| Object: To investigate the role of Hp infection-related genes in gastric adenocarcinoma(STAD)using public databases and to screen for target genes with prognostic value.A combined multi-indicator prediction model was constructed to more accurately predict tumour progression and metastasis and response to immunotherapy.Methods:(1)A total of 876 STAD patients were downloaded from the Gene Expression Omnibus database(GEO;https: //www.ncbi.nlm.nih.gov/geo/,GSE84437)and The Cancer Genome Atlas(TCGA;https://portal.gdc.cancer.gov/)databases for gene expression data and complete clinical annotations,and batch correction and homogenization of the data were performed.A total of 2014 Hp infection-associated genes were downloaded from the Gene Card database(https: //www.genecards.org/)and 73 Hp-associated genes were obtained from the GSEAMSig DB(https: //www.gsea-msigdb.org/gsea/msigdb)website.(2)Differentially expressed gene analysis,survival analysis,one-way Cox regression analysis,Lasso regression analysis and multi-factor Cox regression analysis were then used to screen Hp infection-associated prognostic genes.(3)Combined multi-metric models were used to predict postoperative survival of STAD patients.(4)The efficacy of prognostic models for predicting postoperative survival time of STAD patients was evaluated in external and internal data sets by ROC curves,survival curves,calibration curves and DCA curves.(5)Quantitative real-time PCR(q RT-PCR)and immunohistochemistry(IHC)were used to detect the expression levels of key genes in STAD tissues in the prognostic model;chi-square test and Fisher’s exact test were used to analyze the correlation between the expression of key genes and patients’ clinical characteristics and prognosis.(6)Gene Set Enrichment Analysis(GSEA)and CIBERSORT algorithm were used to explore the mechanisms of key genes affecting STAD development in the STAD prognostic model.Results:(1)Hp-related prognostic markers of STAD were obtained by screening for SERPINE1(plasminogen activator inhibitor-1),CLDN1(Claudin1),CTHRC1(collagen triple helix repeat containing-1),and NRP1(neuropilin 1).(2)A multi-indicator combined prognostic model for STAD was established: age,differentiation,SERPINE1,NRP1,CLDN1 and CTHRC1.the sensitivity of this model to predict the prognosis of STAD patients in the training and validation groups was 83.33%、and 76.92%,respectively,and the specificity was 84.81% and 55.0%,respectively.(3)The prognostic model for STAD was evaluated in the internal and external datasets,and the AUCs of the ROC curves to determine the accuracy of the model at 1,3,and 5 years were 0.678,0.843,and 0.882,respectively;the Kaplan-Meier method showed different and poor survival curves between the high and low mortality risk groups(P <0.001,P = 0.002).The results of calibration curve analysis of the internal and external data sets showed a good fit of the model.DCA curves can show the good clinical benefit of the model.(4)Elevated m RNA expression of SERPINE1,CTHRC1,NRP1 and CLDN1 in STAD tissues was confirmed according to q RT-PCR(P < 0.05)and correlated with Hp infection.(5)Immunohistochemistry observed cytoplasmic and cytomembrane staining of CTHRC1,NRP1 and CLDN1 in STAD tissues and cytoplasmic staining of SERPINE1 in STAD tissues(P < 0.05)and they were associated with Hp infection,and all of them were significantly associated with clinical characteristics and prognosis of patients’ tumor invasion.(6)The high m RNA expression samples of SERPINE1,CTHRC1,NRP1 and CLDN1 were all enriched in the Extracellular Matrix(ECM)receptor interaction pathway and correlated with multiple immune cells.Conclusion: In this study,four Hp-related genes,SERPINE1,CLDN1,CTHRC1,and NRP1,were obtained by bioinformatics analysis and could be potential prognostic markers for STAD patients,and a multi-indicator STAD prognostic model could be established from "clinical factor-protein-tissue".This will help doctors to develop the best treatment plan for STAD patients. |
PDF Full Text Request