基于SEER数据库的原发性纵隔及肺部软组织肉瘤相关生存及预后分析

Survival and Prognosis of Primary Mediastinal and Pulmonary Sarcoma Based on SEER Database

目的利用SEER数据库分析影响原发性纵隔及肺部软组织肉瘤预后的相关因素。方法收集SEER数据库376例患者数据,随机分为训练集(263例)与验证集(113例)。使用Kaplan-Meier法和Cox比例风险回归分析各变量与患者生存及预后的关系,建立列线图预测患者总生存期。采用校准曲线、一致性指数及ROC曲线评价列线图性能。结果组织学类型、手术、化疗、肿瘤大小、肿瘤分期是影响原发性纵隔及肺部软组织肉瘤预后的因素。建立的列线图模型可以预测其6个月、1年及2年总生存率,校准曲线显示与实测值基本一致,训练集与验证集C指数分别为0.754与0.745,ROC的曲线下面积分别为0.849与0.924,说明具有良好的预测准确度。结论本研究建立的列线图可以预测原发性纵隔及肺部软组织肉瘤患者6个月、1年及2年总生存率。

Objective To analyze the factors affecting the prognosis of soft tissue sarcomas originating from the mediastinum and lung using relevant data from the SEER database.Methods The data of 376 patients were collected from the SEER database,and were randomly divided into the train set(n=263)and validation set(n=113).The relationship between each variable and patient survival and prognosis was analyzed using the Kaplan-Meier method and Cox proportional risk regression to establish a nomogram,to predict the overall survival of patients.The calibration curves,consistency index,and ROC curves were used to evaluate the performance of the nomogram.Results Histological type,surgery,chemotherapy,tumor size,and tumor stage were the factors affecting the prognosis of primary mediastinal and pulmonary soft tissue sarcomas.The established nomogram could predict the 6-month,1-year,and 2-year overall survival of patients,and the calibration curves showed good prediction accuracy with measured values.C index of the train set and validation set were 0.754 and 0.745,respectively.The areas under the curve of ROC were 0.849 and 0.924.Conclusion The nomogram established in this study can predict 6-month,1-year,and 2-year overall survival in patients with primary mediastinal and pulmonary soft tissue sarcoma.