论文部分内容阅读
目的:分析非小细胞肺癌(non-smallcell lung cancer,NSCLC)脑转移患者脑脊液的蛋白质谱,建立NSCLC脑转移的蛋白质谱诊断模型。方法:采用表面增强激光解吸电离飞行时间质谱技术(surface-enhanced la-ser desorption/ionization ti me of flight massspectrometry,SELDI-TOF-MS)对29例脑转移NSCLC、23例非肿瘤和10例无脑转移NSCLC患者的脑脊液进行蛋白质谱分析,Biomarker Wizard软件统计分析3组样本的蛋白质峰,Biomarker Patterns软件的决策树模型进行差异蛋白峰的比较和判别,建立分类决策树诊断模型,评价其诊断和应用价值。结果:脑转移组和非肿瘤组之间有8个蛋白质峰表达差异有统计学意义,其中m/z6634.73蛋白质峰可把2组样本分开,灵敏度为89.66%(26/29),特异度为86.96%(20/23)。进一步行交叉验证结果显示其灵敏度为80.00%(8/10),特异度为70.00%(7/10)。脑转移组与无脑转移组有5个蛋白质峰表达差异有统计学意义,运用m/z8698.00、1215.32和1245.70蛋白质峰将2组样本分开,灵敏度和特异度均为100.00%(29/29)。结论:采用SELDI技术可以发现3组患者的脑脊液蛋白质峰表达差异有统计学意义,利用这些差异蛋白质建立的蛋白质谱分类决策树状诊断模型具有较高的灵敏性、特异性和准确率。
Objective: To analyze the protein spectrum of cerebrospinal fluid (CSF) in patients with non-small cell lung cancer (NSCLC) brain metastasis and to establish a protein spectrum diagnosis model of NSCLC brain metastasis. Methods: Twenty-nine patients with brain metastases NSCLC, 23 non-tumor patients and 10 patients without brain were analyzed by surface-enhanced laser desorption / ionization ti me of flight mass spectrometry (SELDI-TOF-MS) The protein profiles of CSF in patients with metastatic NSCLC were analyzed. Biomarker Wizard software was used to statistically analyze the protein peaks in three groups of samples and Biomarker Patterns. The diagnostic and application of classification decision tree value. Results: Eight protein peaks between brain metastasis group and non-tumor group showed statistical significance. The peak of m / z 6634.73 protein could separate two groups of samples with the sensitivity of 89.66% (26/29), the specificity 86.96% (20/23). Further line cross-validation results showed that the sensitivity was 80.00% (8/10) and the specificity was 70.00% (7/10). Five protein peaks in brain metastasis group and no brain metastasis group showed statistically significant difference. The two groups of samples were separated by m / z 8698.00,1215.32 and 1245.70 protein peak, the sensitivity and specificity were 100.00% (29/29 ). CONCLUSIONS: The protein peak expression of cerebrospinal fluid in three groups of patients was found to be statistically significant using SELDI technique. The use of these differential proteins to establish a tree-based diagnosis model of protein spectrum classification has high sensitivity, specificity and accuracy.