Clinical trials: "Registry-embedded clinical trials are the way of the future"

Interview with Prof. Edmund Neugebauer, Head of the Institute of Research in Operative Medicine (IFOM), University Witten/Herdecke

Even medical risk products are not always tested as thoroughly as would be necessary – be it because of criminal energy, lack of know-how or financial reasons. A revision of clinical trial procedures could not only fix loop holes and methodological flaws. Products and methods could also be brought into general medical care more quickly under new rules.


Photo: Prof. Edmund Neugebauer

Prof. Edmund Neugebauer; © private talked to Prof. Edmund Neugebauer. He advocates clinical trials for risk products that consist of several stages. Part of this process would be documentation in registries during the whole process from the beginning and the close integration of manufacturers into the trial.

Professor Neugebauer, in your opinion, when is the clinical trial of medical devices insufficient?

Edmund Neugebauer: This is the case, when the benefit has not been proven for patients. The Federal Joint Committee G-BA provides criteria for the benefit assessment of methods in medicine, which I share. These are primarily patient-relevant criteria such as mortality, morbidity rate and health-related quality of life. The additional benefit of the new method compared to other methods with the same objective should also be demonstrated. Finally, the benefit needs to be weighed against the risks and side effects of the treatment.

How can the risk for patients be minimized as much as possible?

Neugebauer: Studies need to be designed to where the potential for distortions (bias) is as minor as possible. Randomized, controlled studies cause the least bias. A very simple study, such as a case series, causes a major bias, because there is only prospective observation of patients without the explicit comparison with a control group. If randomization is not possible in studies on medical devices, at least a parallel control study should be conducted.

What is behind bias in studies?

Neugebauer: There are several effects. First of all, diseases always have a natural progression. The body also gradually heals itself without requiring any treatment. This can distort the assessment – the result is being interpreted as being too positive. Secondly, there is the Hawthorne effect. Just the fact alone that a patient knows that he/she is part of a study, has a positive effect on the result. The placebo effect can also play a role. It can occur, when patients respond very enthusiastically to the use of the new method in a study and therefore like to communicate a benefit that is perhaps not there. These bias effects, which can account for up to 60 per cent of the overall effect, are being minimized through comparative studies.
Graphic: Four stair landings with doors above each other

This could be the future of clinical trials for medical risk products: a multi-staged process that is completely documented in a registry; © Milosh Kojadinovich

How should clinic trials ideally be conducted?

Neugebauer: I support an implementation in several steps for class II b and III devices. The initial "first in man" step is very critical. A rational and clear description of selected patients and very accurate patient documentation are necessary. Patient safety criteria as well as all details of the new procedure and the team need to be recorded and subsequently evaluated. An ethics committee has to be involved in any case and a detailed case report written.

The second step is the further development of the technology, which needs to take place together with the author of the treatment method. Team effects are being assessed here. This step is meant to improve the process as well as complement data on the process and patient safety. Essentially, you are meant to determine whether the procedure is a feasible alternative to the standard method.

Before the formal evaluation, the intermediate step is the informal comparison with the standard method: in this instance, you compare the process and the short-term effect with the standard method. This is followed by the formal evaluation of the new technology compared to the standard. This best occurs through a prospective controlled comparative clinical study compared to the appropriate standard method. This includes PROs, patient-reported outcomes, and the cost-benefit equation as outcome measures.

The last step is the application of the medical device during normal patient-centered care. Ideally, quality-related and safety data should be collected through structured systematic monitoring and collected in a database. Rare results, which are often not detected in controlled studies, are primarily important here.

It would make sense to also link randomized trials with prospective registries that are created right from the beginning of the trial. The first steps with the product are being documented here and you are able to compare the trial data with previous data. These are so-called "registry-embedded clinical trials", which are the way of the future from my point of view.
Photo: Training of medical staff

Manufacturers can well be integrated into studies. They can support health care providers - hospital staff and physicians - with know-how and train them in the handling of a new device; ©

What role would manufacturers play in this whole process?

Neugebauer: Manufacturers are understandably interested in launching their products on the market as soon as possible and for them to be safely used. They are able to provide valuable support in this case. They should provide the care providers with comprehensive product-specific information and support the participating hospitals and physicians with their expertise. They can develop guidelines for best practices pertaining to the product and train the staff.

How do you rate the chances of the framework for clinical trials changing?

Neugebauer: I really hope that they will change. Part of it is actually already meant to change. In Europe, between 20 and 30 particularly competent sites for high-risk medical devices are meant to be certified in the future. It is a good thing that inspection authorities are obligated to conduct unannounced spot checks at manufacturers. We also need more transparency in the system through a clinical trial registry for medical devices like the ones that already exist for pharmaceutical clinical trials.

As clinical researchers, we are also currently designing structures to conduct trials much more quickly and increase manufacturer involvement. We have built the CHIR-Net, a surgical trials network, where more than 250 hospitals are participating in nationwide. In doing so, we are able to conduct clinical trials and testing very quickly. This is also how we address the concern of manufacturers that their innovation is already outdated before testing has been completed.
Photo: Timo Roth; Copyright: B. Frommann

© B. Frommann

The interview was conducted by Timo Roth and translated from German by Elena O'Meara.