Β-amyloid indirectly affects cognition through tau and hippocampal atrophy in Alzheimer’s disease
Poster No:
247
Submission Type:
Abstract Submission
Authors:
Sofia Fernandez Lozano1, Vladimir Fonov1, Joseph Therriault2, Nesrine Rahmouni2, Stijn Servaes2, Jenna Stevenson2, Nina Marguerita Poltronetti2, Pedro Rosa-Neto2, D Louis Collins1
Institutions:
1McConnell Brain Imaging Centre, Montreal Neurological Institute, Montreal, Quebec, 2Translational Neuroimaging laboratory, McGill Centre for Studies in Aging, Montreal, Quebec
First Author:
Sofia Fernandez Lozano
McConnell Brain Imaging Centre, Montreal Neurological Institute
Montreal, Quebec
McConnell Brain Imaging Centre, Montreal Neurological Institute
Montreal, Quebec
Co-Author(s):
Joseph Therriault
Translational Neuroimaging laboratory, McGill Centre for Studies in Aging
Montreal, Quebec
Translational Neuroimaging laboratory, McGill Centre for Studies in Aging
Montreal, Quebec
Nesrine Rahmouni
Translational Neuroimaging laboratory, McGill Centre for Studies in Aging
Montreal, Quebec
Translational Neuroimaging laboratory, McGill Centre for Studies in Aging
Montreal, Quebec
Stijn Servaes
Translational Neuroimaging laboratory, McGill Centre for Studies in Aging
Montreal, Quebec
Translational Neuroimaging laboratory, McGill Centre for Studies in Aging
Montreal, Quebec
Jenna Stevenson
Translational Neuroimaging laboratory, McGill Centre for Studies in Aging
Montreal, Quebec
Translational Neuroimaging laboratory, McGill Centre for Studies in Aging
Montreal, Quebec
Nina Marguerita Poltronetti
Translational Neuroimaging laboratory, McGill Centre for Studies in Aging
Montreal, Quebec
Translational Neuroimaging laboratory, McGill Centre for Studies in Aging
Montreal, Quebec
Pedro Rosa-Neto
Translational Neuroimaging laboratory, McGill Centre for Studies in Aging
Montreal, Quebec
Translational Neuroimaging laboratory, McGill Centre for Studies in Aging
Montreal, Quebec
Introduction:
Alzheimer's disease (AD) starts with gradual cerebral accumulation of β-amyloid (Aβ) decades before the onset of clinical symptoms. Recent biomarker models describe a sequence of Aβ aggregation, tau pathology, neurodegeneration, and eventual cognitive decline (Selkoe & Hardy, 2016). Medial temporal atrophy, characterized by volume loss in the hippocampus (HC), is an early effect of neurodegeneration in AD disease progression (Rao et al., 2022).
Structural equation modeling (SEM) is a statistical framework that combines factor analysis and multiple regression to examine multiple relationships among variables (Deng et al., 2018). Variables in SEM are represented as latent constructs and the relationships among them are expressed as a system of equations that examines direct and indirect pathways in a single integrated model.
Structural equation modeling (SEM) is a statistical framework that combines factor analysis and multiple regression to examine multiple relationships among variables (Deng et al., 2018). Variables in SEM are represented as latent constructs and the relationships among them are expressed as a system of equations that examines direct and indirect pathways in a single integrated model.
Methods:
We used SEM (lavaan R package (Rosseel, 2012) to explore the complex relationships between the neurobiological factors in the early stages preceding AD dementia in a longitudinal cohort of cognitively healthy (n=147) and mild cognitive impaired (n=64) participants from the TRIAD cohort (Stevenson et al., 2022). Our sample comprised 333 timepoints of neuropsychological evaluation, structural MRI, Aβ- and tau-PET. We explored the direct and indirect effects of Aβ on cognition (Montreal Cognitive assessment, MoCA) with tau pathology and HCl atrophy as potential mediators.
Aβ and tau were measured with [18F]AZD4694 and [18F]MK6240 PET respectively, using the cerebellar grey matter as the reference region for [18F]AZD4694 and the inferior gray matter of cerebellum as the reference for [18F]MK6240 SUVR calculation. Aβ accumulation was quantified by the average SUVR in the neocortex. Tau accumulation was quantified as the average SUVR value for 6 Braak stage masks (Braak & Braak, 1991).
Neurodegeneration was characterized by the HC-to-Ventricle ratio (HVR), a HC integrity measure that leverages the idea of ex-vacuo dilation in a single value composed of the ratio of the HC volume and the sum of the volumes of HC and the temporal horn of the lateral ventricle (Schoemaker et al., 2019). We averaged the HVR values of the left and right hemispheres and inverted the result (1–HVR) to convert the integrity biomarker into a measure of atrophy.
In the SEM, we included age, sex, the number of APOE4 alleles and education as covariates. Finally, we calculated the standardized coefficients and the mediation proportion for the direct and indirect effects in our model.
Aβ and tau were measured with [18F]AZD4694 and [18F]MK6240 PET respectively, using the cerebellar grey matter as the reference region for [18F]AZD4694 and the inferior gray matter of cerebellum as the reference for [18F]MK6240 SUVR calculation. Aβ accumulation was quantified by the average SUVR in the neocortex. Tau accumulation was quantified as the average SUVR value for 6 Braak stage masks (Braak & Braak, 1991).
Neurodegeneration was characterized by the HC-to-Ventricle ratio (HVR), a HC integrity measure that leverages the idea of ex-vacuo dilation in a single value composed of the ratio of the HC volume and the sum of the volumes of HC and the temporal horn of the lateral ventricle (Schoemaker et al., 2019). We averaged the HVR values of the left and right hemispheres and inverted the result (1–HVR) to convert the integrity biomarker into a measure of atrophy.
In the SEM, we included age, sex, the number of APOE4 alleles and education as covariates. Finally, we calculated the standardized coefficients and the mediation proportion for the direct and indirect effects in our model.
Results:
The demographic data of our sample is presented in Table 1. The standardized coefficients for the mediation model are represented in Fig. 1.
MoCA scores were affected by tau (β=-0.64, βstd=-0.42, p<0.01), HC atrophy (β=-7.47, βstd=-0.25, p<0.01), and education (β=-0.09, βstd=-0.13, p<0.01), but not by Aβ (β=-0.3, βstd=-0.06, p=0.31), age (β=-0.03, βstd=-0.08, p=0.14) or sex (β=0.08, βstd=0.02, p=0.73)
Aβ was affected by APOE4 (β=0.29, βstd=0.32, p<0.01) and age (β=0.01, βstd=0.19, p<0.01), but not by sex (β=-0.03, βstd=-0.03, p=0.54).
Tau was affected by Aβ (β=1.97, βstd=0.6, p<0.01), APOE4 (β=0.53, βstd=0.18, p<0.01), and age (β=-0.02, βstd=-0.1, p=0.02), but not sex (β=-0.11, βstd=-0.03, p=0.44).
HC atrophy was affected by Aβ (β=0.02, βstd=0.12, p=0.04), age (β<0.01, βstd=0.49, p<0.01), sex (β=0.04, βstd=0.20, p<0.01), and APOE4 (β=0.02, βstd=0.12, p=0.01), but not tau (β<0.01, βstd=0.08, p=0.17).
The mediation proportion showed that 73.8% of the total effect of Aβ was mediated through the indirect effect of tau and 8.7% through HC atrophy.
MoCA scores were affected by tau (β=-0.64, βstd=-0.42, p<0.01), HC atrophy (β=-7.47, βstd=-0.25, p<0.01), and education (β=-0.09, βstd=-0.13, p<0.01), but not by Aβ (β=-0.3, βstd=-0.06, p=0.31), age (β=-0.03, βstd=-0.08, p=0.14) or sex (β=0.08, βstd=0.02, p=0.73)
Aβ was affected by APOE4 (β=0.29, βstd=0.32, p<0.01) and age (β=0.01, βstd=0.19, p<0.01), but not by sex (β=-0.03, βstd=-0.03, p=0.54).
Tau was affected by Aβ (β=1.97, βstd=0.6, p<0.01), APOE4 (β=0.53, βstd=0.18, p<0.01), and age (β=-0.02, βstd=-0.1, p=0.02), but not sex (β=-0.11, βstd=-0.03, p=0.44).
HC atrophy was affected by Aβ (β=0.02, βstd=0.12, p=0.04), age (β<0.01, βstd=0.49, p<0.01), sex (β=0.04, βstd=0.20, p<0.01), and APOE4 (β=0.02, βstd=0.12, p=0.01), but not tau (β<0.01, βstd=0.08, p=0.17).
The mediation proportion showed that 73.8% of the total effect of Aβ was mediated through the indirect effect of tau and 8.7% through HC atrophy.
·Table 1: Demographic data.
·Figure 1: Mediation model with standardized coefficients.
Conclusions:
In our early AD cohort, we found that the direct impact of Aβ aggregation in the brain on cognitive function is not statistically significant, and rather its influence on cognition is largely explained by its mediating effects of tau pathology and, to a lesser extent, HC atrophy. Our results support Aβ aggregation as an early, upstream event in the development of dementia due to AD.
Disorders of the Nervous System:
Neurodegenerative/ Late Life (eg. Parkinson’s, Alzheimer’s) 1
Modeling and Analysis Methods:
Multivariate Approaches 2
Keywords:
Aging
Modeling
MRI
Positron Emission Tomography (PET)
Other - Alzheimer's disease, Tau, Amyloid, Hippocampus, Structural Equation Modeling, Mediation Models
1|2Indicates the priority used for review
Provide references using author date format
Deng, L., Yang, M., & Marcoulides, K. M. (2018). Structural Equation Modeling With Many Variables: A Systematic Review of Issues and Developments. Frontiers in Psychology, 9, 580. https://doi.org/10.3389/fpsyg.2018.00580
Rao, Y. L., Ganaraja, B., Murlimanju, B. V., Joy, T., Krishnamurthy, A., & Agrawal, A. (2022). Hippocampus and its involvement in Alzheimer’s disease: A review. 3 Biotech, 12(2), 55. https://doi.org/10.1007/s13205-022-03123-4
Rosseel, Y. (2012). lavaan: An R Package for Structural Equation Modeling. Journal of Statistical Software, 48, 1–36. https://doi.org/10.18637/jss.v048.i02
Schoemaker, D., Buss, C., Pietrantonio, S., Maunder, L., Freiesleben, S. D., Hartmann, J., Collins, D. L., Lupien, S., & Pruessner, J. C. (2019). The hippocampal-to-ventricle ratio (HVR): Presentation of a manual segmentation protocol and preliminary evidence. NeuroImage, 203, 116108. https://doi.org/10.1016/j.neuroimage.2019.116108
Selkoe, D. J., & Hardy, J. (2016). The amyloid hypothesis of Alzheimer’s disease at 25 years. EMBO Molecular Medicine, 8(6), 595–608. https://doi.org/10.15252/emmm.201606210
Stevenson, J., Rahmouni, N., Chamoun, M., Benedet, A. L., Stevenson, A., Pallen, V., Therriault, J., Lussier, F. Z., Tissot, C., Bezgin, G., Pascoal, T. A., Vitali, P., Gauthier, S., & Rosa-Neto, P. (2022). TRIAD multi-dimensional biobank for biomarker discovery. Alzheimer’s & Dementia, 18(S6), e067924. https://doi.org/10.1002/alz.067924