COMP 001

COMPASS Pathfinder Limited

3rd Floor, 1 Ashley Road, Altrincham, Cheshire, United Kingdom, WA14 2DT

Guy Goodwin, COMPASS Pathways, Ltd, +44 7443 136539, info@compasspathways.com

Guy Goodwin, COMPASS Pathways, Ltd, +44 7443 136539, info@compasspathways.com

No

No

No

Final

22 Sep 2022

No

Yes

27 Sep 2021

No

The primary objective of this study was to evaluate the efficacy of COMP360 (25 mg and 10 mg) compared to 1 mg, administered under supportive conditions to adult participants with TRD, in improving depressive symptoms, as assessed by the change in the Montgomery-Asberg Depression Rating Scale (MADRS) total score from baseline. Baseline was defined as the assessment score obtained on Day -1. The primary timepoint was Week 3. The MADRS was also analysed for the change from baseline to Day 2, and Weeks 1, 6, 9, and 12.

During the screening period, participants who are on antidepressant medications will be expected to complete the taper at least 2 weeks prior to Baseline (V2). Participants will be given a choice of the tapering rate. During the tapering period all participants will be supported by the study clinician. The designated study team member will be in frequent contact with the participants to monitor for withdrawal and worsening of depression symptoms. All participants will be evaluated for safety at the clinic weekly for a minimum of 3 weeks prior to COMP360 administration to ensure safe discontinuation of current antidepressant therapy required by the protocol. Participants’ companions (friend or family member) will be educated about the signs of worsening of depression and suicidality, and instructed on ways to contact the study team in case of significant worsening of depression. The COMP360 administration session will be supported by a trained therapist who will be present with the participant at all times until discharge. Rescue medications are allowed during the study as described in the protocol. Efficacy and safety assessments including Adverse Events are performed throughout the study up to the end of study.

03 Jan 2019

Yes

Yes

Canada: 6

United States: 96

Netherlands: 50

Portugal: 2

Spain: 16

United Kingdom: 33

Czechia: 7

Denmark: 8

Germany: 4

Ireland: 11

233

98

0

0

0

0

0

0

226

7

0

Overall trial (overall period)

Randomised - controlled

Bottles (each containing 5 capsules for a single dose administration) for each participant will be assigned according to unique identifiers by IWRS programmed with blind-breaking instructions. Study assessments were performed by investigators who were blinded to treatment assignment.

Yes

COMP360

Psilocybin

Capsule

Oral use

COMP360 25mg single dose treatment: 5 × 5 mg capsules

COMP360

Psilocybin

Capsule

Oral use

COMP360 10mg single dose treatment: 2 × 5 mg capsules and 3 × placebo capsules

COMP360

Psilocybin

Capsule

Oral use

COMP360 1 mg single dose treatment: 1 × 1 mg capsule and 4 × placebo capsules

COMP360 25mg

COMP360 10mg

COMP360 1mg

Main Efficacy Analysis Set (Full Analysis Set-FAS)

All participants randomised who receive study drug and have at least 1 post-dose efficacy assessment. Outputs based on this analysis set will use the treatment the study participant was randomised to.

Secondary Efficacy Analysis Set (Per Protocol Analysis Set)

All participants in the FAS who do not have a protocol deviation (PD) that is thought to significantly affect the integrity of the participant’s efficacy data.

Safety Analysis Set

All randomised participants who receive study drug. Outputs based on this analysis set will use the actual treatment received by the study participant.

Randomised Analysis Set

All randomised participants, whether or not they receive study drug.

COMP360 25mg

COMP360 10mg

COMP360 1mg

Main Efficacy Analysis Set (Full Analysis Set-FAS)

All participants randomised who receive study drug and have at least 1 post-dose efficacy assessment. Outputs based on this analysis set will use the treatment the study participant was randomised to.

Secondary Efficacy Analysis Set (Per Protocol Analysis Set)

All participants in the FAS who do not have a protocol deviation (PD) that is thought to significantly affect the integrity of the participant’s efficacy data.

Safety Analysis Set

All randomised participants who receive study drug. Outputs based on this analysis set will use the actual treatment received by the study participant.

Randomised Analysis Set

All randomised participants, whether or not they receive study drug.

Primary

Baseline (Day -1) to 3 weeks post COMP360.

Mixed Model for Repeated Measures model includes treatment, visit, pooled study site, treatment*visit, baseline (day -1) MADRS total score and an unstructured correlation. Data after the start of new treatment for depression or withdrawal (lack of efficacy or adverse event) was imputed 100 times under MNAR. Remaining missing data were imputed under MAR using MCMC methods with a non-informative prior. The imputed datasets were analysed using the MMRM model and estimates pooled using Rubin's rules

COMP360 25mg v COMP360 1mg

158

Pre-specified

-6.6

2-sided

Standard error of the mean

1.86

Mixed Model for Repeated Measures model includes treatment, visit, pooled study site, treatment*visit, baseline (day -1) MADRS total score and an unstructured correlation. Data after the start of new treatment for depression or withdrawal (lack of efficacy or adverse event) was imputed 100 times under MNAR. Remaining missing data were imputed under MAR using MCMC with a non-informative prior. The imputed datasets were analysed using the MMRM model and estimates pooled using Rubin's rules.

COMP360 1mg v COMP360 10mg

154

Pre-specified

-2.5

2-sided

Standard error of the mean

1.87

A MADRS responder is a participant with a ≥ 50% reduction in baseline MADRS total score at Week 3 post COMP 360.

Secondary

Week 3 post COMP360

A GLMM using a binomial distribution and a logit link was fit. The model included treatment, pooled study site, visit, treatment*visit, baseline (day -1) MADRS total score as covariates and an unstructured correlation. Participants starting new treatment for depression or withdrawing (lack of efficacy or adverse event) were imputed as non-responders. Missing MADRS values imputed using MAR methods for the primary endpoint were dichotomised. Results were pooled using Rubin's rules

COMP360 25mg v COMP360 1mg

133

Pre-specified

2.9

2-sided

A GLMM using a binomial distribution and a logit link was fit. The model included treatment, pooled study site, visit, treatment*visit, baseline (day -1) MADRS total score as covariates and an unstructured correlation. Participants starting new treatment for depression or withdrawing (lack of efficacy or adverse event) were imputed as non-responders. Missing MADRS values imputed using MAR methods for the primary endpoint were dichotomised. Results were pooled using Rubin's rules.

COMP360 1mg v COMP360 10mg

123

Pre-specified

1.2

2-sided

Remitters are defined as having a MADRS total score ≤ 10 at Week 3 post COMP360.

Secondary

Week 3 post COMP360

A GLMM using a binomial distribution and a logit link was fit. The model included treatment, pooled study site, visit, treatment*visit, baseline (day -1) MADRS total score as covariates and an unstructured correlation. Participants starting new treatment for depression or withdrawing (lack of efficacy or adverse event) were imputed as non-remitters. Missing MADRS values imputed using MAR methods for the primary endpoint were dichotomised. Results were pooled using Rubin's rules.

COMP360 25mg v COMP360 1mg

133

Pre-specified

4.8

2-sided

A GLMM using a binomial distribution and a logit link was fit. The model included treatment, pooled study site, visit, treatment*visit, baseline (day -1) MADRS total score as covariates and an unstructured correlation. Participants starting new treatment for depression or withdrawing (lack of efficacy or adverse event) were imputed as non-remitters. Missing MADRS values imputed using MAR methods for the primary endpoint were dichotomised. Results were pooled using Rubin's rules.

COMP360 1mg v COMP360 10mg

123

Pre-specified

1.2

2-sided

Sustained responders are participants meeting the MADRS response criteria (a ≥ 50% reduction from baseline in MADRS total score) at any visit up to and including week 3 and also at all visits after week 3 until week 12.

Secondary

At Week 12

A logistic regression with treatment and baseline (day -1) MADRS total score as covariates was fit. Participants starting new treatment for depression or withdrawing (lack of efficacy or adverse event) were imputed as non-responders. Sustained responder status for all other participants was derived from the multiple imputed dataset as obtained for the main analysis of the Composite Strategy estimand of the MADRS responder endpoint. Results were pooled across imputations using Rubin's rules.

COMP360 25mg v COMP360 1mg

90

Pre-specified

2.2

2-sided

A logistic regression with treatment and baseline (day -1) MADRS total score as covariates was fit. Participants starting new treatment for depression or withdrawing (lack of efficacy or adverse event) were imputed as non-responders. Sustained responder status for all other participants was derived from the multiple imputed dataset as obtained for the main analysis of the Composite Strategy estimand of the MADRS responder endpoint. Results were pooled across imputations using Rubin's rules.

COMP360 1mg v COMP360 10mg

88

Pre-specified

0.7

2-sided

From ICG sign-off to last study event

Any AEs occurring before the start of treatment (ie, before the dose of the IP on Day 0 [V3]) was recorded in the medical history. Any AE ongoing at V10 (EOS/ET) was followed until resolution or no longer considered clinically significant by the investigator.

23.0

COMP360 25mg

COMP360 10mg

COMP360 1mg