Measuring drug-related receptor occupancy with positron emission tomography Jan Passchier, a, * Antony Gee, a Antoon Willemsen, b Willem Vaalburg, b and Aren van Waarde b a GlaxoSmithKline, Addenbrooke’s Centre for Clinical Investigation, Addenbrooke’s Hospital, Hills Road, Cambridge CB2 2GG, UK b PET Center, Groningen University Hospital, P.O. Box 30.001, 9700 RB Groningen, The Netherlands Accepted 16 May 2002 Abstract Several techniques can be used to measure indirectly the effect of drugs (e.g., EEG, fMRI) in healthy volunteers and in patients. Although each technique has its merits, a direct link between drug efficacy and site of action in vivo usually cannot be established. In addition, when the specific mode of action of a drug has been determined from preclinical studies, it is often not known whether the administered dose is optimal for humans. Both industry and academia are becoming more and more interested in determining the dose-related occupancy of specific targets caused by administration of drugs under test. Positron emission tomography (PET) and single-photon emission computed tomography (SPECT) are noninvasive imaging techniques that can give insight into the rela- tionship between target occupancy and drug efficacy, provided a suitable radioligand is available. Although SPECT has certain advantages (e.g., a long half-life of the radionuclides), the spatial and temporal resolution as well as the labeling possibilities of this technique are limited. This review focuses on PET methodology for conducting drug occupancy studies in humans. Ó 2002 Elsevier Science (USA). All rights reserved. 1. Introduction Currently, drug dosing regimens for patients are based mainly on the outcome of preclinical and Phase I– II human studies such as in vitro autoradiography, dose–effect relationships, plasma concentration, tolera- bility, EEG, and fMRI [1–6]. Although each of these studies has its own merits, they cannot give direct insight into the relationship between the amount of drug ad- ministered and the occupancy of its target and correlate these results to the effectiveness of the drug in treating disease. For example, a drug that is assumed to exert its action in the central nervous system (CNS) may instead provoke a systemic release of cortisol or noradrenaline, leading to changes in EEG readings or changes in blood flow which can be misinterpreted as the drug having a direct effect in the CNS. Similarly, measuring plasma levels of a drug following a single or repeated adminis- tration and assuming that the target occupancy is lin- early related to the dose or the concentration in plasma may be incorrect. Two cases can easily be identified in which plasma concentration is not a good indication of target occupancy. First, if brain uptake is high and the off-rate of the drug from the target is slow, a significant fraction of the target receptor will remain occupied for prolonged pe- riods, even if the clearance of the drug from plasma is rapid. The pharmacokinetics of the serotonin 5-HT 1A antagonist [carbonyl- 11 C]WAY 100635 provides a good example. This radioligand for PET imaging of 5-HT 1A receptors in the human brain is both rapidly metabo- lized and rapidly cleared from plasma, leading to very low levels of radioactivity in the circulation and a per- centage intact parent of about 10% in plasma at 10 min postinjection [7,8]. Based on the plasma kinetics, this compound would be dismissed as a potential drug. However, the time–activity curves for the regions of interest (e.g., dorsal raphe) show a rapid and high up- take of 11 C into these regions followed by a very slow release [7,9]. In this case, receptor occupancy cannot be deduced from the plasma concentration of the Methods 27 (2002) 278–286 www.academicpress.com * Corresponding author. 1046-2023/02/$ - see front matter Ó 2002 Elsevier Science (USA). All rights reserved. PII:S1046-2023(02)00084-1