Premium
Lower neocortical tau burden distinguishes Alzheimer’s pathology occurring with vs. without primary non‐Alzheimer’s pathology
Author(s) -
Phillips Jeffrey S,
Cousins Katheryn Alexandra Quilico,
Irwin David J,
Wolk David A,
Lee Eddie B,
Gibbons Garrett S,
McMillan Corey T,
Trojanowski John Q,
Grossman Murray
Publication year - 2020
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 - 10.1002/alz.046529
Subject(s) - entorhinal cortex , pathology , hippocampal sclerosis , neurofibrillary tangle , dementia , subiculum , alzheimer's disease , neuropathology , frontotemporal lobar degeneration , medicine , amyotrophic lateral sclerosis , parahippocampal gyrus , hippocampus , temporal cortex , atrophy , psychology , disease , temporal lobe , frontotemporal dementia , neuroscience , senile plaques , dentate gyrus , epilepsy
Background The NIA/AA research framework proposes a biological definition of Alzheimer’s disease (AD) based on the presence of both amyloid and tau. However, AD frequently co‐occurs with other non‐AD pathologies, and it is unclear how these pathologies affect the progression of AD‐associated amyloid‐β and tau. We investigated regional heterogeneity in AD pathology in a postmortem sample of either pure AD or co‐occurring non‐AD and AD pathology. We hypothesized that amyloid‐β and tau would be more likely to co‐localize in pure AD than in the mixed‐pathology group. Method We retrospectively identified two groups with intermediate or high levels of AD pathologic change by ABC criteria: 1) cases with a pathologic diagnosis of AD (AD‐pure; n=92) and minimal co‐pathology; and 2) those with a primary non‐AD pathology (frontotemporal lobar degeneration due to tau or TDP‐43; dementia with Lewy bodies; Parkinson’s disease; or amyotrophic lateral sclerosis) coupled with AD co‐pathology (AD co‐path; n=68). Primary outcomes were ordinal ratings (0‐3) of postmortem amyloid‐β and tau burden in the amygdala, entorhinal cortex, CA1/subiculum, superior/middle temporal gyrus, middle frontal gyrus, and angular gyrus; positivity was assessed by ratings of 2‐3. Result The AD‐pure group was younger at death (M=74.3, SD=12.2) than AD co‐path (M=78.6, SD=7.9) and was more likely to have high AD pathologic change (OR=18.0, p<0.0001). The AD co‐path group had fewer regions positive for both amyloid‐β and tau (M=0.50, SD=0.27) than the AD‐pure group (M=0.87, SD=0.23), independent of age and Braak stage. AD‐singular patients were more likely than AD co‐path patients to be tau‐positive in middle frontal (84% vs. 21%), angular (84% vs. 29%), and superior/middle temporal cortex (88% vs. 41%). A logistic regression model based on tau ratings in these three neocortical regions distinguished AD‐pure and AD co‐path with 82.4% accuracy. When the analysis was stratified by Braak stage, differences in neocortical tau were limited to Braak V/VI cases. Conclusion Autopsy data indicate AD co‐path cases may have lower neocortical tau burden than AD‐singular cases at end of life. These postmortem findings suggest that AD co‐path cases may be identifiable during life by assessing neocortical tau distribution using tau‐binding PET ligands.