A bispecific Tribody PET radioligand for visualization of amyloid-beta protofibrils - a new concept for neuroimaging.

Antibodies are highly specific for their target molecules, but their poor brain penetrance has restricted their use as PET ligands for imaging of targets within the CNS. The aim of this study was to develop an antibody-based radioligand, using the Tribody format, for PET imaging of soluble amyloid-beta (Aβ) protofibrils, which are suggested to cause neurodegeneration in Alzheimer's disease. Antibodies, even when expressed in smaller engineered formats, are large molecules that do not enter the brain in sufficient amounts for imaging purposes. Hence, their transport across the blood-brain barrier (BBB) needs to be facilitated, for example through interaction with the transferrin receptor (TfR). Thus, a Fab fragment of the TfR antibody 8D3 was fused with two single chain variable fragments (scFv) of the Aβ protofibril selective antibody mAb158. Five Tribody proteins (A1-A5) were generated with different linkers between the Fab-8D3 and scFv-158. All proteins bound to TfR and Aβ protofibrils in vitro. Three of the proteins (A1-A3) were radiolabeled with iodine-125 and studied ex vivo in wild-type (wt) and transgenic mice overexpressing human Aβ. The systemic pharmacokinetics were similar with half-lives in blood of around 9h for all three ligands. Brain concentrations at 2h were around 1% of the injected dose per gram brain tissue, which is similar to what is observed for small molecular radioligands and at least 10-fold higher than antibodies in general. At 72h, transgenic mice showed higher concentrations of radioactivity in the brain than wt mice (12, 15- and 16-fold for A1, A2 and A3 respectively), except in the cerebellum, an area largely devoid of Aβ pathology. A3 was then labelled with iodine-124 for in vivo positron emission tomography (PET) imaging. Brain concentrations were quantified in six different regions showing a clear distinction both quantitatively and visually between wt and transgenic mice and a good correlation with Aβ pathology. We have thus produced a recombinant, bispecific protein, actively transported into the brain, for PET imaging within the CNS. In a longer perspective, this technique may enable imaging of other proteins involved in neurodegenerative diseases for which imaging agents are completely lacking today.