Premium
Urine dicarboxylic acids are metabolic biomarkers of early Alzheimer’s disease
Author(s) -
Harrington Michael G.,
Maldonado Salvador,
Sagare Abhay P.,
Zlokovic Berislav V.,
Chui Helena C.,
Fonteh Alfred N.
Publication year - 2021
Publication title -
alzheimer's and dementia
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 6.713
H-Index - 118
eISSN - 1552-5279
pISSN - 1552-5260
DOI - 10.1002/alz.056655
Subject(s) - urine , medicine , cohort , chemistry , alzheimer's disease , cerebrospinal fluid , endocrinology , gastroenterology , disease
Background We previously identified urine dicarboxylic acids (DCA) of carbon length 4‐9 as candidate biomarkers of probable Alzheimer’s disease (AD) and energy dysregulation, and found their levels correlated with hippocampal volume. Our aim was to evaluate these initial findings in a second cohort. Method Study participants (> 65 years) were classified clinically as cognitively healthy (CH, n=188) or with probable AD (n=66) based on Uniform Data System‐3 of the National Alzheimer’s Coordinating Center criteria. Cerebrospinal fluid (CSF) amyloid ß 42 (A) and phospho‐ tau (T) levels were measured by validated immunoassay. Urine was collected in a “spot” sample after a 12 h overnight fast; target DCAs were detected using gas chromatography coupled with negative ion chemical ionization mass spectrometry and each DCA was normalized to deuterated internal standards and to a percent of total fatty acids. Result The mean of short‐chain DCA’s (C4‐C5) were higher in CH (44.38% ±17.79) compared to AD individuals (34.79% ±12.39) (p=0.0002), while the long‐chain DCA’s (C7‐C9) were lower for CH (34.70% ±16.88) when compared to AD individuals (45.22% ±15.11) (p<0.0001). DCA levels in the subset of CH participants with A + /T + were in a transition to the levels of the AD group and C4‐C5 and C7‐C9 differed from those with A ‐ /T ‐ biomarkers (p=0.0001). Combining results into a ratio of short‐ to long‐chain DCAs distinguished the A ‐ /T ‐ (2.330 ±2.144) from the A + /T + group (1.178 ±0.946); p<0.0001. ROC curve of the DCA ratio between CH‐NAT and AD is 82%. Study participant clinical and DCA groups were not significantly correlated with the common confounders of age, sex, education or ApoE genotype. Conclusion Our data extends our initial finding that urine DCA measures have potential as biomarkers of AD pathology, including for cognitively healthy individuals who are positive for CSF AT biomarkers. We propose that measurement of urine DCAs provides a new approach to screen for metabolic fluctuations in early AD and to monitor the effect of novel therapies in clinical trials. While studies are needed, including with other populations, varied clinical classifications, simplified chemistry, this data supports DCA biomarkers for widespread screening since the necessary urine sample is available from everyone.