Abstract
79
Background: Cerebrovascular disease (CVD) is not currently considered a core pathological feature of Alzheimer’s disease (AD), but concurrent CVD may exacerbate AD progression (Brickman et al, 2015).
Objectives: The purpose of this study was first to examine the relationship among amyloid, CVD, and neurodegeneration among individuals with AD, mild cognitive impairment (MCI), and healthy controls, and second, to examine the extent to which amyloid and CVD pathology drive neurodegeneration over time.
Methods: Six hundred eight subjects (222 normal(NL), 291 MCI, 93 AD) from the Alzheimer’s Disease Neuroimaging Initiative underwent longitudinal AV45 PET imaging, as well as T1- and T2-weighted MRI scans to quantify amyloid, white matter hyperintensities (WMHs), and cortical thickness, respectively. AV45 standard uptake value ratio (SUVR) was estimated as the ratio of cortical regions to a composite white matter reference region. White matter hyperintensity volume, an indication of CVD, was derived with T1- and T2-weighted MRI using a Bayesian Markov-Random Field approach with spatial and contextual prior probabilities. Mean cortical thickness in AD-associated regions was measured with the FreeSurfer 5.1 longitudinal processing stream. A series of linear mixed-effects models tested for: 1) differences in AV45 SUVR, WMH volume, and cortical thickness between diagnostic groups 2) the relationship among the three image-derived biomarker measures 3) differences in rates of change over time in biomarkers across diagnostic groups 4) the relationship between longitudinal rate of change in cortical thickness, an indication of neurodegeneration, and baseline amyloid or WMH volume. Covariates included mean-centered age (baseline: 73.4±6.9yrs), APOE-ɛ4 status (359 APOE- ɛ4+), sex (326M), education (16.1±2.1yrs), and intracranial total volume (baseline: 1519.8±159.7cm3). Results: Across diagnostic groups, AV45 SUVR increased (NL<MCI,AD), WMH volume increased (NL<MCI,AD), and cortical thickness decreased (NL>MCI,AD), with or without age adjustment (Table1). In a model including all three imaging biomarkers, increased AV45 SUVR and WMH volume were both significantly associated with decreased cortical thickness, but after full covariate adjustment, AV45 SUVR and age, but not WMH volume, were significant (Figure1). AV45 SUVR rate decreased across diagnostic groups (NL>AD), indicating a possible plateau, whereas WMH volume rate increased on the trend level (Table 1). Cortical thinning rate slightly decreased across diagnostic groups (NL<AD; Table1). In a model including all three imaging biomarkers, the longitudinal cortical thinning rate was significantly associated with higher baseline WMH, but not AV45 SUVR; however, the association between baseline WMH and cortical thinning rate did not survive full covariate adjustment. Conclusion: This multimodal approach indicates that increased amyloid pathology and increased WMH volume are related to lower cortical thickness. On the other hand, increased WMH volume is associated with neurodegenerative/atrophic changes over time in key AD-related brain regions but amyloid is not. While CVD and AD may be etiologically independent, the findings suggest that CVD should be considered explicitly for its effect on AD progression.