MDMA

Original Paper

The safety and efficacy of �3,4-methylenedioxymethamphetamine- assisted psychotherapy in subjects with chronic, treatment-resistant posttraumatic stress disorder: the first randomized controlled pilot study

Michael C Mithoefer1, Mark T Wagner2, Ann T Mithoefer1, Lisa Jerome3 and Rick Doblin3

Abstract Case reports indicate that psychiatrists administered �3,4-methylenedioxymethamphetamine (MDMA) as a catalyst to psychotherapy before recrea-

tional use of MDMA as ‘Ecstasy’ resulted in its criminalization in 1985. Over two decades later, this study is the first completed clinical trial evaluating

MDMA as a therapeutic adjunct. Twenty patients with chronic posttraumatic stress disorder, refractory to both psychotherapy and psychopharmacology,

were randomly assigned to psychotherapy with concomitant active drug (n¼12) or inactive placebo (n¼8) administered during two 8-h experimental

psychotherapy sessions. Both groups received preparatory and follow-up non-drug psychotherapy. The primary outcome measure was the Clinician-

Administered PTSD Scale, administered at baseline, 4 days after each experimental session, and 2 months after the second session. Neurocognitive

testing, blood pressure, and temperature monitoring were performed. After 2-month follow-up, placebo subjects were offered the option to re-enroll in

the experimental procedure with open-label MDMA. Decrease in Clinician-Administered PTSD Scale scores from baseline was significantly greater for the

group that received MDMA than for the placebo group at all three time points after baseline. The rate of clinical response was 10/12 (83%) in the active

treatment group versus 2/8 (25%) in the placebo group. There were no drug-related serious adverse events, adverse neurocognitive effects or clinically

significant blood pressure increases. MDMA-assisted psychotherapy can be administered to posttraumatic stress disorder patients without evidence of

harm, and it may be useful in patients refractory to other treatments.

Keywords combat disorders, MDMA, Posttraumatic stress disorder, psychedelics, PTSD

Introduction

Posttraumatic stress disorder (PTSD) is a debilitating anxiety disorder characterized by re-experiencing, hyperarousal and avoidance symptoms, and is a major worldwide public health

problem. The high incidence of PTSD and the limited effec- tiveness of existing treatments combine to create an urgent need for the development of new treatments. In the United

States, the lifetime prevalence of PTSD in the general popu- lation is between 6% and 10% (Kessler et al., 2005), and is much higher in countries where there is endemic armed con- flict (de Jong et al., 2003; Thabet and Vostanis, 1999; Weine

et al., 1995). In US soldiers returning from service in Iraq and/or Afghanistan, the incidence of PTSD is as high as 18% (Hoge et al., 2004), and it is estimated that those with

PTSD will number between 75,000 and 225,000 (Tanielian and Jaycox, 2008). PTSD is typically a chronic illness (Breslau and Davis, 1992; Kessler et al., 2005) associated

with high rates of psychiatric and medical comorbidity,

disability, suffering, drug abuse, and suicide (Breslau, 2001; Cohen et al., 2009; Frayne et al., 2004; Kessler et al., 2005; Perkonigg et al., 2000).

Existing treatments for PTSD include both pharmacother- apy and psychotherapies. Although a variety of drugs are used to treat symptoms of PTSD, they have limited efficacy.

1Private Practice of Psychiatry and Clinical Research, Mount Pleasant SC,

USA. 2 Medical University of South Carolina, Charleston SC, USA.

3 Multidisciplinary Association for Psychedelic Studies, Santa Cruz, CA,

USA.

Corresponding author: Michael C Mithoefer, MD, 208 Scott Street, Mount Pleasant, SC 29464, USA

Email: [email protected]

Journal of Psychopharmacology

0(0) 1–14

! The Author(s) 2010

Reprints and permissions:

sagepub.co.uk/journalsPermissions.nav

DOI: 10.1177/0269881110378371

jop.sagepub.com

To date, the selective serotonin reuptake inhibitors (SSRIs) sertraline and paroxetine are the only two drugs approved by the Food and Drug Administration (FDA) for this indication

(Brady et al., 2000; Marshall et al., 2001). Nine clinical trials of SSRIs for PTSD published between 1994 and 2007 dem- onstrated a mean between-group effect size (Cohen’s d) of 0.5 for drug effect compared with placebo (Foa et al.,

2009). After identifying 22 individual drugs in seven different drug classes, a 2008 review of PTSD treatment studies by the Institute of Medicine was inconclusive regarding evidence for

use of any of the drugs studied (Committee on Treatment of Posttraumatic Stress Disorder, 2008). Other recent reviews of the literature have reached more favorable conclusions about

pharmacotherapy. Several meta-analyses found that, while some studies did not show a significant drug effect, in general the response rate to pharmacotherapy was 20–22% greater

than the response to placebo, and in SSRI trials 30% of sub- jects can achieve complete remission at 12 weeks. All these reviews have emphasized the need for further research into more effective agents for PTSD (Foa et al., 2009; Ipser et al.,

2006; Stein et al., 2009; Ursano et al., 2004). The most widely recognized methods of psychotherapy

for PTSD are cognitive behavioral therapy (particularly

Prolonged Exposure and Cognitive Processing Therapy), Eye Movement Desensitization and Reprocessing (EMDR), and psychodynamic psychotherapy (Foa et al., 2009; Ursano

et al., 2004). A recent meta-analysis of direct comparison studies of ‘bona fide psychotherapies’ concluded that there is no statistically significant difference in between-group effect size (Benish et al., 2008). A review of cognitive behav-

ioral therapy for PTSD listed 107 studies (Foa et al., 2009). Of the 36 studies using the Clinician-Administered PTSD Scale (CAPS) as an outcome measure and intent-to-treat analysis,

the mean effect size was 0.88 when compared against wait-list controls (Foa et al., 2009). In clinical trials of psychotherapy for PTSD, the dropout rate is typically 20–30%, and the

response rate is between 60% and 95% among subjects who receive active treatment and complete the trials (Cloitre, 2009; Hembree et al., 2003; Rothbaum et al., 2006). Data

about combined psychotherapy and pharmacotherapy are limited to a few small studies with mixed results (Foa et al., 2009; Davis et al., 2006; Rothbaum et al., 2006). Overall, the evidence for pharmacotherapy and psychotherapies in the

above studies indicates that existing therapies for PTSD are ineffective for between 25% and 50% of patients who enroll in clinical trials. An effective treatment that could reduce the

substantial treatment failure rates associated with existing PTSD treatments is needed.

Methylenedioxymethamphetamine (MDMA) is a ring-

substituted phenylisopropylamine derivative with a unique profile of psychopharmacological effects. MDMA was pat- ented in 1914 by the German chemical and pharmaceutical company Merck KGaA as an intermediate compound in the

synthesis of other drugs (Benzenhoefer and Passie, 2006; Freudenmann et al., 2006). Before MDMA was classified in the United States as a Schedule I controlled substance in

1985, a number of psychiatrists and other therapists in the United States and Europe used MDMA as an adjunct to psychotherapy. MDMA was reported to decrease feelings

of fear while maintaining a clear-headed, alert state of

consciousness (Greer and Tolbert, 1998). More recently, Phase I clinical trials have demonstrated that MDMA can be administered without evidence of harm to pre-screened

subjects, and induces a 2–4-h experience typically character- ized by euphoria, increased well being, sociability, self-confi- dence, and extroversion (Cami et al., 2000; Harris et al., 2002; Kolbrich et al., 2008; Liechti et al., 2001; Tancer and

Johanson, 2001; Vollenweider et al., 1998). To test the poten- tial efficacy of the clinical use of MDMA, we present the first rigorous data on its therapeutic application in this pilot Phase

II clinical trial. The decreased fear response induced by MDMA adminis-

tration may be useful in the treatment of PTSD, a condition

that involves exaggerated and uncontrolled fear responses. Many psychotherapies for PTSD involve the induction and extinction of these abnormal autonomic responses through

revisiting traumatic experiences in psychotherapy with an appropriate level of emotional engagement (Foa et al., 2009). Frequently, however, treatment may be ineffective when patients are unable to tolerate feelings associated

with revisiting the trauma, or when emotional numbing during exposure to traumatic memories precludes a level of engagement necessary for extinction (Jaycox and Foa, 1999).

Therefore, if a drug could temporarily reduce fear and increase interpersonal trust, without clouding the sensorium or inhibiting access to emotions, it might prove an effective

catalyst to psychotherapy for PTSD. The use of drugs to catalyze psychotherapy has been discussed in the psychiatric literature since the 1940s and has included the use of barbi- turates, amphetamines, nitrous oxide, LSD, and others

(Sargent et al., 1972). This report contains findings from the first completed pilot study designed to explore the possibility that MDMA could serve as such a catalyst. The hypothesis

tested is that MDMA could be administered without harm to people with chronic, treatment-resistant PTSD and, in con- junction with psychotherapy, would lead to a significant

decrease in PTSD symptoms compared with the same psycho- therapy in conjunction with inactive placebo.

Possible mechanisms

Several possible mechanism of action for MDMA-assisted psychotherapy can be postulated. Learning theory, emotional

processing theory and social-cognitive theories aimed at explaining the therapeutic effects of exposure therapy have been summarized by Foa et al. (Foa et al., 2009). To be effec-

tive, exposure must be accompanied by a degree of emotional engagement or ‘fear activation’ while avoiding dissociation or overwhelming emotion (Foa et al., 2007). This has been

referred to as working within the ‘optimal arousal zone’ or ‘window of tolerance’ (Ogden and Pain, 2006; Siegel, 1999; Wilbarger and Wilbarger, 1997). Patients with PTSD are prone to extremes of emotional numbing or overwhelming

anxiety, and often have a narrow window between thresh- olds of under and over-arousal (Ogden and Pain, 2006). MDMA may exert its therapeutic effect by widening this

window. If MDMA allows patients to stay emotionally engaged without being overwhelmed by anxiety while revisit- ing traumatic experiences, it may thereby catalyze effective

exposure therapy.

2 Journal of Psychopharmacology 0(0)

The pharmacological effects of MDMA include serotonin release, 5HT2 receptor stimulation, and increase in levels of the neurohormones oxytocin, prolactin and cortisol (Dumont

et al., 2009; Grob et al., 1996; Harris et al., 2002; Mas et al., 1999; Thompson et al., 2007; Wolff et al., 2006). Serotonin release plays an important role in producing the subjective effects of MDMA (Farre et al., 2007; Liechti et al., 2000;

Liechti and Vollenweider, 2000; Tancer and Johanson, 2007). Pretreatment with SSRIs reduces most acute subjective and physiological effects of MDMA, including effects on

mood and perception. Serotonin release directly or indirectly leads to an elevation in oxytocin, possibly by stimulating 5HT1A receptors (Baggott et al., 2008; Dumont et al., 2009;

Thompson et al., 2007). Recent findings suggest that oxytocin is involved in affiliation, trust and accurate perception of emotion (Domes et al., 2007; Kirsch et al., 2005; Zak et al.,

2005), so elevated oxytocin might help participants form a therapeutic alliance and revisit traumatic experiences in an emotionally engaged state. In human volunteers, oxytocin reduces activation of the amygdala in response to fear-indu-

cing stimuli (Kirsch et al., 2005) and increases trust (Baumgartner et al., 2008; Kosfeld et al., 2005). A recent study reported that elevation in oxytocin after MDMA was

associated with greater sociability and gregariousness (Dumont et al., 2009). It has been postulated that prolactin release following MDMA administration may contribute to

a post-orgasmic-like sense of relaxation and receptivity (Passie et al., 2005). The extent to which each of these phar- macological mechanisms may play a role in possible thera- peutic effects of MDMA is speculative.

The ‘neurocircuitry model’ of PTSD postulates a deficit in extinction of fear conditioning mediated by the amygdala and the ventral/medial prefrontal cortex (vmPFC) (Rauch et al.,

2006), a model supported by findings of reduced hippocampal activity and volume, increased activity in the amygdala and decreased activation of the medial prefrontal cortex in people

with PTSD (Rauch et al., 2006). A human Positron Emission Tomography (PET) study 75 min following MDMA admin- istration has shown increases in cerebral blood flow in the

ventromedial frontal and occipital cortex, and decreases in the left amygdala (Gamma et al., 2000). MDMA may pro- duce some of its effects through these acute changes in brain activity, possibly reversing abnormalities known to be associ-

ated with PTSD and thereby allowing for effective processing of traumatic material during the therapy sessions.

Methods and materials

Subjects

Subjects were recruited via letters to psychotherapists and internet advertisements. Potential subjects aged 21–70 years were screened using a scripted telephone interview to identify

previously diagnosed medical or psychiatric exclusion criteria. Candidates who passed telephone screening and gave informed consent were evaluated in an outpatient

office by an independent rater and a physician. Of these, 20 met all enrollment criteria and were recruited for the study, with replacement of two dropouts (Figure 1). For

this initial pilot study, a minimum sample size of eight per

group with oversampling for the experimental group was determined adequate to produce useful estimates of effect size. The study was approved by the Copernicus Group

Independent Review Board (IRB), Research Triangle Park, NC, USA, and was conducted according to their regulatory guidance for protection of human subjects and relevant Federal regulations and international standards. Written

informed consent was obtained from each subject by the investigators. Capacity to give informed consent was assessed clinically and with a written quiz testing understanding of

the consent document. Enrollment began in March 2004 and ended in January 2008. Follow-up was completed in September 2008.

MDMA

The MDMA used was from a supply produced by David E Nichols, PhD, for the sponsor, which holds an FDA Drug Master File and Investigational New Drug permit (IND) for the study drug.

Design

After enrollment, subjects were randomized, in double-blind fashion, to receive two experimental sessions of either psychotherapy with concomitant MDMA administration

or the same psychotherapy accompanied by inactive placebo (lactose) administration (psychotherapy-only). The blind was broken for each subject after the follow-up visit 2 months after the second experimental session. All subjects who ini-

tially received placebo were offered participation in an open- label crossover segment (‘Stage 2’) (Figure 2). After the 2-month follow-up, nine subjects were given a third session

of MDMA with psychotherapy, as allowed in a protocol amendment. However, because not all subjects received a third session and placebo subjects received only two sessions,

data related to the third session were omitted from analysis, and for simplicity are omitted from Figure 2.

Subjects were required to taper and abstain from all psy-

chotropic medication during study participation except seda- tive hypnotics or anxiolytics used as-needed between MDMA or placebo sessions (referred to as ‘rescue medications’). After preliminary evidence of safety and efficacy had been estab-

lished, a protocol amendment was approved allowing the last nine subjects to receive a supplemental dose of MDMA or placebo in all experimental sessions. The purpose of this sup-

plemental dose, half the initial dose administered 2 h after- wards, was to prolong the therapeutic window of MDMA effects and gather pilot data about dose for design of future

clinical trials.

Assessments

Study entry screening consisted of a Structured Clinical Interview for Axis I Diagnosis (SCID) (First et al., 1994), the SCID II for personality disorder (First et al., 1997),

CAPS, medical history, physical examination, serum chemis- try profile, complete blood count, thyroid-stimulation hor- mone (TSH), free thyroxine, HIV serology, urinalysis, and

electrocardiogram (ECG). Subjects were required to meet

Mithoefer et al. 3

DSM-IV-R criteria for the diagnosis of crime or war-related chronic PTSD, and to have treatment-resistant symptoms, defined as a CAPS score of �50 (signifying moderate

to severe symptoms) following at least 3 months of prior SSRI or serotonin–norepinephrine reuptake inhibitor (SNRI) treatment in addition to at least 6 months of psycho-

therapy. Exclusion criteria required freedom from any major medical conditions. In addition, psychiatric exclusion criteria included Borderline Personality Disorder or any current Axis I disorder with the following exceptions which were allowed:

anxiety disorders, affective disorders other than bipolar dis- order type 1, substance abuse or dependence in remission for �60 days, and eating disorder without active purging. Urine

drug testing for cocaine, marijuana, amphetamines, MDMA, opiates and benzodiazepines was performed during initial screening and immediately before each experimental session.

All screens were negative in all subjects.

Outcome measures

Primary. The CAPS is a widely used structured interview for quantifying PTSD symptoms that has excellent psychometric properties of reliability and validity (Weathers et al., 2001).

This measure produces a global symptom severity score as well as a categorical ranking as to whether or not a subject meets DSM-IV-R criteria for PTSD.

Secondary. Impact of Events Scale-Revised (IES-R) is a widely used self-report measure that assesses psychological

response to stress. It was revised to parallel the DSM-IV cri- teria for PTSD and found to have acceptable psychometric properties (Weiss, 1997). The Symptom Checklist 90-Revised

(SCL-90-R) is a self-report of symptoms covering a wide

Assessed for eligibility (n = 134)

Excluded (n = 112)

Not meeting inclusion criteria at telephone screen (n = 61) Declined to participate (n = 46)

Not meeting inclusion criteria at full screen (n = 5)

Randomized (n = 23)

Allocated to two sessions MDMA + 11 non-drug therapy (n = 15)

Received allocated intervention (n = 12) Dropped out (n = 2)

relapse of depression n = 1 unwilling to travel n = 1

1 = not treatment resistant

Supplemental dose 2–2.5h after 1st dose (n = 4) (protocol amendment)

A ll

o c

a ti

o n

Allocated to placebo two sessions + 11 non-drug therapy (n = 8)

Received allocated intervention (n = 8)

Drop-outs (n = 0)

Elected to enter open-label arm (n = 7)

O p

e n

-L a

b e

l

Allocated to one additional open-label MDMA-assisted sessions + three non-drug sessions (n = 5) (protocol amendment) Discontinued intervention (n = 0)

Open-label arm Allocated to two sessions MDMA + 11 non- drug (n = 7)

Allocated to one additional MDMA + three non-drug sessions (n = 4) (protocol amendment)

Supplemental dose 2–2.5h after 1st dose (n = 4) (protocol amendment)

Analyzed (n = 12)

Excluded from analysis (n = 0)

Analyzed (n = 8)

Excluded from analysis (n = 0)A

n a

ly s

is E

n ro

ll m

e n

t

Figure 1. Flow Diagram.

4 Journal of Psychopharmacology 0(0)

range of psychiatric categories and yielding nine symptom scales as well as global indices (Derogatis, 1994).

Neurocognitive measures

The Repeatable Battery for the Assessment of Neuropsychological Status (RBANS) tests attention and

processing speed, expressive language, visual-spatial and constructional abilities, and memory (Randolph et al., 1998). The Paced Auditory Serial Addition Task (PASAT)

assesses information processing speed and mental flexibility (Gronwall, 1977). The Rey-Osterrieth Complex Figure (RCFT) assesses visuospatial memory associated with a vari-

ety of neurological conditions that alter effective brain func- tion (Mirushina et al., 1999).

Physiological measures

Blood pressure, pulse, and temperature were monitored during all experimental sessions.

Randomization and blind – Stage 1

Subjects were randomized to either MDMA or placebo, in conjunction with psychotherapy, in a ratio of 60% MDMA (n¼12) to 40% placebo (n¼8), replacing drop-

outs to preserve the group-assignment ratio and double- blinding. On the day of each subject’s first experimental session, an independent randomization monitor assigned a bottle containing either MDMA or placebo, deter-

mined by a computer-generated randomization list. The subjects, investigators and the independent rater who administered the outcome measures were all blinded. The

independent rater did not have access to records of treat- ment sessions.

After screening and enrollment, participants completed

baseline psychological and neurocognitive measures. Outcome measures were repeated approximately 4 days after each of two 8-h experimental sessions and 2 months

after the second experimental session. Participants com- pleted neurocognitive measures at baseline and again 2 months after the second experimental session (Table 1).

Stage 2

Placebo subjects who enrolled in the open-label arm com-

pleted outcome measures 2 months after their last experimen- tal session. The schedule of visits in Stage 2 was nearly identical to the schedule in Stage 1 (Table 1).

Therapeutic intervention

The study was conducted in a comfortable, aesthetically

pleasing outpatient office with facilities for the subject and a blinded consultant nurse to remain at the site overnight. The investigators maintained equipment and drugs for treat-

ment of medical emergencies, and an emergency physi- cian and nurse were on site for all experimental sessions. The experimental sessions lasted 8–10 h, followed by an over-

night stay. One of the investigators maintained daily

Study visit

Stage l - double blind

Informed consent and initial screening

SCID, CAPS, IES, SCL-90-R Neuropsychological measures

Physical exam, blood tests, EKG

Two 90 minute introductory sessions with therapists

All day MDMA or placebo experimental session

Spending night in clinic with nurse on duty

Daily phone contact for a week

Repeat CAPS, IES & SCL-90-R 4 days after

experimental session

Repeat CAPS, IES & SCL-90-R and neuropsychological

measures 2 months after 2nd experimental session

Blind broken

For Stage ll repeat CAPS, IES & SCL-90-R 2 months after 2nd

open lable MDMA session

*Subjects who received placebo in stage l may elect to repeat above cycles for 2 open lable MDMA sessions

Repeat the above cycle for the second

experimental session

There 90 minute non-drug therapy sessions for

integration following each experimental session approximately weekly

Stage ll* (open lablel cross-over)

90 min session with therapists next morning

Figure 2. Study Visits.

Mithoefer et al. 5

telephone contact with the subjects during the week after

experimental sessions. In addition to experimental sessions of MDMA or placebo

during psychotherapy, the therapeutic intervention included

non-drug psychotherapy sessions for preparation and integra- tion. A male and female co-therapist team, one a psychiatrist, the other a psychiatric nurse, was present for all sessions.

Each subject had two 90-min introductory sessions within 6 weeks before the first experimental session to prepare them for the structure of the sessions, the approach to therapy

and possible effects of MDMA. In Stage 1, there were eight integration sessions focused on discussing the experimental sessions, additional emotional processing (Foa and Kozak, 1986) if necessary, and helping subjects incorporate any

insights or new perspectives into their daily lives. One of these integration sessions occurred the morning after each of the two experimental sessions, and three more were sched-

uled during the month following each experimental session. Additional integration sessions were permitted if needed. A final integration session occurred 2 months after the second

experimental session. In Stage 2, the schedule of integration sessions was the same as in Stage 1 except that there were three scheduled follow-up sessions after each experimental session instead of four.

The method of psychotherapy followed principles developed by Stanislav Grof, MD and others for LSD psy- chotherapy (Pahnke et al., 1971) and Holotropic Breathwork

(Grof, 2000), and adapted for MDMA-assisted psychother- apy by Metzner and others (Greer and Tolbert, 1998; Metzner and Adamson, 2001). These methods were further

modified by the investigators for application to PTSD

treatment, and the psychotherapy technique was manualized

to the extent possible prior to this initial pilot study. This draft treatment manual was written by the sponsor and inves- tigators before the study began, and is being refined and

operationalized with quantitative adherence measures based on this pilot study. It is available at http://www.maps.org/ mdma/

During experimental sessions, the subject sat or reclined on a futon bed and the co-therapists sat in chairs on either side. The first dose of MDMA (125 mg) or placebo was

given in a capsule by mouth at 10 a.m. Subjects then rested in a comfortable position with eyes closed or wearing eyeshades, and listened to a program of music that was ini- tially relaxing and later emotionally evocative. The pro-

grams of music used were identical for all subjects, though subjects or investigators could choose to skip over some selections or to replace them with periods of silence.

Throughout the experimental sessions, periods of conversa- tion alternated with periods during which subjects were encouraged to focus on introspection. The schedule for

these alternating periods was flexible and determined by the desires of the subject and the judgment of the therapists. The therapists sought an approximately equal balance between quiet introspection and therapeutic discussion, but

the actual ratio varied among individuals and between sessions.

The optional supplemental dose of 62.5 mg MDMA or

placebo was administered 2–2.5 h after the initial dose if the investigators judged it to be safe and advisable and the subject agreed to it. The supplemental dose was administered

in 22 of the 23 sessions in which it was an option. The

Table 1. Study Visits

Stage 1 Visit number 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 SCID CAPS

X

Medical evaluation

X

CAPS IES SCL-90 RBANS Rey PASAT Preparatory psychotherapy MDMA or placebo Non-drug psychotherapy

Stage 2, Open-label cross-over for placebo subjects Visit number 17.1 2–

18 2– 19

2– 20

2– 21

2– 22

2– 23

2– 24

2– 25

2–26

CAPS IES SCL-90

X**

Non-drug psychotherapy

Non-drug psychotherapy sessions occurred 1–7 days apart.*3–7 days after prior MDMA or placebo session. Grey bars indicate experimental sessions with MDMA or placebo. Visit 5 occurred within a month after visit 1. Visit 11 occurred 3–5 weeks after visit 5. Visit 17 occurred 2 months after visit 11. Non-drug psychotherapy sessions occurred 1–7 days apart. ** Symptom measures repeated only if visit 18 occurred > 30 days after the final symptom measures in the double-blind protocol, visit 17. ***2 months after second MDMA session. Visit 2-22 occurred 3–5 weeks after Visit 2-18.

MDMA

X

X X X X

X X X X

X

X

X

X X X X

X

X X X X

X

X

X X*

X*

X*

X***

6 Journal of Psychopharmacology 0(0)

therapists stayed with the subject until at least 5 p.m. or until the physical and psychological effects of the session had sub-

stantially subsided and the subject was judged to be in stable condition and at baseline mental status. If necessary, zolpi- dem or lorazepam were prescribed for insomnia. During

psychotherapy on the following day, participants and inves- tigators were asked to guess condition assignment and rate the certainty of their guess.

Statistical analysis

Per-protocol analysis with ANOVA with repeated measures

was conducted to test CAPS, IES-R and SCL-90-R scores for differences between groups over time. The Holm’s sequential Bonferroni correction for multiplicity was employed after a

significant ANOVA for the four post-hoc tests for differences between groups.

Results

Table 2 shows participant characteristics. Fifteen of the 20

subjects had previously undergone multiple medication trials (mean 4.2 different psychiatric drugs) and 15 had com- pleted more than one course of psychotherapy. Average dura-

tionofPTSD wasestimatedat19þyears.Atbaseline therewere

no significant differences between groups with the exception of duration of previous therapy. Analysis of co-variance

with repeated measures was conducted for CAPS using number of months of previous therapy as covariate. The covar- iate proved non-significant (CAPS: F¼0.35, p¼0.56). Index

traumas for participants are shown in Table 3. Two subjects, not included in Tables 2 or 3, dropped out before the second experimental session. One did so because she required resump- tion of medication for relapse of depression 42 days after her

one MDMA-assisted session. The other withdrew because he found travel to the study site problematic due to limits on reimbursement of travel expenses set by the IRB, which

were increased at the sponsor’s request after this subject dropped out.

Physiological response and side effects

Onset of MDMA effects occurred 45–75 min after the initial dose. The effects reached a peak at 2–2.5 h and lasted 4–5 h in

the 11 subjects who received a single dose, and 5–6 h in the nine who received a supplemental dose. Effects diminished gradually over several hours. Elevations of blood pressure,

pulse, and body temperature were greater in the MDMA group, and spontaneously returned to baseline at session end in both groups (Table 4). There were no resulting medical

complications or pharmacologic interventions.

Table 2. Participant characteristics

CHARACTERISTIC All Subjects n¼20 MDMA group n¼12 Placebo Group n¼8 p

Mean age (std) 40.4 (7.2) 40.2 (7.6) 40.8 (7.0) 0.79

Gender 0.66

Male 3 15% 2 17% 1 13%

Female 17 85% 10 83% 7 87%

Caucasian 20 100% 12 100% 8 100%

Marital status 0.27

Never married 7 35% 6 50% 1 12.5%

Divorced/separated 2 10% 1 8% 1 12.5%

Widowed 1 5% 0 1 12.5%

Married/living with partner 10 50% 5 42% 5 62.5%

On disability for PTSD 3 15% 2 17% 1 13% 0.85

Hx alcohol abuse/dependency 2 10% 1 8% 1 13% 0.63

Hx other substance abuse/dependency 1 5% 0 1 12% 0.62

Prior MDMA use 9 43% 6 46% 3 38% 0.53

Lifetime MDMA use (number of times) for prior MDMA users 9 (1–5) 6 (1–5) 3 (1–2) 0.32

PTSD mean number of months of duration (std) 248 (173) 232 (201) 273 (126) 0.57

PTSD Crime-related 19 95% 11 85% 8 100% 0.60

PTSD War-related 1 5% 1 8% 0 0.60

Comorbid major depression 16 80% 9* 75% 7** 88% 0.47

Comorbid anxiety disorder 3 15% 2*** 17% 1**** 13% 0.65

Mean # months of prior therapy (std) 58.5 (49.5) 40.6 (38.5) 85.3 (54.2) 0.04

Mean Baseline CAPS score (std) 79.4 (22.4) 79.2 (23.6) 79.6 (22.0) 0.97

Bivariate statistical tests were conducted on all baseline measurements and subject characteristics to determine if significant differences existed initially between the

placebo and MDMA treatment groups. All tests were non-significant except number of months subjects had of previous therapy.

* two in remission

** three in remission

*** one panic disorder in remission, one current generalized anxiety disorder and simple phobia

**** current panic disorder and obsessive compulsive disorder, in remission

Mithoefer et al. 7

Spontaneously reported side effects occurring within 7 days of MDMA or placebo administration are shown in Table 5. The most common side effects that occurred more frequently in

the MDMA group on the day of experimental sessions were: jaw tightness, nausea, feeling cold, dizziness, loss of appetite, and impaired balance. Equally or more common in the placebo group on the day of experimental sessions were: anxiety, insom-

nia, headache and fatigue. In the week following experimental sessions, some of the most common side effects were reported at similar incidence by both groups: fatigue, anxiety, low mood,

headache and nausea, with anxiety being slightly more frequent in the MDMA group and low mood slightly more frequent in the placebo group. During this week, irritability and loss of

appetite were more frequently reported in the MDMA group and insomnia was reported more often in the placebo group. Side effects typically resolved over a period of hours or days,

usually spontaneously; sometimes with short-term

symptomatic treatment such as sedative hypnotics or non-ster- oidal anti-inflammatory drugs following experimental sessions. No medical treatment was required during any exper-

imental sessions. No serious drug-related adverse effects occurred.

Symptom levels Stage 1 – double blind

Figure 3 illustrates that PTSD symptoms, as measured by CAPS, improved over time in both groups (Time:

F(3, 17)¼40.292, p<0.0005), but the MDMA group showed significantly greater improvement (Time*Group interaction F(1, 17)¼7.173, p¼0.015). Mean differences

between ‘group� time’ were examined using independent t-tests with Holm’s sequential Bonferroni correction (a) for multiplicity. Statistical significance (*) is indicated where

p<a: Time 1 p¼0.966, a¼0.050, Time 2* p¼0.013, a¼0.017, Time 3* p¼0.002, a¼0.012, Time 4* p¼0.013, a¼0.025. Similar results were found for the IES-R, shown in Figure 4. PTSD symptoms, as measured by IES-R,

improved over time in both groups (Time: F(3, 17)¼11.003, p<0.0005), but the MDMA group showed signifi- cantly greater improvement (Time*Group interaction

F(1, 17)¼3.290, p¼0.027). Mean differences between ‘group- � time’ were examined using independent t-tests with Holm’s sequential Bonferroni correction (a) for multiplicity. Statistically significance (*) is indicated where p<a: Time 1 p¼0.976, a¼0.050, Time 2* p¼0.016, a¼0.017, Time 3* p¼0.006, a¼0.012, Time 4 p¼0.038, a¼0.025. The CAPS scores of the two subjects who dropped out, both of whom

were randomized to receive MDMA, fell from 110 and 107, respectively, at baseline, to 17 and 27, respectively, 4 days after their only MDMA-assisted psychotherapy session.

These data are not included in the analysis.

Table 4 Physiologic data: All experimental sessions

MDMA Placebo

Systolic BP Mean (St. Dev.) Range Mean (St. Dev). Range

Baseline 117.9 (15.0) 97/149 114.2 (11.6) 92/140

Maximum 143.1 (16.3) 117/179 136.4 (18.8) 102/176

Max Change 25.2 (13.7) 16/50 22.2 (15.3) �3/57

Diastolic BP

Baseline 74.6 (9.4) 56/92 74.4 (8.1) 58/89

Maximum 89.0 (9.2) 74/113 88.2 (10.5) 65/106

Max Change 14.4 (6.5) 1/14 13.8 (7.5) 2/28

Pulse

Baseline 74.3 (11.7) 54/98 69.9 (10.9) 5/92

Maximum 103.1 (15.9) 67/135 92.1 (17.8) 68/141

Max Change 28.8 (11.5) �7/52 22.2 (13.7) 3/57

Temperature

Baseline 36.6 (0.5) 35.6/37.8 36.3 (0.6) 35/37.2

Maximum 37.1 (0.3) 36.7/37.8 37.1(0.4) 36.7/37.8

Max Change 0.5 (0.4) �0.6/1.5 �0.7 (0.6) �0.06/1.9

Group comparisons of vital signs were tested for change pre-session (15 min prior) to highest recorded and pre-session to post-session (6 h post) using t tests. There was a

significantly greater increase in all physiologic measures from pre-session to highest recorded value during experimental sessions for the MDMA group than for the placebo

group (p < 0.05). There were no significant differences when comparing change from pre-session to post-session (p > 0.05). All values returned to pre-session norms by 6 h

after session completion.

Table 3. Index traumas for study participants

Index trauma MDMA (n¼12) Placebo (n¼8) p

Sexual assault 5 (42%) 3 (38%) p > 0.95

Childhood abuse

Sexual 4 (33%) 4 (50%) p > 0.95

Physical, neglect 2 (16%) 0 p¼0.35

Emotional 1 0 p > 0.95

Family violence, other 1 0 p > 0.95

Violence (as stabbing) 0 1 p > 0.95

Combat stress 1 0 p > 0.95

8 Journal of Psychopharmacology 0(0)

Symptom levels Stage 2 – open-label crossover

Seven of the eight placebo subjects chose to enroll in the

crossover arm. One of the placebo responders, whose CAPS score had fallen from 67 to 15 after placebo, elected not to enroll in the crossover because she was satisfied with her improvement. The other placebo responder had a transient

decline in CAPS from 54 to 15 after placebo, but her CAPS increased to 64 in the 3 months prior to enrollment in Stage 2. Paired t-tests were used to analyze change in outcome mea-

sures from Time 1c to Time 4c in the crossover arm. For the seven placebo subjects who completed the open-label cross- over, there were significant decreases in CAPS (Table 6) and

IES-R scores (mean IES-R decrease¼15.9, SD¼12.1,

p¼0.013) from end of the control trial to 4–6 weeks after two MDMA sessions were completed. These decreases are similar in magnitude to the CAPS and IES-R decreases in

the subjects initially randomized to full-dose MDMA.

Neurocognitive measures

At baseline, there were no significant group differences on any of the cognitive measures including the RBANS total score (t¼1.78, p¼0.09), PASAT Trial 1 (t¼0.95, p¼0.35),

PASAT Trial 2 (t¼�0.16, p¼0.88) and the Rey-Osterrieth Figure 30-minute delay (t¼0.06, p¼0.95). To test whether the experimental condition had an adverse impact on cogni-

tion, between-group comparisons were performed at the

Table 5. Side effects: Number of instances of spontaneously reported side effects in association with both experimental sessions in Stage 1, with

instances a function of number of subjects and experimental sessions in which each side effect occurred during the session or within 7 days following

Day of MDMA

sessions (2)

Day of placebo

sessions (2)

Within 7 days

after MDMA sessions

Within 7 days after

placebo session

N (%) N (%) N (%) N (%)

Subjects¼12 Subjects¼8 Subjects¼12 Subjects¼8

Sessions N¼24 Sessions N¼16 Sessions N¼24 Sessions N¼16

Anxiety 14 (58%) 13 (81%) 13 (54%) 7 (44%)

Decreased concentration 3 (13%) 1 (6%) 6 (25%) 0

Derealization, detachment 0 0 1 (4%) 0

Diarrhea 0 0 3 (13%) 0

Dizziness 9 (38%) 2 (13%) 3 (13%) 1 (6%)

Drowsiness 2 (8%) 3 (19%) 0 2 (13%)

Dry mouth 4 (17%) 1 (6%) 1 (4%) 0

Fatigue 11 (46%) 8 (50%) 18 (75%) 12 (75%)

Feeling cold 10 (42%) 3 (19%) 1 (4%) 0

General infection 1 (4%) 0 0 2 (13%)

Headache 14 (58%) 9 (56%) 6 (25%) 4 (25%)

Heavy legs 2 (8%) 0 1 (4%) 0

Impaired balance 6 (25%) 0 0 0

Insomnia 13 (54%) 10 (63%) 9 (38%) 9 (56%)

Irritable 2 (8%) 3 (19%) 8 (33%) 3 (19%)

Loss of appetite 8 (33%) 1 (6%) 9 (38%) 0

Low mood 4 (17%) 2 (13%) 10 (42%) 8 (50%)

Muscle tension 4 (17%) 2 (13%) 2 (8%) 1 (6%)

Nausea 12 (50%) 2 (13%) 7 (29%) 4 (25%)

Need more sleep 0 1 (6%) 5 (21%) 2 (13%)

Nystagmus 1 (4%) 0 0 0

Pain 1(4%) 4 (25%) 1 (4%) 1 (6%)

Panic, re-experiencing 0 0 1 (4%) 1 (6%)

Parasthesias 2 (8%) 0 0 0

Perspiration 4 (17%) 1 (6%) 0 1 (6%)

Restless 5 (21%) 2 (13%) 1(4%) 0

Rumination 1 (4%) 1 (6%) 3 (13%) 2 (13%)

Somatic sensations 1 (4%) 1 (4%)

Thirst 2 (8%) 1 (6%) 0 0

Tight jaw 19 (79%) 3 (19%) 6 (25%) 2 (13%)

Upper GI burning 1 (4%) 0 0 0

Upper respiratory infection 0 0 2 (8%) 2 (13%)

Visual disturbance 1(4%) 0 2 (8%) 0

Weakness 1 (4%) 1 (6%) 6 (25%) 0

Mithoefer et al. 9

study follow-up 2 months after the second experimental ses-

sion. There were no significant group differences on any of the major index scores. This is best captured by the RBANS total score (t¼1.05, p¼0.31), PASAT Trial 1 (t¼0.35, p¼0.19),

PASAT Trial 2 (t¼0.41, p¼0.69) and the Rey-Osterrieth 30-minute delay score (t¼0.98, p¼0.35).

Clinical response

Clinical response was defined as >30% reduction from base- line in CAPS total severity score. In Stage 1, the clinical

response was 83.3% (10/12) in the MDMA group versus 25% (2/8) in the placebo group. Likewise, 10 of the MDMA group no longer met DSM-IV criteria for PTSD

compared with two of the placebo group. In Stage 2, the clinical response rate was 100% in the seven subjects, six of whom had failed to respond to placebo and one of whom had relapsed after an initial placebo response. An unplanned

observation was that all three subjects who reported being unable to work due to PTSD were able to return to work.

Additional psychotherapy sessions

As allowed in the protocol, additional psychotherapy sessions

occurred when the investigators judged them to be necessary

to support integration in subjects who experienced anxiety or other difficulties following experimental sessions. Only one additional session was conducted following placebo sessions, whereas 20 such sessions were provided to seven of 13 sub-

jects following MDMA-assisted sessions. The data does not allow meaningful statistical characterization of the relation- ship between extra visits and changes in symptom measures;

for one to three extra sessions there was a trend toward cor- relation with improved CAPS scores, but any increase beyond three extra sessions was inversely related to improvement in

CAPS scores.

80

70

60

50

40

M e

a n

C A

P S

30

Group

Placebo M = 69.9 SE = 6.9 MDMA M= 43.0 SE = 5.7

20

10

0

1 2 Time

Time 1: Baseline < 4 weeks before first experimental session and after discotinuing any psychotropic medications

Placebo = 74.1 (10.3), MDMA = 37.8 (8.4)

Placebo = 59.1 (9.4), MDMA = 25.5 (7.7)

Placebo = 79.6 (8.1), MDMA = 79.2 (6.6)

Placebo = 66.8 (8.0), MDMA = 29.3 (6.5)

Time 2: 3-5 days after first experimental session

Time 3: 3-5 days after second experimental session

Time 4: 2 months after second experimental session

3 4

Figure 3. CAPS Mean Scores by Group for Time 1-Time 4.

50

40

M e

a n

I E

S -R 30

Group Placebo M = 40.5, SE = 4.4 MDMA M= 25.4, SE = 3.6

20

10

0

1 2

Time

Time 1: Baseline < 4 weeks before first experimental session and after discontinuing any psychotropic medications

Placebo = 46.0 (6.5), MDMA = 23.7 (5.3)

Placebo = 32.6 (6.1), MDMA = 15.0 (5.0)

Placebo = 45.1 (5.3), MDMA = 44.9 (4.3)

Placebo = 38.28 (5.0), MDMA = 18.2 (4.1)

Time 2: 3-5 days after first experimental session

Time 3: 3-5 days after second experimental session

Time 4: 2 months after second experimental session

3 4

Figure 4. IES-R Mean Scores by Group for Time 1-Time 4.

Table 6. Crossover Post-hoc Group Comparisons of CAPS at Time

1c–Time 4c

Crossover

Arm CAPS

Time 1c

(Baseline)* Time 4c**

Change Time

1c - Time 4c

MDMA Mean 65.6 33.9 �31.7***

n 7 7 7

Std. Dev. 24.2 12.8 15.0

*Crossover Time 1c is pre-MDMA and at least 2 months post-placebo. Crossover trial

Time 4¼Time 1c, unless >30 days until starting crossover, in which case baseline

measures repeated <30 days before crossover.

**4–6 weeks after second MDMA session

*** p < 0.05 Post-hoc paired t-test determined statistical significance of mean

difference between times

10 Journal of Psychopharmacology 0(0)

Supplemental doses

Comparing CAPS scores over the four time periods between

the four subjects who received a supplemental dose of MDMA (per protocol amendment) and the eight who did not, there were no significant differences in Group by time (ANOVA with repeated measures F(1, 9)¼0.413,

p¼0.637).

Evaluation of blind

Nineteen of 20 participants (95%) correctly guessed their con- dition assignment, though three were uncertain about their

guess. The therapists guessed correctly in all cases, but were uncertain in three instances.

Rescue medication

Zolpidem was administered following 31 of 51 MDMA- assisted sessions (60.7%) and after 11 of 16 psychotherapy-

only sessions (68.8%) (p¼0.77). Benzodiazepines were administered following (though usually not the same day as) 24 of 51 MDMA-assisted sessions (47.0%) and after six

of 16 psychotherapy-only sessions (37.5%) (p¼0.57 – both with Fisher’s Exact Test). Seventeen of 20 subjects, the major- ity of whom had pre-existing sleep disturbance related

to PTSD, received zolpidem for insomnia during study par- ticipation. In five cases this was limited to one or two nights following MDMA or placebo sessions. Fourteen sub- jects received benzodiazepines during study participation.

Eleven of 14 reported taking benzodiazepines before enroll- ment. Two of the three who had not taken benzodiaze- pines before study enrollment did so for 1 and 7 days,

respectively. The mean decrease in CAPS scores from time 1 to time 4 was nearly equal for the 14 who received benzo- diazepines and the six who did not: 40.3 (SD¼39) and 40.7

(SD¼27.2), respectively (p¼0.98).

Discussion

This pilot study demonstrates that MDMA-assisted psycho- therapy with close follow-up monitoring and support can be used with acceptable and short-lived side effects in a carefully

screened group of subjects with chronic, treatment-resistant PTSD. In this group, MDMA-assisted psychotherapy com- pared with the same psychotherapy with inactive placebo pro-

duced clinically and statistically significant improvements in PTSD symptoms as measured by standard symptom scales. This difference was immediate and was maintained through-

out the time period. There were no drug-related serious adverse events and no evidence of impaired cognitive function as measured by neuropsychological testing. The between- group effect size (1.24) of the study drug compares favorably

with other treatment modalities for PTSD (Foa et al., 2009), particularly given the treatment-refractory nature of the cur- rent sample. The clinical significance of the symptom reduc-

tions is indicated by the high percentage of subjects attaining a >30% reduction in CAPS scores and no longer meeting criteria for PTSD 2 months after MDMA-assisted psycho-

therapy, and by the report that all three subjects who had

been unable to work because of PTSD were able to return to work.

The strengths of this study are its prospective, double-

blind, crossover design, the use of a standardized primary outcome measure (CAPS) that is widely used for PTSD research (Weathers et al., 2001), enrollment of chronic, treat- ment-resistant subjects who had moderate to severe PTSD,

and the use of a blinded, independent rater. Subjects met well- defined inclusion and exclusion criteria. Groups were well matched; at baseline, subjects in both groups had nearly iden-

tical CAPS scores. This study has several limitations and should be consid-

ered only a preliminary step toward exploring MDMA as a

possible therapeutic adjunct. Sample size is small, as is appro- priate in a Phase II pilot study. The majority of participants were female and all were Caucasian. Gender and/or ethnic

differences in response to MDMA-assisted psychotherapy could exist. At baseline, the placebo group had a history of more prior psychotherapy than the MDMA group, which could mean that the placebo group was more treat-

ment-resistant; however, this covariate proved non-signifi- cant. Furthermore, in the open-label phase, the placebo group had a response comparable to the MDMA group, so

in fact, the placebo group proved not to be more resistant to MDMA-assisted psychotherapy.

Another important weakness of this study is the transpar-

ency of the blinding. We chose to use an inactive placebo in this initial trial in order to compare side effects of MDMA with those of placebo in this patient population. Although the independent rater remained effectively blinded because he was

not present during experimental sessions, the novel subjective experience was a strong clue for the subjects, as was the sub- jects’ increase in pulse and blood pressure for the investiga-

tors. We have recently obtained a ‘may proceed’ letter from the FDA for a protocol for a three-arm, dose–response study that we expect will result in successful blinding. An argument

against a placebo response having accounted for between- group outcome differences is the maintenance of treatment effect at 2-month follow-up, and the subjects’ having failed

to respond to prior treatments during a course of PTSD last- ing a mean of nearly 20 years. Nevertheless, a placebo effect cannot be ruled out and future studies must address this limitation.

Additional psychotherapy sessions were conducted more often after MDMA-assisted sessions than after psychother- apy-only sessions. This is a potentially confounding factor,

but it is not a likely explanation for the difference in final outcome. Additional psychotherapy cannot explain the sig- nificant improvements in CAPS scores recorded 4 days after

the first MDMA-assisted session, before any additional psy- chotherapy sessions occurred. In future studies, attempts should be made to limit additional psychotherapy sessions while retaining some flexibility to ensure subject safety while

maintaining therapeutic effect. Another limitation is that measurement of the durability of symptom improvement was limited to 2 months after the second experimental session.

We chose this interval because it is well beyond the acute effects of the drug and the immediate expectancy effects, but short enough to minimize effects from intervening events.

A long-term outcome study is currently being conducted

Mithoefer et al. 11

evaluating subjects from 1–5 years post-treatment. The absence of therapist adherence measures was an unavoidable weakness of this first pilot study. A treatment manual and

adherence measures based on audio and video recordings from this study have now been developed for use in future clinical trials.

The use of zolpidem for sleep or benzodiazepines for

anxiety during the study did not differ between the MDMA and placebo conditions, but deserves some discussion to clar- ify the intent. As is common with PTSD, most of the study

subjects had pre-existing sleep disturbance. Zolpidem was offered frequently, often the night after an experimental ses- sion when subjects spent the night in the relatively unfamiliar

environment of the clinic after a long day of emotional pro- cessing. Benzodiazepines were offered more sparingly so as to avoid suppressing ongoing emotional processing, which is

considered an important element of the integration phase of therapy, while providing some symptomatic relief to help subjects effectively balance emotional processing with rest, work and other daily activities. The vast majority of benzo-

diazepine doses went to people who had used them before, and all subjects had previously taken psychiatric medications with significant anxiolytic properties that were not allowed

during the study. It is important to note that a temporary increase in anxiety was sometimes a side effect of this treat- ment, but the fact that neither zolpidem nor benzodiazepines

were administered significantly more often after MDMA sessions than after placebo sessions suggests that this side effect was caused by the psychotherapy in the setting of chronic PTSD rather than by MDMA administration.

An obvious feature of this treatment model is that it involves an initial period of concentrated patient-therapist contact (31 h over 2 months) including all-day therapy ses-

sions and an overnight stay in the clinic. These are not usual features of psychotherapy practice in the outpatient setting. If MDMA-assisted psychotherapy is ultimately approved

for use in clinical practice, it would likely occur in clinics specifically equipped for longer treatment sessions and over- night stays. This method also involves patient preparation

and close follow-up to support further processing of emotions and integration of cognitive shifts that may occur. Both the preparation sessions and the integration sessions appear to be important for safety and therapeutic effect. In future stud-

ies we recommend that the number of 90-min preparation sessions be increased from two to three. This approach is initially more expensive than other outpatient treatments;

however, given the chronic nature of treatment-resistant PTSD requiring ongoing psychiatric treatment and the asso- ciated high rates of medical comorbidity and disability, it

has the potential to be more cost effective over time for a significant segment of the patient population. In this study the second session appeared to add depth to the overall therapeutic process and it provided the reassurance that

subjects would have more than one opportunity to work through their issues. Nevertheless, the fact that most of the symptom improvement occurred after the first MDMA-

assisted psychotherapy session suggests that it would be worthwhile for future studies to investigate the impact of number of sessions on strength and duration of PTSD symp-

tom reduction.

The promising results of this initial pilot study suggest that further research is warranted to confirm our findings, distin- guish and refine the essential elements of this approach,

enhance the methodology, and elucidate the mechanisms involved.

Funding/Support

All funding was provided by The Multidisciplinary Association for Psychedelic Studies, a non-profit organiza-

tion. The sponsor played a role in study design, data analysis and writing of the report. The investigators performed all data collection. The corresponding author wrote the first

draft of the report, had full access to all the data in the study and had final responsibility for the decision to submit for publication.

Conflict of interest statement

Two of the authors, Lisa Jerome and Rick Doblin, are

employed by the sponsor. Michael Mithoefer is medical mon- itor for other studies of MDMA-assisted psychotherapy currently being conducted by the sponsor. He and Ann

Mithoefer both receive payment from the sponsor for con- ducting this research, working on development of a treatment manual, investigator training program, and protocols for

additional studies planned by the sponsor. Mark Wagner received payment from the sponsor for acting as the indepen- dent rater for this study.

Trial Registration

ClinicalTrials.gov Identifier: NCT00090064

Acknowledgements

We wish to thank the subjects who participated in this study, the

Copernicus IRB and our Data Safety Monitoring Board for their

careful oversight, the emergency physicians and nurses who provided

stand-by medical backup, the psychiatric nurses who attended our

subjects during their overnight stays, Yvonne Michel, PhD, for sta-

tistical analysis and guidance, Nicole Vilardo for formatting and

proofing, Heather Mithoefer for graphics and John H Halpern,

MD, Harvard Medical School, Harrison G Pope, MD, MPH,

Harvard Medical School. Andrew Sewell, MD, VA Connecticut

Healthcare/Yale University School of Medicine and Timothy D

Brewerton, MD, Medical University of South Carolina, who pro-

vided helpful comments on an previous version of the manuscript.

References

Baggott MJ, Galloway G, Jang M, Didier R, Pournajafi-Nazarloo H,

Carter CS, et al. (2008) 3,4-methylenedioxymethamphetamine

(MDMA, ‘Ecstasy’) and prazosin interactions in humans.

70 th

Annual Meeting of the College on Problems of Drug

Dependence. San Juan, Puerto Rico.

Baumgartner T, Heinrichs M, Vonlanthen A, Fischbacher U and

Fehr E (2008) Oxytocin shapes the neural circuitry of trust and

trust adaptation in humans. Neuron 58: 639–650.

Benish SG, Imel ZE and Wampold BE (2008) The relative efficacy

of bona fide psychotherapies for treating post-traumatic stress

disorder: a meta-analysis of direct comparisons. Clin Psychol

Rev 28: 746–758.

12 Journal of Psychopharmacology 0(0)

Benzenhoefer UP and Passie T (2006) The early history of ecstasy.

Nervenartz 77: 95–96 (98–99).

Brady K, Pearlstein T, Asnis GM, Baker D, Rothbaum B, Sikes CR,

et al. (2000) Efficacy and safety of sertraline treatment of post-

traumatic stress disorder: a randomized controlled trial. JAMA

283: 1837–1844.

Breslau N (2001) The epidemiology of posttraumatic stress disorder:

what is the extent of the problem? J Clin Psychiatry 62: 16–22.

Breslau N and Davis GC (1992) Posttraumatic stress disorder in

an urban population of young adults: risk factors for chronicity.

Am J Psychiatry 149: 671–675.

Cami J, Farre M, Mas M, Roset PN, Poudevida S and Mas A (2000)

Human pharmacology of 3,4-methylenedioxymethamphetamine

(‘‘ecstasy’’): psychomotor performance and subjective effects.

J Clin Psychopharmacol 20: 455–466.

Cloitre M (2009) Effective psychotherapies for posttraumatic stress

disorder: a review and critique. CNS Spectr 14: 32–43.

Cohen BE, Marmar C, Ren L, Bertenthal D and Seal KH (2009)

Association of cardiovascular risk factors with mental health

diagnoses in Iraq and Afghanistan war veterans using VA

health care. JAMA 302: 489–492.

Committee on Treatment of Posttraumatic Stress Disorder, Institute

of Medicine (2008) Posttraumatic stress disorder: an assessment of

the evidence. Washington, DC: The National Academies Press.

Davis M, Barad M, Otto M and Southwick S (2006) Combining

pharmacotherapy with cognitive behavioral therapy: traditional

and new approaches. J Trauma Stress 19: 571–581.

de Jong JT, Komproe IH and Van Ommeren M (2003) Common

mental disorders in post-conflict settings. Lancet 36: 2128–2130.

Derogatis LR (1994) SCL-90-R: Administration, Scoring and

ProceduresManual.Minneapolis:NationalComputerSystems,Inc.

Domes G, Heinrichs M, Michel A, Berger C and Herpertz SC (2007)

Oxytocin improves ‘‘mind-reading’’ in humans. Biol Psychiatry

61: 731–733.

Dumont GJ, Sweep FC, van der, Steen R, Hermsen R, Donders AR,

Touw DJ, et al. (2009) Increased oxytocin concentrations and

prosocial feelings in humans after ecstasy (3,4-methylenedioxy-

methamphetamine) administration. Soc Neurosci 4: 359–366.

Farre M, Abanades S, Roset PN, Peiro AM, Torrens M, O’Mathuna

B, et al. (2007) Pharmacological interaction between 3,4-methyle-

nedioxymethamphetamine (ecstasy) and paroxetine: pharmaco-

logical effects and pharmacokinetics. J Pharmacol Exp Ther

323: 954–962.

First MB, Gibbon M, Spitzer RL, Williams J and Benjamin L (1997)

Structured clinical interview for Axis II DSM-IV disorders:

Patient Edition (SCID II). Washington, DC: American

Psychiatric Press, Inc.

First MB, Spitzer RL, Gibbon M and Williams J (1994) Structured

clinical interview for Axis I DSM -IV disorders: Patient Edition

(SCID-I/P). New York: NYSPI.

Foa EB, Hembree EA and Rothbaum BO (2007) Prolonged exposure

therapy for PTSD: emotional processing of traumatic experiences:

therapist guide. New York, NY: Oxford University Press.

Foa EB, Keane TM, Friedman MJ and Cohen JA (2009) Effective

treatments for PTSD, practice guidelines from the international

society for traumatic stress studies, 2nd edn. New York, NY:

Guilford Press.

Foa EB and Kozak MJ (1986) Emotional Processing of Fear:

Exposure and corrective information. Psychol Bull 99: 20–35.

Frayne SM, Seaver MR, Loveland S, Christiansen CL, Spiro A and

Parker VA (2004) Burden of medical illness in women with

depression and posttraumatic stress disorder. Arch Intern Med

164: 1306–1312.

Freudenmann RW, Oxler F and Bernschneider-Reif S (2006) The

origin of MDMA (ecstasy) revisited: the true story reconstructed

from the original documents. Addiction 101: 1241–1245.

Gamma A, Buck A, Berthold T, Liechti ME and Vollenweider FX

(2000) 3,4-Methylenedioxymethamphetamine (MDMA) modu-

lates cortical and limbic brain activity as measured by

[H(2)(15)O]-PET in healthy humans. Neuropsychopharmacology

23: 388–395.

Greer GR and Tolbert R (1998) A method of conducting therapeutic

sessions with MDMA. J Psychoactive Drugs 30: 371–379.

Grob CS, Poland RE, Chang L and Ernst T (1996) Psychobiologic

effects of 3,4-methylenedioxymethamphetamine in humans: meth-

odological considerations and preliminary observations. Behav

Brain Res 73: 103–107.

Grof S (2000) Psychology of the Future. Albany, NY: State University

of New York Press.

Gronwall DM (1977) Paced auditory serial-addition task: a measure

of recovery from concussion. Percept Mot Skills 44: 367–373.

Harris DS, Baggott M, Mendelson JH, Mendelson JE and Jones RT

(2002) Subjective and hormonal effects of 3,4-methylenedioxy-

methamphetamine (MDMA) in humans. Psychopharmacology

(Berl) 162: 396–405.

Hembree EA, Foa EB, Dorfan NM, Street GP, Kowalski J and Tu X

(2003) Do patients drop out prematurely from exposure therapy

for PTSD? J Trauma Stress 16: 555–562.

Hoge CW, Castro CA, Messer SC, McGurk D, Cotting DI and

Koffman RL (2004) Combat duty in Iraq and Afghanistan,

mental health problems, and barriers to care. N Engl J Med

351: 13–22.

Ipser J, Seedat S and Stein DJ (2006) Pharmacotherapy for post-

traumatic stress disorder – a systematic review and meta-analysis.

S Afr Med J 96: 1088–1096.

Jaycox LF and Foa EB (1999) Obstacles in implementing exposure

therapy for PTSD: Case discussions and practical solutions. Clin

Psychol Psychother 3: 176–184.

Kessler RC, Chiu WT, Demler O, Merikangas KR and Walters EE

(2005) Prevalence, severity, and comorbidity of 12-month DSM-

IV disorders in the National Comorbidity Survey Replication.

Arch Gen Psychiatry 62: 617–627.

Kirsch P, Esslinger C, Chen Q, Mier D, Lis S, Siddhanti S, et al.

(2005) Oxytocin modulates neural circuitry for social cognition

and fear in humans. J Neurosci 25: 11489–11493.

Kolbrich EA, Goodwin RS, Gorelick DA, Hayes RJ, Stein EA and

Huestis MA (2008) Plasma pharmacokinetics of 3,4-methylene-

dioxymethamphetamine after controlled oral administration to

young adults. Ther Drug Monit 30: 320–332.

Kosfeld M, Heinrichs M, Zak PJ, Fischbacher U and Fehr E (2005)

Oxytocin increases trust in humans. Nature 435: 673–676.

Liechti ME, Baumann C, Gamma A and Vollenweider FX (2000)

Acute psychological effects of 3,4-methylenedioxymethampheta-

mine (MDMA, ‘‘Ecstasy’’) are attenuated by the serotonin uptake

inhibitor citalopram. Neuropsychopharmacology 22: 513–521.

Liechti ME, Gamma A and Vollenweider FX (2001) Gender differ-

ences in the subjective effects of MDMA. Psychopharmacology

(Berl) 154: 161–168.

Liechti ME and Vollenweider FX (2000) The serotonin uptake inhib-

itor citalopram reduces acute cardiovascular and vegetative

effects of 3,4-methylenedioxymethamphetamine (‘Ecstasy’) in

healthy volunteers. J Psychopharmacol 14: 269–274.

Marshall RD, Beebe KL, Oldham M and Zaninelli R (2001) Efficacy

and safety of paroxetine treatment for chronic PTSD: a fixed-

dose, placebo-controlled study. Am J Psychiatry 145: 1982–1988.

Mas M, Farre M, de la Torre R, Roset PN, Ortuno J, Segura J, et al.

(1999) Cardiovascular and neuroendocrine effects and pharmaco-

kinetics of 3, 4-methylenedioxymethamphetamine in humans.

J Pharmacol Exp Ther 290: 136–145.

Metzner R and Adamson S (2001) Using MDMA in healing, psycho-

therapy and spiritual practice. Ecstasy: the complete guide.

Rochester, VT: Inner Traditions, 182–207.

Mithoefer et al. 13

Mirushina MN, Boone KB and D’Elia LF (1999) Handbook of

normative data for neuropsychological assessment. New York,

NY: Oxford University Press.

Ogden PKM and Pain C (2006) Trauma and the body. New York,

NY: W.W. Norton & Company.

Pahnke WN, Kurland AA, Unger S, Savage C and Grof S (1971) The

experimental use of psychedelic (LSD) psychotherapy. Int J Clin

Pharmacol 4: 446–454.

Passie T, Hartmann U, Schneider U, Emrich HM and Kruger TH

(2005) Ecstasy (MDMA) mimics the post-orgasmic state: impair-

ment of sexual drive and function during acute MDMA-effects

may be due to increased prolactin secretion. Med Hypotheses 64:

899–903.

Perkonigg A, Kessler RC, Storz S and Wittchen HU (2000)

Traumatic events and post-traumatic stress disorder in the com-

munity: prevalence, risk factors and comorbidity. Acta Psychiatr

Scand 101: 46–59.

Randolph C, Tierney MC, Mohr E and Chase TN (1998) The

Repeatable Battery for the Assessment of Neuropsychological

Status (RBANS): preliminary clinical validity. J Clin Exp

Neuropsychol 20: 310–319.

Rauch SL, Shin LM and Phelps EA (2006) Neurocircuitry models

of posttraumatic stress disorder and extinction: human neuroim-

aging research – past, present, and future. Biol Psychiatry 60:

376–382.

Rothbaum BO, Cahill SP, Foa EB, Davidson J, Compton J, Connor

KM, et al. (2006) Augmentation of sertraline with prolonged

exposure in the treatment of posttraumatic stress disorder.

J Trauma Stress 19: 625–638.

Sargent W, Slater E and Kelly D (1972) An Introduction to Physical

Methods of Treatment in Psychiatry. Edinburgh and London:

Churchill Livingstone.

Siegel D (1999) The developing mind. New York, NY: Guilford Press.

Stein DJ, Ipser J and McAnda N (2009) Pharmacotherapy of post-

traumatic stress disorder: a review of meta-analyses and treatment

guidelines. CNS Spectr 14: 25–31.

Tancer ME and Johanson CE (2001) The subjective effects of

MDMA and mCPP in moderate MDMA users. Drug Alcohol

Depend 65: 97–101.

Tancer M and Johanson CE (2007) The effects of fluoxetine

on the subjective and physiological effects of 3,4-methylenediox-

ymethamphetamine (MDMA) in humans. Psychopharmacology

(Berl) 189: 565–573.

Tanielian TL and Jaycox L, the Rand Corporation. (2008) Invisible

wounds of war: psychological and cognitive injuries, their consequences,

and services to assist recovery. Santa Monica, CA: RAND.

Thabet AA and Vostanis P (1999) Post-traumatic stress reactions in

children of war. J Child Psychol Psychiatry 40: 385–391.

Thompson MR, Callaghan PD, Hunt GE, Cornish JL and McGregor

IS (2007) A role for oxytocin and 5-HT (1A) receptors in

the prosocial effects of 3,4-methylenedioxymethamphetamine

(‘‘ecstasy’’). Neuroscience 146: 509–514.

Ursano RJ, Bell C, Eth S, Friedman M, Norwood A, Pfefferbaum B,

et al. (2004) Practice guideline for the treatment of patients with

acute stress disorder and posttraumatic stress disorder. Am

J Psychiatry 161: 3–31.

Vollenweider FX, Gamma A, Liechti M and Huber T (1998)

Psychological and cardiovascular effects and short-term sequelae

of MDMA (‘‘ecstasy’’) in MDMA-naive healthy volunteers.

Neuropsychopharmacology 19: 241–251.

Weathers FW, Keane TM and Davidson JR (2001) Clinician-admi-

nistered PTSD scale: a review of the first ten years of research.

Depress Anxiety 13: 132–156.

Weine SM, Becker DF, McGlashan TH, Laub D, Lazrove S,

Vojvoda D, et al. (1995) Psychiatric consequences of ‘‘ethnic

cleansing’’: clinical assessments and trauma testimonies of newly

resettled Bosnian refugees. Am J Psychiatry 152: 536–542.

Weiss DM (1997) The Impact of Event Scale-Revised. In: Wilson JP

and Keanne TM (eds) Assessing Psychological Trauma and PTSD.

New York: Guilford Press.

Wilbarger P and Wilbarger J (1997) Sensory defensiveness and related

social/emotional and neurological problems. Van Nuys,CA: Wilbarger.

Wolff K, Tsapakis EM, Winstock AR, Hartley D, Holt D, Forsling

ML, et al. (2006) Vasopressin and oxytocin secretion in

response to the consumption of ecstasy in a clubbing population.

J Psychopharmacol 20: 400–410.

Zak PJ, Kurzban R and Matzner WT (2005) Oxytocin is associated

with human trustworthiness. Horm Behav 48: 522–527.

14 Journal of Psychopharmacology 0(0)