18F]AV1451 PET IN RELATION TO ATROPHY ACROSS THE ALZHEIMER’S DISEASE SPECTRUM

Abstract

Neuropathological Braak staging of neurofibrillary tangles has been introduced to capture the spatial spreading of tau. Using [18F]AV1451 PET, it is now possible to visualize these Braak stages in vivo. The aim of this study was to investigate whether and how specific [18F]AV1451 binding in ascending Braak stages relates to regional atrophy patterns across the Alzheimer's disease (AD) spectrum. We included 70 subjects (35 subjective cognitive decline [SCD], 6 mild cognitive impairment [MCI] and 29 AD dementia) who underwent a 130-minute dynamic [18F]AV1451 PET scan and structural MRI (table1). For [18F]AV1451, receptor parametric mapping (RPM) with cerebellar grey matter as reference region was used to calculate binding potential (BPND). Separate regional values for 1) Braak stage I/II (enthorinal), 2) Braak stage III/IV (limbic) and 3) Braak stage V/VI (neocortical) (Schöll et al, 2016) were extracted (figure 1). T1-weighted MR images were used to assess grey matter (GM) volumes. We performed whole-brain voxelwise analyses using the 3 regional Braak [18F]AV1451 BPND as independent variables, and GM volumes as the dependent variable, adjusting for age, sex and total intracranial volume, with a threshold of p<0.001 (uncorrected for multiple comparisons). These analyses were repeated using region-of-interest analyses for both [18F]AV1451 BPND and GM volumes. [18F]AV1451 BPND values per Braak region, according to diagnosis. Abbreviations: BPND= binding potential, SCD = subjective cognitive decline, MCI = Mild Cognitive Impairment, AD = Alzheimer's Disease. In the MCI/AD group, [18F]AV1451 BPND in Braak stage I/II showed limited associations with GM volumes. Greater [18F]AV1451 BPND in Braak stage III/IV, however, was associated with lower GM volumes in temporal, parietal and frontal cortices (figure2). These associations were similar for [18F]AV1451 BPND in Braak stage V/VI, although slightly less widespread (figure2). For SCD, only subtle associations between [18F]AV1451 BPND in Braak III/IV and GM volumes were observed (figure3). Region-of-interest analyses were in line with these findings (table2, figure4). Associations between regional [18F]AV1451 BPND and voxelwise brain GM volumes in MCI/AD. Analyses were adjusted for age, sex and TIV. Displayed are p-values, thresholded at an uncorrected p<0.001. Associations between regional [18F]AV1451 BPND and voxelwise brain GM volumes in SCD. Analyses were adjusted for age, sex and TIV. Displayed are p-values, thresholded at an uncorrected p<0.001. Relationships between [18F]AVIASI BPND and gray matter ROIs. Displayed are relationships between [18F]AV1451 BPND in Braak 1-2 (panel A,D), Braak 3-4 (panel B,E) and Braak 5-6 (panel C,F) and gray matter in medial temporal lobe (panel A-C) and temporoparietal lobe (panel D-F). Data are shown for SCD in blue and MCI/AD in red. Beta coefficients and p-values are displayed in blue boxes for SCD and in red boxs for MCI/AD. Abbreviations: BPND = binding potential GM = gray matter; ROI = region of interest, SCD = subjective cognitive decline, MCI = mild cognitive decline, AD = Alzheimer's disease These data suggest that tau pathology in Braak stages III/IV is most strongly associated with gray matter volume loss in patients with MCI and AD. Higher tau specific binding in Braak stage V/VI is not related to more extensive atrophy. In SCD and in Braak stage I/II in MCI/AD, only subtle associations between tau pathology and gray matter were observed. These findings could possibly be influenced by partial volume effects.