Analysis of pyroptosis-related genes of colon cancer cells based on bioinformatics screening and construction of prognostic model / 肿瘤研究与临床

ABSTRACT

Objective:To explore the characteristics of pyroptosis-related genes in colon cancer cells screened by bioinformatics, and to verify the constructed prognostic model of colon cancer based on differentially expressed pyroptosis-related genes.Methods:Genetic data of RNA sequencing and clinical data of colon cancer patients were downloaded from The Cancer Genome Atlas (TCGA) database. Fifty-two genes associated with pyroptosis were identified by searching the literature and compared with the RNA sequencing gene dataset of colon cancer and normal colon tissues obtained from TCGA database to obtain differentially expressed pyroptosis-related genes in clinical samples. The protein interaction network of differentially expressed pyroptosis-related genes was analyzed by using STRING website and R software. Based on the differential expression of pyroptosis-related genes in clinical samples of TCGA database, colon cancer patients in TCGA database were divided into pyroptosis and non-pyroptosis groups, and genes with significant differential expression between the two groups were screened at P < 0.05 according to gene expression; based on these differentially expressed genes, LASSO Cox regression was used to construct a prognostic model of colon cancer associated with pyroptosis. Patients collected from TCGA database were divided into high risk (≥ median value) and low risk (< median value) groups according to the median value of risk scores calculated by the model, and the overall survival of the two groups was analyzed by Kaplan-Meier survival function. The time ROC package of R software was used to analyze the efficacy of applying risk scores to predict the different survival time of colon cancer patients in TCGA database. Multivariate Cox regression was used to analyze the effects of clinicopathological factors and risk scores calculated by the model on the survival of patients in TCGA database. R software was used to analyze and obtain the differential genes between high and low risk groups of colon cancer patients in TCGA database. R software was used to conduct Gene Ontology (GO), Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis and single sample gene set enrichment analysis of immune cells and immune function (ssGESA) for differentially expressed pyroptosis-related genes. Results:Thirty-eight differentially expressed pyroptosis-related genes between colon cancer tissues and normal tissues of clinical samples were obtained based on data of TCGA database. A prognostic model consisting of 13 pyroptosis-related genes was established by applying LASSO Cox regression, the risk score = 0.118×MID2+0.354×IL20RB+0.083×HOXC11+0.011×TMEM88+0.021×SYNGR3+0.246×UPK3B+0.030×EGFL7+0.109×TMPRSS11E+0.138×IFITM10+0.161×RNF207+0.097×LINGO1+0.202×HEYL+0.025×ROBO3. Survival analysis showed that TCGA database had worse overall survival in the high-risk group than in the low-risk group ( P < 0.001). Receiver operating characteristic (ROC) curve analysis showed that the area under the curve of the prognostic model risk score in predicting the survival of colon cancer patients in TCGA database at 1, 3 and 5 years was all > 0.7. Multivariate Cox regression analysis showed that risk score was an independent influencing factor for survival of colon cancer patients in TCGA database (high risk vs. low risk HR = 3.988, 95% CI 2.865-5.551, P < 0.001). GO and KEGG enrichment analysis showed that the differentially expressed genes between high and low risk groups (SULF1, FBLN2, COL1A1, DES, SFRP2, FNDC1, MYH11, APOE, C3, SPP1, COL1A2, COL10A1, THBS2, AEBP1, CNN1, IGHG1, and SFRP4) were upregulated in the high risk group, which were mainly associated with cellular matrix structural components and extracellular matrix (ECM) receptor interactions. ssGSEA analysis showed that the level of immune cell infiltration was higher in high risk group, especially B cells, macrophages, mast cells, helper T cells, and tumor-infiltrating lymphocytes were higher than those in low risk group; for immune function, chemokine receptors, immune checkpoints, human leukocyte antigens, parainflammation, T cell suppression, T-cell stimulation, and type Ⅱ interferon response in high risk group were higher than those in low risk group. Conclusions:The constructed prognostic model of colon cancer based on pyroptosis-related genes is valuable for predicting the prognosis of colon cancer patients. Pyroptosis-related genes may play an important role in tumor immunity of colon cancer and can be used for prognostic analysis of colon cancer patients.