Every outcome registered for Compass Pathways’ COMP005 and COMP006 trials and the Usona Institute’s uAspire trial is measured at about Week 6, which is Day 43 in uAspire. None of the three ClinicalTrials.gov records lists an outcome measured later, although two say durability will be assessed. A registry is not a protocol, and none of the three records attaches one, so this describes what the public record shows, not what the sponsors will analyze.
Durability is part of the case Compass makes for COMP360. Its September release describes “an option that could provide durable benefit with just a few treatments a year.” The term can mean three different things: how long one dose lasts untreated, how long until relapse, or how patients do under a regimen of repeat treatment. The three designs support different answers, because in each the period after Week 6 can include more treatment. Compass’s trials stay blinded to Week 26, allow retreatment and antidepressants, then go open-label to Week 52. Usona describes a six-week double-blind period followed by a year of monitoring, with open-label re-administration for participants who meet criteria it has not published. This piece compares what the records say. It does not compare results, because uAspire has none public, and the trials enroll different patients: major depressive disorder without a treatment-resistance requirement for Usona, treatment-resistant depression for Compass.
What each registry records
| Trial | Registered outcomes | After the primary endpoint |
|---|---|---|
| uAspire (Usona, NCT06308653) | Day 43: central-rater MADRS (primary); CGI-S and SDS (secondary) | One-year follow-up, no registered outcome; optional open-label 25 mg re-administration |
| COMP005 (Compass, NCT05624268) | Week 6: MADRS, 25 mg vs placebo (primary); SDS (secondary) | Blinded to Week 26 with retreatment option; open-label to Week 52 |
| COMP006 (Compass, NCT05711940) | Week 6: MADRS, 25 mg vs 1 mg (primary); SDS and 25 mg vs 10 mg MADRS (secondary) | Two doses in Part A; blinded to Week 26 with retreatment; open-label to Week 52 |
MADRS is the Montgomery-Asberg Depression Rating Scale, CGI-S a clinician severity rating and SDS the Sheehan Disability Scale. Source: ClinicalTrials.gov records, last posted October 21 (uAspire) and August 10 (COMP005 and COMP006), read November 4, 2026; Compass 8-K of July 7 and releases of September 9 and November 3; Usona statement of March 20, 2024. No record has posted results or attached a protocol or statistical analysis plan, and the registry’s change history for these records was not checked.
What Compass’s design can isolate
COMP005 gave one 25 mg dose or placebo, 2:1. Compass says Part A was blinded through Week 6 and Part B through Week 26, with eligible participants able to receive another dose matching their original assignment, and Part C, from Week 26 to Week 52, was open-label. The COMP005 registry says “Durability of efficacy and long-term safety, and the efficacy and safety of re-treatment will be assessed” in Part B. COMP006 is not a single-dose design: participants received two doses three weeks apart, at 25, 10 or 1 mg, in a nine-week Part A that contains the Week 6 primary, followed by a blinded Part B to Week 26. Compass says participants in both trials could take protocol-allowed antidepressants in Parts B and C.
That structure gives Compass a blinded comparison between arms to Week 26. It does not isolate how long an initial regimen lasts untreated. In COMP006, Compass says 58% of the 25 mg arm was retreated after Week 9, on prespecified criteria, with participants receiving “either what they were randomized into or antidepressant.” Its Week 26 view of Week 6 responders is labeled post hoc, meaning it was not the trial’s prespecified test, and the Part B curves use observed data only, so they describe the participants still being measured. Part C is more limited still. Compass reports an average 13-point reduction at Week 52 for the 90 participants from the COMP005 25 mg arm who received another open-label dose, and about 10 points for the 52 placebo-arm participants after a first 25 mg dose. About 70% of the COMP005 cohort continued past Week 26, and neither figure has a concurrent control.
One number shows how much depends on stated bases. Compass says 39% of the COMP006 25 mg arm had a clinically meaningful response at Week 6 and, on average, maintained it. The July 7 slide behind it lists 116 responders and remitters among 269 participants in a Week 6 analysis set it labels “based on ITT,” meaning intention-to-treat. Out of 269 that is 43%. The same 116 is 39% of the 296 participants dosed in the arm, which fits Compass’s description of the figure as a share of the arm, but Compass has not said which base it used. This desk has covered the open-label Part C data and the COMP006 Week 26 analysis in detail.
What Usona’s registry leaves open
uAspire randomizes participants to one dose of 25 mg, 5 mg or inactive placebo, with a central-rater MADRS primary at Day 43. The registry gives as a purpose to “characterize the durability of initial treatment effect and subsequent response to optional Psilocybin 25 mg re-administration(s)” over the one-year follow-up. Usona’s March 2024 statement describes “an initial 6-week randomized, double-blind treatment period” and a year of monitoring “to characterize the treatment durability and long-term safety.” Participants who meet “pre-defined MDD severity criteria” and re-administration eligibility criteria may be offered open-label 25 mg, repeatedly, whatever arm they came from.
The registry also says full blinding of participants, personnel and the sponsor “will be maintained until database lock at the conclusion of the trial.” That sentence could mean allocation stays concealed through follow-up. Usona’s own description limits the double-blind period to six weeks, and open-label 25 mg is offered to placebo and 5 mg participants who meet criteria. The record does not say how those fit together, whether follow-up assessments are prespecified endpoints, or how participants who cross to open-label 25 mg are handled in an analysis. A comparison of the initial dose alone beyond Day 43 depends on analysis rules that are not public.
Where the FDA guidance sits
FDA’s guidance Psychedelic Drugs: Considerations for Clinical Investigations was finalized in July 2026 (notice at 91 FR 43101, docket FDA-2023-D-1987). It says the most informative design would incorporate blinded long-term follow-up, typically 12 months, with prespecified retreatment criteria that account for symptom ratings and for outside treatment. Its June 2023 draft described an open-label extension for a year beyond Week 12 instead, and the final adds that sponsors should evaluate the 12-week effect double-blind before an initial application. This desk’s coverage of the guidance covers that change. The guidance is nonbinding and names neither company. COMP005 and COMP006 were registered before the draft appeared, in November 2022 and January 2023, and uAspire in March 2024, after it. All three finished primary data collection before the final: COMP005 on May 28, 2025, COMP006 on November 13, 2025 and uAspire on January 5, 2026. Whether any protocol was amended after the draft or the final is not public. The guidance describes a design FDA considers informative. It does not evaluate these trials, and neither does this article.
What the public record does not show
- uAspire results, including whether any follow-up outcome will be reported and whether it will be analyzed by original assignment.
- Whether allocation is concealed during uAspire’s follow-up, and how re-administration enters the analysis.
- Retreatment or re-administration criteria for either company. Compass calls its criteria prespecified and has not published them.
- How Compass’s Phase 3 trials were rated and assessed. The registry says outcomes assessors were masked. The filings reviewed do not say whether central raters were used, whether blinding or expectancy questionnaires were collected, or whether a double-blind assessment at Week 12 exists. No chart in the July 7 presentation plots one.
- Protocols and statistical analysis plans. None is attached to any record.
What to watch
Compass says it will complete its new drug application within two weeks of its November 3 results release. Usona has not said when uAspire results will be shared. When either arrives, three questions will show how much it can establish about durability: which analysis was prespecified, which data come from open-label treatment, and what the denominator is.