Comparing metabolomic and pathologic biomarkers alone and in combination for discriminating Alzheimer%26apos;s disease from normal cognitive aging

作者：Motsinger Reif Alison A; Zhu Hongjie; Kling Mitchel A; Matson Wayne; Sharma Swati; Fiehn Oliver; Reif David M; Appleby Dina H; Doraiswamy P Murali; Trojanowski John Q; Kaddurah Daouk Rima^*; Arnold Steven E

来源：Acta Neuropathologica Communications, 2013, 1(1): 28.

DOI：10.1186/2051-5960-1-28

• Alzheimer%26apos
• s disease
• Dementias
• Metabolomics
• Cerebrospinal fluid biomarkers
• Stepwise logistic regression

Background: A critical and as-yet unmet need in Alzheimer disease (AD) research is the development of novel markers that can identify individuals at risk for cognitive decline due to AD. This would aid intervention trials designed to slow the progression of AD by increasing diagnostic certainty, and provide new pathophysiologic clues and potential drug targets. %26lt;br%26gt;Results: We used two metabolomics platforms (gas chromatography-time of flight mass spectrometry [GC-TOF] and liquid chromatography LC-ECA array [LC-ECA]) to measure a number of metabolites in cerebrospinal fluid (CSF) from patients with AD dementia and from cognitively normal controls. We used stepwise logistic regression models with cross-validation to assess the ability of metabolite markers to discriminate between clinically diagnosed AD participants and cognitively normal controls and we compared these data with traditional CSF Luminex immunoassay amyloid-beta and tau biomarkers. A beta and tau biomarkers had high accuracy to discriminate cases and controls (testing area under the curve: 0.92). The accuracy of GC-TOF metabolites and LC-ECA metabolites by themselves to discriminate clinical AD participants from controls was high (testing area under the curve: 0.70 and 0.96, respectively). %26lt;br%26gt;Conclusions: Our study identified several CSF small-molecule metabolites that discriminated especially well between clinically diagnosed AD and control groups. They appear to be suitable for further confirmatory and validation studies, and show the potential to provide predictive performance for AD.

• 出版日期2013
• 单位UC Davis