目的 比较肝细胞癌(hepatocellular carcinoma,HCC)组织中与细胞焦亡相关基因的表达水平,建立对临床工作有用的预测模型.方法 从癌症基因组图谱(the cancer genome atlas,TCGA)数据库中随机选取HCC患者,并按1∶1比例分为训练组和验证组.然后,从有影响力的研究中检索到33个焦亡相关基因,并分析其表达水平.进行了单变量Cox分析和随后的最小绝对收缩和选择算子(the least absolute shrinkage and selection operator,LASSO)算法,对总体生存期(overall survival,OS)进行分析,以鉴定候选预后基因.最后,利用多变量Cox分析形成预后模型,可以将所有患者区分为高风险或低风险组.结果 5个与焦亡相关的基因(GPX4、CASP8、NLRP6、GSDMC和PLCG1)被纳入预测模型,风险评分的中位截断值可以将训练组和验证组患者分为高风险或低风险组.在训练组中发现低风险组的死亡概率低于高风险组,并在验证组中得到确认,差异有统计学意义(P<0.05).对OS的时间依赖性受试者工作特征曲线(receiver operating characteristic curve,ROC)进行了分析,发现在训练组中,1年、2年和3年时曲线下面积(area under curve,AUC)分别为0.775、0.713和0.677;在验证组中,1年、2年和3年时AUC分别为0.671、0.614和0.624.功能富集分析显示,碳水化合物和脂质代谢途径、巨噬细胞亚群以及免疫反应与风险模型相关联.结论GPX4、CASP8、NLRP6、GSDMC和PLCG1等5个与细胞焦亡相关的基因模型具有显著的HCC预后预测效果.包括细胞焦亡与细胞代谢、巨噬细胞和HCC免疫微环境中的免疫反应之间的相互作用在内的潜在机制,需要进一步探索.
The role of a pyroptosis-related gene-based model in the prognostic prediction of patients with hepatocellular carcinoma
Objective:Hepatocellular carcinoma(HCC)has high genetic heterogeneity,while the survival rate is poor.Recent studies have suggested that pyroptosis is closely connected to the development of tumours,including genesis,invasion and metastasis,but whether these genes are associated with pyroptosis could provide information for prognosis prediction in HCC is unknown.Methods:All HCC patients from The Cancer Genome Atlas(TCGA)database were divided randomly into a training group and a validation group at a 1:1 ratio.Then,we analysed the expression levels of 33 pyroptosis-related genes retrieved from influential studies.Univariate Cox analysis and the following LASSO algorithm of overall survival(OS)were conducted to identify candidate prognostic genes.Finally,multivariable Cox analysis was utilized to form a prognostic model that could distinguish all patients into high-risk or low-risk groups.Results:Five pyroptosis-related genes(GPX4,CASP8,NLRP6,GSDMC and PLCG1)were enrolled in the prediction model,while the median cut-off value of the risk score could stratify the training and validation group patients into the high-risk or low-risk group.A lower probability of death was found in the low-risk group than in the high-risk group in the training group and was confirmed in the validation group(all P<0.05).The time-dependent ROC curves for OS were analyzed,and the area under the curve(AUC)was found to be 0.775,0.713,and 0.677 in the training group and 0.671,0.614,and 0.624 in the validation group at 1,2,and 3 years,respectively.Functional enrichment analyses showed that carbohydrate and lipid metabolism pathways,macrophage subgroups and immune reactions were associated with the risk model.Conclusion:The five pyroptosis-related gene models have a significant prognostic prediction effect.The potential mechanism,including the interaction between pyroptosis and cell metabolism and macrophage and immune reactions in the HCC immune microenvironment needs further exploration.
NETL
NSTL
万方数据
