| Objective:Cuproptosis is a novel and independent copper-dependent programmed cell death that plays a key role in the occurrence and progression of various cancers.However,little is known about the effect of cuproptosis-related genes on hepatocellular carcinoma(HCC).The aim of this study was to explore the cuproptosis-related genes,identify the subtype of hepatocellular carcinoma,develop a novel prognostic model of hepatocellular carcinoma,and explore its relationship with tumor microenvironment infiltration.Methods:Based on The Cancer Genome Atlas(TCGA)database,we evaluated the expression,somatic mutation and copy number variation of 10 cuproptosis-related genes in hepatocellular carcinoma.Consensus clustering was used to identify new molecular subtypes related to copper death in HCC,and enrichment analysis was performed to determine their biological functions.Subsequently,differentially expressed genes between subtypes were identified,and candidate prognostic genes were obtained by univariate Cox regression analysis.Further,using the least absolute shrinkage and selection operator(Lasso)regression analysis to narrow the overfit,a gene risk scoring model for predicting overall survival was established.They were validated in the International Cancer Genome Consortium(ICGC)database.Univariate and multivariate Cox regression were used to determine whether the risk score could be used as an independent predictor.The risk score and clinical characteristics were included to draw a nomogram.The calibration curve,ROC curve,and DCA curve were used to evaluate the predictive value of the nomogram.In addition,the "Estimate" and "ss GSEA" algorithms were used to assess the infiltration of immune cells in the tumor immune microenvironment between high and low risk groups and the correlation with prognostic model genes.At the same time,associations between different risk groups and expression levels of commonly used chemotherapy agents and immune checkpoints were evaluated.Finally,the expression patterns of genes in the risk model were verified by Human Protein Atlas(HPA)database.Results:A total of 10 cuproptosis-related genes were included in this study.Compared with normal tissues,FDX1 was down-regulated in hepatocellular carcinoma tissues,and the remaining 9 genes were up-regulated.CDKN2 A had the highest mutation rate(3%).GO and KEGG analysis showed that copper death-related genes were mainly involved in the citric acid cycle and other pathways.Using the 10 differentially expressed genes,HCC was divided into two subtypes with different prognosis.The prognosis of Cluster2 was worse than Cluster1(P<0.05),and the T stage,TNM stage and tumor subtype were significantly positively correlated.A new risk prediction model was developed based on the LASSO regression coefficients of CBX2,G6 PD,SLC2A1,KPNA2,NEIL3,TAF3,TTK,KIAA1841 and UCK2.Overall survival(OS)was better in the low-risk group than in the high-risk group in both datasets(P<0.001).The 1,3 and 4-year AUC areas of TCGA dataset and ICGC external validation set were 0.788,0.720 and 0.722,respectively,and 0.778,0.782 and 0.776,respectively.At the same time,the constructed model was compared with the model constructed in previous studies,and the model of this study was superior to the previous models,which further illustrates the accuracy of the constructed model.High-risk patients have higher tumor histological grade,higher stage,and worse prognosis.Multivariate Cox regression analysis also confirmed that the risk score was an independent prognostic factor for HCC in TCGA and ICGC groups(P<0.001).In addition,the risk score and five variables of clinical characteristics(gender,age,TNM stage and histological grade)were included to construct a nomogram.The calibration curves showed excellent agreement between the nomogram predictions and the actual observed probabilities.The concordance index(CI)indicated that the CI of the model combined with clinicopathological features was higher than that of individual clinicopathological factors.ROC curve analysis showed that AUC values were higher than those when each of the clinical characteristic variables was used alone to predict OS,indicating the high accuracy of the new nomogram prediction model.The clinical decision curve showed that the clinical benefit rate of the nomogram constructed by combining the five variables was higher than that of each clinical variable.A significant difference was observed between the two risk groups in terms of the "Estimate" score,with the high-risk group having a lower matrix score than the low-risk group,and the risk score was negatively correlated with the matrix score.The results of the ss GSEA analysis revealed differences in seven immune cells and four immune-related functions between the high-risk and low-risk groups.The expression levels of immune checkpoint genes in the high-risk group were generally higher than those in the low-risk group.Treatment sensitivity of sorafenib,5-fluorouracil,doxorubicin,gemcitabine,sunitinib,and mitomycin was determined in patients in the high-risk group.Finally,in the HPA database,the protein expression of CBX2,G6 PD,KPNA2,TTK,and SLC2A1 in HCC tissues was detected to be higher than that in normal tissues,while the protein expression of KIAA1841 was lower.NEIL3,TAF3,and UCK2 are not recorded in the database.Conclusions:In this study,patients with hepatocellular carcinoma were classified based on copper death-related genes,and a new prognostic model was constructed,which can better predict the survival outcome of hepatocellular carcinoma,improve the understanding of tumor immune microenvironment,and is expected to help clinicians identify more aggressive tumors and initiate appropriate individualized therapy. |
PDF Full Text Request