Cargando…

Differential trajectories of hypometabolism across cognitively-defined Alzheimer’s disease subgroups

Disentangling biologically distinct subgroups of Alzheimer’s disease (AD) may facilitate a deeper understanding of the neurobiology underlying clinical heterogeneity. We employed longitudinal [(18)F]FDG-PET standardized uptake value ratios (SUVRs) to map hypometabolism across cognitively-defined AD...

Descripción completa

Detalles Bibliográficos
Autores principales: Groot, Colin, Risacher, Shannon L., Chen, J.Q. Alida, Dicks, Ellen, Saykin, Andrew J., Mac Donald, Christine L., Mez, Jesse, Trittschuh, Emily H., Mukherjee, Shubhabrata, Barkhof, Frederik, Scheltens, Philip, van der Flier, Wiesje M., Ossenkoppele, Rik, Crane, Paul K.
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Elsevier 2021
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8238088/
https://www.ncbi.nlm.nih.gov/pubmed/34153688
http://dx.doi.org/10.1016/j.nicl.2021.102725
Descripción
Sumario:Disentangling biologically distinct subgroups of Alzheimer’s disease (AD) may facilitate a deeper understanding of the neurobiology underlying clinical heterogeneity. We employed longitudinal [(18)F]FDG-PET standardized uptake value ratios (SUVRs) to map hypometabolism across cognitively-defined AD subgroups. Participants were 384 amyloid-positive individuals with an AD dementia diagnosis from ADNI who had a total of 1028 FDG-scans (mean time between first and last scan: 1.6 ± 1.8 years). These participants were categorized into subgroups on the basis of substantial impairment at time of dementia diagnosis in a specific cognitive domain relative to the average across domains. This approach resulted in groups of AD-Memory (n = 135), AD-Executive (n = 8), AD-Language (n = 22), AD-Visuospatial (n = 44), AD-Multiple Domains (n = 15) and AD-No Domains (for whom no domain showed substantial relative impairment; n = 160). Voxelwise contrasts against controls revealed that all AD-subgroups showed progressive hypometabolism compared to controls across temporoparietal regions at time of AD diagnosis. Voxelwise and regions-of-interest (ROI)-based linear mixed model analyses revealed there were also subgroup-specific hypometabolism patterns and trajectories. The AD-Memory group had more pronounced hypometabolism compared to all other groups in the medial temporal lobe and posterior cingulate, and faster decline in metabolism in the medial temporal lobe compared to AD-Visuospatial. The AD-Language group had pronounced lateral temporal hypometabolism compared to all other groups, and the pattern of metabolism was also more asymmetrical (left < right) than all other groups. The AD-Visuospatial group had faster decline in metabolism in parietal regions compared to all other groups, as well as faster decline in the precuneus compared to AD-Memory and AD-No Domains. Taken together, in addition to a common pattern, cognitively-defined subgroups of people with AD dementia show subgroup-specific hypometabolism patterns, as well as differences in trajectories of metabolism over time. These findings provide support to the notion that cognitively-defined subgroups are biologically distinct.