First-in-Human Studies with [11C]COU, a substrate-based Radiotracer for MAO-B

Introduction: Alzheimer’s disease (AD) is characterized by accumulation of tau neurofibrillary tangles and amyloid plaques, and there are FDA-approved radiopharmaceuticals for PET imaging of both misfolded proteins. New immunotherapies to treat AD have also received regulatory approval (e.g. lecanemab) - amyloid-directed antibodies that can slow cognitive decline - and use of amyloid + tau PET were critical to their development. Data suggests these new AD drugs will be more effective if started earlier in AD progression than can be detected by amyloid or tau PET. Evidence suggests that astrocytic dysfunction occurs earlier in the dementia cascade than amyloid can be observed by PET¹ and, since greater expression of MAO-B has been observed in astrocytes in AD, we are interested in MAO-B PET as an imaging biomarker of astrocytosis. While historical tracers like [¹¹C]deprenyl can provide information on MAO-B expression levels, they do not provide functional measurements of enzyme activity. To address this, we hypothesize that a substrate/trapped metabolite approach, akin to FDG and PMP, will give a measure of MAO-B activity. Herein we describe translation and first-in-human studies with [¹¹C]COU, a substrate-based radiotracer for MAO-B.

Methods: [¹¹C]COU was prepared as previously reported by our lab,² and underwent standard QC testing. Rodent and nonhuman primate (NHP) imaging were conducted using a Concorde P4, including baseline studies and MAO-B challenge studies. Biodistribution experiments in rodents were used to estimate human dosimetry using Olinda, and pharm-tox testing was conducted in rats at 100x the max. human dose. Clinical translation was enabled under an approved IND and IRB protocol. Initially, 4 healthy controls (2M/2F) received 18 mCi of [¹¹C]COU i.v. and the subjects’ brains were imaged for 60 min to investigate regional brain kinetics. An additional 30 min whole-body image was acquired for each of the 4 subjects to determine human biodistribution/dosimetry. Venous blood was acquired throughout to conduct metabolite analysis. Test/re-test studies in an additional 8 healthy volunteers are ongoing, which include arterial sampling to determine a metabolite-corrected input function, informing our kinetic modeling strategy (goal is to use k3 as a pseudo-measure of MAO-B activity).

Results: [¹¹C]COU was synthesized in >15% radiochemical yield (140±20 mCi) from [¹¹C]CO₂. Preclinical studies confirmed 40-fold higher affinity for MAO-B than -A. Imaging with [¹¹C]COU confirmed extensive brain uptake in rodents and NHPs that was significantly reduced with irreversible (deprenyl, rodents) and reversible (lazabemide, NHP) MAO-B inhibitors. Pharm-tox testing showed that a single i.v. dose of COU administered to Sprague-Dawley rats at 86 μg/kg resulted in no drug-related clinical findings. The maximum human dose was set at ≤0.14 µg/kg (100x below the human effective dose corresponding to the rodent pharm-tox dose) and 18 mCi (no dosimetry concerns). Initial human studies revealed extremely high penetration of [¹¹C]COU into CNS (>100k counts in brain, see Figure). The tracer gets trapped in brain fast (~10 min), but washout in some areas is apparent suggesting tracer is not flow limited. Whole body scans revealed high uptake in heart (which express MAO-B like brain), and excretion through liver + kidneys, consistent with preclinical findings. There have been no adverse events and we are evaluating strategies to determine optimal kinetic modeling approach.

Conclusions: [¹¹C]COU is a first in class substrate-based radiotracer for MAO-B that functions via a trapped metabolite approach, akin to FDG and PMP. IND was approved by FDA, and human studies commenced in 2023. Future studies will examine differences in [¹¹C]COU signal between controls and dementia patients, and determine sensitivity of the PET signal to deprenyl challenge.

Acks: R21AG073686, PET Techs, Jim Pool.

References: 1. Leclerc et al, Sci World J 2013:589308; 2. Brooks et al, ACS Med Chem Lett. 2020;11:2300.

• Download figure
• Open in new tab
• Download powerpoint