PET study design
In this single centre, open-label study (EudraCT 2017-000670-11), male participants (25–60 years old, body weight 55–100 kg, BMI 18.5–30 kg/m2) received a single oral dose of MR309 across 4 dosing cohorts (200–800 mg). All participants were considered healthy and subject to similar inclusion and exclusion criteria as described for the PK study, with the additional exclusion of any contraindications to radial artery cannulation and magnetic resonance imaging (MRI).
Prior to study treatment, participants underwent a single brain MRI scan to confirm suitability to take part in the study and obtain anatomical information to support PET scan analysis.
Each subject underwent three PET scans with [11C]SA-4503 at baseline, 2h and 8h after a single oral dose of MR309. Participants fasted for ≥8 hours prior to receiving MR309. Arterial blood samples were collected from a radial artery cannula throughout each PET scan to estimate the time-course of total blood and plasma radioactivity and the concentration of unmetabolized [11C]SA-4503.
PET data quantification
The primary objectives were to evaluate brain σ1R occupancy following a single oral dose of MR309 and to describe the relationship between σ1R occupancy and MR309 plasma concentration at the time of the PET scan. The primary quantification parameter was the total volume of distribution (VT) for [11C]SA-4503 at each PET scan. VT represents the partition coefficient for SA-4503 between brain and plasma at equilibrium, and is proportional to the total binding of SA-4503 (displaceable and non-displaceable) in the relevant brain region (Innis et al., 2007). σ1R occupancy was calculated by comparing VT at baseline and post-dose [11C]SA-4503 for each subject, and σ1R occupancy for each scan was related to the plasma concentration of MR309.
Brain MRI images underwent brain extraction, grey matter segmentation and co-registration to a standard reference space (MNI152 template brain image and associated CIC atlas), (Grabner, Janke, Budge, Smith, Pruessner & Collins, 2006; Tziortzi et al., 2011) and were nonlinearly warped to each participant’s MRI image to enable automated definition of regions of interest (ROIs). ROIs included the frontal, temporal and occipital lobes, caudate nucleus, putamen, thalamus, hippocampus and cerebellum. Representative orthogonal cross sections of co-registered PET and MRI images are shown in Figure 1.
Dynamic PET images were corrected for motion using a frame-to-frame registration process with a normalized mutual information cost function, and co-registered to each participant’s MRI scan. ROIs defined on the MRI images were applied to the dynamic PET data to derive regional time-activity curves (TACs), with activity concentrations expressed as standardised uptake values (SUV), calculated by normalising the measured radioactivity concentration to the injected radioactivity and the participant’s bodyweight.
Regional TACs were analysed using the alternative multilinear analysis 1 (MA1) kinetic model with a metabolite-corrected arterial plasma input function, to generate regional VT values. σ1R occupancy for each post-dose PET scan was calculated from the regional VT estimates using an occupancy plot analysis (Cunningham, Rabiner, Slifstein, Laruelle & Gunn, 2010). The occupancy plot method makes the assumption that the volume of distribution of the non-displaceable component (VND) is uniform across an individual’s brain, and that the fractional occupancy of the target in each post-dose scan is uniform across all ROI. VND was constrained to be the same for each post-dose PET scan in each individual.
The relationship between σ1R occupancy and drug exposure was then explored using data-driven model fitting. To characterize the relationship between MR309 plasma concentration and σ1R occupancy, the σ1R occupancy estimates were plotted against MR309 plasma concentration measured at the start of each post-dose PET scan. Four models were fitted and compared: the optimal model to describe the relationship between MR309 plasma concentration and σ1R occupancy was selected based on goodness of fit and AIC values (Table 1). For each model, EC50 , Emax, Hill slope and Akaike information criterion (AIC) values (measure of goodness of fit) were obtained. (Bozdogan, 1987)
PK of single-dose MR309 and its metabolites were assessed as a secondary objective. Plasma concentrations of MR309 and its metabolites were assessed as described in the PK study with the endpoints of Cmax, Tmax, AUCt (area under the plasma concentration time curve from the time of dosing to the last measurable concentration), AUCINF (area under the plasma concentration time curve from the time of dosing to infinity), LambdaZ (estimate of terminal elimination rate constant) and t1/2Z (apparent terminal phase half-life). Safety and tolerability were assessed pre-dose and at regular intervals up to Day 4 post-dose, as described in the PK study.