Neuroimaging biomarkers of glial activation for predicting the annual cognitive function decline in patients with Alzheimer's disease.

ABSTRACT

BACKGROUND: Glial activation is central to the pathogenesis of Alzheimer's disease (AD). However, researchers have not demonstrated its relationship to longitudinal cognitive deterioration. We aimed to compare the prognostic effects of baseline positron emission tomography (PET) imaging of glial activation and amyloid/tau pathology on the successive annual cognitive decline in patients with AD. METHODS: We selected 17 patients diagnosed with mild cognitive impairment or AD. We assessed the annual changes in global cognition and memory. Furthermore, we assessed the predictive effects of baseline amyloid and tau pathology indicated by cerebrospinal fluid (CSF) concentrations and PET imaging of glial activation (11C-DPA-713-binding potential in the area of Braak 1-3 [11C-DPA-713-BPND]) on global cognition and memory using a stepwise regression analysis. RESULTS: The final multiple regression model of annual changes in global cognition and memory scores included 11C-DPA-713-BPND as the predictor. The CSF Aß42/40 ratios and p-tau concentrations were removed from the final model. In stepwise Bayesian regression analysis, the Bayes factor-based model comparison suggested that the best model included 11C-DPA-713-BPND as the predictor of decline in global cognition and memory. CONCLUSIONS: Translocator protein-PET imaging of glial activation is a stronger predictor of AD clinical progression than the amount of amyloid/tau pathology measured using CSF concentrations. Glial activation is the primary cause of tau-induced neuronal toxicity and cognitive deterioration, thereby highlighting the potential of blocking maladaptive microglial responses as a therapeutic strategy for AD treatment.