Preliminary Establishment Of Prognostic Risk Model Of CGAS-STING Signaling Pathway-related Genes In Lung Adenocarcinoma

ObjectiveThe cyclic guanosine phosphate-adenosine monophosphate synthase-interferon gene stimulating factor(cGAS-STING)signaling pathway is a newly identified immune signaling pathway that triggers the body’s innate immune response to disease and exerts anti-tumor immune effects.The aim of this study was to construct a prognostic risk model of lung adenocarcinoma based on genes related to the cGAS-STING pathway,and to investigate the relationship between genes related to the cGAS-STING signaling pathway and the prognosis of lung adenocarcinoma and tumor immune microenvironment.Method1、Download lung adenocarcinoma RNA-Sep gene expression data and clinical data from TCGA(the cancer genome atlas)and GEO(Gene Expression Omnibus)databases.Access cGAS-STING signaling pathway-related genes at the Gene Card website,a comprehensive database of human genes.2、GO(GO Enrichment Analysis)and KEGG(Kyoto Encyclopedia of Genes and Genomes)enrichment analysis verified the enrichment of cGAS-STING signaling pathway in lung adenocarcinoma and paraneoplastic tissues.3、Single-factor COX regression and survival curve analysis were performed to screen prognosis-related genes in lung adenocarcinoma;multi-factor COX regression was performed to determine the prognostic model of cGAS-STING signaling pathway-related genes and form the risk score(RS)formula.The RS was calculated for each patient,and the patient population was divided into two groups,including a high-risk group and a low-risk group,using the median as the cut-off.4、Time-dependent subject work characteristic(ROC)curves were plotted,AUC was calculated,and KM survival curves were plotted.5、Forest plots were drawn to compare the univariate and multifactorial regression results of the prognostic factors of risk score,age,gender,tumor TNM stage,history of radiotherapy,and smoking or not to verify their validity as independent prognostic markers.It was also validated in an external dataset;the accuracy of the constructed prognostic model was assessed using calibration curves and C-index indices.6、The CIBERSORT method was used to estimate 22 percent of transcriptome data of lung adenocarcinoma patients The proportion of immune cell infiltration.Result1、The 308 lung adenocarcinoma samples and 37 paracancer samples in TCGA and the corresponding clinical information were used as the training set to build the prognostic risk model.GSE30129(n=83)in GEO was used as the test set to validate the prognostic risk model.2 、 A prognostic risk model containing 8 genes,which are FEN1,TXNDC15,VDAC1,TRIM38,THEM6,BLM,TM9SF3 and XRCC5,was constructed.3、ROC curves were plotted for 1,3 and 5 years,with area under the curve AUCs of 0.766,0.738 and 0.685,respectively.4 、 Multifactorial Cox regression analysis showed that RS(HR=1.497 [1.346,1.665],P<0.001)was an independent risk factor in the training set.5、Validated in an external dataset(GSE30219),KM survival analysis showed a significantly better prognosis for patients in the low-risk group than in the high-risk group(P=0.002),with a 5-year survival AUC value of 0.813.Calibration curves for constructing column line plots and the C-index(C-index=0.754)both reflected that the model had good accuracy.6、In the tumor immune microenvironment,the proportion of unactivated memory CD4+ T cells infiltrated significantly higher in the low-risk group than in the high-risk group(P<0.01),whereas the proportion of activated CD4+ T cells infiltrated significantly higher in the high-risk group than in the low-risk group(P<0.01).Conclusion1、The prognostic model constructed by screening eight genes related to the cGAS-STING signaling pathway in this study has good accuracy.2、Differences in tumor immune microenvironment may be related to genes related to cGAS-STING signaling pathway,which is worthy of further clinical verification.