In Vivo (18)F-Flortaucipir PET Does Not Accurately Support the Staging of Progressive Supranuclear Palsy

Progressive supranuclear palsy (PSP) is a neurodegenerative disorder characterized by neuroglial tau pathology. A new staging system for PSP pathology postmortem has been described and validated. We used a data-driven approach to test whether postmortem pathologic staging in PSP can be reproduced in vivo with (18)F-flortaucipir PET. Methods: Forty-two patients with probable PSP and 39 controls underwent (18)F-flortaucipir PET. Conditional inference tree analyses on regional binding potential values identified absent/present pathology thresholds to define in vivo staging. Following the postmortem staging approach for PSP pathology, we evaluated the combinations of absent/present pathology (or abnormal/normal PET signal) across all regions to assign each participant to in vivo stages. ANOVA was applied to analyze differences among means of disease severity between stages. In vivo staging was compared with postmortem staging in 9 patients who also had postmortem confirmation of the diagnosis and stage. Results: Stage assignment was estimable in 41 patients: 10, 26, and 5 patients were classified in stage I/II, stage III/IV, and stage V/VI, respectively, whereas 1 patient was not classifiable. Explorative substaging identified 2 patients in stage I, 8 in stage II, 9 in stage III, 17 in stage IV, and 5 in stage V. However, the nominal (18)F-flortaucipir--derived stage was not associated with clinical severity and was not indicative of pathology staging postmortem. Conclusion: (18)F-flortaucipir PET in vivo does not correspond to neuropathologic staging in PSP. This analytic approach, seeking to mirror in vivo neuropathology staging with PET-to-autopsy correlational analyses, might enable in vivo staging with next-generation tau PET tracers; however, further evidence and comparisons with postmortem data are needed.