Premium
P2‐079: AMYLIN DYSHOMEOSTASIS DISRUPTS WHITE MATTER STRUCTURAL INTEGRITY AND MODULATES AMYLOID COMPOSITION AND PATHOLOGY DISTRIBUTION IN BRAINS OF PATIENTS WITH AD AND AD RATS
Author(s) -
Ly Han,
Verma Nirmal,
Lashley Tammaryn,
Hardy John,
Despa Florin
Publication year - 2019
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 - 10.1016/j.jalz.2019.06.2486
Subject(s) - amylin , white matter , amyloid (mycology) , endocrinology , medicine , grey matter , immunohistochemistry , pathology , diabetes mellitus , magnetic resonance imaging , islet , radiology
Background: Network and models at the region of interest (ROI) level have been used to study the complex deposition of amyloid throughout the brain in the early stages of Alzheimer’s disease (AD). While the estimation of deposition of amyloid and networks are not independent, both approaches continue to be studied separately providing an incomplete picture of the relationships between amyloid levels and network organisation. Methods: In this work, amyloid levels and networks are estimated jointly via a Wombling model. Applied on data from the Alzheimer’s Disease Neuroimaging Initiative (ADNI; CN 1⁄4 220, MCI 1⁄4 182 at baseline), all individuals were PET amyloid negative and 30% were CSF positive (Ab42 < 192 pg/ml). Wombling estimated the amyloid levels and amyloid covariance network conditional on each other providing a complete picture of both amyloid regional spatial patterns and level estimates. Results: In our pre-dementia cohort we found the posterior and anterior cingulate gyrus, as well as the banks of the superior temporal sulcus, had significantly higher amyloid levels compared to the frontal and subcortical regions such as the amygdala and hippocampus. Interestingly, despite these differences in amyloid levels, with a posterior probability greater than 0.9, there were second order network connections between ROIs with the highest and lowest amyloid levels. Conclusions: Joint modelling of pre-dementia amyloid levels and network connectivity enable a better understanding of the complex regional patterns and disease pathology in the early stages where Ab42 has decreased in CSF but not yet detected via PET.