Last week's EPIsoDE readout did what honest blinding tends to do to a psychedelic trial: it took the shine off. Germany's triple-blind study, built with an active placebo precisely to stop patients guessing their way to a response, could not separate psilocybin from placebo on its primary endpoint in treatment-resistant depression. That is the indication every commercial programme is built on. So the question worth asking this week is not whether psychedelics help in TRD, but who is carrying that indication all the way through Phase 3, and whether the effect survives a hard look. The registries answer it more plainly than the press releases do.

Compass is already through the gate everyone else is still staring at. Both pivotal COMP360 psilocybin trials in TRD have hit their primary endpoints, the February 2026 COMP006 readout by −3.8 points on the MADRS (p<0.001), a result that is statistically clean and clinically modest in the same breath. Both trials (NCT05711940, enrolment closed at an actual 572 participants; NCT05624268, an estimated 255) now sit ACTIVE_NOT_RECRUITING, and the company is in FDA rolling NDA review. The pivotal trial is US-anchored, 62 of its 116 sites American against 45 European and 9 Canadian, but it is still the programme whose data Europe reads first, and the one that sets the tone for everyone behind it.

Now read what everyone else actually registered. Cybin's CYB003 (NCT06793397) is recruiting, but for major depressive disorder, not TRD, a broader and less battle-scarred population. Compass's own newest Phase 3, the Redefine study (NCT07570654), is not depression at all, it is PTSD. And AtaiBeckley's BPL-003, the TRD hopeful, still has no Phase 3 record on ClinicalTrials.gov at all.

The releases all read "Phase 3 on track." The registries reveal the strategy underneath: the field is quietly routing around resistant depression, into broader MDD, into PTSD, into guidance, while Compass, having already cleared the efficacy bar, is the one programme committed to carrying TRD the whole way. So the thing to watch is no longer whether COMP360 works on the primary endpoint, it does, modestly, but whether the effect lasts: the 26-week COMP006 durability data, due early Q3, and then the NDA decision. That readout lands exactly as EPIsoDE has raised the bar on whether a psilocybin signal in TRD holds up under honest scrutiny. If durability disappoints, the most advanced programme in Europe will also be the most exposed.

One name in that lead needs a footnote. As covered in our H1 review, atai and Beckley Psytech combined in November 2025 to form AtaiBeckley Inc. (NASDAQ: ATAI), folding both into one pipeline: BPL-003 (intranasal 5-MeO-DMT, TRD), EMP-01 (oral R-MDMA, social anxiety) and VLS-01, with cash guided into the early-2029 Phase 3 readouts. The point for this week is narrower than that recap. Of the three TRD-and-adjacent contenders above, AtaiBeckley is the only one without a Phase 3 on the registry yet, which is exactly what makes its BPL-003 timeline worth watching closely (see The Call). (AtaiBeckley Q1)

Cybin takes CYB003 into Phase 3, with European sites in six countries. Cybin's deuterated psilocin analogue is recruiting for a Phase 3 in major depressive disorder, with listed sites in Czechia, Germany, Greece, Ireland, Poland and the UK. That an Irish-incorporated sponsor, Cybin IRL Limited, is running a six-country European Phase 3 is the clearest sign yet that Europe is a primary theatre for late-stage psychedelic development now, not a place sponsors bolt on for extra recruitment. (NCT06793397, registry updated 11 June)

France lines up a psilocybin Phase 3 in alcohol use disorder. Centre Hospitalier Universitaire de Nîmes has registered a Phase 3, not yet recruiting, testing psilocybin for relapse prevention in alcohol use disorder with depressive symptoms. France keeps building an academic-led footprint, and the most interesting EU work, as this is a reminder, does not only come from listed companies. (NCT07638553, registry updated 10 June)

Sweden pushes an oral, slow-release ketamine toward its Phase 2 readout. A Lund University trial led by Daniel Lindqvist, testing add-on oral slow-release ketamine tablets in major depression, is now active and no longer recruiting. It is a shot at a take-home ketamine format, the form factor that would matter most for access and the hardest to get right, if it survives Phase 2. (NCT07396272, registry updated 18 June)

Ten. The number of European countries with listed sites in Compass's COMP360 Phase 3 (NCT05711940): Czechia, Denmark, France, Germany, Ireland, the Netherlands, Poland, Spain, Sweden and the UK. A single psychedelic Phase 3 spanning ten European jurisdictions means ten sets of trial, import and controlled-substance rules at once, the operational tax Europe's fragmentation charges, expressed in one number. The trial is US-anchored overall (62 of 116 sites American, 45 European, 9 Canadian), but ten countries is a real bloc, not a token presence. (NCT05711940)

Germany / BfArM. A psilocybin compassionate-use pathway has been reported, but primary source pending: we are not running it as fact until it is confirmed beyond a 16 June secondary interview. Watching for a BfArM or clinic document.

EU CTR / EMA. Nothing surfaced, and that is a blind spot rather than a finding. We do not yet have a working EMA feed (every RSS path returns an antibot error), so read this as "nothing we can see," not "nothing moved." Closing that gap is on the list.

One falsifiable prediction per issue. Scored in the next one.

Scoring Issue 2. Last issue I called it: AtaiBeckley will not announce a first BPL-003 Phase 3 patient dosed before 30 June, and I said in print that if a dosed-patient release landed this month I would mark myself wrong here. So I checked. There is no such announcement. No press release to that effect, and still no BPL-003 Phase 3 record on ClinicalTrials.gov, not recruiting, not even registered. The most recent guidance (the 10 and 17 March releases, restated in the 12 May Q1 results) says only that Phase 3 "initiation" is "on track for Q2 2026." Initiation is not dosing, and guidance is not a trial. The trigger I set hasn't fired, so the call stands and resolves 30 June when Q2 closes. If a dosing release surfaces in the final week, I mark the miss in Issue 4. As of today, the bet is good.

This week's call. AtaiBeckley will report its first BPL-003 Phase 3 patient dosed in Q3 2026, after 30 June and before 30 September. Same logic that has held so far: the Q2 initiation language stays alive right up to the wire, but the gap between "initiated" and "dosed" slips into the next quarter.

If something here is wrong, tell me and I'll correct it openly.

Thank you so much for reading.

Dhruv

Sources