Background: The relationship between in vivo imaging measures of fibrillar amyloid-beta (Aβ) deposition and postmortem assessments of amyloid load is not fully understood. This study examined correspondence between in vivo [C-11]PiB PET retention measures and postmortem assessments of Aβ plaque load in the precuneus.
Methods: [C-11]PiB PET was performed for 12 subjects representing a range of clinical states from cognitively normal to AD dementia. Subjects later came to autopsy with an average imaging-to-death interval of 36±24 months. Postmortem Aβ plaque load was assessed using Aβ immunohistochemistry (IHC, antibody clone 4G8) and a highly fluorescent derivative of PiB (6-CN-PiB) applied to 12-μm paraffin sections of the precuneus. A systematic anatomical match of postmortem dissected precuneus region was identified on the antemortem MR and co-registered [C-11]PiB PET. PiB SUVR values were corrected for atrophy-related cerebrospinal fluid (CSF) dilution using a 2-component (gray+white matter and CSF) correction. PiB SUVR and histopathology correlations were assessed using Pearson correlations.
Results: In the precuneus region from 12 PiB PET imaged autopsy cases, SUVR PET corresponded with 6-CN-PiB plaque load (R 2 =0.802) and less strongly with Aβ IHC plaque load (R 2 =0.65). However, in cases where total plaque load exceeded 7% area by either Aβ IHC or 6-CN-PiB, SUVR values appeared to plateau. Separate analysis of plaque load with respect to either diffuse or cored plaques revealed that total plaque load was dominated by diffuse plaques (R 2 =0.94), particularly for cases at the SUVR plateau level.
Conclusions: These preliminary results suggest that in brain regions where pathology load exceeds 7% area covered with plaques, and the majority of plaques are diffuse, there is no corresponding increase in [C-11]PiB retention. This could be due to saturation of the in vivo PiB PET signal with very high Aβ plaque load, but more likely is due to: 1) low PiB PET detection sensitivity for diffuse plaques at nM in vivo PiB concentrations and 2) overestimation of fibrillar Aβ content by semi-quantitative plaque load analyses. This observation also needs to be considered when investigating the detection threshold of [C-11]PiB PET positivity in relation to underlying diversity of Aβ plaque pathology.
© 2014 Published by Elsevier Inc.