1 Department of Clinical Medicine - Center of Functionally Integrative Neuroscience, Department of Clinical Medicine, Health, Aarhus University2 McGill University3 Department of Clinical Medicine - Center of Functionally Integrative Neuroscience, Department of Clinical Medicine, Health, Aarhus University
Background: The dynamic biomarker hypothesis predicts that fluid biomarker abnormalities precede brain morphological changes observed in AD by many years. However, the link between early CSF abnormalities and late structural changes remains under debate. Here we investigated the association between regional cortical thinning (CT) measured by Magnetic Resonance Imaging (MRI) and brain amyloidosis (measured by CSF Ab 1-42 concentrations), or tau hyperphosphorylation (tau 181; p-tau) in Alzheimer’s Disease (AD) and Mild Cognitive Impairment (MCI) patients. We test the hypothesis that the association between cortical thinning, amyloidosis or tau hyperphosphorylation depends on cortical regions and clinical stages of AD. Methods: T1-weighed MRIs and associated CSF markers from individuals with mild cognitive impairment (MCI; 16), Alzheimer Disease (AD; n¼7) and age-matched cognitively normal subjects (CN; n¼8) were obtained from the Alzheimer’s Disease Neuroimaging Initiative (ADNI) cohort. Cortical surface reconstruction and group registration were generated using Freesurfer. A general linear model was used to conduct regressions between CSF markers and cortical thickness. Results: Correlation analyses in the MCI group showed a positive correlation between CT and Ab 1-42 measures predominantly in the temporo-parietal regions, namely the precuneus (peak r¼0.67; p<0.01) and the fusiform (peak r¼0.85; p<0.0001). Similar but smaller effects were present in AD, with additional clusters in the frontal cortex (peak r¼0.94; p¼0.0001). In both MCI and AD groups, significant clusters of negative correlations between CTand p-tau were mainly found in medio-temporal and dorso-lateral prefrontal areas. In the CN group, no relationship was observed between cortical thickness and either CSF biomarkers. Conclusions: Although limited by sample sizes, our results suggest a posterior-anterior gradient of structural vulnerability to A b 1-42. Alternatively, weaker temporo-parietal effect between CT and Ab 1-42 in AD relative to MCI, likely results from a ceiling effect of Ab 1-42 levels in the ADgroup. Conversely, a frontal effect can only be found in the AD group, which possibly corresponds to a later stage of Ab 1-42 -driven neurodegeneration.
Alzheimer; cortical thickness; CSF; MCI
Main Research Area:
Alzheimer's Association International Conference, 2013