EMA to Hold Advisory Meeting on Clinical Trials for Innovative Therapies
As part of its June session, the regulator will discuss modern approaches to study design for new drugs, including advanced cell and gene therapies, setting the direction for the entire pharmaceutical industry.
EMA Advisory Meeting on Clinical Trials: The Quiet Revolution That Will Bury Traditional Phases II-III
Insider insight you won't find in official press releases: The EMA advisory meeting in June 2026 is not just a "discussion of design approaches." It is a public execution of the classic three-phase drug development model. For the past 25 years, the pharma industry has lived by the mantra "Phase I – safety, Phase II – efficacy, Phase III – confirmation." This model has broken down for gene and cell therapies (ATMPs). Because when you inject a patient with their own edited stem cells, there is no "Phase I in healthy volunteers" – it's unethical. There is no "double-blind placebo control" because you can't fake a surgical procedure. There is no "long-term follow-up" as a separate phase – it must be embedded in the very fabric of the clinical trial. The EMA meeting in June 2026 is the moment when the regulator officially admits: the old rules don't work. And what replaces them will change the business models of everyone, from Pfizer to garage startups.
[The Gist]: What's Really Happening
In reality, the EMA is not going to "discuss" but to dictate. Three fundamental changes are at stake. First: a revision of requirements for confirmatory trials for accelerated approvals. Currently, the scheme looks like this: get PRIME designation, fast-track assessment, conditional marketing authorization – then, 2-3 years later, conduct a confirmatory study. But in 2025-2026, it turned out that 30% of accelerated ATMP approvals did not confirm clinical benefit in post-marketing studies. Why? Because doctors don't want to randomize patients who are already receiving a working therapy into a placebo group. The EMA realized: you can't require traditional confirmatory trials where they are infeasible. Instead, "real-world evidence" (RWE) and "historical controls" (external control arms) will be introduced. This is not a technical detail. It means that a company that brought a CAR-T to market may not spend $100-200 million on a Phase III, but simply collect data from patient registries. Those who know how to work with real-world data will win.
The second change concerns CMC (chemistry, manufacturing, controls) – production requirements. In the first half of 2026, the FDA issued two Complete Response Letters for gene therapies Kresladi (LAD-I) and another candidate precisely due to CMC issues. The EMA is going further: they want to allow "analytical comparability" instead of clinical bridging studies when changing production lines. What does this mean in plain English? Previously, if you produced CAR-T manually in a university clinic clean room ($100,000 per dose) in Phase I/II and then switched to an automated line ($50,000 per dose) for commercial launch, the EMA required an additional clinical study with 20-30 patients to prove that "the new dose is the same as the old one." This cost $5-10 million and delayed market entry by 12-18 months. The new approach: you show at the biochemical level that the cells have the same phenotype, the same CAR expression, the same cytokine profile – and the regulator believes you. This will save the industry billions.
The third, most radical point is the so-called "platform clinical trials" and "master protocols." In the classic model, each new gene therapy is tested in a separate study with a separate placebo group. This is madness when you have 10 different CAR-T constructs for CD19-positive lymphomas. The platform approach: one protocol, one control group (shared), multiple experimental arms that can be added and removed during the study. The EMA wants to standardize such protocols so that companies don't have to renegotiate the design with the regulator each time. This will increase patient enrollment speed by 3-4 times.
[Timeline and Context]
Let me break down how we got here. December 2024: The EMA launches a pilot project on platform trials for oncology ATMPs. March 2025: The FDA releases a draft guidance on the use of external control arms in gene therapy. May 2025: The first results of this pilot come out – it turns out that platform protocols reduce the time from IND to BLA (biologics license) by an average of 14 months. September 2025: The EMA publishes the deadline for submitting applications for PRIME eligibility in 2026. But the key moment is January 2025: the EU HTA Regulation comes into force, requiring EU countries to jointly assess the clinical benefit of ATMPs, instead of each of the 27 countries doing it separately. This was the trigger. Because if countries can no longer independently negotiate prices based on "their own" efficacy assessment, the burden of proof falls entirely on the single European regulatory window – i.e., the EMA.
February 2026: The CHMP issues a positive opinion for Breyanzi (lisocabtagene maraleucel) for DLBCL, PMBCL, and FL3B. This was the first major CAR-T to go through the new PRIME accelerated dialogue system from the very beginning of development. The entire process – from the first PRIME request to positive opinion – took 26 months. For comparison: Yescarta (first-generation CAR-T) took 41 months. A 1.6-fold acceleration. April 2026: An editorial is published in the ISCT (International Society for Cell & Gene Therapy) journal, where the authors directly state: "Q1 2026 was a turning point for CGT, the industry is moving from a 'promising frontier' to commercial scalability." And now, June 2026, the advisory meeting where the EMA summarizes the results of two years of pilots and announces new permanent rules.
Why is this happening now, not a year earlier or later? Because in November 2025, the European Parliament adopted amendments to Regulation 1394/2007 (the main ATMP regulation), expanding the powers of the Committee for Advanced Therapies (CAT) – now the CAT can give binding, not just advisory, opinions on CMC issues. The EMA no longer needs to wait for member states to agree on a position. The CAT simply says "this meets the standards," and that decision applies throughout the EU. A colossal acceleration.
[Who Wins and Who Loses]
Winner #1: Small biotech companies without their own manufacturing. Previously, such companies (e.g., CRISPR Therapeutics or Intellia Therapeutics) had to either build a plant for $200-300 million or sign onerous contracts with CDMOs (contract manufacturers) like Lonza or Catalent. Now, with the new CMC rules, they can produce clinical batches in university GMP labs (rental for $50,000 per month) and for commercial launch, sign a licensing agreement with a large pharma where "comparability" will be proven analytically, not clinically. This lowers the entry barrier for ATMPs from $150-200 million to $30-40 million. Venture funds (Flagship Pioneering, Arch Venture) are already revising their models.
Winner #2: Companies specializing in real-world evidence (RWE). Flatiron Health (acquired by Roche for $1.9 billion), Palantir (via their Foundry for healthcare), and even startups like Aetion – their services will become mandatory. Because when the EMA allows RWE instead of confirmatory trials, you will need infrastructure to collect and analyze data from thousands of medical records. The cost of one RWE study for an ATMP is $2-5 million. Margins are 40-50%. This is a new gold mine.
Loser: Large CROs (contract research organizations) – IQVIA, ICON, PPD. Their business is built on managing traditional Phase II-III trials with thousands of patients. In the new paradigm, where confirmatory trials are replaced by RWE and platform trials reduce the number of control groups, demand for their services will drop by 30-40% in the ATMP segment. IQVIA has already announced the layoff of 700 employees in Europe in May 2026 – officially "optimization," unofficially adaptation to the new rules.
Silent loser: Insurers and healthcare systems. When the EMA allows accelerated approval based on surrogate markers (e.g., percentage of CAR-T cells in blood at 28 days instead of 5-year survival) and then accepts RWE instead of confirmatory trials, the burden of proof "does it actually work?" shifts to payers – insurance companies. They must decide whether to pay €1.5 million for a therapy that has no Phase III. The French HAS and German G-BA are already raising alarms. They will demand that the EMA tighten RWE quality requirements. The result: RWE studies will become as expensive and lengthy as Phase III, just under a different name.
[What the Media Isn't Saying]
The first dirty secret: "platform trials" are a trap for small companies. Yes, it sounds nice: one protocol, a shared control group. But who controls that protocol? The large player who initiated it. For example, Novartis launches a platform trial for its CAR-Ts. A small company wants to add its arm. Novartis sets a price for access to the platform – $10 million per year plus sales royalties. And you can't refuse, because if you try to run your own separate trial, you'll enroll patients 5 times slower – they'll all go to the Novartis platform. The EMA knows about this problem. At the June 2026 meeting, a mechanism for "open platforms" with regulated tariffs will be proposed. But lobbyists from Novartis, Roche, and Pfizer are already working to make these tariffs "confidential." The result: monopolization of ATMP trials by 3-4 giants by 2028.
The second omission concerns "historical controls" (external control arms). The idea: take data from patients treated 5-10 years ago and use them as a control group. But those patients were treated under old standards, didn't have modern diagnostic methods, and didn't sign informed consent for their data to be used in this way. In its draft guidance (April 2026), the EMA proposed that "anonymization" of data is sufficient to avoid requiring re-consent. The European Data Protection Board (EDPB) has already filed a formal objection. Outcome: within 90 days, the EMA will abandon this idea, and "historical controls" will become unavailable for 90% of ATMPs. Companies that have already incorporated them into their protocols (e.g., Bluebird Bio) will have to rewrite their study designs.
The third layer, which even industry analysts are silent about: the CAT (Committee for Advanced Therapies) is physically unable to handle the volume. In 2025, the CAT received 147 applications for ATMP classification and 89 PRIME requests. The CAT has 25 experts who do this in addition to their main jobs. The average review time for a PRIME application has increased from 60 to 120 days. The EMA will propose increasing the CAT staff to 40 and making them full-time. But the EMA budget is fixed. The money will come from drug registration fees – meaning fees will increase by 15-20% for everyone, including traditional small molecules. This will cause a wave of discontent from generics associations.
[Forecast: Next 30 Days and 90 Days]
Next 30 days:
Expect the publication of an official EMA "Reflection Paper" following the advisory meeting. The document will contain three specific changes effective from September 1, 2026: (1) recognition of RWE from European registries (e.g., EBMT for transplantology) as sufficient for efficacy confirmation if the sample size is >200 patients; (2) a simplified CMC comparability procedure without clinical bridging studies; (3) introduction of an "adaptation period" for platform trials – 6 months during which companies can transition existing trials to the new design.
Also, the CHMP will issue a decision on two new ATMPs: most likely, a gene therapy for hemophilia A (BioMarin) and a CAR-T for multiple myeloma (Janssen). Both will receive positive opinions. But note: in both cases, the EMA will require post-authorization safety studies (PASS) using RWE, not classic confirmatory trials. This will be the first precedent that everyone will copy.
Next 90 days:
September 2026 – a CAT meeting where the new PRIME application format will be officially approved. Key change: no longer required to provide "preliminary efficacy data" (clinical). "Convincing preclinical data on human models" – e.g., organoids – will suffice. This is a direct result of the April work on COPD. Now, to get PRIME, organoid data is enough. This will reduce the cost of preparing an application from $500,000 to $50,000. Expect a flood of applications from academic groups – 30-40 applications in August-September instead of the usual 10-15.
And the final, toughest forecast: the EMA will announce the launch of a "practical guide" for so-called "low-budget ATMPs" – therapies produced in university clinics for orphan diseases (fewer than 100 patients per year). The guide effectively legalizes the "Hospital Exemption" at the European level, removing national barriers. This means that a small clinic in Leiden can produce CAR-T for its patients and "export" it to a clinic in Lille without separate approval in each country. Patients will win. But commercial ATMP manufacturers (Novartis, Gilead) will lose their monopoly on orphan indications. The battle will be fierce. Lobbyists are already gathering signatures. The outcome of this round will determine whether the European ATMP industry becomes competitive or remains a toy for the rich. Keep your finger on the pulse.
— Editorial Team
