Adaptive brain stimulation administers automated on-demand therapy

Alfonso Fasano still remembers the first time he saw a patient treated with deep brain stimulation. He was only 20 years old at the time, studying medicine at the Catholic University of Rome. The patient was a woman with young-onset Parkinson’s disease; the disease struck before she reached age 50.

She experienced prominent tremors, one of the hallmark symptoms of Parkinson’s disease. Tremors significantly affect the quality of patients’ lives, interfering with their abilities to work, engage in hobbies, and eat, as well as causing substantial emotional distress (1).

For treatment, surgeons implanted electrodes into her subthalamic nucleus, a small structure nestled deep within the hemisphere of the brain. That day in the clinic, the physician turned on the deep brain stimulation (DBS) device. Almost immediately, her tremors melted away. “It was mind-blowing to see the effect,” Fasano recalled. This experience shaped his future career path. Today, he is a neurologist and neuromodulation researcher at the University of Toronto who treats patients with DBS.

DBS is currently FDA-approved to treat some movement disorders, including Parkinson’s disease, drug-resistant epilepsy, and obsessive-compulsive disorder (2). While DBS is highly effective in some respects, many researchers think that there’s still room for improvement. The present form of DBS involves continuous stimulation of a specific brain area, which may result in unwanted side effects or, in some cases, cause the treatment to lose efficacy over time. This constant delivery of therapy might not actually be necessary.

“There are some conditions that are episodic,” said Casey Halpern, a neurosurgeon at the University of Pennsylvania. “Patients will spend plenty of their time in a baseline, normal state, until they get an exacerbation.”

Pharmaceutical treatments can’t be responsive to patient needs on these short time scales. Medications take time to travel from the digestive system to the bloodstream to their targets in the brain. Electrical stimulation, however, is subject to no such time constraints; it can be turned on and off just like a light switch.

Researchers are now investigating adaptive DBS, sometimes also called closed-loop DBS, which can record electrical activity in the patient’s brain, identify a problematic signal, then deliver stimulation to correct the problem, providing on-demand treatment precisely when the patient needs it. Dozens of adaptive DBS trials are currently underway as researchers hope to help patients with everything from Parkinson’s disease to treatment-resistant depression.  

A focus on Parkinson’s disease

The efficacy of DBS therapy for a particular disorder relies on researchers finding the correct brain area for intervention. For Parkinson’s disease, these efforts have a surprisingly long history. In James Parkinson’s first description of his eponymous disease in 1817, he noted that one of his patients’ tremors improved after the patient suffered a stroke. Inspired by this observation, early and mid-20^th century researchers lesioned, or intentionally damaged, various motor control brain regions in attempts to ameliorate symptoms of Parkinson’s disease in their own patients (3). Through a process of trial and error that often involved relatively high mortality rates, they eventually identified a specific region of the brain that, when lesioned, reduced tremors. 

In their search for a safer approach, researchers realized that high-frequency electrical stimulation could mimic the effect of a lesion, and DBS eventually replaced surgical lesions as a more adjustable way to reduce tremors in patients with Parkinson’s disease. In 2002, the FDA officially approved DBS for treating Parkinson’s disease (4). 

The technique still wasn’t perfect, though. While some symptoms like tremor often responded well to DBS, Fasano noted that other symptoms, such as freezing of gait (the sudden and somewhat inexplicable inability to take a step forward) or speech difficulties, sometimes worsened.

“In the early days of long-term studies,” said Fasano, “the narrative was that this happens because of disease progression.” But Fasano didn’t think that was the whole story. He had observed long-term DBS-treated patients who showed improvements in certain symptoms if the stimulation was turned off, indicating that the worsening of these symptoms couldn’t solely be due to disease progression.

Fasano hypothesizes that controlling the amount of stimulation by providing it when needed instead of constantly might reduce these side effects. “It's complex because freezing of gait can be improved or worsened by DBS, but maybe adaptive stimulation will just give you the good without the bad,” said Fasano.

Conventional DBS requires knowing where to stimulate; for adaptive DBS, researchers must also know when to stimulate. By recording brain activity from the same electrodes used for DBS stimulation, researchers identified that certain types of activity in the beta band — the brainwaves oscillating at 13-30 hertz — correlated with worse symptoms (5). Therefore, researchers decided to test an adaptive DBS system that kicked in when the abnormal beta activity appeared.

Beginning in the 2010s, preliminary trials in small numbers of patients showed that adaptive stimulation based on beta activity could effectively reduce Parkinson’s disease symptoms, and researchers have been working to finetune this technique ever since (6,7). Several trials of adaptive DBS are currently in progress to study the efficacy of different devices and determine optimal stimulation parameters (8). Fasano is currently preparing to begin a small trial to determine whether gait and speech impairments are reduced when patients are treated with adaptive DBS compared to continuous DBS (9).

In the future, adaptive DBS will likely be further refined, according to Fasano. While beta power has been the most extensively studied, the activity in other frequency bands may be related to other symptoms such as cognitive difficulties or side effects of Parkinson’s disease medications such as the erratic, involuntary movements known as dyskinesia (10,11). He believes that one day, devices may detect and respond to multiple types of brain activity, taking into account the full picture rather than focusing on one frequency band at a time, which will hopefully result in precise control of more symptoms.

Restoring control in other diseases

As a neurosurgeon specializing in DBS, Halpern has seen first-hand how much this treatment can benefit patients with movement disorders, including Parkinson’s disease. Halpern thinks that this transformational therapy could be useful for other diseases as well.

Currently, he is working with patients with binge eating disorder. This disorder affects more than 1 in 50 Americans at some point in their lives, and is associated with an increased risk of metabolic syndrome and hypertension (12,13).

In order to find the appropriate brain area to target, Halpern and his research team began by studying mice to figure out what was happening in their brains right before they engaged in binge eating behavior. They decided to examine activity in the nucleus accumbens, an important part of the brain’s reward circuit that seems to be activated in anticipation of something good.

In mice conditioned to exhibit binge-like eating behavior, researchers observed an increase in delta power (brain waves oscillating at one to four hertz) right before the mice began chowing down on high-fat foods (14). Importantly, they didn’t observe this delta boost when the mice were about to eat their normal, healthy food, indicating that it wasn’t simply a signal associated with normal eating. Using this delta signal to trigger a short bout of brain stimulation turned out to significantly reduce binge eating in mice.

The team was eager to determine whether brain stimulation could provide the same benefits in humans. In a pilot study, they enrolled two patients with binge eating disorder and severe obesity who had not been helped by bariatric surgery (15). After implanting electrodes in the nucleus accumbens of the patients, the researchers recorded the participants’ brain activity during exposure to high-calorie foods.

“We actually found a fairly similar signal within the human nucleus accumbens as we saw in mice,” said Halpern. Researchers saw an increase in low-frequency brain activity right before an episode of loss-of-control eating.

For six months, the device automatically triggered a short bout of stimulation whenever it sensed the low-frequency brain activity that signaled the onset of binge eating behavior. Over this time period, the participants reported an increased feeling of control over eating; one participant’s condition improved so much that she no longer met the criteria for binge eating disorder. “Subjectively, they felt that they could make healthier decisions, that they didn’t crave food in the way that they did before,” said Halpern.

Although the exact mechanisms by which this treatment works are still not fully understood, Halpern said that in humans with binge eating disorder, the connectivity between the nucleus accumbens and cortical control centers seems to be abnormally low. “By delivering this activating episodic intervention, we believe that what we're doing is restoring relatively normal connectivity to this circuit,” he said.

Halpern is planning to validate these early promising results in more patients with obesity and loss of control over eating behavior, as well as exploring the combined effects of adaptive DBS with other therapies, such as cognitive behavioral therapy or pharmaceutical interventions. He also believes that a similar treatment could apply to other types of disorders that involve feelings of loss of control.

“[The goal of this research] is not just to help people with eating disorders. This subjective sense that you can’t control yourself is pervasive in so many mental health conditions: suicidality in depression, compulsive behavior in obsessive compulsive disorder, panic in post-traumatic stress disorder, addiction,” said Halpern. “It’s the pervasiveness that I find so intriguing.”

What the DBS future holds

Many researchers are now wondering just how far adaptive DBS can go, but expansion into a wider variety of neurological and psychiatric disorders could prove challenging. For adaptive DBS therapy to be successful, researchers must not only identify the brain region driving symptoms, but also figure out the specific pattern of brain activity that indicates onset or increase in symptom severity. For some diseases, the appropriate brain region and pattern of brain activity is relatively consistent from patient to patient, so the same surgical procedures and stimulation triggers can be used in most patients, with only the timing of stimulation being personalized to each patient’s brain activity.

For a condition like depression, however, the picture looks much different. Several brain regions and circuits have been implicated in the pathophysiology of depression, with considerable variation from study to study and patient to patient (16). Indeed, conventional DBS has shown mixed results for treatment-resistant depression and in one of the larger randomized controlled trials, patients receiving sham treatment improved just as much as patients treated with conventional DBS (17).

Instead of trying to identify one region or one type of electrical signature for depressive symptoms that works for every patient, a research group at the University of California, San Francisco is testing out an even more personalized approach (18).

Their first patient was a 36-year-old woman with severe childhood-onset depression. In order to identify her individual neural signature of depressive symptoms, the researchers implanted ten electrodes in her brain in various depression-associated brain regions. For ten days, they continually recorded brain activity in these regions while the patient reported on the severity of her symptoms. In this patient, a particular type of activity in her amygdala, a region involved in emotional information processing, was associated with more severe symptoms.

Having identified a neural signature for her depression, they next needed to determine a therapeutic plan of action. Stimulation of the ventral capsule/ventral striatum on the right side of her brain improved her symptoms in most trials, the team noted. So, they implanted a device that would automatically stimulate this brain region whenever it detected the signature of severe depressive symptoms from the other electrode implanted in the amygdala.

For this patient, the results were stunning. After a single day of treatment, the severity of her symptoms had plummeted. More than a year after the device was implanted, this improvement was still going strong.

It’s still early days for this type of treatment. Just because it works for one patient, doesn’t necessarily mean it will work for many others. However, based on these promising preliminary results, the research team has initiated a twelve-person trial to determine if this same approach could provide similarly life-changing results for other patients with treatment-resistant depression.

Fasano noted, however, that as DBS technologies become more advanced and more effective, they generally also become more expensive. Even DBS in its current form, he said, is inaccessible to many people, and the problem may only get worse. “The inequality of adaptive stimulation will become an issue at some point,” he said.

While there are still many ways that adaptive DBS needs to be optimized and made more accessible, this personalized, on-demand treatment may one day provide life-changing relief for patients with Parkinson’s disease, binge eating disorder, depression, and potentially many other neuropsychiatric disorders.

References

1. Louis, E. D. & Machado, D. G. Tremor-Related Quality of Life: A Comparison of Essential Tremor vs. Parkinson’s Disease Patients. Parkinsonism Relat Disord  21, 729–735 (2015).
2. Deep Brain Stimulation (DBS): What It Is, Purpose & Procedure. Cleveland Clinic at <https://my.clevelandclinic.org/health/treatments/21088-deep-brain-stimulation>
3. Hariz, M., Lees, A. J., Blomstedt, Y. & Blomstedt, P. Serendipity and Observations in Functional Neurosurgery: From James Parkinson’s Stroke to Hamani’s & Lozano’s Flashbacks. Stereotact Funct Neurosurg  100, 201–209 (2022).
4. Gardner, J. A history of deep brain stimulation: Technological innovation and the role of clinical assessment tools. Soc Stud Sci  43, 707–728 (2013).
5. Little, S. & Brown, P. Brain Stimulation in Neurology and Psychiatry. Ann N Y Acad Sci  1265, 9–24 (2012).
6. Little, S. et al. Bilateral adaptive deep brain stimulation is effective in Parkinson’s disease. J Neurol Neurosurg Psychiatry  87, 717–721 (2016).
7. Little, S. et al. Adaptive deep brain stimulation in advanced Parkinson disease. Annals of Neurology  74, 449–457 (2013).
8. MedtronicNeuro. Adaptive DBS Algorithm for Personalized Therapy in Parkinson’s Disease. (clinicaltrials.gov, 2023). at <https://clinicaltrials.gov/ct2/show/NCT04547712>
9. Fasano, A. Adaptive/Closed Loop vs. Continuous/Open Loop Deep Brain Stimulation of Subthalamic Nucleus: a Two-Phase, Cross-Over, Double-Blind Trial in Patients With Parkinson’s Disease. (clinicaltrials.gov, 2022). at <https://clinicaltrials.gov/ct2/show/NCT05402163>
10. Wiest, C. et al. Finely-tuned gamma oscillations: Spectral characteristics and links to dyskinesia. Experimental Neurology  351, 113999 (2022).
11. Singh, A., Richardson, S. P., Narayanan, N. & Cavanagh, J. F. Mid-frontal theta activity is diminished during cognitive control in Parkinson’s disease. Neuropsychologia  117, 113–122 (2018).
12. Kessler, R. C. et al. The prevalence and correlates of binge eating disorder in the WHO World Mental Health Surveys. Biol Psychiatry  73, 904–914 (2013).
13. Longitudinal association between binge eating and metabolic syndrome in adults: Findings from the ELSA?Brasil cohort - Solmi - 2021 - Acta Psychiatrica Scandinavica - Wiley Online Library. at <https://onlinelibrary.wiley.com/doi/full/10.1111/acps.13356>
14. Wu, H. et al. Closing the loop on impulsivity via nucleus accumbens delta-band activity in mice and man. Proc Natl Acad Sci U S A  115, 192–197 (2018).
15. Shivacharan, R. S. et al. Pilot study of responsive nucleus accumbens deep brain stimulation for loss-of-control eating. Nat Med  28, 1791–1796 (2022).
16. Pandya, M., Altinay, M., Malone, D. A. & Anand, A. Where in the Brain Is Depression? Curr Psychiatry Rep  14, 634–642 (2012).
17. Holtzheimer, P. E. et al. Subcallosal cingulate deep brain stimulation for treatment-resistant depression: a multisite, randomised, sham-controlled trial. The Lancet Psychiatry  4, 839–849 (2017).
18. Scangos, K. W. et al. Closed-loop neuromodulation in an individual with treatment-resistant depression. Nat Med  27, 1696–1700 (2021).