Abstract:

Objective To construct a prognostic model of lung adenocarcinoma（LUAD）-pyroptosis genes based on bioinformatics，and to investigate the relationship between LUAD and cell pyroptosis. Methods The transcriptomic data of LUAD patients（TCGA cohort） and clinical data were downloaded from the Cancer Genome Atlas（TCGA），and the limma program package was used to identify differentially expressed pyroptosis genes. Cluster typing was obtained from TCGA cohort based on the expression characteristics of differentially expressed pyroptosis genes，and differentially expressed pyroptosis genes between clustering were identified. Using least absolute shrinkage and selection operator（LASSO）-Cox regression method，a prognostic model based on inter-clustering differentially expressed pyroptosis genes was developed，and the risk scores were calculated. TCGA cohort LUAD patients were classified into high and low risk groups using the median risk score. The clinical prognostic role of the prognostic model was analyzed using Kaplan-Meier survival curve and Cox regression analysis in conjunction with clinical characteristics. The LUAD patients from the Gene Expression Omnibus（GEO） were classified into 2 risk groups（high and low） using the median risk score of the TCGA cohort，and the clinical prognostic validation was performed. The signaling pathways and biological functions of differentially expressed pyroptosis genes in TCGA cohort were analyzed using Gene Ontology（GO） and Kyoto Encyclopedia of Genes and Genomes（KEGG） analysis. Single sample gene set enrichment analysis（ssGSEA） and CIBERSORT were used to compare immune cells and immune-related pathways. Results Totally，41 differentially expressed pyroptosis genes were identified in LUAD tissues and paracancer tissues. The results of cluster analysis showed that TCGA cohort samples could be classified into 2 cluster types，and the survival time of type 2 was longer than that of type 1（P<0.05）. Totally，11 differentially expressed pyroptosis genes（ANO1，PKHD1L1，FMO2，TDRD1，TBX4，ABCC12，AQP4，CPXM2，WFIKKN2，ATP1A4 and IGSF10） associated with survival in LUAD patients were identified. Kaplan-Meier survival curve analysis showed that the survival time of TCGA cohort high risk group was shorter than that of low risk group（P<0.05）. The same results were obtained using GEO cohort. Cox regression analysis showed that the risk score derived from the prognostic risk model was an independent risk factor for the prognosis of LUAD patients. Heat map analysis of 11 differentially expressed pyroptosis genes in the TCGA cohort showed that there was statistical significance in sex，clinical stage，T stage and N stage between high risk and low risk group（P<0.05）. The differentially expressed pyroptosis genes between high risk and low risk groups in TCGA cohort were related to chemokines mediating tumor-related signaling pathways，immune response and cell membrane function. Conclusions Pyroptosis genes play roles in tumor immunity，and the constructed 11 pyroptosis genes-related prognostic model can be used to predict the prognosis of LUAD.

Key words: Cell pyrotosis gene, Lung adenocarcinoma, The Cancer Genome Atlas, Prognosis, Immune infiltrating cell