Building a prognostic model based on Cytokine-related genes and exploration of colon cancer prognostic indicators

To explore the potential impact of immune microenvironment and gene mutation on the prognosis of patients with colon cancer. A colon cancer prognostic model was constructed based on cytokine-related genes. Colon cancer transcriptome sequencing data, clinical information, mutation information and cytokine-related gene lists were obtained from TCGA, GEO, UCSC, and ImmPort. The 112 genes of CRDGs were constructed by differential analysis and intersection with cytokine-related genes. Based on this gene set, a risk model was constructed, evaluated and validated, and PPI, GO and KEGG enrichment analyses were performed on this gene set. From the performance of the ROC of the training and validation models, this model has good predictive ability, and the risk score can be used as an independent factor for colon cancer prognosis. Immune infiltration and mutation analysis based on the risk model showed that Bcell, Tcell, and M2 were significantly decreased in the high-risk group, while M0 was increased. From the expression of mutated genes in risk groups, TTN, TP53, KRAS, APC, MUC16, and MUC4 have beneficial or adverse effects on prognosis. Independent prognostic analysis and drug sensitivity analysis revealed a certain clinical value of this model. In summary, The model constructed by CRDGs has good predictive ability and can be used as an independent factor for clinical prognosis. The immune microenvironment and some gene mutations have important effects on tumor prognosis.