CLINICAL EVIDENCE DILEMMAS: HOW LONG SHOULD PATIENT FOLLOW-UP BE?

The Implant Files touched upon one of the toughest dilemma’s designing a clinical study: How long do patients treated with a medical device need to be followed in a clinical study before one can say the device is safe and performing? Patient follow-up can have an enormous impact on the duration of your study, and the tendency is to keep it as short as possible; the shorter your study is, the sooner you will have the results, and thereby your publication or market approval. However, what is required, desirable, and even possible strongly depends on the purpose of the study and the medical device involved.

As a last post in the series of clinical evidence dilemmas, some of the key aspects to consider when deciding on the length of patient follow-up in your clinical study.

Pre-market studies

Besides the sample size and enrolment rate, patient follow-up can have a significant impact on the duration of your trial: imagine the length of a clinical study with an implantable device that is supposed to function for the rest of your life …, but when a study concerns a pre-market study to show safety and performance or efficacy, patient follow-up has to be long enough to prove (together with other existing data) safety and performance, while being short enough to bring the concerning device to market as soon as possible. So the golden question is

How long is long enough?

Treatment duration

Some devices are meant to be used for a short lasting period, so the required patient follow-up duration will be relatively short (or even non-existing for certain diagnostic devices). For example, examining the safety and performance of a new dialyzer during a hemodialysis session of ~4 hours, depending on your exact study objectives and the status of the concerning patients, generally needs a patient follow-up of 1 day. When the treatment, however, involves surgery and/ or general anesthesia it becomes a different story.

Surgery

Device effects may be clear within few days, but when surgery is involved and/ or the application of the device requires general anesthesia (especially when the current standard of care or alternative treatments do not), you also need to take into account the effects of general anesthesia, which are known to last for several months especially in the elderly. Thereby prolonging of the patient follow-up in your clinical study with several months, with a minimum of 30-days after the intervention.

Implantables

The situation becomes even more complex when implantable devices are involved that are supposed to function for several years. As said above: imagine the duration (and cost) of a clinical study where patients are to be followed for the rest of their life’s … By the time you collected the safety and performance data, the concerning device will be outdated. Therefore typically acceleration testing, modelling techniques, and clinical data from comparable devices are used to get a reasonable idea of long term safety and performance. That such is not always good enough has become clear in the last few decades:  Possible deleterious effects of mesh and metal hips implants became only visible after several years after implantation. Which is exactly why MEDDEV 2.7/1 Rev 4 and 2.12-1, and from 2020 on also the MDR, require a much more solid Post-Market Clinical Follow (PMCF) program for the higher risk class of devices and implantables.

Post-market studies

Determining the right patient follow-up duration for a post-market clinical study, so once the device is on the market, tends to lead to even more discussion than for a pre-market study. Besides device durability, a wider range of study objectives other than ‘just’ safety and performance start weighing in, as well as the type of patient involved, but also in a post-market setting one will want to have the study results as fast as possible.

Clinical evidence gap

When, however, you need to perform a PMCF study because the (pre-market) clinical evaluation indicated, for example, there is an uncertainty on long term clinical safety or performance, it is pretty clear that patient follow-up needs to be substantially longer than for the existing data supporting the safety and performance claims at market release. Still the question remains ‘how long is long enough’ and the factors mentioned above apply. Currently I am involved in several post-market clinical follow-up studies following patients for 3 to 5 years after the device was implanted. This is pretty long considering the standard product life cycle is 18 to 24 months, but still may not be long enough for a device that is supposed to last for more than 10 years. It is not for nothing that the FDA recently required an extension of the follow-up period from 3 to 5 years for women who received the Essure birth control devices.

Patient outcomes

Besides data on safety and performance, post-market medical device studies often aim to collect clinical data on patient outcomes due to the fact that in the pre-market setting clinical evidence collection tends to focus on device performance, especially in Europe, and evidence with respect to efficacy or patient benefits often is indirect. Clinical outcomes, such as a reduced probability of mortality, morbidity, and enhanced Quality of Life, however, add yet another complicating factor, since besides the study objective and the type of device also the type of patient involved needs to be taken into account: A clinical study aiming to show the impact on patient survival with a device that has been shown to effectively destroy cancer cells, for example, will require a much longer follow-up period in patients with prostate cancer than in patients with pancreatic cancer, simply because the survival rates are so very different.

Conclusion

In short, selecting the right patient follow-up duration in your medical device study is not an easy process. The general tendency is to keep the study duration as short as possible, but different from medicinal products, the process of determining the minimal follow-up duration is affected by a complex interplay of regulatory requirements,  study objective(s) and endpoints, as well as the type of medical device and patients involved.

Feel free reaching out in case you require support developping your medical device study. You can reach me through the ACS website or the information displayed in the right upper hand corner.

Annet