Reactive astrogliosis is associated with higher cerebral glucose consumption in the early Alzheimer's continuum

Purpose Glial activation is one of the earliest mechanisms to be altered in Alzheimer's disease (AD). Glial fibrillary acidic protein (GFAP) relates to reactive astrogliosis and can be measured in both cerebrospinal fluid (CSF) and blood. Plasma GFAP has been suggested to become altered earlier in AD than its CSF counterpart. Although astrocytes consume approximately half of the glucose-derived energy in the brain, the relationship between reactive astrogliosis and cerebral glucose metabolism is poorly understood. Here, we aimed to investigate the association between fluorodeoxyglucose ([F-18]FDG) uptake and reactive astrogliosis, by means of GFAP quantified in both plasma and CSF for the same participants. Methods We included 314 cognitively unimpaired participants from the ALFA + cohort, 112 of whom were amyloid-beta (A beta) positive. Associations between GFAP markers and [F-18]FDG uptake were studied. We also investigated whether these associations were modified by A beta and tau status (AT stages). Results Plasma GFAP was positively associated with glucose consumption in the whole brain, while CSF GFAP associations with [F-18]FDG uptake were only observed in specific smaller areas like temporal pole and superior temporal lobe. These associations persisted when accounting for biomarkers of A beta pathology but became negative in A beta-positive and tau-positive participants (A + T +) in similar areas of AD-related hypometabolism. Conclusions Higher astrocytic reactivity, probably in response to early AD pathological changes, is related to higher glucose consumption. With the onset of tau pathology, the observed uncoupling between astrocytic biomarkers and glucose consumption might be indicative of a failure to sustain the higher energetic demands required by reactive astrocytes.

Automated summary

This study aimed to investigate the association between GFAP and reactive astrogliosis with glucose consumption. The results showed that plasma GFAP was positively associated with glucose consumption in the whole brain, while CSF GFAP was only associated with glucose uptake in specific smaller areas. These associations persisted after accounting for Aβ pathology, but became negative in Aβ and tau-positive participants in similar areas of AD-related hypometabolism.