Psychedelics - A skeptical approach to MDMA aka Ecstasy

- - In the world of psychotropic medication, the question is not just whether it works or not, but whether it works better than a placebo and whether the effect size is clinically significant and the benefits outweigh the risks.  In the case of MDMA (aka molly or ecstasy), the effect size for improving post-traumatic stress disorder symptoms reported by researchers has been very large.  Often it is found to be two to three times higher than is often found for serotonin reuptake inhibitors, which are currently the first line therapy for PTSD.  On the surface this is super exciting.  A deeper look will reveal why excitement may need to be tempered, and why an independent panel recommended that the FDA not approve this medication.

Please leave feedback at https://www.psydactic.com or send any comments to feedback@psydactic.com.

References and readings (when available) are posted at the end of each episode transcript, located at psydactic.buzzsprout.com. All opinions expressed in this podcast are exclusively those of the person speaking and should not be confused with the opinions of anyone else. We reserve the right to be wrong. Nothing in this podcast should be treated as individual medical advice.

Transcript

Welcome to psydactic, I am Dr. O’Leary and today is…  I am a psychiatrist and a fellow in child and adolescent psychiatry in the national capital region. This is a podcast about psychiatry and neuroscience.  When I first started this podcast, I imagined it as a structured learning guide for residents, but over time, I realized that resources like that are rather abundant.  What is not abundant are podcasts that ask psychiatrists like me to really think deeply about what I do and how I do it.  I want this podcast especially to challenge the listener to evaluate not just what they think, but how they think and why they think that way.

I also want to be clear that everything I say here is my own opinion and no one else's, which includes anyone I work for, work with, or whatever.  I also want to be clear that I am not anti-psychiatry or anti-anything-else.  I will likely question just about everything I have been taught, but please don’t take it personally if I am questioning something that you feel strongly about.  I question things only to try to understand what is more certain and what is more assumption.  Being uncertain doesn’t mean something is wrong, but it means that it does not have the support that people have claimed.  Today, especially, I am going to be putting on my skeptical hat and exploring psychedelics, especially MDMA, which goes by many other names, the best known and descriptive of which is ECSTAsY.

I need to define terms here, because the word skeptic will likely conjure different images and meaning in different minds.  If I were to ask a philosopher what it meant to be a skeptic, she might say, “A skeptic proposes that it is impossible to know anything, even whether or not we exist.”  That is a little too off in right-field for me, if right-field even exists.  Someone might also think of a skeptic as someone who tries to make it seem like they are smart by poking holes in everyone else’s arguments without contributing anything substantial to the conversation themselves.  This is a contrarian, better known as an a-hole.  I don’t want to invoke that image when I use the word skeptic, although I do feel like it is critically important to point out non-sequiturs when someone or some institution is trying to convince me of something.

A skeptic is not someone who effectively believes nothing and cannot be convinced of anything.  A skeptic, according to my present formulation, is someone who isn’t caught up in hype, either supportive or contrarian, and who insists on a sober discussion of the plausibility of claims and the strength of the evidence.  Skepticism that is informed by science and framed in terms of probabilities is what I am talking about.  Now that I have convinced you to stop listening, let me apply this way of thinking to the decision of an FDA advisory committee to reject approval of MDMA (aka ecstasy) in combination with a particular kind of psychotherapy for the treatment of PTSD.

Before this decision, from what I had read and heard from my colleagues, it seemed all but certain that MDMA would be approved.  In fact, my initial reaction was one of disappointment, like I had something personally invested in that decision.  I was hoping to be able to tell my PTSD patients that a new highly effective therapy was just approved, but it went beyond that.  I had attended two seminars from representatives from MAPS (the Multidiscplinary Association for Psychedelic Studies), that had framed psychedelics like MDMA and psilocybin in combination with therapy as shoo-ins for approval, with large positive effect sizes and virtually no significantly worrisome side-effects.  I feel confident in saying that the presentations I saw from MAPS were decidedly not skeptical in their approach.  They did contain some preliminary data showing very large effect sizes in uncontrolled trials and some apologetics; for example, explaining that they believed MDMA was much safer than the literature suggested because prior research into its toxicity had used samples that contained regular methamphetamine instead of 3,4-Methyl​enedioxy​methamphetamine (aka MDMA).  This presentation also contained many anecdotal accounts of massive improvements experienced by some early study participants.

Lycos therapeutics, which is merely a rebrand of MAPS pharmaceutical division, has consistently expressed high levels of confidence that MDMA would be approved, and have been able to secure over 100 million in funding as a result.  The FDA panel, however, was not convinced.  The panel appeared to disagree both with the proposed effectiveness of MDMA and with claims that the risks of using it are very low.  Proponents of MDMA are framing this as a major set-back and while they are careful not to overly criticize the decision, they make it clear that they feel like the decision was a particularly biased rendering of the facts.  The FDA panel was concerned that Lycos was presenting them a particularly biased rendering of the facts.

I can imagine a room full of people who take sides in this issue arguing incessantly over the minute details of every data point.  What I want to do is take a broader approach and discuss what a skeptical approach would look like and, to put the biases that I am aware I have front and center, argue that a skeptical approach is the best way to evaluate this situation.

When beginning to think about a question such as, “Can MDMA help relieve the symptoms of PTSD?,” there is an even more fundamental question.  “Is it plausible that a pharmaceutical such as MDMA could treat PTSD?”  These may seem like the same question, the second one explicitly asks us to look through the lens of plausibility.  Plausibility not mere possibility.  It is possible that the moon will crash into the earth next week, but given the fact that we have a huge amount of data predicting something else, it is not plausible.  I wouldn’t waste my limited brain resources anticipating the moon crashing into the earth, unless I am writing a novel.

Another idea related to plausibility is something called face validity.  Face validity is when we propose a construct (which is merely an explanation for something we have not or cannot directly measure) that appears to make sense on its face.  If our construct is plausible, then it has face validity.  However, science and reason have demonstrated over and over again that plausibility or face validity is not enough to draw conclusions.  However, it is enough to start asking questions.

MDMA is a psychotropic medication that has been used as a party drug because, like its cousin methamphetamine, it works as a stimulant, increasing the users perceived motivation and energy.  However, it is not as potent of a stimulant as meth, and it has other distinct properties including making the user feel a general sense of well-being and increasing prosocial behaviors.  It has been said that MDMA increases a person’s empathy and feelings of connectedness to others. All of this sounds very nice on the surface, but there is also a dark side to feeling a profound sense of connectedness with others.

First of all this increase in empathy is likely just illusory.  The user does not necessarily have any actual deeper insight into what others are feeling.  They just assume that they do.  Even though the user feels as if they are able to connect more with others, this may be a false positive connection, an illusion brought on by the drug.  This sense of connectedness with others is not necessarily a good thing, because it lowers a person's normal defenses against undue influence by others and may result in a person being taken advantage of, coaxed into a financial commitment or sexual experience that they would not have otherwise agreed to because they feel such a profound sense of trust.

MDMA affects this sense of profound connectedness by making our Raphe nuclei (those areas of the brain that produce and ship serotonin) to release far more serotonin than they normally would.  MDMA does this by resulting in a reverse of flow of the serotonin transporter.  Normally, serotonin in the synapse would be transported back into the neuron that dumped it and then packaged into vesicles.  MDMA inhibits the packaging of serotonin and also makes the transporter function backwards.  This results in a lot more serotonin floating freely in the cell with serotonin being pumped into the synapse instead of from the synapse into the cell body.  It also increases the levels of norepinephrine (aka adrenaline) and to a lesser extent, dopamine, by blocking their transporters and inhibiting the VMAT2 transporter which contributes to its stimulating properties.  Overall, the thing that makes MDMA or ecstasy special is its effects on serotonin.

So it is plausible that MDMA could have effects on conditions like PTSD that often result in negative cognitions about the world, feeling isolated, trapped, or targetted, or in generally not feeling safe.  On its face, MDMA appears to be a good target for exploration as a psychotropic medication.  But plausibility is not enough to justify belief.

The next thing to consider is the prior probability that something will be found to be effective, that MDMA would be effective.  Everything we do to demonstrate efficacy is some kind of statistical test, so the prior probability that something will be effective is important to include in the calculation of our confidence in that thing.  The vast majority of agents under investigation for treating psychiatric disorders are never found to be substantially effective.  Something around 90% of all drugs that make it to serious investigation fail in further clinical trials, whether that is due to lack of efficacy, toxic side effects, or pharmacokinetic effects (which means that the way the drug is processed by the body results in an insurmountable problem).  For MDMA until the last decade we could add in the fact that it was determined to be a drug of abuse with no medical use and made a schedule 1 substance, so research into the drug was lacking until recently when restrictions on research into psychedelics were relaxed.

We could simply apply the 90% fail rule to MDMA and say that it has a 10% prior probability of being found to be effective.  However, I feel like that number is a little exaggerated.  Drugs under investigation may be found efficacious, but have unfavorable pharmacokinetic effects or have toxic side effects.  If we subtract the probability of poor pharmacokinetic effects from MDMA (because we know that it has favorable pharmacokinetic properties), then the prior probability that it will be approved can be estimated higher: at about 25%.

So here I have proposed that clear thinking about whether MDMA will be found to be effective is about 1 out of 4.  This is not the rosy picture that has been promulgated by the media or by groups researching MDMA.  For example,  on May 9, 2021, The New York Times published an article with the title “The Psychedelic Revolution Is Coming. Psychiatry May Never Be the Same.”  This is not a skeptical rendering.  It is reflective of the excitement that proponents of psychedelics have promulgated.  This has been typical of popular media accounts.  They basically assume these drugs will change everything.  Every time a new class of drugs breaks onto the scene, a similar excitement occurs.  Of course, the societies that advocate for the use of psychedelics also have framed their substances as miraculous.  I’m not saying this because of media accounts.  I have heard it from the mouths of psychedelics researchers.  The question is not if, but when psychedelics will enact this revolution.

When it comes to raising money for a cause, a lack of doubt is necessary.  When it comes to making sober choices, an appropriate amount of doubt is wisdom.

Something else I have not yet defined is what it means to be QUOTE effective.  In the world of psychotropic medication, the question is not just whether it works or not, but whether it works better than a placebo and whether the effect size is clinically significant and the benefits outweigh the risks.  In the case of MDMA, the effect size has been reported by researchers in many studies to be large.  Often it is found to be around 1 or more.  This is 2 to three times higher than is often found for SSRIs, which are currently the first line therapy for PTSD.  On the surface this is super exciting.  A deeper look will reveal why excitement may need to be tempered.

Here I need to introduce a concept called the Likelihood.  A likelihood is an estimate of how strongly data supports a hypothesis.  Generally, when variability in the data is low and effect sizes are large, then the likelihood will be large too.  The data appears to strongly support an effect.  In the case of MDMA assisted therapy for PTSD, I am going to estimate here, without doing any actual math, that the likelihood, given the data, that MDMA assisted therapy is effective is very high because the effect size is so large.  It appears that everyone agrees that the likelihood considering only the current data that MDMA is effective is very high.

Why then did the FDA panel reject it?  Only 2 out of the 11 members of the panel agreed that it had been shown to be effective.  Only 1 out of 11 that the risks were outweighed by the benefits.  It is unlikely that they wrote out Bayes Theorem and plugged in the numbers.  Instead, they made mental estimates of the bias in the studies and concluded that it canceled out the likelihood of the large effect.  The posterior probability they estimated was very low that MDMA was effective, given the biases in the data.  Estimates of probabilities can be conceptualized as levels of confidence, and the FDA panel appears to have less than 10% confidence that the risk of taking MDMA is greater than the benefits and less than 20% confidence that it is effective.  This level of confidence is even less than the prior probability that I proposed that MDMA would be found to be effective in the first place.

In addition to prior probabilities and likelihoods, when calculating posterior probabilities (or what I called levels of confidence earlier), a mathematical model could incorporate other estimates of bias.   It appears that the committee must have heard evidence that caused them to incorporate substantial estimates of bias into their reasoning.  What were those biases?

One of the biases was the unblinding effect.  The second bias was something I am going to call a priming effect.

Ideally studies should be blinded, meaning that the person being tested and the person testing them have no idea whether they received the drug or not.  For some medications, this is impossible.  MDMA, at the doses given in the study, is known for gifting its user a sense of euphoria among other things, such as an increase in heart rate, so a patient, their therapist and prescriber are likely to know from the response whether they received the active treatment.  When a study is unblinded, we expect to have a higher effect size than if it is not.  That is because of the placebo effect, which is complicated, but it predicts that a patient who knows they received a treatment is more likely to report feeling better than one who knows they did not.  This especially complicates psychotropic medication studies because all we have to measure effects are the self-reports of the patient and their behaviors.

There was something else that I called the priming bias, and conceptually it is related to unblinding because it gives the study participant information about a treatment that encourages them to answer in a certain way.  For example, someone could tell a patient that their answers are extremely important and that how they answer will affect people’s ability to get this treatment in the future.  This sets up a power differential, where the participant is asked to wield power over others and can result in the participant overstating a benefit or reporting a benefit where none occurred in order to make sure that future patients will be able to receive this treatment.  If you combine a state like that with another statement to the effect of “This treatment has the potential to help millions of people,” or “This treatment promises to change the way we treat PTSD and help millions of people,” then the priming bias is amplified even more.  Apparently patients participating in Lycos clinical trials report that statements like this were part of the protocol that they experienced.

For a study to minimize biases, a participant should know as little about the potential treatment as is ethical.  Because the placebo effect relies in part on a participant's expectation of benefit, anything that makes them believe that this treatment is particularly special, ground-breaking, or miraculous can increase the placebo component of a treatment.  The priming effect is also important with treatments that can be blinded.  The difficulty for fully blinded studies with a large placebo effect is that there will be smaller differences between the treatment and placebo group, This will make the treatment seem less effective than it is.  Priming can reduce the effect size.  If the treatment group cannot be blinded then priming bias can make the active treatment group report much higher benefit that they otherwise would have.  It will increase the effect size.

We certainly have to consent participants to the risks versus benefits of any treatment, but the less they know, generally the less biased their reports will be.  Participants in Lycos studies reported that they had been primed to respond in a certain way, so the independent committee were concerned that the effect sizes measured were heavily biased by the protocol.

Based on the large potential for bias in the reported numbers, the committee had low confidence that these effects are real.

Another major concern of the independent committee was safety.  They criticized Lycos for not providing more data concerning the long-term safety of using MDMA.  Will the treatment predispose participants to unhealthy use of this or other substances?  After all, ecstasy is not that difficult to find on the street.

Another safety issue is that ecstasy places patients in a particularly vulnerable condition.  Because it can create a sense of deep connection with others, those who take it can be easily taken advantage of whether it be financial or sexual or to gain personal information from them.  In fact, there is an investigation into Lycos due to allegations that at least one of the therapists took sexual advantage of at least one participant.

There was also a dearth of long term follow-up data because few participants were followed for more than a few months.  We don’t have strong evidence that the effects are enduring.  In addition to this, the FDA does not approve psychotherapies, and so they could really only focus on the drug portion of the treatment.

To summarize, Lycos had a tough hill to climb to get FDA approval.  MAPS has lobbied aggressively for psychedelics as potential therapeutic agents.  The data seems to show that psychotherapy assisted MDMA treatment can have a large reduction in PTSD symptoms.  However, due to research practices that are known to inflate effect sizes, lack of abundant safety data, little long-term follow-up and allegations against one of the therapists, the committee recommended against approval both on grounds of efficacy and on grounds of risk versus benefit.

There is one pattern that has seared itself into my brain and in some ways has left me a little cynical.  Actually, maybe it is not cynicism, maybe it is realism.  Anyway, there is no lack of historical claims that a drug or a device or a procedure or a new therapy is going to revolutionize treatment of just about everything.  Almost never is there an actual revolution.  There are frequently step-wise improvements, such a finding agents that are safer or have fewer side effects, but none of the claims of substance induced psychic healing proposed in the past has ever lived up to the hype.  I am beginning to doubt that MDMA will be the first.


https://www.asbmb.org/asbmb-today/opinions/031222/90-of-drugs-fail-clinical-trials#:~:text=It%20takes%2010%20to%2015,candidates%20in%20clinical%20trials%20fail. 

https://www.nytimes.com/2021/05/09/health/psychedelics-mdma-psilocybin-molly-mental-health.html

https://www.vox.com/future-perfect/353752/mdma-fda-approval-hearing-lykos-blinding-misconduct-allegations 

https://www.nytimes.com/2024/06/04/health/fda-mdma-therapy-ptsd.html?searchResultPosition=4