Significant decreases in frontal and temporal [11C]-raclopride binding after THC challenge

Abstract

Δ9-tetrahydrocannabinol (THC) increases prefrontal cortical dopamine release in animals, but this is yet to be examined in humans. In man, striatal dopamine release can be indexed using [11C]-raclopride positron emission tomography (PET), and recent reports suggest that cortical [11C]-raclopride binding may also be sensitive to dopaminergic challenges. Using an existing dataset we examined whether THC alters [11C]-raclopride binding potential (BP_ND) in cortical regions. Thirteen healthy volunteers underwent two [11C]-raclopride PET scans following either oral 10 mg THC or placebo. Significant areas of decreased cortical [11C]-raclopride BP_ND were identified using whole brain voxel-wise analysis and quantified using a region of interest (ROI) ratio analysis. Effect of blood flow on binding was estimated using a simplified reference tissue model analysis. Results were compared to [11C]-raclopride test–retest reliability in the ROIs identified using a separate cohort of volunteers. Voxel-wise analysis identified three significant clusters of decreased [11C]-raclopride BP_ND after THC in the right middle frontal gyrus, left superior frontal gyrus and left superior temporal gyrus. Decreases in [11C]-raclopride BPND following THC were greater than test–retest variability in these ROIs. R1, an estimate of blood flow, significantly decreased in the left superior frontal gyrus in the THC condition but was unchanged in the other ROIs. Decreased frontal binding significantly correlated to catechol-o-methyl transferase (COMT) val108 status. We have demonstrated for the first time significant decreases in bilateral frontopolar cortical and left superior temporal gyrus [11C]-raclopride binding after THC. The interpretation of these findings in relation to prefrontal dopamine release is discussed.

Introduction

Δ9-tetrahydrocannabinol (THC), the main psychoactive component of cannabis sativa, increases prefrontal cortical dopamine levels in animal microdialysis studies (Chen et al., 1990, Jentsch et al., 1997, Pistis et al., 2002) and also increases the firing rate of dopaminergic prefrontal cortical neurones in vitro (Diana et al., 1998). Prefrontal cognitive efficiency is in part dependent on the maintenance of stable prefrontal dopamine levels (Williams and Castner, 2006) and stable prefrontal dopamine levels may also contribute to higher brain functions relevant to motivation and emotion (Nieoullon and Coquerel, 2003). Taken together, modulation of processes associated with prefrontal cortical function, such as verbal recall, attention and motivation (D'Souza et al., 2008), following THC may occur as a result of the effect of THC on dopaminergic neurones (Wilson and Nicoll, 2002). These studies predict that THC administration increases dopamine release in the human prefrontal cortex. However this has not yet been examined in the human brain in vivo.

In humans, change in striatal dopamine levels can be estimated in vivo using [11C]-raclopride positron emission tomography (PET). Extracellular dopamine release is thought to increase competition between [11C]-raclopride and dopamine for dopamine D_2/3 receptor occupancy and promotes receptor internalisation, thus decreasing [11C]-raclopride binding (Goggi et al., 2007, Laruelle, 2000, Skinbjerg et al., 2010). Whilst significant but small decreases in striatal [11C]-raclopride binding have been reported after acute administration of inhaled THC (Bossong et al., 2009), this result was not replicated in a larger cohort of volunteers following an acute challenge with oral THC (Stokes et al., 2009).

In addition to decreases in [11C]-raclopride binding potential in the striatum, decreases in [11C]-raclopride binding potential have also been observed in extrastriatal regions following drug or behavioural challenges which may increase extracellular dopamine. Specifically, intravenous methamphetamine decreases [11C]-raclopride binding in the prefrontal and anterior cingulate cortices (Piccini et al., 2003), and decreased binding has also been reported in frontal cortical motor areas during a motor sequence learning task (Garraux et al., 2007), and in the anterior cingulate cortex during a spatial working memory task (Sawamoto et al., 2008) and a Tower of London planning task (Egerton et al., 2009). There is however some debate about the validity of using [11C]-raclopride PET to index changes in cortical dopamine levels (Egerton et al., 2009), as [11C]-raclopride binding is far lower in cortical than striatal areas due to the lower density of D₂ receptors, resulting in low signal to noise ratios and poor accuracy. Whilst decreases in cortical [11C]-raclopride binding potential may reflect increased dopamine, it is also possible that they may simply reflect false positive findings, due to low within-subjects reliability of extrastriatal measurement. Alternatively, the observed decreases in [11C]-raclopride binding potential may be ‘true’ decreases, but arising consequential to non-dopaminergic processes—such as differences in radiotracer washout or delivery under challenge compared with control conditions due to blood flow effects.

In an exploratory voxel level, whole brain analysis of an existing dataset (Stokes et al., 2009), we investigated whether THC administration was associated with decreases in [11C]-raclopride binding potential in extrastriatal brain regions. Following detection of significant clusters of decreased [11C]-raclopride binding potential in the THC compared to placebo condition, we compared the magnitude of change detected to the test–retest reliability of [11C]-raclopride in identical regions of interest in a separate cohort of volunteers, to estimate to what extent these findings differed from chance-level observations. For estimation of change in [11C]-raclopride binding potential following THC administration, we compared results of two analytical methods with differing resilience to concomitant changes in blood flow, the equilibrium ratio method (Carson et al., 1997) and kinetic modelling using the simplified reference tissue model (SRTM) (Lammertsma and Hume, 1996). To further explore the extent to which there may be a dopaminergic basis to the observed changes, we also determined whether there was any relationship between THC induced change in binding potential and catechol-o-methyl transferase (COMT) val^108/158met polymorphism.