基于多组学数据构建肺腺癌预后相关风险预测模型
作者:
作者单位:

作者简介:

通讯作者:

中图分类号:

基金项目:

国家重点研发计划(2018YFC1315000,2017YFC0907900,2016YFC1302500,2016YFC0905400);国家自然科学基 金(81871885,81673265);中央保健专项资金(W2017BJ39);北京协和医学院协和青年科研基金(2017320013);北京市科技计 划首都临床特色应用研究(Z181100001718212),中国医学科学院医学与健康科技创新工程(2017?I2M?1?005);中国医学科学 院肿瘤医院院所科研课题(LC2017A01,LC2017D01);北京市优秀人才培养资助(2017000021223TD05)


Prediction model of prognosis risk of lung adenocarcinoma based on multi genomic data
Author:
Affiliation:

Fund Project:

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
  • |
  • 文章评论
    摘要:

    目的:通过整合肺腺癌(lung adenocarcinoma,LUAD)临床信息组、基因组和转录组的多组学信息,构建预后相关风险预测模型,为肺腺癌患者的预后风险预测提供依据。方法:通过肿瘤基因图谱(the cancer genome atlas,TCGA)网站下载522例肺腺癌患者数据,完成肺腺癌患者多组学数据库构建。对经过质量控制的数据库,采用Cox比例风险模型评价多组学数据对LUAD预后的影响,构建基于临床信息、基因分型和基因表达多组学信息的预后风险预测模型,并使用受试者工作特征(receiver operating characteristic,ROC)曲线和曲线下面积(area under curve,AUC)评估预后模型风险识别的能力。结果:共386例肺腺癌患者纳入多组学数据库,分别筛选出1个临床信息组、5个基因组单核苷酸多态性(single nucleotide polymorphisms,SNP)和4个转录组mRNA预后相关特征因素与肺腺癌的预后相关。与肺腺癌低风险组相比,高风险组患者较低风险组的预后更差(HR=3.67,95%CI=2.46~5.49,P=1.96×10-10),联合基因组、转录组与临床信息构建多组学预后指数后,其ROC曲线下面积为0.694(95%CI=0.633~0.754)。结论:采用联合临床信息组、基因组和转录组特征的多组学信息构建的预后风险预测指数可有效区分肺腺癌患者预后好坏,其研究结果将为肺腺癌患者的精准治疗提供有效支撑。

    Abstract:

    Objective:By integrating the clinical information group,genome and transcriptome of lung adenocarcinoma,a prognostic risk prediction model was constructed to provide evidence of lung adenocarcinoma patients. Methods:The data of 522 patients with lung adenocarcinoma were downloaded from TCGA website,and the multi group database of lung adenocarcinoma was completed.Cox proportional hazard model was used to evaluate the effect of multigroup data on the prognosis of LUAD. A prognostic risk prediction model based on clinical information,genotyping and multigroup information of gene expression was constructed. ROC curves and AUC was used to assess the ability of risk recognition in prognostic models. Results:A total of 386 patients with lung adenocarcinoma were included in the multi group database. One clinical information group,five genomes and four transcription groups for prognostic factors were detected to associate with the prognosis of lung adenocarcinoma.HR(95%CI)of prognosis of lung adenocarcinoma for high-risk group was 3.67(95% CI=2.46~5.49),as compared with low-risk group. After combining genome,transcriptome and clinical information,the ROC curve is 0.694(95%CI=0.633~0.754). Conclusion:Prognostic risk prediction index combining clinical information group,genome and transcriptome features can effectively distinguish the prognosis of lung adenocarcinoma patients. The results will provide effective support for accurate treatment of lung adenocarcinoma patients.

    参考文献
    相似文献
    引证文献
引用本文

李 鑫,魏锣沛,吕章艳,冯小双,李 江,陈宏达,温 艳,高禹舜,薛 奇,高树庚,谭锋维.基于多组学数据构建肺腺癌预后相关风险预测模型[J].南京医科大学学报(自然科学版),2018,(12):1816-1822

复制
分享
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:2018-09-03
  • 最后修改日期:
  • 录用日期:
  • 在线发布日期: 2019-01-02
  • 出版日期:
通知关闭
郑重声明