University of California, San Francisco (UCSF) scientists have developed a form of neurological pacemaker that adapts in real time to a patient’s walking and could address one of the most disabling and hard-to-treat symptoms of Parkinson’s disease.
Over 10 million people worldwide live with Parkinson’s disease. Deep brain stimulation (DBS) has been shown to improve certain symptoms, like tremor, stiffness, and slowness, but has limited effects on walking.
Researchers think this is because walking is a highly dynamic behaviour that requires precise timing across both sides of the body. And because a person’s gait is constantly changing, the fixed stimulation pattern delivered by conventional DBS doesn’t reflect the rapid coordination among the brain, spinal cord, and muscles required to walk.
Just like a cardiac pacemaker responds to the heart’s rhythm, the team’s implanted personalised adaptive DBS (aDBS) responds to the rhythm of walking. It detects specific neural signals corresponding to different phases of gait, including whether the left or right leg is swinging, and automatically adjusts stimulation within fractions of a second.
“The early results are encouraging,” UCSF neurosurgeon Doris D Wang tells Refractor.
In laboratory testing, the aDBS system improved gait symmetry and reduced variability in walking patterns, and participants reported fewer falls while maintaining overall control of Parkinson’s symptoms when using the system in their daily lives.
“Perhaps the strongest endorsement came from the patients themselves – after experiencing both therapies at home in a blinded fashion, the participants who remained in the study chose to continue using adaptive DBS when given the option for more than a year after the trial ended,” Wang adds.
One of the most interesting findings was that while many patients exhibited changes within similar frequency ranges, the optimal signals and recording locations varied substantially between individuals.
“In some patients, the most informative signals came from the cortex, while in others they came from the basal ganglia,” Wang says. “This highlights an important principle for future neuromodulation therapies: there is unlikely to be a single biomarker that works for everyone. Instead, we will probably need personalized approaches that identify and respond to each patient's unique neural signatures.”
The study involved just five people, so larger studies are needed to understand how these algorithms perform across diverse patient populations, different stages of Parkinson's disease, and varying gait impairments, Wang says, but “the ultimate goal is to create adaptive systems that can continuously personalize therapy as patients' symptoms and needs change over time.”
The aDBS also relied on additional research electrodes and a specialized investigational device that are not part of standard clinical DBS therapy.
“Future versions will need to achieve similar performance using commercially available hardware and signals that can be recorded from routinely implanted electrodes,” Wang says.
Additionally, patient-specific biomarkers were identified through detailed in-clinic testing. “For this technology to become broadly practical, devices will need to automatically discover and adapt to an individual's neural signatures with minimal clinician intervention,” she adds.
The next step is to translate these findings into technologies that are more practical for widespread clinical use. This includes identifying gait-related biomarkers that can be detected without additional electrodes, developing automated methods for biomarker discovery, and conducting larger multicentre studies to evaluate long-term safety, efficacy, and patient benefit.
And the underlying concept extends “far beyond walking,” Wang says. “Parkinson's disease affects many domains of function, including sleep, cognition, mood, and other motor symptoms. In principle, adaptive stimulation could be designed to detect neural signatures associated with these states and deliver therapy only when needed.”
The platform also provides a framework for treating brain disorders using real-time feedback from the nervous system itself, Wang adds.
“Similar approaches are already being explored for conditions such as obsessive-compulsive disorder, depression, epilepsy, chronic pain, and other neuropsychiatric disorders. We believe this study represents part of a larger shift toward intelligent neuromodulation systems that continuously adapt to a patient's changing brain state.”
This study was published in Nature Medicine.
Fact-checked by Mike McRae.
