US oncology and advanced-therapy sponsors have long understood that approval by the Food and Drug Administration does not buy market access in Europe, and that a European marketing authorization is only a permit to sell, not a decision to pay. What is new, and what most US development teams still underweight, is that a distinct clinical layer now sits between the European Medicines Agency (EMA) and the national bodies that fund medicines: the Joint Clinical Assessment. The EU Health Technology Assessment Regulation makes that assessment mandatory wherever a centralized marketing-authorization application was submitted after 12 January 2025 for a medicinal product containing a new active substance whose therapeutic indication is the treatment of cancer, or for a medicinal product regulated as an advanced therapy medicinal product (ATMP), and the first year of operation has produced enough of a record to say what the assessment demands and where it bites.1Regulation (EU) 2021/2282 on health technology assessment [2021] OJ L458/1; joint clinical assessment as the first of four cooperation streams; date of application 12 January 2025. The lesson of that year is uncomfortable for a team working from an FDA-anchored playbook: the outcome of a Joint Clinical Assessment is largely fixed by trial-design decisions taken years before the dossier is filed, and a package built to satisfy one regulator against one comparator will not answer the questions the assessment poses on behalf of many.
1. The Assessment US Sponsors Cannot Read Off the FDA Map
There is no federal health technology assessment body in the United States, and that absence shapes how US sponsors reason about evidence. The FDA decides safety and efficacy; coverage and payment are settled downstream and in fragments, by the Centers for Medicare & Medicaid Services for federal programs, by commercial payers for everyone else, and, in an advisory and non-governmental capacity, by the Institute for Clinical and Economic Review. None of these actors sits between approval and launch as a single gate, and none requires the sponsor to prove relative effectiveness against a defined standard of care before the product can be sold.2The United States has no centralized federal HTA gate: FDA approval under the FDCA is distinct from coverage, which is fragmented across CMS, commercial payers, and the non-governmental ICER. A US sponsor therefore tends to build one pivotal evidence package, optimized for the FDA, frequently against a placebo or a single active control, and to treat Europe as a later, separable set of national conversations.
The EU Health Technology Assessment Regulation collapses part of that separability. It creates four permanent streams of Member-State cooperation: the Joint Clinical Assessment of a technology's relative clinical effectiveness and safety, the Joint Scientific Consultation through which developers align evidence plans with assessors before pivotal trials, the identification of emerging health technologies, and voluntary cooperation on the questions the mandatory streams do not reach.3HTA Regulation (n 1), Chapter II, Sections 1 to 4 (the four streams of joint work), Art. 7(1) and (2) (scope and staggered phase-in), Art. 7(4) (Commission selection of certain high-risk medical devices and IVDs on a Coordination Group recommendation), with device classes under Art. 51 MDR and Art. 47 IVDR and the expert-panel gates in Art. 54 MDR and Art. 48(6) IVDR. The Joint Clinical Assessment is the load-bearing one, and its scope widens on a fixed schedule that a US planner should read as a calendar rather than a footnote: from oncology medicines with a new active substance and ATMPs on 12 January 2025, to orphan medicinal products from 13 January 2028, and to the remaining medicinal products within the centralized route from 13 January 2030. Certain high-risk medical devices and class D in vitro diagnostics are drawn in selectively rather than swept in automatically, and the choice is not the Coordination Group's to make: the Commission adopts it by implementing act, at least every two years and after seeking a recommendation from the Coordination Group. The eligible pool is narrower than the device classes suggest, because a class IIb or class III device qualifies only where an expert panel has given a scientific opinion in the clinical evaluation consultation procedure under Art. 54 MDR, and a class D diagnostic only where an expert panel has given its views under Art. 48(6) IVDR.
The consequence is that the clinical evidence question a US sponsor could once defer becomes a Union-level exercise timed to the EMA review itself. The assessment does not judge cost-effectiveness, and it does not set a price; it produces a scientific report on how the technology performs against the relevant comparators, and it leaves the value judgment to the Member States. That division of labor is the feature US teams most often misread, because it means the Joint Clinical Assessment can be demanding and consequential without ever producing the reimbursement answer the sponsor actually needs.
2. How the Joint Clinical Assessment Actually Runs
The mechanics are set out in the Commission Implementing Regulation of 23 May 2024, which lays down the detailed procedural rules for joint clinical assessments of medicinal products at Union level and the way they interlock with the EMA timetable.4Commission Implementing Regulation (EU) 2024/1381 of 23 May 2024 laying down, pursuant to Regulation (EU) 2021/2282 on health technology assessment, procedural rules for … joint clinical assessments of medicinal products for human use at Union level, as well as templates for those joint clinical assessments, OJ L, 2024/1381, 24.5.2024. The assessment runs in parallel with the centralized marketing-authorization review, not after it. When a developer submits to the EMA, it also enters the joint clinical track: the Member State Coordination Group, working through a dedicated subgroup and a Commission-run secretariat, appoints an assessor and a co-assessor drawn from national HTA authorities in different Member States, and the scoping phase begins.5HTA Regulation (n 1), Art. 3 (Member State Coordination Group), Art. 8–12 (joint clinical assessment procedure, report, finalization and publication), and Art. 16–21 (joint scientific consultation). Scoping is where the assessment defines the questions the dossier must answer, and it is national in origin: each participating Member State can specify the population, intervention, comparator, and outcomes, the PICO, that reflects its own standard of care.
The developer then compiles a single Joint Clinical Assessment dossier that must address the consolidated set of PICOs, and it must do so inside a window measured against the regulatory clock rather than the sponsor's convenience. The Implementing Regulation fixes that window at 100 days from notification of the Commission's first request, the request that carries the finalized assessment scope, and at 60 days where the marketing-authorization application runs under the accelerated procedure or where the assessment concerns a variation corresponding to a new therapeutic indication; the secretariat may extend it only in justified cases, and never past the outer limit of 45 days before the envisaged opinion of the Committee for Medicinal Products for Human Use. Either period is short for analyses that may require indirect comparisons or subgroup work that were never designed into the trial.6On PICO consolidation, the Coordination Group's Guidance on the scoping process V1.0 (13 November 2024); on the dossier deadlines, Art. 12(2) and (3) of the JCA Implementing Regulation (n 4); industry commentary in EFPIA (4 November 2025) and ICON (13 March 2026). The assessor and co-assessor then produce a draft assessment report, which is refined through comment and finalized by the assessment subgroup at the latest on the date the Commission adopts the decision granting the marketing authorization, with endorsement by the Coordination Group falling no later than 30 days after that decision, after which the Commission publishes the endorsed report on the publicly accessible webpage of the health technology assessment IT platform.
What the Member States must then do with the report is the provision US counsel should read most carefully. Under Art. 13 HTA Regulation, Member States must give the published joint report "due consideration" in the health technology assessments they carry out nationally, must annex it and the developer's dossier to the national file, and must refrain from asking the developer at national level for information already submitted at Union level. What Art. 13 HTA Regulation does not say is that a Member State may not weigh the clinical evidence again: it expressly preserves each Member State's competence to draw its own conclusions on the overall clinical added value of the technology in its own healthcare context, and the reservation of exclusive national competence over pricing and reimbursement sits in Art. 1(2) HTA Regulation rather than in Art. 13 HTA Regulation.7HTA Regulation (n 1), Art. 13(1) (Member States give the published joint clinical assessment report due consideration and do not re-request evidence already submitted at Union level); Art. 1(2) (exclusive national pricing and reimbursement competence); Art. 9(1) (no value judgment in the report). The report is therefore an input each Member State must take into account and a conclusion none of them is bound by: it removes the duplicated request for the same clinical evidence across national bodies, but it does not remove the twenty-seven national decisions that follow, and it does not stop a national authority from reaching a different conclusion on value once the shared clinical facts are in front of it.
3. The PICO Problem, One Year In
The single hardest feature of the first year was the one the pre-implementation simulations had flagged: the PICO. Because the comparator in a PICO is the treatment a given health system actually uses, and because standards of care in oncology differ across Europe, one product can attract a large and heterogeneous set of PICOs, each demanding a comparison the pivotal trial may not have made. Assessors consolidate overlapping PICOs, but the industry account of that consolidation is that it retains considerable flexibility, that the rationale behind the initial proposals and the consolidation decisions following them is not visible enough for a developer to anticipate the scope, and that a consolidated PICO can still set an evidence standard the dossier cannot satisfy.
The Joint Clinical Assessment measures the evidence a sponsor chose to generate against the evidence twenty-seven health systems wish it had generated. The PICO exposes the gap between the two, and that gap is fixed at trial design, not at submission.
For a US-anchored program the gap is structural rather than incidental. A trial designed to secure FDA approval against placebo, or against the single comparator most relevant to the US market, will not carry head-to-head data against the several active therapies that European systems treat as standard, and no volume of dossier drafting inside the statutory dossier window can retrofit a comparison that the protocol never contemplated. The available answers, indirect treatment comparisons built from the published literature, are the analyses assessors scrutinize hardest, and they carry recurring evidentiary weaknesses of their own: risk of bias in single-blinded trials, questions over the transferability of non-European trial populations, and the handling of missing data. The Joint Scientific Consultation exists to close exactly this gap by aligning the evidence plan with assessors before the pivotal trial locks, but it is rationed by design: it is open only while the clinical studies are still in the planning stage, it is offered in published request periods, and where eligible requests exceed the planned number the Coordination Group selects among them against the criteria in Art. 17(3) HTA Regulation. Its outcome document gives rise to no legal effect on the Member States, the Coordination Group, or the developer, and does not prejudice the joint clinical assessment that follows. A sponsor that reaches it after the protocol is fixed has arrived too late; a sponsor that reaches it in time has an opinion, not an assurance.
4. The First Year's Record and the Runway Ahead
The volume of the first year was modest and the substance was not. The Member State Coordination Group reported that it started thirteen joint clinical assessments on new oncology products and ATMPs in 2025, selected seven joint scientific consultations across two request periods and completed four of them, and recorded thirty-eight patients, caregivers, or clinicians involved in the joint clinical assessments and ten in the joint scientific consultations by year end.8Member State Coordination Group on Health Technology Assessment, Annual Report 2025 (adopted pursuant to Art. 6(4) of the HTA Regulation (n 1); published 16 February 2026): 13 joint clinical assessments started; 7 joint scientific consultations selected across two request periods, 4 completed. The number of assessments sat below the Coordination Group's own pre-launch estimate, which had put 2025 at seventeen assessments for medicinal products with new active substances indicated for the treatment of cancer and eight for advanced therapy medicinal products, cancer-indicated advanced therapies counting inside the seventeen. The Coordination Group published no cause for the gap, and the final figure in any year turns on how many valid marketing-authorization applications the European Medicines Agency receives. Even at that volume, the operational strain of the PICO and dossier process was the dominant theme in the practitioner commentary, and the transparency of the reasoning behind published scopes was a recurrent complaint.
The strain will grow as volume and scope increase. The 2026 work program adopted by the Coordination Group on 28 November 2025 projects a substantial increase: around thirty-five assessments of medicinal products with new oncology active substances, around fifteen of advanced therapy medicinal products, three on new-indication variations, and approximately five on a first selection of high-risk medical devices and in vitro diagnostics, the first of those expected to start in June 2026; the orphan expansion then lands in January 2028 and the all-products expansion in January 2030.9Member State Coordination Group on Health Technology Assessment, Annual Work Program 2026 (adopted 28 November 2025): around 35 oncology and 15 ATMP assessments, 3 on variations, and approximately 5 on a first selection of medical devices and IVDs. For a US sponsor the practical reading of that calendar is that the assessment stops being an oncology-and-ATMP curiosity and becomes the default European gate for every new active substance within the planning horizon of a molecule now entering the clinic.
5. Strategic Considerations for US Oncology and ATMP Sponsors
For a US sponsor, the question the Joint Clinical Assessment raises is how early the evidence the dossier will need has to be designed. If the comparators that satisfy the European PICOs are chosen in the pivotal-trial protocol, and if the window to compile the dossier is too short to generate anything the trial did not, then the decisive move is the one made at protocol design, when a US team is usually reasoning about the FDA alone. So the first question is whether the trials being planned today for a molecule that will reach the EMA in 2028 or later are being built with European standard-of-care comparators and the Joint Scientific Consultation in view, or whether the European evidence problem is still being treated as a post-approval task that a market-access function will pick up after the fact.
A second question runs through the deal base rather than the development plan. When a US biotech out-licenses an oncology or cell-and-gene asset to a partner with European reach, or in-licenses one, who bears the cost and the responsibility of the Joint Clinical Assessment dossier, who controls the comparator strategy and the timing of any scientific consultation, and how are the reps and warranties about the adequacy of the clinical evidence framed when the standard against which adequacy is measured is set by national bodies rather than by the FDA? Milestones and valuations keyed to European launch increasingly depend on an assessment whose difficulty the parties may not have priced, and diligence on an inbound asset now has to ask not only whether the pivotal data support approval but whether they can survive a PICO-driven comparison the originator never ran.
A third question is one of expectation management about what the assessment actually delivers. Because the joint report harmonizes the clinical dossier but not the reimbursement decision, a favorable Joint Clinical Assessment does not guarantee a favorable national outcome, and divergent decisions across Member States on the same shared evidence are not a malfunction of the system but a designed-in consequence of Art. 13 HTA Regulation. How should a launch sequence, a pricing corridor, and an investor narrative be built so that they survive the possibility that the same clinical report yields access in one Member State and refusal in another, and that the two happen at different times?
A fourth question concerns the European markets that the joint process does not reach. Europe is not the European Union for these purposes. Switzerland is outside the EU and the EEA and outside the Joint Clinical Assessment entirely; Swiss marketing authorization runs through Swissmedic under the HMG, and Swiss pricing and reimbursement through the Spezialitätenliste, the list of medicines reimbursed by compulsory health insurance, administered by the BAG under the KVV, on a timetable and an evidentiary logic of their own.10Switzerland is outside the EU HTA framework: Swissmedic authorization under the Heilmittelgesetz (HMG, SR 812.21) and the Spezialitätenliste, drawn up by the BAG under Art. 52(1)(b) KVG (SR 832.10) on the admission conditions in Art. 65 ff. KVV (SR 832.102), run on a separate track. The United Kingdom runs separate appraisals, principally through the National Institute for Health and Care Excellence (NICE), with Scotland advised separately by the Scottish Medicines Consortium. A US sponsor that treats "Europe" as a single assessment market will build one comparator story and find that it answers the EU joint process, misses the Swiss and UK appraisals, and still faces twenty-seven national reimbursement decisions behind the joint report. Whether a global evidence plan optimized for the EMA also serves the Swiss and UK routes, and whether the comparators chosen for the European PICOs are the ones those separate systems will demand, are questions that cannot be read off the FDA map and that grow harder to answer once the pivotal trial has locked.