What the first year of EU JCA reveals about comparative effectiveness and value assessment: An interview with Huseyin Naci, LSE Health

As Joint Clinical Assessments (JCAs) under the EU Health Technology Assessment Regulation become publicly available, attention is turning to what this early experience means for future evidence generation and value assessment. These early assessments raise questions about how developers should plan comparative evidence when evidence at the time of assessment is limited and treatment pathways, comparators and clinical practice may evolve before a medicine reaches patients.

This potential for JCAs to influence evidence planning is explored in a recent publication by Huseyin Naci (Department of Health Policy, London School of Economics and Political Science [LSE]) and Beate Wieseler (Institute for Quality and Efficiency in Health Care [IQWiG]). Drawing on the first assessment, they argue that making comparative evidence gaps explicit at the time of approval could encourage developers to generate relevant data earlier, informing subsequent national judgments about the added therapeutic benefit of a medical product.

In this interview, Radek Wasiak (Adigens Health) speaks with Huseyin about the lessons for developers and assessors from the first set of assessments, how to document value early with a forward-looking mindset, and how to address the limited description of comparative effectiveness strategies in the early assessments.

On January 27, 2027, Miguel Hernán (CAUSALab, Harvard T.H. Chan School of Public Health; Adigens Health) will join Radek and Huseyin for our webinar, After the first JCA reports: How should comparative effectiveness be assessed?, developed in association with Adigens Health. Building on this interview, the panel will discuss how developers can use lessons from the early assessments to plan comparative studies, explain how they will address evidence gaps and support future judgments about a medicine’s value.

REGISTER HERE

Thank you for joining me, Huseyin. Looking across the JCAs published so far, what lessons are emerging for developers and assessors about comparative effectiveness and the evidence needed to support future judgments of value?

Only a few JCA reports have been published so far, so any lessons are preliminary. But one thing is already clear: the assessments stick closely to the methodological guidance published by the Member State Coordination Group on HTA.

Three dimensions matter most:

1. Validity of clinical studies – guidance on study design treats randomized trials as the gold standard and says single-arm trials, on their own, can’t support an assessment of relative effectiveness.
2. Outcomes – guidance on outcomes asks member states to request final patient-centered outcomes, such as mortality, morbidity and quality of life, and to rely on surrogates only where their validity is clearly established.
3. Evidence synthesis – guidance on evidence synthesis prioritizes direct comparisons and, where indirect comparisons are needed, accepts only anchored ones that respect randomization.

That consistency is itself an important lesson. Developers may be used to a degree of flexibility, whether from regulators adapting evidentiary expectations case by case or from HTA bodies reaching different judgements on similar evidence. JCAs, so far, are more consistent: the assessors have applied the published methods with fidelity. Developers can know in advance what will count as evidence.

“That is a real shift. Applied consistently and published in full, these methods mean evidence gaps are documented before pricing and reimbursement decisions are made, rather than going unnoticed at the time of approval.”

You’ve highlighted how consistently assessors have applied the published methods. I’d like to turn to what that means for study design. The population, intervention, comparator and outcomes (PICO) framework specifies which comparisons a JCA should address. When randomized trials don’t cover them all, how should developers decide how to make those comparisons?

The Coordination Group’s scoping guidance is a useful starting point. PICOs translate Member State policy questions into research questions, and they are set before the dossier is submitted. Because the questions are defined by European health systems rather than by the developer, there is a real risk that they won’t align with the pivotal trial.

The fundamental shift for evidence generation is that the most consequential decisions for the JCA are made years earlier, when the pivotal trial is designed. Two decisions are particularly important in this planning process.

The first is the comparator. Where practice varies, Member States can define an “individualized treatment” comparator, which requires one effect estimate against a bundle of options, and a physician’s choice arm can deliver exactly that. A comparator that reflects widely used standard of care also connects the trial to the wider evidence base, which makes anchored indirect comparisons possible later.

The second is data access. Credible non-randomized comparisons require full individual patient data. In the first report, for tovorafenib, differences between study populations couldn’t be adjusted for because those data weren’t available for the comparator study. Securing access to comparator data needs to start early.

You mentioned that some of the most important decisions are made when the pivotal trial is designed. But the comparator that matters then may have changed by the time the medicine reaches patients. How should developers plan for that uncertainty and make clear which assumptions underpin their view of the medicine’s potential value?

This is where early advice is most valuable. Joint Scientific Consultations (JSCs) let developers test their planned comparators and outcomes against what Member States are likely to need, years before the JCA. The scoping guidance tells assessors to consult the outcome of any JSC when drafting the assessment scope.

But the guidance is also realistic. It acknowledges that consultation advice may no longer apply if the therapeutic landscape has changed, and that PICOs will reflect the situation at the time of the survey. So, developers can’t treat early advice as a fixed target, and they need evidence that can adapt.

Prospectively planned network meta-analyses are one way to do that. My collaborators and I proposed the approach in 2013, and we returned to it in our 2020 Lancet series on comparative effectiveness. The idea is simple: if trials in a therapeutic area are designed with similar populations, outcomes and follow-up, they can be synthesized as new comparators emerge, rather than pieced together after the fact.

The Coordination Group’s methods guideline makes similarity across trials a precondition for any valid synthesis, and that is much easier to achieve prospectively. Platform trials offer something similar within a single study, since their control arm can evolve as the standard of care changes. A prospective plan of this kind also makes the assumptions and evidence gaps explicit from the start, which is the most credible way to describe a medicine’s potential value.

I’d like to pick up on your point about planning studies prospectively. If we take a comparison that a randomized trial won’t cover, how could setting out the study protocol early help developers decide what data to collect and how to address potential bias?

Early planning is key. The Coordination Group’s validity guidance points to the target trial framework, which defines the estimand, eligibility criteria, exposure and outcomes of the ideal randomized trial that a non-randomized comparison is trying to approximate. Writing that protocol early, before data collection, can be helpful in turning evidence needs into concrete decisions about what to measure and in whom.

This approach also aligns with what JCA assessors will likely look for. The methods guideline expects covariate selection to be prespecified and based on transparent criteria. Because non-randomized comparisons can be highly uncertain, it also expects a large effect, which can be tested formally using shifted hypotheses.

The same logic applies to randomized trials. In both small cell lung cancer JCAs for tarlatamab and lurbinectedin, quality-of-life and safety data were collected only until the end of treatment, which shortened the observation period compared with survival. Planning with the JCA in mind could have addressed this issue as well.

Looking beyond the assessment itself, when important comparative questions remain unresolved at the time of JCA, how should developers plan to address them? What would national HTA bodies and payers need to understand about those plans when making decisions that affect patient access?

JCAs bring much greater scrutiny to the uncertainties that remain at the time of approval. Although this is not their intended purpose, I expect that scrutiny to extend to whether post-marketing evidence generation is likely to resolve them. At the moment, those plans are mostly handled at the national level, with individual HTA bodies and payers negotiating their own data requirements, often through managed entry agreements. In my view, this is unfortunate, because it fragments evidence generation. Different countries ask for different studies to address what are often the same gaps, and the overall evidence base ends up less coherent than it could be.

The JCA offers a way out of that, because it sets out where the evidence falls short. Developers may increasingly present post-marketing plans that explicitly address the gaps documented in the JCA, specifying which questions each study will answer, with what comparator and design, and on what timeline. Such plans could provide national bodies with a common reference point, which might make it more likely that many of them will be satisfied with the same plan, rather than each requiring something different.

National bodies have a responsibility too. Member states are required to give due consideration to published JCA reports, and I’d think it’s in their own interest for the system to work. Some variation in added-benefit judgements based on the same report is expected, given differences between health systems. But if national bodies impose their own additional evidence requirements, that might ultimately undermine the central aim of the JCA, which is to reduce duplication.

Finally, Huseyin, I’d like to return to a point you and Beate raised in your publication: the relevance of these assessments beyond Europe. How might US payers and other decision-makers use JCA findings to inform their own assessments and future evidence requirements? And what limitations would they need to consider?

Shortly after a new medicine is approved, a JCA report is likely to be the one place where that medicine’s comparative benefits and harms have been examined independently and systematically. That makes the reports interesting and relevant well beyond Europe. The US has no national HTA body, so JCA reports could be useful for US payers by showing what is known and not known about the relative effectiveness and safety of new medicines. In the longer term, the methods behind them might also inform how comparative evidence is weighed in Medicare’s drug price negotiations.

Transparency is also important. Each report is published alongside the developer’s full dossier and the exchange between the developer and the assessors. Decision-makers elsewhere can therefore see not only the conclusions but the evidence and analyses behind them. Over time, the evidence gaps documented in JCAs could also help other payers set clearer expectations for the evidence they will require.

There are important limitations, however. JCA reports describe the evidence but deliberately stop short of judging how much added benefit a medicine offers, and they leave the question of whether results apply to a particular setting to each member state. Users outside Europe will have to make those judgements for themselves. The comparators reflect European clinical practice, which may differ from treatment patterns in the US.

Ultimately, the opportunity is for JCAs to inform not only how we assess the evidence available today, but also which comparative evidence we generate next, in Europe and beyond.

About the interviewees

Huseyin Naci
Professor, Department of Health Policy, London School of Economics and Political Science (LSE)

Huseyin Naci is a Professor in the Department of Health Policy at the LSE and a Lecturer in the Department of Population Medicine at the Harvard Medical School. He is the founder and Director of the Pharmaceutical Policy Lab and serves as one of the Deputy Directors of LSE Health.

Radek Wasiak
Co-founder, Managing Director, Adigens Health

Radek, an expert in real world data assets and generation and real-world evidence, has provided research services to life sciences for more than two decades. His research spans real-world data assessment, outcome measurement and their real-world data applications. He held research and executive roles at top consulting and clinical-research organizations, driving innovation and team growth.

Disclaimers

The opinions expressed in this feature are those of the author and do not necessarily reflect the views of The Evidence Base® or Becaris Publishing Ltd.