@article{oai:repo.qst.go.jp:00046627,
 author = {Harada, Ryuichi and Okamura, Nobuyuki and Furumoto, Shozo and Tago, Tetsuro and Maruyama, Masahiro and Higuchi, Makoto and et.al and 原田 龍一 and 岡村 信行 and 古本 祥三 and 多胡 哲郎 and 丸山 将浩 and 樋口 真人},
 issue = {1},
 journal = {European Journal of Nuclear Medicine and Molecular Imaging},
 month = {Oct},
 note = {Purpose
Extensive deposition of senile plaques and neurofibrillary tangles in the brain is a pathological hallmark of Alzheimer disease (AD). Although several PET imaging agents have been developed for in vivo detection of senile plaques, no PET probe is currently available for selective detection of neurofibrillary tangles in the living human brain. Recently, [18F]THK-523 was developed as a potential in vivo imaging probe for tau pathology. The purpose of this study was to compare the binding properties of [18F]THK-523 and other amyloid imaging agents, including PiB, BF-227 and FDDNP, to synthetic protein fibrils and human brain tissue.
\nMethods
In vitro radioligand binding assays were conducted using synthetic amyloid beta42 and K18 deltaK280-tau fibrils. Nonspecific binding was determined by the addition of unlabelled compounds at a concentration of 2 micro M. To examine radioligand binding to neuropathological lesions, in vitro autoradiography was conducted using sections of AD brain.
\nResults
[18F]THK-523 showed higher affinity for tau fibrils than for Abeta fibrils, whereas the other probes showed a higher affinity for Abeta fibrils. The autoradiographic analysis indicated that [18F]THK-523 accumulated in the regions containing a high density of tau protein deposits. Conversely, PiB and BF-227 accumulated in the regions containing a high density of Abeta plaques.
\nConclusion
These findings suggest that the unique binding profile of [18F]THK-523 can be used to identify tau deposits in AD brain.},
 pages = {125--132},
 title = {Comparison of the binding characteristics of [(18)F]THK-523 and other amyloid imaging tracers to Alzheimer's disease pathology.},
 volume = {40},
 year = {2012}
}