构建基于凋亡相关基因的膀胱癌预后模型

Construction of prognostic model of bladder cancer based on apoptosis-related genes

目的细胞凋亡在膀胱癌(BC)发生、发展以及化疗等生物学进程中发挥着重要作用。文中通过分析差异表达的凋亡相关基因(ARGs)以及临床特征,构建预测BC患者预后的模型。方法分析肿瘤基因组图谱(TCGA)数据库中161个ARGs,筛选出差异表达基因。通过GO和KEGG分析差异表达ARGs在BC中潜在的分子机制。Cox回归分析ARGs在预后的价值,构建评价患者的以后风险模型。使用Kaplan-Meier(KM)曲线和受试者工作特征(ROC)曲线以确定模型的预后价值。结果与正常膀胱组织相比,BC组织中41个差异表达ARGs。GO和KEGG分析发现差异表达ARGs与铂类耐药、细胞凋亡、细胞周期等生物学进程有关,同时也与P53等一些肿瘤发生经典途径调控信号通路相关。多因素Cox回归分析发现4个基因(AIFM3、TAP1、CASP6、EMP1)与BC患者生存和预后密切相关。基于人类蛋白质图谱网站数据库验证指出4个基因在肿瘤组织和癌旁正常组织之间也存在差异性表达,并且与患者预后密切相关。KM分析显示高风险评分组的预后明显差于低风险评分组(P=6.002e-06<0.0001)。风险评分与相应的临床变量相结合,构建预测患者生存率的诺模图,预测C指数为0.694,预测1年、3年和5年BC生存率的ROC曲线下面积分别为0.652、0.648和0.684。结论基于ARGs的预后模型可以用于BC患者的预后预测。

Objective Apoptosis plays an important role in the occurrence,development and chemotherapy of bladder cancer(BC).A model is constructed for predicting the prognosis of patients with BC by differentially expressed apoptosis-related genes(ARGs)and clinical characteristics.Methods The differentially expressed genes were screened by analyzing 161 ARGs in the Cancer Genome Atlas(TCGA).GO and KEGG were used to analyse the potential molecular mechanism of differentially expressed ARGs in BC.The prognostic value of ARGs was analysed by Cox regression and a future risk model was constructed to evaluate patients.Kaplan-Meier(KM)curve and receiver operating characteristic(ROC)curve were used to determine the prognostic value of the model.Results Compared with normal bladder tissues,41 ARGs were differentially expressed in BC tissues.GO and KEGG analysis showed that the differential expression of ARGs were related to the platinum drug resistance,apoptosis and cell cycle,as well as to some classical signal pathways regulating tumorigenesis,such as p53 pathway.Multivariate Cox regression analysis showed that four genes(AIFM3,TAP1,CASP6,EMP1)were closely related to the survival and prognosis of BC patients.The verification based on the database of Human Protein Atlas website showed that there were differential expressions of the four gene between tumor tissues and normal tissues adjacent to cancer,and were closely related to the prognosis of patients.KM analysis showed that the prognosis of the high-risk group was significantly worse than that of the low-risk group(P=6.002e-06<0.0001).Combining the risk score with the corresponding clinical variables,a nomogram was constructed to predict the survival rate of patients.The predicted C index was 0.694.The areas under the ROC curve for predicting 1-,3-and 5-year BC survival rates were 0.652,0.648 and 0.684 respectively.Conclusion The prognostic model based on ARGs could be used to predict the prognosis of patients with BC.