基于CT征象联合血清肿瘤标志物的列线图对胃癌淋巴结转移风险的预测价值

Risk prediction of lymph node metastasis in gastric cancer based on the nomogram of CT features combined with serum tumor markers

目的:探讨基于CT征象联合血清肿瘤标志物的列线图对胃癌淋巴结转移风险的预测价值。方法:回顾性分析176例经手术病理证实的胃癌患者资料,按手术时间顺序分为训练集123例(淋巴结转移77例,无淋巴结转移46例)和验证集53例(淋巴结转移35例,无淋巴结转移18例)。分析血清肿瘤标志物和CT征象与淋巴结转移风险的相关性。采用多因素logistic回归分析建立CT征象联合血清肿瘤标志物的淋巴结转移风险预测模型,利用ROC曲线的AUC评价模型预测效能,并绘制列线图实现模型可视化。结果:糖类抗原72-4(CA72-4)、淋巴结大小、是否突破浆膜层及动脉期CT值是胃癌淋巴结转移的独立预测因子。联合模型在训练集中的AUC为0.822(95%CI 0.746~0.897),验证集AUC为0.805(95%CI 0.685~0.925)。校准曲线显示预测模型具备良好的拟合度,决策曲线表明模型具有较高的临床应用价值。结论:基于CT征象和血清肿瘤标志物构建的列线图可为术前预测胃癌淋巴结转移风险提供无创、可视化的方法。

Objective:To explore the value of a nomogram model based on CT features combined with serum tumor markers in the risk prediction of lymph node metastasis(LNM)in gastric cancer.Methods:The data of 176 patients with gastric cancer confirmed by surgery and pathology were retrospectively analyzed.They were divided into the training cohort(123 cases,77 cases with LNM and 46 cases without LNM)and the validation cohort(53 cases,35 cases with LNM and 18 cases without LNM).The correlation of serum tumor markers,CT features and LNM was analyzed.Multifactor logistic regression analysis was used to establish a prediction model of LNM risk based on CT features combined with serum tumor markers.ROC curves were used to evaluate the prediction efficiency of the model,and a nomogram was drawn to visualize the model.Results:CA72-4,lymphnode size,serosal penetration and CT value in the arterial phase were the independent predictors of LNM.The AUC of the combined model in the training cohort was 0.822(95%CI 0.746~0.897),that in the validation cohort was 0.805(95%CI 0.685~0.925).The calibration curve showed that the model had a good fitting degree,and the decision curve showed that the model had a high clinical value.Conclusions:The combined model based on CT features and serum tumor markers can provide a noninvasive visualization method for the preoperative risk prediction of LNM in gastric cancer patients.