Spatiotemporal correlation between amyloid and tau underlies cognitive changes in aging
Poster No:
1162
Submission Type:
Abstract Submission
Authors:
Chanmie Kim1, Ibai Diez1, Elisenda Bueichekú1, Sung Ahn1, Victor Montal2, Jorge Sepulcre1
Institutions:
1Massachusetts General Hospital, Boston, MA, 2Universitat Autonoma de Barcelona, Barcelona, Barcelona
First Author:
Co-Author(s):
Introduction:
It is still largely unknown how the two hallmarks of Alzheimer's disease (AD) - amyloid-beta (Aβ) plaques and tau neurofibrillary tangles - interact and propagate spatiotemporally to produce synaptic dysfunction and neuronal death at the large-scale level [1, 2]. This study aimed to identify the spatiotemporal cortical patterns of Aβ-and-tau and longitudinal cognitive changes in cognitively normal older adults (CN).
Methods:
A total of 91 participants, all deemed CN, from the Harvard Aging Brain Study (HABS), who completed cognitive assessment, T1 MRI, 11C Pittsburg Compound B (PiB) PET, and 18F Flortaucipir (tau) PET at both baseline and two-year FU visits, were included [3].
T1 MRIs were preprocessed by FreeSurfer recon-all procedure for spatial normalization, anatomical segmentation, reconstruction of cortical surfaces, and calculation of cortical thickness [4-6]. PiB PET and tau PET images were co-registered to corresponding T1 MRIs and then spatially normalized to MNI/ICBM space. Both PET images were scaled by a mean value in the cerebellar gray reference region to calculate the standardized uptake value ratio (SUVR) [7]. We applied partial volume correction (PVC) in both PET images by using an extended Müller-Gärtner (MG) method to estimate a true concentration of radiotracer in GM [8]. After that, we resampled the PET SUVRs into the standard cortical surface.
We measured vertex-wise correlations within and across PET modalities between baseline and FU. Each PET(BASE)-to-PET(FU) correlation was measured by the partial correlation between a z-score of baseline PET SUVR in vertex b and a z-score of FU PET SUVR in the paired vertex d across all possible pairs of vertices within a cortical surface while controlling for age, sex, and a z-score of FU PET SUVR in the vertex d that was not used for the correlation as a FU PET SUVR. Then, we calculated the weighted degree (WD) in each PET(BASE)-to-PET(FU) correlation matrix to identify which baseline cortical region displays hubness properties between baseline and FU PET images. The WD of each correlation matrix at column b was calculated as the sum of the significant correlation coefficients between baseline PET SUVR in vertex b and FU PET SUVRs in all possible paired vertices. Finally, the WDs of each correlation matrix were mapped to the cortical surface at the group level.
To identify the contribution of baseline & FU PET SUVRs to cognitive changes, we measured partial correlations between the merged PET SUVRs, which were calculated by averaging between baseline PET SUVRs and FU PET SUVRs across all possible vertex pairs, and FU PACC-96 scores in each correlation combination. Then, we calculated the WDs and mapped these to the standard cortical surface to visualize the spatial patterns at the group level.
T1 MRIs were preprocessed by FreeSurfer recon-all procedure for spatial normalization, anatomical segmentation, reconstruction of cortical surfaces, and calculation of cortical thickness [4-6]. PiB PET and tau PET images were co-registered to corresponding T1 MRIs and then spatially normalized to MNI/ICBM space. Both PET images were scaled by a mean value in the cerebellar gray reference region to calculate the standardized uptake value ratio (SUVR) [7]. We applied partial volume correction (PVC) in both PET images by using an extended Müller-Gärtner (MG) method to estimate a true concentration of radiotracer in GM [8]. After that, we resampled the PET SUVRs into the standard cortical surface.
We measured vertex-wise correlations within and across PET modalities between baseline and FU. Each PET(BASE)-to-PET(FU) correlation was measured by the partial correlation between a z-score of baseline PET SUVR in vertex b and a z-score of FU PET SUVR in the paired vertex d across all possible pairs of vertices within a cortical surface while controlling for age, sex, and a z-score of FU PET SUVR in the vertex d that was not used for the correlation as a FU PET SUVR. Then, we calculated the weighted degree (WD) in each PET(BASE)-to-PET(FU) correlation matrix to identify which baseline cortical region displays hubness properties between baseline and FU PET images. The WD of each correlation matrix at column b was calculated as the sum of the significant correlation coefficients between baseline PET SUVR in vertex b and FU PET SUVRs in all possible paired vertices. Finally, the WDs of each correlation matrix were mapped to the cortical surface at the group level.
To identify the contribution of baseline & FU PET SUVRs to cognitive changes, we measured partial correlations between the merged PET SUVRs, which were calculated by averaging between baseline PET SUVRs and FU PET SUVRs across all possible vertex pairs, and FU PACC-96 scores in each correlation combination. Then, we calculated the WDs and mapped these to the standard cortical surface to visualize the spatial patterns at the group level.
Results:
We found significant patterns in uni- and multi-modal PET(BASE)-to-PET(FU) analyses, indicating positive spatiotemporal relationships between a given local pathology at baseline and distributed pathology accumulations at FU (Fig. 1).
The temporal accumulations of interlinked Aβ and tau pathology display distinctive spatiotemporal correlations associated with early cognitive decline (Fig. 2). Notably, we observed that baseline Aβ deposits -Thal amyloid phase Ⅱ- related to future increase of tau deposits -Braak stage Ⅰ-Ⅳ-, both displaying linkage to the decline in multi-domain cognitive scores (Fig. 2A). We also found unimodal tau-to-tau and cognitive impairment associations in broad areas of Braak stages Ⅰ-Ⅳ (Fig. 2B).
The temporal accumulations of interlinked Aβ and tau pathology display distinctive spatiotemporal correlations associated with early cognitive decline (Fig. 2). Notably, we observed that baseline Aβ deposits -Thal amyloid phase Ⅱ- related to future increase of tau deposits -Braak stage Ⅰ-Ⅳ-, both displaying linkage to the decline in multi-domain cognitive scores (Fig. 2A). We also found unimodal tau-to-tau and cognitive impairment associations in broad areas of Braak stages Ⅰ-Ⅳ (Fig. 2B).
·Figure 1
·Figure 2
Conclusions:
AD-related pathology is considered to affect the human brain via disconectomic processes, and interdigitated spatial correlations of Aβ and tau seem to be upstream factors that might promote the breakdown of brain circuits and produce, in turn, cognitive decline in older adults. Our findings suggest that those spatiotemporal network relationships between Aβ and tau contribute to cognitive changes in the trajectory toward AD.
Disorders of the Nervous System:
Neurodegenerative/ Late Life (eg. Parkinson’s, Alzheimer’s)
Lifespan Development:
Aging 1
Modeling and Analysis Methods:
Connectivity (eg. functional, effective, structural)
Multivariate Approaches 2
PET Modeling and Analysis
Keywords:
Aging
Cognition
Multivariate
Neurological
Positron Emission Tomography (PET)
Other - amyloid, tau, graph theory, network analysis
1|2Indicates the priority used for review
Provide references using author date format
[2] Busche MA, Hyman BT. Synergy between amyloid-β and tau in Alzheimer’s disease. Nature neuroscience. 2020;23:1183-93.
[3] Dagley A, LaPoint M, Huijbers W, Hedden T, McLaren DG, Chatwal JP, et al. Harvard Aging Brain Study: Dataset and accessibility. Neuroimage. 2017;144:255-8.
[4] Dale AM, Fischl B, Sereno MI. Cortical surface-based analysis. I. Segmentation and surface reconstruction. Neuroimage. 1999;9:179-94.
[5] Fischl B, Sereno MI, Tootell RB, Dale AM. High-resolution intersubject averaging and a coordinate system for the cortical surface. Human brain mapping. 1999;8:272-84.
[6] Desikan RS, Segonne F, Fischl B, Quinn BT, Dickerson BC, Blacker D, et al. An automated labeling system for subdividing the human cerebral cortex on MRI scans into gyral based regions of interest. Neuroimage. 2006;31:968-80.
[7] Sepulcre J, Grothe MJ, d’Oleire Uquillas F, Ortiz-Terán L, Diez I, Yang H-S, et al. Neurogenetic contributions to amyloid beta and tau spreading in the human cortex. Nature medicine. 2018;24:1910-8.
[8] Müller-Gärtner HW, Links JM, Prince JL, Bryan RN, McVeigh E, Leal JP, et al. Measurement of radiotracer concentration in brain gray matter using positron emission tomography: MRI-based correction for partial volume effects. Journal of Cerebral Blood Flow & Metabolism. 1992;12:571-83.