MDMA-Assisted Therapy for Major Depressive Disorder: A Seven-Month Follow-Up Proof of Principle Trial

Major depressive disorder (MDD) is a common, recurrent and disabling illness for which many patients do not achieve sustained benefit from existing pharmacological and psychological treatments. The authors note high non-response and relapse rates in routine care and trials, and discuss MDMA-assisted therapy (MDMA-AT) as a candidate intervention whose subjective effects (for example, increased self-compassion and emotional processing) may be favourable for psychotherapeutic work. Although pooled long-term data from Phase II MDMA-AT studies for PTSD suggest durable effects, the long-term course of outcomes after MDMA-AT in people with MDD remained uncertain. The present study aimed to provide preliminary long-term follow-up data on safety, retention and sustained efficacy of MDMA-AT in people with moderate to severe MDD. Specifically, the researchers assessed clinical and self-reported outcomes at baseline, after a short treatment course (two MDMA dosing sessions plus preparatory/integration psychotherapy) and at a single follow-up visit seven months after baseline (approximately six months after the final MDMA session and four months after end of treatment). This proof-of-principle study sought to test whether improvements observed shortly after treatment were maintained at that follow-up point.

This was a follow-up study of a previously reported proof-of-concept MDMA-AT trial. Individuals who had taken part in the trial were invited to a single follow-up visit seven months after baseline. Twelve participants who had received open-label MDMA-AT were included. Main inclusion criteria for the original trial were age ≥18 years, MDD of at least moderate severity at inclusion and episode duration between more than three and less than 24 months. Key exclusions included psychosis, mania, personality disorder, significant suicide risk and cardiac disease. The follow-up study received ethics approval and all participants provided written consent. Treatment comprised two day-long MDMA dosing sessions about one month apart, embedded within a course of nine 90-minute psychotherapy sessions for preparation and integration. Dosing was 80 mg with a 40 mg supplemental dose in the first session, and 120 mg with a 60 mg supplemental dose in the second session. Psychotherapists were trained and followed an MDMA Treatment Manual similar to those used in MDMA-AT PTSD protocols. The follow-up visit occurred seven months after baseline (six months after the final MDMA session and four months after the end-of-trial post-treatment assessment). Primary and secondary outcomes were the clinician-rated Montgomery–Åsberg Depression Rating Scale (MADRS) and the self-reported Sheehan Disability Scale (SDS), respectively. The MADRS was administered by raters described as independent of the study team and unaware of participants' psychiatric history; responder status was defined as ≥50% reduction in MADRS and remission as MADRS ≤12. Sustained response and relapse were prespecified (sustained response = initial response maintained at follow-up; relapse = initial remission then MADRS ≥18 at follow-up). Exploratory self-report measures repeated at follow-up included BDI-II (depression), GAD-7 (generalised anxiety), PCL-5 (PTSD symptoms), Bergen Insomnia Scale (BIS), AUDIT and DUDIT (alcohol/substance use). Suicidality from post-treatment through follow-up was assessed with the Columbia–Suicide Severity Rating Scale (C-SSRS). For analysis, the researchers used mixed-effects regression models to account for repeated measures within participants. Outcomes judged approximately normally distributed (MADRS, SDS total, BDI-II, GAD-7, PCL-5, BIS) were analysed with linear mixed-effects models; non-normally distributed measures (AUDIT, DUDIT, SDS days lost/underproductive days) were dichotomised at prespecified cut-points and analysed with mixed-effects logistic regression. Age, sex and a time variable were included as explanatory variables. Because the number of days between visits varied, three functional forms of time (linear, second-order polynomial, reciprocal) were compared using Akaike's information criterion to select the best fit.

All 12 participants who received MDMA-AT completed the follow-up visit. The sample had a mean age of 44.1 years (SD 9.2); seven were women, most were white, and most had recurrent MDD with a mean time since first episode of about 16 years. Participants had previously tried on average about three antidepressants. Clinician-rated depression (MADRS) showed a statistically significant reduction from baseline to follow-up; no significant change was detected between the post-treatment visit and follow-up. The abstract reports p<0.001 for the MADRS change. At the post-treatment visit 9 of 12 participants (75.0%) met responder criteria and 9 of 12 (75.0%) met remission criteria; at follow-up, sustained response was observed in 8 of 12 participants (66.7%) and sustained remission in 8 of 12 (66.7%). One participant (8.3%) relapsed between post-treatment and follow-up. The three non-responders at post-treatment remained non-responders at follow-up. Functional impairment measured by the SDS total score also showed a significant reduction from baseline to follow-up (reported p=0.001), with no significant change from post-treatment to follow-up. For SDS subscales, mixed-effects logistic regression indicated an overall significant decrease in days lost (model testing details reported; a Wald test gave p=0.077 while a likelihood-ratio test supported inclusion of reciprocal days since baseline, p=0.001). SDS underproductive days showed a significant reduction (coefficient 2.86, z=2.82, p=0.005); sex and age did not have significant effects, although there was a non-significant indication that men had lower underproductive-day scores than women (p=0.134). Exploratory self-report measures (BDI-II, PCL-5, GAD-7, BIS) showed significant reductions from baseline to follow-up on mixed-effects modelling, with no significant change between post-treatment and follow-up. Alcohol and substance use did not change, but baseline levels were low in line with exclusion criteria. The analyses did not find changes attributable to sex or age for these measures. Suicidality: at baseline all participants reported lifetime suicidal ideation, 5/12 reported lifetime serious ideation and 4/12 had prior suicidal behaviour. No new cases of serious suicidal ideation or suicidal behaviour occurred through to the post-treatment visit. At follow-up the mean C-SSRS ideation score was 0.9 (SD 1.3) compared with 1.7 (SD 1.3) in the six months prior to screening; mean SI intensity rating at follow-up was 4.5 (SD 6.8) versus 7.2 (SD 5.9) pre-screening. No participant showed worsening of SI relative to any point in the treatment period, although two non-responders experienced increases in SI intensity at follow-up; one of those with a pre-study suicide attempt reported preparatory/abortive suicidal behaviour at follow-up. The MADRS suicidality item did not exceed a score of 2 for any participant at follow-up (mean 0.3, SD 0.7). No participants dropped out of the study.

M. and colleagues interpret these findings as evidence that MDMA-AT produced sustained reductions in clinician-rated depression and in self-reported functional impairment, depression, insomnia, generalized anxiety and PTSD symptoms at seven months after baseline (four months after the end of treatment and six months after the last MDMA dose). Most participants who responded or remitted at post-treatment maintained those clinical gains at follow-up (8 of 9 responders/remitters), with a single relapse. The three initial non-responders showed no spontaneous improvement by follow-up. The absence of dropouts and the low relapse rate (8.3%) are highlighted in comparison with higher dropout and relapse rates reported in standard pharmacological and psychotherapeutic studies, although the authors note these comparisons are limited by differing study designs and follow-up durations. The authors relate their results to long-term follow-up data from MDMA-AT studies in PTSD, suggesting that durable benefits may extend across diagnoses and that trauma processing during MDMA sessions could represent a transdiagnostic mechanism. The observed improvement in insomnia is noted as clinically relevant given its role in MDD and PTSD. The authors also discuss the possibility that adding a third dosing session—an approach associated with higher response in PTSD trials—might increase efficacy and safety, as increases in suicidality were observed only in non-responders. Key limitations acknowledged by the authors include the small, demographically limited sample, the open-label design without a control group, and lack of formal blinding of outcome raters (raters were affiliated with the hospital rather than the study team but were not fully blinded), all of which can inflate estimates of effect and leave room for regression to the mean. The authors therefore recommend conducting randomised controlled trials to address these limitations and to more rigorously evaluate the durability, mechanisms and safety of MDMA-AT in MDD.