On 7 July, Compass Pathways reported six-month data from COMP006, the second of its two Phase 3 psilocybin trials in treatment-resistant depression, and for the first time attached a hard commercial date to a classic psychedelic. It now says a COMP360 launch is anticipated in the first half of 2027 subject to FDA approval and following DEA rescheduling. For Europe, that date matters more than the efficacy line does. Compass is the furthest-along program the whole European field measures itself against, and "the first half of 2027" is the first concrete you-could-actually-prescribe-this milestone anyone in the West has put on the record.

The data underneath is more modest than "rapid and durable" implies. COMP006's headline was already known and we covered it when the topline landed: a 3.8-point MADRS separation at Week 6, 25mg over the trial's own 1mg comparator. The new part is persistence. Compass says 39% of the 25mg arm achieved a clinically meaningful response, defined as a 25% or greater MADRS reduction, at Week 6, and that this group maintained benefit through at least Week 26; nearly 30% of those Week-6 responders later reached remission after retreatment. Read that carefully: the 39% is a Week-6 response rate whose benefit is said to persist, not a fresh count of who was still responding at six months. Even so, on the trial's own framing fewer than four in ten of the top-dose patients cleared even a soft response bar at Week 6. "Durable" is carrying a lot of weight.

And the comparator is the thread this newsletter keeps pulling. COMP006 had no inert placebo. Its 3.8-point advantage is measured against a 1mg dose of the same drug, a low active dose chosen partly because a patient cannot easily be sure they got the small one. That is the same expectancy-management design we flagged in Definium's Panorama call two issues ago, now sitting directly under the durability claim of the only psychedelic with a 2027 launch date. Watch two things next: whether the final rolling NDA submission actually closes in Q4, and whether DEA rescheduling, which has no public start, can realistically clear in time for a first-half-2027 launch.

Two Phase 3s, and they are not the same test. Compass ran COMP005 and COMP006, and it is worth keeping them apart before anyone nets the numbers. COMP005 is the placebo-controlled pivotal, a 3.6-point MADRS separation against an inert placebo. COMP006, the one that just read out at six months, is a dose comparison: 25mg versus 10mg versus 1mg, dosed three weeks apart, with the Week 9 to Week 26 window blinded and a retreatment option built in. Its 3.8-point result is 25mg over 1mg, not over placebo. That design is defensible, even clever, because a fixed inert placebo makes unblinding trivial in a drug patients can plainly feel. But it changes what "confirmed" means. COMP006 confirms the effect is dose-related and that it persists; it does not, on its own, show the drug beats doing nothing. For a reimbursement conversation, and Europe's payers are tougher than the FDA, six-month durability is exactly the number that matters. It is also the number most exposed if the response bar turns out to be soft.

AtaiBeckley moves VLS-01 toward Phase 3. The company dosed the last patient in the VLS-01 Phase 2b in TRD and plans a Phase 3 in MDD. The field-wide bet is shrinking the session toward something a clinic can schedule, and a short-duration candidate going late-stage is that bet reaching the pivotal stage. It gives AtaiBeckley a second late-stage asset behind BPL-003.

Cybin is building its own durability dataset. A Phase 3 long-term extension, EXTEND (NCT06605105), for CYB003, is the adjunctive camp's counterpart to the durability question Compass just answered for the washout camp. Cybin's EU exposure runs through what appears to be the EMBRACE trial (NCT06793397). Both of last week's two camps now race the same six-month endpoint EU payers will price against.

1mg. The dose Compass used as the comparator in COMP006, the trial now anchoring a six-month durability claim and a 2027 launch. There was no inert placebo: the 3.8-point MADRS advantage that cleared the primary endpoint is 25mg over this 1mg dose of the same drug (arms of roughly 296, 142 and 143 patients across the 25mg, 10mg and 1mg groups, 581 dosed in total). What it reveals: the pivotal that "confirms" the durable profile was built to compare psilocybin against a smaller amount of psilocybin. That manages unblinding, which is a real problem, but it cannot separate how much of the effect is the drug from how much is the expectation of it. For a payer weighing a six-month benefit, that distinction is close to the whole ballgame.

US / FDA + DEA: Compass expects to complete its rolling NDA submission in Q4 2026 and guides a COMP360 launch "anticipated in the first half of 2027 subject to FDA approval and following DEA rescheduling." The DEA step is the soft spot: rescheduling psilocybin is a separate federal process that, on the public record, carries no formal notice as of this writing. An H1 2027 launch needs both an FDA approval and a DEA reschedule to land inside roughly eighteen months.

UK / MHRA: COMP360's ILAP designation stands (context, not news).

EU / EMA: Compass has disclosed no EMA filing for COMP360; its European exposure is COMP006 trial sites, not a regulatory dossier. We still have no working EMA feed our end, so read this as "nothing disclosed," not "nothing moving."

Germany / reimbursement: The Bundestag passed a system-wide statutory health-insurance cost-containment reform this week, raising mandatory manufacturer rebates and tightening reimbursement, which drugmakers warn will squeeze prices in the EU's largest market. It is not psychedelics-specific and does not target the category, but it is the reimbursement climate a 2027 COMP360 launch would enter in Europe, and it is moving the wrong way for a premium-priced, clinic-administered therapy.

Germany / BfArM: The psilocybin compassionate-use pathway is still pending a primary source beyond secondary reporting. Carried from Issues 3 to 5, not yet confirmed.

One falsifiable prediction per issue, scored in the next.

Scoring the open calls: three calls are live and none has resolved. Issue 3 (first BPL-003 Phase 3 patient dosed in Q3 2026) resolves 30 September, open. Issue 4 (Panorama posts a smaller placebo-adjusted effect than Voyage) resolves once both GAD toplines are public, neither has read out, open. Issue 5 (BPL-003's Phase 3 registers as adjunctive, no SSRI washout) resolves when the protocol posts, expected H2 2026, open. Nothing due this week, nothing to score.

This week's call: COMP360 will not be commercially available to prescribe in the US by 30 June 2027. This is our call, an analyst scenario and our opinion, not confirmed timing and not something Compass has forecast. The company guides to a launch in the window, and we are calling that it slips. In plain terms: the "first half of 2027" target does not hold. Compass's own gating language is "subject to FDA approval and following DEA rescheduling," and the final NDA piece only lands in Q4 2026. DEA rescheduling of psilocybin is a distinct process that, as of this writing, carries no public notice, and it has historically run slowly. Our thesis is that both cannot be cleared in time. Falsifiable and checkable: if COMP360 is approved and commercially available to prescribe in the US on or before 30 June 2027, we are wrong. Resolve: 30 June 2027, or earlier if launch happens.

If something here is wrong, tell us and we will correct it in the open.

Thank you so much for reading.

Dhruv Shekhawat