Construction of a prognostic risk model based on pyroptosis-related genes and comprehensive analysis of key genes and tumor immune microenvironment for colon cancer.

Abstract

Pyroptosis-related genes have great potential for prognosis, an accurate prognostic model based on pyroptosis genes has not been seen in Colorectal adenocarcinoma (COAD). Furthermore, understanding the mechanisms of gene expression characteristics and the Tumor Immune Microenvironment associated with the prognosis of COAD is still largely unknown. Constructing a prognostic model based on pyroptosis-related genes, and revealing prognosis-related mechanisms associated with the gene expression characteristics and tumor microenvironment. 59 pyroptosis-related genes were collected. The gene expression data and clinical data of COAD were downloaded from The Cancer Genome Atlas. External validation datasets were downloaded from the Gene Expression Omnibus database. 10 characteristic genes with prognostic values were obtained using univariate and LASSO Cox. 10-gene Riskscore prognostic model was constructed. Both gene set enrichment analysis and network propagation methods were used to find pathways and key genes leading to different prognostic risks. The area under the ROC curves were used to evaluate the performance of the model to distinguish between high-risk and low-risk patients, the results were 0.718, 0.672, and 0.669 for 1-, 3-, and 5-year survival times. A nomogram based on Riskscore and clinical characteristics showed the probability of survival at 1, 3, and 5 years, and the calibration curves showed good agreement between the predicted and actual observations, its C-index is 0.793. The decision curves showed that the net benefit of the nomogram was significantly superior to that of the other single variables. Four key pathways leading to different prognostic risks were obtained. Six key genes with prognostic value, significant expression differences (P < .05) and significant survival differences (P < .05) between high/low risk groups were obtained from the gene set of all 4 key pathways. This study constructed a prognostic model for COAD using 10 pyroptosis-related genes with prognostic value. This study also revealed significant differences in specific pathways and the tumor immune microenvironment (TME) between the high-risk group and the low-risk group, highlighted the roles of ALDH5A1 and Wnt signaling in promoting COAD and the suppressive effects of the IL-4/IL-13 pathway and RORC on COAD. The study will be helpful for precision therapy.