Prediction Clinical Prognosis And Immunotherapy Response Of Gastric Adenocarcinoma Based On Cuproptosis-Related LncRNAs Molecular Model

Objective: Gastric cancer is one of the most common malignant tumors in the world,while adenocarcinoma of stomach(STAD),derived from malignant transformation of gastric gland cells,is the most common histological type of gastric cancer,with an incidence of 95% of gastric cancer and the third highest death toll from gastric cancer in China.Although surgery,chemotherapy and other treatments have been shown to be effective against gastric adenocarcinoma,new cancer treatments still need to be explored.Cuproptosis is a newly discovered mode of cell death with a mechanism different from that of ferroptosis,pyroptosis and necroptosis.By screened and identified Long non-coding RNAs(lncRNAs)associated with cuproptosis,we constructed a prognostic assessment model.It is used to predict the prognosis survival period and tumor immunoinvasive state in patients with STAD,concurrently screened immunotherapy drug,this study provided basis model support for future clinical treatment and prognosis evaluation of gastric cancer.Methods: In this study,transcriptome data and clinical data files of STAD tumor samples(n=343)and normal tissues(n=30)were downloaded from The Cancer Genome Atlas(TCGA)database.Firstly,we distinguished m RNAs and lncRNAs expression data by Practical Extraction and Report Language(Perl),and further obtained lncRNAs associated with cuproptosis by co-expression analysis.The differences of cuproptosis-related genes and cuproptosis-related lncRNAs were analyzed.We further combined the expression data of cuproptosis-related lncRNAs with survival data,and screened cuproptosis-related lncRNAs by Cox regression and minimum absolute shrinkage method to construct the STAD risk model.Then the model was validated and evaluated by Kaplan-Meier analysis,univariate and multivariate Cox regression,C-index analysis,constructed nomogram,progression free survival analysis,and time-dependent receiver operating characteristic(ROC)curve.In addition,Gene Ontology(GO)and Kyoto Encyclopedia of Genes and Genomes(KEGG)were used to analyzed the lncRNAs differences genes in the model genomes jointly.The TME and immunotherapeutic drugs were further analyzed.To further distinguish the differences in immune microenvironment among different STAD cluster subtypes,we divided the entire STAD data into three clusters based on the risk score file of prognostic related lncRNAs,and conducted a series of survival analysis,t SNE cluster,immune microenvironment and immunotherapy drug analysis.Results: In this study,we successfully constructed a risk prognosis assessment model consisting of 7 lncRNAs associated with cuproptosis(HAGLR,LINC00963,AC008915.2,LINC01094,AL157371.2,LINC00571,AC129507.1).This model can better distinguish between high and low risk groups,and the high risk group has a higher level of tumor immune invasion and is more sensitive to immune checkpoint genes.Cluster analysis of STAD patients showed that different STAD clusters had different immunotherapy responses.1.we distinguish the transcriptome data m RNAs and lncRNAs,get 16876 lncRNAs,of which 995 lncRNAs associated with cuproptosis(correlation coefficient | R | > 0.4,P < 0.001).Then,the co-expression network of cuproptosis-related genes and lncRNAs was constructed,then the differences between cuproptosis genes and lncRNAs were analyzed,a total of 652 cuproptosis-related lncRNAs differentially expressed in normal and tumor samples(| Log2 fold change | > 1,P < 0.05),Among them,85 lncRNAs were down-regulated and 567 were up-regulated in STAD.Among56 genes in the co-expression network,44 genes were differentially expressed in STAD,among which 40 genes were up-regulated and 4 genes were down-regulated in STAD.2.We used univariate Cox regression,LASSO regression and multivariate Cox regression analysis successfully constructed a 7 lncRNAs prognostic model(HAGLR,LINC00963,AC008915.2,LINC01094,AL157371.2,LINC00571,AC129507.1).Riskscore=Exp(HAGLR)*(0.1814)+Exp(LINC00963)*(-0.6175)+Exp(AC008915.2)*(0.5707)+Exp(LINC01094)*(0.5049)+Exp(AL157371.2)*(0.9685)+Exp(LINC00571)*(-1.5388)+Exp(AC129507.1)*(0.8004).Among them,included 2 protective lncRNAs and 5 risk lncRNAs(P< 0.05).The 1-year,3-year and 5-year area under ROC curve(AUC)of the prognostic risk assessment model was 0.722,0.771 and0.843.The AUC were all greater than 0.7,showed that the model had moderate predictive power.3.The results of K-M survival analysis,tumor stage analysis and progression-free survival analysis in the high-low risk group proved that,as expected,the high-risk group had poor survival in train group and test group,riskscore had the better predictive power compared with other clinical indicators such as age and gender.GO and KEGG enrichment analysis of target genes of model lncRNAs was further conducted.Risk lncRNAs-related differential genes are mainly concentrated in extracellular matrix and neutrophil-related pathways,ECM promoted the pathogenesis of cancer by interfering with the communication between cancer and immune cells,and tumor-related neutrophils enhanced the proliferation of tumor cells by releasing neutrophil extranuclear traps,both suggest poor prognosis in high-risk patients.While Protective lncRNAs-related differential genes are mainly concentrated in nucleotide catabolism and exosome related pathways.At the same time,we analyzed the difference of TME based on the prognostic riskscore.The high risk group had higher levels of immune infiltration,and the immune checkpoint genes were highly expressed in the high risk group.The immune modulators Belatacept and Abatacept could acted on the CD86.4.Based on the riskscore,the tumor samples were divided into 3 clusters.The survival curves of Cluster1,Cluster2 and Cluster3 were significantly different,and Cluster1 had the worst prognosis.t SNE analysis could significantly distinguish the three clusters.Cluster1 had the highest ESTIMATE score,and the expression of immune checkpoint genes was relatively high in Cluster1.Through further analysis of drug and gene interaction,we screened out the immunosuppressants Belasipl and Abacipl acted on CD80 and CD86 genes.Conclusion: 1.In this study,a risk prognosis assessment model composed of 7lncRNAs associated with cuproptosis(HAGLR,LINC00963,AC008915.2,LINC01094,AL157371.2,LINC00571,AC129507.1)was successfully constructed.2.The model can better distinguish between high and low risk groups,and predict the prognosis of STAD patients.The enrichment pathways of risk and protective lncRNAs-related differential genes are different,and the level of tumor immune invasion is higher in the high risk group,and we selected the immunosuppressants Belasipl and Abacipl.3.Cluster analysis of STAD patients showed that different STAD cluster subtypes showed different immunotherapy responses.