Combined studies of a medicine with a device or diagnostic
A combined study tests a medicinal product together with a medical device or an in vitro diagnostic, so in the European Union it falls under the clinical trials regulation and under the device or diagnostic regulation at once. Whether a trial is a combined study depends on how each device in it is used. A sponsor that uses a device outside its intended purpose takes on the manufacturer's obligations, and until a coordinated procedure exists the study needs two separate applications.
What a combined study is
In the European Commission's informal definition, a combined study is a clinical trial of a medicinal product together with a performance study of an in vitro diagnostic or a clinical investigation of a medical device. A study that runs an investigational medicinal product in parallel with both is a combined study too. A performance study, or performance evaluation study, is by definition any study undertaken to establish or confirm the analytical or clinical performance of a device.
The regulations and actors involved
The three regulations that govern a combined study are fairly new: the medical device regulation came into force in 2021, the in vitro diagnostic regulation (IVDR) in 2022, and by 2025 the clinical trials regulation was fully implemented. The pharmaceutical legislation is much older, dating from the early 2000s, and as a directive it is much more established. On the device side the actors are the notified body, the national competent authorities and the European Commission; on the medicine side they are the national competent authorities, the European Medicines Agency and the Commission; and a combined medical device and medicine faces the interface of the two. The regulations are designed around high-level principles, and the difficulty lies in the detail. The problems raised are administrative complexity with redundant assessment, similar but not identical questions, ambiguities in interpreting the regulations, and gaps in harmonisation. There is a need for streamlining, and the Commission has heard the plea.
Devices used in a medicine trial
Three categories of device turn up in a medicine trial. The first is a CE-marked device or in vitro diagnostic used within its intended purpose, which the manufacturer defines: the population and the use the device is for. That purpose has to be read carefully, because a device labelled for adults cannot be used in a paediatric population without becoming investigational. Within their intended purpose, such devices are normally used for drug delivery, such as a prefilled syringe or an inhaler, for monitoring or measuring parameters, such as an oximeter or an ECG machine, for collecting data, or for a diagnostic use such as a pregnancy test. The second is the investigational device or the device for a performance study, whose safety and performance are assessed at the same time as the investigational drug. The third is the in-house device, mostly software or tests needed to perform the study, developed by a health institution established in the Union and used without CE marking.
Whether a device makes a trial a combined study depends on how the device is used. A device used within the intended use its manufacturer states does not make one. An investigational device or a device for a performance study can lack a CE mark, or it can carry one and be used outside its intended use, which makes it investigational, just as a drug studied for a new indication becomes investigational. A trial that uses such a device falls into one of two scenarios. In one, the device delivers the drug or measures an endpoint, so it is part of the trial and in its scope. In the other, rare, scenario the device is used only for an exploratory purpose, collecting data with a technology that has nothing to do with what the study sets out to prove, and the study is not automatically a combined study.
Each category has its own regulatory requirements for use in a trial. A CE-marked device used within its intended use has to be CE-marked under the old directives or the new regulations, and because the transition phase is still running, the marks on such products are mixed. It needs at least a CE certificate for high-risk products and a declaration of conformity. An investigational device needs its design validation and analytical performance in place before the trial. The general safety and performance requirements that relate to the trial are validated during it, but the rest, usability among them, have to be done beforehand. Its safety and performance statement has to be signed by the person responsible for regulatory compliance, under Article 15 of the medical device regulation. An in-house device needs no CE marking but still has to comply with some general safety and performance requirements, and with validation and verification. The devices and diagnostics used in a trial often come to the pharmaceutical company from a third-party supplier, so on top of everything the trial requires, the sponsor has to manage the supplier and how far its documentation and compliance have got.
Which of a trial's tests are in vitro diagnostics
A trial uses many tests, and they serve different purposes, such as screening for inclusion or exclusion, monitoring safety during the trial, evaluating efficacy against primary and secondary endpoints, supporting exploratory research, and developing a companion diagnostic. The tests also differ in status. Some are already commercially available and CE-marked, such as the clinical chemistry assays; some are still under evaluation; and some are laboratory-developed tests, as they are called in the United States, or in-house developed devices. The pathway a test has to follow depends on which of those it is, so establishing that is the first step.
In Europe the guidance document MDCG 2022-10 settles which tests are in vitro diagnostics (IVDs) in a trial: only tests that meet the definition of an IVD are IVDs in the context of a clinical trial, so not every assay used in a trial is one. A trial test can be considered an IVD if it has a specific medical purpose and affects the management of an individual patient, and such a test has to undergo specific steps.
The guidance illustrates the criterion with examples of processes, sorted into those with and those without an impact on patient management. Processes with an impact on patient management, such as patient selection or exclusion, allocation of patients to treatment arms, and monitoring of patient safety or of treatment efficacy, fall under the IVDR. Processes without such an impact, such as patient stratification, data-analysis endpoints and exploratory endpoint analysis, fall outside it.
One condition brings IVDR Annex I back in for a test that would otherwise fall outside. For exploratory endpoint analysis, the guidance states that where it is foreseeable that the IVD will in future be used with an IVD purpose, the full general safety and performance requirements of Annex I apply. One IVD regulatory trainer remarks that nobody knows what "foreseeable" means, but that the condition is very important to keep in mind and needs to be taken into account from the start of developing a clinical study with IVDs. Put as a question, the condition asks whether it is predictable that the assay will be used with an impact on patient management in future clinical trials, and where the answer is yes, Annex I also applies.
The sponsor's obligations
The sponsor of the clinical trial is responsible for its initiation and management and for compliance with the clinical trials regulation and with the device or diagnostic regulation as applicable. In the context of clinical trials the IVDR regulates only the development and manufacture of IVDs, not their use, so where an IVD is used outside its intended purpose the clinical trial sponsor assumes the manufacturer's obligations under Article 16(1). The sponsor then has to apply the whole of the regulation. To avoid this, the sponsor should discuss with the supplier beforehand whether the supplier will take over the whole development of the IVD and act as its sponsor, because changing anything about a device's intended use in the trial without clarifying that first can bring the sponsor under Article 16. The sponsor of a trial that uses IVDs is also responsible for the other products used in the trial and must document the competence of the testing laboratories to support the reliability of the results.
The performance study alongside the trial
Using an unapproved IVD in a trial brings the IVDR into play. For a trial that needs a companion diagnostic, the ideal is that the diagnostic is co-developed with the drug and used in the trial, perhaps as fit for purpose at first and then under clinical validation, though this is unrealistic, even if it may sometimes happen. More realistically, the trial runs with a clinical trial assay that is later bridged to the companion diagnostic. Either way an unapproved device is in use, so a performance study has to run in parallel. Such a study is needed in all phases of the trial where a companion diagnostic is used, and the trial and the study together form a combined study.
Performance studies with companion diagnostics must be designed, authorised, conducted and recorded in accordance with IVDR Articles 58 to 77 and Annex XIV. A study using leftover samples only is the exception and needs only to be notified, though not in every EU country. An EU trial that uses a device for medical decision-making without approval for that purpose needs the parallel study wherever the testing takes place, so it is needed even when the samples are tested in the United States.
Authorisation is national. Every EU country with a clinical trial site must authorise the performance study, and a site where blood is merely drawn is a performance study site too. The requirement covers companion diagnostics in development and also assays used as a companion diagnostic in the trial only for inclusion and exclusion, which no party intends to commercialise.
A performance study needs a sponsor; an EU legal representative where the sponsor, usually the laboratory running the test, is established outside the Union; a study identifier; and objectives and endpoints. The combined study needs its separate approvals in place before it starts, and objectives and endpoints of its own that cannot be copied from the clinical trial.
Applications for a combined study
As of 2025, only one of the three regulations is coordinated: clinical trial applications run through a centralised pathway in the Clinical Trials Information System, while for devices and in vitro diagnostics there is no coordination yet, and although it is foreseen it is some way off. There is no common EU procedure for a combined study, so the sponsor files two applications: one for the drug in the Clinical Trials Information System, and one for the clinical investigation or performance study in the competent authority's database where one is available, pending the module of the European device database for clinical investigations and performance studies. The documentation for the two applications overlaps heavily, and filing them in parallel normally makes the work easier. A single identification number is to come, and it will also have to be communicated to the Clinical Trials Information System. The regulation requires an interface between that system and the device database, which could greatly reduce the workload, but as of 2025 work on it had not started. In practice the sponsor faces separate applications, no coordination of the review process or of the two timelines, actors with different knowledge and different perspectives and expectations, no limited validation period, requests for additional information from national competent authorities on the device side, a lack of harmonisation between competent authorities' positions, and delay to the device database module that could reduce the burden and centralise submission.
What the authorities evaluate in a performance study
The competent authority and the ethics committee evaluate a performance study to answer three questions. The first is whether the device is safe to use in the trial, which they judge from the risk management and the analytical performance data. The second is whether the rights of the study subjects are protected, which they judge from the informed consent process. The third is whether the study has meaningful objectives and endpoints.
The competent authorities' opinions on the third question have evolved over the years, particularly for devices that are not companion diagnostics, because a device never meant to be commercialised can sometimes make meaningful endpoints hard to state. By 2025 more and more countries allowed a pragmatic approach to it.
To show that the three questions can be answered, a study first has to fulfil the applicable general safety and performance requirements of the IVDR: a compliant intended purpose statement; a risk management system per ISO 14971, with a positive benefit-risk ratio; a device state of the art in design and manufacture, with design control; and the specific labelling requirements. Analytical performance must be established before the study starts. That can sometimes be a problem for a companion diagnostic under development, because not all its analytical performance steps may have been done in the early-phase trials, and the IVDR does not allow a lighter analytical validation for a phase 1 trial. The performance study plan must meet the IVDR's requirements, be designed to ISO 20916 and comply with the General Data Protection Regulation (GDPR). The plan is not the clinical trial protocol. One year after the performance study, a report on it is due to Europe.
The core package and the country package
The national competent authority and the ethics committee evaluate a performance study from what they have at hand, its submission package. The package is evaluated on its merits, because it is the language in which the authorities are addressed. It consists of a core package that every country requires, plus member-state documents and translations. Besides the package's quality, structure and transparency, knowing what each country needs is key to success.
The core submission, more or less what every country wants, is an application form, an investigator's brochure, the performance study plan, a statement of conformity with the general safety and performance requirements, proof of insurance, the informed consent form, and a description of how the study complies with the GDPR. Two references for writing the performance study plan define safety reporting with a subtle difference between them, and the IVDR's definitions are the ones to follow.
On top of the core, individual countries often ask for documents the IVDR does not require: a cover letter in the local language, a study synopsis in the local language, instructions for use in the local language, CVs of the principal investigators, a list of the other member states involved, and a description of how participants will be compensated. The IVDR also permits a country to request full details of the available technical documentation. In 2025 certain countries asked for the performance evaluation plan, the analytical performance report or the general safety and performance requirements checklist, full risk management detail or more on scientific validity, and insisted on a post-market performance follow-up plan, which to one practitioner does not always seem useful. The IVDR gives theoretical timelines, but in practice, with the sponsor, its legal representative and the member state review process all involved, getting a study authorised took between five and eleven months.
Protocols and other documents
Whether a combined study has one protocol or two can turn on the countries it runs in. One multinational study ran in the United States, China, Japan, Argentina and other countries but needed a clinical investigation only in Europe, so its sponsor wrote one protocol for the drug and one for the device, keeping the device content out of a protocol seen by every country not running an investigation. For another study, confined to Germany, the same sponsor could write a single combined protocol, because only one country was involved.
How to write the protocol, and later the report, are questions a combined study raises. The messaging and content of the two parts have to be consistent: what the clinical study protocol says has to align with what the clinical investigation plan says.
The clinical documents for a drug–device combination line up across the medicinal product and device sides, some with potential for overlap.
| Medicinal product documentation | Overlap potential | Medical device documentation |
|---|---|---|
| Clinical trial application | Clinical investigation application | |
| Clinical evaluation plan | ||
| Clinical trial protocol | Protocol | Clinical investigation plan |
| Medicinal product investigator's brochure | Investigator's brochure | Medical device investigator's brochure |
| Clinical study report | Study report | Clinical investigation report |
| Clinical summaries (overview as part of the Common Technical Document) | Clinical evaluation report (part of the technical file) |
For a medical writer, understanding where the clinical documents for a drug–device combination can overlap would be an invaluable asset.
Whether the investigator's brochure can be a single one for the drug, with a little device information added, and what the informed consent form for a combined study should look like are other questions such a study raises.
The COMBINE programme
The European response to the problems of combined studies is the COMBINE initiative, led by the European Commission with the support of member states and initiated under the Accelerating Clinical Trials in the EU initiative. Its vision is to keep Europe a relevant place for clinical research, with a regulatory framework fit for purpose for sponsors and regulators, so that patients ultimately benefit. Without the initiative's work, patients risk no longer being able to join relevant trials of technically advanced products. Europe is also trying to defend its place in clinical research as somewhere sponsors can run their trials fairly smoothly.
COMBINE began in June 2023 as a project to understand the challenges, ran for about nine months, and then became a programme with a full strategy adopted in December 2024. The project was to explain and analyse the challenges and possibly develop solutions aimed at aligning the interfaces between the regulations, starting with the three that govern a combined study, though many other regulations interfere with the device regulation, such as the AI Act for AI-based medical devices. The analysis, broadly divided into four parts and led by Ireland, drew in fifteen member states and more than fifty-five experts. It began by drawing up an issue list and identifying the relevant activities of every stakeholder, namely industry, the competent authorities and the European Medicines Agency, and its results were reported in 2024. It looked for the challenges that caused delay in starting combined studies, mapped the EU landscape, in which national agencies are not all set up the same way, and the relevant activities where the three regulations come together, and proposed solutions. Almost eighty issues were collected, and fifty of them were judged critical. For one participant in the project, the difficult phase was the first, in which all the issues had to be analysed.
As a programme, COMBINE acquired formal governance, with ownership resting with the Commission, the medical device coordination group and the clinical trials coordination group among others, and a steering board for oversight. Its work is split into cross-sector projects delegated to groups that have to work together, each with a chair and a manager. Its centrepiece is coordinated assessment, planned for certain medicines combined with a device or diagnostic: the idea is one submission covering the clinical trial and the device or diagnostic study, one coordinated assessment by the various parties at national level, one set of requests for information back to the sponsor, one package returned, and at the end a decision satisfactory to all the parties. The other projects are reporting of serious adverse events, where the regulations impose different requirements that need comparing so that guidance can clarify the gaps; the modalities of using devices in medicinal-product trials, which seeks to define in advance the scenarios in which a device or diagnostic is combined with a medicine, so that the framework for each is clarified ahead of time, most likely through new or updated guidance; clarification of the interfaces; training; and exploring the need for advice, for the exchange of best practice among national assessors and for a platform where sponsors can get clarification. Changes to the legislation may be needed as a result of the work.
The coordinated assessment pilot
The pilot of coordinated assessment starts with medicinal products combined with in vitro diagnostics, as industry requested, because most of the studies it conducts involve them. Thirty-four participants had signed up across the clinical trial, device and diagnostic groups, and more member states were joining. The project is to analyse the challenges the pilot shows and define the next steps. The pilot brings the clinical trials regulation's part one and part two together with the competent authority application and the ethics committee opinion on the device side in a single procedure, and where they have synergies, where they differ and where something new is needed is still to be worked out. It runs in the Clinical Trials Information System because there is no other option for now. The device part of a study is often handled by the device supplier and the drug part by the drug company, and the device world does not use that system and does not know how to, though this is expected to be workable for the pilot. Where the procedure runs in future is undecided.
The Commission was expected to announce the pilot in June 2025, and the pilot to start after the summer. Planned alongside it were a workshop from 29 May on the challenges of reporting serious and other adverse events in combined studies, meetings outlining the device scenarios and their regulatory challenges with a recap in June, a status report on the pilot and communication to sponsors in mid-June, and an external webinar planned by the Commission. The Accelerating Clinical Trials in the EU initiative has taken COMBINE under its wing, which brings closer monitoring and opportunities for dialogue. Companies that join the pilot pave the way for those that come after them, and it can be a huge learning for both. The Commission's website carries information on the initiative and will show when the pilot is published.
What the work has shown and still needs
For the European Medicines Agency, which supports the programme, the work is not easy, but it is worth sticking at for the benefit of patients, and progress is slow but is progress. Delays are costly for industry, and even coordinated assessment through the Clinical Trials Information System is not plain sailing. To one participant, the collaboration between industry, competent authorities and the other parties was very constructive and well managed.
The priorities the work identified were coordinated assessment between the trial and device sides, together with the IT infrastructure for it, since the device database module was not yet in production and an alternative had to be found; alignment of member states' positions, with procedures agreed on a common approach; sponsor awareness of what is needed from them; guidance and clarity on combined studies; and clear dialogue and scientific or technical advice. The device side lacks scientific advice of the kind the pharmaceutical side has, and a notified body cannot currently be involved in it. As of 2025 there is no stakeholder platform of the kind the clinical trial side has, and the European Medicines Agency tries to facilitate informal and formal dialogue. A focus group looked at scientific advice and good practice to help sponsors navigate the field, and its publication was expected shortly. What the work still needs is experts from both sides, more resources for both sides, device expertise, open dialogue, exchange of best practice, training for sponsors and for the other stakeholders, and cross-functional national teams in member states.