基于SEER数据库低级别胶质瘤预后列线图的构建  

Construction of prognostic nomogram for low-grade glioma based on SEER database

作　　者：齐泽迪 屈一晨 张明达 胡昌晨 齐国章 陈胜利[1] QI Zedi;QU Yichen;ZHANG Mingda;HU Changchen;QI Guozhang;CHEN Shengli(Department of Neurosurgery,the Fifth Clinical Medical College of Shanxi Medical University,Shanxi Taiyuan 030012,China;Trigeminal Neuralgia Hospital of Anyang,Henan Anyang 455000,China)

机构地区：[1]山西医科大学第五临床医学院神经外科,山西太原030012 [2]安阳三叉神经医院,河南安阳455000

出　　处：《现代肿瘤医学》2023年第11期2031-2036,共6页Journal of Modern Oncology

基　　金：国家自然科学基金(编号:30901774);山西省人民医院省级专项配套经费科研项目(编号:sj20019003)。

摘　　要：目的:通过监测、流行病学及预后(surveillance,epidemiology,and end result,SEER)数据库开发列线图来分析低级别胶质瘤(low-grade glioma,LGG)患者的预后因素并且预测其生存率。方法:通过SEER数据库收集LGG患者5439例,并统计其人口统计学信息及临床特征。随机抽取其中1001例作为模型的内部验证集,并收集2010-2017年间就诊于山西省人民医院的LGG患者67例作为外部验证集。采用单因素、多因素Cox回归及Lasso回归分析LGG患者的独立危险因素,并考虑其临床效用性。将这些独立预测因素整合在一起,绘制预测LGG患者1年和3年生存率的列线图。通过内部验证集数据及外部验证集数据绘制ROC曲线和校准曲线图来评估列线图的性能。结果:纳入训练集患者4438例,内部验证集患者1001例,外部验证集患者67例。一般情况人群分布无显著统计学差异。通过单因素、多因素Cox回归及Lasso回归分析联合生存分析结果选择独立危险因素,纳入年龄、病理学分型、手术方式、肿瘤大小、婚姻状况、放化疗及发病部位为独立预测因素(P<0.001)。由上述7种因素构建预后预测模型,结果以列线图形式呈现。内部验证集验证列线图的ROC曲线下面积为0.841和0.804;外部验证集验证列线图的ROC曲线下面积为0.703和0.742,表明该模型的区分度与准确度较高。校准曲线显示其具有较好的一致性。结论:本列线图可用于预测LGG患者1年和3年生存率,并且拥有较高的临床价值,可以为LGG的个体化治疗提供参考。Objective:To establish a nomogram from SEER database to analyze prognostic factors and predict survival rate of patients with low-grade glioma.Methods:A total of 5439 patients with low-grade glioma were collected through SEER database,and their demographic information and clinical characteristics were analyzed.1001 cases were randomly selected as the internal validation set of the model,while 67 low-grade glioma patients treated in The Fifth Clinical Medical College of Shanxi Medical University from 2010 to 2017 were collected as the external validation set.Independent risk factors were analyzed by Cox regression analysis and Lasso regression analysis.These independent risk predictors were combined to exploit a nomogram predicting 1-year and 3-year survival rate for patients with low-grade glioma.The performance of nomogram was evaluated by ROC curve and calibration curve from internal validation set data and external validation set data.Results:A total of 4438 patients in the training set,1001 patients in the internal validation set and 67 patients in the external validation set were included.There was no significant difference in the distribution of general population.Age,pathological type,surgical method,tumor size,marital status,chemotherapy,radiotherapy and location were included as independent predictors(P<0.001).The prognostic prediction model was constructed by the above 7 factors,and the results were presented in the form of nomogram.The area under ROC curves of the nomogram verified by the internal validation set were 0.841 and 0.804.The area under the ROC curves of the nomogram verified by the external validation set were 0.703 and 0.742.It shows that the model has good discrimination and accuracy.The calibration curve shows that it has good consistency.Conclusion:This nomogram can be used to predict the overall survival of patients with low-grade glioma,and has high clinical value,which can provide a reference for individualized treatment of low-grade glioma.

关 键 词：低级别胶质瘤 列线图 总体生存率 SEER数据库 

分 类 号：R739.4[医药卫生—肿瘤]