基于SEER数据库建立颅咽管瘤预后预测模型列线图  

Establishment of a nomogram model for predicting the prognosis of craniopharyngioma based on SEER database

作　　者：张雅婕 张啸虎 张硫悦 梁君 ZHANG Yajie;ZHANG Xiaohu;ZHANG Liuyue;LIANG Jun(Department of Neurosurgery,the Affiliated Hospital of Xuzhou Medical University,Xuzhou,Jiangsu 221002,China)

机构地区：[1]徐州医科大学附属医院神经外科,江苏徐州221002

出　　处：《徐州医科大学学报》2024年第11期830-836,共7页Journal of Xuzhou Medical University

摘　　要：目的利用SEER数据库分析颅咽管瘤患者预后风险因素。方法基于SEER数据库收集2000—2019年所有颅咽管瘤患者的临床数据,用Kaplan-Meier法绘制总生存曲线,用Log rank检验确定显著性。使用Cox比例风险模型进行单因素和多因素回归分析,确定潜在的预测因素。并利用该因素建立列线图预测颅咽管瘤患者1、2、5和10年的生存率。采用一致性指数(C指数)、受试者操作特征(ROC)曲线、曲线下面积(AUC)和校准曲线评估列线图的辨别能力。结果根据训练集多因素分析筛选变量并建立列线图,训练集和验证集C指数分别为0.77和0.75,训练集和验证集预测1、2、5、10年生存率的AUC均大于0.75。根据校准曲线得出预测生存率和实际生存率之间有良好的一致性。外部验证组1、2年的ROC曲线下面积分别为0.942和0.830,建立的列线图有良好的预测作用。结论是否放疗、肿瘤大小、年龄、种族、婚姻和家庭收入状况是颅咽管瘤患者预后的重要影响因素,根据这些因素建立的列线图具有可靠的辨别能力和准确性,对预测颅咽管瘤患者的预后有重要的临床意义。Objective To analyze the prognostic factors of patients with craniopharyngioma based on the Surveillance,Epidemiology,and End Results(SEER)database.Methods Clinical data were collected from patients who were diagnosed with craniopharyngioma between 2000 and 2019 from SEER database.The overall survival curves were plotted using the Kaplan-Meier method and significance was determined by Log rank test.Univariate and multivariate regression analyses were performed using the Cox proportional hazards model to identify potential predictors.Based on those factors,a nomogram model was established to predict 1-year,2-year,5-year and 10-year survival rates of craniopharyngiomas patients.The prediction efficiency was evaluated by the concordance index(C-index),receiver operating characteristic(ROC)curve,area under the curve(AUC)and calibration plots.Results The nomogram model was established through multivariate analysis using the training set.The C-index value of the nomogram model was 0.77 for the training set and 0.75 for the validation set.According to the training and validation sets,the AUC of 1-year,2-year,5-year and 10-year survival rates were greater than 0.75.The calibration curves also showed satisfactory consistency between predicted survival and observed outcomes.The areas under the ROC curve of the external verification group for 1 year and 2 years were 0.942 and 0.838.Therefore,the established nomogram model had good predictive effect.Conclusions The influencing factors for the prognosis of craniopharyngioma patients include radiotherapy,tumor size,age,race,marriage and family incomes.The resultant nomogram model has reliable discrimination and accuracy,and is important for prediction prognosis of craniopharyngioma patients.

关 键 词：颅咽管瘤 SEER数据库 COX回归 列线图 预后 

分 类 号：R739.4[医药卫生—肿瘤]