Construction of a prognostic model of prostate cancer based on pyroptosis gene and analysis of tumor microenvironment
Objective To construct a prognostic model of prostate cancer(PCa)based on pyroptosis gene and ana-lyze the tumor microenvironment.Method The transcriptome and mutation data of 502 PCa patients and 51 healthy sub-jects were obtained from The Cancer Genome Atlas(TCGA)database,and the differentially expressed pyroptosis genes were extracted.The clinical features and expression data of pyroptosis gene in 96 PCa patients were obtained from Gene Expression Omnibus(GEO)database,and the prognostic-related pyroptosis genes were screened out.The immune cell da-ta were downloaded from the ImmPort database.The optimal subtypes of PCa samples were determined by cluster analy-sis and verified by principal component analysis(PCA).The differential intersection genes of different subtypes of PCa patients were annotated by Gene Ontology(GO)function and enriched by Kyoto Encyclopedia of Genes and Genomes(KEGG).The significantly differentially expressed genes obtained by univariate Cox analysis were incorporated into Las-so regression model to obtain the genes associated with prognosis.PCa patients were randomly and equally divided into training set and test set and Lasso regression analysis was performed respectively.The training set and test set were divid-ed into high risk group and low risk group according to the median value of the risk score of the training set.Multivariate Cox regression model was used to analyze the independent prognostic factors of PCa patients and to construct a prognos-tic model.R language package was used to analyze the correlation between pyroptosis genes and immune cell content,stem cell content and risk score and the difference of tumor mutation load in high and low risk groups.Result The differ-ential analysis obtained 37 significantly differentially expressed pyroptosis genes.According to the expression of pyropto-sis genes and clinical data analysis,10 genes related to PCa prognosis were obtained.According to the K value of cumula-tive distribution function(CDF)diagram,PCa were divided into three subtypes A,B and C,and the classification results were verified by PCA analysis to be reliable.GO and KEGG enrichment analysis showed that the 109 differentially inter-secting genes mainly activated protein tyrosine kinase(PTK)and transmembrane receptor protein kinase(TRPK)and oth-er molecular functions through extracellular matrix organization(EMO),extracellular structure organization(ESO),pepti-dyl-tyrosine phosphorylation(PTP),peptidyl-tyrosine modification(PTM)and other pathways.Lasso regression model screened out 9 prognostic related genes including B-cell leukemia/lymphoma 3(BCL3),bone morphogenetic protein 2(BMP2),C-C motif chemokine receptor 5(CCR5),chitinase 3 like 2(CHI3L2),deleted in liver cancer 1(DLC1),major histocompatibility complex,class Ⅱ,DQ alpha 2(HLA-DQA2),secretoglobin family 3A member 1(SCGB3A1),serpin family E member 1(SERPINE1),solute carrier family 14 member 1(SLC14A1).Analysis of differences showed that out-comes in the training and test sets were roughly the same,and that the number of patients who died increased with the in-crease in risk scores.Multivariate Cox analysis showed that age,tumor stage(T stage,N stage)and risk degree were inde-pendent factors influencing the prognosis of PCa patients.The prognosis model predicted the survival of PCa patients at the 6th,9th and 12th year,which was consistent with the actual survival of patients.Correlation analysis showed that there was a correlation between immune cells and pyroptosis genes(P＜0.01).The difference between high-risk group(n=376)and low-risk group(n=126)was analyzed,and the results showed that the stromal cell score and tumor mutation load of high-risk group were significantly higher than those of low-risk group,and the differences were statistically significant(P＜0.01).Stem cell content was positively correlated with risk score(P＜0.01).Conclusion The PCa prognostic model based on pyroptosis genes is successfully constructed in this study,which finds that 9 pyroptosis genes,including BCL3,BMP2,CCR5,CHI3L2,DLC1,HLA-DQA2,SCGB3A1,SERPINE1,SLC14A1,can be used to predict the prognosis and immunotherapy response of PCa patients.It is helpful to further guide clinical practice.
万方数据
维普
