基于转移淋巴结比例的N3期胃癌新分期系统的建立  

Construction of a new staging system for stage N3 gastric cancer based on the metastatic lymph node ratio

作　　者：张红雨[1] 刘光辉[1] 叶延伟[1] 赵春临[1] 符洋[1] Zhang Hongyu;Liu Guanghui;Ye Yanwei;Zhao Chunlin;Fu Yang(Department of Gastrointestinal Surgery,the First Affiliated Hospital of Zhengzhou University,Zhengzhou 450001,China)

机构地区：[1]郑州大学第一附属医院胃肠外科,郑州450001

出　　处：《中华普通外科杂志》2025年第2期123-130,共8页Chinese Journal of General Surgery

摘　　要：目的探索N3期胃癌转移淋巴结比例阈值并建立新的TNM分期系统以判断预后。方法收集2004年1月至2020年12月SEER数据库4291例N3期胃癌患者和2016年1月至2020年12月郑州大学第一附属医院567例N3期胃癌患者的临床资料,基于最佳转移淋巴结比例阈值建立新的TNrM分期及列线图模型并与第8版TNM分期模型对比预后区分度、预测准确性及临床实用价值。结果0.5为最佳转移淋巴结比例阈值。根据转移淋巴结比例的Nr分期结合T分期建立TNrM分期,单因素和多因素COX回归分析显示TNrM分期是影响预后的显著性因素(均P<0.05)。在训练集中构建基于TNM分期和TNrM分期的列线图模型,并在验证集中进行外部验证。与TNM分期模型相比,TNrM分期模型的C指数和时间依赖的ROC曲线下面积(AUC)更大(训练集3年AUC∶66.5比74.4,5年AUC∶68.9比75.3;验证集3年AUC∶62.3比73.1,5年AUC∶62.6比75.8),在校准曲线中其总生存预测曲线更接近理想曲线,在决策曲线中其临床净获益更多。结论转移淋巴结比例超过0.5的N3期胃癌患者预后较差,基于转移淋巴结比例建立的TNrM分期列线图模型较第8版TNM分期列线图模型具有更好的预后区分能力和预测准确性以及更多的临床净获益。Objective:To explore the cut-off value of metastatic lymph node ratio(LNR)for stage N3 gastric cancer and construct a new TNM staging system to predict prognosis.Methods:Clinical data of 4291 patients from Jan 2004 to Dec 2020 in the SEER database and 567 patients from Jan 2016 to Dec 2020 in the First Affiliated Hospital of Zhengzhou University with stage N3 gastric cancer were collected.A new TNrM staging system and a nomogram model were constructed based on the optimal LNR cut-off value and compared with the 8th TNM staging model in terms of prognostic discrimination,prognostic prediction accuracy,and clinical usefulness.Results:The optimal cut-off value of LNR was 0.5.A TNrM staging system was constructed by combining the Nr stage with the T stage.Univariate and multivariate COX regression analyses showed that the TNrM staging system was a significant prognostic factor(all P<0.05).Based on the 8th TNM and TNrM staging system,two nomograms were constructed in the training set and externally validated in the validation set.Compared with the TNM staging model,the TNrM staging model had a larger C-index and area of time-dependent ROC curve(AUC)(training set:3-year AUC:66.5 vs.74.4,5-year AUC:68.9 vs.75.3;validation set:3-year AUC:62.3 vs.73.1,5-year AUC:62.6 vs.75.8),its overall survival prediction curves were closer to the ideal curve in the calibration curve,and its clinical net benefit was greater in the decision curves.Conclusions:Stage N3 gastric cancer patients with a metastatic lymph node ratio>0.5 have a poor prognosis.The TNrM staging nomogram model constructed based on the lymph node ratio has better prognostic discrimination ability and prediction accuracy and more clinical net benefits compared to the 8th edition of TNM staging nomogram model.

关 键 词：胃肿瘤 淋巴转移 肿瘤分期 预后 

分 类 号：R735.2[医药卫生—肿瘤]