. 18F-THK523: a novel in vivo tau imaging ligand for Alzheimer's disease. Brain. 2011 Apr;134(Pt 4):1089-100. PubMed.


Please login to recommend the paper.


Make a Comment

To make a comment you must login or register.

Comments on this content

  1. This paper by Fodero-Tavoletti and her colleagues from Melbourne and Sendai represents a landmark in the search for in vivo tau imaging agents. Although it is not yet clear that this agent will succeed in human studies, it certainly represents the most advanced tau imaging agent reported in the literature to-date. There are several promising characteristics of THK523 including: 1) a high affinity for synthetic K18 tau preparations (1.67 nM) and a lower affinity for synthetic Aβ1-42 fibrils (20.7 nM); 2) THK523 appears to specifically label neurofibrillary tangles in human brain tissue without appreciable labeling of diffuse Aβ plaques; 3) THK523 is fairly lipophilic and has a low molecular weight, allowing it to cross the blood-brain barrier fairly well with an initial brain level of 2.75 percent injected dose (ID)/g in mice; and 4) THK shows ~50 percent higher in-vivo retention in tau transgenic mouse brain than in wild-type mouse brain. While these data are very promising and certainly justify in vivo human studies, pending appropriate toxicology studies, several things must be kept in mind.

    First, is that the overall retention in human brain is determined by a ratio of the total binding sites (Bmax) over the affinity (Kd). The data in this paper suggest that ratio would be much higher for tau (1.3) than it is for Aβ1-42 (0.06), but this in vitro data may not be predictive of human studies since it is well-known that synthetic Aβ fibrils possess only 1/500 the number of binding sites that are present on Aβ deposits in human brain (Klunk et al., 2005) and the same may be true of tau. This must be combined with the fact that in many human brains there will be a large excess of fibrillar Aβ compared to deposited tau. Thus, the affinity differences may need to be significantly higher than 12-fold. The histological data showing selectivity in human brain is much more promising, but would have been more useful if frontal cortex samples containing compact Aβ deposits were included in addition to the diffuse deposits seen in hippocampus. This is because the Aβ plaques present in the hippocampus are often diffuse and are not labeled well by PiB and other similar Aβ selective ligands. This may not be relevant for specific detection of tau in the hippocampus, but would be relevant for detection of tau in frontal cortex in the tauopathies.

    The pharmacokinetics of THK523 should be put in context by comparison to other tracers such as PiB. The initial brain uptake (2.75 percent ID/g) is on the edge of acceptability for a PET agent, where values above 4.0 percent ID/g are common in mice (e.g., PiB has a initial uptake of over 8 percent ID/g in mice). Brain clearance is even more of a concern with THK523. The 2-to-30 min ratio in control mouse brain has been used as a predictor of the noise level that will compete with specific signal in human studies. This ratio is 1.8 for THK523. This is significantly lower than most successful PET ligands (e.g., PiB has a ratio of 12) (Mathis et al., 2003).

    The data from transgenic tau mice (Tg4510) are promising and are significantly better than any similar data obtained with any of the human Aβ ligands in transgenic Aβ mice (e.g., APP/PS1). However, this lack of binding of the Aβ ligands to APP/PS1 mice should caution against over-interpreting the specificity of THK523 when comparing the Tg4510 and APP/PS1 models.

    Thus, while the promise of THK523 for in vivo tau assessment can only be determined by human studies at this point, the in vitro and animal data certainly justify such studies while, at the same time, pointing to specific tracer characteristics which leave room for improvement if THK523 does not prove successful in initial human trials.


    . Binding of the positron emission tomography tracer Pittsburgh compound-B reflects the amount of amyloid-beta in Alzheimer's disease brain but not in transgenic mouse brain. J Neurosci. 2005 Nov 16;25(46):10598-606. PubMed.

    . Synthesis and evaluation of 11C-labeled 6-substituted 2-arylbenzothiazoles as amyloid imaging agents. J Med Chem. 2003 Jun 19;46(13):2740-54. PubMed.