NEUROFIBRILLARY TANGLES PREDICT IN VIVO CORTICAL ATROPHY IN PRIMARY PROGRESSIVE APHASIA WITH ALZHEIMER’S DISEASE

Abstract

The neuropathologic basis of in vivo cortical atrophy remains poorly understood in neurodegenerative diseases that cause dementia. Primary progressive aphasia (PPA) is a language-based dementia syndrome characterized by asymmetric cortical atrophy and associated with multiple proteinopathies, including the plaques and tangles of Alzheimer's disease (AD). Tangles are the strongest correlate of atrophy in those with amnestic AD, and a clinicopathologic concordance has been established between tangles and the aphasic phenotype of PPA patients with AD (PPA-AD). However, the relationship between AD pathology and in vivo cortical atrophy has not been fully elucidated in PPA-AD. The current study included five PPA-AD patients with magnetic resonance imaging (MRI) <2.5 years before death. Cortical thickness and volume data were measured with FreeSurfer software, and then converted to metrics of cortical atrophy by deriving z-scores based on a normal, age-matched control group (n=35). Neurofibrillary tangles and dense-cored amyloid-ß plaques were quantified using stereology after visualization with the Thioflavin-S stain. The in vivo and ex vivo data were examined in the same 14 language and non-language regions per case, with anatomic correspondence between MRI and brain sections. The distribution of AD pathology, cortical atrophy, and the relationship between them were determined using linear mixed models that included fixed effects for hemisphere, region, and pathologic variable, and a random effect for patient. PPA-AD displayed more cortical atrophy in the left compared to right hemisphere, and more in language compared to non-language regions (p<0.05). The accumulation of tangles, not plaques, mirrored the atrophy findings such that tangle densities showed left greater than right asymmetry and greater densities in the language compared to non-language regions (p<0.05). Larger tangle densities predicted greater magnitudes of cortical atrophy (p<0.05), while plaques showed no relationship with cortical atrophy. This investigation revealed that the regional selectivity for tangles and plaques varied based on the degree of cortical atrophy. More specifically, only tangles showed a positive relationship with cortical atrophy in PPA-AD, which is consistent with the neuropathologic basis of atrophy in amnestic AD.