Abstract
BACKGROUND: While episodic memory decline is the most common cognitive symptom of Alzheimer's disease (AD), changes in attentional control have also been found to be sensitive to early AD pathology. The relations between longitudinal trajectories of these specific cognitive domains, especially attentional control, and biomarkers of tau and neurodegeneration have not been thoroughly examined. OBJECTIVE: We examined whether baseline tau positron emission tomography (PET) and cortical thickness, relatively later markers within the AD cascade, predicted cross-sectional and longitudinal changes in episodic memory and attentional control. METHODS: Cognitively normal individuals ([Clinical Dementia Rating CDR(®)] = 0; n = 249) at baseline completed a magnetic resonance imaging (MRI), tau PET, and multiple assessments of episodic memory and attentional control. Generalized additive mixed-effects models examined whether tau PET summary measure and cortical thickness signature predicted cross-sectional and longitudinal trajectories of attentional control and episodic memory. RESULTS: Higher tau PET and lower MRI cortical thickness were generally associated with worse cross-sectional cognitive performance. Our exploratory analyses found cortex-wide associations between tau PET and episodic memory, with limited suggestions of region-specific associations with attentional control. On longitudinal follow-up, higher tau PET was associated with a greater decline in episodic memory. CONCLUSIONS: These results indicate that tau PET is particularly sensitive to detecting longitudinal changes in episodic memory. This further informs relevant endpoints for clinical drug trials in cognitively normal individuals. Future studies might consider longer follow-ups and lag associations between changes in AD biomarkers and changes in cognition.