Constructing And Analyzing Pancreatic Cancer CircRNA-associated CeRNA Prognostic Network And Prognostic Model Based On Bioinformatics

[Objective] The Gene Expression Omnibus(GEO)and The Cancer Genome Atlas T(TCGA)were applied to construct a prognosis-related circ RNA-mi RNA-m RNA ce RNA network and prognosis model for pancreatic cancer based on bioinformatics approach and analyze the ce RNA network in pancreatic cancer and explore the clinical application value of the prognostic model.[Methods] The gene microarray expression profile gene microarray datasets of circ RNA,mi RNA,m RNA and clinical information data of pancreatic cancer were retrieved and downloaded from GEO and TCGA databases,and gene differential expression analysis was performed using R language software based on bioinformatics approach,and circ RNA-mi RNA,mi RNA-m RNA interaction pairs were predicted using online databases.Based on ce RNA principle to construct circ RNA-mi RNA-m RNA ternary network,through GO and KEGG database for gene function enrichment analysis to understand the gene function and the related pathways involved.The survival analysis of m RNAs in the ce RNA network was performed by combining the gene expression profile of pancreatic cancer downloaded from TCGA database and clinical data using Kaplan-Meier method to construct a prognosis-related ce RNA network.Subsequently,genes with strong prognostic relevance were screened by LASOO regression analysis,and finally,prognostic models were constructed by univariate COX and multivariate COX regression analysis,and pancreatic cancer samples were divided into high-risk and low-risk groups according to the median value of the prognostic models.The accuracy of the prognostic model was verified using time-dependent ROC curves and clinical multi-indicator ROC curves.Next,the independence of the prognostic model was further evaluated by combining the prognostic model with clinical parameters for univariate COX and multivariate COX regression analysis and forest plotting.Tumor microenvironment(TME)and immune cell infiltration were analyzed using the R packages "ESTIMATE" and "CIBERSORT".Tmb tumor mutation load(TMB)analysis was performed on samples from high-risk and low-risk groups using the R language package "maftools",and finally the sensitivity of several common antitumor drugs in high-risk and low-risk groups was analyzed by the Genomics Database of Anticancer Drug Sensitivity(GDSC).[Results] By differential analysis of pancreatic cancer microarray expression profiling gene microarray dataset,457 circ RNAs,231 mi RNAs and 1186 m RNAs differentially expressed genes were obtained,and a ce RNA network containing 9circ RNAs,13 mi RNAs and 49 m RNAs was constructed based on the principle of ce RNA network,which was analyzed by GO and KEGG.The results suggest that the genes in the ce RNA network are involved in cell regeneration,differentiation,smooth muscle cell proliferation regulation,various protein kinase activity regulation,cell cycle regulation,DNA transcription factor activity regulation and other biological processes and are related to P53 signaling pathway,oxytocin signaling pathway,c GMP-PKG signaling pathway,etc.Combined with clinical information on the genes in the ce RNA network for survival The 16prognosis-associated m RNAs were analyzed to construct a prognosis-associated ce RNA network consisting of 9 circ RNAs,11 mi RNAs and 16 m RNAs.The m RNAs in this network were extracted for further analysis,and finally a prognostic model consisting of 7 genes(EMP1,ABHD2,PDE3 B,S100A16,PKM,NCAPG,RELN)was constructed,and the pancreatic cancer data could be divided into a high-risk group consisting of 89 samples and a low-risk group consisting of 89 samples according to the median risk score of the prognostic model.The Kaplan-Meier method survival analysis revealed that the overall survival time of the high-risk group was significantly shorter than that of the low-risk group.The accuracy of the prognostic model was assessed by plotting ROC curves with values of area under the curve(AUC values)of 0.699,0.704 and 0.709 for 1-year,2-year and 3-year survival,respectively),and AUC results for clinical parameter-related ROC curves were(prognostic model 0.74,age 0.656,sex 0465,pathological classification 0.603,clinical stage 0.59),and the results demonstrated that the prognostic model had better accuracy and predictive performance than other clinical indicators.The independence of the prognostic model was assessed by forest plots,and the results of the analysis showed that the prognostic model had higher independence compared with other clinical indicators in both univariate and multifactorial COX analyses,and could be used as an independent prognostic factor for the prognostic assessment of pancreatic cancer.Subsequently,the prognostic model was further analyzed in relation to TME and immune cell infiltration,and among the TME,the proportion of immune cells,the proportion of stromal cells in the TME of the low-risk group was higher and the tumor purity was lower,suggesting an association with a better prognosis in the low-risk group.By further analysis,four immune cells(primitive B cells,resting-state CD4+ memory T cells,M0-type macrophages,M1-type macrophages,and M2-type macrophages)were found to have a significantly higher composition in the high-risk group compared with the low-risk group,suggesting that the poor prognosis in the high-risk group may be associated with the infiltration of these immune cells.In the results of TMB analysis,the TMB scores were higher in the high-risk group than in the low-risk group,suggesting that the high-risk group may have a better response to immunotherapy.Finally,we analyzed to obtain the sensitivity of some commonly used antitumor drugs between different subgroups in the prognostic model,with 10 drugs having higher sensitivity in the high-risk group and 2 drugs having higher sensitivity in the low-risk group.[Conclusions] A multi-node circ RNA-mi RNA-m RNA prognosis-associated ce RNA network consisting of 9 circ RNAs,11 mi RNAs and 16 m RNAs and a prognosis model consisting of genes in 7 prognosis-associated ce RNA networks were constructed to provide a new theoretical basis for the study of pancreatic cancer development,progression mechanisms and treatment.