Construction and validation of neutrophil-related gene based risk models for assessing colon cancer prognosis and guiding immunotherapy

Abstract

Abstract Purpose Colon cancer is one of the most common digestive tract malignancies. Studies have shown that neutrophils can interact with immune cells and immune factors to affect the prognosis of patients. Methods We first determined the infiltration level of neutrophils in tumors using CIBERSORT and identified key genes in the final risk model by Spearman correlation analysis and subsequent Cox analysis. The risk score of each patient was obtained by multiplying the Cox regression coefficient by the gene expression level, and patients were divided into two groups according to the median. Differences in OS and PFS were assessed by KM survival analysis, and model accuracy was validated in another independent dataset. Finally, the differences in immune infiltration and immunotherapy were evaluated by immunoassay. Results We established and validated a risk scoring model based on neutrophil-related genes in two independent datasets; the patients in the high-risk group had a poorer prognosis than those in the low-risk group. A new nomogram was constructed and validated by combining clinical characteristics and the risk score model to better predict patient OS and PFS. Immune analysis showed that patients in the high-risk group had immune cell infiltration level, immune checkpoint levels, and tumor mutational burden and were more likely to benefit from immunotherapy. Conclusion The low-risk group had relatively better OS and PFS than the high-risk group in the neutrophil-related gene-based risk model. Patients in the high-risk group presented higher immune infiltration levels and tumor mutational burden and thus may be more responsive to immunotherapy.