非小细胞肺癌脑转移风险预测的诺模图评分模型应用探究  

A study of nomogram scoring model to predict brain metastases risk of non-small cell lung cancer

作　　者：孙艺媛 丁广成[2] 陈鑫[2] 时昌立 孟三彦 SUN Yiyuan;DING Guangcheng;CHEN Xin;SHI Changli;MENG Sanyan(Department of Oncology,Gongli Hospital of Shanghai Pudong New Area,Shanghai 200135,China)

机构地区：[1]上海浦东新区公利医院肿瘤科,上海200135 [2]郑州大学第五附属医院伽马刀科,450003

出　　处：《临床肿瘤学杂志》2019年第10期914-918,共5页Chinese Clinical Oncology

摘　　要：目的对可能影响非小细胞肺癌(NSCLC)脑转移的临床因素进行收集和分析,并建立一个有效的诺模图评分模型,实现对NSCLC脑转移的预测。方法收集2010年1月至2015年1月352例NSCLC患者的临床资料,采用Cox回归模型分析患者发生脑转移的危险因素,建立NSCLC确诊后3、5年发生脑转移概率的诺模图模型。结果Cox单因素回归分析初步筛选性别、病理类型、分化程度、T分期、淋巴结转移数目、N分期、TNM分期、化疗方案数、EGFR基因突变、神经元特异性烯醇化酶(NSE)、CA125和CA199共12个预测变量,再代入Cox多因素回归模型,结果显示,NSE(HR=8.89,95%CI:3.34~25.77,P<0.001)、病理类型(HR=1.95,95%CI:1.32~3.62,P=0.026)、淋巴结转移数目(HR=3.9,95%CI:2.28~7.56,P<0.001)、分化程度(HR=2.99,95%CI:1.67~5.45,P=0.001)和TNM分期(HR=1.78,95%CI:1.12~2.98,P=0.035)为影响NSCLC发生脑转移的独立因素,依此建立预测NSCLC脑转移风险诺模图的一致性指数为0.83。结论诺模图可以用于预测NSCLC确诊后3﹑5年发生脑转移的概率,从而识别发生脑转移风险的高危患者,为制定临床决策提供参考依据。Objective To collect and analyze the clinical factors that may affect brain metastases of non-small cell lung cancer(NSCLC)and establish an model of nomogram to predict the brain metastases of NSCLC.Methods All 352 cases were presented from January 2010 to January 2015.Combined with patients'clinical and follow-up data,Cox regression model was used to analyze the risk factors of brain metastases and a nomogram was developed for predicting 3-and 5-year brain metastases rates.Results Univariate Cox regression preliminarily screened 12 predictive variables including gender,pathological types,differentiation,T stage,number of metastatic lymph nodes,N stage,TNM stage,number of chemotherapeutic regimens,EGFR mutation and the level of neuron-specific enolase(NSE),CA125 and CA199.Cox multivariate analysis identified NSE(HR=8.89,95%CI:3.34-25.77,P<0.001),pathological types(HR=1.95,95%CI:1.32-3.62,P=0.026),number of metastatic lymph nodes(HR=3.9,95%CI:2.28-7.56,P<0.001),differentiation(HR=2.99,95%CI:1.67-5.45,P=0.001)and TNM stage(HR=1.78,95%CI:1.12-2.98,P=0.035)were independent factors affecting brain metastases of NSCLC.The nomogram showed a coefficient coincidence of 0.83.Conclusion The nomogram can predict the incidence of 3-and 5-year brain metastases of NSCLC and can be used to identify high-risk patients with brain metastases in order to provide a reference for making clinical decisions.

关 键 词：非小细胞肺癌(NSCLC) 脑转移 危险因素 诺模图 

分 类 号：R734.2[医药卫生—肿瘤]