Urine dicarboxylic acids reflect loss of energy capacity and hippocampal volume in pre‐symptomatic Alzheimer's disease Biomarkers (non‐neuroimaging) / plasma/serum/urine biomarkers

• Published in: Alzheimer's & dementia Vol. 16; no. S4
• Main Authors: Castor, Katherine, Tran, Thao, King, Kevin, Chui, Helena C., Fonteh, Alfred N., Harrington, Michael G
• Format: Journal Article
• Language: English
• Published: 01.12.2020
• DOI: 10.1002/alz.046021
• ISSN: 1552-5260

Abstract Background Non‐invasive biomarkers will expedite widespread screening and early diagnosis of Alzheimer’s disease (AD). We hypothesized that the considerate loss of brain tissue in AD will result in detection of brain lipid components in urine, and that these will change in concert with CSF and brain biomarkers of AD. Methods Study participants (> 70 years) classified as cognitively healthy (CH, n=64) or AD (n=26) after neuropsychological assessment were recruited into HMRI’s Brain Aging Study. Urine dicarboxylic acids (DCA) of carbon length 3‐10 that reflect oxidative damage and energy generation or balance that may reveal changes in brain function were quantified using gas chromatography coupled with negative ion chemical ionization mass spectrometry. Brain volumes were obtained using a GE 3 or 1.5 Tesla MRI scanner. MRI data was analyzed using Freesurfer 6.0 to obtain left and right volumes (mm 3 ) of seven brain regions including the hippocampus. Results Mean C4‐C5 DCAs were lower and mean C7‐C10 DCAs were higher in the urine from AD compared to cognitively healthy (CH) individuals. Moreover, mean C4‐C5 DCAs were lower and mean C7‐C9 were higher in urine from CH individuals with abnormal compared to normal CSF amyloid and Tau levels; i.e., the apparent urine changes in AD also appeared to be present in CH individuals that have CSF risk factors of AD pathology. In examining the relationship between urine DCAs and AD biomarkers, we found short chain DCAs positively correlated with CSF Aβ 42 , while C7‐C10 DCAs negatively corelated with CSF Aβ 42 and positively correlated with CSF Tau levels. Furthermore, we found a negative correlation of C7‐C10 DCAs with hippocampal volume (p<0.01), while there was no significant correlation with the occipital volume. Urine measure of DCAs have an 82% ability to predict cognitively healthy participants with normal CSF amyloid/Tau. Conclusions These data suggest that urine measures of increased lipid oxidation and dysfunctional energy balance reflect early AD pathology manifest by brain and CSF biomarkers. Measures of urine DCAs may contribute to personalized healthcare by indicating AD pathology and may be utilized to explore population wellness and monitor the efficacy of therapies in clinical trials.