Neuroimaging Predictors of Cognitive Resilience against Alzheimer's Disease Pathology - PubMed (original) (raw)
Neuroimaging Predictors of Cognitive Resilience against Alzheimer's Disease Pathology
McKenna E Williams et al. Ann Neurol. 2025 Jun.
Abstract
Objective: Some individuals demonstrate greater cognitive resilience-the ability to maintain cognitive performance despite adverse brain-related changes-through as yet unknown mechanisms. We examined whether cortical thickness in several brain regions confers resilience against cognitive decline in amyloid-positive adults by moderating the effects of thinner cortex in Alzheimer's disease (AD)-related brain regions and of higher levels of tau.
Methods: Amyloid-positive participants from the Alzheimer's Disease Neuroimaging Initiative with relevant imaging data were included (n = 160, observations = 473). Risk factors included an AD brain signature and cerebrospinal fluid phosphorylated tau. Cognitive measures were episodic memory and executive function composites. Mixed effects models tested whether region-specific cortical thickness moderated relationships between markers of AD risk and memory or executive function.
Results: Cross-sectionally, thicker cortex in 8 regions minimized the negative impact of thinner cortex/smaller volume in AD signature regions on executive function. Longitudinally, higher baseline thickness in a composite of these 8 regions predicted less memory decline (p = 0.007) and weakened negative effects of phosphorylated tau on memory decline (p = 0.014), independent of baseline cognition and risk markers.
Interpretation: We identified 8 cortical regions that appear to confer cognitive resilience cross-sectionally and longitudinally in the face of established indicators of AD pathology. Brain regions fostering executive function may enable compensation in later memory performance and confer cognitive resilience against effects of phosphorylated tau and AD-related cortical changes. These "resilience" regions suggest the value of focusing on brain regions beyond only those determined to be AD-related and may partially explain variability in AD-related cognitive trajectories. ANN NEUROL 2025;97:1038-1050.
© 2025 The Author(s). Annals of Neurology published by Wiley Periodicals LLC on behalf of American Neurological Association.
Conflict of interest statement
Nothing to report.
Figures
FIGURE 1
Distributions of cerebrospinal fluid (CSF), imaging, and cognitive measures by diagnostic group. ****p < 0.0001, ***p < 0.001, **p < 0.01, ns = not significant. [Color figure can be viewed at
]
FIGURE 2
Cortical thickness regions of interest (ROIs) that significantly moderate relationship between Alzheimer's disease (AD) signatures and executive functioning performance in cross‐sectional models. Models control for p‐tau, age, and repeated assessments. Figures display model estimates for higher cortical thickness values (+1 standard deviation from the mean) depicted in blue/solid line, mean cortical thickness values in purple/large dashed line, and low cortical thickness values (−1 standard deviation from the mean) in orange/small dashed line. Lower AD signature values indicate lower cortical thickness in AD‐related regions. Standardized mean difference (SMD) effect sizes are reported for each ROI. [Color figure can be viewed at
]
FIGURE 3
Cortical thickness regions of interest (ROIs) that significantly moderate relationship between Alzheimer's disease signatures and executive functioning performance. These ROIs comprise the resilience composite used in longitudinal analyses. Bilateral ROIs are used, only left hemisphere is displayed. Stronger effect sizes (standardized mean difference) of interaction terms for each ROI are depicted in darker shades of red. [Color figure can be viewed at
]
FIGURE 4
Higher baseline resilience composite scores are associated with slower memory decline. Model controls for the effects of risk (Alzheimer's disease [AD] signatures, p‐tau, age) and education over time as well as repeated assessments within participants (model 5). Estimates for higher cortical thickness values (+1 standard deviation from the mean) are depicted in blue/solid line, mean cortical thickness values in purple/large dashed line, and low cortical thickness values (−1 standard deviation from the mean) in orange/small dashed line. Lower AD signature values indicate lower cortical thickness in AD‐related regions. [Color figure can be viewed at
]
FIGURE 5
Baseline resilience composite and baseline cerebrospinal fluid (CSF) p‐tau interact on memory performance over time. A and B depict the same interaction term, with differing primary moderators. Mean and +/− 1 standard deviation values of CSF p‐tau for the longitudinal sample are reported. [Color figure can be viewed at
]
References
- Bennett DA, Schneider JA, Arvanitakis Z, et al. Neuropathology of older persons without cognitive impairment from two community‐based studies. Neurology 2006;66:1837–1844. -PubMed
- Riley KP, Snowdon DA, Desrosiers MF, Markesbery WR. Early life linguistic ability, late life cognitive function, and neuropathology: findings from the nun study. Neurobiol Aging 2005;26:341–347. -PubMed
MeSH terms
Substances
Grants and funding
- R01 AG064955/AG/NIA NIH HHS/United States
- U19 AG024904/AG/NIA NIH HHS/United States
- R01 AG076838/AG/NIA NIH HHS/United States
- K01 AG063805/AG/NIA NIH HHS/United States
- T32 MH018399/MH/NIMH NIH HHS/United States
- R01 AG050595/AG/NIA NIH HHS/United States
LinkOut - more resources
Full Text Sources
Medical