肺腺癌预后甲基化基因生物标志物鉴定

Identification of methylated gene biomarkers associated with prognosis of lung adenocarcinoma

目的寻找一种有效的甲基化基因生物标志物,用于肺腺癌的预后预测。方法从癌症基因组图谱(The Cancer Genome Atlas,TCGA)数据库下载219例肺腺癌和29例正常肺组织样本DNA甲基化数据及临床信息,采用Cox比例风险回归模型进行DNA甲基化生物标志物的鉴定,使用时间依赖性受试者工作特征(receiver operating characteristic,ROC)曲线评估该模型的性能,通过基因本体论(Gene Ontology,GO)进行功能注释,探讨DNA甲基化标志物的生物学功能。结果基因差异甲基化分析,鉴定了160个差异甲基化基因。通过Cox回归分析,筛选出生存相关的4个DNA甲基化基因(AF131215.8、AL021918.1、RP11.386G11.3和ZNF382)作为肺腺癌预后生物标志物。Kaplan-Meier和ROC分析显示,患者总生存期比较低风险组(1.97年)明显高于高风险组(0.54年),差异比较具有统计学意义(P<0.001),并且具有很好的预测能力(AUCSignature=0.793)。多因素Cox回归分析结果表明筛选出生存相关的生物标志物是独立预后因子。GO功能注释显示,这些基因主要参与转录激活活性、RNA聚合酶Ⅱ启动子转录、蛋白质异源二聚体活性和RNA聚合酶Ⅱ启动子转录的正调控等。结论鉴定出4个DNA甲基化基因作为预测肺腺癌预后的生物标志物并且是独立预后因子。

Objective Lung adenocarcinoma(LUAD)is one of the most serious tumors.Most LUAD patients are diagnosed as advanced stage and have poor prognosis.There is an urgent need for new and effective biomarkers to predict overall survival(OS)of LUAD patients.The purpose of this study was to search for an effective gene methylation biomarker to predict the prognosis of LUAD.Methods Two hundred and nineteen LUAD and 29 normal lung tissue DNA methylation profile data were downloaded from The Cancer Genome Atlas(TCGA)database.Univariate and multivariable Cox proportional hazards regression were applied to identify DNA methylated genes biomarker in the samples randomly assigned to the training group and the other samples as the test subset.The receiver operating characteristic(ROC)curve were used to evaluate the performance of the model.Gene Ontology(GO)annotation were done to explore the biological function of DNA methylated genes signature.Results In total 160 different methylation genes were identified in the 219 LUAD data.Finally,four methylation genes(AF131215.8,AL021918.1,RP11.386G11.3 and ZNF382)significantly associated with survival time were identified as biomarkers by univariate and multivariable Cox proportional hazards regression analysis.Kaplan-Meier and ROC analyses showed that the OS of low-risk patients(1.97 years)was longer than that of high-risk patients(0.54 years)(P<0.001),and the biomarkers had a powerful predictive ability(AUC Signature=0.793),indicated that these four methylation biomarkers can predict the prognosis of LUAD patients accurately.Multivariate Cox regression analysis verified that our biomarker was an independent prognostic factor,independent of clinical factors such as age,gender and TNM stage.GO functional notes showed that these genes were mainly involved in transcriptional activation activity,RNA polymeraseⅡpromoter transcription,protein heterodimer activity and positive regulation of RNA polymeraseⅡpromoter transcription.Conclusion We identified a prognosis biomarker for LUAD by combing the DNA methylation signatures of 4 methylation genes.