This systematic review and meta-analysis of four randomised, double‑blind trials found MDMA‑assisted psychotherapy produced significant reductions in PTSD severity (CAPS‑IV) at 75 mg and 125 mg versus active placebo, with mainly transient adverse effects and minimal neurocognitive or physical risk, but little consistent benefit on depressive symptoms (BDI); larger, better‑powered RCTs are required.
Rationale
Novel, evidence-based treatments are required for treatment-resistant post-traumatic stress disorder (PTSD). 3,4-Methylenedioxymethamphetamine (MDMA) has beneficially augmented psychotherapy in several small clinical trials.
Objective
To review the use of MDMA-assisted psychotherapy in treatment-resistant PTSD.
Methods
Systematic searches of four databases were conducted from inception to February 2020. A meta-analysis was performed on trials which were double-blinded, randomised, and compared MDMA-assisted psychotherapy to psychotherapy and placebo. The primary outcomes were the differences in Clinician Administered PTSD Scale (CAPS-IV) score and Beck’s Depression Inventory (BDI). Secondary outcome measures included neurocognitive and physical adverse effects, at the time, and within 7 days of intervention.
Results
Four randomised controlled trials (RCTs) met inclusion criteria. When compared to active placebo, intervention groups taking 75 mg (MD −46.90; 95% (confidence intervals) CI −58.78, −35.02), 125 mg (MD −20.98; 95% CI −34.35, −7.61) but not 100 mg (MD −12.90; 95% CI −36.09, 10.29) of MDMA with psychotherapy, had significant decreases in CAPS-IV scores, as did the inactive placebo arm (MD −33.20; 95% CI −40.53, −25.87). A significant decrease in BDI when compared to active placebo (MD −10.80; 95% CI −20.39, −1.21) was only observed at 75 mg. Compared to placebo, participants reported significantly more episodes of low mood, nausea and jaw-clenching during sessions and lack of appetite after 7 days.
Conclusion
These results demonstrate potential therapeutic benefit with minimal physical and neurocognitive risk for the use of MDMA-assisted psychotherapy in TR-PTSD, despite little effect on Beck’s Depression Inventory. Better powered RCTs are required to investigate further. International Prospective Register of Systematic Reviews: CRD42019109132 available online at www.crd.york.ac.uk/prospero .
Papers cited by this study that are also in Blossom
Amoroso, T., Workman, M. · Journal of Psychopharmacology (2016)
Bahji, A., Forsyth, A., Groll, D. et al. · Progress in Neuro-Psychopharmacology and Biological Psychiatry (2020)
Dumont, G., Sweep, F., van der Steen, R. et al. · Social Neuroscience (2009)
Liechti, M. E., Gamma, A., Vollenweider, F. X. · Psychopharmacology (2001)
Post-traumatic stress disorder (PTSD) is a chronic and disabling condition characterised by intrusive memories, hyperarousal, avoidance and mood and cognitive changes following exposure to traumatic events. The disorder has a substantial individual and societal burden, lifetime prevalence estimates around 3.6–3.9% worldwide, and higher rates in conflict-affected populations. A sizeable proportion of patients do not remit with standard treatments; the authors note that definitions of treatment-resistance vary but often denote failure to respond to at least two evidence-based therapies. Against this background, Illingworth and colleagues set out to evaluate whether MDMA (3,4-methylenedioxymethamphetamine) administered in a supervised clinical setting as an adjunct to psychotherapy provides benefit for treatment-resistant PTSD (TR-PTSD). Earlier phase 1 and phase 2 work suggested MDMA can facilitate psychotherapy through pro-social, anxiolytic and fear-reducing effects, but trials are small and blinding is challenging. This systematic review and meta-analysis therefore aimed to identify double-blind randomised trials comparing MDMA-assisted psychotherapy to psychotherapy plus placebo and to quantify effects on clinician-rated PTSD severity (CAPS-IV), depressive symptoms (BDI) and adverse events, including session and 7-day follow-up events.
The review was prospectively registered on PROSPERO (CRD42019109132) and followed PRISMA guidance. The investigators searched PUBMED, PsycINFO, EMBASE and MEDLINE from September 2018 to February 2020 for double-blind randomised controlled trials of MDMA-assisted psychotherapy in TR-PTSD (search terms included MDMA, 3,4-methylenedioxymethamphetamine, psychotherapy, PTSD). Two reviewers independently screened titles/abstracts, assessed full texts for eligibility and extracted trial characteristics. Discrepancies were resolved by discussion, contact with study authors, or a third reviewer. Eligible studies required a PTSD diagnosis and randomised comparison between MDMA plus psychotherapy and placebo plus psychotherapy; active placebo was defined as a lower MDMA dose and inactive placebo as a non‑psychoactive substance. Trials conducted in inpatient or outpatient settings were eligible. Primary outcomes pre-specified were change in Clinician Administered PTSD Scale (CAPS-IV) at more than 3 weeks after blinded sessions and Beck Depression Inventory (BDI). Secondary outcomes included psychological (anxiety, low mood, insomnia) and physical adverse events (jaw clenching, insomnia, nausea, headache, anorexia) during sessions and at 7-day follow-up. Risk of bias for each trial was assessed using the Cochrane Risk of Bias tool. Review Manager 5.3 was used for analysis. Given heterogeneity, a random-effects inverse-variance model was applied for primary outcomes; fixed-effects (Mantel–Haenszel) models were used for many secondary outcomes. When event rates were very low, Peto odds ratios were employed. The authors contacted original trial investigators to clarify or obtain missing data when necessary.
The search yielded 688 references; after deduplication and screening, six RCTs reporting 109 participants initially met criteria but two were excluded (one lacked appropriate follow-up, one duplicated participant data), leaving four randomised trials totalling 85 participants for the primary analyses. Individual trial sizes ranged from 12 to 28 participants. All included trials reported the primary outcome of CAPS-IV change after at least 3 weeks. For the primary PTSD outcome, intervention groups receiving MDMA plus psychotherapy showed significant reductions in CAPS-IV scores at certain doses. Compared with active placebo, the 75 mg and 125 mg MDMA groups had statistically significant decreases in CAPS-IV; the 100 mg group did not reach statistical significance (MD = -12.90; 95% CI -36.09 to +10.29; p = 0.28). When compared to inactive placebo, the 125 mg MDMA plus psychotherapy group also showed a statistically significant reduction in CAPS-IV (reported MD -33.20; the extracted text reports the 95% confidence interval as -0.53, -25.87 and p < 0.00001, but the ordering of the interval in the extraction is unclear). Regarding depressive symptoms, only the 75 mg MDMA plus psychotherapy arm showed a significant improvement in Beck Depression Inventory scores (MD -10.80; p = 0.02). The extracted Results text did not clearly report the 95% CI for this estimate, although the abstract presents it as -20.39 to -1.21. No significant BDI differences were reported for the 100 mg and 125 mg comparisons. Secondary outcomes demonstrated increased rates of several physical adverse events in MDMA arms. Jaw clenching occurred significantly more often at several dose levels: 75 mg (Peto OR 13.46; 95% CI 1.44, 125.80), 125 mg per patient (Peto OR 7.34; 95% CI 1.98, 27.71), 125 mg per session (Peto OR 4.08; 95% CI 1.04, 15.99) and 150 mg (RR 8.67; 95% CI 1.21, 61.91); the 100 mg comparison did not reach significance (RR 1.67; 95% CI 0.47, 5.96). Nausea during sessions and lack of appetite within 7 days were more frequent with 125 mg MDMA compared with inactive placebo (reported RR 4.00; 95% CI 1.03, 15.53; other CI reporting in the extraction is incomplete). The authors summarised session-level frequencies across trials: nausea occurred in 58% of sessions, low mood in 17% of sessions, jaw clenching in 30% of sessions and lack of appetite within 7 days in 25% of sessions; jaw clenching was particularly common at the highest reported dose (67% of 150 mg sessions). Four serious adverse events were reported across studies, including episodes of suicidal ideation and one hospital admission related to suicidal thoughts. The authors also noted compromised blinding in included trials, citing evidence (from Mithoefer et al.) that most participants and therapists could correctly guess treatment assignment after higher doses.
Illingworth and colleagues interpret their findings as indicating that MDMA-assisted psychotherapy is associated with clinically meaningful reductions in clinician-rated PTSD severity (CAPS-IV) in treatment-resistant PTSD, at doses of 75 mg and 125 mg when compared with active placebo and at 125 mg versus inactive placebo; the 100 mg comparison did not reach significance. The authors emphasise that depressive symptoms, as measured by BDI, showed limited improvement overall, with a significant change observed only at the 75 mg dose. The discussion situates these results within earlier phase 2 work showing feasibility and promising signals of efficacy, and notes the biological and psychological rationale for MDMA as a psychotherapy adjunct (reduced fear, increased trust and openness). Safety was described as largely favourable for neurocognitive and physical risk at therapeutic doses under clinical supervision, but notable transient somatic adverse effects and some instances of increased suicidal ideation were observed. The reviewers highlight that some adverse events (for example jaw clenching) showed a dose–response relationship, arguing against adopting higher doses without clear additional benefit. Key limitations acknowledged by the authors include the small number of trials and small sample sizes, inconsistent reporting of secondary outcomes across studies, and likely suboptimal blinding because doses produced perceptible psychoactive effects. These factors increase uncertainty and limit generalisability. The authors recommend the development of a core outcome set for PTSD trials to harmonise reporting (for example standardising measures such as BDI/PHQ-9/GAD-7 and CAPS-IV measurement intervals) and to capture adverse events up to 12 months. Finally, the paper notes that ongoing Phase 3 trials will be important to establish efficacy and safety more definitively, that subgroup analyses may identify populations most likely to benefit, and that the intervention should be seen as a combined drug-plus-psychotherapy treatment requiring skilled therapeutic management rather than as a pharmacological or psychotherapeutic intervention in isolation.
The authors conclude that MDMA-assisted psychotherapy is associated with significant reductions in CAPS-IV scores in subjects with treatment-resistant PTSD but has little consistent effect on Beck Depression Inventory scores. They describe a potential therapeutic benefit with minimal neurocognitive risk when administered in a controlled clinical setting, while emphasising that findings must be interpreted cautiously given small sample sizes and limited trial numbers. Larger-scale studies are required to clarify efficacy, safety and optimal dosing before clinical implementation can be recommended.
Post-traumatic stress disorder (PTSD) is a psychiatric disorder triggered by exposure to a psychologically traumatic event, and is characterised by anxiety, sleep disturbance, hypervigilance, avoidance behaviour, and intrusive thoughts and memories about the trauma. To satisfy diagnostic criteria, symptoms must continue at least 1-month after exposure and cause significant distress and social, occupational or other dysfunction. Newer diagnostic criteria (DSM V) have also included changes in mood and cognition as part of this disorder (American PsychiatricExamples of commonly associated traumatic events include sexual assault, particularly at a young age, and witnessing or being the victim of violence, particularly in conflict-zones. PTSD is a disabling condition and inflicts a large individual and societal burden. It is reported to have a lifetime prevalence of between 1.0-9.2% in the general population (based on DSM criteria), with more robust epidemiological data placing the estimated figure at 3.64-3.9% worldwide. PTSD is particularly prevalent in conflict-affected populations such as refugees (mean: 30.6%). It is estimated that PTSD tends to remit in only about half of individuals after a period of more than 3 years. The definition of treatment-resistance varies across the literature and between the studies analysed within this review, though it is often considered as failure to respond to at least two evidence-based therapies. Further complexity arises because of inconsistency in the definition of 'treatment response', for example one commonly used and validated benchmark is a 10-point reduction in CAPS-IV, however many others exist. Because of this heterogeneity in the literature, for the purposes of this review all trials describing their subjects as 'treatment resistant' were included, regardless of the criteria used to determine said treatment resistance.
There are close links between PTSD and physical health: not only are those who have had life-threatening illnesses, such as strokes and heart attacks, more likely as a result to develop PTSD, those with PTSD are also at greater risk of future morbidity and mortality from a range of physical medical conditions. PTSD populations also have higher rates of psychiatric comorbidity than non-affected populations. Rates of self-harm and suicide in those with PTSD are significantly higher than those in the general population, and this effect is even more pronounced in the subset of patients for whom the majority of relevant trauma was experienced during childhood (van der. Without intervention PTSD is frequently a chronic condition, with the median self-reported 'duration of symptoms' for those not seeking professional treatment over 4 years. Even for those seeking treatment, the median duration of symptoms is approximately 3 years.
Management of PTSD generally consists of pharmacological treatments, primarily serotonin reuptake inhibitors, as well as psychotherapies that involve exposure to and processing of traumatic events. Many types of psychotherapy are used in the management of PTSD including exposure therapies, trauma-focused cognitive behavioural therapy (CBT), interpersonal therapy (IPT) and eye-movement desensitisation and reprocessing. Of these, IPT is notable in that it does not require exposure to, or processing of, traumatic experiences. Psychotherapy is first line management in the UK (National Institute for Health and Care Excellence (NICE), 2019). There is good evidence for the efficacy of a range of psychotherapies for the treatment of PTSD. Of the pharmacotherapy options, the strongest evidence is for selective serotonin reuptake inhibitors (SSRIs) and venlafaxine, a serotonin and norepinephrine reuptake inhibitor (SNRI). Psychotherapy alone appears to be more effective than pharmacotherapy alone although methodological difficulties, for example with blinding and sampling, in trials of psychotherapies means their effectiveness may be overestimated. On the whole, pharmocotherapy has not been shown to be effective at augmenting psychotherapy in clinical trials, however the two treatment modalities are nonetheless frequently used in combination. Despite this, a significant proportion (33-60%) of patients do not achieve remission regardless of the regimen used. Because of its chronicity and the number of patients who do not achieve satisfactory response with currently available therapies, there is a pressing need for new treatment options for PTSD. Proposed benefits of MDMA for augmentation of psychological therapies 3,4-methylenedioxymethamphetamine (MDMA) is a psychoactive drug that has been widely used as a recreational drug for its entactogenic (promoting feelings of empathy and relatedness), pro-social and euphoric effects. It may elicit these effects through modulation of monoamine neurotransmitter systems. Administration of MDMA also results in increased circulating concentrations of oxytocin, which may mediate interpersonal bond-forming and trust, and prolactin, thought to be responsible for relaxation and receptivity to novel ideas and experiences. Reductions in fear and negative emotions, as well as increases in empathy, extroversion and confidence are among the other positive subjective effects of MDMA. Similar effects have been demonstrated in SSRIs in the context of. Because of these subjective and appreciable effects of MDMA, active placebo is frequently the comparator of choice in trials. Active placebos are control interventions that mimic some of the psychophysiological effects but not therapeutic effects of the intervention under investigation, in an attempt to reduce the risk of unblinding. Examples include using an alternative medication that produces similar side effects or alternatively using a lower dose of the drug under investigation. In those affected by PTSD, excess fear and aversion may impair access to traumatic memories. If they are accessed, associated states of hyper-arousal or dissociation may prevent psychological resolution. By moderating fear responses, MDMA may have utility in positively augmenting psychotherapy for PTSD by inducing a psychological state more conducive to accessing and processing trauma.
Side effects of MDMA include, but are not limited to, dry mouth, nausea, anxiety, inability to urinate, bruxism, restlessness and increases in heart rate, blood pressure and body temperature. Higher doses can cause a potentially fatal serotonin syndrome. Deaths have also rarely been reported with seemingly modest doses of 'street' MDMA; for example as an apparent consequence of polydipsia and/or syndrome of inappropriate anti-diuretic hormone (SIADH), with resultant hyponatraemia sometimes leading to cerebral oedema. In 2018 there were 92 deaths associated with MDMA use in England and Wales (The Office for National Statistics, 2019). However at the doses used in therapy, and without the impurities and adulterants often present in illicitly-sourced MDMA, side effects are usually transient, mild and unlikely to pose any significant risk to participants under close clinical supervision.
The intervention investigated in this review is the administration of MDMA in an expertly supervised clinical setting as an adjunct to a limited number of psychotherapy sessions. The feasibility of administering MDMA during supervised psychotherapy has been demonstrated in phase 1and phase 2 clinical trials. Further phase 3 trials are planned or ongoing in the USA (Identifier NCT03537014)and Europe. There have been previous systematic reviews in this area published in 2016and 2019. This present review has several notable differences, updating and complementing this previous research.used robust statistics to compare the effect sizes of MDMA assisted psychotherapy and prolonged exposure (PE) therapy, finding both treatments had comparable outcomes. However, PE effect sizes were based on several studies comparing PE to wait-list controls, while MDMA assisted psychotherapy was compared to equivalent psychotherapy plus active placebo. As a result, the effect-sizes for each intervention are not well suited to direct comparison and may relatively overstate PE's effect size. This paper also used half the data (two trials with combined n = 37) that we have included in this more up-to-date study (four trials with combined n = 85).differs in its analysis of the primary outcome, reporting risk-ratios for 'loss of PTSD diagnostic status' and 'clinical response' (with three different benchmarks for clinical response used among the five included studies). This incorporates into the primary outcome some indication of the likelihood of clinically significant, as well as statistically significant change. Our analysis complements this review by reporting the mean change in CAPS-IV score at individual dosages, which will provide the reader with more information on the diversity of response magnitude, that may be lost if the complexity of continuous data is reduced to a binary outcome (e.g. response v. no-response; if 15% improvement is considered 'response', improvements of 20 and 90% are given equal weight). The present review also provides more in-depth analysis of adverse events (AE) and secondary outcomes.
The protocol was registered on PROSPERO: International Prospective Register of Systematic Reviews. The registration number is CRD42019109132 and is available online (www.crd. york.ac.uk/prospero). We followed the reporting guidelines for meta-analyses and systematic reviews of randomised controlled trials, as outlined by the Preferred Reporting Items for Systematic Reviews and Meta-Analyses statement. We searched PUBMED, PsychINFO, EMBASE and MEDLINE from September 2018 to February 2020 to identify published double-blinded, randomised controlled trials evaluating the efficacy and safety of MDMAassisted psychotherapy treatment in TR-PTSD patients (search terms: MDMA; 3,4-methylenedioxymethamphetamine; psychotherapy; PTSD; post-traumatic stress disorder). Two review authors (BJGI and DJL) independently screened titles and abstracts. They critically reviewed the full text of selected randomised trials to assess eligibility. We included studies where participants had been diagnosed with PTSD and were treated with MDMA and psychotherapy. We only included randomised control trials. In all studies included for analysis in this review where the intervention was compared to placebo, 'active placebo' should be taken to mean a lower dose of MDMA than used in the intervention arm. An inactive placebo should be taken to mean a substance with no psychoactive effects. Trials conducted in inpatient or outpatient settings were considered eligible. Two review authors (BJGI and DJL) independently extracted study characteristics including trial design, setting, participants and outcomes. We extracted all relevant data from each randomised trial report. Two review authors (BJGI and DJL) independently assessed the risk of bias for each study using the Cochrane Risk of Bias tool. Each trial was evaluated based on its approach to minimizing selection, performance, detection, attrition, and reporting bias. Two review authors (BJGI and DJL) independently extracted data from each randomised trial report related to pre-specified primary and secondary outcomes. The primary outcomes were Clinician Administered PTSD Scale (CAPS-IV) score at greater than 3 weeks and the Beck Depression Inventory (BDI). Secondary outcomes included episodes of anxiety and low mood and physical adverse effects (jaw clenching, insomnia, nausea, headache and anorexia), reported both during the session and at 7-day follow-up. We contacted authors to seek clarification and requested missing data or additional data to complete our analysis. Discrepancies between the reviewers were resolved through discussion, by contacting the authors, or by consultation with a third reviewer (JMND). Review Manager 5.3 (Cochrane Collaboration, UK) was utilised to analyse results. We used a random-effects model (inverse variance) to calculate the summary estimates given the statistical heterogeneity of our data. For secondary outcomes, we used a fixed-effects (Mantel-Haenzsel method) to calculate summary estimates. When there were few or no events we used the peto odds ratio (POR) as this is a less biased and more powerful method when comparing studies with low event rates. For secondary outcomes where events were reported as both per session and per patient in different studies at the same dosage, we calculated separate summary estimates because these denominators are not directly comparable.
We imported 688 references for screening. After excluding 573 duplicate records, 115 titles and abstracts were screened which identified 13 potentially relevant studies. Six randomised controlled trials reporting data from 109 participants met our inclusion criteria. Of these trials one was excluded for not containing appropriate follow up timesanother was excluded as the participant data was previously used in another trial included in the analysis. Four trials using 85 participants were used in the final analysis (Figure). Included randomised trials evaluated the reduction of PTSD symptoms in subjects with treatment-resistant PTSD. The included randomised trials were relatively small, ranging from 12 to 28 participants. (The authors' judgments regarding the risk of bias for the included studies are detailed in Figure)
One primary outcome was the change in CAPS-IV score from baseline to greater than 3 weeks after the blinded sessions. When compared to active placebo, the intervention groups taking 75 mg and 125 mg of MDMA in conjunction with psychotherapy had significant decreases in CAPS-IV scores (Figuresand Table). The decrease in CAPS-IV score at 100 mg, however, did not reach significance (MD = -12.90; 95% CI (-36.09, +10.29), p = 0.28) (Figure). When compared to inactive placebo, a statistically significant reduction in PTSD symptoms was also noted in the intervention groups taking 125 mg of MDMA in conjunction with psychotherapy (Figureand Table) (MD -33.20; 95% CI -0.53, -25.87, p = < 0.00001). As the 150mg dose was only administered in open-label sessions we have not included the data in our analysis of the primary outcome. The other primary outcome measurement was the Beck's Depression Inventory (BDI). When compared to active placebo, there was no significant difference noted in BDI at higher strengths of MDMA (100 and 125 mg) when taken in conjunction with psychotherapy and only the intervention group taking 75 mg MDMA in conjunction with psychotherapy saw a significant improvement in BDI (MD -10.80; 95% CI), p = 0.02),and Table).
We calculated risk ratios and PORs for both physical (jaw clenching, headache, loss of appetite, nausea) and psychological adverse effects (anxiety, low mood, insomnia) at intervention and within the 7 days subsequently (Tablesand). Jaw clenching occurred significantly more often when compared to inactive placebo) and active placebo when taken at 75 mg (POR 13.46; 95% CI 1.44, 125.80), 125 mg (per patient) (POR 7.34; 95% CI 1.98, 27.71), 125 mg (per session) (POR 4.08; 95% CI 1.04, 15.99), 150 mg (RR 8.67; 95% CI 1.21, 61.91) but not 100 mg (RR 1.67; 95% CI 0.47, 5.96). Nausea during the session and lack of appetite within 7 days occurred significantly more often when taking therapeutic doses of MDMA at 125 mg when compared to inactive but not active placebo (RR 4.00; 95% CI 1.03, 15.53) (POR 8.14; 95%
The results of this meta-analysis suggest that the use of MDMA in conjunction with psychotherapy is associated with a significant decrease in CAPS-IV scores at greater than 3 weeks when compared to both active (except at 100 mg) and inactive placebo groups in the treatment of treatment-resistant PTSD. MDMA-assisted psychotherapy showed no significant decreases in BDI except for at 75 mg. Subjects in these trials experienced significant physical side effects from the intervention. These included nausea (58% of sessions), low mood (17% of sessions) and jaw clenching in session (30% of sessions) and also lack of appetite within 7 days of intervention (25% of sessions). Jaw clenching was more prevalent at higher doses (67% of 150 mg sessions). Four serious adverse effects (SAE), including suicidal ideation, were recorded during one study.
The strengths of this prospectively registered systematic review include its comprehensive search strategy, methodological design and statistical analysis. All the studies included reported the primary outcome, reduction in CAPS-IV score after 3 weeks of intervention. Another strength of this paper as compared to other reviews and the pooled analysis performed by, is that it looks in-depth at the secondary outcomes and adverse events. This paper has attempted to compare the intervention arm against active and inactive placebo separately. It has also assessed the dose differences on effect size. Pooling of data, as used by previous studies, may be inappropriate as it assumes that different MDMA doses or inactive and active placebos are comparable or interchangeable. This study has several limitations. The comprehensive search strategy only identified four randomized trials that reported data from a relatively small number of participants. The sample sizes that were included in the trials were low. As numbers within a sample decrease, so the magnitude of chance variation within the sample increases, which may lead to errors of inference and comparison. Our results should, therefore, be treated with caution. A further limitation is that outcome measures (such as BDI) are not reported consistently throughout all the papers. To address this, and to ensure that there is adequate evidence in order to inform clinical practice, a core outcome set for use in all future PTSD trials would be helpful. This might include measures such as the BDI, PHQ-9 and Generalised Anxiety Disorder 7 score (GAD-7), intervals at which the CAPS IV should be measured (1 month, 2 months, 12 months), and a set of 'adverse events of special interest', occurring up to 12 months after treatment. The doses used in these studies were large enough to produce noticeable psychoactive and physical effects. This implies suboptimal blinding of both participants and investigators in all of the studies included. This failure of the blind is supported by evidence from Mithoefer et al., where 95% of participants were able to guess which treatment arm they were in after receiving a 125 mg dose of MDMA. Therapists in this trial were also able to identify which treatment arm the patients were on with complete accuracy.
MDMA-assisted psychotherapy has been shown to have a largely favourable safety profile and have some effect in reducing CAPS-IV scores in participants with treatment-resistant PTSD in phase 2 trials. Ongoing phase 3 trials will provide further evidence on whether efficacy and safety parameters are sufficiently favourable to warrant an application for a marketing authorisation. Certain symptomatology clusters or trauma aetiologies (such as combat or domestic violence) may be more amenable to MDMA-assisted psychotherapy. Sub-group analysis may identify those most likely to respond, although this has proven challenging with other treatments. This meta-analysis adds to a literature that shows that MDMAassisted psychotherapy may have a role in the reduction of symptoms of PTSD, although our results also highlight that MDMA-assisted psychotherapy may have only limited effects on depressive symptoms. Our analysis shows, overall, that patients undergoing MDMA-assisted psychotherapy did not experience an increase in reported depressive or anxious episodes in the week following the session, in either the active or inactive arms of the trial. Some participants experienced a transient increase in suicidal ideation and one participant was admitted to hospital because of suicidal thoughts 13 days after their second 30mg session. Transient increases in suicidal ideation were more common in the active MDMA group, although no suicidal behaviour was reported. Larger trials will be required to investigate this phenomenon more definitively. To an extent, the emergence of low mood and suicidal ideation may be interpreted as a necessary part of engaging with and re-processing long-avoided traumatic experiences. On the other hand, such processes could be understood as a biological effect of MDMA itself. The truth likely lies in-between, bringing the skilled management of such processes by the attending therapist and psychiatrist into necessary focus. This treatment is not just drug, nor is it just psychotherapy. In this sense, it is rather more complex and 'real world' than current paradigms of clinical trial design are able to interrogate. Part of the purpose of phase 2 trials is dose finding. Doses of MDMA as low as 75mg had a significant effect on the symptoms of TR-PTSD when compared to active placebo and were as effective as 125 mg. Some somatic side effects were more prevalent at lower doses of MDMA; however, our analysis detected no significant difference. At higher doses, some physical adverse events such as jaw clenching were significantly more prevalent in the experimental group and followed a dose-response relationship, suggesting a causative aetiology. On this basis it would be difficult to justify increased doses of MDMA, however it may be fruitful to examine the effects of lower doses, particularly in individuals who otherwise appear sensitive to the effects of MDMA.
In summary, this meta-analysis has demonstrated that use of MDMA-assisted psychotherapy is associated with significant decreases in CAPS-IV scores but little effect in BDI in subjects with TR-PTSD, indicating a potential therapeutic benefit with minimal physical and neurocognitive risk. Care must be taken when interpreting these results due to the small sample sizes involved and data from larger scale studies are needed to provide further evidence on whether efficacy and safety parameters are sufficiently favourable to warrant use in a clinical setting.
Create a free account to open full-text PDFs.
Mithoefer, M. C., Feduccia, A. A., Jerome, L. et al. · Psychopharmacology (2019)
Papers in Blossom that reference this study
Sevchik, B. L., Singleton, S. P., Lahey, A. et al. · MedRvix (2026)
Kachmarik, J. E., Loftis, J. M., Stauffer, C. S. · Frontiers in Neuroscience (2026)
Van Dongen, N. N. N., Zijlmans, J., Vermetten, E. et al. · European Journal of Psychotraumatology (2024)
Colcott, J., Guerin, A. A., Carter, O. et al. · Neuropsychopharmacology (2024)
Van Vugt, A. S., Zijlmans, J., Lindauer, R. et al. · Drug Science Policy and Law (2023)
Green, W. M., Raut, S. B., James, F. L. J. et al. · MedRvix (2023)
Belser, A. B. · Frontiers in Psychology (2022)
Earleywine, M., Low, F., Lau, C. et al. · Journal of Humanistic Psychology (2022)
Bird, C. I. V., Modlin, N. L., Rucker, J. · International Review of Psychiatry (2021)