select search filters
roundups & rapid reactions
before the headlines
Fiona fox's blog

expert reaction to phase 2 trial comparing psilocybin and escitalopram for depression

A study published in the New England Journal of Medicine looks at phase 2 trail comparing psilocybin, the active compound in magic mushrooms, and escitalopram, a leading antidepressant, for depression.

This Roundup accompanied an SMC Briefing.


Prof Kevin McConway, Emeritus Professor of Applied Statistics, The Open University, said:

“There’s been a lot of interest over the years in the potential for using psychedelic drugs in the treatment of serious mental illness, and this trial is a real advance in the search for evidence as to whether such treatments can be safe and efficacious. The trial compares the effects of the psychedelic substance psilocybin with the effects of a conventional antidepressant drug, escitalopram, on people with depression. The trial was not large, involving only 59 patients, and in several ways the results were rather inconclusive. But it’s important to understand that this trial was probably not meant to be definitive about whether psilocybin can have a role in treating patients with depression. It is a Phase 2 trial. Such trials are intended to provide some evidence on whether a treatment is efficacious, that is, whether it can influence a relevant outcome of the disease, and also to provide more data on safety and (in many cases, including this trial) on how the treatment works – but they are also usually intended to provide information that will see whether it is worth continuing to a much more definitive Phase 3 trial, which would typically involve considerably more patients and attempt to tie down whether and how the new treatment would be useful in clinical practice, as opposed to in an experimental situation. Very often the experience of Phase 2 trials can help in designing Phase 3 trials in the most appropriate way.

“The report of this trial does explain that the patients were recruited by a quite complicated process, so that it’s not clear how far they might be typical of patients with serious depression, but that’s not necessarily a huge problem for a phase 2 trial where an important aim is to see whether the treatment has an effect in patients. Also, because the end of the trial ran into the period of the current pandemic, a few difficulties arose in obtaining full data on a few patients. But these things happen at various levels on all trials, and they do not concern me hugely. What makes the interpretation of the results most problematic is that the evidence turned out not to be entirely clear. People running clinical trials have to declare a lot in advance about how the trial will be run, what kind of patients will be included, what outcomes are going to be measured, and how the resulting data will be analysed. This process is called registration. There are very good reasons for this. If it were not necessary to register trials before they begin, researchers could record a huge list of different things that could be taken to be measures of the efficacy of their new treatment, and then, at the end, present results only for the measure or measures on which the treatment seemed to have ‘worked’, and it could be very difficult to tell what was really going on. The trouble is that, if you measure enough different things, it becomes pretty likely that the new treatment will show up as successful on at least one of the measures simply by chance, even if in fact the new treatment really has no effect at all.

“These researchers did indeed register their trial in advance, though one thing that I found confusing is that there seem to be two different registrations and they don’t both say the same thing in terms of outcomes. One, on the standard American site recording registrations, and mentioned in the research paper (presumably because the journal is American), is at . The other one that I found is on a website run by the EU clinical trials register, at . In the American registration, the researchers declare outcomes from fMRI brain imaging as the primary outcome that will be recorded, and the results on a standard scale for measuring depressive symptoms, the QIDS-SR16, as the secondary outcome. In the EU registration, the QIDS-SR16 is listed as the primary outcome, and the fMRI measures and a long list of other clinical and psychological measures are also given as secondary outcomes. The difference between the two registrations does seem slightly strange, though these things do happen, and in particular it is odd that the NEJM paper lists only the American registration but gives no results on the fMRI imaging, although the American registration says that is the primary outcome measure. However, I don’t believe this represents any sharp practice on the part of the researchers – the QIDS-SR16 results are indeed registered as the primary outcome on the EU registration, and the other secondary outcomes listed in the research paper are indeed also registered in the EU registration. So no real problem, just a strangeness.

“On the primary registered outcome (the change in QIDS-SR16), the researchers report that the difference between the two treatment groups (psilocybin and escitalopram) was not statistically significant. It’s very important to be clear on what that actually means. On average, the patients in the trial who were assigned to take psilocybin did have a better outcome on this measure than patients who were assigned to the conventional antidepressant drug, but the difference between the two groups was small enough that it might possibly be due solely to chance. For example, the randomisation process might just have happened to assign the patients, who were going to do better anyway whichever treatment they got, to the psilocybin group. The lack of statistical significance means that we can’t rule this possibility out. But, just because it can’t be ruled out, that doesn’t mean that it’s definitely the case. It could also be that there’s a real difference between the two treatments, on this measure, and the trial simply happened not to be able to confirm that real difference. So, on this pre-registered primary outcome, the results are inconclusive. But you also have to bear in mind that the efficacy of psilocybin was being compared with the efficacy of an established drug, escitalopram, which might not be perfect (no antidepressant drug is) but which has been established to be better for serious depression than a placebo (an inactive substance) in previous clinical trials. So what this primary outcome comparison, in the new trial, means is that psilocybin hasn’t been definitely shown to be better than a drug which is known, in a certain sense, to work in serious depression. Psilocybin also has not been shown to be less efficacious than the established drug, and in fact on average, across the patients in the trial, it was more efficacious, but not sufficiently more efficacious to rule out the possibility that the efficacy is the same as that of the established drug. A larger, phase 3, trial, if one happens, could clarify the position.

“The researchers also report that the secondary outcome measures also generally favoured psilocybin over escitalopram. The details are mainly in the supplementary appendix to the main research paper, and in many cases the difference between the results on the two treatments for these alternative measures of efficacy are large enough to be statistically significant. However, it would be wrong to favour these results over the pre-registered primary outcome, where the difference was not statistically significant. These requirements of registration are there for a reason, to prevent researchers choosing the outcome measure they want to concentrate on after they have seen the data, and thus potentially distorting what went on by picking the comparison that is most favourable to their ideas and possible prejudices or biases about what might work. These researchers, properly, did not do that. The researchers also, rightly, report that these other outcome comparisons were not adjusted to take account of the fact that a lot of different statistical comparison were made. Such adjustments are done for the following reason. The more comparisons that are made, the greater the chance that the ones where the difference just happens to be greatest will produce a statistically significant result just by chance, even if really there’s no difference between the treatments being tested. The adjustment for multiple comparisons takes care of this possibility, at the cost of making it harder for any one of the comparisons to come out as statistically significant if in fact there really is a difference. But the researchers in this new study did not fall into that trap of not adjusting for multiple comparisons and, as a result, claiming too much – they behaved properly, and did not make an adjustment for multiple comparisons because they had not declared in advance that they would do so, and so they are quite restrained in the research paper in what they say about these secondary results. The fact they now have results on a number of different psychological and clinical measures, not just the primary outcome, could well help them to design any phase 3 trial more appropriately. However, the lack of adjustment does mean that again, we can’t take these secondary measures to show that psilocybin is more efficacious than the comparator drug escitalopram.

“I’ll finish by mentioning one interesting aspect of this publication process, which is that the lead author, Robin Carhart-Harris, has been tweeting extensively about the trial for almost a week now, in a thread starting at He does not break the embargo, in that he does not reveal the detailed results, but he does give useful extra information about some of the details, and I think it’s pretty obvious, from what he writes about both the primary outcome measure and all the secondary ones, which direction the results go in. (I think it’s also reasonably clear how he would interpret the results himself.) The one aspect of his Twitter thread that I really don’t like is where he describes the analysis and presentation of the results in the NEJM paper as being particularly conservative, because they follow established conventions of giving primacy to the pre-registered primary outcome measure, and of taking great care in interpreting secondary outcome measures when there is a large number of them. Those conventions exist for good reasons, as I’ve explained (and as indeed he also explains). But when he writes, in relation to the comparisons of the secondary measures, that “Values are just values, free of any narrative”, he’s simply wrong, in my view. The values are numbers that emerged from particular choices, made by him and the other researchers on this trial, on how to design and register a clinical trial, and on further choices about how to analyse the data. Looking at the numbers in isolation as if the numbers can stand for themselves is seriously misleading.”


Prof David Owens, Professor Emeritus of Clinical Psychiatry, University of Edinburgh, said:

“The accompanying commentary by Jeff Liberman is considered and, in my view, spot-on. This is a small, exploratory study with numbers too small to analyse fully. The population is not recruited randomly from, for example consecutive admissions or presentations, and screening of volunteers was by telephone, not face-to-face. One might say this is an ‘interested’ population, willing to go for novel approaches and with no placebo group, the extent of the placebo response cannot be assessed.

“I must emphasise that the results, in that they were not negative and, apparently, no deterioration in mental state was observed (a substantial theoretical risk), are interesting but, as Lieberman points out, do not necessarily point us in new directions. They give a little fuel to continue down the road but no more. I think more of the same will be required before even Phase 3 can be contemplated. Phase 2 studies are about first-in-human effects – pharmacokinetics, tolerability and, especially broad signs of efficacy. They are a necessary prerequisite for progressing to the large, multicentre studies that comprise the data for regulatory submission. The latter are very expensive to undertake so companies need to make sure there are no hitherto unrecognised problems with the product.

“In that psychiatry has squeezed to death the old fashioned aetiological and treatment theories for mood disorders, I fully support industry (and others) pursuing different routes, such as traditionally utilised compounds such as magic mushrooms. But this is not the study to get excited about in my view. If it does, however, impress the markets and the group gets funding to pursue Phase 3 development, something positive coming out of that might well be front page news!”


Prof Guy Goodwin, Emeritus professor of Psychiatry, University of Oxford, said:

“Remedicalising psilocybin and related drugs is the most interesting project in contemporary psychiatry. The present study is not a quantum leap: it is under-powered and does not prove that psilocybin is a better treatment than standard treatment with escitalopram for major depression. However, it offers tantalizing clues that it may be. In particular, the differential and beneficial effects on the well-being of patients in this study are interesting. It underlines the broader point that research in depression has been too driven by ratings of particular symptoms rather than the return of positive mood and patient well-being. Academia, pharma and the regulators have a responsibility to review the criteria whereby we judge advances in treatment for mood disorder.”


Prof Anthony Cleare, Consultant Psychiatrist; Professor of Psychopharmacology and Affective Disorders, Institute of Psychiatry Psychology & Neuroscience, King’s College London (IoPPN), said:

“Rigorous clinical studies on the efficacy of psychedelic treatments for depression are tricky to do, with many potential pitfalls, some of which are unavoidable, such as patients knowing which treatment they have received given how profound the psychedelic effects are. Notwithstanding these difficulties, this study provides some of the most powerful evidence to date that psychedelics may have a role to play in the treatment of depression. The lack of serious side effects is also extremely encouraging, and it does seem that with careful administration and monitoring these treatments can be given safely.

“We need much more data before these treatments could be considered ready for use outside of carefully controlled research studies. In particular, we do not yet know which types of patients, and which types of depression, may be best suited for psychedelic treatments.  We also do not know how long benefits last and whether or how often treatments may need to be repeated.

“Psychedelic treatments are not a panacea, and will not replace existing treatments for depression, but this study is a further step in the pathway that may lead to them becoming one of the options for patients in the future”


Dr James Rucker, Lead for the Psychedelic Trials Group @ King’s College London, NIHR Clinician Scientist Fellow,, and Consultant Psychiatrist, The Institute of Psychiatry, Psychology and Neuroscience (IoPPN), King’s College London, said:

Is this good quality research?  Are the conclusions backed up by solid data?

“It is good quality research. This is a single centre randomised controlled trial in which the authors sought to investigate the safety and efficacy of psilocybin (the active ingredient of so-called ‘magic’ mushrooms and a Schedule 1 restricted drug) when compared with escitalopram (a widely available antidepressant). Because these treatments are delivered in different ways, the design of the trial is necessarily unusual. However it is basically good quality research and a valid and interesting addition to this nascent field of research. 

“The data seems to be solid. The conclusion is that the primary outcome (the participant rated Quick Inventory of Depressive Symptoms) did not differ significantly between conditions at 6 weeks follow up. That is to say, at 6 weeks there was no statistically significant difference in participant-reported depression between the group that received psilocybin and the group that received escitalopram. It is possible that the study was not adequately powered to detect a difference, or that this represents a true finding that the treatments are equivalent in terms of patient reported outcomes, when delivered in this context. This is important, because other trials have indicated very large effect sizes for psilocybin therapy and the interpretation of this by some is that psilocybin will be more effective than established treatments for depression.

“This trial reported a number of secondary outcomes. These included analyses that interpreted the primary outcome measure (participant reported depression symptoms) in a different way, participant rated measures of various aspects of psychological and social functioning not directly related to depression, as well as other outcome measures that were rated by the trial team themselves. Generally, these secondary measures did report a difference that favoured psilocybin over escitalopram. However, the conclusions that can be drawn here are limited as the more measurements are taken, the higher the threshold for statistical significance becomes. In addition, many of these measures were taken by the study team themselves, which brings in to play the possibility of unconscious bias in how the measures were rated. The authors, correctly, present these results as comparisons of the means between both groups, as well as 95% confidence intervals, without much further interpretation. These secondary analyses are interesting and worthy of further follow up. 

“As the authors point out, no clinically significant conclusions can be drawn at this stage from this research. However it does make a good case for further investigation of psilocybin’s potential safety and efficacy in this patient group. These studies are now underway.

How does this work fit with the existing evidence?

“This work provides additional evidence to other early phase clinical trials suggesting that psilocybin, when given with appropriate psychological and medical support, has an antidepressant effect in participants with simple major depressive disorder. However, it also confirms that escitalopram, when given in a similar context, also has this effect. It is already known that escitalopram is an effective antidepressant. It isn’t clear from the design what the effect of the psychological support was to the treatment effect, because no placebo group was included (in fact, this had been intended, but was not possible due to the cost implications).

“The authors conducted an earlier pilot trial in 2015-6 in which they focussed on participants with so-called ’treatment resistant’ major depressive disorder. Here, the authors have chosen to focus on a different patient group in this trial (simple (or non-treatment resistant) major depressive disorder). The rationale for this change in clinical focus is not clear in the manuscript. 

“The work substantially fits with other clinical trials suggesting that when psilocybin is delivered in a medically controlled and psychologically supportive setting, that it is relatively safe. There were no serious adverse events in this trial (in either treatment condition). The overall proportion of participants reporting an adverse effect of some nature was similar between the groups. This is reassuring in terms of the ongoing development of psilocybin as a potential treatment. 

Have the authors accounted for confounders?  Are there important limitations to be aware of?

“There are several limitations here. 

“There was no placebo control group in this trial. This would have been useful because it would have given us an idea of the extent to which the psychological support on offer as part of this trial contributed to the overall treatment effect. 

“The sample was likely to be self-selecting in favour of the psilocybin intervention. The authors state that ‘most’ participants self-referred to the trial. ‘Most’ is not further defined. News about trials like this often spreads organically, meaning that those with pre-existing interests in psilocybin as a treatment are more likely to know about it, and volunteer, than those who are agnostic or against the treatment. Of course, this would happen anyway and to an extent it is hard to account for. Patients must be free to make informed choices about treatment in situations like this. In mitigation of this, the authors did collect collateral confirmation of diagnosis from primary care and did thoroughly assess each participant prior to inclusion. 

“The sample was predominantly men (66%), whereas depression predominates in women. 39% of participants who were deemed eligible were not taking a medical treatment for depression, despite being moderately clinically depressed. It’s unclear to what extent these factors may make the sample unrepresentative. It is known that men are more likely to volunteer for drug trials. It is also ethically problematic in studies of experimental treatments to recruit truly representative samples (and major depressive disorder is known to be highly heterogeneous in nature). Thus the results of this study may not generalise to the wider MDD community.

“This trial used a form of randomisation in which blocks of eight participants were randomised 1:1 to receive psilocybin or escitalopram. Whilst this approach is valid given the context, it introduces the possibility that researchers are able to guess the allocation of participants in the block from their observations of previous participants. 

“As with many trials of this nature, blinding is a problem. The authors did not take account of the high probability of subjects being able to tell which condition they were allocated to. That is to say, they did not ask participants which condition they thought they were allocated to. It is known that the subjective effects of 25mg of psilocybin are strong enough to be identifiable by most people. Similarly, the side effects of escitalopram are well known, particularly at the higher dose the authors used. If the sample were already likely to be in favour of psilocybin treatment over escitalopram, this will likely introduce a further form of bias that favours psilocybin.

What are the implications in the real world?  Is there any overspeculation?  

“The implications for the real world are currently limited. This trial suggests that we should proceed to a more exhaustive phase of clinical research, which is a multi-centre (preferably international) clinical trial. 

“From the perspective of medicine’s regulators it is the non-superiority of psilocybin over escitalopram that is likely to be of interest, because the delivery of licensed psilocybin therapy is likely to be significantly more expensive than the delivery of escitalopram therapy. This may suggest that if psilocybin therapy is proven to be a safe and effective therapy for depression, that its use may be reserved for those with so-called ‘treatment resistant depression’, who have failed conventional treatments like escitalopram. 

“Overall I think the authors, peer reviewers and editors have done a good job in keeping this article well balanced. This is important as the psychedelic research field has a certain ‘fever’ in it at the moment that is threatening objectivity. 

What still needs to be done before we know if psilocybin is a suitable therapy for depression? What will the next stage of trials look like?

“We need multi centre phase 2b and, then, phase 3 trials. There are two multi centre trials underway to my knowledge. One is Sponsored by the Usona Institute in the United States and another is Sponsored by Compass Pathways in the UK, Europe and United States. These trials will likely report initial results in 2022. 

What would you want to say someone reading this paper who thinks they may have depression/ is diagnosed with depression but worried about their current treatment etc.?

“Discuss with your GP or mental health care provider about your options. It is often the case that if one treatment does not suit you, then another might. There are lots of approaches to treating depression, not just psychotherapy and medications!

Any other comments?

“I think what this trial tells us is, partly, what we already know. And that is combining a drug treatment with a psychotherapy has a good antidepressant effect that is usually better than either alone. Antidepressants are often prescribed in a context where timely access to good quality psychological therapy isn’t available. That’s a problem of funding that has not been adequately addressed yet by governments.

“The trial was funded by the Alex Mosely charitable trust and by several entrepreneurial donations to the Imperial Centre for Psychedelic Research. This is declared at the end.

“The psilocybin was produced by Compass Pathways and supplied to the trial team by them (I believe at no charge). The lead author (RCH) is a paid member of the Scientific Advisory Board of Compass Pathways. This has been declared in the manuscript at the end. Similarly, many of the other authors declare consultancy work for companies interested in developing psychedelics into treatments. In this small and very nascent field, such things are not unusual, however I think as with all papers the narrative and results should be interpreted in this context (see declarations of interest at the bottom of the roundup for further comments on Dr Rucker’s declarations). 


Dr Paul Keedwell, Consultant Psychiatrist, Cardiff University, said:

“Depression is the leading cause of disability worldwide and existing treatments are often ineffective or cause unacceptable side effects.

“Psilocybin, the psychedelic constituent of “magic mushrooms”, has potential as an effective and long lasting treatment for depression that is not adequately treated by conventional antidepressants. 

“This study provides some tantalising “secondary outcome” results, suggesting that psilocybin might be more effective in the treatment of depression than the antidepressant escitalopram, a common first line treatment for depression. However, using their main outcome measure, no differences were found. The study was not big enough to draw firm conclusions, and there was no control group. 

“Also, the authors admit that the study period was too short to catch the late responders to the escitalopram, which could have biased the findings.

“Nevertheless, these results are encouraging.

“Bigger and longer studies are needed to test the potential of this exciting psychedelic, which is thought to produce “emotional breakthroughs” in depression sufferers. An “active placebo” will be essential, because individuals know when they have been given the psychedelic. 

“Larger studies are already underway in the commercial sector, where millions have been raised in Pharma Startups like Compass Pathways, who have patented their own synthetic psilocybin. The results are keenly anticipated.”


Dr Sameer Jauhar, Senior Clinical Lecturer in Mood Disorders and Psychosis, Institute of Psychiatry, Psychology and Neuroscience (IoPPN), King’s College, London, said:

It is quite something for a UK trial, particularly in the field of psychiatry, to be published in the New England Journal of Medicine.

“We know that Major Depressive Disorder has a significant impact on people’s lives, and that our treatments, both pharmacological (predominantly antidepressant medications) and psychological are not effective for all people, and can have major side effects.

There is little doubt that we need a better range of treatments.

“The use of Psilocybin in the Westerm World is not new-it was synthesised in1959, and used for a variety of psychiatric and medical conditions, though use was curtailed in the 1960s after a number of adverse effects were noted.

“The thinking behind this trial is reasonable-does psilocybin work as an antidepressant, compared to a conventional antidepressant?

“To test this the authors examined psilocybin in 59 people with a clinical diagnosis of Major Depressive Disorder, whose illness was moderate in severity. They gave psilocybin at a reasonable dose and placebo to one group of 30 people and a tiny dose of psilocybin plus an established antidepressant (escitalopram) to 29 people. Therefore they compared psilocybin to an active comparator.

“They found that people’s self-report of depression for both groups decreased significantly, but that there was no statistically significant difference in people’s ratings of depression.

“They examined other measures of depression and wellbeing and though they found results seemed to favour the psilocybin group this was not apparent when controlling for the number of tests performed.

“What are we to make of this trial?

“We know that antidepressants outperform placebo and therefore the design is sound- testing psilocybin versus an active comparator, and both groups did show a good response to treatment.

“The primary outcome was a self-report measure and not a clinician rated measure (though these were secondary measures).

“It is difficult to blind participants to whether they receive psilocybin or not, and those taking part in the trial were a self-selected sample, ie may have had a preference for psychedelics and therefore expected to have more of an improvement.

These caveats are explained in the paper, and are just part of the science-you cannot blame the trial authors for this. If less clinician rated measures had been used it would have been interesting to see if there was a difference when controlling for multiple tests, but one has to remember this is only a Phase 2, i.e. early, trial.

“Of course it is a small sample but we have to start somewhere and this is a good, honest and clear presentation of results.

“The findings on some outcomes are to be cautiously welcomed, as showing a possible signal for psilocybin in those recruited to the trial-and  a lot of secondary measures are related to each other, i.e. if they had focused on less measures they may well have shown a statistically significant difference.

“Any discussion of underlying mechanism of action is purely theoretical-we still do not really know precisely how antidepressants themselves work, but modulation of the serotonin system makes some sense-though notions of one receptor target and increase/decrease of serotonin is too simplistic.

“What does this mean for people with Major Depression? It suggests psilocybin may offer benefits, though we are quite a long way from establishing this in the same manner as shown for antidepressants and other therapies, where we have RCT data from 1000s of people.

“One cannot recommend this as a treatment by any means, but definitely a step forward in terms of the science, and the authors are to be congratulated for conducting such a trial and presenting the results in a conservative fashion, in a premier Medical Journal.”


Dr Andreas Reif, Head of Department of Psychiatry, University Hospital Frankfurt, Goethe University, said:

“In the recent years, the use of psychedelics as antidepressant compounds experienced a renaissance. However, well-controlled studies were largely lacking. This very important and well-controlled study significantly contributes to our knowledge on the efficacy of psilocybin in depression. A clever design was chosen to compare psilocybin against a standard antidepressant, escitalopram. Unfortunately, the primary outcome was negative, which means that regarding a pre-defined endpoint (the QIDS-SR-16 score at week 6) was not significantly different between both treatment conditions, i.e., psilocybin did not seem to be more effective than escitalopram. However, several secondary endpoints were favoring psilocybin. A closer look at the data shows that likely the number of patients in each arm was too small to reveal superiority of psilocybin. Importantly, the safety profile of psilocybin was very good. Given the promise that this completely novel mechanism of action holds, the huge prevalence of depression and considering the large number of patients not benefiting from first-line treatments, we clearly now need a confirmatory international phase III study testing the efficacy of psilocybin in a large number of patients. Such a study could substantially transform depression treatment algorithms; at the moment however, it is too early to recommend the use of psilocybin in routine care.”


Prof Gitte Moos Knudsen, Professor of Neurobiology and Chair of Department of Neurology and Neurobiology Research Unit, Copenhagen University Hospital, Denmark, said:

“Psilocybin, a serotonergic psychedelic drug, has re-emerged as a treatment for different brain disorders, in particular Major Depressive Disorder (MDD). Four randomized, controlled clinical trials of psilocybin as treatment for MDD and anxiety have shown reductions in depressive symptoms after the administration of one or two doses of psilocybin, in combination with some psychological support before, during and after the intervention, which typically lasts up to 8 hours or more.

“The current study compares the drug efficacy of first-line antidepressant treatment with SSRI (here escitalopram) versus assisted psilocybin intervention. The SSRI treatment was not just standard treatment, but was supplemented with a non-hallucinogenic dose of psilocybin (1 mg). Although psilocybin did not turn out to be superior to SSRI treatment, at least relative to the defined primary outcome measure (a self-questionnaire named QIDS SR-16), a number of issues still means that more data is needed. Importantly, a larger sample size could have helped to settle the matter better. In addition, a number of questions still remain to be addressed:

  • Is washout of previous antidepressive treatment required? In the study, at least 2 weeks went before starting trial medication and psychotherapy was stopped at least 3 weeks before.
  • Which patients can safely undergo an intervention with a psychedelic dose and how much screening is needed? Of the 1000 patients considered for inclusion, 891 did not meet inclusion criteria. Is there a sex-difference in drug response? Whereas two-thirds of MDD patients are female, only one-third of the patients were female in this study. To what extent and in which form is building a therapeutic alliance prior to, during and after the intervention needed?
  • Had the non-hallucinogenic dose of psilocybin added to the SSRI treatment any effect on patient expectations in the present study? And is repeated dosing of psilocybin necessary? In the study, psilocybin intervention was repeated after 3 weeks, but the unblinded studies suggest that there are long-lasting effects of one single dose.

“So although the study outcome is a bit disappointing, there are still many factors that need to be scrutinized and tuned before we get a full understanding of the potential of psilocybin for treatment of MDD.”


Prof Phil Cowen, Professor of Psychopharmacology, University of Oxford, said:

Is this good quality research?  Are the conclusions backed up by solid data?

“This was a well conducted study which compared escitalopram (a conventional SSRI antidepressant) and psilocybin in depressed patients. The primary end point (Change in QIDS-SR score at 6 weeks) did not distinguish the two treatment groups but the psilocybin treated patients did generally better on other outcomes. For example at six weeks about twice as many psilocybin treated patients achieved remission (asymptomatic status) on the QIDS SR  as those receiving escitalopram.

How does this work fit with the existing evidence?

“This study is consistent with other psilocybin trials in depression showing that in carefully selected patients psilocybin can produce significant antidepressant effects in the short-term.

Have the authors accounted for confounders?  Are there important limitations to be aware of?

“The main limitation is the difficulty of blinding a study like this. It would have been pretty obvious to most patients whether or not they had received an active dose of psilocybin. This coupled to the fact that the patients in the study self-referred and generally wanted to try psilocybin means that expectancy effects cannot be excluded.

What are the implications in the real world?  Is there any overspeculation?  

“It’s the Buch Cassidy question- ‘Who are these guys?’ The diagnosis of clinical depression covers a broad range of mental health problems. It is not yet clear which particular population of depressed patients, psilocybin will be suitable for. The patients in this study generally have mild to moderate depression. They are mostly male (which is unusual in clinical samples) and the majority are employed and University educated.

What still needs to be done before we know if psilocybin is a suitable therapy for depression? What will the next stage of trials look like?

“A key issue is how far the effect of psilocybin endures. For example if  the patients in the current study who achieved remission to psilocybin are still found to be in remission at six months (having had no further treatment), this will be an important finding.

What would you want to say someone reading about this paper who thinks they may have depression/ is diagnosed with depression but worried about their current treatment etc.?

“Psilocybin is not currently an approved or licensed treatment for depression and self-medication is not recommended. Patients interested in this treatment and interested in being involved in future trials should look to see how far their condition reflects those in the population of patients treated successfully with psilocybin so far. It might be better to wait until longer term follow-up data are available. The psychedelic effects of psilocybin appear to be blocked in patients taking conventional antidepressants.

Any other comments?

“The participants received a two doses of psilocybin three weeks apart but there was little evidence that the second dose produce much additional effect.”


Dr Eric Ruhe, Psychiatrist and Associate Professor at the Department of Psychiatry and Principal Investigator (PI) in Radboudumc (Nijmegen the Netherlands), said:

Is this good quality research?  Are the conclusions backed up by solid data?

“This is a well designed clinical trial in a difficult area of psychedelic research, where psilocybin is combined with psychotherapy. This is difficult to properly blind in two ways and although the researchers did not formally ask participants about blinding, their design is aimed to blind participants and raters, including psychotherapeutic sessions in both treatment arms. The research team managed to obtain most of the data-points despite COVID-19 and strictly adhered to the initial protocol regarding outcomes.

How does this work fit with the existing evidence?

“Previous work in psilocybin research was improperly blinded or in open label studies, so this trial is a first step in order to obtain good quality evidence for the application of psilocybin for Major Depressive Disorder (MDD).

Have the authors accounted for confounders?  Are there important limitations to be aware of?

“Yes, regarding the differences in alcohol consumption, this was corrected for. In addition confounding is also covered by the randomization. An important limitation is the self-referral of most of the participants as mentioned in the manuscript. Most of the patients were suffering from long term -rather unspecified- courses of Major Depressive Illness, and previous treatments were not specified. So the level of treatment resistance remains unclear but could be high or heterogeneous. With respect to the treatment resistance and heterogeneity, this is understandable that a Phase II trial might not address this, but must be addressed in next Phase III studies.

What are the implications in the real world?  Is there any overspeculation? 

“This trial did not show superiority of psilocybin over the control on the primary outcome. However, that seems to be a matter of an underpowered study, suggested by the secondary outcomes as well. A replication of this study, with a larger sample size avoiding this power-issue would be relevant and could provide more clear evidence when results are pooled in a meta-analysis. The authors are modest in their presentation of the conclusion of this study without any overspeculation. Given the fact that we urgently need new treatment modalities for MDD and especially Treatment Resistant Depression, this study should be followed up by other randomized studies that need to better clarify the place of psilocybin as a treatment for MDD. This study shows that this is possible in the field. This trial also shows that adverse effects of the treatment are acceptable, although rare adverse effects might not have been encountered with the small sample sizes.

What still needs to be done before we know if psilocybin is a suitable therapy for depression? What will the next stage of trials look like?

“See above, 1. replication of this study with a larger sample size; 2. analysis and publication of long-term follow-up data; 3. Phase III trials, preferably in larger and more widespread application, including long-term efficacy.

What would you want to say someone reading about this paper who thinks they may have depression/ is diagnosed with depression but worried about their current treatment etc.?

“Realize that this is not yet a definite answer on whether psilocybin treatment is a clear solution for MDD and that if one considers to be willing to take this drug do so while participating in a trial that will properly investigate the effects of psilocybin.



‘Trial of Psilocybin versus Escitalopram for Depression’ by Robin Cahrat-Harris et al. was published in the New England Journal of Medicine at 22:00 UK time on Wednesday 14 April.

DOI: 10.1056/NEJMoa2032994



Declared interests

Prof Kevin McConway: “I am a Trustee of the SMC and a member of its Advisory Committee.  I am also a member of the Public Data Advisory Group, which provides expert advice to the Cabinet Office on aspects of public understanding of data during the pandemic.  My quote above is in my capacity as an independent professional statistician.”

Prof Guy Goodwin: “I have acted as an advisor to Compass pathfinder Ltd and other companies involved in the development of drugs to treat major depression.”

Prof Anthony Cleare: “Prof Cleare has in the last three years received honoraria for speaking from Lundbeck and Janssen; honoraria for consulting from Livanova, Allergan, Janssen and NICE; sponsorship for conference attendance from Janssen; and research grant support from the Medical Research Council (UK), Wellcome Trust (UK), the National Institute for Health Research (UK) and Protexin Probiotics International Ltd. One of Prof Cleare’s research interests is the efficacy of psychedelic treatments.”

Dr James Rucker: “I lead a trials group at King’s College London in which we also investigate the safety and efficacy of psilocybin. Two of those trials are funded by Compass Pathways. I don’t benefit personally from those grants and I am not on the Scientific Advisory Board of Compass. Compass also supply the psilocybin for a UK government funded study I am the Principal Investigator of (again, free of charge). Otherwise, I have received consultancy fees from two companies interested in developing psychedelics – they are called Clerkenwell Health and Beckley PsyTech. My salary is paid for by the National Institute for Health Research, which is a UK government funded body.”

Dr Sameer Jauhar: “No personal conflucts here

IoPPN is conducting RCTS of Psilocybin, though I have no role or funding associated with this.”

Dr Andreas Reif: “I have no conflict of interest to disclose.”

Prof Gitte Moos Knudsen: “No conflict of interest.”

Dr Eric Ruhe: “I am a psychiatrist specialized in the treatment of (treatment resistant) affective disorders, especially unipolar depression (MDD) in the Radboudumc, Nijmegen Netherlands. I have no conflicts of interests regarding this study, nor do I have a connection with the authors.”

None others received.

in this section

filter RoundUps by year

search by tag