Psilocybin with psychological support for treatment-resistant depression: An open-label feasibility study

Abstract

Background:
Psilocybin is a serotonin receptor agonist that occurs naturally in some mushroom species. Recent studies have assessed the therapeutic potential of psilocybin for various conditions, including end-of-life anxiety, obsessive-compulsive disorder, and smoking and alcohol dependence, with promising preliminary results. Here, we aimed to investigate the feasibility, safety, and efficacy of psilocybin in patients with unipolar treatment-resistant depression.

Methods:
In this open-label feasibility trial, 12 patients (six men, six women) with moderate-to-severe, unipolar, treatment-resistant major depression received two oral doses of psilocybin (10 mg and 25 mg, 7 days apart) in a supportive setting. There was no control group. Psychological support was provided before, during, and after each session. The primary outcome measure for feasibility was patient-reported intensity of psilocybin's effects. Patients were monitored for adverse reactions during the dosing sessions and subsequent clinic and remote follow-up. Depressive symptoms were assessed with standard assessments from 1 week to 3 months after treatment, with the 16-item Quick Inventory of Depressive Symptoms (QIDS) serving as the primary efficacy outcome. This trial is registered with ISRCTN, number ISRCTN14426797.

Findings:
Psilocybin's acute psychedelic effects typically became detectable 30-60 min after dosing, peaked 2-3 h after dosing, and subsided to negligible levels at least 6 h after dosing. Mean self-rated intensity (on a 0-1 scale) was 0·51 (SD 0·36) for the low-dose session and 0·75 (SD 0·27) for the high-dose session. Psilocybin was well tolerated by all of the patients, and no serious or unexpected adverse events occurred. The adverse reactions we noted were transient anxiety during drug onset (all patients), transient confusion or thought disorder (nine patients), mild and transient nausea (four patients), and transient headache (four patients). Relative to baseline, depressive symptoms were markedly reduced 1 week (mean QIDS difference -11·8, 95% CI -9·15 to -14·35, p=0·002, Hedges' g=3·1) and 3 months (-9·2, 95% CI -5·69 to -12·71, p=0·003, Hedges' g=2) after high-dose treatment. Marked and sustained improvements in anxiety and anhedonia were also noted.

Interpretation:
This study provides preliminary support for the safety and efficacy of psilocybin for treatment-resistant depression and motivates further trials, with more rigorous designs, to better examine the therapeutic potential of this approach.

Funding:
Medical Research Council.

Full text

www.thelancet.com/psychiatry Published online May 17, 2016 http://dx.doi.org/10.1016/S2215-0366(16)30065-7
1
Articles
Lancet Psychiatry 2016
Published Online
May 17, 2016
http://dx.doi.org/10.1016/
S2215-0366(16)30065-7
See Online/Comment
http://dx.doi.org/10.1016/
S2215-0366(16)30087-6
*Contributed equally
Centre for
Neuropsychopharmacology,
Division of Brain Sciences,
Faculty of Medicine, Imperial
College London, London, UK
(R L Carhart-Harris PhD,
M Bolstridge MD,
J Rucker MD, C M J Day MD,
D Erritzoe MD, M Kaelen BSc,
Prof D J Nutt DM); Department
of Pharmacy and Pathology,
South London and Maudsley
NHS Foundation Trust,
London, UK (Prof D Taylor PhD);
The Institute of Psychiatry,
Psychology and Neuroscience
(J Rucker) and Institute of
Pharmaceutical Science
(Prof B Forbes PhD), King’s
College London, London, UK;
Department of Psychiatry
(M Bloomfield MD), Clinical
Psychology and Clinical
Effectiveness
(Prof S Pilling PhD), and Clinical
Psychopharmacology Unit
(Prof V H Curran PhD),
University College London,
London, UK; Barts Health
Pharmaceuticals, Barts Health
NHS Trust, The Royal London
Hospital, London, UK
(J A Rickard PhD); and The
Beckley Foundation, Beckley
Park, Oxford, UK (A Feilding)
Correspondence to:
Dr Robin L Carhart-Harris, Centre
for Neuropsychopharmacology,
Division of Brain Sciences,
Faculty of Medicine, Imperial
College London, London
W12 0NN, UK
r.carhart-harris@imperial.ac.uk
Introduction
Psilocybin is a naturally occurring plant alkaloid
found in the Psilocybe genus of mushrooms. Psilocybe
mushrooms have been used for millennia for healing
purposes, but were only discovered by modern science
in the late 1950s.1,2 Psilocybin is a prodrug of psilocin
(4-hydroxy-dimethyltryptamine), a serotonin receptor
agonist and classic psychedelic drug whose principal
psychoactive eff ects are mediated by serotonin 2A
(5-HT2A) receptor agonism.3 Psilocybin therefore has a
novel pharmacology in the context of currently available
antidepressant medications, because selective serotonin-
reuptake inhibitors are not direct 5-HT2A receptor
agonists.
Enhanced cognitive fl exibility,4 associative learning,5
cortical neural plasticity,6 and antidepressant responses
have been reported with 5-HT2A receptor agonism in
animals,7 and increased and sustained improvements in
wellbeing8 and optimism9 have been observed after
psychedelic experiences in human beings. Findings
from human imaging studies with psilocybin have
supplemented these discoveries, showing changes in
brain activity suggestive of antidepressant potential; for
example, a range of eff ective antidepressant treatments
have been found to normalise hyperactivity in the
medial prefrontal cortex and we found reduced blood
fl ow in this region with intravenous psilocybin.10
Moreover, data obtained from large-scale population
studies have recently challenged the view that
psychedelics negatively aff ect mental health,11–13 with one
study’s fi ndings showing lower rates of psychological
distress and suicidality among people who had used
Psilocybin with psychological support for treatment-resistant
depression: an open-label feasibility study
Robin L Carhart-Harris, Mark Bolstridge, James Rucker*, Camilla M J Day*, David Erritzoe, Mendel Kaelen, Michael Bloomfi eld, James A Rickard,
Ben Forbes, Amanda Feilding, David Taylor, Steve Pilling, Valerie H Curran, David J Nutt
Summary
Background Psilocybin is a serotonin receptor agonist that occurs naturally in some mushroom species. Recent
studies have assessed the therapeutic potential of psilocybin for various conditions, including end-of-life anxiety,
obsessive-compulsive disorder, and smoking and alcohol dependence, with promising preliminary results. Here, we
aimed to investigate the feasibility, safety, and effi cacy of psilocybin in patients with unipolar treatment-resistant
depression.
Methods In this open-label feasibility trial, 12 patients (six men, six women) with moderate-to-severe, unipolar,
treatment-resistant major depression received two oral doses of psilocybin (10 mg and 25 mg, 7 days apart) in a
supportive setting. There was no control group. Psychological support was provided before, during, and after each
session. The primary outcome measure for feasibility was patient-reported intensity of psilocybin’s eff ects. Patients
were monitored for adverse reactions during the dosing sessions and subsequent clinic and remote follow-up.
Depressive symptoms were assessed with standard assessments from 1 week to 3 months after treatment, with the
16-item Quick Inventory of Depressive Symptoms (QIDS) serving as the primary effi cacy outcome. This trial is
registered with ISRCTN, number ISRCTN14426797.
Findings Psilocybin’s acute psychedelic eff ects typically became detectable 30–60 min after dosing, peaked 2–3 h
after dosing, and subsided to negligible levels at least 6 h after dosing. Mean self-rated intensity (on a 0–1 scale)
was 0·51 (SD 0·36) for the low-dose session and 0·75 (SD 0·27) for the high-dose session. Psilocybin was well
tolerated by all of the patients, and no serious or unexpected adverse events occurred. The adverse reactions we
noted were transient anxiety during drug onset (all patients), transient confusion or thought disorder (nine patients),
mild and transient nausea (four patients), and transient headache (four patients). Relative to baseline, depressive
symptoms were markedly reduced 1 week (mean QIDS diff erence –11·8, 95% CI –9·15 to –14·35, p=0·002, Hedges’
g=3·1) and 3 months (–9·2, 95% CI –5·69 to –12·71, p=0·003, Hedges’ g=2) after high-dose treatment. Marked and
sustained improvements in anxiety and anhedonia were also noted.
Interpretation This study provides preliminary support for the safety and effi cacy of psilocybin for treatment-resistant
depression and motivates further trials, with more rigorous designs, to better examine the therapeutic potential of
this approach.
Funding Medical Research Council.
Copyright © Carhart-Harris et al. Open Access article distributed under the terms of CC BY.

Articles
2
www.thelancet.com/psychiatry Published online May 17, 2016 http://dx.doi.org/10.1016/S2215-0366(16)30065-7
psychedelics within their lifetime than among those
who used no psychedelics but an equivalent amount of
other drugs.11 In modern trials, psychedelics have been
found to reduce anxious,14,15 depressive,15,16 and obsessive-
compulsive symptoms,17 as well as addictive
behaviours,18,19 often for several months after just one or
two exposures. Extensive historical and modern
evidence now supports the view that, administered in a
controlled environment with appropriate support,
psychedelics have a favourable safety profi le.20
Depression is a major public health problem; it is a
leading contributor to the global burden of disease,
aff ecting hundreds of millions of people worldwide, and
costing the USA alone more than US$200 billion each
year.21 Antidepressant medications and cognitive behav-
ioural therapy can be eff ective for some patients, but
around 20% do not respond to any intervention, and many
of those who do respond, eventually relapse.22 We aimed to
investigate the safety and feasibility of psilocybin in
patients with treatment-resistant depression, and to
establish an initial impression of its effi cacy. We postulated
that the treatment would be well tolerated and depressive
symptoms would be substantially reduced from baseline at
all assessment points, for up to 3 months after treatment.
Methods
Study design and participants
This was an open-label feasibility study in patients with
treatment-resistant depression; there was no control
group. Patients, invest igators, raters, and statisticians
were not masked to treatment assignment, and all
participants received the study intervention (psilocybin
administered in two dosing sessions; an initial safety
[low] dose and a subsequent treatment [high] dose). The
inclusion criteria were major depression of a moderate to
severe degree (17+ on the 21-item Hamilton Depression
Rating scale [HAM-D]), and no improvement despite two
adequate courses of antidepressant treatment of diff erent
pharmacological classes lasting at least 6 weeks within
the current depressive episode.23 Exclusion criteria were:
current or previously diagnosed psychotic disorder;
immediate family member with a diagnosed psychotic
disorder; medically signifi cant condition rendering
unsuitability for the study; history of serious suicide
attempts (requiring hospitalisation); history of mania;
blood or needle phobia; positive pregnancy test at
screening or during the study; and current drug or
alcohol dependence.
Information about the study’s recruitment was sent to
general practitioners via the North West London Clinical
Research Network. However, patients were also allowed
to self-refer to the study if they were UK residents. In
every case, patients initiated contact with the research
team (via email, letter, or telephone), were sent a study
information sheet, and a subsequent telephone screening
was arranged, during which the lead psychiatrist on the
trial (MBo) obtained information about the patient’s
demographics, medical and psychiatric history, and other
key inclusion or exclusion criteria. The patient’s
general practitioner or psychiatrist provided written
documentation of the patient’s diagnosis and mental
health background in every case.
This trial received a favourable opinion from the
National Research Ethics Service London—West London,
was sponsored and approved by Imperial College London’s
Joint Research and Compliance Offi ce (JRCO), and was
adopted by the National Institute for Health Research
Clinical Research Network. The National Institute for
Health Research/Wellcome Trust Imperial Clinical
Research Facility gave site-specifi c approval for the study.
Research in context
Evidence before this study
We searched PubMed up to Jan 30, 2016, using the terms
“psilocybin”, “hallucinogens”, “psychedelics”, and
“depression”. We did not fi nd any clinical trials assessing
psilocybin as a treatment for depression, but we did fi nd
population analyses, review articles, and imaging studies
lending support to this approach. We also found one report
documenting enduring decreases in depressive symptoms
after a single dose of psilocybin in a randomised controlled
trial of psilocybin-assisted psychotherapy for end-of-life
anxiety, one report on an open-label trial showing rapid
decreases in depressive symptoms that endured for up to
21 days after a single dose of ayahuasca, and two early reports
or case studies on the eff ects of lysergic acid diethylamide on
“neurotic” and depressive symptoms describing
“improvements”, albeit without validated measures of
symptom severity.
Added value of this research
To our knowledge, this is the fi rst investigation of the safety and
effi cacy of psilocybin as a treatment for major depression.
Our fi ndings imply that psilocybin might have value as a treatment
option in the management of treatment-resistant depression.
Single oral administrations of 10 mg (safety dose) and 25 mg
(treatment dose) psilocybin were well tolerated and led to
enduring reductions in symptom severity after the two sessions.
Implications of all the available evidence
The results of this small-scale feasibility study should help to
motivate further research into the effi cacy of psilocybin with
psychological support for major depression. Larger-scale
randomised controlled trials are warranted to better examine the
potential of psilocybin as a treatment option for this highly
prevalent, disabling, costly, and diffi cult-to-treat disorder.
More broadly, the present study should help to catalyse the
re-emergence of a promising research area in psychiatry.

Articles
www.thelancet.com/psychiatry Published online May 17, 2016 http://dx.doi.org/10.1016/S2215-0366(16)30065-7
3
The study was reviewed and approved by the Medicines
and Healthcare products Regulatory Agency (MHRA). All
participants provided written informed consent. Study
and data monitoring was carried out independently by the
Imperial Clinical Research Facility and JRCO.
Procedures
Psilocybin was obtained from THC-pharm (Frankfurt,
Germany) and formulated into the investigational
medicinal product (5 mg psilocybin in size 0 capsules)
by Guy’s and St Thomas’ Hospitals’ Pharmacy
Manufacturing Unit (London, UK). A Home Offi ce
Licence for storage and dispensing of Schedule One
drugs was obtained.
Screening consisted of written informed consent, a
thorough evaluation of the patient’s physical and mental
health background, a psychiatric interview (Mini-
International Neuropsychiatric Interview), clinician
assessments of depression severity (the 21-item HAM-D
and the Montgomery-Åsberg Depression Rating Scale
[MADRS], and Global Assessment of Functioning [GAF];
all assessed by MBo), and additional patient-rated scales
(16-item Quick Inventory of Depressive Symptoms
[QIDS], Beck Depression Inventory [BDI—original
version], Spielberger’s State-Trait Anxiety Inventory
[form 2, trait version only; STAI-T], and the Snaith-
Hamilton Pleasure Scale [SHAPS]). Patients also received
a thorough physical health check, consisting of an
electrocardiogram, routine blood tests, blood pressure,
heart rate, and physical examination. At the end of
screening, eligible patients were given an opportunity to
meet with the two clinical psychiatrists who would
support them through the remainder of the trial.
Eligible patients attended a subsequent visit involving a
baseline functional MRI (fMRI) scanning session lasting
60 min, followed by an extensive preparatory session
with their allocated psychiatrists; fMRI data will be
reported elsewhere. This preparatory session involved
inviting the patient to talk openly about their personal
history (including thoughts on the origins of their
depression), a discussion of psilocybin’s psychological
eff ects, and simulation of aspects of the dosing session
itself, such as listening to a sample of the session music
while wearing eyeshades. The preparatory session
typically lasted for 4 h, with lunch and breaks provided.
Patients enrolled in the study attended two subsequent
dosing sessions that were separated by 7 days. No more
than one patient was dosed on any given day. Patients
arrived at the research facility (Imperial Clinical Research
Facility) at 0900 h, gave a urine sample for drugs of abuse
(including amphetamines, benzodiazapines, opiates,
and cannabinoids), performed a breathalyser test for
alcohol use, and completed interim QIDS, BDI, and
STAI-T assessments to ensure no substantial deviation
from baseline measures. They were then taken to a
dosing room that was pre-decorated (eg, with low
lighting). Patients were invited to relax on a ward bed in a
supine or reclined position and music was played
through high-quality stereo speakers and earphones. The
two psychiatrists sat on either side of the bed. Patients
were supervised at all times by at least two staff members.
Dosing commenced at 1030 h in every case. Patients
received a low oral dose of psilocybin 10 mg (two 5 mg
capsules) on a fi rst dosing day and a high oral dose of
psilocybin 25 mg (fi ve 5 mg capsules) on a second dosing
day, separated by 1 week. Blood pressure, heart rate, and
observer ratings of the intensity of psilocybin’s acute
psychoactive eff ects (0–4, with 0 signifying no eff ects and
4 signifying extreme eff ects8) were measured at baseline
(typically 5 min before dosing) and 30, 60, 120, 180, 240,
300, and 360 min after dosing. Subjective ratings of the
acute altered state of consciousness using the revised
Figure 1: Schedule of study interventions
Follow-up
Enrolment and treatment
Recruitment
Unspecified time period
Low
psilocybin
dose
High
psilocybin
dose
Remote screening or follow-up
Clinic screening or follow-up
Psilocybin dosing session
1 week 1 week 2 weeks 7 weeks
Each bar represents 1 day
Ongoing support from
study psychiatrists
if required
Time
Telephone
screening
Screening visit
and baseline
assessment
Patients
contacted
for remote
assessment
Post-treatment
fMRI and assessment
(interim questionnaires)
Baseline
fMRI and
preparatory
session
Interim
questionnaires
1 week follow-up
at research facility
2 week follow-up
(remotely)
3 week follow-up
(remotely)
5 week follow-up
(remotely)
3 month follow-up
(remotely)

Articles
4
www.thelancet.com/psychiatry Published online May 17, 2016 http://dx.doi.org/10.1016/S2215-0366(16)30065-7
11 dimension altered states of consciousness questionnaire
(11D ASC)24 were completed 6–7 h after dosing.
Psychiatrists adopted a non-directive, supportive
approach, allowing the patient to experience a mostly
uninterrupted inner “journey”. Check-ins (ie, asking the
patient how they are feeling) occurred at the same
timepoints as the physiological recordings. Tranquilising
medications (oral lorazepam and risperidone) were
available if necessary. The phenomenology of the acute
experience, including accounts of the nature of the
therapeutic support provided before, during, and after
the experience, and considerations related to the music
selection and other aspects of the clinical setting, will be
discussed in separate publications.
Return transport from the research facility was
organised ahead of dosing sessions. Patients were taken
to and from the sessions accompanied by a close friend
or relative, and had the option of staying overnight in
accommodation adjacent to the hospital. Emergency
contact details were provided, and patients confi rmed
their safe return from the research facility.
Patients were contacted via telephone 1 day after their
low-dose session to check on their wellbeing and monitor
for any adverse events. Patients returned to the research
facility 1 day after their high-dose session for a post-
treatment fMRI scan lasting 60 min. After the fMRI
scan, patients completed interim questionnaires (QIDS,
STAI-T, and HAM-D), and were invited back to the
research facility where they were met by their
psychiatrists to discuss their experience the previous day.
Patients attended one further study visit to the research
facility 1 week after their high-dose session, during which
all baseline questionnaires and assessments were repeated
and an opportunity was provided for further psychological
debriefi ng (the 1 week follow-up visit). Assessments of
HAM-D, MADRS, and GAF were again done by MBo.
Subsequent assessments of clinical progress were done via
email 2, 3, and 5 weeks after the high-dose session; we
assessed only QIDS during subsequent follow-up, so as
not to overload the patient. Final follow-up was done
remotely at 3 months after the high-dose session, and
included QIDS, BDI, STAI-T, and SHAPS. Patients were
made aware that they could contact the study psychiatrists
at any time if their depression deteriorated. Figure 1
summarises the screening, intervention, and follow-up
procedures in this study.
Outcomes
The main objective of this study is to optimise the protocol
for the administration of oral psilocybin in this patient
group, while gaining an initial impression of treatment
effi cacy. The primary outcome measure to assess feasibility
was patient-rated subjective intensity of psilocybin’s
eff ects, which we report on a 0–1 scale. We assessed the
safety of the intervention through clinical monitoring
during and after dosing sessions, and during 3 months of
face-to-face and remote follow-up. We also aimed to assess
the preliminary effi cacy of psilocybin in patients with
treatment-resistant depression; the primary outcome
measure for this endpoint was mean change in the severity
of self-reported depressive symptoms (with the 16 item
QIDS) from baseline to 1 week after the high-dose
psilocybin session. The QIDS was chosen as the primary
outcome measure due to its brevity, increasingly
widespread use, and validity at 1 week intervals.25 We chose
to assess the primary effi cacy endpoint at 1 week after the
high-dose session to allow comparison with previous
studies of ketamine infusion for treatment-resistant
depression;26 the low-dose session was conceived a priori
as a safety session rather than a treatment session. We also
assessed change in BDI, STAI-T, and SHAPS between
baseline and 1 week and 3 months of follow-up, and
change in HAM-D, MADRS, and GAF between baseline
and 1 week of follow-up.
Statistical analysis
In this feasibility study, we did not perform a formal
power calculation. We planned to recruit 12 patients to
provide an initial impression of the tolerability and
effi cacy of this novel treatment approach. A subsequent
protocol amendment (Oct 6, 2015) increased the
recruitment to 20 patients to provide statistical power
for fMRI imaging. Here, we report fi ndings for the
12 patients initially enrolled; outcome and fMRI data
for all 20 patients will be reported separately.
Due to the small population, two-tailed Wilcoxon
signed ranks tests were performed for non-parametric
data. Two-tailed t tests were also performed and the
relevant t values are provided in the appendix. We provide
95% CIs around the mean diff erences. We calculated
Figure 2: Trial profi le
72 individuals expressed an interest in
participating in the trial
34 excluded because they did not meet the
entry criteria
38 telephone screened
20 excluded because they did not meet the
entry criteria
18 attended screening visit
6 excluded because of insufficiently severe
depression (HAM-D)
12 recruited to the study and fully compliant
with protocol
See Online for appendix

Articles
www.thelancet.com/psychiatry Published online May 17, 2016 http://dx.doi.org/10.1016/S2215-0366(16)30065-7
5
eff ect sizes using the Hedges’ g formula, which is more
appropriate for small sample sizes. Hedges’ g values are
very similar to Cohen’s d values for dependent data.
This trial is registered with the ISRCTN registry,
number ISRCTN14426797. The registration was initiated
on March 30, 2015, and fi nalised on July 7, 2015 (delay
caused by administrative issues); recruitment started on
April 21, 2015, after initiation of public registration.
Role of the funding source
The study funder had no role in the design, data collection,
analysis, interpretation, or writing of the report. The
corresponding author had full access to all of the data in
the study and had fi nal responsibility for the decision to
submit for publication.
Results
Enrolment started on May 1, 2015, and fi nished on Aug 25,
2015. 72 people were initially considered for the study,
most of whom self-referred after hearing about this trial
through public outreach work (eg, public presentations by
the investigators and media reports). 38 were considered
appropriate for a telephone screen, from which 18 were
invited for a formal screening visit, and 12 were ultimately
recruited for the trial (fi gure 2), of whom ten were self-
referrals. Patients’ demographic and clinical characteristics
are shown in table 1. Nine of the 12 patients met criteria
for severe or very severe depression at baseline
(BDI score ≥30), with the remaining three patients meet-
ing criteria for moderate depression (BDI score 19 to <30).
11 patients had received some form of psychotherapy
before participation in the study.
The acute effects of psilocybin were well tolerated by
all of the patients and no serious or unexpected
adverse events occurred. Mean self-rated intensity of
psilocybin experience was 0·51 (SD 0·36) for the low-
dose session and 0·75 (0·27) for the high-dose session
(difference 0·24 [95% CI 0·06–0·41], Z –2·4, p=0·019).
Sex Age,
years
Ethnic
origin
Employment
status
Estimated
illness
duration,
years
Baseline scores Past unsuccessful
medications*
Past psychotherapy Education Weekly
alcohol
intake,
units
Previous
psilocybin use
(time since last
use)
BDI HAM-D STAI-T
1 Female 43 Black
Caribbean
Employed 30 36 19 72 SSRI (two), SNRI
(two), NDRI,
NSSRI, MAOI
None Postgraduate 1 None
2 Male 40 Hispanic Unemployed 25 33 28 76 SSRI (two), SNRI,
NDRI, NSSRI, Na+
channel blocker
(two), ketamine
infusion, TCA
Cognitive narrative
therapy
Postgraduate 0 None
3 Male 37 White Employed 17 22 18 63 SSRI (two), SNRI Cognitive behavioural
therapy, group therapy
Postgraduate 0 None
4 Female 30 White Studying 10 26 18 67 NDRI, NSSRI Cognitive behavioural
therapy
Postgraduate 0 One use
(6 months)
5 Male 34 White Unemployed 12 38 25 71 SSRI (three), TCA Cognitive and
mindfulness
behavioural therapy
Undergraduate 0 None
6 Female 57 White Unemployed 29 39 23 78 SSRI (four), SNRI,
SARI
Counselling Secondary
education
2 Two uses
(45 years)
7 Male 52 White Unemployed 27 33 22 57 TCA, SARI Counselling,
mindfulness
Secondary
education
0 Three uses
(30 years)
8 Female 37 White Employed 17 39 17 71 SSRI (two), TCA Counselling Undergraduate 2 None
9 Male 37 White Unemployed 15 32 26 71 SSRI (three), SNRI Counselling, cognitive
behavioural therapy
Postgraduate 6 None
10 Female 36 Black
Caribbean
Unemployed 8 47 28 75 SSRI (two), NSSRI Counselling Undergraduate 18 Three uses
(14 years)
11 Female 64 White Employed 15 24 17 72 SSRI (four), SNRI
(two), NDRI, MAOI,
Na⁺ channel
blocker, SARI, DRI
Cognitive behavioural
therapy
Postgraduate 1 Three uses
(48 years)
12 Male 45 White Employed 8 35 17 68 SSRI, TCA Cognitive behavioural
therapy
Undergraduate 0 None
BDI=Beck Depression Inventory. HAMD-D=Hamilton Depression Rating scale. STAI-T=State-Trait Anxiety Inventory. SSRI=selective serotonin-reuptake inhibitor. SNRI=serotonin–noradrenaline reuptake
inhibitor. NDRI=noradrenaline–dopamine-reuptake inhibitor. NSSRI=noradrenaline and specifi c serotonin-reuptake inhibitor. MAOI=monoamine oxidase inhibitor. TCA=tricyclic antidepressant. SARI=serotonin
antagonist and reuptake inhibitor. DRI=dopamine-reuptake inhibitor. *One medication from each class, unless otherwise stated.
Table 1: Baseline and demographic characteristics, by patient

Articles
6
www.thelancet.com/psychiatry Published online May 17, 2016 http://dx.doi.org/10.1016/S2215-0366(16)30065-7
No patients required tranquilising medications (oral
lorazepam and risperidone) during the dosing
sessions. Psilocybin’s acute psychedelic effects
typically became detectable between 30 min and
60 min after dosing, peaked between 2 h and 3 h after
dosing, and subsided to negligible levels at which the
patient could be assessed for discharge at least 6 h
after dosing (appendix). Self-rated experiences on the
11D-ASC questionnaire from the two sessions are
shown in the appendix. Results from interim patient
questionnaires (QIDS, BDI, and STAI-T), done
immediately before the low-dose session to monitor
for substantial changes since enrolment, did not
differ from baseline (data not shown). Interim
questionnaires done the day after the high-dose
session showed some reduction in depressive
symptoms (data for HAM-D in appendix; data for
QIDS and STAI-T not shown).
Severity Timing or onset Duration
Patient 1
Transient anxiety Mild Onset of both sessions 60 min
Transient headache Mild Day after high-dose session One day only
Transient confusion Mild (core drug eff ect) Peak of both sessions 60–120 min
Patient 2
Transient anxiety Mild Anticipatory anxiety only (both sessions) 30 min
Patient 3
Transient anxiety Mild Anticipatory anxiety only (both sessions) 30 min
Transient confusion Mild (core drug eff ect) Peak of both sessions 60–180 min
Patient 4
Transient anxiety Mild (low dose), moderate (high dose) Onset of both sessions and peak of high dose 60 min (low dose), 120 min (high dose)
Transient nausea Moderate Onset phase of high-dose session Arose and subsided within 60 min
Transient confusion Mild (core drug eff ect) Peak of both sessions 60–180 min
Transient paranoia Mild Peak of high-dose session Arose and subsided within 30 min
Patient 5
Transient anxiety Moderate (low dose), severe (high dose) Onset of both sessions and peak of high dose 60 min (low dose), 150 min (high dose)
Transient headache Mild Day after high-dose session One day only
Transient confusion Mild (core drug eff ect) Peak of both sessions 60–120 min
Patient 6
Transient anxiety Mild Anticipatory anxiety only (both sessions) 30 min
Patient 7
Transient anxiety Mild Anticipatory anxiety only (both sessions) 30 min
Transient confusion Mild (core drug eff ect) Peak of both sessions 60–180 min
Patient 8
Transient anxiety Mild or negligible Anticipatory anxiety only (both sessions) 30 min
Patient 9
Transient anxiety Mild (low dose), moderate (high dose) Onset of low-dose and high-dose session 60 min (low dose), 150 min (high dose)
Transient headache Mild Day after high-dose session One day only
Transient confusion Mild (core drug eff ect) Peak of both sessions 60–180 min
Patient 10
Transient anxiety Mild Onset of both sessions 60 min
Transient nausea Mild Onset and peak of low-dose session Subsided after 90 min
Transient headache Mild or moderate Day after high-dose session 2 days
Transient confusion Mild (core drug eff ect) Peak of both sessions 60–180 min
Patient 11
Transient anxiety Moderate (both sessions) Onset phase and peak of both sessions 150 min (both sessions)
Transient nausea Mild (high dose) Onset phase of high-dose session Arose and subsided within 60 min
Transient confusion Mild (core drug eff ect) Peak of both sessions 60–180 min
Transient paranoia Mild Peak of low-dose session Arose and subsided within 60 min
Patient 12
Transient anxiety Mild Anticipatory anxiety only (both sessions) 30 min
Transient confusion Mild (core drug eff ect) Peak of both sessions 60–180 min
Table 2: Adverse events by patient

Articles
www.thelancet.com/psychiatry Published online May 17, 2016 http://dx.doi.org/10.1016/S2215-0366(16)30065-7
7
The most common adverse events were transient
anxiety (mostly mild) during drug onset (n=12),
transient confusion or thought disorder (n=9), mild and
transient nausea (n=4), and transient headache (n=4;
table 2). These adverse events were expected psycho-
logical eff ects of psilocybin. Subacute headache typically
presented 1 day after the psilocybin session, and
subsided after 1–2 days. Paranoia presented in only one
patient, but this was mild and transient. No prolonged
psychotic symptoms were observed in any of the
patients. One patient contacted the study psychiatrists
during the 3 months of follow-up due to deterioration of
their depression, and was referred to their general
practitioner.
QIDS depression scores were signifi cantly reduced
from baseline to 1 week and 3 months post-treatment,
with the maximum eff ect at 2 weeks (fi gure 3, table 3).
BDI and clinician-administered ratings confi rmed
these results (fi gure 4, table 3). All patients showed
some reduction in depression severity at 1 week that
was sustained in the majority for 3 months (appendix).
According to standard criteria for determining
remission (eg, a score of ≤9 on the BDI), eight (67%) of
the 12 patients achieved complete remission at 1 week
and seven patients (58%) continued to meet criteria for
response (50% reduction in BDI score relative to
baseline) at 3 months, with fi ve of these (42%) still in
complete remission (fi gure 4, table 3). STAI-T anxiety
scores were also signifi cantly reduced at 1 week and
3 months post-treatment, as were SHAPS anhedonia
scores for 1 week and 3 months post-treatment (table 3).
Discussion
In this open-label, single-arm pilot study, we sought to
examine the feasibility of administering psilocybin to
patients with treatment-resistant depression as a prelude
to a larger randomised controlled trial. Our results support
Figure 3: Mean depression severity (QIDS) over time
Depression severity determined by self-rated 16-item QIDS. QIDS scores of
16–20 are considered to refl ect severe depression, scores of 11–15 are considered
moderate depression, scores of 6–10 are considered mild depression, and scores
of 5 and less are considered absent depression. All post-treatment assessments
were obtained after the high-dose session (ie, 1 week post-treatment refers to
1 week after the high-dose session). Hedges’ g values versus baseline are shown.
QIDS=Quick Inventory of Depressive Symptoms.
0
5
10
15
20
25
Baseline 1 week 2 weeks 3 weeks 5 weeks 3 months
QIDS score
Hedges’ g
3·1
p=0·002
Hedges’ g
3·2
p=0·002
Hedges’ g
3·2
p=0·002
Hedges’ g
2·7
p=0·003
Hedges’ g
2·0
p=0·003
QIDS BDI STAI-T SHAPS HAM-D MADRS GAF
Base-
line
1 week 2 weeks 3 weeks 5 weeks 3 months Base-
line
1 week 3 months Base-
line
1 week 3 months Base-
line
1 week 3 months Base-
line
1 week Base-
line
1 week Base-
line
1 week
Mean (SD) 19·2
(2·0)
7·4
(4·9)
6·3
(4·6)
6·4
(5·1)
8·2
(5·4)
10·0
(6·0)
33·7
(7·1)
8·7
(8·4)
15·2
(11·0)
70·1
(5·8)
40·6
(14·2)
54·8
(14·5)
7·5
(3·7)
1·4
(2·7)
2·8
(3·7)
21·4
(4·5)
7·4
(6·9)
31·0
(5·0)
9·7
(9·8)
50·3
(9·2)
77·7
(13·0)
Diff erence
versus
baseline
(95% CI)
·· –11·8
(–9·15 to
–14·35)
–12·9
(–10·64
to
–15·16)
–12·8
(–9·9 to
–15·6)
–11·0
(–7·7 to
–14·2)
–9·2
(–5·69 to
–12·71)
·· –25·0
(–20·1
to
–29·9)
–18·5
(–11·8 to
–25·2)
·· –29·5
(–22·03
to
–36·97)
–15·3
(–7·77 to
–22·83)
·· –6·1
(–4·46
to
–7·74)
–4·7
(–3·29 to
–6·11)
·· –14·0
(–9·6 to
–18·4)
·· –23·3
(–17·1 to
–29·5)
·· 27·3
(18·0 to
36·6)
Z·· –3·1 –3·1 –3·06 –2·9 –3·0 ·· –3·1 –3·1 ·· –3·1 –2·9 ·· –3·1 –3·1 ·· –3·0 ·· –3·1 ·· 3
Hedges’ g* ·· 3·1 3·2 3·2 2·7 2·0 ·· 3·2 2·0 ·· 2·7 1·4 ·· 1·9 1·3 ·· 2·4 ·· 2·7 ·· 2·4
p value* ·· 0·002 0·002 0·002 0·003 0·003 ·· 0·002 0·002 ·· 0·002 0·004 ·· 0·002 0·002 ·· 0·003 ·· 0·002 ·· 0·003
Follow-up refers to the period starting after the second (high-dose) administration of psilocybin. Clinician-administered ratings (HAM-D, MADRS, and GAF) were completed only at baseline and 1 week after the high-dose session. QIDS=Quick
Inventory of Depressive Symptoms. BDI=Beck Depression Inventory. STAI-T=State-Trait Anxiety Inventory. SHAPS=Snaith-Hamilton Pleasure Scale. HAM-D=Hamilton Depression Rating scale. MADRS=Montgomery-Åsberg Depression Rating Scale.
GAF=Global Assessment of Functioning. *Compared with baseline.
Table 3: Clinical ratings at baseline and follow-up

Articles
8
www.thelancet.com/psychiatry Published online May 17, 2016 http://dx.doi.org/10.1016/S2215-0366(16)30065-7
the view that, done with appropriate safeguards (eg, careful
screening and adequate therapeutic support), psilocybin
can be safely administered to this patient group.
Because this was a small-scale feasibility study with an
open-label design, strong inferences cannot be made
about the treatment’s therapeutic effi cacy. However, the
data do suggest that further research is warranted. The
response rate to psilocybin was 67% (n=8) at 1 week
after treatment (HAM-D and BDI), and seven of these
eight patients also met criteria for remission. Moreover,
58% (n=7) of the patients maintained their response for
3 months, and 42% (n=5) remained in remission. It is
also worth noting that psilocybin has a favourable
toxicity profi le and is not associated with compulsive
drug-seeking behaviours in animals or human beings.
The side-eff ects that we noted were minor, and expected
in light of previous studies of psilocybin.27
Spontaneous recovery in refractory depression is rare,
and many of the patients in the present study reported
having depression for much of their adult lives
(mean estimated illness duration 17·8 years [SD 8]). Key
questions for future research therefore should address why
the therapeutic eff ect observed in the present study is so
large, and if it can be replicated when tighter experimental
controls are introduced. Because the treatment in our study
consisted of not just two psilocybin administrations but
also psychological support before, during, and after these
sessions, as well as a positive therapeutic environment for
the sessions, the relative eff ects of these factors need to be
determined, which can only be done by conducting further
trials with appropriate control conditions.
A logical next step would be to carry out a placebo-
controlled randomised trial in which the level of therapist
contact is consistent between conditions. This would
enable any between-group diff erences in clinical
outcomes to be attributed to psilocybin rather than the
psychological support provided. However, a positive
interaction between these variables seems likely, and
inert placebo-based blinds are known to be ineff ective
in studies involving conspicuous experimental inter-
ventions, because patients can easily discern whether
they are in the active condition or not. Use of an active
placebo for the control condition might therefore be
worth considering. Additionally, randomised comparative
effi cacy trials (eg, with an optional crossover component)
incorporating another treatment for refractory depression
(eg, ketamine infusion) could also be explored.
The magnitude and persistence of the antidepressant
eff ects observed here are not incongruent with what has
been observed previously with psilocybin in chronic
psychiatric conditions. For example, 80% of long-term
heavy tobacco smokers demonstrated abstinence from
smoking 6 months after two treatment sessions with
psilocybin.18 Alcohol-dependent patients demonstrated
signifi cantly reduced drinking behaviours over 8 months
after one or two psilocybin sessions.19 Signifi cantly
decreased anxiety and depression scores were observed
3 and 6 months after a single dose of psilocybin in
patients with anxiety related to end-stage cancer,15 and
improve ments in wellbeing lasting for more than 1 year
were observed in healthy individuals given a single dose
of psilocybin.8 Rapid and enduring decreases in depressive
symptoms were also recently found in a small-scale
feasibility trial involving the psychedelic brew, ayahuasca.16
It is important to consider the limitations of this pilot
study; for example, although all patients showed some
clinical improvements for at least 3 weeks after treatment,
and no serious or unexpected adverse reactions were
observed, enduring improvements beyond 3 weeks were
not observed universally, and fi ve of the 12 patients
showed a degree of relapse at 3 months.
One should be cautious of the potential for infl ated
eff ect sizes in early trials, particularly when the sample
size is small. That all patients showed some improvement
in their depressive symptoms for up to 3 weeks after
treatment could be suggestive of an expectancy bias.
It may also be relevant that most patients in this trial were
self-referring and, thus, actively sought this treatment.
Psychedelics are known to promote suggestibility,28 which
might have further enhanced positive outcomes. Future
double-blind randomised controlled trials could address
the role of expectancy and suggestibility by measuring and
controlling these variables. For example, patients could be
asked about their pre-treatment expectations, suggestions
could be controlled between conditions, and outcomes
from self-referred patients could be compared with those
from patients referred via clinicians. From a more
pragmatic perspective, if expectancy or suggestibility are
found to be infl uential in the context of psychedelic
therapy, they could be treated as exploitable components
of the treatment model rather than confounding variables.
Figure 4: Depression severity (BDI) over time, by patient
Figure shows depression severity (BDI) over time plotted for each of the 12 patients. Mean values (SD) are shown
as well as the relevant eff ect sizes (Hedges’ g) versus baseline. BDI=Beck Depression Inventory.
0
10
20
30
40
50
Baseline 1 week 3 months
BDI score
33·7 (7·1)
15·2 (11)
g=2
Severe depression (≥30)
Mild or moderate
depression (>9 to >30)
No depression (≤9)
8·7 (8·4)
g=3·2
P1
P2
P3
P4
P5
P6
P7
P8
P9
P10
P11
P12
High dose

Articles
www.thelancet.com/psychiatry Published online May 17, 2016 http://dx.doi.org/10.1016/S2215-0366(16)30065-7
9
Serotonergic antidepressants have been found to down-
regulate the primary receptor target of psilocybin (the
5-HT2A receptor) and attenuated subjective responses to
psychedelics have previously been reported in individuals
chronically medicated with serotonergic antidepressants.29
Thus, patients may be required to withdraw from
concurrent antidepressant medication before receiving
psilocybin and this should only ever be done with care.
In conclusion, we sought to assess the safety and
tolerability of psilocybin plus psychological support in
patients with unipolar treatment-resistant depression.
Our fi ndings support the feasibility of this approach and
the magnitude and duration of the post-treatment
reductions in symptom severity motivate further
controlled research. Psilocybin has a novel pharma-
cological action in comparison with currently available
treatments for depression (ie, 5-HT2A receptor agonism)
and thus could constitute a useful addition to available
therapies for the treatment of depression.
Contributors
RLC-H and DJN designed the study and RLC-H wrote the report. RLC-H
coordinated the study, and collected and analysed the data. MBo was the
lead psychiatrist on the trial. MBo, JR, CMJD, DE, and MBl provided
psychological support for the patients. All authors critically revised the
report or contributed important intellectual content.
Declaration of interests
DT has received research funding and lecture honoraria from Servier,
and lecture honoraria from Lundbeck. The other authors declare no
competing interests.
Acknowledgments
This study was funded by an MRC clinical development scheme grant
(MR/J00460X/1). MK was supported by the Beckley Foundation and this
work was carried out as part of the Beckley/Imperial Research
Collaboration. The research was carried out at the NIHR/Wellcome
Trust Imperial Clinical Research Facility. We would like to thank
Robert Sullivan (Meridian West London, London, UK) for provision of
high-quality audio equipment for the dosing sessions.
References
1 Hofmann A, Frey A, Ott H, Petr Zilka T, Troxler F. Elucidation of
the structure and the synthesis of psilocybin. Experientia 1958;
14: 397–99.
2 Hofmann A, Heim R, Brack A, Kobel H. Psilocybin, a psychotropic
substance from the Mexican mushroom Psilicybe mexicana Heim.
Experientia 1958; 14: 107–09.
3 Halberstadt AL, Geyer MA. Multiple receptors contribute to the
behavioral eff ects of indoleamine hallucinogens. Neuropharmacology
2011; 61: 364–81.
4 Boulougouris V, Glennon JC, Robbins TW. Dissociable eff ects of
selective 5-HT2A and 5-HT2C receptor antagonists on serial spatial
reversal learning in rats. Neuropsychopharmacology 2008;
33: 2007–19.
5 Harvey JA. Role of the serotonin 5-HT(2A) receptor in learning.
Learn Mem 2003; 10: 355–62.
6 Vaidya VA, Marek GJ, Aghajanian GK, Duman RS. 5-HT2A
receptor-mediated regulation of brain-derived neurotrophic factor
mRNA in the hippocampus and the neocortex. J Neurosci 1997;
17 (8): 2785–95.
7 Buchborn T, Schroder H, Hollt V, Grecksch G. Repeated lysergic
acid diethylamide in an animal model of depression: normalisation
of learning behaviour and hippocampal serotonin 5-HT2 signalling.
J Psychopharmacol 2014; 28: 545–52.
8 Griffi ths R, Richards W, Johnson M, McCann U, Jesse R.
Mystical-type experiences occasioned by psilocybin mediate the
attribution of personal meaning and spiritual signifi cance
14 months later. J Psychopharmacol 2008; 22: 621–32.
9 Carhart-Harris RL, Kaelen M, Bolstridge M, et al. The paradoxical
psychological eff ects of lysergic acid diethylamide (LSD).
Psychol Med 2016; 46: 1379–90.
10 Carhart-Harris RL, Erritzoe D, Williams T, et al. Neural correlates of
the psychedelic state as determined by fMRI studies with
psilocybin. Proc Natl Acad Sci USA 2012; 109: 2138–43.
11 Hendricks PS, Thorne CB, Clark CB, Coombs DW, Johnson MW.
Classic psychedelic use is associated with reduced psychological
distress and suicidality in the United States adult population.
J Psychopharmacol 2015; 29: 280–88.
12 Krebs TS, Johansen PO. Psychedelics and mental health: a population
study. PLoS One 2013; 8: e63972.
13 Bouso JC, Gonzalez D, Fondevila S, et al. Personality, psychopathology,
life attitudes and neuropsychological performance among ritual users
of Ayahuasca: a longitudinal study. PLoS One 2012; 7: e42421.
14 Gasser P, Holstein D, Michel Y, et al. Safety and effi cacy of lysergic
acid diethylamide-assisted psychotherapy for anxiety associated with
life-threatening diseases. J Nerv Ment Dis 2014; 202: 513–20.
15 Grob CS, Danforth AL, Chopra GS, et al. Pilot study of psilocybin
treatment for anxiety in patients with advanced-stage cancer.
Arch Gen Psychiatry 2011; 68: 71–78.
16 Osorio Fde L, Sanches RF, Macedo LR, et al. Antidepressant eff ects
of a single dose of ayahuasca in patients with recurrent depression:
a preliminary report. Rev Bras Psiquiatr 2015; 37: 13–20.
17 Moreno FA, Wiegand CB, Taitano EK, Delgado PL. Safety, tolerability,
and effi cacy of psilocybin in 9 patients with obsessive-compulsive
disorder. J Clin Psychiatry 2006; 67: 1735–40.
18 Johnson MW, Garcia-Romeu A, Cosimano MP, Griffi ths RR.
Pilot study of the 5-HT2AR agonist psilocybin in the treatment of
tobacco addiction. J Psychopharmacol 2014; 28: 983–92.
19 Bogenschutz MP, Forcehimes AA, Pommy JA, Wilcox CE, Barbosa P,
Strassman RJ. Psilocybin-assisted treatment for alcohol dependence:
a proof-of-concept study. J Psychopharmacol 2015; 29: 289–99.
20 Johnson M, Richards W, Griffi ths R. Human hallucinogen research:
guidelines for safety. J Psychopharmacol 2008; 22: 603–20.
21 Greenberg PE, Fournier AA, Sisitsky T, Pike CT, Kessler RC.
The economic burden of adults with major depressive disorder in
the United States (2005 and 2010). J Clin Psychiatry 2015; 76: 155–62.
22 Gaynes BN. Identifying diffi cult-to-treat depression:
diff erential diagnosis, subtypes, and comorbidities. J Clin Psychiatry
2009; 70 (suppl 6): 10–15.
23 Sackeim HA. The defi nition and meaning of treatment-resistant
depression. J Clin Psychiatry 2001; 62 (suppl 16): 10–17.
24 Studerus E, Gamma A, Vollenweider FX. Psychometric evaluation
of the altered states of consciousness rating scale (OAV). PLoS One
2010; 5: e12412.
25 Rush AJ, Trivedi MH, Ibrahim HM, et al. The 16-Item Quick Inventory
of Depressive Symptomatology (QIDS), clinician rating (QIDS-C), and
self-report (QIDS-SR): a psychometric evaluation in patients with
chronic major depression. Biol Psychiatry 2003; 54: 573–83.
26 Ibrahim L, DiazGranados N, Franco-Chaves J, et al. Course of
improvement in depressive symptoms to a single intravenous
infusion of ketamine vs add-on riluzole: results from a 4-week,
double-blind, placebo-controlled study. Neuropsychopharmacol 2012;
37: 1526–33.
27 Johnson MW, Sewell RA, Griffi ths RR. Psilocybin dose-dependently
causes delayed, transient headaches in healthy volunteers.
Drug Alcohol Depend 2012; 123: 132–40.
28 Carhart-Harris RL, Kaelen M, Whalley MG, Bolstridge M,
Feilding A, Nutt DJ. LSD enhances suggestibility in healthy
volunteers. Psychopharmacology 2015; 232: 785–94.
29 Bonson KR, Buckholtz JW, Murphy DL. Chronic administration of
serotonergic antidepressants attenuates the subjective eff ects of
LSD in humans. Neuropsychopharmacology 1996; 14: 425–36.