18F-FIBT may expand PET for ß-amyloid imaging in neurodegenerative diseases.

18F-FIBT, 2-(p-Methylaminophenyl)-7-(2-[18F]fluoroethoxy)imidazo-[2,1-b]benzothiazole, is a new selective PET tracer under clinical investigation to specifically image ß-amyloid depositions (Aß) in humans in-vivo that binds to Aß with excellent affinity (Kd 0.7 ± 0.2) and high selectivity over tau and α-synuclein aggregates (Ki > 1000 nM). We aimed to characterize 18F-FIBT in a series of patients with different clinical-pathophysiological phenotypes and to compare its binding characteristics to the reference compound PiB. Six patients (mild late-onset and moderate early-onset AD dementia, mild cognitive impairment due to AD, intermediate likelihood, mild behavioral variant of frontotemporal dementia, subjective memory impairment without evidence of neurodegeneration, and mild dementia due to Posterior Cortical Atrophy) underwent PET imaging with 18F-FIBT on PET/MR. With the guidance of MRI, PET images were corrected for partial volume effect, time-activity curves (TACs) of regions of interest (ROIs) were extracted, and non-displaceable binding potentials (BPnd), standardized uptake value ratios (SUVR), and distribution volume ratio (DVR) were compared. Specific binding was detected in the cases with evidence of the AD pathophysiological process visualized in images of BPnd, DVR and SUVR, consistently with patterns of different tracers in previous studies. SUVR showed the highest correlation with clinical severity. The previous preclinical characterization and the results of this case series suggest the clinical usefulness of FIBT as a selective and highly affine next-generation 18F-labeled tracer for amyloid-imaging with excellent pharmacokinetics in the diagnosis of neurodegenerative diseases. The results compare well to the gold standard PiB and hence support further investigation in larger human samples.

Correction: 18F-FIBT may expand PET for ß-amyloid imaging in neurodegenerative diseases.

The author listing has been updated to indicate that Timo Grimmer and Kuangyu Shi are equally contributing authors.

Mapping the Binding Interface of PET Tracer Molecules and Alzheimer Disease Aß Fibrils by Using MAS Solid-State NMR Spectroscopy.

Positron emission tomography (PET) tracer molecules like thioflavin T specifically recognize amyloid deposition in brain tissue by selective binding to hydrophobic or aromatic surface grooves on the ß-sheet surface along the fibril axis. The molecular basis of this interaction is, however, not well understood. We have employed magic angle spinning (MAS) solid-state NMR spectroscopy to characterize Aß-PET tracer complexes at atomic resolution. We established a titration protocol by using bovine serum albumin as a carrier to transfer hydrophobic small molecules to Aß(1-40) fibrillar aggregates. The same Aß(1-40) amyloid fibril sample was employed in subsequent titrations to minimize systematic errors that potentially arise from sample preparation. In the experiments, the small molecules 13 C-methylated Pittsburgh compound B (PiB) as well as a novel Aß tracer based on a diarylbithiazole (DABTA) scaffold were employed. Classical 13 C-detected as well as proton-detected spectra of protonated and perdeuterated samples with back-substituted protons, respectively, were acquired and analyzed. After titration of the tracers, chemical-shift perturbations were observed in the loop region involving residues Gly25-Lys28 and Ile32-Gly33, thus suggesting that the PET tracer molecules interact with the loop region connecting ß-sheets ß1 and ß2 in Aß fibrils. We found that titration of the PiB derivatives suppressed fibril polymorphism and stabilized the amyloid fibril structure.

Toward Novel [18F]Fluorine-Labeled Radiotracers for the Imaging of α-Synuclein Fibrils.

The accumulation of α-synuclein aggregates (α-syn) in the human brain is an occurrence common to all α-synucleinopathies. Non-invasive detection of these aggregates in a living brain with a target-specific radiotracer is not yet possible. We have recently discovered that the inclusion of a methylenedioxy group in the structure of diarylbisthiazole (DABTA)-based tracers improves binding affinity and selectivity to α-syn. Subsequently, complementary in silico modeling and machine learning (ML) of tracer-protein interactions were employed to predict surface sites and structure-property relations for the binding of the ligands. Based on this observation, we developed a small focused library of DABTAs from which 4-(benzo[d][1,3]dioxol-5-yl)-4'-(3-[18F]fluoro-4-methoxyphenyl)-2,2'-bithiazole [ 18 F]d 2, 6-(4'-(3-[18F]fluoro-4-methoxyphenyl)-[2,2'-bithiazol]-4-yl)-[1,3]dioxolo[4,5-b]pyridine [ 18 F]d 4, 4-(benzo [d][1,3]dioxol-5-yl)-4'-(6-[18F]fluoropyridin-3-yl)-2,2'-bithiazole [ 18 F]d 6, and 6-(4'-(6-[18F]fluoropyridin-3-yl)-[2,2'-bithiazol]-4-yl)-[1,3]dioxolo[4,5-b]pyridine [ 18 F]d 8 were selected based on their high binding affinity to α-syn and were further evaluated. Binding assay experiments carried out with the non-radioactive versions of the above tracers d 2, d 4, d 6, and d 8 showed high binding affinity of the ligands to α-syn: 1.22, 0.66, 1.21, and 0.10 nM, respectively, as well as excellent selectivity over ß-amyloid plaques (Aß) and microtubular tau aggregates (>200-fold selectivity). To obtain the tracers, their precursors were radiolabeled either via an innovative ruthenium-mediated (SNAr) reaction ([ 18 F]d 2 and [ 18 F]d 4) or typical SNAr reaction ([ 18 F]d 6 and [ 18 F]d 8) with moderate-to-high radiochemical yields (13% - 40%), and high molar activity > 60 GBq/µmol. Biodistribution experiments carried out with the tracers in healthy mice revealed that [ 18 F]d 2 and [ 18 F]d 4 showed suboptimal brain pharmacokinetics: 1.58 and 4.63 %ID/g at 5 min post-injection (p.i.), and 1.93 and 3.86 %ID/g at 60 min p.i., respectively. However, [ 18 F]d 6 and [ 18 F]d 8 showed improved brain pharmacokinetics: 5.79 and 5.13 %ID/g at 5 min p.i.; 1.75 and 1.07 %ID/g at 60 min p.i.; and 1.04 and 0.58 %ID/g at 120 min p.i., respectively. The brain uptake kinetics of [ 18 F]d 6 and [ 18 F]d 8 were confirmed in a dynamic PET study. Both tracers also showed no brain radiometabolites at 20 min p.i. in initial in vivo stability experiments carried out in healthy mice. [ 18 F]d 8 seems very promising based on its binding properties and in vivo stability, thus encouraging further validation of its usefulness as a radiotracer for the in vivo visualization of α-syn in preclinical and clinical settings. Additionally, in silico and ML-predicted values correlated with the experimental binding affinity of the ligands.

Characterization and first human investigation of FIBT, a novel fluorinated Aß plaque neuroimaging PET radioligand.

Imidazo[2,1-b]benzothiazoles (IBTs) are a promising novel class of amyloid positron emission tomography (PET) radiopharmaceuticals for diagnosis of neurodegenerative disorders like Alzheimer's disease (AD). Their good in vivo imaging properties have previously been shown in preclinical studies. Among IBTs, fluorinated [(18)F]FIBT was selected for further characterization and advancement toward use in humans. [(18)F]FIBT characteristics were analyzed in relation to Pittsburgh compound B (PiB) as reference ligand. [(18)F]FIBT and [(3)H]PiB were coinjected to an APP/PS1 mouse for ex vivo dual-label autoradiographic correlation. Acute dose toxicity of FIBT was examined in two groups of healthy mice. Preexisting in vivo stability and biodistribution studies in mice were complemented with analogous studies in rats. [(18)F]FIBT was titrated against postmortem human AD brain homogenate in a saturation binding assay previously performed with [(3)H]PiB. Binding of [(18)F]FIBT to human AD brain was further analyzed by in vitro incubation of human AD brain sections in comparison to [(11)C]PiB in relation to standard immunohistochemistry. Finally, [(18)F]FIBT was administered to two human subjects for a dynamic 90 min PET/MR brain investigation. Ex vivo autoradiography confirmed good uptake of [(18)F]FIBT to mouse brain and its excellent correlation to [(3)H]PiB binding. No toxicity of FIBT could be found in mice at a concentration of 33.3 nmol/kg. As in mice, [(18)F]FIBT was showing high in vivo stability in rats and comparable regional brain biodistribution dynamics to [(3)H]PiB. Radioligand saturation binding confirmed at least one high-affinity binding component of [(18)F]FIBT around 1 nM. Good binding of FIBT relative to PiB was further confirmed in binding assays and autoradiographies using post-mortem AD brain. First use of [(18)F]FIBT in humans successfully yielded clinical [(18)F]FIBT PET/MR images with very good contrast. In summary, [(18)F]FIBT has been characterized to be a new lead compound with improved binding characteristics and pharmacokinetics on its own as well as in comparison to PiB. A pilot human PET investigation provided high-quality images with a plausible tracer distribution pattern. Detailed clinical investigations are needed to confirm these first results and to explore the specific qualities of [(18)F]FIBT PET for dementia imaging in relation to established ligands.

Regional expansion of hypometabolism in Alzheimer's disease follows amyloid deposition with temporal delay.

BACKGROUND: Cross-sectional imaging studies suggest that patterns of hypometabolism (measured by [(18)F] fluorodeoxyglucose positron emission tomography [FDG-PET]) and amyloid deposition (measured by [(11)C] Pittsburgh Compound B [PiB]- PET) in Alzheimer's disease (AD) show some overlap with each other. This indicates that neuronal dysfunction might spread within the anatomical pattern of amyloid deposition. The aim of this study was to examine longitudinal regional patterns of amyloid deposition and hypometabolism in the same population of mild AD subjects and to establish their regional relationship to each other. METHODS: Twenty patients with mild AD underwent baseline (BL) and follow-up (FU) examination with [(18)F] FDG-PET and [(11)C] PiB-PET. Voxel-by-voxel statistical group comparison (SPM5) was performed between patient BL- and FU-PET data as well as between patients and 15 PiB-negative elderly control subjects, who had undergone identical imaging procedures. To obtain objective measures of regional overlap, Dice similarity coefficients (DSC) between the imaging findings were calculated. RESULTS: Compared with elderly control subjects, AD patients showed typical patterns of BL hypometabolism and BL amyloid deposition, with a similarity of 40% (DSC). Amyloid deposition was more extended than hypometabolism at BL and showed only minor changes over time, whereas significant expansion of hypometabolism was observed, almost exclusively within areas already affected by BL amyloid deposition. Thus, increased similarity of FU hypometabolism with BL amyloid deposition was found (DSC: 47%). CONCLUSIONS: Longitudinal regional expansion of cerebral hypometabolism, as a measure of neuronal dysfunction in AD, seems to follow the anatomical pattern of amyloid deposition with temporal delay. This indicates that amyloid-based disruption of neuronal integrity might contribute to the regional expansion of neuronal dysfunction.