食管鳞状细胞癌患者生存预后列线图预测模型构建及验证  

Construction and verification of nomogram prediction model for survival prognosis of patients with esophageal squamous cell carcinoma

作　　者：郭嘉仪 宋正宇 张海平[1,2] 张力为[1,2] 伊地力斯·阿吾提[1,2] GUO Jiayi;SONG Zhengyu;ZHANG Haiping;ZHANG Liwei;EDRIS-Awut(Department of Thoracic Surgery,The First Affiliated Hospital of Xinjiang Medical University,Urumqi,830054,P.R.China;State Key Laboratory of Xinjiang Medical University,Urumqi,830054,P.R.China)

机构地区：[1]新疆医科大学第一附属医院胸外科,乌鲁木齐830054 [2]新疆医科大学国家重点实验室,乌鲁木齐830054

出　　处：《中国胸心血管外科临床杂志》2024年第5期718-727,共10页Chinese Journal of Clinical Thoracic and Cardiovascular Surgery

基　　金：国家自然科学基金(81960498);省部共建与国家重点实验室(SKL-HIDCA-2021-21)。

摘　　要：目的 探讨术前血清白蛋白与球蛋白比值(albumin-to-globulin ratio,AGR)及中性粒细胞与淋巴细胞比率(neutrophil-lymphocyte ratio,NLR)对食管鳞状细胞癌(鳞癌)患者总体生存(overall survival,OS)的预后价值,建立个体化的列线图模型并评价其效能,以期为食管癌患者的临床治疗及术后预后随访评估提供可能的评价依据。方法 回顾性分析2010—2017年新疆医科大学第一附属医院胸外科经病理诊断为食管鳞癌患者的AGR、NLR、临床病理及随访资料。分析NLR、AGR与各临床病理资料的相关性;采用Kaplan-Meier生存曲线及log-rank检验进行生存分析。通过X-tile软件确定AGR、NLR的最佳截断值并根据最佳截断值将患者分为高水平组和低水平组。同时经单因素和多因素Cox回归分析明确影响食管鳞癌患者OS的独立危险因素,并构建列线图预测模型同时经内部验证。通过受试者工作特征(receiver operating characteristic,ROC)曲线和校准曲线评估模型的诊断效能,通过决策曲线分析评估临床应用价值。结果 本研究共纳入150例患者,其中男105例、女45例,年龄(62.3±9.3)岁,随访时间1~5年。高水平AGR组患者的5年OS率明显高于低水平组(χ^(2)=6.339,P=0.012),两组中位OS分别为25个月、12.5个月。高水平NLR组患者5年OS率明显低于低水平组(χ^(2)=5.603,P=0.018),两组中位OS分别为18个月、39个月。多因素Cox回归分析提示,AGR、NLR、T分期、N分期、伴随淋巴结转移以及分化程度是食管鳞癌患者OS的独立危险因素。列线图模型经内部验证后C-index为0.689[95%CI(0.640,0.740)],预测1、3、5年OS率的ROC曲线下面积分别为0.773、0.724、0.725。同时校准曲线和决策曲线提示模型在生存预后的预测上有一定的效能。结论 术前AGR、NLR是食管鳞癌患者的独立危险因素,高水平AGR和低水平NLR可能与较长OS相关;基于AGR、NLR及临床病理特征的列线图模型,可作为预测食管�Objective To investigate the prognostic value of preoperative serum albumin-to-globulin ratio(AGR)and neutrophil-lymphocyte ratio(NLR)in the overall survival(OS)of patients with esophageal squamous cell carcinoma(ESCC),and to establish an individualized nomogram model and evaluate its efficacy,in order to provide a possible evaluation basis for the clinical treatment and postoperative follow-up of ESCC patients.Methods AGR,NLR,clinicopathological and follow-up data of ESCC patients diagnosed via pathology in the Department of Thoracic Surgery,The First Affiliated Hospital of Xinjiang Medical University from 2010 to 2017 were collected.The correlation between NLR/AGR and clinicopathological data were analyzed.Kaplan-Meier analysis and log-rank test were used for survival analysis.The optimal cut-off values of AGR and NLR were determined by X-tile software,and the patients were accordingly divided into a high-level group and a low-level group.At the same time,univariate and multivariate Cox regression analyses were used to identify independent risk factors affecting OS in the ESCC patients,and a nomogram prediction model was constructed and internally verified.The diagnostic efficacy of the model was evaluated by receiver operating characteristic(ROC)curve and calibration curve,and the clinical application value was evaluated by decision curve analysis.Results A total of 150 patients were included in this study,including 105 males and 45 females with a mean age of 62.3±9.3 years,and the follow-up time was 1-5 years.The 5-year OS rate of patients in the high-level AGR group was significantly higher than that in the low-level group(χ^(2)=6.339,P=0.012),and the median OS of the two groups was 25 months and 12.5 months,respectively.The 5-year OS rate of patients in the high-level NLR group was significantly lower than that in the low-level NLR group(χ^(2)=5.603,P=0.018),and the median OS of the two groups was 18 months and 39 months,respectively.Multivariate Cox analysis showed that AGR,NLR,T stage,lymph node metast

关 键 词：食管鳞状细胞癌 列线图模型 血清白蛋白/血清球蛋白比值 中性粒细胞/淋巴细胞比值 生存分析 

分 类 号：R735.1[医药卫生—肿瘤]