基于乳酸代谢基因的结肠腺癌风险模型构建及其预后价值

Construction of a risk model for colon adenocarcinoma based on lactate metabolism genes

目的基于乳酸代谢的差异表达基因(DEGs)构建结肠腺癌(COAD)的风险模型,并评价其预后价值。方法自GeneCards数据库获取乳酸代谢相关基因信息,自癌症基因组图谱(TCGA)获取COAD转录组数据及临床信息。采用“limma”包筛选乳酸代谢DEGs,进行单因素Cox回归及Lasso回归构建风险模型,通过Kaplan-Meier曲线及受试者工作特征(ROC)曲线验证风险模型的预测能力。对临床性状和风险模型进行单因素、多因素Cox分析,独立的预后因素用于构建列线图模型。分析风险模型与免疫细胞浸润程度的相关性。对高、低风险组进行差异性分析获取风险DEGs,用于基因本体(GO)及京都基因组基因百科全书(KEGG)富集分析。结果成功构建包含17个乳酸代谢DEGs的风险模型。高风险组较低风险组的总生存期(OS)及无进展生存期(PFS)明显缩短,均P<0.05。风险模型的ROC曲线下面积(AUC)>0.60。年龄(HR=1.031,95%CI为1.010~1.053,P=0.004)、临床分期(HR=2.193,95%CI为1.674~2.873,P<0.001)和风险模型(HR=1.959,95%CI为1.333~2.877,P<0.001)均是独立的预后因素,列线图模型的AUC值为0.758。风险模型与免疫细胞(质细胞、未活化的CD4记忆性T细胞、未活化的自然杀伤细胞、M0巨噬细胞、活化及未活化的树突状细胞、嗜酸性粒细胞)浸润程度相关,均P<0.05。GO功能富集分析结果显示,共有397个风险DEGs(65个下调、332个上调),主要集中于细胞外基质组织及结构组织。KEGG通路分析前3位的依次为局灶性粘连、PI3K-Akt信号通路、蛋白消化与吸收。结论风险模型是COAD的独立预后因素,可以评估COAD预后和免疫细胞的浸润程度,为COAD预后提供潜在的预测标志物。

Objective A risk model of colon adenocarcinoma(COAD)was constructed based on differentially expressed genes(DEGs)of lactate metabolism,and its prognostic value was evaluated.Methods Lactate metabolism-related gene information was obtained from GeneCards,and COAD transcriptome data and clinical information were obtained from the cancer genome atlas(TCGA).The"limma"package was used to screen DEGs of lactate metabolism,and the risk model was constructed by single factor Cox regression and Lasso regression,and the predictive ability of the risk model was verified by Kaplan-Meier curve and receiver operating characteristic(ROC)curve.Univariate and multivariate Cox analyses were performed on clinical traits and risk models,and independent prognostic factors were used to construct nomogram models.The correlation between the risk model and the degree of immune cell infiltration was analyzed.Differential analysis was performed on the high and low risk groups to obtain risk DEGs,which were used for enrichment analysis of gene ontology(GO)and kyoto encyclopedia of genomes(KEGG).Results A risk model including 17lactic acid metabolism DEGs was successfully constructed.The overall survival(OS)and progression-free survival(PFS)of the high-risk group were significantly shortened,compared to the low-risk group(P<0.05),and the area under the ROC curve(AUC)of the risk model was>0.60.Age(HR=1.031,95%CI:1.010-1.053,P=0.004),clinical stage(HR=2.193,95%CI:1.674-2.873,P<0.001)and risk model(HR=1.959,95%CI:1.333-2.877,P<0.001)were independent prognostic factors,and the AUC of the nomogram model was 0.758.Risk score and immune cells(plasma cells,non-activated CD4 memory T cells,non-activated natural killer cells,M0macrophages,activated and non-activated dendritic cells,eosinophils,all P<0.05)significantly correlated with the degree of infiltration.The GO functional enrichment analysis results showed there were a total of 397risk DEGs(65down regulated and 332up-regulated),mainly concentrated in extracellular matrix and structural tissues.The top three pathways in KEGG pathway analysis were focal adhesion,PI3KAkt signaling pathway,protein digestion,and absorption.Conclusionﾟ>@ﾟ>€ﾟRisk model is an independent prognostic factor for COAD,which can evaluate the prognosis of COAD and the degree of immune cell infiltration,and provide potential predictive markers for the prognosis of COAD.