| Literature DB >> 29948244 |
Hong Wei1,2,3, Xiaolan Zhu4,5, Yuefeng Li6,7.
Abstract
Although the FDA already has approved two types of drug therapies for Alzheimer's disease, in regard to moderate AD, there is no clear research to support the best choice of drug treatment. The goal of this study was to examine the levels of serum biomarkers in moderate-AD patients and to explore the value of these serum biomarkers for the diagnosis of memantine sensitivity in AD patients who are significantly affected by MEM. In our study, 177 patients with moderate AD were enrolled and divided into memantine-sensitive AD (n = 90) and memantine-insensitive AD (n = 87) groups. The sera from all patients were collected, and seven serum biomarkers were analysed. Then, 120 patients were used to establish a diagnostic model that was built with a binary logistic regression analysis, and 57 patients were used to validate our model. In addition, the area under the receiver operating characteristic (ROC) curve was established. From the seven serum biomarkers, the four serum biomarkers that were selected in to establish the regression model were VEGF, BDNF, IL-6 and IL-1β. The ROC curve of best combined detection was 0.899. The diagnostic ratio of the logistic model was 0.825. This study suggests that the logistic regression model (LRM) and the ROC curve based on patients' serum levels of VEGF, BDNF, IL-6 and IL-1β is a promising research for diagnosing and choosing the best course of treatment for moderate AD.Entities:
Keywords: Alzheimer’s disease; Logistic regression; Memantine; Moderate AD
Mesh:
Substances:
Year: 2018 PMID: 29948244 DOI: 10.1007/s00415-018-8926-4
Source DB: PubMed Journal: J Neurol ISSN: 0340-5354 Impact factor: 4.849