PET Imaging of brain muscarinic receptors with 18F-Fluorobenzyl-Dexetimide: A first in human study

https://doi.org/10.1016/j.pscychresns.2021.111354Get rights and content

Highlights

  • This is the first-in-human use of FDEX PET.

  • FDEX is the first fluorine-18 (F-18) labelled PET tracer for muscarinic receptors. This allows it to be produced more efficiently and distributed more widely than existing carbon-11 labelled tracers that require an on-site cyclotron.

  • FDEX provides high quality images of brain mACh receptors.

  • FDEX has low variance in the level of binding to mACh receptors across normal young adults suggesting it may be sensitive for detection of changes in receptor density or binding in disease states.

  • Imaging time and measurement of tissue ratios using the cerebellum as the reference region are suitable for application to patients with schizophrenia or dementia.

Abstract

M1 and M4 muscarinic receptor (mAChR) agonists are under development for the treatment of schizophrenia, Alzheimer's and Parkinson's disease. We performed first-in-human PET imaging of mAChR with 18F-Fluorobenzyl-Dexetimide (FDEX) in 10 healthy participants (29.4±4.3yrs). Four underwent dynamic brain scanning for 240 min, and then six underwent static brain scans at 120 and 160-min post injection of 250 MBq of FDEX. Gjedde-Patlak graphical analysis was applied to determine the influx constant (Ki). Regional tissue ratios (SUVR) were calculated using the cerebellar cortex as the reference region. No adverse events were observed. The tracer showed good brain entry (∼4.2% ID at 5 min) but irreversible distribution kinetics over four hours in regions of high mAChR. Binding was consistent with the distribution of mAChR receptors with striatum > cortex > hippocampus >> thalamus >>> cerebellum with low variance in regional binding between subjects. Ki was 0.42±0.04 in the putamen, 0.27±0.01 in frontal cortex, 0.25±0.02 in the hippocampus and 0.10±0.01 in the thalamus. SUVR at 120 and 240 min. were highly correlated with these Ki values with R2 of 0.91 and 0.99 respectively. FDEX yields high quality brain images with uptake in the known distribution of mAChR with remarkably little variance between normal subjects.

Introduction

Cholinergic circuits, particularly within the basal forebrain complex, are vital to cognitive processes including attention and memory (Mufson et al. 2003, Ballinger et al. 2016) and have been implicated in psychiatric disorders particularly schizophrenia (Scarr et al., 2013). Notably, the synthesis of drugs that can target individual muscarinic receptors with increasing specificity at their orthosteric and allosteric binding sites have made the mAChR family of 5 receptor subtypes of increasing interest with respect to treating disorders of the human CNS.

Muscarinic AChR are most abundant in the striatum, cortex and hippocampus, with lower abundance in the thalamus and very low abundance in the cerebellum with respective concentrations relative to cerebellum of approximately 13, 10, 8 and 4 reported by Toyohara et al. (2013) and 8, 7, 6 and 3 reported by Ehlert et al. (1995). Of the five mAChR subtypes (M1-5) in the human CNS, levels of the M1 receptor are the most abundant in the frontal, temporal and parietal cortex and the hippocampus where they constitute 45-55% of total mAChR. The M1 has similar abundance to M2 in the occipital cortex (35% each) and lower abundance than M4 in the striatum (35% vs 45%) (Levey et al. 1991, Flynn et al.1995, Volpicelli and Levey, 2004, Toyohara et al. 2013). M2 predominates in the thalamus, brainstem and cerebellum (Levey et al. 1991). M3 and M5 are expressed in low abundance in the brain (Levey et al. 1991, Flynn et al.1995, Volpicelli and Levey, 2004, Toyohara et al. 2013). The M1 and M4 receptors are of increasing interest to psychiatry because postmortem studies suggest that a sub-group of patients with schizophrenia (∼25%) has lost a large proportion (∼ 75%) of these cortical M1 and M4 receptors (Dean et al. 2002, Dean et al. 2008, Scarr et al. 2009). Moreover, both animal and human studies suggest that drugs that target M1 and M4 receptors have cognitive enhancing effects in Alzheimer's disease (Mufson et al. 2008) and provide symptom improvement in schizophrenia (Shekhar, et al. 2008, Miller et al. 2016). In-vivo detection of muscarinic receptors in human CNS would be useful in establishing target engagement and provide information on clinical responsiveness vs. receptor levels and their occupancy in drug trials.

Since the 1990’s, the SPECT tracer I-123 iododexetimide has been used for in-vivo imaging of mAChRs (Boundy et al. 1995, Boundy et al. 2005) and a recent study has reported a degree of M1 subtype selectivity at the nanomolar radioligand concentrations achieved in SPECT and PET imaging (Bakker et al. 2015). Due to its higher spatial resolution and quantitative accuracy, PET is the preferred modality to investigate alterations in receptor density. To date, other than F-18-FP-TZTP for imaging M2 subtype mAChR, PET tracers used for imaging mAChR in humans have been labeled with carbon-11 (C-11), restricting access to sites with an on-site cyclotron (Toyohara et al. 2013). Currently, there is no available F-18 labeled PET tracer to image mAChRs.

Discovered in the 1960’s, Dexetimide is a high potency, non-subtype selective muscarinic antagonist at pharmacologically active doses. A derivative, F-18 labeled 4-fluorodexetimide (FDEX), was produced and tested in rodents in the early 1990’s with good results (Hwang et al. 1991, Wilson et al.1991), however it was never trialed in humans.

We implemented FDEX production (Fig. 1), confirmed mAChR specificity on a broad receptor screening panel, demonstrated marked reduction in FDEX binding using autoradiography in post-mortem brain tissue from persons with schizophrenia where low M1 receptor density had been determined by [3H] pirenzepine binding, and found at 5nM concentration, as typically achieved during PET studies, marked reduction in FDEX binding in M1 knock-out mice membrane preparations. We then demonstrated no toxicity up to 1000 times the expected dose from a FDEX PET study (RDDT final report, 2020). We now report the first-in-human FDEX studies performed with the aims to assess safety and binding characteristics in healthy volunteers.

Section snippets

FDEX synthesis

4-[18F]FDEX (FDEX) was produced at Austin Health, Melbourne, Australia (Fig. 1). The radiosynthesis was implemented on the iPhase Flexlab module with a yield of 10-12%. FDEX specific activity was 900 mCi/µmol and radiochemical purity was greater than 95%. Radiosynthesis of 4-[18F]FDEX was carried out by reductive amination of (S)nordexetimide (precursor 2) with 4-[18F]fluorobenzaldehyde. No-carrier-added-4-[18F]fluoride was produced from irradiation of [18O]water. [18F]fluoride was then eluted

Dose and safety

The mean administered activity was 246.3 ± 6.09 MBq (range, 240-262 MBq). The specific activity was 43.06 ± 16.75 GBq /umol. There were no adverse or clinically detectable pharmacologic effects in any of the ten participants. No significant changes in vital signs or the electrocardiogram were observed.

Tracer metabolism

Metabolite analysis of venous sampling was determined for 8 of the subjects (5 Female, 3 Male). There was some variability in the rate of metabolism between subjects, however all subjects showed

Discussion

This first-in-human study was performed to assess the safety and binding characteristics in healthy volunteers of FDEX, the first F-18 labelled PET tracer for mAChR to be studied in human subjects. An F-18 labelled mAChR PET tracer has major efficiency and cost advantages compared to C-11 labelled tracers for the study of large patient populations and clinical application due to the radioactive half-life of almost two hours compared to 20 minutes for C-11 permitting wider distribution and

Conclusion

FDEX gives high quality brain images with uptake in the expected distribution of mAChR with very little variance across normal participants. This suggests that FDEX may be a robust tool to detect variations in the muscarinic receptor in the diseased brain and could prove useful for drug development.

CRediT authorship contribution statement

Christopher C. Rowe: Formal analysis, Funding acquisition, Supervision, Writing – original draft, Conceptualization. Natasha Krishnadas: Data curation, Investigation. Uwe Ackermann: Methodology, Resources. Vincent Doré: Formal analysis, Data curation. Rachel Y.W. Goh: Methodology, Resources. Rodney Guzman: Project administration. Lee Chong: Methodology, Resources. Svetlana Bozinovski: Project administration. Rachel Mulligan: Methodology. Richard Kanaan: Formal analysis. Brian Dean:

Declaration of Competing Interest

There are no conflicts of interest for the authors of this paper in relation to this work. Funding for the work was supplied from an Australian Government grant through the CRC for Mental Health. There was no commercial funding for this work and there are no patents relating to this work.

Acknowledgements

Funding for this study was provided by the CRC for Mental Health Ltd., a not-for-profit company established by a grant from the Australian Federal Government.

References (26)

  • FJ Ehlert et al.

    Molecular biology, pharmacology, and brain distribution of subtypes of muscarinic receptor

    Psychopharmacology - 4th Generation of Progress

    (1995)
  • DD Flynn et al.

    Differential regulation of molecular subtypes of muscarinic reeptors in Alzheimer’s disease

    J. Neurochem.

    (1995)
  • AI Levey et al.

    Identification and localization of muscarinic acetylcholine receptor proteins in brain with subtype-specific antibodies

    J. Neurosci.

    (1991)
  • Cited by (8)

    View all citing articles on Scopus
    View full text