Resilience in Alzheimer’s disease

Alzheimer’s disease (AD) is characterized by considerable heterogeneity in the relationship between neuropathology, brain structure and cognitive impairment. Some individuals maintain cognitive function despite substantial pathology, a phenomenon commonly attributed to reserve and resilience. This thesis aimed to improve understanding of resilience in AD by investigating statistical measurement methods, determinant factors and underlying mechanisms across the AD continuum. First, we evaluated the conceptualization, implementation and utility of residual-based operationalizations of resilience-related constructs. Although residual measures captured clinically meaningful variability, their interpretation was limited by substantial methodological heterogeneity, inherent statistical dependencies and limited incremental value over more complete explicit models. We therefore recommend direct modelling of determinants and moderation effects within explicitly specified models (Chapters 1 and 2). Second, using longitudinal data from large clinical cohorts, we investigated modifiable lifestyle and behavioral, demographic, brain and biological factors associated with cognitive and brain resilience. Individuals with more years of formal education showed higher cognitive performance at similar levels of AD pathology; however, once dementia symptoms were present and tau burden or subsequent atrophy was high, cognitive decline accelerated in those with higher educational attainment (Chapters 3 and 5). Greater engagement in past and current cognitively stimulating activities, and greater current physical activity, were associated with better cognitive performance at comparable levels of brain atrophy, but showed no consistent association with longitudinal cognitive decline or clinical progression (Chapter 4). Larger intracranial volume was associated with higher baseline cognition independent of brain atrophy (Chapter 3), while greater cortical thickness was associated with slower cognitive decline independent of tau burden (Chapter 5). Finally, biomarkers reflecting co-existing biological processes, particularly vascular integrity and axonal degeneration, modified the relationship between tau pathology and subsequent brain atrophy and cognitive decline (Chapter 6).