Tau accumulation and its spatial progression across the Alzheimer's disease spectrum

Abstract

The accumulation of tau abnormality in sporadic Alzheimer's disease is believed typically to follow neuropathologically defined Braak staging. Recent in-vivo PET evidence challenges this belief, however, as accumulation patterns for tau appear heterogeneous among individuals with varying clinical expressions of Alzheimer's disease. We, therefore, sought a better understanding of the spatial distribution of tau in the preclinical and clinical phases of sporadic Alzheimer's disease and its association with cognitive decline. Longitudinal tau-PET data (1370 scans) from 832 participants (463 cognitively unimpaired, 277 with mild cognitive impairment and 92 with Alzheimer's disease dementia) were obtained from the Alzheimer's Disease Neuroimaging Initiative. Among these, we defined thresholds of abnormal tau deposition in 70 brain regions from the Desikan atlas, and for each group of regions characteristic of Braak staging. We summed each scan's number of regions with abnormal tau deposition to form a spatial extent index. We then examined patterns of tau pathology cross-sectionally and longitudinally and assessed their heterogeneity. Finally, we compared our spatial extent index of tau uptake with a temporal meta-region of interest-a commonly used proxy of tau burden-assessing their association with cognitive scores and clinical progression. More than 80% of amyloid-beta positive participants across diagnostic groups followed typical Braak staging, both cross-sectionally and longitudinally. Within each Braak stage, however, the pattern of abnormality demonstrated significant heterogeneity such that the overlap of abnormal regions across participants averaged less than 50%, particularly in persons with mild cognitive impairment. Accumulation of tau progressed more rapidly among cognitively unimpaired and participants with mild cognitive impairment (1.2 newly abnormal regions per year) compared to participants with Alzheimer's disease dementia (less than 1 newly abnormal region per year). Comparing the association of tau pathology and cognitive performance our spatial extent index was superior to the temporal meta-region of interest for identifying associations with memory in cognitively unimpaired individuals and explained more variance for measures of executive function in patients with mild cognitive impairments and Alzheimer's disease dementia. Thus, while participants broadly followed Braak stages, significant individual regional heterogeneity of tau binding was observed at each clinical stage. Progression of the spatial extent of tau pathology appears to be fastest in cognitively unimpaired and persons with mild cognitive impairment. Exploring the spatial distribution of tau deposits throughout the entire brain may uncover further pathological variations and their correlation with cognitive impairments.

Keywords: Alzheimer’s disease; positron emission tomography; spatial extent; tau.

Graphical Abstract

Figure 1

Spatial extent methodology. For each cortical region of the Desikan atlas and the bilateral amygdalae, we extract the SUVR of our participants (A). Then, a two-component Gaussian mixture modelling technique is applied to the SUVR values in each region (B and C). The second distribution is considered to reflect abnormally high SUVR tau values. We extract the probability that each participant belongs to the ‘abnormal’ distribution and establish a threshold that individuals with over 50% probability are considered positive for the given region (D). Once thresholds are derived across all regions, we derive the spatial extent index for each participant by summing the number of positive regions across the brain (E). We also apply the same methodology to the average SUVR within each aggregate composing Braak stages I and III through VI (F). To compare our spatial extent index in the cognition analyses, we also compute the average SUVR in a classic temporal meta-ROI. (G). Figure adapted from sihnpy’s documentation (https://sihnpy.readthedocs.io/), available with a CC-BY licence.

Figure 2

Amyloid and tau status in the cohort. (A) Aβ/tau status in the included participants from ADNI. Aβ positivity was established using ADNI’s tracer-specific recommendations for both florbetapir and florbetaben. Tau positivity was defined as having at least one region positive for tau pathology (spatial extent index of one and above). (B) Scatterplot of the probability of having at least one positive tau region (i.e. spatial extent index equal to or higher than one) as a function of the Aβ load (in centiloid). The probability was extracted from a logitistic regression. The odds ratio (and confidence interval) derived from a logistic regression is presented at the bottom of the graph. Note that the points were jittered by a factor of 0.065 × 0.065 for visualization purposes.

Figure 3

Spatial extent of abnormal tau deposition in amyloid-positive participants of the ADNI cohort. (A) Based on the method discussed in Fig. 1, abnormality thresholds were determined for each I Braak stages (except stage II) and for each II region of the cortical mantle and the bilateral amygdalae (70 regions). One row on the heatmap corresponds to an individual participant, while each column represents a distinct cortical region. Within each diagnostic group, participants were sorted from individuals with the lowest to highest spatial extent index. Regions on the x-axis in II are sorted by Braak stages. (B) Regional average SUVR, by diagnostic status. (C) Brain maps representing the percentage of participants having abnormal levels of tau in each region, by diagnostic status.

Figure 4

Spatial localization of abnormal tau accumulation over time in amyloid-positive participants of the ADNI cohort. (A) Abnormal accumulation is presented by (I) Braak stages and (II) all 70 individual brain regions of the Desikan atlas. Colours denote the change in the region between the baseline and the last available visit. A stable region (negative or positive) did not change status during the follow-up. A progressing region was originally negative and subsequently became positive over time. A regressing region was originally positive and became negative over time. (B) Brain maps presenting the average SUVR change per region per year. (C) Brain maps representing the percentage of participants becoming tau-positive in each region annually. In both B and C, values in the bilateral amygdalae are represented by small coloured circles in the medial view of the brain, and the annual change is calculated in each region using linear mixed-effect models with random slopes and intercepts. Only participants with at least three tau scans (n = 100) were kept for B and C to ensure a constant sample across the longitudinal follow-ups.

Figure 5

Association between tau-PET measures, and memory performance and decline. (A) Memory performance closest in time to the tau-PET scan and (B) memory decline computed across the study period were associated with both temporal meta-ROI SUVR and spatial extent index in Aβ-positive participants using linear regressions. Cognitive decline was computed for each participant with more than two cognitive time points using linear mixed-effect models with random slopes and intercepts. In each panel, columns represent a diagnostic group (leftmost: whole sample, second from the left: cognitively unimpaired, second from the right: mild cognitive impairment, right-most: Alzheimer’s disease). Simple and standardized β coefficients, adjusted R² and Akaike information criterion, controlled for age sex and education, are shown on the graphs. P-value of models are indicated next to the simple beta coefficients. (°: P < 0.1, *: P < 0.05, **: P < 0.01, ***: P < 0.001) Results remained significant after a multiple comparison FDR correction.

Figure 6

Region-wise associations between regional tau-PET SUVR and cognitive performance and decline in participants with MCI and Alzheimer’s disease. Association between tau-PET SUVR and cognitive performance (A) and cognitive decline (B) in participants with MCI and with Alzheimer’s disease across four cognitive domains (memory, executive functioning, language and visuospatial). Cognitive decline was computed for each participant with more than two cognitive time points using linear mixed-effect models with random slopes and intercepts. The standardized β coefficients of the associations between tau-PET SUVR in a specific region and each cognition measure are displayed if it survives adjustment for age, sex and education and a multiple comparison FDR correction (P_corrected < 0.05).