蛋白指纹图谱蛋白芯片技术检测诊断早期胃癌的临床意义

Detection and evaluation of serum proteomic patterns by SELDI-TOF-MS in early gastric cancer

目的:应用SELDI蛋白质芯片检测胃癌患者血清蛋白质指纹图谱,筛选候选肿瘤标志物以建立诊断模型,并探讨其诊断早期胃癌的临床意义。方法:表面加强激光解吸电离-飞行时间质谱(SELDI-TOF-MS)技术及其配套蛋白质芯片检测80例胃癌患者(Ⅰ/Ⅱ期40例与Ⅲ/Ⅳ期40例)、80例良性胃病患者(胃溃疡40例与慢性萎缩性胃炎40例)和80例健康人的血清蛋白质质谱。将部分研究对象随机分为训练集(40例胃癌、20名良性胃病与20名健康人群)和验证集(40例胃癌、20名良性胃病与20名健康人群),前者用于筛选胃癌差异蛋白标志物并建立人工神经网络诊断模型,后者用于模型诊断效度的盲法验证。结果:质荷比(m/z)分别为2927、3217、3236、3287的4个蛋白质峰组合所构建的诊断模型能达到诊断胃癌患者的最佳诊断效果,灵敏度90.0%,特异度92.5%。结论:SELDI蛋白芯片技术在胃癌的诊断尤其是早期诊断、术前分期及候选肿瘤标志物筛选等方面具有一定价值,值得进一步研究。

Objective：To detect the serum proteomic patterns using SELD1-TOF-MS ProteinChip array technology in gastric cancer,to screen biomarker candidates,to build diagnostic models and to evaluate its clinical significance in early gastric cancer.Methods：SELDI-TOF-MS ProteinChip was used to detect the serum proteomic patterns of 80 patients with gastric cancer （40 cases of Ⅰ/Ⅱ stage and 40 cases of Ⅲ/Ⅳ stage）,80 patients with benign gastropathy （40 cases of gastric ulcer and 40 cases of chronic atrophic gastritis）, and 80 healthy people.Some of the serum samples were randomized into training set （40 cases of gastric cancer, 20 cases of benign gastropathy and 20 cases of healthy people） and test set （40 cases of gastric cancer, 20 cases of benign gastropathy and 20 cases of healthy people）. At first, the training set of samples was detected using SELDI mass spectrometry.Using a multi-layer artificial neural network with a back propagation algorithm, a proteomic pattern was identified,which could discriminate cancer from other samples in the training set.The discovered patern was then used to determine the accuracy of the classification system in the test set. Results：The model composed by 4 protein peaks 2927m/z,3217m/z ,3236rn/z and 3287m/z could do the best in the diagnosis of gastric cancer.The sensitivity and specificity of it were 90.0% and 92.5% ,respectively.COnclusion：This method shows great potential for the early detection,staging before operation and screening novel and better biomarkers of early gastric cancer.