Amyloid imaging demonstrates the in vivo presence of amyloid-β (Aβ) deposits in the aging human brain but it is still unknown which structural forms and modifications of Aβ are detected. In Alzheimer's disease, most amyloid deposits are predominantly composed of Aβ ending at amino acid residues Val40 or Ala42. It has been reported that Aβ40 is largely restricted to neuritic plaques while Aβ42 may be deposited in amyloid plaques of all types, and is often the sole component of diffuse plaques. The distinction is important as it is mainly the neuritic plaques that correlate with cognitive impairment while diffuse plaques may be the initial type of Aβ deposited. Whether PET amyloid ligands such as florbetapir-18F (Amyvid) are partially or wholly selective for brain deposits of Aβ40 or Aβ42 is currently unknown. We compared antemortem florbetapir PET cortical/cerebellar signal intensity (SUVr) of 55 subjects with postmortem biochemical (ELISA) measurements employing specific antibodies against Aβ40 and Aβ42. Spearman's univariable correlations were significant for both Aβ40 and Aβ42, but were much stronger for Aβ42. Multiple linear regression showed significance only for Aβ42. These results suggest that florbetapir binds only weakly, if at all, to Aβ40. This may be in part due to the higher likelihood for Aβ42 to be present in a β-pleated sheet tertiary structure, or to differences between Aβ40 and Aβ42 in β-pleated sheet tertiary or quaternary structure.
Keywords: Alzheimer’s disease; autopsy; diagnosis; diffuse plaque; neuritic plaque.