上海交通大学学报（医学版）

上海交通大学学报（医学版） ›› 2017, Vol. 37 ›› Issue (7): 950-.doi: 10.3969/j.issn.1674-8115.2017.07.011

• 论著（基础研究） • 上一篇    下一篇

基于血浆免疫及炎症相关蛋白的阿尔茨海默病筛查标志物的研究

何 璐 1，王瑛 1, 2，王瑛 1，徐玮 1，陈生弟 1，赵简 2，丁健青 1   

  1. 1. 上海交通大学 医学院附属瑞金医院神经内科，上海交通大学 医学院神经病学研究所，上海 200025；2. 同济大学附属东方医院干细胞工程转 化医学中心，上海 200123
  • 出版日期:2017-07-28 发布日期:2017-08-25
  • 通讯作者: 丁健青，电子信箱：jqding18@163.com
  • 作者简介:?何璐（1990—），女，医学学士；电子信箱：s_hlxj@163.com
  • 基金资助:
    国家重点基础研究发展计划（“973”计划）（2014CB965002）；国家自然科学基金（81171200）；上海市科学技术委员会项目（13JC1401502, 13140904000）

Study on detection biomarkers of Alzheimer’s disease based on plasma immune and inflammatory proteins#br#

HE Lu1, WANG Ying1, 2, WANG Ying1, XU Wei1, CHEN Sheng-di1, ZHAO Jian2, DING Jian-qing1   

  1. 1. Department of Neurology and Institute of Neurology, Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, Shanghai 200025; 2. Translational Medical Center for Stem Cell Therapy, Shanghai East Hospital, School of Medicine, Tongji University, Shanghai 200123
  • Online:2017-07-28 Published:2017-08-25
  • Supported by:
    National Program on Key Basic Research Project of China, “973”Program,  2014CB965002; National Natural Science Foundation of China, 81171200; Project from Science and Technology Commission of Shanghai Municipality, 13JC1401502, 13140904000

PDF (PC)

851

摘要： 目的 · 探索可作为阿尔茨海默病（AD）潜在筛查标志物的血浆免疫及炎症相关蛋白。方法 · 纳入对照组19 例，AD 组 19 例，收集血浆样本并检测 70 种免疫及炎症相关蛋白。运用 Mann-Whitney U 检验、偏相关分析筛选与 AD 强相关的免疫炎症蛋白。采 用 Wilks’ lambda 逐步分析法建立多蛋白联合判别算法，建立受试者工作特征（ROC）曲线评价判别算法的诊断效能。结果 · 70 种 蛋白中，23 种在AD 患者血浆中表达显著升高（P<0.05），其中19 种与AD 强相关（P<0.05）。用 Wilks’ lambda 逐步分析法建立 的多蛋白联合判别算法显示，由11 种血浆免疫及炎症蛋白（EGF、GRO、MDC、MCP-1、MCP-2、MCP-4、TARC、SCF、TRAIL、 CTACK、GCP-2）联合建立的判别方程具有最优的诊断效能（AUC=0.994），其最佳截断值为 -0.609。取最佳截断值时，方程诊断敏 感度可达 100%，特异度可达 94.7%。结论 · 由上述 11 种血浆免疫炎症相关蛋白组成的判别方程具有辅助 AD 筛查的潜力。

关键词: 阿尔茨海默病, 生物标志物, 血浆, 免疫及炎症相关蛋白

Abstract:

 Objective · To explore plasma immune and inflammatory proteins that could serve as potential screening markers for Alzheimer's disease (AD).  Methods · Healthy controls (n=19) and AD patients (n=19) were enrolled. Plasma samples were collected and 70 kinds of immune and inflammatory proteins were detected. The immune and inflammatory proteins associated with AD were screened by Mann-Whitney U test and partial correlation analysis. Discriminant analysis was used to develop multi-protein combined algorithm to distinguish plasma samples of AD patients from those of healthy controls. Receiver operating characteristic (ROC) curve was used to evaluate the diagnostic efficacy for the multi-protein combined algorithm.  Results · Among the 70 proteins analyzed, 23 were significantly higher in AD patients (P<0.05), among which 19 were strongly correlated with AD (P<0.05). These 19 proteins were analyzed with Wilks' lambda stepwise analysis to develop discriminant algorithm for detecting plasma samples of AD. Finally, the discriminant algorithm established by 11 plasma immune and inflammatory proteins (EGF, GRO, MDC, MCP-1, MCP-2, MCP-4, TARC, SCF, TRAIL, CTACK, GCP2) was found to have an optimal diagnostic efficacy (AUC=0.994). The optimal cutoff value of the algorithm was -0.609. When the optimal cutoff value was obtained, the sensitivity of the equation could reach 100% and the specificity could reach 94.7%.  Conclusion · The discriminant equation composed of the above 11 plasma immune and inflammatory proteins has the potential to assist AD screening.

Key words:  Alzheimer's disease, biomarker, plasma, immune and inflammatory proteins