Affordable Access

Access to the full text

Multifactorial Analysis of a Biomarker Pool for Alzheimer Disease Risk in a North Indian Population

Authors
  • Talwar, Puneet
  • Grover, Sandeep
  • Sinha, Juhi
  • Chandna, Puneet
  • Agarwal, Rachna
  • Kushwaha, Suman
  • Kukreti, Ritushree
Type
Published Article
Journal
Dementia and Geriatric Cognitive Disorders
Publisher
S. Karger AG
Publication Date
Jun 21, 2017
Volume
44
Issue
1-2
Pages
25–34
Identifiers
DOI: 10.1159/000477206
PMID: 28633142
Source
Karger
Keywords
License
Green
External links

Abstract

Background: Alzheimer disease (AD) is a progressive neurodegenerative disease with a complex multifactorial etiology. Here, we aim to identify a biomarker pool comprised of genetic variants and blood biomarkers as predictor of AD risk. Methods: We performed a case-control study involving 108 cases and 159 non-demented healthy controls to examine the association of multiple biomarkers with AD risk. Results: The APOE genotyping revealed that ε4 allele frequency was significantly high (p value = 0.0001, OR = 2.66, 95% CI 1.58-4.46) in AD as compared to controls, whereas ε2 (p = 0.0430, OR = 0.29, CI 0.07-1.10) was overrepresented in controls. In biochemical assays, significant differences in levels of total copper, free copper, zinc, copper/zinc ratio, iron, epidermal growth factor receptor (EGFR), leptin, and albumin were also observed. The AD risk score (ADRS) as a linear combination of 6 candidate markers involving age, education status, APOE ε4 allele, levels of iron, Cu/Zn ratio, and EGFR was created using stepwise linear discriminant analysis. The area under the ROC curve of the ADRS panel for predicting AD risk was significantly high (AUC = 0.84, p < 0.0001, 95% CI 0.78-0.89, sensitivity = 70.0%, specificity = 83.8%) compared to individual parameters. Conclusion: These findings support the multifactorial etiology of AD and demonstrate the ability of a panel involving 6 biomarkers to discriminate AD cases from non-demented healthy controls.

Report this publication

Statistics

Seen <100 times