摘要
目的基于竞争风险分析开发一种易于使用的个体生存预测工具,用来预测非肌层浸润性膀胱癌(NMIBC)患者1年、2年和3年复发的风险。方法获取来自西安交通大学第二附属医院NMIBC患者的随访资料。共纳入419名患者,将其随机分为训练组(n=293)和验证组(n=126)。主要变量包括诊断时的年龄、性别、吸烟史、肿瘤数量、肿瘤大小、病理分级、病理分期、膀胱灌注药物。使用训练组所有变量估计复发的累计发病率函数(CIF),通过Gray s检验发现潜在的预后变量,使用Fine-Gray子分布比例风险方法作为多因素竞争风险分析以确定独立预后变量。构建竞争风险列线图(nomogram)来预测膀胱癌患者的复发的风险。使用受试者工作特征曲线下面积(AUC)、校准曲线(calibration curve)和Brier score评估竞争风险模型的性能。结果成功将年龄、肿瘤数量、肿瘤大小、病理分级和病理分期5个独立预后因素构建竞争风险模型。验证组中竞争风险模型1年、2年和3年的AUC分别为0.895(95%CI:0.831~0.959)、0.861(95%CI:0.774~0.948)和0.827(95%CI:0.721~0.934),初步证实该模型具有较高的预测性能。结论我们成功开发了预测NMIBC患者1年、2年和3年复发风险的竞争风险模型。该预测模型有助于临床医生改善患者的术后管理。
Objective To construct an easy-to-use individual survival prognostic tool based on competing risk analyses to predict the risk of 1-,2-and 3-year recurrence for patients with non-muscle invasive bladder cancer(NMIBC).Methods The follow-up data of 419 NMIBC patients were obtained.The patients were randomly divided into training cohort(n=293)and validation cohort(n=126).The variables included age at diagnosis,sex,history of smoking,tumor number,tumor size,histolo-gic grade,pathological stage,and bladder perfusion drug.The cumulative incidence function(CIF)of recurrence was estimated using all variables in the training cohort and potential prognostic variables were determined with Gray s test.The Fine-Gray subdistribution proportional hazard approach was used as a multivariate competitive risk analysis to identify independent pro-gnostic variables.A competing risk nomogram was developed to predict the recurrence.The performance of the competing risk model was evaluated with the area under the receiver operating characteristic curve(AUC),calibration curve,and Brier score.Results Five independent prognostic factors including age,number of tumors,tumor size,histologic grade and pathological stage were used to construct the competing risk model.In the validation cohort,the AUC of 1-,2-and 3-year recurrence were 0.895(95%CI:0.831-0.959),0.861(95%CI:0.774-0.948)and 0.827(95%CI:0.721-0.934),respectively,indicating that the model had a high predictive performance.Conclusion We successfully constructed a competing risk model to predict the risk of 1-,2-and 3-year recurrence for NMIBC patients.It may help clinicians to improve the postoperative management of patients.
作者
王莉
彭子赫
雒启东
黄山珑
孙云
张敏
贾鸾
赵乐
李洪亮
WANG Li;PENG Zihe;LUO Qidong;HUANG Shanlong;SUN Yun;ZHANG Min;JIA Luan;ZHAO Le;LI Hongliang(Department of Urology,The Second Affiliated Hospital of Xi an Jiaotong University,Xi an 710004;Medical Science Center,Xi an Jiaotong University,Xi an 710061,China)
出处
《现代泌尿外科杂志》
CAS
2023年第6期487-492,共6页
Journal of Modern Urology
关键词
非肌层浸润性膀胱癌
竞争风险模型
无病生存期
复发
列线图
non-muscle invasive bladder cancer
competing risk model
disease-free survival
recurrence
nomogram
