Enantio-selective cognitive and brain activation effects of N-ethyl-3,4-methylenedioxyamphetamine in humans

With the advent of cognitive neuroscience and functional brain imaging, substituted amphetamine derivatives such as MDMA and the closely related N-ethyl-3,4-methylenedioxyamphetamine (MDE) can be studied in humans with a combination of subjective, neuropsychological and brain-imaging measures. Earlier research indicated that these agents produce entactogenic subjective effects (changes in affect, perception and cognitive style) and act on serotonergic and catecholaminergic systems. Animal work has shown enantiomer-specific effects for MDE, with behavioural differences between the (R)- and (S)-forms, but human data on stereoselective effects were lacking. Spitzer and colleagues set out to characterise the enantio-selective effects of (R)-MDE and (S)-MDE in healthy volunteers. The study aimed to compare subjective experience, neuropsychological performance and functional MRI activation during low-level visual and higher-level semantic tasks after a single oral dose of each enantiomer, using a randomised double-blind crossover design. The investigators hypothesised that the two enantiomers would produce dissociable effects on mood, cognition and regional brain activation, reflecting differential actions on serotonergic and dopaminergic systems and on cortical areas such as primary visual cortex and frontal regions.

Five healthy right-handed male physicians without history of psychiatric, neurologic disorders or drug abuse were recruited and gave written informed consent. The study used a randomised double-blind crossover design with at least a two-week washout between sessions. Each subject received a fixed oral dose of 70 mg MDE hydrochloride (59 mg base) of one enantiomer per session; after the washout the other enantiomer was administered. Baseline measures were taken at 09:00 and fMRI measurements occurred about 90 minutes after drug intake. Neuropsychological testing and psychopathometric scales were administered immediately after scanning in a randomised order. Blood samples for enantioselective plasma pharmacokinetics were collected frequently up to 34 hours post-dose; concentrations were measured by an enantioselective HPLC method and pharmacokinetic parameters were derived using a non-compartmental model (TopFit 2.1). Behavioural paradigms outside the scanner included a visual pop-out search task (set sizes 4–16; 96 trials), a computerised modification of the Wisconsin Card Sorting Test (48-card pack; switching cost derived from reaction times), and a semantic priming lexical-decision task with directly associated, indirectly associated, non-associated words and pseudo-words. In-scanner tasks were block-designed semantic and colour similarity decisions using 120 word pairs and 120 coloured-asterisk pairs (40 trials per category), with activation epochs of 21 s (six whole-brain volumes) alternated with controlled rest. Subjects responded by button press and presentation times were subject-controlled up to specified maxima. Imaging data were acquired on a 1.5 Tesla Siemens Magnetom VISION using T2*-weighted echo-planar imaging (38 slices, TR=3.480 s, TE=50 ms, voxel interpolation to 2×2×2 mm in normalisation). Preprocessing and statistical analyses were performed with SPM99 (realignment, coregistration, normalisation, 8 mm FWHM smoothing). A box-car general linear model convolved with the haemodynamic response was used; low-frequency drifts were removed and images globally scaled. Drug-dependent contrasts for semantic and colour tasks were thresholded at voxelwise p<0.001 uncorrected and cluster-level p<0.05 uncorrected. Statistical analyses of behavioural and psychometric data used repeated-measures MANOVA with Newman–Keuls post-hoc tests where appropriate; Wilcoxon tests for paired samples were applied for small-sample comparisons.

Pharmacokinetics: Both enantiomers were measurable in plasma and had comparable times to maximum concentration and elimination half-lives, but plasma concentrations of the (R)-enantiomer were significantly larger due to higher clearance of the (S)-enantiomer. Subjective measures: The two enantiomers produced markedly different subjective profiles. On the Actual Subjective Mental State Scale (BfS) there was a significant difference with (R)-MDE producing depressive mood and (S)-MDE producing manic/elevated mood (P<0.05, Wilcoxon). The Depression Scale (DS) showed a numerical increase under (R)-MDE and a decrease under (S)-MDE but this contrast did not reach significance (P=0.22). Somatic complaints (BL) trended higher after (R)-MDE and lower after (S)-MDE (P≈0.07). Scores on the Altered States of Consciousness OAV instrument showed a large and significant difference on the 'oceanic boundlessness' scale ((S)-MDE: 86.1; (R)-MDE: 3.9; P<0.05), with (S)-MDE producing pronounced entactogenic-type experiences. Neuropsychology: In the visual pop-out search task a repeated-measures MANOVA found a main effect of drug (F(2,38)=9.16; P<0.001). Post-hoc tests showed a significant decrease in visual search time under (R)-MDE compared with pre-drug (P=0.001) and compared with (S)-MDE (P=0.001). The small increase in search time after (S)-MDE versus baseline was not significant. For WCST switching costs there was a main effect of drug (F(2,42)=5.45; P<0.01). Switching costs increased significantly after (S)-MDE relative to baseline (P<0.05) and relative to (R)-MDE (P<0.01); the decrease in switching costs under (R)-MDE versus baseline was not significant. Semantic priming showed a trend for the indirectly related condition (P=0.067): (S)-MDE produced a 38% decrease in reaction time while (R)-MDE produced a 5% increase, suggesting greater availability of remote associations with (S)-MDE. fMRI: Direct contrasts between enantiomers revealed distinct activation patterns. During the semantic judgement task (R)-MDE significantly activated right visual cortical areas and left frontal regions, whereas (S)-MDE produced right frontal and bilateral temporoparietal activation. In the colour discrimination task (R)-MDE again increased occipital visual activation; (S)-MDE produced no increased activation in visual regions. Imaging statistics were reported at voxelwise p<0.001 uncorrected with cluster-level p<0.05 uncorrected. No severe hallucinatory visual phenomena were reported by subjects at the doses used.

Spitzer and colleagues interpret the pattern of results as evidence for stereoselective pharmacological effects of MDE in humans. They propose that (R)-MDE facilitates bottom-up visual information processing, as reflected by occipital activation and faster parallel visual search, and that this effect is accompanied by depressive subjective changes and activation of left frontal areas that correlated with numerically lower switching costs. The authors link these observations to psychological models in which negative mood narrows attentional scope and promotes systematic processing. They also note that although (R)-MDE activated visual cortex, subjects did not report prominent visual hallucinations at the administered dose; higher doses or experimental paradigms with sensory stimulation might be required to evoke hallucinatory phenomena. In contrast, the (S)-enantiomer produced entactogenic subjective effects (elevated mood, talkativeness), increased switching costs on the WCST and right frontal and temporoparietal activation, which the investigators relate to reduced context-driven conceptual control and greater availability of remote associations. On this basis the authors suggest that the entactogenic properties of racemic MDE (and by extension MDMA-like effects) are likely mediated by the (S)-enantiomer. With respect to neurotoxicity, the discussion highlights three converging observations that make (R)-MDE a candidate mediator of deleterious effects reported after repeated substituted-amphetamine use: preferential activation of visual cortical regions with dense serotonergic projections, engagement of cortical areas implicated in cognitive functions known to be impaired in heavy users, and substantially higher plasma concentrations of the (R)-enantiomer under the dosing regimen used. The investigators therefore recommend that any future therapeutic applications consider using the (S)-enantiomer alone. The authors acknowledge dose-dependency as an important consideration (higher doses might yield different sensory phenomena) and argue that repetition effects do not explain their findings because the order of enantiomer administration was balanced across subjects. They conclude that stereo-specific drug-challenge studies combining experiential, behavioural, neuropsychological and functional imaging methods are a valuable approach for probing the neurobiology of psychoactive drugs and suggest further work along these lines.