Last week we wrote that assessment requires a filing and filing requires an approval. That is true of Germany and not true generally, and we should have said so.

Germany's §35a benefit assessment is bound to medicines with new active substances being placed on the market, so no procedure can open for a psilocybin or MDMA product today. That part holds. Britain works differently. NICE's own methods manual states that a medicine "is only evaluated if it has or is expected to have regulatory approval (or appropriate regulatory signal) by the planned draft or final guidance publication date," and that its scope "includes information about the regulatory approval of the technology, and the stage of approval for technologies that have not yet received approval." Germany also has a second route we did not mention. Assessments of examination and treatment methods run under a different part of the code, without the same starting condition.

Our Issue 8 call stands. The reasoning behind it was narrower than the sentence we published.

Issue 9 / Week of 28 July to 3 August 2026

Britain built its assessment system to start early. NICE begins scoping a medicine before a licence exists, working against an expected approval rather than an existing one. The register of psychedelic therapies in that process is empty.

We assumed that was Britain's doing. It isn’t.

Compass Pathways holds the most advanced psilocybin programme in the world. Two completed Phase 3 trials, a rolling NDA with final submission guided for the fourth quarter of this year, and launch guided to the first half of 2027. Its most recent quarterly filing, covering the period to 31 March, states that the company "has not previously submitted a new drug application, or NDA, to the FDA, or a marketing authorisation application, or MAA, to the EMA or the MHRA."

That is a statement of record, not of intent. It sits in a risk factor; it covers the American filing as well as the European ones, and it was true four months ago. What it establishes is narrow and worth having: as of the last document Compass filed, nothing had gone to either European regulator.

What the company does say about geography is this, from its annual report: if it obtains FDA approval for COMP360, its first launch market will be the United States. First implies a second. The filing names none, dates none, and discloses no European submission.

Read that carefully, because the absence is specific rather than general. Compass retains global development and commercialisation rights, so the silence cannot be explained away by a partner holding Europe. The company is headquartered in London. It holds a UK regulatory designation it recites as a credential in its own boilerplate. It has run trial sites across Europe for years. Every element of a European strategy is present except the strategy.

Then there is the detail that tells you the most. Twelve analysts took questions on the last earnings call. Advisory committee timing, DEA scheduling, competition, CPT codes, REMS staffing, retreatment, labelling, centre capacity, buy-and-bill economics, room turnover, the priority voucher clock, PTSD endpoints. Not one asked about Europe.

That is what a market priced at zero sounds like from inside the room. Not a decision anyone announced. Not a door anyone closed. A question nobody thought was worth the airtime.

We have spent three issues on this. Lilly wrote no European milestone into a $3.8 billion deal. Europe pays for clinic-administered psychiatry but not for psychedelics, because no manufacturer has asked it to. And now the field's most advanced programme has reached the edge of approval with no European application on file. Those are not three stories. They are one, getting clearer.

One honest limit. Not having filed is not the same as having decided not to. Compass has said nothing either way about future European intent, and a company under no obligation to disclose a pre-submission conversation could be having one today. We are reporting what the record shows.

Compass reports second-quarter results tomorrow, 5 August. If the company names a UK or EU filing on that call, this lead is wrong, and we will say so in Issue 10.

The machinery is not the obstacle, which is the part most coverage gets backwards.

NICE's manual is explicit that evaluation runs ahead of authorisation. Scoping normally happens during topic selection, before a licence is granted, against an expected approval. But entry is not a company's to trigger on its own. Topics reach appraisal through NICE's prioritisation process and by direction from the Department of Health and Social Care, with companies feeding in through horizon scanning and submission-readiness information.

That matters for how you read the empty register. NICE evaluates a medicine only if it has, or is expected to have, regulatory approval by the guidance publication date. With no application on file and no European filing signalled, there is no expected approval to scope against. The register is not empty because Britain declined. It is empty because the manual's own precondition has not been met.

The German picture differs in mechanism and matches in result. A benefit assessment attaches to a medicine with a new active substance being placed on the market, which presupposes authorisation. The Netherlands shows the same emptiness. We queried Zorginstituut Nederland directly and found no psilocybin or MDMA entry, with a control query on esketamine returning ten results, which is how we know the search works rather than the system being opaque.

Three countries, three different mechanisms, the same answer. Nobody has filed.

The registers are now closed, all three. We treated the Issue 8 call as provisional because we had checked Germany and Britain but not the Netherlands. All three are now verified with working controls. NICE's guidance-in-development register returns no results for psilocybin, MDMA, psilocin, COMP360 or MM120, while a control search on zuranolone returns a live procedure. The G-BA's list of live benefit assessments carries thirty procedures and none of them is a psychedelic. Zorginstituut Nederland returns nothing, with the esketamine control confirming the query functions.

Nothing moved at the regulators either. We swept EMA and Swissmedic publications for the window. Thirty-two EMA items, four Swissmedic items, not one touching psychedelics. We report that as a verified negative rather than a quiet week, because our first attempt at this sweep failed on page rendering rather than returning genuinely empty, and we would rather tell you the difference.

One thing we cannot resolve. The MHRA relaunched its Innovative Licensing and Access Pathway in January 2025 with more selective criteria, aimed at medicines that have not yet demonstrated their potential. COMP360 holds a designation from the earlier scheme and recites it in every release. What that legacy designation confers under the new pathway is not addressed in the MHRA's published guidance, and we are not going to guess at it.

Zero. Newly registered interventional psychedelic trials with a site in the EU, EEA, UK or Switzerland during the week of 28 July to 3 August. Four studies matched on terms and dates. All four were excluded, one observational and three with no European site.

We are going to be careful with this number rather than dramatic about it. We wrote in Issue 8 that one zero week is noise and two is worth writing down. We have since built the base rate, and it does not support that. Across 135 weeks since January 2024, this series averages 0.41 registrations a week, and 67 percent of all weeks are zero. Runs of three or more zero weeks have happened sixteen times. Late July and August run at roughly half the rate of the rest of the year, and August 2025 was itself a zero month.

So three consecutive zeros is the most ordinary thing this measure does. It is not a standstill, and we will not sell it as one. The number is worth publishing because the base rate is worth knowing, and because a measure you only cite when it is dramatic is not a measure.

UK. No psychedelic therapy in NICE guidance development. The pre-authorisation route exists and is unused, for the reason set out above. As of its last quarterly filing, Compass had not submitted an application to the MHRA.

Germany. Thirty live benefit assessments, no psychedelic among them. Structurally closed until a product holds authorisation. The treatment-method route remains theoretically open and unused.

Netherlands. No psilocybin or MDMA entry at Zorginstituut Nederland, control-verified.

EU. No psychedelic item published by the EMA in the window, and as of its last quarterly filing Compass had not submitted an application there either. The division of labour is unchanged and remains the underlying problem. Central approval is union-wide from the day it lands. Pricing and reimbursement stay national.

Scoring the open calls. Six are live, none resolved. Issue 3, that AtaiBeckley reports a first BPL-003 Phase 3 patient dosed in the third quarter, resolves 30 September and is fifty-eight days out with no Phase 3 in the registry, verified three ways this week. We are planning for a miss. Issue 4, that Panorama posts a smaller placebo-adjusted effect than Voyage, is the nearest scoreable call. Voyage's primary completion passed in July and Panorama's falls this month, so both toplines are imminent. Issue 5, that BPL-003's Phase 3 registers as adjunctive without SSRI washout, waits on the same registry posting as Issue 3. Issue 6, that COMP360 is not commercially prescribable in the United States by 30 June 2027, is open, and we are recording lower confidence than when we made it. COMP360 holds a Commissioner's National Priority Voucher, which the company describes as something that may streamline review once the rolling submission completes, so the call now rests more heavily on DEA rescheduling being slow. Issue 7, on Lilly's European footprint, is open and unfalsified with the deal still unclosed. Issue 8, that no psilocybin or MDMA therapy enters a formal assessment in Germany, the Netherlands or the United Kingdom before the end of 2027, is open and now non-provisional across all three countries.

Worth naming a structural weakness in our own board. Three of six open calls resolve on the same event, a single registry posting for BPL-003's Phase 3. Half the ledger moves together or not at all, and if that trial never posts, Issue 5 becomes permanently unscoreable. We are diversifying deliberately from here.

This week's call. Compass Pathways will not publicly disclose a marketing authorisation application to the MHRA or the EMA for COMP360 before 30 June 2027, the end of its own guided US launch window. This is our read and not Compass's stated plan, since the company has said nothing about future European intent in either direction. The reasoning is the whole lead. A US-first launch, no application on file as of the last quarterly filing, no analyst pressure to change that, and a commercial organisation being built in one market. Falsifiable and checkable: if Compass discloses a UK or EU application, or a regulator's register shows one, before 30 June 2027, we are wrong.

Resolve: 30 June 2027, or earlier if disclosed.

If something here is wrong, please tell me, and I will correct it.

Thank you so much for reading.

Dhruv Shekhawat

Founder, Psychedelic Brief