AbstractBackgroundBrain arteriolosclerosis, which involves thickening of the vessel wall and stenosis of arterioles, is one of the main pathologies of cerebral small vessel disease, and often co‐occurs with other pathology indicative of Alzheimer’s disease and related disorders (ADRD). A novel classifier of arteriolosclerosis, ARTS, was recently developed by the MarkVCID consortium for in vivo studies of arteriolosclerosis. We hypothesized that higher levels of arteriolosclerosis are associated with worse ADRD biomarker levels.Method109 participants from the Wake Forest Alzheimer’s Disease Research Center’s Clinical Core underwent clinical and cognitive evaluation, 3T brain MRI, and PiB (amyloid) PET (n=92). Participants were adjudicated by NIA‐AA criteria as having normal cognition (n=69), mild cognitive impairment (n=25), or dementia (n=15). Raw T1, FLAIR, and TOPUP and Eddy Current Corrected DTI images along with age and sex were inputs to a machine learning based classifier generating ARTS scores; higher ARTS indicates higher likelihood of arteriolosclerosis. For variables of interest, T1 MRI were processed (FreeSurfer v5.3) to calculate cortical gray matter (GM), hippocampal, and intracranial (ICV) volumes, and temporal meta‐ROI cortical thickness. White matter hyperintensity (WMH) volumes were calculated on FLAIR with LST in SPM12. NODDI were processed using AMICO to generate mean Free Water (FW) in hippocampus and global GM and white matter (WM). A set of AD‐sensitive cortical ROIs were used to extract mean PET SUVr signal. Separate multivariable general linear models (GLM) examined relationships between ARTS and variables of interest adjusted initially for age and total ICV (Table 1), and then again after adjusting for additional covariates (sex, race, education, APOE‐e4 carrier status, and clinical diagnosis).ResultAmong 109 participants (71±7 years, 60% women), ARTS ranged from ‐.80 to .29 (mean±SD = ‐.37±.20), suggesting a low overall likelihood of arteriolosclerosis in our sample. Higher ARTS was significantly associated with lower cortical thickness, higher WMH, higher WM FW, and greater amyloid deposition (Table 1). When adjusting for additional covariates, only amyloid no longer significantly associated with ARTS (p=0.19).ConclusionHigher ARTS scores are cross‐sectionally associated with other abnormalities in ADRD biomarkers. Future directions will examine the relationship between ARTS and Flortaucipir (tau) uptake.
Read full abstract