Logistic回归和ROC曲线评价肿瘤标志物在肺癌淋巴结转移中的价值

Diagnostic Value of Tumor Markers for Limph Nodes Metastasis in Lung Cancer Analysis with ROC Curve and Logistic Regression

目的:探讨Logistic回归和ROC曲线综合分析血清癌胚抗原(CEA)、鳞状细胞癌抗原(SCC)和铁蛋白(SF)检测在肺癌淋巴结转移中的应用价值。方法:采用电化学发光法检测CEA,酶免法检测SCC,免疫比浊法检测SF,检测100例肺癌患者血清中三种肿瘤标志物水平,其中转移组65例,未转移组35例。结果:转移组CEA、SCC的阳性率显著高于未转移组的阳性率(P<0.01,P<0.05),而SF在两组间的阳性率比较差异无统计学意义。建立回归模型Y=1/[1+EXP(1.584X1+0.935X2-0.425)],新变量Y的AUC高于3种单一肿瘤标志物的AUC。结论:CEA和SCC对肺癌淋巴结转移的诊断具有较高的价值,综合运用Logistic回归和ROC曲线分析的提高临床判断肺癌淋巴结转移的准确性。

Objective：To investigate the diagnostic value of serum CEA,SCC and SF as analyzed with Logistic vegression and ROC curve in patients with lung cancer lymph nodes metastasis. Method：Serum CEA,SCC,and SF were detected in 100 patients with lung cancer,and in those cases,65 cases were identified as lymph nodes metastasis group and 35 as non-metsatasis group. Result：The positive rate of CEA and SCC of metastais group was evidently higher than that of non-metastasis（P〈0.01,P〈0.05）,but there was no significant difference for SF between the two groups. According to regression equation Y=1/[1＋EXP（1.584X1＋0.935X2-0.425）],and the AUC of variable Y was higher than any of the three tumor markers. Conclusion：The combination of CEA and SCC is useful as indicator of lung cancer lymph nodes metastasis. Applications of Logistic regression and ROC curve increase diagnostic accuracy in lung cancer.