用于预测低级别胶质瘤患者生存率的列线图模型构建  

Nomogram model construction for predicting survival rate of patients with low-grade glioma

作　　者：林明健 王文波[1] 夏学巍[1] 丁乙夫 廖红展 周配权 李文才 夏少怀 谭洁[2] LIN Mingjian;WANG Wenbo;XIA Xuewei;DING Yifu;LIAO Hongzhan;ZHOU Peiquan;LI Wencai;XIA Shaohuai;TAN Jie(Department of Neurosurgery,Affiliated Hospital of Guilin Medical University,Guilin,Guangxi 541001,China;Guangxi Key Laboratory of Brain and Cognitive Neuroscience,Guilin Medical University,Guilin,Guangxi 541001,China)

机构地区：[1]桂林医学院附属医院神经外科,广西桂林541001 [2]桂林医学院广西脑与认知神经科学重点实验室,广西桂林541001

出　　处：《重庆医学》2021年第13期2283-2288,共6页Chongqing medicine

基　　金：国家自然科学基金项目(81860449,81560413);回国基金项目(一般项目)(2016GXNSFCA380028);中青年教职工科研能力提升项目(桂医科[2018]5号);2017年度广西脑与认知神经外科学重点实验室开放课题(GKLBCN-20170105-04)。

摘　　要：目的构建列线图模型预测低级别胶质瘤(LGG)患者的生存率。方法从癌基因数据库(TCGA)数据库中下载患者临床数据,构建临床预测模型比较临床参数与LGG患者的预后关系。结果Kaplan-Meier生存分析结果显示诊断年龄>65岁、KPS<80分、肿瘤分级G3、分子亚型为IDH野生型、首次治疗后肿瘤有进展、化疗史与LGG患者中位生存时间降低相关(P<0.05),COX因素分析结果显示诊断年龄>65岁、肿瘤分级G3级、分子亚型为IDH野生型、首次治疗后肿瘤有进展是LGG患者的独立危险因素(P<0.05),并以相关危险因素绘制列线图,其ROC曲线下面积(AUC)分别为0.810、0.850、0.874,且预测生存率与实际生存率曲线较吻合,列线图校准度与区分度均较好。结论利用列线图可预测低级别胶质瘤患者生存率。Objective To construct the nomographic model to predict the survival rate of the patients with low-grade glioma(LGG).Methods The patients clinical data were downloaded from the TCGA database to construct the clinical prediction model for comparing the relationship between the clinical parameters with the prognosis of LGG patients.Results The Kaplan-Meier analysis results showed that the age>65 years old,KPS<80 points,tumor grade G3,molecular subtype IDH wild type,tumor progression after first treatment and history of chemotherapy were correlated with the decrease of the median survival time(OS)in the patients with LGG(P<0.05),the COX factor analysis results showed that the age>65 years old,tumor grade G3,molecular subtype IDH wild type,and the tumor progression after the first treatment were the independent risk factors for the patients with LGG(P<0.05),and the nomograph was drawn by using the related risk facors,the areas under their ROC curves were 0.810,0.850 and 0.874 respectively,moreover the predicted survival rate was in good agreement with the actual survival rate curve,the nomogram calibration and the discrimination degree of the nomogram were good.Conclusion The nomogram can be used to predict the survival rate of the patients with LGG.

关 键 词：胶质瘤 列线图 生存率 TCGA数据库 临床预测模型 

分 类 号：R739.41[医药卫生—肿瘤]