Change in Amyloid‐β Deposition and Diffusion Tensor Image Analysis along the Perivascular Space Index in Incident Neurodegenerative Disease

OBJECTIVE: We investigated the longitudinal association between cognitive decline, amyloid-β deposition, and perivascular diffusivity using diffusion tensor image analysis along the perivascular space (DTI-ALPS) index. METHODS: This 2-year longitudinal study included 211 participants from the Alzheimer's Disease Neuroimaging Initiative, categorized into sustained cognitively normal, incident mild cognitive impairment, persistent mild cognitive impairment, and incident Alzheimer's disease (AD). The DTI-ALPS index and amyloid-β standardized uptake value ratio (SUVR) from amyloid positron emission tomography were obtained at baseline and follow-up. Linear mixed-effects models were used to examine longitudinal changes in the DTI-ALPS index and their associations with cognitive decline and amyloid-β accumulation. Multiple mediation analyses were performed to evaluate whether amyloid burden and gray matter atrophy mediated the relationship between changes in the DTI-ALPS index and cognitive performance. Additional analyses assessed the effect of baseline amyloid-β positivity. RESULTS: The incident mild cognitive impairment and incident AD groups showed greater longitudinal declines in the DTI-ALPS index than the sustained cognitively normal group. Longitudinal reductions in the DTI-ALPS index were associated with greater cognitive decline and increased amyloid-β accumulation. Mediation analyses showed no significant indirect effects through SUVR or gray matter volume. Although baseline amyloid-β positivity alone was not associated with longitudinal DTI-ALPS changes, significant interactions with clinical progression were observed. INTERPRETATION: Longitudinal alterations in the DTI-ALPS index are associated with cognitive decline and amyloid-β accumulation. These findings support the DTI-ALPS index as a non-invasive imaging marker reflecting disease progression in AD. ANN NEUROL 2026.