MDMA-induced changes in within-network connectivity contradict the specificity of these alterations for the effects of serotonergic hallucinogens

This analysis of data from two clinical trials (n=45) found that MDMA induced similar (neuronal) changes as classical (serotonergic) psychedelics.

Abstract

“It has been reported that serotonergic hallucinogens like lysergic acid diethylamide (LSD) induce decreases in functional connectivity within various resting-state networks. These alterations were seen as reflecting specific neuronal effects of hallucinogens and it was speculated that these shifts in connectivity underlie the characteristic subjective drug effects. In this study, we test the hypothesis that these alterations are not specific for hallucinogens but that they can be induced by monoaminergic stimulation using the non-hallucinogenic serotonin-norepinephrine-dopamine releasing agent 3,4-methylenedioxymethamphetamine (MDMA). In a randomized, placebo-controlled, double-blind, crossover design, 45 healthy participants underwent functional magnetic resonance imaging (fMRI) following oral administration of 125 mg MDMA. The networks under question were identified using independent component analysis (ICA) and were tested with regard to within-network connectivity. Results revealed decreased connectivity within two visual networks, the default mode network (DMN), and the sensorimotor network. These findings were almost identical to the results previously reported for hallucinogenic drugs. Therefore, our results suggest that monoaminergic substances can induce widespread changes in within-network connectivity in the absence of marked subjective drug effects. This contradicts the notion that these alterations can be regarded as specific for serotonergic hallucinogens. However, changes within the DMN might explain antidepressants effects of some of these substances.”

Authors: Felix Müller, Friederike Holze, Patrick Dolder, Laura Ley, Patrick Vizeli, Alain Soltermann, Matthias E. Liechti & Stefan Borgwardt

Summary

INTRODUCTION

Several recent studies have assessed how serotonergic hallucinogens exert their typical effects on the brain. They found that hallucinogenic drugs decrease functional connectivity within resting-state networks, which may explain their profound effects on the psyche.

The selective serotonin reuptake inhibitor sertraline induces changes in within-network FC that closely resemble those seen after LSD. Although these substances are clearly distinct on the level of subjective effects, it is obvious that they share monoaminergic stimulation as a common factor.

We have already hypothesized elsewhere that changes in within-network FC after administration of hallucinogens might be an epiphenomenon, induced by unspecific serotonergic stimulation, and therefore possibly not specific or explanatory for the effects of these drugs.

SSRIs have been reported to induce changes in within-network FC and degree centrality in thalamic and cerebellar regions, and these alterations have also been linked to the mechanism of action of hallucinogens.

fMRI findings that were interpreted as reflecting typical hallucinogenic drug effects might actually be induced by unspecific monoaminergic stimulation. MDMA, a potent serotonin – norepinephrine-dopamine releasing agent, was used to further investigate this possibility.

METHODS

This analysis is based on pooled data sets of two clinical trials conducted in Basel, Switzerland.

Both studies tested the effects of different psychoactive substances on healthy participants using a randomized, placebo-controlled, double-blind, crossover design. They included identical assessments of subjective drug effects, physiological parameters, and plasma drug concentrations.

Participants were recruited by advertisement on the University of Basel website and by word of mouth. They had to provide urine samples to rule out illicit drug use, and women underwent a pregnancy test.

A total of 52 participants received 125 mg MDMA or placebo. Eleven participants had used MDMA before, 12 had used stimulants, five had used hallucinogens, seven had used sedatives, and one subject had used an opioid.

The imaging data were collected using a 3 Tesla MRI system with a 20-channel phased-array radio frequency head coil. A 35-axial slice with a slice thickness of 3.5 mm was used for functional MRI acquisition.

The CONN toolbox 19c was used to process the fMRI data, which was then normalized into a standard stereotactic space and smoothed. The noise was corrected using ART, linear detrending, and linear regression of the six motion parameters.

Pharmacological fMRI studies are prone to biases induced by the respective pharmacological agent. The CompCor approach was used to remove physiological fluctuations and then band-pass filtered the resulting functional images.

Quality assessment comprised three stages: first, all scans were assessed considering the percentage of scrubbed volumes; second, mean and maximal head motion after the scrubbing procedure were assessed using the measure framewise displacement (FD).

Mean and maximum FD after scrubbing were tested for significant differences between conditions. There was no significant difference for mean FD (p = 0.33) but average maximum FD was significantly higher under the drug condition (p = 0.02).

Independent component analysis (ICA) was performed using group-ICA procedures implemented in the CONN toolbox. The number of dimensions was restricted to 20 factors to allow comparisons with 10 established resting-state networks described by Smith et al.

Within-network functional connectivity analysis was performed using ICA components identified as RSNs. A voxel threshold of p 0.001 (uncorrected) and a cluster-size threshold of p 0.005 (Bonferroni corrected; FWE) was used to assess significant clusters.

Degree centrality was calculated by averaging all correlation coefficients between a voxel and all other voxels across the brain and then thresholding at r > 0.15. The resulting images were z-transformed and compared by a two-tailed paired t test.

All physiological parameters were elevated under the drug condition and differed significantly compared with placebo. All parameters were tested for potential confounding effects on FC and were significant at p 0.05 (two-tailed).

Confounding was assessed using two different approaches: first, by correlation analyses of FC values with changes in maximum FD and physiological parameters, and second, by entering demeaned values of all parameters as covariates of no interest in the second-level analyses.

MDMA, LSD, and sertraline primarily stimulate the serotonin system, but also the dopamine system and the noradrenaline system. We performed a systematic literature search to investigate the possibility that results obtained in this and other fMRI studies might be attributable to other neurotransmitter systems than serotonin.

RESULTS

Independent component analysis was used to identify the visual, cerebellar, auditory, frontoparietal, and executive control networks. The default mode network was identified as the posterior part of the network and the sensorimotor network as the anterior part.

MDMA decreased within-network connectivity in the visual networks 1 and 2, the DMN, the cerebellar network, and the sensorimotor network compared with placebo. There were no significant alterations in within-network connectivity in the auditory network, the executive control network, or the frontoparietal networks.

MDMA decreased degree centrality in the cerebellum, occipital regions, and the brain-stem compared with placebo. It increased degree centrality in several cortical regions, including the post- and precentral gyrus.

Using an uncorrected threshold of p 0.10, there was no evidence that maximal FD or any of the physiological parameters were systemically associated with FC measures in our ICA analyses or in our degree centrality analysis.

Nine studies investigated effects of monoaminergic drugs on the dopamine system, including levodopa, haloperidol, escitalopram, methylphenidate, d-amphetamine, bupropion, tryptophan depletion/loading, and citalopram. Sertraline, LSD, and MDMA were not investigated in other studies.

DISCUSSION

We tested the acute effects of MDMA on within-network FC and degree centrality. Our results did not support the notion that alterations in within-network FC are explanatory for the typical acute effects of hallucinogenic drugs.

MDMA acutely decreased functional connectivity within two visual networks, the default mode network (DMN), the sensorimotor network, and the cerebellar network. However, the cerebellar network was not affected, which is in line with previous studies on sertraline and LSD.

Although the within-network FC of the DMN does not seem to be specific for acute hallucinogenic effects, it is possible that the decreased FC of the posterior DMN is important for therapeutic long-term effects of these substances.

We found no evidence that MDMA induced similar changes in degree centrality as reported for the SSRI escitalopram, but minor changes were observed in a cluster covering small portions of the thalamus and parts of the left caudate.

This study underlines the difficulties in interpreting fMRI data induced by hallucinogens, such as findings induced by unspecific drug effects or various sources of confounders. Moreover, comparison with related psychoactive substances might also be worthwhile.

Although studies were induced by monoaminergic stimulation, it would be unclear how equivalent doses across different substances could be defined, and there are some overlaps with studies induced by LSD. Moreover, sample sizes and statistical thresholds varied across studies and there are additional differences.

In this study, low-dimensional ICA was used to analyze hallucinogenic networks. It is possible that hallucinogenic specific effects within subnetworks could be detected using high-dimensional approaches.

MDMA, LSD, and the SSRI sertraline stimulate the serotonin system, but they also stimulate the dopamine system to some extent, and MDMA additionally simulates the noradrenaline system. Therefore, the observed effects on within-network FC might be mediated through other systems than the serotonin system.

This study demonstrated that MDMA induces within-network FC alterations similar to those seen after administration of LSD. These alterations are not associated with potential sources of bias.

FUNDING AND DISCLOSURE

This work was supported by the Swiss National Science Foundation and the University Hospital Basel and was licensed to Mind Medicine Inc. after study completion.