Clinical evaluation – more process than paper

The new version of the European regulatory framework for medical devices in the form of the MDR has resulted in a dynamization that affects existing medical devices as well as new products.

The central elements “Clinical Evaluation Plan & Clinical Evaluation” should not be considered as static documents, but rather as a dynamic and ongoing process. This process involves the systematic and continuous collection, analysis and evaluation of clinical data to ensure the safety and performance of a medical device throughout its life cycle.

The purpose of this process is not only to meet regulatory requirements, but also to define a clear “roadmap to CE”. This will determine the time and effort required to obtain the necessary clinical evidence:

These questions – posed in the clinical evaluation plan – set the course, especially for innovative medical devices. The implementation of the plan leads to the clinical evaluation and is more of a network than a straight road. The above questions must be asked during the development phase. Milestones, efforts and goals are derived from these questions.

“Sugar Sentry” – A hypothetical example of a Class IIa product

A hypothetical example is the “SugarSentry IIa”, an app that scans the barcode of groceries and displays the amount of sugar that is ingested with the consumption of this food.

The manufacturer formulates the intended use in such a way that it corresponds to the definition of a medical device and the app is classified as a Class IIa product in accordance with MDR Classification Rule 11.

The hypothetical “SugarSentry IIa” monitors the amount of sugar ingested. The intended user group includes adults and adolescents who want to reduce their sugar consumption in order to avoid health risks.

So far, so good. But now it’s time to take this medical device through the conformity assessment and demonstrate its safety and performance. The performance and safety parameters of this app can be easily reduced to technical aspects, as the use of the app in terms of its intended use does not require any interaction with the user in the first instance.

According to MDR Article 2(22), performance refers to the ability of a device to fulfill its intended use as stated by the manufacturer. The direct performance and safety of the “SugarSentry IIa” is defined by its ability to scan the barcode and correctly display the sugar content and the amount of sugar consumed. The proof of these “claims” should be achievable by purely technical means through appropriate tests, comparison with standards and software verification and validation.

The need for clinical data on performance and safety

The goal of reducing sugar consumption formulated in the further definition of the intended use can only be proven at this point by means of clinical data from test subjects. The source of this data is diverse. For Class IIa products, the MDR does not initially impose any major restrictions here.

Option 1:

If your own company has data on sugar reduction using a predecessor app or if there is published data on sugar reduction using a competitor app, this data can be used as evidence.

However, the precondition is that the predecessor app or the competitor app is equivalent to “SugarSentry IIa” in technical, biological and clinical terms. If this approach is pursued, this must be taken into account during the definition and development of the app (e.g. same application conditions, same target group, same technical implementation, etc.).

Option 2:

We assume that the app is truly innovative and that there are no comparable competing systems. In this case, separate data must be collected as part of a “clinical investigation”. The definition of “clinical investigation” according to MDR §2(45) initially only states that this is a systematic investigation involving one or more human subjects and conducted for the purpose of evaluating the safety or performance of a device. Whether this clinical investigation includes only a treatment group or also a control group, what type of participants are included or excluded and where and under what circumstances this investigation takes place depends on the context of use, the intended use, the target group and the risk profile, etc. of the medical device to be tested.

Again, the Clinical Evaluation Plan and the Clinical Evaluation that align the required evidence with the available evidence, with the context of use and with the risk profile of the device is at the center of the action – and are once again within the developers’ authority to interpret. The clinical evaluation process captures changes, adaptations and developments in the specifications of the product and identifies gaps in the evidence and the strategies needed to close them.

For example, restricting the target group to adults would simplify the approval and implementation process of a clinical study, as the inclusion of adolescent test subjects creates higher ethical and organizational hurdles. On the other hand, this also eliminates potential users in the context of marketing, as CE certification then excludes use by adolescents for the time being.

The intended use – keyword sugar reduction – provides the rough framework for a possible two–arm clinical study. Treatment group with app and control group without app, each of which is then questioned about sugar consumption after a certain period of time (with potential for uncertainty), tested directly by blood glucose (invasive; makes the test more complex) or indirectly by surrogate markers (e.g. weight loss). This results in the quantifiable clinical performance of sugar reduction.

How to prove the clinical benefit of “Sugar Sentry”?

Note the definition of the term “clinical performance” according to MDR §2 (52):

The clinical benefit would be much more difficult to demonstrate at this point. It would take a long–term study under very controlled conditions to prove directly that the use of the “SugarSentry IIa App” can lead to a long–term reduction in the risk of diabetes, obesity or tooth decay (which are also, but not only, caused by excessive sugar consumption).

A multi–stage concept is also conceivable. The verification of the direct (clinical) performance in the context of the controlled trial described and the clinical benefit derived from this in the context of a subsequent broader register (possibly also in post–market observation). The actual clinical benefit (in this case prevention of diseases caused by high sugar consumption) is often much more difficult to record than the direct benefit. This often requires a different type of study, such as the above–mentioned post–market registry.

However, the clinical benefit can also be derived from the so–called state of knowledge of medical science (based on the current clinical–medical literature and published guidelines/recommendations of medical societies). For example, it is almost common medical knowledge that reducing sugar consumption on an individual basis and in the context of public health has preventive and health–promoting effects.

It is certainly appropriate that a manufacturer can fall back on general, reputable medical knowledge if the intended use and clinical performance are successfully demonstrated. This must then be adequately presented as part of the clinical evaluation (and also followed up through regular revisions, as required by the MDR, in the pre– and post–market phases).

Evidence for marketing claims

Beyond the terms performance, clinical performance and clinical benefit, the so–called clinically relevant “claims” now come into play. These are often driven by marketing but, if present, are certainly subject to clinical evaluation/evidence. A “claim” such as “…SugarSentry IIa halves sugar consumption within the first week of use…” also needs to be proven.

In turn, the manufacturer has the power of definition and can compare this with the available data, make it more precise or even reject it.

In the end, the clinical evaluation plan dynamically summarizes these scenarios into a consistent strategy, which leads to the evidence presentation of the clinical evaluation and thus to CE conformity.