Abstract
OBJECTIVES: To examine trajectories of depression and apathy over a 5-year follow-up period in (prodromal) Alzheimer's disease (AD), and to relate these trajectories to AD biomarkers. METHODS: The trajectories of depression and apathy (measured with the Neuropsychiatric Inventory or its questionnaire) were separately modeled using growth mixture models for two cohorts (National Alzheimer's Coordinating Center, NACC, n = 22 760 and Alzheimer's Disease Neuroimaging Initiative, ADNI, n = 1 733). The trajectories in ADNI were associated with baseline CSF AD biomarkers (Aβ(42,) t-tau, and p-tau) using bias-corrected multinomial logistic regression. RESULTS: Multiple classes were identified, with the largest classes having no symptoms over time. Lower Aβ(42) and higher tau (ie, more AD pathology) was associated with increased probability of depression and apathy over time, compared to classes without symptoms. Lower Aβ(42) (but not tau) was associated with a steep increase of apathy, whereas higher tau (but not Aβ(42) ) was associated with a steep decrease of apathy. DISCUSSION: The trajectories of depression and apathy in individuals on the AD spectrum are associated with AD biomarkers.