表面增强激光解析电离飞行时间质谱技术对周围型肺癌差异蛋白表达的研究

The study of differential proteins in patients with peripheral lung cancer by surface-enhanced laser desorption/ionization time of flight mass technology

目的 筛选周围型肺癌患者血清、肺组织活检前后BALF中的差异蛋白,探讨其表达特点、影响因素及其临床意义.方法 选择2008年3月至2009年11月周围型肺癌及肺部良性病变患者各20例,应用表面增强激光解析电离飞行时间质谱（SELDI-TOF-MS）技术分别检测患者血清、活检前后BALF中的蛋白质质谱并初步建立分类树诊断模型.结果 （1）两组患者血清中检出6个差异蛋白峰,以质荷比为6637的差异蛋白建模诊断周围型肺癌的敏感度为70%（14/20）,特异度为90%（18/20）,正确率为80%（32/40）,阳性预测值为88%（14/16）,阴性预测值为75%（18/24）,受试者工作曲线（ROC曲线）下而积（AUC）为0.73;（2）两组活检前BALF中检出11个差异蛋白峰,以质荷比为7982的差异蛋白建模诊断周围型肺癌的敏感度为85%（17/20）,特异度为90%（18/20）,正确率为88%（35/40）,阳性预测值为89%（17/19）,阴性预测值为86%（18/21）,AUC为0.94;（3）两组活检后BALF中检出14个差异蛋白峰,以质荷比为7671的差异蛋白建模诊断周围型肺癌的敏感度为85%（17/20）,特异度为100%（20/20）,正确率为93%（37/40）,阳性预测值为100%（17/17）,阴性预测值为87%（20/23）,AUC为0.93.结论 周围型肺癌活检前后BALF中检出的差异蛋白较血清中多,活检后BALF中的差异蛋白较活检前多.活检前质荷比为7982的差异蛋白与活检后质荷比为7671的差异蛋白在周围型肺癌患者BALF中的AUC值均＞0.9,高于血清;这2种差异蛋白可作为较好的早期诊断周围型肺癌的标志蛋白.

Objective To scan the protein mass spectra in the sera and bronchoalvolar lavage fluid （BALF） from patients with peripheral lung cancer, screen out the differential proteins, and explore the clinical significance of the differential proteins. Methods SELDI-TOF-MS was used to detect the protein mass spectra and to screen out the differential proteins in the sera and BALF collected before and after lung biopsy in 20 patients with peripheral lung cancer and 20 patients with benign pulmonary diseases. The differential proteins were analyzed and the initial diagnostic models were set up. Results （ 1 ） There were 6 differential protein peaks in the sera of the 2 groups （ P ＜0. 05 ＝. The protein with a mass/charge ratio （ M/Z） of 6637 was selected to establish the diagnostic model. The sensitivity of diagnosing peripheral lung cancer was 70% （14/20） ,the specificity 90% （ 18/20）, the accuracy 80% （32/40）, the positive predictive value （ PV ＋ ） 88% （ 14/16）, the negative predictive value（ PV - ） 75% （ 18/24）, and the area under the ROC curve （AUC）was 0.73. （2） There were 11 differential protein peaks in the BALF collected before lung cancer biopsy of the 2 groups （ P ＜ 0. 05 ＝. The protein with a M/Z of 7982 was selected to establish the diagnostic model. The sensitivity of diagnosing peripheral lung cancer was 85% （ 17/20 ）, the specificity 90% （ 18/20）, the accuracy 88% （35/40）, the PV ＋ 89% （ 17/19）, the PV - 86% （ 18/21 ）, and the AUC was 0. 94. （3） There were 14 differential protein peaks in the BALF collected after lung cancer biopsy of the 2 groups （ P ＜0. 05 ＝. The protein with a M/Z of 7671 was selected to establish the diagnostic model.The sensitivity of diagnosing peripheral lung cancer was 85% （17/20） ,the specificity 100% （20/20）, the accuracy 93% （37/40）, the PV ＋ 100% （17/17）, the PV- 87% （20/23）, and the AUC was 0. 93. Conclusions There were more differential proteins in BALF as compared with sera. There were more differential proteins in the BALF collected after lung biopsy as compared to that before lung biopsy. The AUC of the diagnostic models set up by proteins in BALF collected before and after lung biopsy were all above 0. 9 and showed higher efficiency for the diagnosis of peripheral lung cancer as compared to proteins in sera.These differential proteins may be better tumor markers for the diagnosis of peripheral lung cancer at the early stage.