Why do we need clinical trials for medical devices?
As part of our review of the new European Medical Devices Regulation (MDR) earlier this year, one of the topics that caught our eye was the expansion and clarification of the rules regarding clinical data for medical devices.
The difference between a clinical investigation and a clinical evaluation
Many people colloquially refer to “clinical trials” as the source for clinical data for efficacy of treatments, where the trial is the activity of testing a medical device with patients to confirm that it provides clinical benefits. In reality, this is what is defined as a clinical investigation within the regulations. The regulations also define a clinical evaluation as a wider scope activity, taking into account all scientific data as well as the data from a clinical investigation as a subset activity.
Where have we come from versus where we are headed
Previously, the European Union (EU) Medical Device Directive (MDD) acknowledged that clinical data is required to show that a device will comply with the essential requirements. It briefly outlined (in only nine paragraphs) the expectations of clinical investigations.
The new MDR expands on this and takes up a whole chapter on the subject. Both the MDD and the MDR dedicate an annex to how clinical trials should be run but, again, the MDD only provided minimal guidance. Under the MDD, many people chose to look to the EU Clinical Trials Directive (and subsequent regulation) and the associated Good Clinical Practice guidelines from the International Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use. Although intended for medicinal products, these gave a greater indication of the expectations for clinical investigations.
Here at Cambridge Design Partnership we were pleased to see that clinical investigation of medical devices has been recognised as an activity in its own right, and that equivalent instruction is now provided in the MDR. In particular, the MDR appears to have been well aligned with the Clinical Trials Regulation with the intention of using the same electronic systems to help with centralised processing of applications. This factor is particularly relevant when considering drug delivery devices or combination devices where both aspects (medicinal product and medical device) need to be considered.
Who needs to do what?
One particular aspect of the expansion and clarification of the requirements for clinical data which has been central to many discussions between us in the quality team are the articles and clauses pertaining to whether a clinical investigation is required.
The MDR is very clear that all products require a clinical evaluation, which comprises a review of relevant scientific literature relating to the safety, performance, design characteristics and intended purposes of the device; a review of all available clinical investigations data (i.e. not just your own but any published data); and a consideration of any currently available alternative treatment options for that purpose.
The MDR then goes on to lay out the rules to determine which classes of devices also require clinical investigations to be carried out. When you analyse the list of requirements, this results in quite a small subset of products starting from Class III and Class IIb (implantables only) and then narrowing down after that with various modifiers.
Compared with the MDD’s requirements – which covered long-term invasive devices as well as implantables, and Class IIa devices as well as Class IIb – we initially considered the MDR’s requirements a simplification. In addition, the MDD requirements were quite vague and didn’t actually specify which devices had to have a clinical investigation – and the listed devices had to wait 60 days before starting their investigation to allow for the competent authorities to have their say, which infers that they were required to have clinical investigation.
However, when we started to apply this to our projects and give guidance to our clients, we noticed something interesting. Although most of the products involved were not required to have a clinical investigation due to their classification, the reality was that as they were usually novel devices, there was not sufficient pre-existing clinical investigation data or scientific literature which the client was aware of to allow for a clinical evaluation to be carried out without a clinical investigation.
This is something that should be considered early in the project when developing the regulatory strategy to plan how this supporting information can be gathered and allow time for this activity.
It will be interesting to see what impact this has on an industry which is already predisposed to look to predicate technologies to allow 510k applications in the US rather than the expanded effort required for premarket approval (PMA) and how this will be resolved for products brought to the market in the EU.