THK-523

• Molecular Formula: C17H15FN2O
• Molecular Weight: 282.31 g/mol
• InChI: InChI=1S/C17H15FN2O/c18-9-10-21-15-6-8-17-13(11-15)3-7-16(20-17)12-1-4-14(19)5-2-12/h1-8,11H,9-10,19H2
• InChIKey: ZGJOXECWZKCGBV-UHFFFAOYSA-N

THK-523: First-Generation Tau PET Tracer

THK-523 (CAS 1354653-91-0) is a first-generation arylquinoline derivative developed as an ¹⁸F-labeled positron emission tomography (PET) radiotracer for the in vivo detection of tau neurofibrillary pathology in Alzheimer's disease (AD) [1]. The compound (6-(2-fluoroethoxy)-2-(4-aminophenyl)quinoline) exhibits selective binding to paired helical filament (PHF)-tau aggregates, forming the foundational chemotype from which later-generation THK-family tracers (THK-5105, THK-5117, THK-5351) were optimized [2]. Preclinical characterization demonstrated that THK-523 crosses the blood–brain barrier in mice, binds recombinant K18ΔK280-tau fibrils with high affinity at two distinct binding sites, and shows significantly higher in vivo retention in tau transgenic mouse brains relative to wild-type or APP/PS1 controls [1]. Although its clinical utility is limited by high white matter retention that complicates visual image interpretation, THK-523 remains a critical reference radioligand for in vitro competitive binding assay development and a benchmark compound in the tau tracer chemical space [3].

Reference radioligand for tau binding site characterization

Foundational arylquinoline chemotype (THK family)

In vitro competitive binding assay standard

• [1] Fodero-Tavoletti MT, Okamura N, Furumoto S, Mulligan RS, Connor AR, McLean CA, Cao D, Rigopoulos A, Cartwright GA, O'Keefe G, Gong S, Adlard PA, Barnham KJ, Rowe CC, Masters CL, Kudo Y, Cappai R, Yanai K, Villemagne VL. 18F-THK523: a novel in vivo tau imaging ligand for Alzheimer's disease. Brain. 2011 Apr;134(Pt 4):1089-100. doi: 10.1093/brain/awr038. View Source
• [2] Okamura N, Furumoto S, Harada R, Tago T, Yoshikawa T, Fodero-Tavoletti M, Mulligan RS, Villemagne VL, Akatsu H, Yamamoto T, Arai H, Iwata R, Yanai K, Kudo Y. Novel 18F-labeled arylquinoline derivatives for noninvasive imaging of tau pathology in Alzheimer disease. J Nucl Med. 2013 Aug;54(8):1420-7. doi: 10.2967/jnumed.112.117341. View Source
• [3] Villemagne VL, Furumoto S, Fodero-Tavoletti MT, Mulligan RS, Hodges J, Harada R, Yates P, Piguet O, Pejoska S, Doré V, Yanai K, Masters CL, Kudo Y, Rowe CC, Okamura N. In vivo evaluation of a novel tau imaging tracer for Alzheimer's disease. Eur J Nucl Med Mol Imaging. 2014 May;41(5):816-26. doi: 10.1007/s00259-013-2681-7. View Source

Why Generic Tau Tracers Cannot Substitute THK-523

Tau PET tracers cannot be treated as interchangeable commodities because they occupy pharmacologically distinct binding sites on the tau fibril, exhibit widely divergent selectivity profiles against off-target amyloid-β (Aβ) and monoamine oxidase enzymes, and display different pharmacokinetic properties that fundamentally affect image contrast and quantitative accuracy [1]. THK-523 binds to a distinct high-affinity site on neurofibrillary tangles that is separate from the binding site recognized by T807 (AV-1451/flortaucipir), meaning that these two tracers report on different molecular epitopes of tau pathology [2]. Furthermore, THK-523 is one of the few tau tracers that shows preferential binding to tau over Aβ fibrils—a property that contrasts sharply with amyloid tracers such as Pittsburgh Compound B (PiB), BF-227, and FDDNP, which bind Aβ with higher affinity than tau [1]. These molecular-level differences translate directly into divergent autoradiographic staining patterns, PET signal distributions, and clinical interpretation outcomes, making generic substitution scientifically unsound and potentially misleading for research conclusions [3].

Binding site THK-523 occupies a distinct NFT site orthogonal to T807; binding site mismatch may alter competitive assay interpretation.

Selectivity direction Tau-over-Aβ selectivity profile opposite to amyloid tracers; using Aβ-preferring tracers may misattribute tau signal.

Pharmacokinetics Slower brain clearance and higher white matter retention than optimized successors may affect image contrast and research interpretation.

• [1] Harada R, Okamura N, Furumoto S, Tago T, Maruyama M, Higuchi M, Yoshikawa T, Arai H, Iwata R, Kudo Y, Yanai K. Comparison of the binding characteristics of [18F]THK-523 and other amyloid imaging tracers to Alzheimer's disease pathology. Eur J Nucl Med Mol Imaging. 2013 Jan;40(1):125-32. doi: 10.1007/s00259-012-2261-2. View Source
• [2] Cai L, Qu B, Hurtle BT, Dadiboyena S, Diaz-Arrastia R, Pike VW. Candidate PET Radioligand Development for Neurofibrillary Tangles: Two Distinct Radioligand Binding Sites Identified in Postmortem Alzheimer's Disease Brain. ACS Chem Neurosci. 2016 Jul 20;7(7):897-911. doi: 10.1021/acschemneuro.6b00051. View Source
• [3] Villemagne VL, Furumoto S, Fodero-Tavoletti MT, Mulligan RS, Hodges J, Harada R, Yates P, Piguet O, Pejoska S, Doré V, Yanai K, Masters CL, Kudo Y, Rowe CC, Okamura N. In vivo evaluation of a novel tau imaging tracer for Alzheimer's disease. Eur J Nucl Med Mol Imaging. 2014 May;41(5):816-26. doi: 10.1007/s00259-013-2681-7. View Source

THK-523 Quantitative Evidence Guide

Tau Fibril Binding Affinity Comparison

THK-523 binds recombinant K18ΔK280-tau fibrils with dual-site kinetics: a high-affinity site (Kd₁ = 1.67 nM) and a lower-affinity site (Kd₂ = 21.74 nM), while competitive displacement against [¹⁸F]THK-5105 in AD brain homogenates yields a Ki of 59.3 nM [1]. In contrast, the optimized successor THK-5105 binds with Kd₁ = 1.45 nM, Kd₂ = 7.40 nM, and Ki = 7.80 nM, representing an approximately 7.6-fold improvement in inhibitory potency at the same binding site [1]. THK-5117 shows a Ki of 10.5 nM against tau fibrils, approximately 5.6-fold more potent than THK-523 [1]. This quantitative affinity gradient (THK-523 << THK-5117 < THK-5105) directly reflects the compound optimization trajectory and establishes THK-523 as the lower-affinity reference standard for in vitro competitive binding assay calibration [2].

Tau Fibril Binding Affinity

Head-to-head

Kd₁ 1.67 nM, Kd₂ 21.74 nM tau fibrils; Ki 59.3 nM AD homogenates

Supports moderate-affinity tau binding reference; competitive assay context

THK-5105 Ki 7.80 nM; THK-5117 Ki 10.5 nM provide higher affinity

Evidence Dimension	Tau fibril binding affinity (Kd and Ki, nM)
Target Compound Data	Kd₁ = 1.67 nM, Kd₂ = 21.74 nM (K18ΔK280-tau fibrils); Ki = 59.3 nM (AD brain homogenate, competitor: [¹⁸F]THK-5105)
Comparator Or Baseline	THK-5105: Kd₁ = 1.45 nM, Kd₂ = 7.40 nM, Ki = 7.80 nM. THK-5117: Ki = 10.5 nM.
Quantified Difference	THK-5105 Ki is 7.6-fold lower (more potent) than THK-523; THK-5117 Ki is 5.6-fold lower.
Conditions	In vitro saturation binding (Kd) using synthetic K18ΔK280-tau fibrils; competition binding (Ki) using AD brain homogenates with [¹⁸F]THK-5105 as radioligand.

Why This Matters

Users requiring a reference radioligand with a well-characterized, moderate-affinity binding profile for competitive screening of novel tau ligands should select THK-523; researchers pursuing the highest tau affinity for in vivo imaging should select THK-5105 or THK-5117.

• [1] Leuzy A, Chiotis K, Lemoine L, Gillberg PG, Almkvist O, Rodriguez-Vieitez E, Nordberg A. Tau PET imaging in neurodegenerative tauopathies—still a challenge. Mol Psychiatry. 2019 Aug;24(8):1112-1134. Table 2: Tau affinity and selectivity data compiled from original references. Data originally from: Fodero-Tavoletti et al. Brain 2011; Okamura et al. J Nucl Med 2013. View Source
• [2] BindingDB entry BDBM247408. US Patent US9452985. Ki: 59.3 nM for THK-523 binding to K18-ΔK280 tau aggregates at 25°C, competitive binding assay using [¹⁸F]THK-5105. View Source

Tau vs. Amyloid-β Selectivity

In competitive radioligand binding assays using synthetic Aβ₁₋₄₂ and K18ΔK280-tau fibrils, [¹⁸F]THK-523 demonstrated higher affinity for tau fibrils (Kd = 1.67 nM) than for Aβ fibrils (Kd = 86.50 nM), yielding a tau-over-Aβ selectivity ratio of approximately 10-fold (reported as Kd(Aβ)/Kd(tau) = 10a) [1]. In stark contrast, the amyloid imaging tracers PiB, BF-227, and FDDNP all showed higher affinity for Aβ fibrils than for tau fibrils in the identical assay system [2]. Autoradiographic analysis of AD brain sections confirmed that [¹⁸F]THK-523 accumulated in regions with high tau deposit density, while PiB and BF-227 accumulated in Aβ plaque-rich regions [2]. This inverted selectivity profile is the fundamental criterion distinguishing THK-523 from the amyloid imaging agent class and establishes its utility for tau-specific signal isolation in brain regions with co-occurring Aβ pathology [3].

Tau vs. Aβ Selectivity

Head-to-head

Tau Kd 1.67 nM, Aβ Kd 86.50 nM; selectivity ratio ≈10 (tau-preferring)

Supports tau-specific signal isolation; inverted vs. amyloid tracers

PiB, BF-227, FDDNP show Aβ-preferring direction

Evidence Dimension	Tau versus Aβ fibril binding selectivity ratio
Target Compound Data	Kd(tau) = 1.67 nM; Kd(Aβ) = 86.50 nM; selectivity ratio (Kd_Aβ/Kd_tau) ≈ 10; tau-preferring.
Comparator Or Baseline	PiB: Aβ-preferring; BF-227: Aβ-preferring; FDDNP: Aβ-preferring (direction inverted relative to THK-523).
Quantified Difference	THK-523 shows ~52-fold higher affinity for tau than for Aβ fibrils (86.50/1.67); comparator tracers show higher affinity for Aβ than tau.
Conditions	In vitro radioligand binding assays with synthetic Aβ₁₋₄₂ and K18ΔK280-tau fibrils; in vitro autoradiography on AD brain sections.

Why This Matters

For studies requiring tau signal isolation without Aβ cross-contamination—particularly in brain regions with mixed pathology—THK-523 offers a fundamentally different selectivity direction from the widely used amyloid tracers, making it scientifically critical for differential PET signal attribution.

• [1] Leuzy A, Chiotis K, Lemoine L, Gillberg PG, Almkvist O, Rodriguez-Vieitez E, Nordberg A. Tau PET imaging in neurodegenerative tauopathies—still a challenge. Mol Psychiatry. 2019 Aug;24(8):1112-1134. Table 2: Selectivity tau/Aβ column, compound 16 (THK-523). View Source
• [2] Harada R, Okamura N, Furumoto S, Tago T, Maruyama M, Higuchi M, Yoshikawa T, Arai H, Iwata R, Kudo Y, Yanai K. Comparison of the binding characteristics of [18F]THK-523 and other amyloid imaging tracers to Alzheimer's disease pathology. Eur J Nucl Med Mol Imaging. 2013 Jan;40(1):125-32. doi: 10.1007/s00259-012-2261-2. View Source
• [3] Fodero-Tavoletti MT, Okamura N, Furumoto S, Mulligan RS, Connor AR, McLean CA, Cao D, Rigopoulos A, Cartwright GA, O'Keefe G, Gong S, Adlard PA, Barnham KJ, Rowe CC, Masters CL, Kudo Y, Cappai R, Yanai K, Villemagne VL. 18F-THK523: a novel in vivo tau imaging ligand for Alzheimer's disease. Brain. 2011 Apr;134(Pt 4):1089-100. View Source

Distinct NFT Binding Site: Orthogonal to T807

Using ³H-labeled THK-523 and ³H-labeled T807 as reference radioligands in competitive binding assays with AD brain homogenates, Cai et al. (2016) demonstrated that THK-523 and T807 (flortaucipir/AV-1451) bind to two pharmacologically distinct, non-competing sites on neurofibrillary tangles [1]. Neither radioligand was displaced by the other cold compound at relevant concentrations, confirming site orthogonality [1]. Furthermore, both sites were shown to be distinct from the thioflavin-T binding site occupied by amyloid PET tracers such as PiB [1]. This finding has direct practical implications: THK-523 and T807 report on different structural or conformational epitopes of tau pathology, meaning that binding site occupancy measured by one tracer cannot be extrapolated to the other [2]. THK-523 thus serves as a site-specific probe for one of the at least two distinct high-affinity small-molecule binding pockets on AD tau aggregates [3].

Binding Site Orthogonality

Head-to-head

³H-THK-523 not displaced by cold T807; distinct site confirmed

Supports orthogonal binding site probe for tau epitope mapping

Two distinct NFT sites; T807 site separate

Evidence Dimension	Tau aggregate binding site identity and orthogonality
Target Compound Data	³H-THK-523 binds to Site 1 (THK-523-preferring site) on NFTs in AD brain homogenates.
Comparator Or Baseline	³H-T807 (AV-1451) binds to Site 2 (T807-preferring site); ³H-PiB binds to thioflavin-T site (Site 3, distinct from both).
Quantified Difference	Two distinct, non-competing binding sites confirmed by cross-displacement assays: ³H-THK-523 not displaced by cold T807; ³H-T807 not displaced by cold THK-523.
Conditions	In vitro radioligand binding with AD brain homogenates (postmortem frontal cortex); ³H-labeled THK-523 and ³H-T807 as reference radioligands; confirmed via autoradiography and immunohistochemistry co-localization.

Why This Matters

For laboratories developing or screening novel tau ligands, THK-523 provides access to a distinct binding site that is orthogonal to the T807/flortaucipir site, making it an essential reference for competitive displacement assays aimed at identifying site-specific binders and for orthogonal validation of tau pathology in tissues.

• [1] Cai L, Qu B, Hurtle BT, Dadiboyena S, Diaz-Arrastia R, Pike VW. Candidate PET Radioligand Development for Neurofibrillary Tangles: Two Distinct Radioligand Binding Sites Identified in Postmortem Alzheimer's Disease Brain. ACS Chem Neurosci. 2016 Jul 20;7(7):897-911. doi: 10.1021/acschemneuro.6b00051. View Source
• [2] Murugan NA, Chiotis K, Rodriguez-Vieitez E, Lemoine L, Ågren H, Nordberg A. Cross-interaction of tau PET tracers with monoamine oxidase B: evidence from in silico modelling and in vivo imaging. Eur J Nucl Med Mol Imaging. 2019 Jun;46(6):1365-1377. (Cites Cai et al. finding of distinct THK-523 and T807 binding sites.) View Source
• [3] Molaid compound page for T807 (AV-1451/Flortaucipir). Describes competitive binding assay with ³H-THK523 and ³H-T807 identifying two distinct small-molecule binding sites on NFTs. View Source

Brain Uptake and Clearance Kinetics Comparison

In biodistribution studies in normal mice, [¹⁸F]THK-523 demonstrated an initial brain uptake of 2.72% ID/g at 2 min post-injection, declining to 1.47% ID/g at 30 min [1]. In contrast, the optimized derivatives [¹⁸F]THK-5105 and [¹⁸F]THK-5117 achieved substantially higher initial brain uptake (9.20% and 6.06% ID/g at 2 min, respectively) and showed significantly faster washout kinetics from normal mouse brain [1]. The brain₂ min/brain₃₀ min clearance ratio for THK-523 is approximately 1.85, compared to 2.55 for THK-5105 and 10.27 for THK-5117, indicating that THK-523 exhibits markedly slower clearance of non-specifically bound tracer from brain tissue [1]. An independent study confirmed that initial brain uptake of THK-523 (2.75% ID/g) is at the lower margin of PET tracer acceptability, where values above 4.0% ID/g are considered standard for adequate signal-to-noise ratios in mice [2]. A follow-up analog study reported THK-523 brain uptake as 2.62 ± 0.39% ID/g at 2 min, significantly lower than its ¹¹C-labeled analog TKF (3.23 ± 1.25% ID/g) [3].

Brain Uptake & Clearance

Cross-study

2 min: 2.72% ID/g; 30 min: 1.47% ID/g; clearance ratio ~1.85

Reported brain uptake and slower clearance; may support in vitro assay use

THK-5105 2 min 9.20% ID/g; faster washout

Evidence Dimension	Initial brain uptake (2 min) and 30-min retention (% ID/g) in normal mice
Target Compound Data	2 min: 2.72% ID/g; 30 min: 1.47% ID/g; clearance ratio (2min/30min): ~1.85.
Comparator Or Baseline	THK-5105: 2 min 9.20, 30 min 3.61, ratio ~2.55. THK-5117: 2 min 6.06, 30 min 0.59, ratio 10.27. PiB: initial brain uptake >8% ID/g. TKF analog: 2 min 3.23 ± 1.25% ID/g.
Quantified Difference	THK-523 initial brain uptake is 3.4-fold lower than THK-5105 and 2.2-fold lower than THK-5117; 30-min clearance is 5.8-fold slower than THK-5117.
Conditions	Biodistribution studies in normal ddY or C57BL/6 mice; % injected dose per gram tissue; data compiled from multiple published studies.

Why This Matters

THK-523's lower brain uptake and slower clearance make it less suitable for high-contrast in vivo PET imaging compared to THK-5105/THK-5117, but its slower washout kinetics may be advantageous for in vitro autoradiography and homogenate binding assays where prolonged target engagement is desirable.

• [1] Leuzy A, Chiotis K, Lemoine L, Gillberg PG, Almkvist O, Rodriguez-Vieitez E, Nordberg A. Tau PET imaging in neurodegenerative tauopathies—still a challenge. Mol Psychiatry. 2019 Aug;24(8):1112-1134. Table 2: Brain uptake column, compounds 16 (THK-523), 17 (THK-5105), 18 (THK-5117). View Source
• [2] Alzforum commentary on Fodero-Tavoletti et al. (2011). 18F-THK523: a novel in vivo tau imaging ligand for Alzheimer's disease. Notes brain uptake of 2.75% ID/g is on the edge of acceptability; PiB reference >8% ID/g. View Source
• [3] Kong Y, Guan Y, Hua F, Zhang Z, Lu X, Zhu T, Zhao B, Zhu J, Li C, Chen J. Optimization and Biodistribution of [11C]-TKF, An Analog of Tau Protein Imaging Agent [18F]-THK523. Molecules. 2016 Aug 5;21(8):1019. doi: 10.3390/molecules21081019. Reports THK-523 brain uptake 2.62 ± 0.39% ID/g at 2 min vs. TKF 3.23 ± 1.25% ID/g. View Source

White Matter Retention Benchmark

In the first-in-human PET study of [¹⁸F]THK-523 in AD patients (n=10) and healthy controls (n=10), white matter retention was significantly higher than grey matter retention across all participants, a pattern that precluded simple visual inspection of PET images and was cited as a critical barrier to clinical or research adoption [1]. This elevated white matter signal was subsequently shown to correspond to specific THK-523 binding to PHF-tau aggregates present in white matter tracts of AD patients, as confirmed by Western blot and in vitro binding studies [2]. Importantly, later-generation THK-family tracers, notably THK-5351, were specifically optimized to reduce white matter retention and achieve faster kinetics with higher grey-to-white matter contrast [3]. The white matter retention problem thus represents a quantifiable differentiation point: THK-523 serves as a 'benchmark' compound representing the pre-optimization state, against which improved white matter specificity of newer tracers is measured [3].

White Matter Retention

Class-level

Higher white matter than grey matter retention in human PET; precludes simple visual interpretation

Benchmark for white matter binding; not recommended for in vivo imaging

THK-5351 optimized for lower white matter signal

Evidence Dimension	White matter retention relative to grey matter in human PET imaging
Target Compound Data	White matter retention significantly higher than grey matter retention in all participants (AD and controls); precludes simple visual inspection.
Comparator Or Baseline	THK-5351 (next-generation THK tracer): reduced white matter retention, faster kinetics, higher signal-to-noise ratio compared with THK-523 (qualitative improvement noted in reviews).
Quantified Difference	Qualitative contrast limitation documented; quantitative white-to-grey matter ratios not numerically reported in primary studies but described as 'very high' and 'preventing clinical use.'
Conditions	Human PET imaging: 10 AD patients, 10 healthy controls, 3 semantic dementia patients. [¹⁸F]THK-523 administered intravenously; partial volume-corrected tissue ratios calculated.

Why This Matters

THK-523 should not be selected for human in vivo PET imaging studies where high grey-to-white matter contrast is required; it is instead optimally positioned as an in vitro reference standard and a historical benchmark compound for evaluating white matter binding improvements in novel tau tracer development programs.

• [1] Villemagne VL, Furumoto S, Fodero-Tavoletti MT, Mulligan RS, Hodges J, Harada R, Yates P, Piguet O, Pejoska S, Doré V, Yanai K, Masters CL, Kudo Y, Rowe CC, Okamura N. In vivo evaluation of a novel tau imaging tracer for Alzheimer's disease. Eur J Nucl Med Mol Imaging. 2014 May;41(5):816-26. doi: 10.1007/s00259-013-2681-7. View Source
• [2] Fodero-Tavoletti MT, et al. Detectable levels of white matter PHF-tau in Alzheimer's disease. Alzheimer's & Dementia: The Journal of the Alzheimer's Association. 2013 Jul;9(4):P872-P873. In vitro binding study confirming THK-523 binding to white matter PHF-tau. View Source
• [3] Harada R, Okamura N, Furumoto S, Furukawa K, Ishiki A, Tomita N, Tago T, Hiraoka K, Watanuki S, Shidahara M, Miyake M, Ishikawa Y, Matsuda R, Inami A, Yoshikawa T, Funaki Y, Iwata R, Tashiro M, Yanai K, Arai H, Kudo Y. 18F-THK5351: A Novel PET Radiotracer for Imaging Neurofibrillary Pathology in Alzheimer Disease. J Nucl Med. 2016 Feb;57(2):208-14. Reports THK-5351 with reduced white matter retention relative to predecessor THK-523. View Source

In Vivo Tau Selectivity vs. Amyloid PET

In the dual-tracer clinical PET study of Villemagne et al. (2014), the same 10 AD patients and 10 healthy controls underwent both [¹⁸F]THK-523 and [¹¹C]PiB PET scans [1]. The cortical distribution of [¹⁸F]THK-523 retention did not correlate with the Aβ distribution pattern assessed by [¹¹C]PiB; instead, it followed the known neuropathological distribution of PHF-tau in AD brain, with higher signal in temporal and parietal cortices than in frontal regions [1]. Furthermore, unlike [¹¹C]PiB, hippocampal [¹⁸F]THK-523 retention was significantly correlated with cognitive performance parameters and with hippocampal atrophy measured by MRI [1]. This in vivo double-dissociation—THK-523 signal correlates with tau-related cognitive decline and atrophy but not with PiB-measured Aβ burden—provides direct clinical evidence that [¹⁸F]THK-523 selectively reports on tau pathology in the living human brain, independently of co-occurring amyloid pathology [2].

In Vivo Tau vs. Amyloid

Head-to-head

THK-523 retention follows tau pathology, not PiB Aβ; hippocampal signal correlates with cognitive decline (p

Supports in vivo tau signal isolation independently of amyloid burden

Dual-tracer dissociation in AD patients

Evidence Dimension	Correlation of PET signal distribution with Aβ (PiB) versus tau pathology pattern; correlation with cognitive scores and hippocampal atrophy
Target Compound Data	[¹⁸F]THK-523 retention pattern: temporal > parietal > frontal; signal does not correlate with [¹¹C]PiB distribution; hippocampal THK-523 retention correlates with cognitive deficits and hippocampal atrophy.
Comparator Or Baseline	[¹¹C]PiB retention pattern follows Aβ plaque distribution; hippocampal PiB retention does not correlate with cognitive parameters in the same cohort.
Quantified Difference	Significantly higher [¹⁸F]THK-523 retention in temporal, parietal, orbitofrontal cortices and hippocampi of AD vs. controls (p<0.05); correlation between hippocampal THK-523 and cognitive scores reported as significant (specific r and p values behind paywall).
Conditions	Dual-tracer PET imaging in 10 AD patients, 10 healthy elderly controls, 3 SD patients. Partial volume-corrected tissue ratios; correlational analysis with composite memory/non-memory scores and MRI volumetry.

Why This Matters

This in vivo evidence confirms that THK-523 can isolate tau-specific signal in the human brain even in the presence of abundant fibrillar Aβ, a critical capability for studies aiming to dissociate tau-mediated neurodegeneration from amyloid-driven pathology in AD.

• [1] Villemagne VL, Furumoto S, Fodero-Tavoletti MT, Mulligan RS, Hodges J, Harada R, Yates P, Piguet O, Pejoska S, Doré V, Yanai K, Masters CL, Kudo Y, Rowe CC, Okamura N. In vivo evaluation of a novel tau imaging tracer for Alzheimer's disease. Eur J Nucl Med Mol Imaging. 2014 May;41(5):816-26. doi: 10.1007/s00259-013-2681-7. View Source
• [2] Abstract of Villemagne et al. 2014 on PubMed. States: '18F-THK523 does not bind to Aβ in vivo, while following the known distribution of paired helical filaments (PHF)-tau in the brain.' Confirms dual-tracer dissociation. View Source

THK-523 Application Scenarios

Reference Radioligand for Competitive Binding Assays

THK-523, as ³H-labeled or ¹⁸F-labeled radioligand, serves as a well-characterized reference standard occupying a distinct binding site (Site 1) on neurofibrillary tangles that is orthogonal to the T807/flortaucipir binding site [1]. Its moderate affinity (Kd₁ = 1.67 nM; Ki = 59.3 nM in AD brain homogenates) enables sensitive detection of competitive displacement by novel ligands without the ultra-high-affinity masking effects that can occur with tighter-binding reference tracers [2]. Procurement of THK-523 for this purpose is scientifically indicated when the screening objective is to identify compounds targeting the THK-523-preferring tau binding site specifically [1].

Tau/Amyloid Differential Binding in AD Brain Tissue

THK-523 is one of the few tau PET tracer chemotypes that preferentially binds tau fibrils over Aβ fibrils by approximately 10-fold [3]. In autoradiography experiments using AD brain sections, [¹⁸F]THK-523 accumulates selectively in tau-rich regions, while PiB and BF-227 highlight Aβ plaques in the same tissue [3]. This inverted selectivity makes THK-523 a critical tool for laboratories performing dual-label or sequential autoradiography studies that require unambiguous biochemical differentiation of tau and amyloid pathology distributions within the same tissue samples [3].

White Matter Binding Benchmark for Tracer Development

THK-523's well-documented high white matter retention—identified as the primary obstacle to its clinical translation in the first-in-human study [4]—establishes it as the appropriate negative benchmark for preclinical white matter specificity screening. Novel arylquinoline or non-arylquinoline tau tracer candidates can be compared head-to-head against THK-523 in biodistribution, autoradiography, and in silico docking studies to quantify improvements in grey-to-white matter binding ratios [5]. Procurement of THK-523 as a reference standard is therefore recommended for any medicinal chemistry program targeting reduced off-target white matter signal.

In Vivo PET Imaging in Tau Transgenic Mouse Models

In tau transgenic mice (rTg4510 model), microPET imaging with [¹⁸F]THK-523 demonstrated approximately 48% higher tracer retention compared with wild-type littermates (P < 0.007), with no increased retention in APP/PS1 amyloid-depositing mice [6]. While THK-523's brain uptake (2.72% ID/g at 2 min) is lower than that of optimized successors, this level is sufficient for small-animal PET studies where longitudinal within-subject comparisons are performed and absolute quantification is achievable through arterial input function measurement [6]. Researchers using well-established tau transgenic models may procure THK-523 for protocols where access to the historical validation data in these specific model systems provides continuity with published preclinical findings.

Application

Selection Property

Validation Focus

Competitive binding assay reference

Distinct NFT binding site (orthogonal to T807)

Site-specific binder identification

Differential tau/amyloid autoradiography

Tau-over-Aβ selectivity direction

Tau-specific signal isolation in mixed pathology

White matter binding benchmark

Documented high white matter retention profile

Grey-to-white matter contrast improvement screening

Small-animal tau PET imaging

Preclinical brain uptake and retention

Longitudinal tau pathology monitoring

• [1] Cai L, Qu B, Hurtle BT, Dadiboyena S, Diaz-Arrastia R, Pike VW. Candidate PET Radioligand Development for Neurofibrillary Tangles: Two Distinct Radioligand Binding Sites Identified in Postmortem Alzheimer's Disease Brain. ACS Chem Neurosci. 2016 Jul 20;7(7):897-911. doi: 10.1021/acschemneuro.6b00051. View Source
• [2] Leuzy A, Chiotis K, Lemoine L, Gillberg PG, Almkvist O, Rodriguez-Vieitez E, Nordberg A. Tau PET imaging in neurodegenerative tauopathies—still a challenge. Mol Psychiatry. 2019 Aug;24(8):1112-1134. Table 2: Tau affinity data for THK-523 and comparators. View Source
• [3] Harada R, Okamura N, Furumoto S, Tago T, Maruyama M, Higuchi M, Yoshikawa T, Arai H, Iwata R, Kudo Y, Yanai K. Comparison of the binding characteristics of [18F]THK-523 and other amyloid imaging tracers to Alzheimer's disease pathology. Eur J Nucl Med Mol Imaging. 2013 Jan;40(1):125-32. doi: 10.1007/s00259-012-2261-2. View Source
• [4] Villemagne VL, Furumoto S, Fodero-Tavoletti MT, Mulligan RS, Hodges J, Harada R, Yates P, Piguet O, Pejoska S, Doré V, Yanai K, Masters CL, Kudo Y, Rowe CC, Okamura N. In vivo evaluation of a novel tau imaging tracer for Alzheimer's disease. Eur J Nucl Med Mol Imaging. 2014 May;41(5):816-26. doi: 10.1007/s00259-013-2681-7. View Source
• [5] Okamura N, Furumoto S, Harada R, Tago T, Yoshikawa T, Fodero-Tavoletti M, Mulligan RS, Villemagne VL, Akatsu H, Yamamoto T, Arai H, Iwata R, Yanai K, Kudo Y. Novel 18F-labeled arylquinoline derivatives for noninvasive imaging of tau pathology in Alzheimer disease. J Nucl Med. 2013 Aug;54(8):1420-7. doi: 10.2967/jnumed.112.117341. View Source
• [6] Fodero-Tavoletti MT, Okamura N, Furumoto S, Mulligan RS, Connor AR, McLean CA, Cao D, Rigopoulos A, Cartwright GA, O'Keefe G, Gong S, Adlard PA, Barnham KJ, Rowe CC, Masters CL, Kudo Y, Cappai R, Yanai K, Villemagne VL. 18F-THK523: a novel in vivo tau imaging ligand for Alzheimer's disease. Brain. 2011 Apr;134(Pt 4):1089-100. doi: 10.1093/brain/awr038. View Source