四种指标联合检测在肺癌鉴别诊断中的价值  

Value of combined detection of 4 indicators in differentiation diagnosis of lung cancer

作　　者：徐鹏[1] 胡冬[1] 

机构地区：[1]四川省绵阳市中心医院检验科,621000

出　　处：《检验医学与临床》2016年第10期1367-1369,共3页Laboratory Medicine and Clinic

摘　　要：目的探讨血清中癌胚抗原(CEA)、糖类抗原(CA)125,细胞角蛋白19片段(CYFRA21-1)和胃泌素释放前肽(Pro-GRP)四种指标联合检测在肺癌鉴别诊断中的价值。方法选择2014年8月至2015年9月该院收治的92例肺癌患者为肺癌组,选择同期142例肺部良性疾病患者作为对照组,采用化学发光法和酶联免疫吸附试验(ELISA)测定血清中CEA、CA125、CYFRA21-1和Pro-GRP水平。运用线性判别分析(LDA)法建立肺癌诊断与预测模型,并与并联试验结果的准确性进行比较。采用内代法,随机选取2/3的患者为训练集,1/3的患者为测试集,考察模型的预测效果。结果肺癌组血清中CEA、CA125、CYFRA21-1和Pro-GRP水平均显著高于对照组,差异均有统计学意义(P<0.05)。LDA模型能正确诊断81.5%的肺癌患者和83.1%的肺部良性疾病患者,模型总的诊断准确率为82.5%。该模型能对76.5%的肺癌和80.0%的肺部良性疾病进行预测,其预测准确率为80.8%。并联试验结果表明,CEA、CA125、CYFRA21-1和Pro-GRP联合检测诊断肺癌的灵敏度、特异度和准确率分别为80.4%、76.1%和77.8%。结论联合检测血清中CEA、CA125、CYFRA21-1和Pro-GRP提高了肺癌诊断的灵敏度。基于血清中CEA、CA125、CYFRA21-1和Pro-GRP的LDA模型能较好地诊断和预测肺癌,该模型为肺癌的鉴别诊断和预测提供了一种全新的思路。Objective To evaluate the value of 4-indicator combined detection of serum carcinoembryonic antigen（CEA）,carbohydrate antigen 125（CA125）,cytokeratin 19fragments（CYFRA21-1）and pro-gastrin-releasing peptide（Pro-GRP）in the differentiation diagnosis of lung cancer.Methods Ninety-two cases of lung cancer in our hospital from August 2014 to September 2015 were selected as the lung cancer group and contemporaneous 142 cases of benign lung disease were taken as the control group.The chemiluminescence and ELISA assays were used to determine the levels of serum CEA,CA125,CYFRA 21-1and Pro-GRP.The linear discriminant analysis（LDA）method was applied to establish the diagnosis and prediction model for lung cancer.Then its accuracy was compared with that in the parallel test.The inner test was adopted to randomly select 2/3of patients as the training set and 1/3of patients as the testing set for the investigating the prediction effect of the model.Results The levels of serum CEA,CA125,CYFRA21-1,and Pro-GRP in the lung cancer group were significantly higher than those in the control group with statistical difference（P0.05）.The LDA model could correctly diagnose 81.5% of patients with lung cancer and 83.1% of patients with benign lung disease,reaching the total accuracy of 82.5%.This model could predicted76.5% of patients with lung cancer and 80.0% of patients with benign diseases,reaching the predictive accuracy rate of 80.8%.The parallel test results showed that the sensitivity,specificity and accuracy of the combined detection of serum CEA,CA125,CYFRA21-1and Pro-GRP for diagnosing lung cancer were 80.4%,76.1% and 77.8% respectively.Conclusion The combined detection of serum CEA,CA125,CYFRA21-1and Pro-GRP increases the sensitivity for the diagnosis of lung cancer.The LDA model based on serum CEA,CA125,CYFRA21-1and Pro-GRP can better diagnose and predict lung cancer and provides a brand-new thinking for the differentiation diagnosis and prediction of lung cancer.

关 键 词：肺癌 癌胚抗原 糖类抗原125 胃泌素释放前肽 线性判别分析 

分 类 号：R734.2[医药卫生—肿瘤]