A Briefer on MDMA Toxicity

1. Introduction

Despite the promise of MDMA’s role in the future of clinical psychiatry and its demonstrated efficacy in treating PTSD, concerns about the safety of MDMA have continued to make the rounds. Despite well-known cases (https://pubmed.ncbi.nlm.nih.gov/12351788/) where MDMA neurotoxicity was falsely reported, subsequent nonclinical studies with racemic MDMA have shown consistent markers of toxicity, in some cases at human therapeutic and recreational doses. This briefer will describe some of these findings and will discuss possible strategies for overcoming these challenges so that MDMA’s scope of possible clinical use can be widened.

2. Neurological Adverse Effects of MDMA

2.1 MDMA Neurotoxicity in Animal Models

Incidence of MDMA neurotoxicity has been demonstrated most clearly in preclinical animal studies, as is standard practice for small-molecule pharmaceutical research. Nonhuman primates have been commonly used as a model organism for MDMA neurotoxicity research given rough approximations in CNS structure to humans. In an excellent review, (Capela et al.) found that in a general sense, MDMA neurotoxicity is broadly observable in three distinct phenomena: serotonergic terminal loss, neuronal degeneration in several brain regions, and acute neuroinflammation (https://pubmed.ncbi.nlm.nih.gov/19373443/). (Hatzidimitriou et. al) found abnormal brain 5-HT innervation patterns were still evident in MDMA-treated monkeys seven years after a single dose of administration (https://pubmed.ncbi.nlm.nih.gov/10366642/). (Ricaurte et. al) administered s.c doses of MDMA to squirrel monkeys at doses approximating human therapeutic strength MDMA and found only partial recovery in the number of serotonergic presynaptic markers (https://pubmed.ncbi.nlm.nih.gov/1374470/). (Ricaurte et al.) in an earlier paper also demonstrated that MDMA selectively damages serotonergic neurons in nonhuman primates (https://pubmed.ncbi.nlm.nih.gov/2454332/). These results suggested that MDMA produces lasting effects on serotonergic neurons in nonhuman primates, with most brain regions showing evidence of persistent denervation and some showing signs of reinnervation (thalamus) or possibly even hyperinnervation (hypothalamus). (Mueller et al.) administered single oral doses of MDMA to squirrel monkeys and found that even the lowest single doses (equivalent to 1.6 mg/kg in humans or ~120 mg) produced significant serotonergic neurochemical deficits (https://academic.oup.com/ijnp/article/16/4/791/790204).

Significant evidence of MDMA neurotoxicity has also been demonstrated in rodent species. (Touriño et al) demonstrated notable hyperthermia, DA terminal loss, and glial activation in MDMA-treated mice at 20 mg/kg (https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0009143), though the authors found that these effects could be attenuated with the co-administration of THC. (Mueller et al.) found significant signs of monoamine terminal loss at room temperature in both rat and mouse (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3558829/).
(Herndon et. al) found a robust microglial response within 12 hours of MDMA administration, however found that direct injection of MDMA did not significantly activate microglia, suggesting a significant contribution of MDMA metabolites to MDMA neurotoxicity (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3930364/). (Cadoni et. al) took a different approach and examined the effects of a low dose repeat regimen of MDMA on adolescent rodents. Groups treated at 5 and 10 mg/kg showed reduced density of tyrosine hydroxylase positive neurons and a reduction of basal dialysate DA in the NAc core. MDMA-pretreated rats also showed behavioral sensitization to an MDMA challenge at adulthood and potentiation of MDMA-induced increase of dialysate DA in the NAc core, but not in the NAc shell. In addition, MDMA-treated rats displayed a deficit in recognition memory (https://pubmed.ncbi.nlm.nih.gov/28603026/). (Frau et al.) examined the neurotoxicity of racemic MDMA, as well as its two optical isomers, R-MDMA and S-MDMA. Racemic MDMA consistently increased levels of CD11b and GFAP, both markers of neuroinflammation and microglial activation, however this effect was found to be mediated solely by the S-enantiomer. S-MDMA was also associated with hyperlocomotion and hyperthermia, whereas R-MDMA did not contribute to these effects (https://onlinelibrary.wiley.com/doi/epdf/10.1111/jnc.12060).

Cognitive and behavioral batteries have also been evaluated on MDMA-treated animals to evaluate potential effects on learning and memory. (Sprague et. al) compared untreated rodents to those treated with s.c MDMA on spatial learning and memory using the Morris water maze test. No statistical differences were found in MWM platform acquisition latency or pathlength between controls and MDMA-treated animals. Probe trials revealed significantly higher proximity score averages and significantly reduced preference for the target quadrant in the MDMA-treated animals. The study demonstrates that hippocampal serotonergic lesions induced by MDMA may be ostensibly linked to a reference memory deficit in rats tested with the MWM (https://pubmed.ncbi.nlm.nih.gov/12834800/).

(Moyano et al.) found that acute MDMA administration to rats resulted in reduced NR1 and N2RB protein levels and impaired passive avoidance training (https://pubmed.ncbi.nlm.nih.gov/14985918/). (Gouzoulis-Mayfrank et. al) evaluated the long-term neuropsychological effects of adolescent MDMA exposure on rodents, through the performing of repeated learning and memory tasks during the rodents lifespan. The performance of drug treated rats was notably worse on tasks requiring cognitive flexibility, such as the water maze task. While some persistent cognitive deficits were found, no significant group differences in serotonin or dopamine levels were found in any of the measured regions of the brain changes, cortical or subcortical (https://www.scirp.org/reference/referencespapers?referenceid=2113205). Lastly, (Fernandez-Castillo et al.) conducted a wide ranging study on MDMA’s impact on gene expression patterns, finding that genes related to inflammation, cytotoxicity-associated neuroadaptation, and immune responses (https://www.researchgate.net/publication/304817206_MDMA_Ecstasy_and_Gene_Expression_in_the_Brain).

2.2 MDMA-Metabolite Neurotoxicity Studies

In addition to results demonstrating the neurotoxicity of MDMA in animals, research has been conducted to better elucidate mechanisms of this toxicity. Much of this has centered on the metabolic profile of MDMA and on the exacerbating properties of hyperthermia, which will be discussed in a later section. (Jones at al) found that glutathione (GSH) and N-acetylcysteine conjugates of alpha-methyldopamine, which are second-order metabolites of MDMA and selective neurotoxins, were present in rat striatum after subcutaneous injection of MDMA (https://jpet.aspetjournals.org/content/313/1/422?ck=nck). (Bai et. al) had previously shown that both of these conjugates were selective neurotoxins in the mouse brain (https://pubs.acs.org/doi/abs/10.1021/tx990084t). Using mouse brain synaptosomes, (Barbosa et al.) evaluated the prooxidant effects of the A-MeDA, N-Me-a-MeDA, 5-GSH-a-MeDA, and other thiolated metabolites of MDMA. They found that the metabolites time- and dose-dependently increased H2O2 and quinoprotein production, though notably having no effect on mitochondrial polarization (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3346242/). (Capela et al) further demonstrated that thioether MDMA metabolites are strong neurotoxins, significantly more than their correspondent parent catechols and induced a concentration dependent delayed neuronal death, accompanied by activation of caspase 3, which occurred earlier in hyperthermic conditions. Furthermore, thioether MDMA metabolites time-dependently increased the production of reactive species, concentration-dependently depleted intracellular GSH and increased protein bound quinones (https://pubmed.ncbi.nlm.nih.gov/17467183/).

2.3 Cognitive effects of MDMA use in humans

Though accurate measurements of the extent of neurotoxicity in humans are difficult to collect, pilot studies have been conducted in test subjects to better elucidate the effects of MDMA use on cognitive tasks, learning, and memory. (Halpern et al.) assessed cognitive function in 52 MDMA users with no prior drug experience and only found moderate detrimental cognitive effects across all subjects (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3053129/pdf/nihms-247953.pdf). (Schilt et al.) examined the relationship between MDMA use and cognitive performance in 188 subjects, split into MDMA and non-MDMA using groups. Though immediate follow-ups did not show a statistically significant discrepancy in the battery of neurocognitive tests, change scores on immediate and delayed verbal recall and verbal recognition were significantly lower in the group of incident MDMA users compared with persistent MDMA-naive subjects (https://pubmed.ncbi.nlm.nih.gov/17548754/). (Wagner et. al) found that in a data set of 96 naïve non-polydrug MDMA users, cognitive performance declined with repeat use as well as with later onset of MDMA use (https://www.sciencedirect.com/science/article/abs/pii/S0278584614002413#:~:text=Cognition%20and%20psychopathology%20were%20associated,attention%20and%20information%20processing%20speed). (Reay et. al) showed that MDMA polydrug users showed impairments in set shifting, memory updating, and social judgement processes (https://journals.sagepub.com/doi/abs/10.1177/0269881106063269). Finally, (Hanson et. al) also found notable signs of cognitive impairment in MDMA users (in memory, attention, and motor function), however this was primarily limited to users with clinically dysfunctional MDMA use (https://psycnet.apa.org/record/2004-12175-005). Lastly, in the most comprehensive neuroimaging study to-date [Kish et. al 2010] concluded that the serotonergic deficits of MDMA users remained present after “controlling for every potential confound we could address.”

3. MDMA effects on the cardiovascular system and body temperature

MDMA has also been extensively studied with respect to its effects on the cardiovascular system and body temperature. (Liechti et al) found that MDMA produced an acute and dose-dependent increase in core body temperature in humans, consistent with hyperthermia. At higher doses, Liechti found that MDMA induced hyperpyrexia, which can lead to intravascular coagulation, rhabdomyolysis, and renal or other organ failure (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5008716/#:~:text=The%20data%20show%20that%20MDMA,in%20a%20controlled%20laboratory%20setting). (Fantegrossi et. al) compared MDMA to its optical isomers, R- and S-MDMA, in their capacity to drive hyperthermia and locomotor activity. Notably, racemic MDMA and S-MDMA produced locomotion and hyperthermic effects in mice, while R-MDMA did not. Cage crowding was found to adversely affect lethality, with greater crowding resulting in a higher percentage of deaths. (https://pubmed.ncbi.nlm.nih.gov/12563544/).

(Koczor et. al) set out to explore the acute cardiotoxicity associated with MDMA, including tachycardia and arrhythmia, which are associated with cardiomyopathy. Extensive MDMA use caused DNA hypermethylation and hypomethylation that was independent of gene expression as well as differential gene expression in 558 genes, including changes to MAPK and circadian rhythm genes (Per3, CLOCK, ARNTL, and NPAS2) (https://pubmed.ncbi.nlm.nih.gov/26251327/). These cardiovascular effects were further studied by (Bexis et. al) who found that MDMA produced a prolonged increase in systolic and diastolic blood pressure in rodents, as well as isomeric contractions. The authors found that these effects were mediated by MDMA’s agonism at the alpha-adrenoreceptors (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2189797/). (Badon et. al) studied the effects of binge doses of MDMA on cardiovascular parameters in rodents. Subsequent MDMA doses were accompanied by hypotension and bradycardia. The hearts of treated rats contained foci of inflammatory infiltrates (lymphocytes and macrophages), some of which contained necrotic cells and/or disrupted cytoarchitecture. MDMA produced cardiac arrhythmias in some rats (https://pubmed.ncbi.nlm.nih.gov/12183645/). Lastly, although understudied compared to other serotonergic agents, MDMA’s agonism at the 5-HT2B receptor subtype has been evaluated in the context of valvular heart disease risk. (Setola et. al) found that MDMA produced fenfluramine-like proliferative actions on valvular interstitial cells in-vitro. The group also demonstrated that MDMA and MDA, like fenfluramine and its N-deethylated metabolite norfenfluramine, elicit prolonged mitogenic responses in human valvular interstitial cells via activation of 5-HT2B receptors. These results suggest that long-term MDMA use could lead to the development of fenfluramine-like VHD (https://pubmed.ncbi.nlm.nih.gov/12761331/)). (Baumann et al.) similarly found that MDMA had potential for inducing VHD, especially with repeat dosing (https://pubmed.ncbi.nlm.nih.gov/19897081/)

4. Additional Toxicity Information

Beyond the first-order neurological and cardiovascular risks of MDMA, research has also been conducted on the toxicity of MDMA in other organ systems as well as developmental effects. Though individual cases of hepatic injury have been reported (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9671195/), a comprehensive review of MDMA’s effects on liver function was lacking until (Cajanding), who demonstrated wide ranging toxic effects in liver (https://pubmed.ncbi.nlm.nih.gov/31462520/) across three clinical studies. Furthermore, (Shahraki et. al) demonstrated changes in the lipoprotein of rats after MDMA administration, potentially showing a hepatoxic profile for MDMA (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4331654/).
Developmental Toxicity was demonstrated by (Barenys et. al), who conducted a review of teratogenic and developmental effects of MDMA. The primary results showed that MDMA exposure during pregnancy impairs neuromotor function in infants. In rats, postnatal exposure to MDMA also caused locomotor deficits as well as impaired spatial learning that might be associated with decreased serotonin levels in the hippocampus (https://iris.uniroma1.it/retrieve/e383532c-48ce-15e8-e053-a505fe0a3de9/Barenys_Developmental_2020.pdf).
Lastly, though a relatively underexplored space, the role of mitochondrial dysfunction in MDMA’s neurotoxic effects should be noted. (Quinton and Yamamoto) found evidence implicating impaired metabolic fluxes in MDMA-induced neurotoxicity (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4497800/#R124). (Alves et. al) found that ALCAR provided some protection against these effects (https://pubmed.ncbi.nlm.nih.gov/19015003/).

5. Mitigation of risks/ REMS

Despite these risks, they should be weighed in the context of the demonstrated clinical benefit of MDMA, which is substantial. This is especially relevant given the serious consequences of non-treatment/ineffective treatment of PTSD. In the case of the MAPS protocol, the limited number of sessions (3) as well as the capped total dose (125 mg {t=0} + 62 mg {t=2hr}) can reduce the total liability of the drug from a cardiovascular and neurological standpoint. Given that existing evidence shows that accumulation of selective neurotoxic metabolites and hyperthermia are both contributors to MDMA’s direct CNS toxicity, limiting the number of sessions and lowering the temperature in the therapy setting may dampen the extent of neuroinflammation that is commonly seen in accompanying animal studies.

However, these steps may come at a cost for treatment resistant cases of PTSD or cPTSD, where the index trauma and onset of symptoms may be highly complex or may have happened in childhood. Cases like these may require more than 3 sessions before symptoms fully subside, and the patient can recover. Furthermore, a strict restriction in the number of sessions will likely limit the clinical use of MDMA outside of trauma therapy, such as for indications like mood and substance disorders where a more regular dosing protocol may be required. Therapists could consider co-administered of agents such as Vitamin-C, ALCAR, and ALA, which are often used alongside MDMA in a clandestine setting to supposedly reduce the neurotoxicity of MDMA, though there are few well-controlled studies on their efficacy.

Additionally, as previously alluded to in Section 2, non-racemic or chiral forms of MDMA with an increased fraction of the R-isomer seem to reduce or eliminate the neurotoxicity and hyperthermia of racemic or S-MDMA while preserving MDMA’s prosocial effects in rodents. Further pharmaceutical development may extend the potential reach of MDMA into other indications within mental health therapeutics. Lastly, de novo empathogens or those developed as non-neurotoxic analogues of MDMA may provide an alternative route for using entactogenic substances to treat mental illness.

6. Nuance and context for evaluating existing data

It is important to recognize the limitations of the summary and analysis of the data above. Firstly, there are inherent limitations to drawing conclusions from canine, rodent, and non-human primate studies due to ADME and other differences between animal models and humans. Furthermore, certain ‘markers’ of neurotoxicity such as reduced levels of monoamines may not establish a causal link with direct neuronal damage or may be transient, with subsequent rebounds to pre-administration levels. Furthermore, existing neurotoxicity literature on MDMA may have been biased by the political and social taboo of entactogens and hallucinogens during the 1990s, when much of this work was performed. In human studies, polydrug effects and inability to control API quality and purity may have also exaggerated the cognitive effects of MDMA neurotoxicity. Still, these markers of toxicity are still stubbornly present in newer studies, which have focused on single drug users and more robust animal neurotoxicity studies on racemic MDMA. The need for safer and more effective empathogens continues to exist.