Could tau-PET imaging contribute to a better understanding of the different patterns of clinical progression in Alzheimer's disease? A 2-year longitudinal study.

BACKGROUND: Monitoring the progression of Tau pathology makes it possible to study the clinical diversity of Alzheimer's disease. In this 2-year longitudinal PET study, we aimed to determine the progression of [18F]-flortaucipir binding and of cortical atrophy, and their relationships with cognitive decline. METHODS: Twenty-seven AD patients at the mild cognitive impairment/mild dementia stages and twelve amyloid-negative controls underwent a neuropsychological assessment, 3 T brain MRI, and [18F]-flortaucipir PET imaging (Tau1) and were monitored annually over 2 years with a second brain MRI and tau-PET imaging after 2 years (Tau2). We analyzed the progression of tau standardized uptake value ratio (SUVr) and grey matter atrophy both at the regional and voxelwise levels. We used mixed effects models to explore the relations between the progression of SUVr values, cortical atrophy, and cognitive decline. RESULTS: We found an average longitudinal increase in tau SUVr values, except for the lateral temporoparietal cortex where the average SUVr values decreased. Individual analyses revealed distinct profiles of SUVr progression according to temporoparietal Tau1 uptake: high-Tau1 patients demonstrated an increase in SUVr values over time in the frontal lobe, but a decrease in the temporoparietal cortex and a rapid clinical decline, while low-Tau1 patients displayed an increase in SUVr values in all cortical regions and a slower clinical decline. Cognitive decline was strongly associated with the progression of regional cortical atrophy, but only weakly associated with SUVr progression. CONCLUSIONS: Despite a relatively small sample size, our results suggest that tau-PET imaging could identify patients with a potentially "more aggressive" clinical course characterized by high temporoparietal Tau1 SUVr values and a rapid clinical progression. In these patients, the paradoxical decrease in temporoparietal SUVr values over time could be due to the rapid transition to ghost tangles, for which the affinity of the radiotracer is lower. They could particularly benefit from future therapeutic trials, the neuroimaging outcome measures of which deserve to be discussed.