| Background:Gastric cancer is one of the most common malignancies worldwide,associated with high mortality rates and a 5-year survival rate of less than 40%.Ferroptosis is a form of programmed cell death driven by the accumulation of lipid peroxidation and reactive oxygen species,while cuproptosis is a form of programmed cell death that directly induces cell death by the esterification of copper ions with components of the tricarboxylic acid cycle.Studies have shown that as metal-related forms of programmed cell death,ferroptosis and cuproptosis are closely associated with the occurrence and development of tumors.Furthermore,genes associated with ferroptosis and cuproptosis have been identified as novel biomarkers for predicting the prognosis of cancer patients.However,the relationship between genes related to ferroptosis and cuproptosis and the prognosis of gastric cancer patients requires further investigation.Therefore,this study aims to screen differentially expressed genes related to ferroptosis and cuproptosis,construct an individualized prognosis prediction model,and provide a theoretical basis for individualized treatment of gastric cancer patients.Objectives:To use bioinformatics methods to screen differentially expressed genes related to ferroptosis and cuproptosis that are associated with the prognosis of gastric cancer patients,perform unsupervised clustering subtype analysis,and establish a prognostic risk scoring model to accurately predict patient outcomes.The expression of relevant genes in gastric cancer tissues and adjacent tissues will be validated using immunohistochemical staining experiments,and the prognostic risk based on the research results will be assessed.Methods:Transcriptome data and clinical data from a total of 412 gastric adenocarcinoma patients’ tissues and 36 adjacent tissues were downloaded from the TCGA-STAD database.Prognosis-related differentially expressed genes related to ferroptosis and cuproptosis in gastric cancer patients were obtained using univariate Cox regression.The obtained differentially expressed genes were subjected to mutual correlation network analysis,protein-protein interaction network analysis,and copy number variation analysis.The gene chip data and clinical information of 433 gastric cancer patients from the sequence GSE84437 in the GEO database were downloaded and standardized.The TCGA-STAD data and GEO-GSE84437 data were combined,and unsupervised clustering analysis using K-means was performed to obtain subtypes.Gene differential analysis,PCA analysis,survival analysis,GSEA analysis,and ss GSEA analysis were performed on the subtypes.The TCGA-STAD data and GEO-GSE84437 data were merged and randomly divided into training and validation sets in a 1:1 ratio.Lasso regression and multivariate Cox regression were used to further screen prognostic-related differentially expressed genes and construct a prognostic risk scoring model based on the training set data.The samples were divided into high-risk and low-risk groups based on the median risk score,and the model was validated using the validation set data.The predictive performance of the model was evaluated through survival analysis,receiver operating characteristic(ROC)curve analysis,and independent prognostic analysis,and a forest plot was constructed.Differential gene analysis and pathway enrichment analysis were performed on the high-risk and low-risk groups to observe which gene pathways were enriched with differential genes.The differences in immune cells and immune functions between the high-risk and low-risk groups were evaluated through tumor mutation burden analysis and ss GSEA analysis,and the association between the model and tumor microenvironment and immunotherapy was predicted.The expression of key genes in the model in gastric cancer tissues and adjacent tissues was validated using external data,clinical samples,and immunohistochemical experiments.Results:1.A total of 12 prognostic-related differentially expressed genes in gastric cancer were screened.Among them,11 genes were related to ferroptosis,including NOX4,GLS2,MYB,PRKAA2,SLC7A11,ZFP36,DUSP1,ANGPTL7,TSC22D3,ATP6V1G2,and NNMT,and 1 gene was related to cuproptosis,which was GCSH.2.Based on the 12 prognostic-related differentially expressed genes,gastric cancer patients were divided into four subtypes through unsupervised clustering analysis.Subtype B exhibited high expression of genes such as GCSH,NOX4,PRKAA2,ZFP36,DUSP1,ANGPTL7,TSC22D3,ATP6V1G2,and NNMT,and low expression of genes such as GLS2,MYB,and SLC7A11.It had poorer clinical features and shorter survival compared to other subtypes.Subtype B also showed differences in immune cell infiltration and immune function compared to other subtypes.3.A prognostic risk scoring model for gastric cancer was constructed using LassoCox regression.The model consisted of four ferroptosis-related genes(NOX4,GLS2,MYB,and NNMT)and one cuproptosis-related gene(GCSH).The area under the ROC curve(AUC)for 1,3,and 5-year survival predictions was 0.641,0.658,and 0.667,respectively.A forest plot incorporating age,T stage,N stage,and risk score was generated based on univariate and multivariate independent prognostic analyses to predict the 1,3,and 5-year survival rates of patients.4.Pathway enrichment analysis of the high-risk and low-risk groups revealed significant differences in the extracellular matrix pathway.The high-risk group showed lower tumor mutation burden,higher TIDE value,and lower likelihood of benefiting from immunotherapy compared to the low-risk group.Significant differences were observed in immune cell infiltration,immune function,and tumor microenvironment scores,indicating higher tumor purity and poorer prognosis in the high-risk group.5.Immunohistochemical experiments validated the high expression of GCSH in gastric cancer tissues and low expression in adjacent tissues,as well as the low expression of GLS2 in gastric cancer tissues and high expression in adjacent tissues,consistent with the results of the prediction model.Conclusion:1.Genes related to ferroptosis and cuproptosis are closely associated with the prognosis of gastric cancer patients.The identified 12 potential tumor biomarkers can be used to predict the prognosis of gastric cancer patients and provide potential targets for diagnosis,immunotherapy,and prognosis monitoring.2.We have established a gastric cancer prognosis risk model consisting of five genes that accurately predicts the prognosis and immunotherapy response of gastric cancer patients,guiding personalized treatment.3.Genes related to ferroptosis and cuproptosis may affect the occurrence and development of gastric cancer and patient prognosis by modulating the function of extracellular matrix and immune cells.These findings reveal important regulatory pathways in the pathogenesis of gastric cancer and provide a theoretical basis for understanding the molecular mechanisms of gastric cancer and developing new treatment strategies. |
PDF Full Text Request