Two new long-term studies, one an extension trial and the other an analysis of real-world experience, show that the RNS System direct brain responsive neurostimulator leads to reduction of seizure frequency in most epilepsy patients who had it implanted. Both studies showed that the benefit from the devices increased over time.
That accruing benefit may be because of improved protocols as clinicians gain experience with the device or because of network remodeling that occurs over time as seizures are controlled. “I think it’s both,” said Martha Morrell, MD, a clinical professor of neurology at Stanford (Calif.) University and chief medical officer at NeuroPace, the company that has marketed the device since it gained FDA approval in 2013.
In both studies, the slope of improvement over time was similar, but the real-world study showed greater improvement at the beginning of treatment. “I think the slopes represent physiological changes, but the fact that [the real-world study] starts with better outcomes is, I think, directly attributable to learning. When the long-term study was started in 2004, this had never been done before, and we had to make a highly educated guess about what we should do, and the initial stimulatory parameters were programmed in a way that’s very similar to what was used for movement disorders,” Dr. Morrell said in an interview.
An Alternative Option
Medications can effectively treat some seizures, but 30%-40% of patients must turn to other options for control. Surgery can sometimes be curative, but is not suitable for some patients. Other stimulation devices include vagus nerve stimulation (VNS), which sends pulses from a chest implant to the vagus nerve, reducing epileptic attacks through an unknown mechanism. Deep brain stimulation (DBS) places electrodes that deliver stimulation to the anterior nucleus of the thalamus, which can spread initially localized seizures.
The RNS device consists of a neurostimulator implanted cranially and connected to leads that are placed based on the individual patient’s seizure focus or foci. It also continuously monitors brain activity and delivers stimulation only when its signal suggests the beginning of a seizure.
That capacity for recording is a key benefit because the information can be stored and analyzed, according to Vikram Rao, MD, PhD, a coinvestigator in the real-world trial and an associate professor and the epilepsy division chief at the University of California, San Francisco, which was one of the trial centers. “You know more precisely than we previously did how many seizures a patient is having. Many of our patients are not able to quantify their seizures with perfect accuracy, so we’re better quantifying their seizure burden,” Dr. Rao said in an interview.
The ability to monitor patients can also improve clinical management. Dr. Morrell recounted an elderly patient who for many years has driven 5 hours for appointments. Recently she was able to review his data from the RNS System remotely. She determined that he was doing fine and, after a telephone consultation, told him he didn’t need to come in for a scheduled visit.
In the real-world analysis, researchers led by Babak Razavi, PhD, and Casey Halpern, MD, at Stanford University conducted a chart review of 150 patients at eight centers who underwent treatment with the RNS system between 2013 and 2018. All patients were followed at least 1 year, with a mean of 2.3 years. Patients had a median of 7.7 disabling seizures per month. The mean value was 52 and the numbers ranged from 0.1 to 3,000. A total of 60% had abnormal brain MRI findings.
At 1 year, subjects achieved a mean 67% decrease in seizure frequency (interquartile range, 50%-94%). At 2 years, that grew to 77%; at 3 or more years, 84%. There was no significant difference in seizure reduction at 1 year according to age, age at epilepsy onset, duration of epilepsy, location of seizure foci, presence of brain MRI abnormalities, prior intracranial monitoring, prior epilepsy surgery, or prior VNS treatment. When patients who underwent a resection at the time of RNS placement were excluded, the results were similar. There were no significant differences in outcome by center.
A total of 11.3% of patients experienced a device-related serious adverse event, and 4% developed infections. The rate of infection was not significantly different between patients who had the neurostimulator and leads implanted alone (3.0%) and patients who had intracranial EEG diagnostic monitoring (ICM) electrodes removed at the same time (6.1%; P = .38).
Although about one-third of the patients who started the long-term study dropped out before completion, most were because the participants moved away from treatment centers, according to Dr. Morrell, and other evidence points squarely to patient satisfaction. “At the end of the battery’s longevity, the neurostimulator needs to be replaced. It’s an outpatient, 45-minute procedure. Over 90% of patients chose to have it replaced. It’s not the answer for everybody, but the substantial majority of patients choose to continue,” she said.
The open-label extension trial, led by Dileep Nair, MD, of the Cleveland Clinic Foundation and Dr. Morrell, followed 230 of the 256 patients who participated in 2-year phase 3 study or feasibility studies, extending device usage to 9 years. A total of 162 completed follow-up (mean, 7.5 years). The median reduction of seizure frequency was 58% at the end of year 3, and 75% by year 9 (P < .0001; Wilcoxon signed rank). Although patient population enrichment could have explained this observation, other analyses confirmed that the improvement was real.
Nearly 75% had at least a 50% reduction in seizure frequency; 35% had a 90% or greater reduction in seizure frequency. Some patients (18.4%) had at least a full year with no seizures, and 62% who had a 1-year seizure-free period experienced no seizures at the latest follow-up. Overall, 21% had no seizures in the last 6 months of follow-up.
For those with a seizure-free period of more than 1 year, the average duration was 3.2 years (range, 1.04-9.6 years). There was no difference in response among patients based on previous antiseizure medication use or previous epilepsy surgery, VNS treatment, or intracranial monitoring, and there were no differences by patient age at enrollment, age of seizure onset, brain imaging abnormality, seizure onset locality, or number of foci.
The researchers noted improvement in overall Quality of Life in Epilepsy Inventory–89 scores at 1 year (mean, +3.2; P < .0001), which continued through year 9 (mean, +1.9; P < .05). Improvements were also seen in epilepsy targeted (mean, +4.5; P < .001) and cognitive domains (mean, +2.5; P = .005). Risk of infection was 4.1% per procedure, and 12.1% of subjects overall experienced a serious device-related implant infection. Of 35 infections, 16 led to device removal.
The extension study was funded by NeuroPace. NeuroPace supported data entry and institutional review board submission for the real-world trial. Dr. Morrell owns stock and is an employee of NeuroPace. Dr Rao has received support from and/or consulted for NeuroPace.
This article originally appeared on MDedge.com.