Assessment of Tau Tangles and Amyloid-β Plaques Among Super Agers Using PET Imaging

This cross-sectional study examines positron emission tomography (PET) imaging to investigate the burden of tau tangles and amyloid β plaques in super agers, normal agers, and patients with mild cognitive impairment vs younger amyloid-negative controls.


Introduction
Little is known about the presence and extent of amyloid-β plaques and tau tangles in individuals who preserve exceptional cognitive function despite advanced age, 1 also known as super agers.
Although lower expression of these hallmarks of neurodegeneration may be expected in this group, [2][3][4] in vivo evidence of tau tangles in particular is still lacking. Using positron emission tomography (PET) imaging data, we therefore studied the tau tangle and amyloid-β plaque burdens in a super ager (SA) group, normal ager (NA) group, and a patient group with mild cognitive impairment (MCI). Because tau tangles are closely associated with cognitive decline, we hypothesized that the SA group would have fewer tau tangles than the NA and MCI groups.

Methods
In this cross-sectional study, we included data (retrieved in June 2019) of 3 age-and educationmatched patient groups of 25 SAs, 25 NAs, and 25 patients with MCI, all aged 80 years or older (Table) from the Alzheimer's Disease Neuroimaging Initiative (ADNI) database. A group of younger amyloid-negative controls (YC group) served as the reference. Ethical approval was obtained by the ADNI investigators at each participating site. All participants provided written informed consent.
We categorized the SA, NA, and MCI groups according to the ADNI memory score from the ADNI neuropsychological test battery. 5 We focused on at least 4 ADNI memory score measurements within a 4-year period leading back from the [ 18 F]AV-1451 PET acquisition. Individuals with a mean ADNI memory z score greater than 1.25 were defined as the SA group, those with a mean z score between 0.5 and 1.25 during the study period were defined as the NA group, and those with a mean z score of less than 0 were defined as the MCI group.
Regional group differences in tau tangles and amyloid-β plaques were compared between the YC group and the other 3 groups using normalized and intensity-standardized (reference: cerebellum) [ 18 F]AV-1451 (tau) and [ 18 F]AV-45 (amyloid) PET scans in a voxelwise (P < .0001, uncorrected) and a region-of-interest (ROI) approach, including sex as covariate. The ROI approach included 5 meta-ROIs (entorhinal cortex, inferior temporal, middle occipital, precuneus, and Author affiliations and article information are listed at the end of this article. The results of the voxelwise analysis (Figure, A) yielded no differences in tau tangles and amyloid-β plaques when comparing the SA group with the YC group. In contrast, the NA group presented with higher tau burden in medial temporal regions but no differences in amyloid burden compared with YC group. The MCI group demonstrated both elevated amyloid and tau burden.  A and B, All brain projections represent the contrast of the respective group against the younger healthy cognitively normal group (P < .0001, uncorrected). C and D, Box plots show significant differences between regional standard uptake value ratios (SUVRs) of the 4 groups. Lines within boxes denote medians, tops and bottoms of boxes denote 75th and 25th percentiles, error bars denote 95% CIs, and circles denote outliers.
Benjamini Hochberg corrected P value threshold was P Յ .022 for tau burden and P Յ .007 for amyloid burden.

Discussion
The in vivo findings of this cross-sectional study suggest that the phenomenon of super aging may be associated with higher brain resistance against the buildup of both tau tangles and amyloid-β plaques, which could prevent neurodegeneration, as previously hypothesized. 1,3 Normal aging, in contrast, appears to be associated with tau tangles but not amyloid plaques, pointing to a role of isolated tau accumulation in age-related cognitive decline, whereas synergistic effects of both proteinopathies seem to accelerate the unsuccessful aging process as seen in MCI.
A limitation of this study was the small sample size. Despite the small sample size and the crosssectional design, this study may stimulate future longitudinal assessments in larger, less selective cohorts also examining the role of lifestyle and molecular pathways, to decipher causal factors associated with successful aging. Overall, the characterization of individuals who remain resistant to these aging-associated proteinopathies, may inspire novel concepts for cognitive preservation in older age and therapy of neurodegeneration.