兰州大学学报（医学版）2024，Vol.50Issue(5)：41-48,55,9.DOI:10.13885/j.issn.1000-2812.2024.05.006
基于SEER数据库构建肝细胞癌患者预后预测新模式
Construction of a new prognostic model for hepatocellular carcinoma patients based on the SEER database
摘要
Abstract
Objective To construct a new predictive model to study the factors influencing the overall survival of patients with hepatocellular carcinoma,thereby providing theoretical support and reference for individual-ized and precise treatment plans for patients.Methods Data of patients diagnosed with hepatocellular carcino-ma from 2011 to 2015 were obtained from the SEER database.Univariate and multivariate Cox regression analysis was used to identify prognostic factors and develop prediction models.A validation cohort consisting of sample data from 280 hepatocellular carcinoma patients at the Second Hospital of Lanzhou University was collected for external validation.The model efficacy was evaluated via the consistency index,subject work characteristic curve,calibration curve and decision curve,which were compared with traditional TNM stag-ing.Risk stratification scores calculated and violin plots drawn.Result Regression analysis revealed that age,sex,N stage,M stage,alpha-fetoprotein expression,operation,radiotherapy and chemotherapy were indepen-dent prognostic factors affecting the overall survival of patients with hepatocellular carcinoma.The training group's concordance index was 0.739(95%CI:[0.730,0.748]),the area under curve for 1,3,and 5 years were 0.806,0.811,and 0.812 respectively,and the internal validation group's concordance index was 0.732(95%CI:[718,0.746])and the area under curve for 1,3,and 5 years were 0.787,0.803,and 0.810,respectively.The consistency index of the external verification group was 0.719(95%CI:[0.684,0.753])and the area under the curve for the 1,3,and 5years periods were 0.794,0.841,and 0.899 respectively.Additionally,the consistency index of the traditional TNM model was found to be 0.704(95%CI:[0.691,0.714])and the areas under the curve for years 1,3,and 5years were determined to be 0.716,0.687,and 0.677 correspondingly.The decision curve analysis curve revealed that the nomogram had a superior prediction efficiency than the traditional TNM staging.Conclusion The nomogram developed in this study can predict overall survival in hepatocellular carcinoma patients and outperform the traditional TNM staging.Moreover,the risk stratification system based on the new predictive model demonstrates a greater precision in distinguishing patients'survival periods,thereby having clinical reference value for the individualized treatment of hepatocellular carcinoma patients.关键词
肝细胞癌/总生存期/SEER数据库/预测模型/列线图/预后Key words
hepatocellular carcinoma/overall survival/surveillance,epidemiology and end results database/predictive model/nomogram/prognostic分类
医药卫生引用本文复制引用
王霄,杨波,马晨哲,汪营营,杨庆娥,李玉民..基于SEER数据库构建肝细胞癌患者预后预测新模式[J].兰州大学学报（医学版）,2024,50(5):41-48,55,9.基金项目
甘肃省重大科技资助项目(20ZD7FA003,22ZD6FA050,22JR9KA002) （20ZD7FA003,22ZD6FA050,22JR9KA002）
