Multicentric F-18-PI-2620 PET for in vivo Braak staging of tau pathology in Alzheimer’s disease

1053 Introduction: Tau aggregates accumulate in the Alzheimer’ disease (AD) brain according to the established Braak staging scheme and spread from transentorhinal over limbic regions to the neocortex. To impact AD patient management, an in vivo tool for tau Braak staging is desired. Early-generation tau tracers have limited performance in detecting early tau stages. Thus, we tested the respective capability of the next-generation tau tracer F-18-PI-2620. We analyzed F-18-PI-2620 PET data of beta-amyloid positive 37 AD dementia patients (69 ± 9 years, 20 females, MMSE scores: 20 ± 6) and those of 19 healthy controls (63 ± 10 years, 10 females, MMSE scores: 29 ± 1). The data were acquired in four different centers (Leipzig, Germany; Melbourne, Australia; Munich, Germany; New Haven, USA). We applied kinetic modeling of 0-60min p.i. PET data using MRTM2 with lower cerebellum as reference region. We used the tau Braak staging atlas of Schwarz et al. 2016 to extract respective DVRs. Controls were used to define stage-dependent PET positivity (> mv + 2.5 sd). In addition, we condensed the six-stage model data into an established four-class model (0, I + II, III + IV and V + VI). Stage-dependent PET positivity widely followed the Braak scheme (except for Braak stage III). The PET positivity frequency declined from Braak I (43 %), II (35 %), III (59 %), IV (30 %), V (24 %) to VI (19 %). A hierarchical model (allowing a stage>I to be defines as positive only in case the lower stage(s) is/are positive) was fulfilled by 54 % AD dementia cases for the six-stage model, while this was the case in 78% of the cases for the four-class Braak staging model. Six cases (16 %) showed a “hippocampal sparing” tauopathy pattern. F-18-PI-2620 PET appears to be capable to perform Braak tau staging of AD in vivo. The results should benefit from further analysis such as correction for partial volume effect.