Abstract
The main inhibitory neurotransmitter in the central nervous system is γ-aminobutyric acid (GABA). GABA transporter type 1 (GAT-1) is the principal GABA transporter in the brain, and it plays a crucial role in modulating GABA signaling. Its potential role in several neuropsychiatric disorders makes it an important target to study. Although PET radiotracers exist for the GABA receptors, none have been successful for imaging GAT-1. The focus of this work was to evaluate the kinetic behavior of 4 novel (18)F-labeled PET radiotracers ((18)F-GATT-31, (18)F-GATT-34, (18)F-GATT-39, and (18)F-GATT-44) for imaging GAT-1 in nonhuman primates and to select the best radiotracer to progress to human studies. Methods: Twenty scans were acquired from 4 rhesus monkeys (Macaca mulatta). Each monkey received 0.5 mg/kg of tiagabine given approximately 20 min before radiotracer injection and underwent baseline and blocking scans with (18)F-GATT-31, (18)F-GATT-34, (18)F-GATT-39, or (18)F-GATT-44 on a small-animal PET scanner. During each scan, arterial blood was collected for measurement of the input function. Kinetic analysis was performed using a 1-tissue compartment model, 2-tissue reversible model (k (4) > 0), and 2-tissue irreversible model (k (4) = 0), including a blood volume fraction term and a time-delay term. Results: All radiotracers exhibited good, albeit slow, brain uptake within the cortical and subcortical gray matter regions and cerebellum. Peripheral metabolism was slow for (18)F-GATT-34, (18)F-GATT-39, and (18)F-GATT-44, with greater than 75% remaining as the parent compound, but was somewhat faster for (18)F-GATT-31 (63%) over the 3-h scans. The 1-tissue compartment model delivered a reliable performance on the basis of the overall lowest Akaike information criterion and an SE of less than 10% for the volume of distribution. (18)F-GATT-39 and (18)F-GATT-34 were eliminated from progressing to human studies because of low brain uptake or low specific binding. The 2 remaining radiotracers had similar characteristics, with (18)F-GATT-44 showing slightly superior performance over (18)F-GATT-31, with more consistent tiagabine blocking results (65%-71%) and with nondisplaceable binding potential (BP(ND)) values ranging from 1.2 to 4.2 across gray matter structures. Conclusion: We successfully developed 4 GAT-1 selective radiotracers and evaluated them in nonhuman primates with kinetic analysis and blocking studies with tiagabine. Of these compounds, (18)F-GATT-44 exhibited consistent results and reasonable BP(ND) values and will progress to human studies.