基于转移性淋巴结比的胃癌患者根治术后预后模型建立与验证

Establishment and validation of prognostic nomograms based on metastatic lymph node ratio in patients with radical resected gastric cancer

目的旨在基于转移性淋巴结比评估无远处转移胃癌患者接受D2根治术后的总体生存情况并建立列线图模型。方法回顾性分析2014年6月-2020年2月在兰州大学第二医院行D2根治术的Ⅰ~Ⅲ期胃癌患者的术后临床病理资料,按8∶2将其随机分为训练队列和验证队列。将单因素Cox回归分析差异有统计学意义的变量再纳入多因素Cox回归分析,筛选出生存预后的独立风险因素绘制列线图,并通过计算一致性指数,绘制校准曲线、受试者操作特征曲线验证模型。结果共纳入497例患者,训练队列398例,验证队列99例。单因素Cox回归分析显示发病年龄、浸润深度、转移性淋巴结比、脉管内瘤栓、肿瘤大小、神经浸润、淋巴管或血管浸润对胃癌患者术后总生存期的影响差异有统计学意义。多因素Cox回归分析显示,影响患者预后的独立因素有发病年龄、原发肿瘤浸润深度、转移性淋巴结比、脉管内瘤栓、淋巴管或血管浸润。训练队列的一致性指数为0.806(95%CI:[0.749,0.864]),验证队列的一致性指数为0.902(95%CI:[0.813,0.991])。受试者操作特征曲线、校准曲线均显示该胃癌预后模型有良好性能。结论转移性淋巴结比是影响患者预后的独立因素,基于转移性淋巴结比的列线图模型可以较好地预测胃癌患者术后的总体生存情况。

Objective This study aimed at evaluating the overall survival of patients with gastric cancer without distant metastases after D2 radical resection based on the ratio of metastatic lymph nodes and to establish a nomogram model.Methods The postoperative clinicopathological data of patients with stageⅠ-Ⅲgastric cancer who underwent D2 radical gastrectomy in The Second Hospital of Lanzhou University from June 2014 to February 2020 were retrospectively analyzed.The patients were randomly divided into the training cohort and the validation cohort according to the ratio of 8∶2.The variables with statistical significance in a univariate Cox regression analysis were included in a multivariate Cox regression analysis to screen out independent risk factors for survival prognosis and to draw a nomogram.The model was verified by calculating the consistency index,drawing the calibration curve and the receiver operating characteristic curve.Results Ultimately,a total of 497 patients were included,who were randomly assigned to the training cohort(n=398)and validation cohort(n=99).A univariate Cox regression analysis showed that the significant differences affecting the prognosis of patients were age of onset,depth of invasion,metastatic lymph node ratio,intravascular tumor thrombus,tumor size,nerve invasion,and lymphatic or vascular invasion.A multivariate Cox regression analysis showed that the independent factors affecting the prognosis of patients were age of onset,depth of primary tumor invasion,metastatic lymph node ratio,intravascular tumor thrombus,lymphatic or vascular invasion.The consistency index of the training cohort was 0.806(95%CI:[0.749,0.864]),and that of the validation cohort was 0.902(95%CI:[0.813,0.991]).The receiver operating characteristic curve and calibration curve showed that the gastric cancer prognostic model had a good performance.Conclusions Metastatic lymph node ratio is an independent factor affecting the prognosis of patients.The nomogram model based on the metastatic lymph node ratio can better predict the overall survival of patients with gastric cancer after surgery.