The EU Joint Clinical Assessment process produced its first completed output in June 2026, with the assessment of Ojemda (tovorafenib) for relapsed or refractory paediatric low-grade glioma. The result was not a verdict on the therapy itself. It was a detailed signal about what the new EU HTA framework expects from manufacturers in terms of comparative evidence quality, PICO coverage, and indirect comparison methodology. Of the eight PICOs set by Member States, only one was meaningfully addressed. This analysis examines what went wrong with the Ojemda dossier and what the lessons mean for any manufacturer preparing a European launch under EU HTAR.
The publication of the first Joint Clinical Assessment (JCA) under the EU Health Technology Assessment Regulation (HTAR) marks a significant milestone for manufacturers preparing for launch in Europe. While the inaugural assessment focused on Ojemda (tovorafenib), a treatment for relapsed or refractory paediatric low-grade glioma, the most important lesson extends far beyond a single product.
The assessment provides an early indication of how the new EU HTA framework may approach one of the most challenging issues in market access: demonstrating comparative clinical value when robust comparative evidence does not exist.
The challenge: a promising therapy with limited comparative evidence
Why did the EU Joint Clinical Assessment flag FIREFLY-1 as insufficient for comparative conclusions?
Ojemda’s clinical evidence package was primarily based on FIREFLY-1, a single-arm study evaluating efficacy and safety in paediatric patients with BRAF-altered low-grade glioma. Such evidence generation approaches are not unusual in rare diseases and paediatric oncology, where conducting large randomised controlled trials may be impractical or ethically challenging.
The first JCA demonstrates that while the assessment framework recognises these challenges, it does not eliminate the need to assess comparative effectiveness. The JCA explicitly noted that:
As FIREFLY-1 is a non-comparative study, no direct comparative data are available to inform any of the PICOs and no data are available to inform an analysis within a connected network of evidence.
As a result, assessors were unable to draw conclusions based on direct comparative evidence against the relevant treatment alternatives identified through the PICO process. Comparative evidence was available only for a subset of patients, through an unanchored Matching-Adjusted Indirect Comparison (MAIC). The assessment highlighted substantial uncertainty associated with this approach, reflecting concerns around residual confounding, population differences, and the absence of randomisation.
There is an important distinction here that manufacturers should not miss. The framework may acknowledge why comparative trials are difficult to run in rare and paediatric settings, but it does not lower the bar for the quality and documentation of whatever evidence is submitted in their place. Recognising the circumstance is not the same as excusing weak or poorly documented evidence. As the assessors put it directly, disease rarity and evidence paucity are not valid excuses for an incomplete or underpowered comparative case.
How the assessors scrutinised the dossier
What did assessors find when they examined the Ojemda evidence package across all eight PICOs?
The detail of the assessment is where the real lessons sit. The scope set eight PICOs (population, intervention, comparator, outcome). Of those eight, only one was meaningfully addressed.
| PICO(s) | Evidence submitted | Outcome |
| PICO 5 | Unanchored MAIC (FIREFLY-1 subpopulation vs Bouffet et al. 2023) | Included in the JCA report |
| PICO 7 | Unanchored MAIC | Excluded: insufficient information to assess the comparator study (TRAM-01) |
| PICOs 1 to 4, 6 and 8 | No comparative evidence submitted | Excluded: no comparator data available |
The criticisms ran across the whole evidence base. Because FIREFLY-1 was non-comparative, it produced no direct comparative data to inform any PICO and no data to inform an analysis within a connected network of evidence. The single-arm descriptive outcomes from the pivotal study were not informative of relative effects against the defined comparators, so they did not feature in the assessment’s conclusions. TRAM-01, the comparator study behind the PICO 7 MAIC, was reported through conference abstracts only, which did not provide enough documentation for assessors to evaluate study methods and results, including the patient characteristics needed to confirm a relevant subpopulation and adjust for confounders.
On the construction of the evidence base, the developer included studies with an individualised treatment comparator only where the comparator covered all treatment options, an approach the assessors deemed inadequate. Inconsistent definitions of objective response rate (ORR) across studies were expected to bias the comparison in favour of tovorafenib, and a comparison using an alternative response definition was requested by assessors but not provided.
The credibility of the MAIC itself drew the sharpest scrutiny. As non-randomised evidence, an unanchored MAIC carries an inherently greater risk of bias and several major uncertainties, and its effect estimates cannot necessarily be read as causal. Population differences between the studies could not be adjusted within the method, and unidentified or unadjusted prognostic and effect-modifying variables left residual confounding, creating a risk of bias of unknown magnitude and direction across all effect estimates. The small FIREFLY-1 sample limited the number of variables that could be adjusted for, and a small effective sample size (ESS) introduced statistical imprecision. That reduced ESS also pointed to poor covariate overlap between the studies, which compounds the risk to both bias and coverage.
Understand how your evidence package is likely to perform across EU HTAR PICOs before you submit.
GPI’s Horizon framework evaluates pre-launch assets against the evidence domains most scrutinised in EU Joint Clinical Assessments: trial design quality, indirect comparison methodology, endpoint robustness, and PICO coverage. It gives market access and HEOR teams a structured view of where their dossier is likely to be challenged, before those challenges become visible to all Member States at once.
A single European assessment does not mean a single European comparator
Does the EU HTAR create a single standard of care for evidence generation?
Perhaps the most important strategic lesson relates to comparator selection.
One of the ambitions of the EU HTAR is to reduce duplication of clinical assessments across Member States. The Ojemda process illustrates that divergence in clinical practice and treatment pathways continues to shape evidence requirements. The assessment incorporated eight separate PICOs proposed by Member States, reflecting differing views on relevant comparators and clinical management across Europe.
This exposes a common misconception about the JCA process. Manufacturers now face a single clinical assessment, but they do not face a single European standard of care. Evidence generation strategies will therefore need to accommodate multiple national perspectives at once. Comparator choices that satisfy one Member State may not address the expectations of another. PICO development is likely to become one of the most critical activities in launch preparation under the EU HTAR framework.
What the assessment process tells us
The Ojemda assessment also offers a clear view of how the JCA process works in practice. The review was led by the National Centre for Pharmacoeconomics (NCPE) in Ireland and IQWiG in Germany, following the selection of PICOs by Member States. The manufacturer submitted evidence against the agreed PICOs, which was then evaluated through the JCA process before adoption by the HTA Coordination Group.
The JCA does not make reimbursement recommendations. It establishes a common European assessment of relative clinical effectiveness and safety, which Member States can then use to inform their own national pricing and reimbursement decisions. In theory, this reduces duplication and improves efficiency. In practice, manufacturers will still need to manage evidence uncertainty carefully, because national decision-makers will continue to interpret the common clinical evidence base through their own reimbursement frameworks.
Several practical patterns from the assessment are worth carrying into launch planning:
One or more PICOs may go unaddressed at the JCA stage. The gap will be flagged in the JCA report, but the consequences tend to surface later, during national-level assessments. Here, most of the PICOs and outcomes requested in the assessment scope were not included. Where a developer chooses to omit a PICO, the justification will be scrutinised, so it needs to be clear and defensible.
Relative effectiveness on health-related quality of life (HRQoL) and patient-reported outcomes (PROs) is actively sought. HRQoL and PROs, including fatigue, were part of the required outcomes here. HRQoL data were submitted as exploratory analyses only, and the results did not make the JCA report because relative effectiveness results were not available.
Differences between the claimed and final indication are checked, and may or may not trigger a new assessment scope. In this case, the interpretation was unaffected and the JCA subgroup agreed no new scope was needed, with a clinical expert noting that the two populations were largely overlapping but potentially not strictly identical.
For PICOs with an individualised treatment comparator, a developer can submit comparator data covering only a selection of the requested treatment options. The suitability of that data is then assessed, and it may end up describing effects for only a subpopulation of the original PICO population.
Unanchored MAICs are accepted, but heavily scrutinised. Rarity does not excuse poor evidence such as conference abstracts, and sufficient documentation is required to allow assessment of study methods and results. Unmet methodological requirements that can introduce bias will be flagged, and the evidentiary standard will be maintained. Transparent reporting is expected throughout, and here transparency was lacking around the methods used to identify the final list of potential prognostic and effect-modifying variables and the reasons behind study exclusions.
Early advice matters. A Joint Scientific Consultation (JSC) can help flag core evidence gaps early, giving a developer the chance to address them before submitting the JCA dossier. No JSC was held in this case.
The downstream implications run in one direction. Uncertainty is scrutinised at EU level first, then cascades into national HTA and pricing negotiations. JCA criticism is explicit and visible to all Member States at the same time, so where the EU evidence bar falls is now a shared reference point rather than a series of private national conversations. Where that bar cuts matters a great deal.
Looking ahead: evidence strategy becomes more important than ever
How should manufacturers adapt their evidence generation strategy for the EU Joint Clinical Assessment process?
The first JCA suggests that the future challenge for manufacturers is not simply demonstrating efficacy, but demonstrating comparative value across multiple European healthcare systems at once.
For products targeting rare diseases, oncology, and advanced therapies, this will require evidence strategies that are developed earlier and with far greater consideration of comparative evidence requirements. External control arms, real-world evidence, long-term follow-up, and robust indirect comparison methodologies are likely to become increasingly important components of launch planning.
The most important lesson from the first JCA is not about Ojemda itself. It is that comparative evidence remains at the heart of European value assessment. The EU HTAR may create a single clinical assessment process, but it has not removed the need for manufacturers to justify how their product performs relative to existing treatment options. If anything, the Ojemda assessment confirms that comparator strategy and evidence planning will be even more decisive determinants of market access success in the years ahead.
References
- European Commission. Joint Clinical Assessment Report: Tovorafenib (Ojemda), June 2026
- European Commission. Joint Clinical Assessments under Regulation (EU) 2021/2282
- IQWiG. First Joint Clinical Assessment under the European HTA Regulation, June 2026
- NCPE. Participation in the first Joint Clinical Assessment of Ojemda (tovorafenib), 2026
FAQs
A: The EU Joint Clinical Assessment (JCA) is a process under the EU Health Technology Assessment Regulation (HTAR, Regulation EU 2021/2282) that establishes a common European evaluation of a medicine’s relative clinical effectiveness and safety. The assessment is led by two Member State health technology assessment bodies (in the Ojemda case, NCPE in Ireland and IQWiG in Germany), and evaluates evidence submitted by the manufacturer against a set of PICOs (population, intervention, comparator, outcome) agreed by participating Member States. The JCA does not make reimbursement recommendations. It provides a shared clinical evidence base that Member States can then use to inform their own national pricing and reimbursement decisions.
A: Of the eight PICOs set by Member States for the Ojemda assessment, only one was meaningfully addressed in the JCA report. The primary evidence base, FIREFLY-1, was a single-arm non-comparative study that produced no direct comparative data to inform any PICO. The MAIC submitted for PICO 7 was excluded because the comparator study (TRAM-01) was reported through conference abstracts only, which did not provide sufficient documentation for assessors to evaluate study methods and results. PICOs 1 to 4, 6, and 8 had no comparative evidence submitted at all. Assessors also identified inconsistent endpoint definitions, inadequate comparator coverage, and significant residual confounding risk in the unanchored MAIC methodology.
A: The first EU Joint Clinical Assessment confirms that the EU HTAR framework acknowledges the evidence challenges in rare disease and paediatric settings but does not reduce the evidential standard for whatever evidence is submitted. Disease rarity and evidence paucity are not accepted as justifications for incomplete or poorly documented comparative cases. Manufacturers targeting rare disease or paediatric indications will need to invest earlier in PICO development, indirect comparison methodology, and evidence documentation quality. A Joint Scientific Consultation (JSC) before dossier submission is strongly advisable to identify and address evidence gaps in advance of the formal JCA process.
Preparing a European launch under EU HTAR?
GPI helps pharmaceutical and biotech teams assess comparative evidence strategy, PICO planning, and market access positioning for EU launches. Our Horizon framework provides a structured, pre-launch view of how your asset is likely to be assessed across EU HTAR criteria and national reimbursement frameworks.
