Construction of a prognostic model and column chart for clear cell renal cell carcinoma based on vesicle-mediated transport-related genes
Objective To investigate the relationship between vesicle-mediated transport genes(VMTGs)and the prognosis of clear cell renal cell carcinoma(ccRCC),aiming to develop a prognostic risk scoring model.Methods We obtained complete clinical and transcriptomic data of 601 and 55 ccRCC cases from the cancer genome atlas(TCGA)and the gene expression omnibus(GEO)databases,respec-tively.Subsequently,722 VMTGs were downloaded from the Reactome database.The differentially ex-pressed genes(DEGs)analysis was performed using the"DESeq2"R package to identify DEGs between tumor and normal samples in TCGA.Intersection of these DEGs with VMTGs revealed 142 differentially ex-pressed VMTGs,including 96 upregulated and 46 downregulated genes.A total of 59 prognostic-related genes were selected through single-factor Cox analysis.Models were then constructed using Least Absolute Shrinkage and Selection Operator(LASSO)regression,stepwise regression,and multi-factor Cox analysis.Patients were stratified into high and low-risk groups based on the model scores,and Kaplan-Meier survival curves were utilized to compare their survival differences.Subsequently,the risk scoring model was valida-ted using the GEO dataset to analyze survival discrepancies between high and low-risk groups.The prognos-tic value of the risk scoring model was evaluated through univariate and multivariate Cox analysis.A col-umn chart was constructed by combining the risk model score with clinical features.Results The analysis of 7 prognosis-related VMTGs[Kinesin family member 18B(KIF18B),Sortilin 1(SORT1),Solute Carrier Family 18 Member A3(SLC18A3),Kinesin family member 13(KIF13B),Serum Amyloid A1(SAA1),Ankyrin 3(ANK3),and Gap Junction Protein Beta 1(GJB1)]was conducted to construct a risk scoring model.The prognosis of the low-risk group was superior to that of the high-risk group,with a statistically significant difference(x2=57.0,P＜0.01).The area under receiver operating characteristic(ROC)curve(AUC)at 1,3,and 5 years were 0.785,0.740,and 0.739,respectively,indicating good predictive per-formance of the risk scoring model.In the GEO cohort,the prognosis of the low-risk group was superior to that of the high-risk group,with a statistically significant difference(x2=5.6,P＜0.05),and the AUCs at 1,3,and 5 years(0.914,0.873,and 0.763,respectively)suggested a good prognostic performance.Both univariate[hazard ratio(HR)=2.718,95％confidence interval(CI)=2.253-3.280,P＜0.01]and multivariate Cox analysis(HR=2.100,95％CI=1.717-2.570,P＜0.01)results indicated that the risk scoring model was an independent prognostic factor.The AUCs at 1,3,and 5 years(0.869,0.812,and 0.783,respectively)showed good prognostic performance of nomagram.Calibration curves and Deci-sion Curve Analysis(DCA)demonstrated the good predictive accuracy of the nomagram.Conclusion The risk scoring model and column chart based on VMTGs can accurately predict the prognosis of ccRCC pa-tients and effectively stratify the patients.
NETL
NSTL
万方数据
