317例老年口咽癌患者的临床特征和预后分析:一项单中心15年回顾性研究  

Clinical Characteristics and Prognostic Analyses of 317 Elderly Patients with Oropharyngeal Cancer:A Single Center Retrospective Study Over 15 Years

作　　者：史婷婷 李晨曦[2,3] 龚忠诚[2] 崔文丽[4] 买买提吐逊·吐尔地[1] 张维娜[5] SHI Tingting;LI Chenxi;GONG Zhongcheng;CUI Wenli;Maimaitituxun·TUERDI;ZHANG Weina(Department of Maxillofacial Trauma and Orthognathic Surgery,School/Hosptital of Stomatology,the First Affiliated Hospital of Xinjiang Medical University,Urumqi 830054,China;Department of Oral and Maxillofacial Oncology&Surgery,School/Hosptital of Stomatology,the First Affiliated Hospital of Xinjiang Medical University,Stomatological Research Institute of Xinjiang Uygur Autonomous Region,Urumqi 830054,China;Clinical Medicine Postdoctoral Scientific Research Station,Xinjiang Medical University,Urumqi 830054,China;Department of Pathology,the First Affiliated Hospital of Xinjiang Medical University,Urumqi 830054,China;Department of Otorhinolaryngology,the First Affiliated Hospital of Xinjiang Medical University,Urumqi 830054,China)

机构地区：[1]新疆医科大学第一附属医院(附属口腔医院)颌面创伤正颌外科,新疆乌鲁木齐830054 [2]新疆医科大学第一附属医院(附属口腔医院)口腔颌面肿瘤外科,新疆维吾尔自治区口腔医学研究所,新疆乌鲁木齐830054 [3]新疆医科大学临床医学博士后科研流动站,新疆乌鲁木齐830054 [4]新疆医科大学第一附属医院病理科,新疆乌鲁木齐830054 [5]新疆医科大学第一附属医院耳鼻喉科,新疆乌鲁木齐830054

出　　处：《口腔医学研究》2025年第2期101-108,共8页Journal of Oral Science Research

基　　金：国家自然科学基金(批号:82360481);口腔颌面发育与再生湖北省重点实验室开放课题基金(批号:2022kqhm008);“天山英才”医药卫生领军人才项目(批号:TSYC202301A001)。

摘　　要：目的:本研究基于回顾性分析老年口咽癌患者的人口统计学、临床病理资料和生存模式,构建列线图模型并进行预后分析。方法:收集新疆医科大学第一附属医院2000年1月~2015年12月收治的老年口咽癌患者317例。按7∶3随机分成训练队列(n=221)和验证队列(n=96)两组。利用R语言进行Bonferroni校正后的χ2检验或Fisher精确概率检验比较训练集和验证集两组间变量分布;单因素和多因素Cox比例风险回归分析评估患者预后的独立危险因素并建立预测模型;采用后向归纳法结合绩效评价模型,并绘制受试者工作特征曲线和校准曲线分析模型识别及校准效能。结果:Cox比例风险回归分析表明,影响老年口咽癌患者预后生存的独立危险因素包括年龄、民族、婚姻状况、肿瘤分化程度、TN分期、手术、放化疗(P值均<0.05)。基于多元回归模型的输出概率,按照0.5界定高危、低危两组患者并绘制生存曲线,Kaplan-Meier分析表明低危组患者(<0.5)生存率均显著高于高危组(>0.5)患者(P<0.001)。在模型预测能力方面,训练队列受试者工作特征的曲线下面积(area under curve,AUC)为AUC1年=0.873,AUC2年=0.829,AUC3年=0.795,验证队列AUC1年=0.823,AUC2年=0.806,AUC3年=0.768,表明模型具有良好的判别能力。校准曲线也显示出模型整体拟合度良好,模型预测概率与实际观察结果匹配程度较高。结论:本研究系统分析了影响老年口咽癌患者的危险因素,并基于这些危险因素开发、验证老年口咽癌患者的临床预后。该预测工具可以帮助临床医生识别高危患者,并提前制定个性化的治疗方案。Objective:To construct a nomogram model to perform the disease-specific analysis,based on demographics,clinicopathological data,and overall survival of the elderly patients with oropharyngeal cancer.Methods:A total of 317 patients admitted to Xinjiang Medical University Affiliated First Hospital,between January 2000 and December 2015,were divided into training(n=221)and validation cohorts(n=96)with a ratio of 7∶3.Statistics was performed by R language program.Association and distribution of clinicopathologic variables among two groups were analyzed using Fisher's exact test or Chi-square withBonferroni correction.Cox proportional hazards regressionanalysis was used to assess independent risk factors impacting patients’prognosis and construct prediction model.Through backward induction and performance evaluation modeling,the identification and calibration efficacy of the definitive predictive model were appraised by receiver operator characteristic curve and calibration curve analysis.Results:Cox analysis revealed that age,race,marital status,grade of tumor differentiation,T and N stage,surgery,radiation,and chemotherapy were independent risk factors(all P<0.005).Based on the probability output by multiple regression model,high-risk and low-risk groups of patients were defined according to probability of 0.5 and survival curves were then plotted.Kaplan-Meier analysis showed that the survival rate of low-risk patients(less than 0.5)was significantly higher than that of high-risk patients(greater than 0.5)(all P<0.001).In terms of model’s predictive ability,the area under curve(AUC)of receiver operator characteristic curve in training queue was AUC 1-year=0.873,AUC 2-year=0.829,and AUC 3-year=0.795,respectively;and AUC values in test queue was AUC 1-year=0.823,AUC 2-year=0.806,and AUC 3-year=0.768,respectively,indicating that the model had good discriminative ability.The calibration curves also showed that the overall fitting degree of the model was good,and the predicted probability of the model matched the a

关 键 词：老年患者 口咽鳞状细胞癌 临床预后 列线图模型 

分 类 号：R73[医药卫生—肿瘤]