局部晚期喉癌手术联合术后放疗预后模型的构建与验证

Construction and validation of a prognostic model for locally advanced laryngeal cancer surgery combined with postoperative radiotherapy

目的探讨手术联合术后放疗的局部晚期喉癌患者预后的危险因素,构建患者预后列线图预测模型并验证。方法回顾性分析2016年1月—2020年1月徐州医科大学附属医院放疗科收治的206例患者临床资料,按7∶3将患者随机分为训练集和验证集,在训练集确定纤维蛋白原/白蛋白比值(FAR)、全身免疫炎症指数(SII)的最佳截断值,统计分析独立危险因素,构建基于FAR、SII及其他危险因素的列线图并验证。结果训练集中依靠受试者工作特征(ROC)曲线确定术前FAR、SII预测生存的最佳截断值分别为6.62和774。最终纳入模型的变量为肿瘤分化程度、淋巴结转移、FAR和SII(均P<0.05)。训练集和验证集预测总生存期的C指数为0.878、0.836,时间依赖ROC曲线预测死亡的曲线下面积均高于0.8,校准曲线示预测与实际情况的拟合度较好,临床决策曲线示大部分患者能临床获益。结论基于FAR和SII构建的列线图可以有效地预测局部晚期喉癌患者的总生存期。

Objective To explore the risk factors of surgery combined with postoperative radiotherapy for the prognosis of patients with locally advanced laryngeal cancer,and to construct and validate a nomogram prediction model for prognosis.Methods A total of 206 patients who were admitted to Department of Radiotherapy,the Affiliated Hospital of Xuzhou Medical University from January 2016 to January 2020 were enrolled and their clinical data were retrospectively analyzed.The patients were randomly divided into a training set and a validation set(at a ratio of 7∶3).The optimal cut-off values of fibrinogen to albumin ratio(FAR)and systemic immune-inflammation index(SII)were determined in the training set.The independent risk factors were analyzed,and a nomogram based on FAR,SII,and other risk factors was constructed.Results In the training set,according to the receiver operating characteristics(ROC)curves,the optimal cut-off value of preoperative FAR and SII in predicting survival was 6.62 and 774,respectively.The final variables included in the model were the degree of tumor differentiation,lymph node metastasis,FAR,and SII(all P<0.05).The C-index of the training and validation sets for predicting overall survival was 0.878 and 0.836,respectively,and the area under the curve of the time-dependent ROC curve for predicting death was higher than 0.8.The calibration curves showed that the predictions were well fitted to the actual situation,and the clinical decision curves demonstrated that the majority of the patients achieved clinical benefits.Conclusions The nomogram based on FAR and SII can effectively predict the overall survival of patients with locally advanced laryngeal cancer.