Psychiatry 3.0

From Our Neurons to Yours Wu Tsai Neuro Podcast

In this episode, psychiatrist Nolan Williams discusses transcranial magnetic stimulation for major depression and its implications for the future of psychiatry.

Listen to the full episode below, or SUBSCRIBE on Apple Podcasts, Spotify, Google Podcasts, Amazon Music or Stitcher. (More options)

 

View all episodes

 

Transcranial magnetic stimulation (TMS) is a technology that uses magnetic fields to stimulate or suppress electrical activity in brain circuits. It's part of a transformation in how psychiatrists are thinking about mental health disorders that today's guest calls psychiatry 3.0.

Nolan Williams has recently pioneered a new form of TMS therapy that has just been approved by the FDA to treat patients with treatment-resistant depression. That actually describes a lot of people with serious depression, somewhere between a third to a half. At some point talk therapy doesn't work, drugs don't work, and for most people, there's not much else to try.

TMS has been used for depression before, but Nolan's team has taken a new, more targeted approach. It's called SAINT, which stands for Stanford Accelerated Intelligent Neuromodulation Therapy. Basically, it uses MRI brain imaging to precisely target intensive TMS stimulation to tweak the function of specific circuits in each patient's brain.

Remarkably, after just one week in Nolan's SAINT trial, 80% of patients went into full remission. The stories these patients tell about the impact this has had on their lives are incredible.

We talked to Nolan, who is a faculty director of the Koret Human Neurosciences Community Laboratory at Wu Tsai Neuro, about what makes this approach unique and what it means for the future of psychiatry.

Additional Reading

 

Episode Credits

This episode was produced by Webby award-winning producer Michael Osborne, with production assistance by Morgan Honaker, and hosted by Nicholas Weiler. Art by Aimee Garza.

 

Episode Transcript

Nicholas Weiler:

This is From Our Neurons to Yours, a podcast from the Wu Tsai Neurosciences Institute at Stanford University. On this show, we crisscross scientific disciplines to bring you to the frontiers of brain science. I'm your host, Nicholas Weiler. Here's the sound we created to introduce today's episode.

Perhaps that is the sound of transcranial magnetic stimulation. If you've never heard of transcranial magnetic stimulation or TMS, it's a way to use magnetic fields to stimulate or suppress electrical activity in brain circuits. It's part of a transformation in how psychiatrists are thinking about mental health disorders that today's guest calls psychiatry 3.0. Nolan Williams is a Stanford psychiatry professor, and he's been in the news recently for pioneering a new form of TMS therapy that has just been approved by the FDA to treat patients with treatment-resistant depression. To be clear, that actually describes a lot of people with serious depression, somewhere between a third to a half. At some point talk therapy doesn't work, drugs don't work, and for most people, there's not much else to try. Now, TMS has been used for depression before, but Nolan's team has taken a new, more targeted approach.

It's called SAINT, which stands for Stanford Accelerated Intelligent Neuromodulation Therapy. Basically, it uses MRI brain imaging to precisely target intensive TMS stimulation to tweak the function of specific circuits in each patient's brain. Remarkably, after just one week in Nolan's SAINT trial, 80% of patients went into full remission. The stories these patients tell about the impact this has had on their lives are incredible. So, when I talk to Nolan about what makes this approach unique and what it means for the future of psychiatry, he told me that part of his inspiration for building new and better tools for treating brain disorders has been the sense of jealousy he sometimes feels when he thinks of all the tools we have for treating other critical organs like the heart.

Nolan Williams:

So, that was really the impetus for doing a lot of the work that we're doing is we don't have the sort of suite of treatments that cardiology has. You can have a lot of jealousy with cardiologists. If you have a cardiac problem, man, they've got multiple devices and drugs they can try that will work really rapidly and tell you an answer within a couple of days, and that's really the playbook that I'm playing off of. I'm just taking my cardiology jealousy and trying to make treatments for psychiatry and neuroscience generally.

Nicholas Weiler:

And I'm sure I'll hear from some cardiologists about this, but they have the advantage of understanding how the heart works. I've heard you talk about what seems like a shift in psychiatry that you're really at the forefront of which I think you called psychiatry 3.0. I wonder if you could tell me a little bit about what you mean by that.

Nolan Williams:

Psychiatry is a very unique specialty in many ways, and really what you see is you see this kind of moving target of definitions of what psychiatric, and then you've got these epochs of understandings of treatments and what is a treatment for a psychiatric illness. The first epoch is really around the kind of Freud period where Freud looked at psychiatric illness and said, "Okay, we can do something with this by conducting a series of psychotherapy sessions with the patient," and psychotherapy's very effective for some conditions, simple phobia, some less treatment-resistant forms of PTSD, cognitive behavioral therapy for mild forms of depression. So, therapy has really played out over the years to be an effective treatment, but it's not complete. Psychotherapy for severe depression where somebody's in the hospital or high levels of suicidal ideation doesn't really work in most cases, and so then that prompted what we call psychiatry 2.0 with the advent of Thorazine and the idea that you could give a medication for what ends up being a problem, as they perceived it, of thought content.

Nicholas Weiler:

So, now it's a biological, chemical problem?

Nolan Williams:

Yeah, this kind of concept of the chemical imbalance. He was born out of this sort of way of thinking and really psychiatry 2.0 didn't completely reject psychotherapy, but rejected its set of assumptions that you had schizophrenia because your mother did certain behaviors it caused you to have. There was a rejection from that to say, "Okay, if we can give Thorazine, it's probably not the schizophrenic mother. It's probably something that's biological," but the problem with drugs is they too don't work in completion and especially the ones that we think about. 29% of people after all the steps really kind of get there as far as remission goes. You get walked through a whole host of different treatments, psychotherapy is in there, and if you're talking about success as remission or absence of symptoms, you don't get that the majority of people. So then psychiatry gets pushed again, and the beginnings of psychiatry 3.0 is rooted in the beginnings of circuitry, understandings even for neurological illness, which is this idea of focal brain stimulation.

If you can stimulate one brain region and that brain stimulation distributes into a greater neural network and this is a therapeutic effect, then it's likely that it's not just stimulating in that one brain region, it's really treating what we think about as a circuitopathy, this idea that there's an kind of a disease circuit from the standpoint of its electrical properties. That idea was born in the '80s in a small way and has expanded over time with really, I think the crescendo being kind of nowish in this idea that we're entering in, and it's pretty clear, everybody kind of buys into this, into a third era of psychiatry, what I and others have called Psychiatry 3.0, which is this idea that we're really focused in on the circuit, and what's useful about that kind of era is that it gets us back with our neurology brethren and gets us on the same language, and that's important.

If you have a shared language across all of the brain specialties, the neurosurgeons, the neurologists, the psychiatrists, the neuroradiologists, you can at least start talking about the problems in the same way. 30 years ago, you put a psychiatrist and a neurologist into the room and you've got one person speaking Mandarin and one person speaking English, and then of course they were like, "What is going on with the other?" But now we're on this page of psychiatry 3.0 are thinking about things at the level of the brain circuit, and what's interesting is as that has evolved, the psychedelic scene has come back online, but under this frame of treating and measuring circuit dynamics and measuring circuit changes that are related to the mental illness. And so they too, interestingly, just to give kind of the listeners an understanding, this isn't about brain stimulation, it's about thinking about it as far as engaging circuits.

They too are on that same page of we're engaging a brain circuit. So whether it be this idea of psychiatry moving forward in the psychedelic space, psychiatry moving forward in the neuromodulation space, which are kind of the two views that people take these days about what's the next steps or even like I said, neurology and neurosurgery is really on that same frame, what we're tossing out is the chemical imbalance idea. What we're tossing out is the schizophrenic mother idea, and we're really getting on the same page that whether it be Tourette syndrome or Parkinson's or dystonia or OCD or depression, it's all circuit disorders that need some sort of circuit intervention.

Nicholas Weiler:

And as we now come to understand the brain's circuitry a little bit better, that certain circuits are involved in emotional states or mood or in regulation or taking how you feel and giving it the appropriate context, we have the ability to start saying, "Okay, well are there people where there are particular circuits that aren't doing their job? And are there ways we can start tweaking those or modifying them or asking them politely to start working in the way that would be more beneficial to the person?" Is that the big picture here? It's we're starting to understand the brain circuits, and so now you have a little idea of what the brain circuits are doing, you can start developing treatments that specifically target those circuits.

Nolan Williams:

That's really the goal is once you have a sense of the neuroanatomy, you have a sense of the brain circuitry, then you can kind of engage that system with what we can validate in various ways as a biologically relevant signal, and it's kind of like the early days of Morse code. One person would be in one room, whatever, a couple streets down, you've got a wire running all the way to the second person, and you can click away at a number of different sequences of taps and lines and get information from one individual to the second, and I think that's the same idea with neurostimulation. That's really what we're doing early Morse code, but the receiver is the brain, not the person per se. The person has no conscious understanding of what the information is, but their brain does. The goal of SAINT was really to figure out where in each person's brain to stimulate, which is like an old deep brain stimulation mantra, everybody that gets deep brain stimulation for Parkinson's, they're getting an individualized mapping of their subthalamic nucleus or their globus pallidus interna or whatever target they're going after.

Nicholas Weiler:

It's not just knowing which circuit it is, it's knowing how it's affected in that person.

Nolan Williams:

It's about knowing the sub circuits. So, you can't just rely on in anatomical landmarks, you've got to actually, to your point, kind of perturb and measure the system, and that's really what we're emulating with SAINT.

Nicholas Weiler:

Before we jump into SAINT, I'd love to take just a quick step back and make sure people understand what brain stimulation is. So, if you recognize psychiatric disorders as circuit problems, the next step of course is you need to have circuit solutions. How do we rewire these circuits? How do we send those signals to those circuits to work better? And so, your work is using this technology, transcranial magnetic stimulation, or TMS, to treat depression. Could you describe, just very briefly at a high level, the setup of TMS? What does the device look like and how do we think it's stimulating the brain?

Nolan Williams:

So, the idea of neuromodulation is to cause action potentials. That's really kind of the goal, if that makes sense, and so the reason why it's a magnet, not an electrical pad is because in order for you to get enough power into the system to depolarize cortical neurons, you need to be able to have a power intensity that wouldn't burn the person. In the case of direct electrical stimulation, like these devices out there, they don't have the power to do that, and you don't want them to because if you turned it up enough, it would burn the skin in many cases. And so in the '80s, what we figured out is that what we really need to do is we need to bypass the skin and to be able to stimulate with a magnet using Faraday's law. And so, the idea of TMS is that if you pulse a high strength magnet, you can generate current and electrically conducting substances for that period of time that you pulsed it.

And so it's not forever, but what it does is it sends a single signal. So if I did that over your motor cortex, Nick, what I would see is I'd see a motor movement. So I'd do a single pulse of TMS, and if I did it in the exact right spot in your homunculus, on your motor strip, you get a thumb twitch. Now, your brain is unchanged. That's a probe, and we figured that out in the mid-80s that you can send a single signal. So it's like in that telegraph, it's like sending a signal to the recipient, ping, and they get it.

Nicholas Weiler:

Right.

Nolan Williams:

They know that the telegraph works, but you didn't give them any information.

Nicholas Weiler:

You've got a connection.

Nolan Williams:

You got a connection, and as long as the person's motor cortex is connected all the way through their cortical spinal tract, it gets down into the nerves in the arm, it gets to the muscle, and all that's intact, you get a motor movement, you have a break in any of that, you don't. I had a stroke and I have severing in my cortical spinal tract. It won't work if you have a spinal cord lesion, it won't work, on and on. But the second thing that they figured out was if you send a series of signals in the brain, what we call repetitive transcranial magnetic stimulation, it's still a probe in the sense that you still are changing everything all the way down, but it also is a system that can actually change the network that you are sending a signal to.

And so in the case of motor cortex, and this was the big TMS invention in the mid-90s, it was sending a very inefficient signal, the first version of repetitive TMS into the motor cortex, and then showing based off of using EMG, muscle measurements from a sensor over the thumb muscle that you can get a change in the cortical excitability. So, if I put an excitatory parameter set into your motor cortex for about an hour, the amount of power it takes for me to move your thumb goes down, you need less.

Nicholas Weiler:

So, you're starting to modify the circuit by repeatedly stimulating it.

Nolan Williams:

And if you send an inhibitory or deep potentiating signal onto that system, which you can also do, then you need more power at the end of that to move the thumb.

Nicholas Weiler:

So, you can tune the circuit either direction, make it stronger-

Nolan Williams:

Either direction, and the sanity check. There's a third set of parameters and probably a much larger, although this hasn't been fully explored, that is inert, that just doesn't do anything, that the system is the same measurement before and after, and that's really what you want because in this case, our Morse code is really go, go, go or stop, stop, stop, or gibberish. In the case of depression, what we're trying to do is say, go, go, go, but in the case of OCD, we're trying to say stop, stop, stop, and so you want that control. Now, there's a world in the future where you're actually telling a story through Morse code where you're actually making sentences where it's not just stop, stop, stop, go, go, go.

Nicholas Weiler:

Right, you can start to actually talk to the brain in the future.

Nolan Williams:

But that's where we're at now, we're in this kind of go or stop information. And so, that was the kind of advent of the '90s. Now you zoom up to the 2000s and the invention in the 2000s was we can send a signal into the brain that mimics the hippocampus' own signaling within the brain, and we can make this way more efficient. So, a paper published in Neuron came out of John Rothwell's lab in 2005 demonstrating what they call theta-burst or this kind of theta-gamma coupled signal that the hippocampus uses that we can actually play back through a TMS coil, and when you do that, you're able to take something that normally takes about an hour to work and you can dial it down to three minutes.

Nicholas Weiler:

And that's because that's sort of the frequency of plasticity in a sense, right?

Nolan Williams:

Yeah, that's the learning signal. You're sending the information in that the brain's used to. You're not just sending in a 10 hertz signal that the brain-

Nicholas Weiler:

It's like a resonance frequency for the brain.

Nolan Williams:

Yeah, that's right, that's the idea.

Nicholas Weiler:

Interesting.

Nolan Williams:

So, now we kind of enter into the 2010s when we started working on things, and then the next jump we think from this story is to take what now is a learning signal and optimize it because think about these classic psychology experiments where I give you somebody's phone number and ask you to remember it. I can give you somebody's phone number once you can sit there and try to remember it, a few people can, most people can't, or whatever, and then you get into this question of, well, how do you actually really learn?

If you're kind of hijacking the learning system and you're exogenously learning through getting information played into the cortex and telling systems to now remember to stay on that in their native forms stay on, but we're sending a signal back in to tell it to stay on, then how do you optimize that signal? Well, what you need to do is you need to emulate what we know in psychology and actually also in basic science experiments, this idea of space learning theory. And so, what space learning theory tells you is the optimal exposure, re-exposure ends up being about 60 to 90 minutes.

Nicholas Weiler:

So, you need to know when you need to spin that wheel again to get the thing really moving. Again, it's like you've got your resonance frequency of stimulation, but then you also have when do you need to give it that new stimulation to repeat it and really lay in the information that you're trying to get stored in the circuit.

Nolan Williams:

If you take somebody and you give them a single note card and you have them look at it 10 times, then you have a hard time being able to remember that. So, we don't typically do that where we give people one note card and we have them look at that same note card over and over again, and then they set it down. Nobody does that when they study with note cards. What do they do? They write out about 60 to 90 note cards, and then they look at it for about a minute, and then they go to the next note card, and then about an hour to an hour and a half later, you get back to the first note card. If we all remember kind of cramming for tests, this is how we did it. There's no accident about why we all do it like that, and that's really the principle that SAINT is based off of. It's this idea that you need to have this optimal interaction between each stimulation to build the system up over time.

Nicholas Weiler:

So TMS, you've got these sort of magnetic coils that sit over a particular part of the head, these sort of two coils. They're like about the size of your hand, right, each one?

Nolan Williams:

Yeah.

Nicholas Weiler:

And so, we know that this can stimulate or inhibit brain circuits, and it can even cause changes in those brain circuits to make them stronger or weaker. If I understand correctly, that's been approved for treating depression since at least 2008, but your SAINT trial has really taken that technology to the next level, and I think you've gone from six to 12 weeks for this to be effective to just one week. So, what is the key thing that makes this more effective and rapid to help people who maybe are experiencing suicidal ideation or something you really need to help these people quickly?

Nolan Williams:

So, it's just that principle that I was talking about earlier, which is this idea that you really have to kind of lay down multiple optimally space stimulation approaches, and you have to do it in an individual level stimulating within a given brain network, that's the idea. You can't just stimulate once a day every day in an average skull spot and expect most people to get better, but if you can get into the right spot and you can stimulate with a right biologically relevant signal for the right dose, then you can get most people well, and that's really what we see as the three innovations. It's a reorganization in time, a reorganization in space, which is that person's unique connectivity pattern and dose. You need to give a sufficient dose.

Nicholas Weiler:

And you've shown that you can give much higher doses than what people have done before, and there are no harmful side effects as far as you can tell.

Nolan Williams:

No, that was the question is, can we get some sort of plateau of efficacy or hit some sort of toxicity point? And to no surprise to us, we didn't really see much toxicity because nobody's ever seen much toxicity with transcranial magnetic stimulation, but we did see this increasing efficacy over time with giving more and more dose. So, we're giving about seven and a half months worth of conventional TMS over five days.

Nicholas Weiler:

And you saw remarkable efficacy. Was it 80% of these people for whom no other treatments work, you were seeing remission after just a week?

Nolan Williams:

Yep, so in our blinded trial, it was 78.5, so around 80% of people went into remission at some point in the four-week follow up. In our open label trial, it was 90. There's some hit for placebo effect, but not much in the way. It's about 10% it looks like is driven by placebo effect of the overall effect, which is really dramatic. Most of the treatments out there, the delta between the active and the placebo is very small, and a lot of what the treatment effect is placebo. We picked really severe people to drive down the placebo effect, and then we had this very powerful treatment effect.

Nicholas Weiler:

It is amazing and wonderful. I'll include some of these stories in the show notes, and I hope listeners will go and read some of the personal stories that people have told about just how severe their depression was and how it feels to not have that experience anymore after, in some cases, decades, and it points to really the big picture mission of neuroscience. If we can just understand what these circuits are, how they learn, how they change with time, what are the mechanisms that the brain operates on, then maybe we actually have a chance of helping people with these conditions. And I wonder if you could speak to that. How do you see depression treatment or psychiatry looking in the future as we develop this psychiatry 3.0?

Nolan Williams:

I think this is the first FDA-cleared MRI for anything in psychiatry. So, it gives us this platform approach to be able to look at circuitopathies more generally. Depression is comorbid with everything, with PTSD, with OCD. So if you're treating somebody for depression and they have OCD, very likely, they also have differences from standard MDD as far as the circuit's affected.

Nicholas Weiler:

Right, depression is not the same from one person to another.

Nolan Williams:

And it's not the same from one person to another, so you've got this ability to really get a big database of depressed individuals and really get a handle on how to use MRI to understand this, and then I think there's going to be a whole host of innovations in the stimulation space, in the neuroimaging analysis space to get this to the next level. I think that the goal is over time, we're going to evolve better ways of doing this, but this is a platform to start with.

Nicholas Weiler:

And just as a point to close on, what are some of the things we should be thinking about as we learn to stimulate and modify brain circuits with greater precision?

Nolan Williams:

I think that you have to value these things first, and you have to show that there's a utility to doing it, and we can show that the pre... And we have data coming out in the proceedings of the National Academy of Science the next couple of weeks on this, but the pre-treatment MRI, we now have some predictive capabilities too, so we can tell who's going to be a really good responder or not before they ever get stimulated. And so, that tells us that if we can use that sort of technology in addition to what we think about as the surgical technology of where do you put this versus the who, who do you stimulate? If we can really marry those two and be able to actually do that on a scan prior to the person ever getting treated, and that vision is realized, then I think it'll change everything about psychiatry because it'll give us the ability to know what to do for the first time.

And getting back to why I'm cardiology jealous, it's because those guys, they get to see KG, they get a readout. They're able to know based off of that readout, "Okay, I'm going to use this drug, or I'm going to use this device," and all that stuff. We have a more complex system, that's why it's taken us longer. That's much more nuanced and seems to be different across different parts of the brain. If we can kind of realize what the cardiologists have over the last 100 years, this 100 years, then we're going to have a completely different set of specialties.

Nicholas Weiler:

Well, it'll be a great world where we can relieve some of the suffering that's caused by these psychiatric disorders, really offer hope to people for whom we haven't had much to offer. Well, thank you so much for the work that you're doing and for coming on the show to tell us about it.

Nolan Williams:

Thanks for having me and I'm looking forward to seeing it. Thank you.

Nicholas Weiler:

Thanks so much again to our guest, Nolan Williams. For more information about his work and the SAINT trial, check out the links in the show notes. There are some amazing stories from the patients who have been treated in this trial that really helps understand the transformation this represents. This episode was produced by Michael Osborne with production assistance by Morgan Honaker. I'm Nicholas Weiler, see you next time.