[18F]Florbetapir Pet Uptake Differences in White Matter Across the Alzheimer's Disease Continuum: From Cognitively Normal to MCI and Dementia

Alzheimer's disease (AD) is widely recognized for hallmark gray matter (GM) changes, including extracellular amyloid plaques and neurofibrillary tangles. However, recent neuroimaging research suggests that white matter (WM) aberrations may also play a significant role in AD risk and development. This study investigates AD dementia by examining regional covariance patterns of WM changes, expanding beyond the conventional focus on GM. Using [18F]Florbetapir ([18F]FBP) Positron Emission Tomography (PET) data from 716 participants in the Alzheimer's Disease Neuroimaging Initiative (ADNI), we analyzed inter-subject covarying patterns to reveal latent WM sources associated with AD dementia. Employing a model order of 100 and independent component analysis (ICA), we identified 13 highly reliable and distinct WM components, along with 80 GM components, enabling a comparative analysis of group effects across the AD continuum and highlighting the differential significance of WM integrity. We assessed the discrimination ability of WM components between cognitively normal (CN), mild cognitive impairment (MCI), and AD dementia diagnostic groups. Statistical analysis on 13 WM component weights highlighted significant diagnostic contrasts, with 11 components differentiating AD dementia from CN, 10 distinguishing AD dementia from MCI, and 8 distinguishing MCI from CN. Notably, results were replicable, reinforcing the robustness of latent WM sources. Our findings underscore the importance of WM integrity in AD pathology, suggesting WM degeneration as a potentially integral feature of the AD continuum. This work demonstrates the potential of combining ICA with [18F]FBP PET to illuminate nuanced neurobiological mechanisms underlying AD-related WM degeneration, offering valuable insights that may inform future diagnostic approaches across the AD continuum.