Entactogens

An SSRI blocks the serotonin transporter so that serotonin already in the synapse stays there longer. An entactogen does something more forceful: it is a substrate for the transporter rather than a blocker, so it is carried into the neuron and, once inside, reverses the pump. The transporter runs backwards and empties the cell's serotonin into the synapse at once ^[4]. MDMA is a substrate at all three monoamine transporters, with the strongest action at the serotonin transporter and meaningful action at the norepinephrine and dopamine transporters, which is where the stimulant component comes from ^[5].

The evidence that serotonin release is the cause rather than a correlate is a single clean human experiment. Pretreating volunteers with the SSRI citalopram, which occupies the transporter and denies MDMA its route in, attenuated most of MDMA's psychological effects ^[6]. Take away the entry point and the state does not appear. This is also the reason people already taking an SSRI often report that MDMA does much less for them.

Oxytocin is the second half of the popular explanation, and it deserves a more careful statement than it usually gets. MDMA does raise plasma oxytocin in humans, and the rise correlates with self-reported prosocial feeling ^[7]. But when intranasal oxytocin was given on its own in a controlled comparison against MDMA, it did not reproduce MDMA's social and emotional effects ^[8]. Oxytocin is part of the picture and is not sufficient on its own; the serotonin release is doing more of the work.

Two further pieces complete the mechanism. Norepinephrine release drives the cardiovascular effects, the raised temperature and much of the physical arousal ^[4]. And the aftermath is built into the mechanism: emptying the vesicles faster than they can be refilled leaves serotonin transiently depleted, which is the pharmacological basis of the low mood some people report in the following days ^[2][4]. That depletion is a supply problem and is not the same claim as damage; the difference is the subject of the neurotoxicity section below.

The useful way to place any compound in this corner is by its release ratio: how much serotonin it releases relative to dopamine. That single number predicts character better than any structural feature.

Push the ratio toward serotonin and you get the entactogen profile: warmth, emotional openness, low compulsivity, and comparatively little drive to redose. Push it toward dopamine and the compound becomes a stimulant with a warm edge, with the compulsive redosing and the crash that stimulants bring. The substituted cathinones sit on that side; mephedrone is often described as entactogen-like and behaves in use much more like a stimulant, while MDPV and a-PVP are transporter blockers with essentially no serotonin component and belong to a different discussion entirely ^[10].

This is also why the class boundary with the psychedelics is real rather than cultural. Classic psychedelics work through 5-HT2A receptor agonism; entactogens work through transporter reversal, which is a different molecular event with a different signature. Some compounds do both, and MDA is the standard example: it releases serotonin like MDMA and carries genuine psychedelic character on top ^[4]. A compound that does both is not evidence that the categories are the same; it is evidence that a molecule can carry two mechanisms.

One historical footnote worth keeping. MDAI was made specifically as a research tool: a serotonin releaser with very little dopamine release, built to test whether the entactogen effect could be separated from stimulation and from the neurotoxic profile ^[10]. It is frequently sold as a safe MDMA substitute. Its actual human safety record is a handful of case reports, and "designed to be less toxic in rats" is not a safety claim about a person.

The family branches off MDMA in three directions: the ring-substituted amphetamines it came from, the aminoindanes built as research tools, and the benzofurans that replaced the methylenedioxy ring with a fused furan.

The benzofurans deserve particular attention because they are the largest recent wave and they carry a specific hazard. Pharmacological profiling found the aminopropylbenzofurans to be monoamine transporter substrates with broadly MDMA-like release profiles, and also potent agonists at the 5-HT2B receptor ^[9]. Chronic 5-HT2B agonism is the established cause of the valvular heart disease produced by fenfluramine and pergolide ^[22], so a compound that combines MDMA-like effects with strong 5-HT2B activity carries a theoretical risk with repeated use that MDMA carries only weakly. No human case has demonstrated it for these compounds; the mechanism is well enough characterised that the inference deserves stating.

The other thing every table in this area understates is that duration varies enormously across the family, and the longer members are where the errors happen. A compound that takes two hours to arrive invites a second dose before the first has fully landed.

The same fear-lowering, trust-raising quality that made MDMA a social drug is the reason it was taken into clinical trials for post-traumatic stress disorder, where the obstacle to treatment is that revisiting the trauma is intolerable. In the protocol the drug is given during a small number of long, supported sessions with two therapists present, embedded in a course of non-drug therapy.

Two phase 3 trials have reported. The first randomised 90 participants with severe PTSD and found a large advantage over placebo on the clinician-administered PTSD scale, with a majority of the treated group no longer meeting diagnostic criteria at the end of the study ^[11]. The second, in 104 participants with moderate to severe PTSD, reproduced the result ^[12]. On their own numbers these are among the strongest results reported for any PTSD treatment.

They did not lead to approval. In 2024 the US regulator declined the application and asked for a further trial. The published criticisms are worth understanding because they are methodological rather than about the numbers: functional unblinding, since almost everyone can tell whether they received MDMA and expectancy in this field is strong ^[24]; the inseparability of the drug from the many hours of therapy around it, which no trial in the programme was designed to disentangle; and concerns about how adverse events were collected. On that last point, a systematic review of psychedelic and MDMA trials found that harms were recorded unsystematically across most of the literature and that long-term effects were rarely tracked at all ^[13].

The honest summary is that MDMA-assisted therapy has produced real, large effects in two controlled trials and has not yet cleared the bar that would make those effects reliable enough to license. That is a very different statement from either "it works" or "it failed", and both simpler versions are in circulation.

This is the part of the topic most distorted in both directions, and it is worth walking through carefully because the history contains a genuine scientific scandal.

In animals, high or repeated doses of MDMA produce long-lasting reductions in serotonergic markers: less serotonin, fewer serotonin transporter sites, and loss of serotonergic axon terminals in cortex. That finding is robust and reproducible ^[4]. The arguments are about whether it transfers to human patterns of use, since the animal work generally uses higher doses, different schedules and species that metabolise the drug differently.

In humans the evidence is imaging plus cognition, and it is genuinely mixed. Early PET work reported reduced serotonin transporter binding in users ^[14]. A later, more careful PET study using a better tracer alongside structural imaging found decreased cortical serotonin transporter binding but no subcortical change and no structural abnormality, and its authors emphasised how heavily polydrug use confounds every study in this area ^[15]. A cognitive study that went to unusual lengths to recruit long-term users with minimal exposure to other drugs, and to control for sleep, socioeconomic status and prior IQ, found little residual impairment ^[16]. Those three results do not contradict each other so much as they narrow the claim: something measurable happens to cortical serotonin markers, and the functional consequence in people who use occasionally is smaller than the early literature implied.

The scandal. In 2002 a widely publicised paper reported that a common recreational dose regimen produced severe dopaminergic neurotoxicity in primates, with two of ten animals dying ^[17]. It was retracted the following year: the vials had been mislabelled and the animals had received methamphetamine, not MDMA ^[18]. The retraction is cited here alongside the original because the episode is the clearest available lesson in how a single dramatic result can shape a field's public understanding for years after it stops being true. Reference 17 in the list below is the retracted paper and is marked as such.

What can be said with reasonable confidence is in the table. What cannot be said is the more useful half: nobody knows the threshold, nobody has a prospective human study, and the confound of hyperthermia is unresolved, since the animal damage is strongly temperature dependent and hot environments are exactly where recreational use happens.

The acute hazards are better characterised than the chronic ones, and they are mostly problems of physiology rather than psychology.

Hyperthermia is the classic and the most dangerous. MDMA impairs thermoregulation while releasing norepinephrine, and in a warm, crowded, physically active setting core temperature can rise far enough to trigger rhabdomyolysis, disseminated intravascular coagulation, liver failure and death. The full syndrome is a recognised medical emergency with an established management pathway ^[19]. It is dose related but not only dose related; the environment does much of the work.

Hyponatraemia is the mirror-image harm and kills people who did everything they were told. MDMA promotes vasopressin release, so water is retained; drinking large volumes on top of that dilutes blood sodium and can cause cerebral oedema and seizures. Young women appear to be at higher risk ^[19]. Both too little water and too much water are dangerous, which is why simple hydration advice is a poor substitute for understanding the mechanism.

Serotonin syndrome. Combining an entactogen with an MAO inhibitor is the hardest rule in this area; blocking the enzyme that clears the released serotonin while forcing a large release is the exact recipe, and the combination has been fatal ^[19]. Combinations with other strongly serotonergic drugs carry the same risk to a lesser degree; the interactions and stacks tool flags them. See the serotonergic system.

Cardiovascular load. Pooled data from controlled administration in healthy volunteers shows acute increases in blood pressure and heart rate that are transient and, in a screened laboratory population, usually uneventful ^[20]. That is a statement about screened volunteers at known doses in a cool room; it does not extend to people with undiagnosed cardiac disease, in the heat, alongside stimulants.

What is in the bag is frequently not what the label says. Hair testing of self-reported ecstasy users found a substantial proportion positive for synthetic cathinones that they did not know they had taken ^[23]. The entactogen market is one of the most heavily substituted, and the substitutes are usually more stimulant, more compulsive and less well characterised than the thing they replace. See research chemicals for how that market works. This page is educational and is not medical advice.