the low-risk group possessed a stronger immune surveillance function. Additionally, single-cell analysis uncovered a lower abundance of immune cells and an increased proportion of epithelial cells within tumor samples. The frequency of interactions among epithelial cells, endothelial cells, and fibroblasts was found to be elevated in tumor samples. Subsequently, validation of the prognostic genes revealed that COTL1 expression was significantly elevated in CRC tissues compared to adjacent normal tissues (P < 0.05). This study presents a prognostic model for colorectal cancer that integrates 12 immune-related genes. The model holds potential clinical value in predicting patient prognosis and guiding personalized treatment strategies. These findings also provide insights into the molecular mechanisms of immune cell dynamics in colorectal cancer and highlight potential therapeutic targets. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1038/s41598-025-33862-2.