COMPASS Pathways Reports Positive Phase 3 Results for Psilocybin in Treatment-Resistant Depression

COMPASS Pathways reported positive results from two Phase 3 studies evaluating COMP360 psilocybin in treatment-resistant depression, with statistically significant improvements in depression scores, rapid onset of effect, and signs of durability through 26 weeks in a subset of patients. The company has submitted Phase 3 data to the FDA and is seeking a rolling NDA submission and expedited review under Breakthrough Therapy designation.

The Phase 3 program included two studies: COMP005, a placebo-controlled trial evaluating a single 25 mg administration of COMP360 with a primary endpoint at six weeks, and COMP006, a three-arm study evaluating 10 mg and 25 mg doses with two administrations given three weeks apart and a six-week primary endpoint. TRD in the studies was defined as failure of two to four prior treatments in the current depressive episode, with moderate-to-severe symptoms (MADRS total score >20 at screening and baseline) and discontinuation of existing antidepressants. COMP005 enrolled U.S. patients while COMP006 enrolled patients across Europe, the U.S., and Canada.

In COMP005, the company reported a 3.6-point difference on the Montgomery–Åsberg Depression Rating Scale (MADRS) between the 25 mg group and placebo at week six. The separation was statistically significant at every reported time point, starting the day after administration (day 2) and continuing through the six-week primary endpoint. In COMP006 Part A, the company reported statistically significant separation at every time point starting the day after administration through week six. At week six, the 25 mg group showed a 3.8-point difference on MADRS compared to a 1 mg active comparator.

The COMP006 design used 1 mg as an active comparator rather than an inert placebo to reduce unblinding and strengthen claims of pharmacologic effect. MADRS assessments were conducted remotely by blinded raters. The company said 25 mg showed statistically significant differences versus the 1 mg comparator at every time point beyond baseline, with a numerical dose-response pattern (25 mg > 10 mg > 1 mg) at one, three, and six weeks.

Beyond the six-week endpoint, longer-term follow-up from COMP005 Part B extended to week 26 and allowed for up to one retreatment while maintaining blinding. The 26-week data indicated that a subset of patients maintained clinically meaningful benefit with one or two treatments through at least six months. Among patients meeting a "clinically meaningful benefit" threshold of at least a 25% reduction in MADRS who received a second administration, 40% subsequently entered remission. The trial showed numerical separation at all time points up to 26 weeks, with most retreatments occurring between weeks 10 and 14.

Across trials, COMPASS highlighted the proportion achieving what it defined as a clinically meaningful reduction in MADRS at week six. In COMP005, 25% of patients in the 25 mg arm achieved a clinically meaningful reduction at week six, sustained through 26 weeks. In COMP006, 39% achieved a clinically meaningful reduction at week six, which the company linked to the fixed second dose in the trial design.

The company defended the use of a 25% MADRS reduction as a marker of clinical meaningfulness, stating it was based on published "crosswalk" work in an esketamine dataset linking changes in MADRS with other measures such as the Clinical Global Impressions-Severity (CGI-S), PHQ-9, and Sheehan Disability Scale. Using an average baseline MADRS around 32, the company calculated that a six-point improvement is about 19%, and said a 25% threshold sits "comfortably above that."

COMP360 was generally well-tolerated, with most adverse events occurring on the dosing day and resolving within a day. Serious suicidal-ideation events were under 1% and blood-pressure elevations were transient and manageable. The dataset now includes more than 1,000 participants across the two Phase III studies and a prior Phase IIb trial, with the company meeting its primary endpoint "three for three" with high statistical significance.

The company framed the opportunity around what it called a large unmet need in TRD, citing an estimate that more than 4 million U.S. adults live with TRD each year. The company said many therapies have failed in this population and noted that only one marketed medicine is currently approved and used for TRD. COMPASS expects 26-week durability data from COMP006 in early Q3 as the last gating item while preparing launch operations.