Deep Brain Stimulation: A Neuroscientist on New, Non-Invasive Tech for Memory Disorders

The concept of sending electrical waves through the brain sounds like it’s straight out of a science fiction novel. However, many scientists are investigating this type of technology as a means to treat neurodegenerative diseases like Alzheimer’s. Deep brain stimulation, a process already in use for diseases like Parkinson’s’, often requires implanting a device into the brain to help address symptoms. However, some researchers are investigating ways to use deep brain stimulation without the need for invasive surgery. 

Award-winning dementia researcher and educator Dr. Nir Grossman of the UK Dementia Research Institute is developing a new, non-invasive form of deep brain stimulation called temporal interference. The technology is now in clinical trials, and he received a prestigious prize from the Science Journal and the American Association for the Advancement of Science (AAAS) for developing the technology. His research has been published in scientific journals like Cell, Science, and Nature Neuroscience.  

To shed light on his research and what it could mean for the future, Grossman joins Being Patient founder Deborah Kan to discuss deep brain stimulation as an emerging approach to treating neurodegenerative diseases like Alzheimer’s. 

Read or watch the conversation below:

Being Patient: When you think of brain stimulation, it does sound like something out of a science fiction novel. What does it mean to stimulate our brains?

Nir Grossman: That’s a great question. I think the way to think about it is to understand that behind our behavior and function, there are a lot of activities there. It’s like if you open a computer, there’s a lot of movement and signaling and moving around to facilitate the computation or operation. That [is] exactly the same thing in our brain, billions of cells that communicate with each other, sending signals to one [another.]

“When we talk about brain stimulation or neuromodulation, that means that we aim to modulate or change the activity in the brain.”

When we talk about brain stimulation or neuromodulation, that means that we aim to modulate or change the activity in the brain. In terms of therapy, we want to correct abnormal activity that often underpins the symptoms, as well as the pathophysiology of the disease.

Being Patient: Tell me a little bit about neuroplasticity and how this may relate. This is a basic explanation, but we’re changing the patterns in which the neurons travel in our brain if we learn something new, like a musical instrument or something, and we know that’s good for our brains. Is this a way kind of to adapt that type of plasticity in our brain, but instead use it for a treatment?

Grossman: Excellent question. Neuroplasticity lets us conceptualize that. If you think about movement, let’s say of cars in the street. You can monitor where there are a lot of cars, where there are traffic jams. You want to increase the size of the roads to facilitate more traffic in the next days, weeks, years, or more in the future. That’s essentially what neuroplasticity means. 

It means that the brain or the neurons in the brain sense where there’s a lot of communication, that seems to be activity, that is signaling that something very important is happening there. That will further amplify or increase the bandwidth if you think of this activity. Now, that assumes that activity is good, but this can also go exactly in the wrong way. 

Plasticity can be used to mark adaptation of the way the brain communicates, so that often leads to persistent change and leads to some symptoms. The symptoms can manifest as a tremor, as psychiatric disorders, as well as cognitive or memory disorders. What we were trying to do here is to correct, first of all, the activity itself. By repeating this corrected activity, we can tap into this plasticity property of the brain to cement this to improve activity, overwriting the abnormal activity or the pathological activity. 

Being Patient: Tell me about the difference between invasive and non-invasive deep brain stimulation. Invasive is obviously implanting something into the brain, so is non-invasive, like wearing a helmet or something to modulate brain stimulation from the outside?

Grossman: First of all, we need to understand that when we talk about the brain signals and communication, the signal and communication is electrical or electrochemical. That means, because of that, [it] is sensitive to external electric fields. So, if we can introduce electric fields, particularly [in] the target region, we can change the activity. With that, also in the correct model, adapted activity, or induce more physiological activity. 

Now, the brain is a three-dimensional structure. We have the cortex and the sulcus, and we have a lot of deep structures. Each one of them has a unique function in terms of cognition, memory, in terms of movement, and so on. If we want to stimulate that with electric fields, we need to apply [an] electric field. 

The electric field amplitude decays with the distance from the source, i.e., the electrons. So, if we would like to target a superficial structure, we can put electrodes on the scalp on the skin and apply an active field, and that will do the job. It will stimulate however, the neurons that are close to it, which will be on the sulcus at the deepest structure, will not be affected because the one on the sulcus will be affected first. 

[First,] there is a lot of benefit of targeting deep structures, because a lot of the more complex computation or the [language] system of the information in the brain sits deep inside. Until now, the way to do that [is] we had to insert electrodes into the brain, literally drilling a hole in the skull and putting electrodes inside. It sounds a little bit science fiction, but it’s actually an FDA-approved therapy for several disorders, including severe motor disorders and severe psychiatric disorders. 

The challenge is because it involves brain surgery, it can only be relevant for those patients [where] nothing [else has helped.] So, the scenario, if we can target those deep structures, is that if we change the activity there, we know it will have benefits, but without the need for surgery, that will have enormous benefits. 

“It sounds a little bit science fiction, but it’s actually an FDA-approved therapy for several disorders, including severe motor disorders and severe psychiatric disorders.”

This is what we develop. We developed a technology that allows us to do exactly that, to target deep structures without the need to surgically insert electrodes. This new technology is called temporal interference brain stimulation. That’s essentially what we are using now to test its therapeutic benefits for people without some disease.

Being Patient: Tell us a little bit about the non-invasive deep brain stimulation that you’re testing. What does that look like?

Grossman: Let’s first understand that will help us to get the concept of where brain stimulation fits in terms of the therapy in comparison to other more traditional therapeutic approaches. One of the main symptoms of dementia is memory loss, and 70 percent of the causes of dementia are due to Alzheimer’s disease. Hence, many of us [are on] the road investigating genetics or the molecular underpinnings of Alzheimer’s disease as the cause for dementia and how these changes affect the health of individual neuron cells.

Neuron cells are the building blocks of our brain and computation. However, a single neuron cell cannot carry complex cognitive operations in isolation. Hence, it makes the path from changes at the molecular level to actual benefits to the patient very elusive or abstract. What we understand today is that the transition from Alzheimer’s disease to dementia goes through changes at the neural network level. 

It is this change in communication between thousands and millions of cells that are responsible for the symptoms of memory loss that we see in dementia. We also know that the change of the neural network, [that the] growth of thousands and millions of cells affect the level of activity of individual cells and has many genetic processes that depend on activity. 

Typically, we’ll try to intervene by developing drugs that affect the genetic and molecular underpinning of the disease; we use brain stimulation, or [as you call] neuromodulation, to intervene with abnormal neural network activity. When we think about brain stimulation, as you mentioned, there are two types of brain stimulation. 

One of them is deep brain stimulation. We insert electrodes deep into the brain to tag a small structure in a very focused way. It’s as you say; it involves brain surgery. This is a therapy that has helped thousands of people around the world, but because of the need for surgery, they are limited. On the other hand, we have a non-invasive brain stimulation that targets the superficial structure of the [brain] in a very dispersed way. 

Until now, it was impossible to target deep structures in a non-invasive way. This is exactly what we did. We address these long-standing technological gaps by developing a new type of brain stimulation called temporal interference brain stimulation. In temporal interference brain stimulation, we apply two currents to the brain or showing via one card to a dependent using one pair of electrodes shown here in blue, at a frequency that is very fast, so the neurons cannot follow it. 

Let’s say one kilohertz or one thousand hertz. At the same time, we apply another current to the brain via another pair of electrodes shown here, black, at the kilohertz frequency is slightly different from the first one. If we apply one thousand hertz with the other, it could be 1,010 hertz. These two currents create an electric field inside the brain, shown here in blue and black arrows, as the neurons inside the brain superimposed those electric fields. If we look at them, on those electric fields in the time domain, the alternating the combined field hasn’t shown here in red, as an amplitude, the change periodically at a different frequency at the peak frequency. 

“Until now, it was impossible to target deep structures in a non-invasive way.”

What we discovered is that the neurons can respond to that, and because of this bit frequency, the overlap can be targeted to deep structural insight. In the past, we reported the technology, and we demonstrated that in animals [that] we can target a deep structure called the hippocampus [that] is critical for memory. When we measure that by using immunohistochemistry, it’s lighting up everything that we stimulate, and you can see we stimulated the deep structure. This is the hippocampus, essentially, without the overlying layer. 

Being Patient: I just want to clarify something that I’m not completely clear on. We know that the presumed pathology of the disease is plaque, beta-amyloid plaque, and tau tangles inflammation. When we’re talking about this in the context of Alzheimer’s dementia, at what stage would neuromodulation be applied? Is this a strategy for prevention of the disease or to slow down the disease? When would someone use this treatment?

Grossman: It’s an excellent question. The way to think about it is to understand the interactions that exist between the neural network activity, the symptoms, and the pathology. These are all interconnected. We know there is an accumulation of toxin aggregates, such as amyloid beta plaques and taus, and they are affecting the health of the cells, but they themselves do not cause the memory loss. 

The amount of aggregates that we have there and the neuroinflammation processes—they are the drivers of the pathology that is sensitive to the activity of the cells. So, if we can change the activity of the cells by brain stimulation, we can potentially slow down or even prevent or reverse those aggregates that risk the health and causing the neurodegeneration. 

Being Patient: Could a healthy person with a high risk for Alzheimer’s, like a homozygous APOE4, use this treatment to prevent the disease from progressing through plaque formation? Or what could you possibly use stimulation to, if you already have plaque in your brain, to break down the plaques and slow down the course of the disease?

Grossman: This needs to be tested and validated. It’s only speculation at the moment. What we know is that the next level of activity sits in between. Imagine [somebody who] has plaques developed, but it hasn’t affected all its community performance, which we know that’s the case because patients can live with a brain full with plaques without a single symptom. 

It starts to affect everyday life when there is a disruption. So, it could be that we just prevent the development of symptoms associated with mild cognitive impairment and dementia for those that develop the plaques, and that means this mechanism of action. 

Being Patient: Your study is at a very early stage, at stage one, right? Are you trying to figure out if the device works and what does it look like? How do you stimulate noninvasively with your specific study?

Grossman: Right, so those are two different questions. I will just briefly show that we target the human hippocampus using a memory test; I just skimmed through that. Importantly, what we showed was that we can noninvasively reduce the effort of the hippocampus, which is a deep structure that is critical for memory and learning. That is the starting point, or location of Alzheimer’s disease, that we can reduce the overall activity in the hippocampus and the efforts required to perform memory operation. 

We can also improve memory when we did it with [our human subjects.] This is an example of how the hippocampus looks like in Alzheimer’s disease, which eventually disappears. With that in mind, we are now finished this year. We wanted to test the safety with the first time to go from a single dose, i.e., a single session of stimulation to 10 days of stimulation. 

We wanted to establish that [it] is safe. The reason we need to go for such a low for the longest immersion session is partially what you were saying about neuroplasticity because we need to repeat that stimulation or augmented real patterns of activity in the brain so the neural network can do that without us stimulating. 

We are starting the next study, the second study, which is also phase one in Alzheimer’s disease patients this year, with the aim to optimize the way we do the therapy and look for cause-dependent memory improvement in those patients. This is where we are. So, the fact that we show that there is memory improvement that means there is early-stage evidence that we can affect the symptoms, the key symptoms of the disease. 

The fact that I showed you that we reduced hippocampal activity alludes to the fact that we may be able to affect the underlying pathology, and indeed, in power to that I’m not showing it here, we do clinical studies in animal models of the disease, where we show or what we see is that actually there is reduction of the plaques when we do the stimulation. Essentially, we can affect those drivers of the disease. 

Being Patient:  This is still really early days, but in the context of what you’re talking about, why not just use this as a health tool? Could this be something that becomes part of a future health routine to get your brain stimulated in order to prevent things from going wrong down the road? Talk to me a little bit about the prevention aspect of this.

Grossman: When you ask why not, I mean, I’m here to prove why yes. That’s hard, and to prove why, yes, in terms of prevention is even harder, just in terms of the magnitude and length of the study. This is what we do. We do a proper, rigorous scientific investigation to understand both the therapeutic mechanism of action and to validate that it can actually benefit patients, and also, via this process, optimize the way we do that and understand which patient subgroup it can help them and if not. 

That by itself is an enormous undertaking. Of course, once we do that, that could create rationale for healthy subjects; elderly people that have a risk of developing mild cognitive impairments and Alzheimer [could have rationale] to use that. But that needs to be proven, or people take it on their own. 

“When you ask why not, I mean, I’m here to prove why yes.”

I think in terms of safety, the safety profile of that would potentially enable us to envision that the therapy will be done by the patients themselves at home. And that’s, in fact, one of the important insights that we got from a focus group that we have that includes patients with Alzheimer’s disease, carers, and healthcare providers. They clearly say to us that the only way these things will be adopted is when it will be operationalized at home, and this is what we’re here to do.

Being Patient: Absolutely. What are some of the risks? I know it’s still early days here, but are there any risks in applying this type of therapy?

Grossman: Any therapy, any intervention has a level of risk. Until now, we’ve deployed this procedure to hundreds of people and now dozens of patients also with Alzheimer’s disease. It’s all well tolerated. Most of the time, [people] do not feel anything. Because sensation that exists during the stimulation, or during transcranial or non-invasive electrical stimulation, is when one stimulates the skin on the way to the brain, and then one can think tingling. 

If the stimulation is very strong, one can even feel a burning sensation, although there’s no burning. Just because we stimulate the somatosensory neurons there, that does not exist so much in this stimulation because it occurs underneath the skin. If one causes a very, very strong stimulation on the brain because we are synchronizing brain activity, there is any pathetical risk of seizure induction, but this requires orders of magnitude a hundred times to one thousand times stronger stimulation than what we’re doing. 

Being Patient: We’re getting some questions coming from our audience that I wanted to ask you. Do you envision the external stimulation as being maintenance in that a patient might need to continuously use it in an ongoing fashion? And how long is it, like ten days per year, or would be used only when symptoms develop? Another part of this question is separate—do you see artificial intelligence perhaps as a tool to adjust and understand how much you will need in terms of this type of intervention?

Grossman: These are excellent, excellent questions that we asked ourselves. I don’t have the answer, but they are exactly the questions that we ask. We can help the audience to conceptualize what would give an example for existing therapies. There is an existing FDA-approved therapy for depression that is based on the electromagnetic stimulation of the cortex. It’s transcranial magnetic stimulation, FDA-approved. 

The therapy involves six weeks, five days a week, one hour a day stimulation, [then] by the end of that most of the patients benefit from them. Then they need, after a couple of years, some intense stimulation. That’s one example.

“I hope that by leveraging on neuroplasticity, we can correct the activity in the brain and, with that, create sustained benefit to the patient.”

Here’s the second example of those patients with severe motor disorders who receive this deep brain stimulation that we discussed before. There if the moment is stopped, the simulation, after a second, the symptoms return. So, you have to stimulate [it] all the time. I hope that by leveraging on neuroplasticity, we can correct the activity in the brain and, with that, create sustained benefit to the patient.

Being Patient: Someone else is asking if you have done any work with people with a history of traumatic brain injury (TBI) or people with both a history of TBI and APOE4 status?

Grossman: No, we have not. I want to emphasize to the audience that we are very susceptible to the unbelievable suffering and needs that patients have, especially because we engage with patients and their families all the time. We’ve been trying to develop a therapy for a disease that, for the last decades, we failed as a society in the sciences; it is very hard. Yes, there is new improvement now coming out with FDA-approved drugs.

I think we’re seeing more and more of them. It could be most likely they will be working in a synergetic way. Even if there is a drug that clears amyloid, the beta plaques in the brain, how that is actually converted into improved memory is not clear, but this is exactly what neuromodulation can fit in because we are correcting the circuit that’s related to the function. 

So, that’s one thing. The second thing to say is that because we are trying to prove scientifically that there is a benefit here. We need, somehow, with limited resources by definition. It’s not just money; it’s also time. We need to be able to select a subset of patients and, first of all, demonstrate that and that’s the reason why we don’t test many, many things. 

Being Patient: We got some questions from the audience about if there are any clinical trials for this in the US since you’re based in the UK. I know there are other studies on brain stimulation going on. I also want to ask, you are in phase one now, but how many years away do you think you’ll be from this being a treatment?

Grossman: I will say two things. First of all, in the technology. By the way, there are many people who do neuromodulation, but there are much less ones that can target deeper [areas noninvasively.] The technology, the temporal interference stimulation technology, was developed when I was at MIT and Harvard in Boston. There are groups around the world, including the US, that are using that. We together, in groups as well in Boston, spearhead the development toward therapy for Alzheimer’s disease. We support other groups around the world to develop therapy for other brain disorders.

What I want to say is that there are mechanisms for patients in the US to engage in during the clinical development stage and not waiting for the approval. Under clinical studies, of course, we have the question of privacy. We also very much encourage input from patients. We always benefit enormously from discussions with patients to understand their expectations, their concerns, and so on. I’m encouraging patients and carers to contact me, and I’ll be very happy to facilitate discussions and so on. 

In terms of the timeline, I know what are the next steps that we need to do. The timeline of them, of course, there is always a huge uncertainty, what will be the outcome of the study?  The next immediate step for us is to conduct the study that will run the next two years, where we will want to demonstrate and test those dependent benefits that will give us that will teach us a lot about therapy. 

“We always benefit enormously from discussions with patients to understand their expectations, their concerns, and so on.”

We are also going to measure the brain activity during the stimulation and use machine learning AI, as you mentioned, to identify those markers or biomarkers in the brain activities that will allow us to predict in the future, which patient would benefit from that. 

That is [what we need to do] in the next two years. In parallel to that we need to develop the capability to do everything not in my lab or the clinic, but to do it at home. At that point, if that’s successful, we’ll be ready to deploy pivotal studies that will test the benefit. We go into phase two and immediately to phase three essentially.

Being Patient: We got a question about this from the audience. Is this different from light therapy?

Grossman: Let’s think about it. Light therapy is not the direct stimulation of the neurons in the brain. It’s essentially giving a sensory input. They combine it with light and audio. It’s a sensory input to the brain that has some periodicity in it. Then, it affects those regions in the brain, the design, and years for this sensory input. The ability to affect other regions in the brain is very indirect. 

So, if we want to affect, for example, the hippocampus, where everything starts in Alzheimer’s, there’s no direct. The hippocampus is not a sensory organ, [it is not a] site for light or for audio input. There could be some indirect signaling that eventually will go to the hippocampus there, but they are very much muted. 

In fact, our brain is very good, as we all know, to attenuate any constant input so we can look for the changes. There’s no doubt there’s benefit in terms of sensory input for elderly people to prevent neurodegeneration. Any sensory, as well as social input and interactions. It could be that a particular frequency or periodicity of the sensory input is also very beneficial. That’s exactly [what the] therapy [can do.] What we do is different in the sense that we go straight to the region in the brain that is affected by the disease and change the activity directly, and not via sensory input.

Katy Koop is a writer and theater artist based in Raleigh, NC.