Objective To investigate the gastric cardia adenocarcinoma (GCA ) related tumor markers and establishrnGCA diagnosis model, to screen the proteins which can distinguish GCA with or without lymph node metastases or not, and create a new way for early diagnosis and prognosis of GCA. Methods The proteins were detected by CM10 chips combined surface-enhanced laser desorption/ionization time of flight mass spectrometry ( SELDI-TOF-MS ) in of serum samples from 101 cases of GCA patients and 118 cases of healthy controls. The biomarker Wizard software and biomarker patterns system( BPS )were used to analyze distinct proteins in two groups, and establish a decision tree model for diagnosing GCA which was blindly validated by random selected samples. The protein fingerprint compared patterns were compared between patients with and without lymph nodes metastasis to obtain distinct proteins. Results 56 distinguished protein peaks in total were detected( P <0. 05 ),four of which were used to build a diagnostic model with the most significant differen(P <10-8 )protein of 5 480. 49 as the primary root node, and the diagnostic model differentiated the patients with GCA from the healthy people with a sensitivity of 95. 05% and specificity of 95. 76% in the blinded sample set. The protein whose M/Z peak was 3 820. 87 , was found to differentiate the patients with lymphatic metastasis from the ones without metastasis; the sensitivity and specificity was 84. 4% and 82. 1% respectively. Conclusion SELDI-TOF-MS technique can be used to establish a GCA diagnostic model with high sensitivity and specificity, and can be used in screening the proteins which may indicate lymph nodes metastasis and provide new method for prognosis of GCA.%目的 寻找贲门癌肿瘤标记物并建立诊断模型.筛选提示发生淋巴转移的差异蛋白,为临床早期诊断贲门癌及判断预后提供新方法.方法 CM10蛋白芯片结合表面增强激光解析电离飞行时间质谱(SELDI-TOF-MS)检测101例贲门癌患者及118例正常对照者血清蛋白,Biomarker Wizard 软件和Biomarker Patterns System(BPS)分析得差异蛋白并建立贲门癌诊断分类树,双盲法验证其准确性.进一步分析对比有、无淋巴转移的贲门癌患者血清蛋白指纹图谱,得到差异蛋白.结果 对照分析贲门癌患者和正常对照组的血清蛋白指纹图谱得56个差异蛋白(P<0.05),经BPS软件筛选,以质荷比为5 480.49(P<10-8)的蛋白为主根结点建立贲门癌诊断模型,其灵敏度为95.05%,特异度为95.76%.得到可提示淋巴转移的差异蛋白3 820.87(P<0.05),其灵敏度为84.4%,特异度为82.1%.结论 SELDI-TOF-MS技术可建立有较高灵敏度和特异度的贲门癌诊断模型,筛选出能够提示发生淋巴转移的差异蛋白,有助于贲门癌的早期筛查、诊断和预后判断.
展开▼
