Tau and Aβ imaging, CSF measures, and cognition in Alzheimer's disease

Abstract

Alzheimer's disease (AD) is characterized by two molecular pathologies: cerebral β-amyloidosis in the form of β-amyloid (Aβ) plaques and tauopathy in the form of neurofibrillary tangles, neuritic plaques, and neuropil threads. Until recently, only Aβ could be studied in humans using positron emission tomography (PET) imaging owing to a lack of tau PET imaging agents. Clinical pathological studies have linked tau pathology closely to the onset and progression of cognitive symptoms in patients with AD. We report PET imaging of tau and Aβ in a cohort of cognitively normal older adults and those with mild AD. Multivariate analyses identified unique disease-related stereotypical spatial patterns (topographies) for deposition of tau and Aβ. These PET imaging tau and Aβ topographies were spatially distinct but correlated with disease progression. Cerebrospinal fluid measures of tau, often used to stage preclinical AD, correlated with tau deposition in the temporal lobe. Tau deposition in the temporal lobe more closely tracked dementia status and was a better predictor of cognitive performance than Aβ deposition in any region of the brain. These data support models of AD where tau pathology closely tracks changes in brain function that are responsible for the onset of early symptoms in AD.

Fig. 1. Mean tau and Aβ topographies

Mean PET tau and Aβ topographies in participants with and without clinical AD. Mean tau SUVR images averaged across participants with CDR0 (top row) or CDR>0 (bottom row). Images are presented in radiological convention. These images show relatively low pathology in cognitively normal individuals and increased pathology in the cognitively impaired group.

Fig. 2. PET tau and Aβ topographies are associated with disease severity

(A) Right singular vectors (V) of the SVD of the regional PET tau and Aβ data represented as topographies. The dimensionality of both tau and Aβ was estimated to be two. The first topography for both PET species broadly represents the mean (that is, all regions are homogeneously weighted). The second tau topography is present within the temporal lobe, and the second Aβ topography is largely seen in frontal and parietal lobes. Color bar represents regional weights within each singular component. (B) The representations of these topographies in individuals (left singular vectors, U) (that is, the original data projected onto these topographies) varied with CDR for both PET tau topographies and the second PET Aβ topography. Increasing disease severity (measured using the CDR) is associated with increasing representation of the present topographies in individuals. However, particularly in the Aβ data, there was significant heterogeneity. (C) Similar graphs as (B), but only participants with CSF analysis were included. Color in the CDR0 group indicates preclinical disease status: blue corresponds to healthy aging (that is, no Aβ or tau pathology), green corresponds to stage 1, and red corresponds to stage 2 of the CDR.

Fig. 3. Tau and Aβ in distinct topographies are strongly correlated

(A) Topographies for PET tau and PET Aβ imaging derived from canonical correlations that maximize the correlation between PET tau and Aβ deposition. (B) Scatter plot of the weights on each region demonstrates that PET tau and Aβ topographies are not significantly related. This is consistent with the typical topographies of tau and Aβ pathology being distinct. (C) The data projected onto the canonical variables revealed a robust positive relationship.

Fig. 4. CSF and neuropsychological performance are predicted by tau and Aβ topographies

Each row represents a single penalized regression model where either CSF protein or neuropsychological performance is predicted. Σ|β_tau| and Σ|β_Ab| represent the total predictive weight of tau and Aβ topographies, respectively. Gray regions have no predictive weight. (A) Penalized regression models that predict CSF tau and Aβ₄₂ using tau and Aβ topographies. Because CSF Aβ₄₂ is inversely related to amyloid burden, negative weights indicate regions where topographies predict worse CSF Aβ₄₂ pathology. (B) Penalized regression models that predict neuropsychological performance in each examined domain: global, episodic, semantic, and visuospatial (visuospat). Regions with negative values (displayed in cool colors) are where more PET pathology predicts lower cognitive performance.