免疫治疗时代肺腺癌的预后相关因素分析及预后预测模型建立

Prognostic factors and prognostic modeling of lung adenocarcinoma in immunotherapy era

目的寻找肺腺癌预后相关的独立影响因素,并依据这些因素建立列线图。方法基于美国SEER数据库,按照入排标准筛选患者,COX回归结合LASSO回归筛选出来变量并绘制列线图。通过C指数(C-Index)和校准曲线评价列线图的预测能力,并通过临床决策曲线(Decision Curve Analysis,DCA)评价临床实用性。结果经过分析处理后我们最终确立9个预后独立影响因素,包括年龄、性别、人种、手术、放疗、化疗、婚姻状况、肿瘤大小和肿瘤数量。根据这些因素绘制列线图,C指数0.736,校准曲线斜率接近1,DCA曲线显示该列线图临床实用性良好。结论本研究建立的预测模型预测能力良好,能够快速便捷的预测患者1年生存率和3年生存率,并且临床实用性良好。

Objective To search for independent prognostic factors based on the Surveillance,Epidemiology,and EndRe⁃sults Database(SEER),and to establish a more convenient and accurate prognostic tool for lung adenocarcinoma.Methods Patients with lung adenocarcinoma were selected according to the U.S.SEER database.COX regression and LASSO regression analysis was performed to identify the independent prognostic factors.Screened variables were used to establish clinical prediction model,namely,to draw a line graph.The predictive power of the line graph was evaluated by c-index(C-index)and calibration Curve,and the clinical utility was evaluated by Decision Curve Analysis(DCA).Results Nine independent prognostic factors were identified.According to these factors,a nomogram the C-index was 0.736 and the slope of the calibration curve was close to 1.The DCA curve showed that the rosette had good clinical prac⁃ticability.Conclusion The prediction model established in this study has good prediction ability,can quickly and conven⁃iently predict the 1-year survival rate and 3-year survival rate of patients,and has good clinical practicability.