Robert Fisher

Listen as Robert Fisher MD, PhD, editor emeritus of and director of the Stanford Comprehensive Epilepsy Program, discusses trials for epilepsy devices.

Recent years have seen adoption of several therapeutic devices for epilepsy, such as the vagus nerve stimulator, responsive neurostimulation, deep brain stimulation, and Visualase laser surgery. More, are on the way? Behind each of these approvals lie one or more clinical trials.

Although device clinical trials are effortful, people with epilepsy are to be encouraged to volunteer and medical personnel to participate. Participation in a successful clinical trial not only advances therapy for the epilepsy community, but also gives participants use of the device years before it becomes generally available.

What Happens During a Device Trial

The preferred first step of a device trial is testing in animal models to confirm a preliminary hypothesis and develop information on approximate device parameter settings. However, animal models also have limitations, so several devices have been tested first in people, with later refinement in animal models.

A trial should focus on a relatively homogeneous population, not on all people with epilepsy, because variability of response of different seizure types may obscure efficacy in the most appropriate sub-population. Several neuromodulation devices require precise targeting. Choosing the target, implanting accurately in the target and verifying proper placement all pose special issues different from any issues with drug trials.

Of particular practical concern is documentation that a device is “MR conditional,” meaning that it has no significant hazards in a specified and tested MRI environment. Sacrificing the ability to perform an MRI would be a large price to pay for an MR unsafe device. Most clinically used neurostimulation devices have been shown to be MR conditional, but this needs to be verified before performing an individual MRI.

Drug trials pre-specify dose and dosing interval. Neurostimulation trials entail many more possible parameters, including pulse frequency, shape, width, train duration, intensity, constant current versus constant voltage, bipolar versus long-distance referential stimulation, and whether stimulation is delivered continuously, on a schedule or contingently, for instance, in response to a detected seizure. Knowledge has so far been insufficient to select optimal parameters, and intelligent guesswork has been required before most device trials.

The Benefits of a Pilot Study

Pilot studies are useful to confirm a preliminary sense of efficacy, safety and tolerability of a therapeutic device and to provide information for design of a larger more definitive trial. Approval of a significant-risk device usually requires Class I evidence derived from a large, randomized, placebo-control trial.

Since some devices produce detectable side effects, it is important to carefully consider proper blinding procedures. Effective devices can produce many benefits, such as reduction of seizure number compared to baseline, increased responder rates, diminished seizure severity, less postictal morbidity, fewer injuries and improved quality of life. Secondary outcomes are important, but investigators will be obliged to pre-specify one primary outcome around which the trial will succeed or fail.

The story does not stop with completion of a successful device trial, but continues with efforts to satisfy regulatory authorities and third-party payers.

New Devices as a Treatment Option

Invasive devices will not likely replace medications or even become first-line therapy, but they do provide an alternative to medications for treating people with epilepsy who do not respond to medications or are not tolerant of medication side effects. People with epilepsy and clinicians will be excited to have something different from medications and potentially useful to try with the new devices.

This content was created through a partnership between the Epilepsy Foundation and American Epilepsy Society.

Authored By: 
Robert Fisher MD, PhD
Authored Date: