Neuropathologic, genetic, and longitudinal cognitive profiles in primary age-related tauopathy (PART) and Alzheimer's disease

Abstract

Introduction: Primary age-related tauopathy (PART) is a recently described entity that can cause cognitive impairment in the absence of Alzheimer's disease (AD). Here, we compared neuropathological features, tau haplotypes, apolipoprotein E (APOE) genotypes, and cognitive profiles in age-matched subjects with PART and AD pathology.

Methods: Brain autopsies (n = 183) were conducted on participants 85 years and older from the Baltimore Longitudinal Study of Aging and Johns Hopkins Alzheimer's Disease Research Center. Participants, normal at enrollment, were followed with periodic cognitive evaluations until death.

Results: Compared with AD, PART subjects showed significantly slower rates of decline on measures of memory, language, and visuospatial performance. They also showed lower APOE ε4 allele frequency (4.1% vs. 17.6%, P = .0046).

Discussion: Our observations suggest that PART is separate from AD and its distinction will be important for the clinical management of patients with cognitive impairment and for public health care planning.

Keywords: Aging; Alzheimer disease (AD); Dementia; Mild Cognitive Impairment (MCI); Neurofibrillary tangles; Primary Age-Related Tauopathy (PART); Public Health Planning.

Fig. 1.

Histopathologic features of primary age-related tauopathy (PART). Low-power digitized scanning view micrograph of Hirano Silver-stained hippocampus of 102-year-old female with definite PART (image scanned at 100X) (A). Semiquantitative categorical scoring of neurofibrillary tangle (NFT) densities on PHF-1 immunohistochemistry and Hirano silver, 200 × showing sparse (B), moderate (C), and frequent (D) NFTs.

Fig. 2.

Comparison of primary age-related tauopathy (PART) and Alzheimer’s disease (AD) pathology and associated neurodegenerative features. Heatmaps showing the semiquantitative densities and distribution of neurofibrillarytangle (NFT) in 42 PART subjects (A—left [blue]) and 62 AD subjects (B—left [Red]). Each row represents a subject with PART or AD. Columns represent brain regions. The densities of NFT, that is, 0: none, 1: sparse, 2: moderate, 3: frequent, are color-coded according to the scales depicted in A-below and B-below. The Braak score, presence of associated Lewy body disease (yellow), and TDP-43 proteinopathy (green) in PART and AD are shown on A-right and B-right, respectively. Clinical dementia rating score (CDR) and Clinical dementia rating score sum of boxes (CDR-SB) and tau haplotypes (TAU) are shown in PART with vascular lesions sufficient to cause or contribute to cognitive impairment (red) or without (white) (A). The CERAD A, B, and C grouping of the AD cases corresponds to sparse, moderate, or frequent neuritic plaques [16] (white = none; X = not examined). Abbreviations: ERC, entorhinal cortex; HIPPO, hippocampus; AMY, amygdala, SMTG, superior and middle temporal gyri, IP, inferior parietal lobule; MFG, middle frontal gyrus; OCC, visual cortex (Brodmann’s areas 17 and 18); CERAD, Consortium to Establish a Registry for Alzheimer’s Disease.

Fig. 3.

Regional distribution and density of neurofibrillary tangles (NFT) throughout the brain in PART (N = 42 subjects). In primary age-related tauopathy (PART), tau lesions in the form of NFTs show a distribution largely limited to the parahippocampal gyrus (entorhinal cortex, transentorhinal cortex, and medial occipitotemporal or fusiform gyrus), hippocampus, amygdala and basal forebrain, with less severe extension to the brainstem and other neocortical regions. The semiquantitative density [16] of NFT is color-coded and is most severe in medial temporal lobe structures.

Fig. 4.

Comparison of longitudinal rates of change of cognitive performance in 5 domains. Longitudinal rates of cognitive performance change examined in 5 domains (Memory, Attention, Executive Function, Language, and Visuospatial Ability) and Mini-Mental State Examination (MMSE) were found to show a highly significant difference between subjects with primary age-related tauopathy (PART) and Alzheimer’s disease (AD) pathology in memory, language, visuospatial ability, and MMSE raw score (prorated for visual impairment) (P = .0034, P = .0019, P = .0051, and P = .016, respectively). In all these domains, AD subjects showed a faster rate of decline than PART subjects. No difference in longitudinal rates of cognitive performance was seen in attention and executive function domains (P = .33, P = .32, respectively).