Comparison of regional flortaucipir PET SUVr values to quantitative tau histology and quantitative tau immunoassay in patients with Alzheimer’s disease pathology: A clinico-pathological study

Objectives: PET tracers for tau pathology in Alzheimer’s disease (AD) are of increasing interest as diagnostics and as markers of disease severity. However, there have been few studies validating tau PET imaging relative to the standard of pathological assessment at autopsy. The objective of this study was to make a quantitative comparison of 18F-flortaucipir (F18-AV-1451) PET uptake across a range of brain regions obtained in end-of-life subjects to density of tau pathology observed at autopsy in these same regions.

Methods: Patients at end-of-life were recruited for this pilot phase of a pivotal clinical trial. Participants had PET imaging consisting of an injection of ~370 mBq flortaucipir F18 followed by imaging from 80-100 minutes after injection. After death, blocks of brain tissue from representative neocortical and limbic regions were obtained for assessment of the density and extent of tau pathology. Subject-specific PET regions of interest corresponding to pathology blocks were constructed for 14 regions in each hemisphere (28 regions per subject). Regions were hand drawn on the PET images of each subject using photographs of the gross pathology blocks to identify the corresponding anatomy on the PET scan. SUVr values were calculated using a cerebellar reference region. Tau pathology was assessed using a quantitative antibody capture assay (Histelide) designed for paraffin-embedded tissue using the AT8 phospho-tau antibody (Postupna et al; Brain Pathol. 2012). All image and pathology assessments were conducted blinded relative to the other measure.

Results: 3 participants with a clinical diagnosis of Alzheimer’s disease were included in this study. The intervals between imaging and death for the three cases were 4, 12 and 27 days. Two cases were found to be Braak stage 5 at autopsy and one was Braak stage 6, but examination of individual tissue blocks showed that tau pathology ranged from absent in some blocks to frequent others. Regional SUVr values ranged from 1.02 to 2.60 and quantitative immunoassay values, normalized to gray matter (GM) area, ranged from -0.51 to 83.01 ug p-nitrophenol/cm2 of GM. There was a highly significant correlation across all patients between regional SUVr and quantitative tau immunoassay (pearson r = 0.81; p<0.0001). There was also a highly significant correlation between regional flortaucipir SUVr and tau immunoassay for each individual case (r = 0.54; 0.73; 0.75; p<0.005 for all).

Conclusion: The results of this pilot study support a quantitative relationship between flortaucipir SUVr values and quantitative measures of tau pathology at autopsy. Additional research including more cases are needed to confirm the generalizability of these results. Research Support: This study was supported by Avid Radiopharmaceuticals, a wholly owned subsidiary of Eli Lilly and Company.