Well, you stuck it out to the last session. Thank you and congratulations on your stamina. You know there's a phrase in Hebrew, which means the last of the last is the most beloved. So thank you for being here. Dr. Nolan Williams is Associate Professor within the Department of Psychiatry and Behavioral Sciences and Director of the Stanford Brain Stimulation Lab. Dr. Williams has broad background in clinical neuroscience and is triple board certified in general neurology, general psychiatry, as well as behavioral neurology and neuropsychiatry. In addition, he has additional specific training in clinical expertise and development of brain stimulation methodologies, which you're going to hear about today. This is hardly a complete description of his bio, but it'll get us started. We are going to be live streamed as well. And at the end of the presentation, we will have a Q&A splitting between a live audience and the live stream questions. So Dr. Nolan. Thanks, Ron. Thanks. I'll ditto that sticking it out thing. I know this is the very end of the meeting, so I appreciate everybody staying around to hear this. So Ron got in touch a few months ago and asked to do kind of a full talk on what TMS is, asking us what kind of the newer rapid acting approaches are, and to try to kind of get a good kind of intro to that whole scene. So that's what this is about. So I'll get pretty deep into some of the technical issues, and then into the clinical data, and then into some of the new stuff we're doing. Here are my disclosures. I'm an advisor for various companies. So I'm going to give you guys a background on what TMS is, what ThetaBurst is, and then what our kind of new rapid acting stuff we've been working on is. So you all know all of the drug classes that we use in psychiatry are all discovered, like they're not really engineered for the conditions that we're using them for. So we've largely discovered these drug classes, and we've always been under the viewpoint that engineering a solution to that, to a given psychiatric problem, is important, and trying to really think about what the engineering principles are. TMS is one and kind of the first example of an engineered solution, right? So I'll describe a little bit about what that is generally, and then we'll get into kind of the therapeutic aspects of it. So TMS is a device that can produce treatments out of it. What I want you to think about when you think about a TMS device is not that it itself is a treatment. It is like a pharmaceutical company for you, right? So the reason for that is because there's a whole host of things, actually an infinite number of things that you can do with these devices, because you have an infinite parameter space in the sense of various frequencies, and we'll talk about all that, dose, position is finite, but various positions all over the head, and ultimately in the brain, and it was initially developed as a probe. So it was a tool for perturbing cortical neurons, right? The idea is, you know, first, can we push on the system and see that this entire track gets activated, that there's a depolarization in the cortical neurons, and ultimately in this case, for the first set of experiments, a thumb movement, right, because they were looking at motor cortex, and so the first use of this was really just to kind of perturb these areas and understand brain behavior relationships, right? And that really grew out of a whole literature of electrical stimulation where that was the same goal. The problem with electrical stimulation, and the reason why we're talking about magnets today is because the amount of electricity that you need to use, right, to depolarize cortical neurons, to actually have action potentials in the brain, is the same amount that you start to get into burning, right? So if you go to that high level of electricity, then you start to get into risk of causing folks to have, you know, burning in their skin, right? And so you, they learn pretty early on that you can't really go that route, or you're going to have kind of adverse events and all that sort of thing, and so the reason to use magnets is what we call Faraday's Law. Anybody familiar? Remember Physics 101, 102, Faraday's Law? Faraday's Law is simple, right? It's been around for hundreds of years. If you pulse a magnet, you induce current in electrically conducting substances. Skin is not an electrically conducting substance, neither is scalp, neither is skull, right? And so using a magnet allows you to bypass the risk areas as far as what we were seeing with direct electrical stimulation and really just get into the electrically conducting substances in our brain, which are neurons and nerves and the CSF around the brain, right? And so it gives us an opportunity to really, you know, in many ways, a really revolutionary opportunity to, you know, affect the brain in the same way the neurosurgeons could for now a hundred years with direct electrical stimulation during epilepsy surgery, but do it without a scalpel, without having to open up the head, without having an incision, right? We're able to interact with the brain, you know, non-invasively and selectively, right? Anybody, you know, most of you are probably familiar with the history of the telegraph, right? So the first experiment, right, is the experiment of I send a signal from where I'm at to somebody in the audience. Let's say somebody in the audience. I have a wire, you have a wire, I do a ping, you receive the ping. There's no inherent information in that ping, right? I'm not sending a message with that ping, but what I can do is I can get a readout that it happened. Does that make sense? So if I have a stroke in my corticospinal, you know, somewhere in the corticospinal tract and I'm stimulating in the hand representation, there is no ping. The signal is broken there and the information is lost, right? Does that make sense? You have to have an intact system to have an output to, in this case, a muscle. But if the system's intact, then you get that ping, just like the telegraph wire is intact if you receive the ping on the other end of the telegraph. And what we do first when we get a patient in for TMS, whatever type of TMS it is, is that we try to figure out what the minimal intensity is to get the information, to get the ping. And then we use that information to personalize the intensity of the stimulation. And really the idea there is that you're, you know, and it makes a lot of sense, right? We want, in this case, we want to prevent risk, but we want the most efficient amount of information, energy going to the brain without too much. And that's that kind of minimal amount that is able to produce a motor movement. And then you know really that's the cortical excitability threshold for essentially the whole brain with the caveat that if there are differences in depth, you may have to adjust for that. And so, great. I'm in California. I sent a message to New York City saying, you know, ping, and they got it. And then I did it again, and then they got it again. What would be the next step in the evolution of the telegram? We'd want to send a message, right? But we can't speak into, not yet, we can't speak into the wire, right? We can only send pings, maybe long pings, short pings. So that's Morse code, right? So it's this idea of being able to send a signal from one person to another. In this case, you as the operator of the device are sending a signal directly into the brain, and you're bypassing the sensory organs, right? So TMS is sending information to the brain that bypasses the sensory system. I mean, you hear clicking and stuff like that, but really, the actual information itself is bypassing the sensory system, the eyes, the ears, so on and so forth. And what you're getting is you're getting information that's like Morse code. And so, just like Morse code, you can send Morse code dots and dashes that are, that make sense, that have information in them. You know, if my three-year-old tried to play with a telegram, you know, she would ping at it and she'd send something, but it wouldn't be, you know, it wouldn't be translatable by Morse code standards on the other end, right? And that's another important principle. The principle is there are ways of stimulating in the brain that actually convey something, and then there are ways of stimulating in the brain that don't do anything. And that's really an important base idea, because if you don't, if you, if everything sent information into the brain, that would make no sense, right? Just like gibberish, I can't just stand up here and say gibberish, and then somehow you understand it, right? There are words, there's syntax, there's language, and it's the same thing with the brain. The brain needs information in a certain way in order to receive it and do anything with it, right? And so how do we know that that message is real, that that message is biologically relevant? And what's really useful, and this is, these methods have been around since the early 90s, this is not at all controversial to motor physiologists, right? Like this is known, known, known stuff. So if I stimulate a motor cortex, I'm going to, for each of you, at a given range of amplitudes, or at a set amplitude, I'm going to get, in the case of a range of amplitudes, a recruitment curve in the setting of a, you know, a, the same amplitude, a consistent kind of EMG readout, a muscle readout, right, where I can basically have this really reproducible readout of what amplitude you get with the amount of power that it takes to do that. And so everybody has a threshold, as I was saying earlier. Above that threshold, you can recruit more and more, you know, muscles in the hand, let's say if you're targeting the hand, and you get more and more of an amplitude as you go up with the power settings on the device, right? And so you can get a sense of how excitable the brain is for a given person through processes like this, right? And what you can do in that situation is you can then move it around, right? There's plenty of ways to move it around. One way to move it around is with psychotropic drugs, right? And there's a big literature on this. If you give somebody amphetamines, you're going to move it around. If you give somebody certain anti-epileptics or benzodiazepines, you're going to move it around, right? So that's kind of a sanity check for you guys. What would you imagine that a benzodiazepine would do? Would it increase or reduce cortical excitability? An anti-epileptic, would it increase or reduce cortical excitability? It would bring it down, right? And what would an amphetamine do? It brings it up, right? Coffee brings it up. Totally logical, right? Like everybody kind of gets that, right? So with stimulation, what's different about that is you're only really changing the cortical excitability in the place that you're stimulating, right? Not in the whole brain. And so we can put, in motor cortex, we can put through parameter sets that change cortical excitability just there and just in that spot. So we can drive up how excitable the brain is just in that spot without driving up excitability in the rest of the brain or in the non-network-connected parts of the brain. Some other places will get affected, but not in the, say, some of the limbic circuitry if you're stimulating a motor cortex, right? And so you're able to do circuit-specific modulation of systems without moving around other systems. Does that make sense? Why is that important? Well, you know, what if, let's say in schizophrenia, you want to turn certain regions up, let's say cognitive control, and you want to turn certain regions down, let's say auditory cortex and auditory hallucinations. If the stimulation approaches were global in the brain, then you'd see if you turned up cognitive control, then the auditory hallucinations would go up, right? But that's not what you see. You see these circuit-specific modulations, and that's another kind of important principle that you're able to affect specific systems in the brain alone. And in the case of motor cortex, we're able to generate new, and I'm going to put this in quotes. It's not actually a drug. New drugs within the motor cortex parameter set. Now it's not really drugs, right? It's new treatments, it's new parameter sets that are like new therapies, and it can be totally borne out of these motor physiology experiments, right? We can develop new parameter sets, new Morse code languages through these readouts. Does everybody understand that? And so that's what we did in the early 90s. We figured out that if you go below 4 hertz or below, you're going to inhibit or depotentiate the brain. You're going to turn down the electrical activity in the brain. If you go 5 hertz or above, you're going to turn up the electrical activity in the brain. Everybody get that? So the idea there is that folks did what they call parametric experiments, so simple experiments. I'm going to go one cycle per second, two cycles per second, three cycles per second. It's the number of pulses per second. They kept going up in these parametric experiments, and at some point, instead of seeing the motor twitch get smaller and smaller, at some point as they went from 4 to 5 hertz, the motor twitch reversed and started going bigger and bigger, right? So you can actually, you have a line in the sand when it comes to standard TMS about what frequencies will turn brain regions down, what frequencies will turn brain regions up. So the fun thing in the lab that folks will do, normal healthy control patients or participants, study coordinators or whatever, is that you can put it over Broca's area, so you can actually cause transient aphasia. It's not really aphasia, but it looks like aphasia, right? You're turning down the speech area, right? So you can use 1 hertz, 2 hertz, 3 hertz, and shut that area down transiently. You instruct them to speak, right? And they'll start garbling their speech. You do 10 hertz, and you don't get that, right? So frequency-dependent effects. And so Mark George, who received the research award here this year and was one of the inventors of the technology back at NIMH in the mid-'90s, was, you know, was, like me, trained as a neurologist and psychiatrist. So he was up at NINDS talking to the folks doing motor experiments, and then he was down in NIMH doing depression neuroimaging studies with PET and SPECT scans. And he had this idea, well, if I can turn up cortical excitability with this excitatory stimulation, 5 plus hertz stimulation, and I can see in depression that there's a reduction in activity in the prefrontal cortex, maybe in a motor cortex I can turn up, you know, blood flow and metabolic activity with TMS, maybe I can drive this prefrontal cortical system back online. And that was the real aha moment, right? It's this idea that this is a motor probe giving you basic information about how the motor system works, and essentially how cortical excitability works over here, and then here's first neuroimages of depressed patients, and then together that's information that will allow for us to make a new treatment. One second. I thought this was charging, and now it's telling me it's not. Hey, Brandon, you were just grabbing the charger, sorry. So that's the idea. So basically it's putting these two findings together. They were really new at the time. They were like, you know, essentially two or three years old, and he came up with this idea of kind of combining the two, and with that stimulating in prefrontal cortical areas that are involved with mood regulation. So that was kind of the aha moment, and that allowed for us to start thinking about TMS as a therapeutic approach. So there you go. All right. Yeah. How many psychiatrists does it take? You know, so there were a number of trials that were the early trials using that first stimulation approach, right? That first, you know, kind of parametrically derived approach, and, you know, they involved getting treated once a day, five days a week for six weeks. You know, early on it was three, four weeks, and then it kind of evolved to longer studies. And what they found was that, you know, in this big pivotal trial, some of the primary outcomes were, some of the outcomes were positive, some were negative. The primary was negative. The secondary was positive, and it led ultimately to an FDA clearance because when they looked at folks that had one and only one med failure, there was actually statistical significance, but when you put the whole population together, you didn't see it, right? And so there was this FDA clearance that happened after, you know, kind of appealing with more data, and the field kind of didn't know what to make of it, right? Like, what is a patient with one and only one med failure, like, you know, is that the sweet spot that we're really treating at? This is pretty early in the game, you know, how do I think about this relative to ECT, that whole sort of thing? Then OPTMS trial came out where there was statistical significance, but the number needed to treat was, you know, relatively high, right, meaning the number of people you need to treat in order for one of them to respond and one of them to remit, you know, so it wasn't kind of a groundbreaking amount of people that were getting better. So you know, folks looked at this, and you know, maybe rightfully so, we're a little bit skeptical about how to, you know, interpret this, right? So if you interpret this as a drug, this isn't the greatest drug, right? If you think about this like this is the end, and this is what this drug is, then it's not the greatest drug for this problem, right? But if you think about it like an engineering experiment, like we were talking about earlier with designing things, it's a great success, right? Because you know, they took a shot in the dark not knowing anything about this very infinite parameter space, very little, and really not that much about the brain anatomy of mood regulation. This was the mid-90s when we were just barely understanding all this stuff, and yet with a lot of those insights, they were able to get a win. It was a, you know, close call, but it was still a win, you know? And then the brain sway trial came out a little bit better as far as separation goes, a little bit better as far as the numbers go. This is a deep coil that kind of, in the case of TMS and magnets, one of the principles you have to understand is this idea of depth focality trade off. So what the brain sway coil does is it says, okay, you don't know where to stimulate within this brain region. So we're just gonna make the field a lot bigger so that the kind of active level of stimulation is broader in order to cover more ground. And it's a reasonable way of thinking about it, right? Because you're able to kind of, if you don't know exactly where to go, then you can kind of get across a much larger span of brain region and you can hit more of, in this case, the name brain region or the dorsolateral prefrontal cortex. And so with the conventional RTMS approval, these are the parameters, the brain sway is a little bit different, but essentially the same sort of thing, right? You're talking about what we call fixed frequency approach where it's really a frequency and then a certain break in between to essentially reduce seizure risk because in the very early days before they were giving what we call inter-train intervals or breaks in between, there were folks having seizures. What's important to understand again is this is, again, the first shot of an engineering approach. With that same vein, you have the brain sway trial that was positive for OCD and that led to an FDA clearance. Now this is a different position in the brain. It's not the dorsolateral prefrontal cortex. This is the kind of dorsomedial ACC area. So it's kind of right here. Think about it, you know, normal hairline is like right a little bit above a normal hairline in the center and you're really getting these medial structures that have been implicated in OCD. And with this protocol, which is useful to understand, there's a, unlike with depression, there's a provocation. So you really have to kind of provoke, in this case, show OCD images and then the patient, you know, finds relevant and then stimulate. And the idea there is that turning the brain on by inducing the brain circuit is a useful way of then modifying or modulating that same system, right? Smoking, same idea. You know, in this case, smoking images and they were able to, you know, have a statistically significant separation between active and sham, led to a clearance for smoking cessation. So that's really kind of the first generation of brain stimulation approaches with those conventional parameter sets. Then, you know, in 2005, we, you know, through John Rothwell's work, we became aware that you can use a different signal in the brain and that signal in the brain has different properties to produce the same output or outcome. So if I send a much more biologically relevant signal, I should be able to do it in a much more efficient way, right? If I'm doing Morse code and I'm sending the Morse code as, let's say, Mandarin, you know, translated into Morse code, translated into English, right? In this, you know, hypothetical example, it's gonna be a lot slower and a lot more inefficient than if I'm just using one language, right? Does that make sense? And so here, what we're doing is we're playing the brain's own native rhythms, the hippocampal rhythms into the brain. And when we do that in humans, we're able to actually dramatically reduce the number of pulses it takes to get the same message across, in this case, the same amount of cortical excitability. We're also able to reduce the amount of dose that we need, the amount of pulses that we need in order to get to the same change in cortical excitability. And that's exciting, right? Like this, I don't think there was like a big, you know, big buzz on the street when this came out. It was probably pretty under the radar that this, but it's actually a really big deal, right? Because that tells you that there's not just one way to do this, or another way of thinking about it, there's not just one drug in this drug company, right? It really kind of shows us that there are other, there are other things that this device can make that would, other parameter sets that would have a biological effect. And that's important because it opens up this idea that, you know, again, more proof, this is kind of an infinite parameter set where you can make, you know, in theory, an infinite number of stimulation approaches with the right, essentially the right equipment. And so what is it? What is theta burst stimulation? So it's a pattern in a pattern. And in the case of excitatory stimulation, it's a pattern in a pattern in a pattern, right? So the first pattern is 50 Hertz bursts. There are three of them, in the case of theta burst. The second one is that happens every fifth of a second for two seconds in the case of excitation. And then there's an eight second, what we call inner train interval. Now this may also sound like, you know, pick your other language, you don't speak, right? It's a little bit, you know, seemingly a little bit arbitrary. It's seeming a little bit unusual. But what it really is, is it's really this phase amplitude coupling you see with hippocampal rhythms in the brain. So there's a high frequency piece that turns on local neuronal populations, and then a long range piece that gets further. And it's kind of like AM and FM radio, if that makes sense. So you're trying to engage the short term areas of the brain that kind of immediate areas of the brain, you're also trying to send signals across networks, right? And so you're really just trying to emulate that process. And so when you do that with excitatory theta burst or intermittent theta burst, as they call it, you change cortical excitability and make the brain more excitable. And so there was a 400 patient non-inferiority trial that was published in the Lancet in 2018. And with that non-inferiority trial came a second FDA clearance for depression with this new parameter set. So take off the like, this is device brain stimulation. Think about it like a head to head drug study. So drug one is in red, the kind of comparator drug that we kind of quote unquote, no works. And drug two is the drug that we're looking to see if it works in non-inferiority. And what's so amazing about this and why it got into the Lancet and why it got FDA clearances, that it did, it was completely non-inferior. If you look at every time point other than week five is basically superimposed means, you know, anybody that's familiar with looking at non-inferiority studies, this is very impressive. It's truly just straight up non-inferior, right? It's the same outcome. And so if you're in the audience and you're thinking, well, I think all of this is just the interactions with the treaters and whatnot, and it's a nonspecific effect. Think about it like this. How much psychotherapy can you do in three minutes? I know there's like these brief psychotherapeutic interventions, but not that brief, right? Like, what'd you have for breakfast this morning? All right, your TMS is over, right? You're not going deep in three minutes, right? And so if that was the reason why you're seeing an effect, you should see a dramatic reduction in the comparator, the new drug, if you will, in its efficacy because the other one is 40 minutes. You know, you can get a lot done in 40 minutes. Three minutes, not so much, right? And we saw a complete non-inferiority. And that's what led to this FDA clearance. And this is the parameter set there. And so it'll be important to talk about some of this later on, but, you know, 18,000 pulses per TMS course, so 18,000 pulses for a six-week course of ThetaBurst to be able to get some improvement. So this is a video. And so what to understand about conventional RTMS is this idea that when you are targeting conventional RTMS, you're targeting using skull-based landmarks, which is pretty good. You know, you can get, you know, reasonably accurate within what I like to say is the city, but you don't really know where you're at within the city. You're basically in kind of a general area. And for some people, they may miss a given brain target they're intending to get. They may kind of barely skim it. For some people, they're gonna hit straight on it, right? You don't know because you're not measuring based off of what you know about the brain. You're measuring based off of what you know about the skull surface, right? And so, you know, that study I showed earlier shows that ITBS is biologically active for TMS, for depression, I'm sorry. So, you know, the next question was, how we've got plenty of good TMS treatments for outpatient, right? For folks that are essentially in a non-crisis mode. And for people who are essentially able to take what in some cases is two hours a day off of work, five days a week for six weeks. How many of us can take two days, two hours a day every day, five days a week for six weeks? At least not me. It's complicated, right? But if you can do that, you know, you've got a treatment that you can do and, you know, do in kind of your timeframe and you get well in kind of like a month and a half or two months. I got very focused on, you know, the folks that can't really do that. The folks that really can't get, you know, can't wait two months. The folks that are kind of in more of a crisis mode right now, right? And, you know, admittedly this was inspired by conceptualization of thinking about depression in a different way, you know, through the ketamine trials that were done in the early 2000s, right? Because what ketamine, probably there's pro-ketamine people, some neutral people, some anti-ketamine people. I'm not advocating ketamine necessarily in what I'm saying here. What I am saying is ketamine told us that we could get depression treated in a faster timescale. And that's what we learned from ketamine, right? That you can get people well in hours and, you know, clearly well in a day or something like that, right? And that was useful as a conceptual breakthrough because it told us that this six week timeframe that conventional TMS is on, that drugs are on, that, you know, psychotherapy is on, isn't the only speed of improvement. And I think that part is where the ketamine story was quite useful. It kind of instructed us in that way. And so, you know, in 2014, we were really thinking about, you know, what we were gonna do in the first couple projects to the lab, you know, I'm very much kind of focused on trying to solve clinical problems that, or problems that, you know, maybe folks are thinking about it at a level, but it's not a huge focus of the current, like armamentarium of experimental therapeutics. And so, you know, this is one of these problems, right? This idea that if you're talking about patients who are, you know, hospitalized in the hospital, then you're talking about patients who, you know, don't really have the same treatments as outpatients, right, and so, in fact, because you escalate care, you, on average, lose treatments because most hospitals don't have ECT, right? And no hospital has TMS, and some of these other novel therapeutics, I guess, ketamine and ketamine, they're not really consistently available, if at all, in inpatient settings. And so, on average, you actually lose treatment opportunities in contrast to the way that rest of medicine works, right? In the rest of medicine, if I start having a heart attack right here, I hope one of you will stay and hand me some aspirin and call 911, but that's about all you're gonna be able to do, and then somebody's gonna get me to the ER, in which case there are more tests and more treatments, and then somebody, if they think I'm having a heart attack, will bring me up to the ICU, where there are more tests and more treatments, right? We don't have the tests in psychiatry from a biological standpoint, and like I said, you lose treatment options. And so we decided this was a good problem, and that information is coupled with this other piece that you all know, which is that the suicide risk for a mood disorder patient in their first psychiatric admission peaks at the point they're discharged out of the hospital, right, so the highest suicide risk lifetime is in that first period after they're discharged out of the hospital, and so being able to affect that and change that risk profile portfolio is important, I think, for our patients, and being able to really think about folks exiting and having a lowered risk. And I'm sure I'm not the only person that's seen beds in ERs like this. Some of you have seen this, right? And it's really kind of a pragmatic issue as far as getting people into the right spots. So we went back to kind of first principles and asked the question of how do you engineer a solution? Now, as I said earlier, there's no singular right answer to this, so what I'm about to describe isn't the only way to do this, likely. I don't know another way at this point, but it's not the only way to do it, probably, given the parameter space that we are operating in, but it's the one where we had the most information, especially in 2015, and so the first thing that you have to think about is kind of first principles. What is the principle, what is the properties of the pulse that I'm putting into the brain? And so how many and what intensity? And so there have been a number of studies to look at this. Again, parametric studies looking at motor cortex, excitation, these are super boring studies. Somebody had to do them, but very useful and very, very informative to the field, but 90% motor threshold, 100% motor threshold, you go up and then you vary the number of pulses, and what this group found, Charlotte Nettenkoven's group found, is that there's a sweet spot of excitation where if you give 1,800 pulses of stimulation at 90%, you produce the greatest change in cortical excitability, and that was confirmed with single daily theta bursts where you apply a single application of 1,800 pulses of ITBS and then look at depression outcomes in a standard schedule. The second question, once you know what the unit is and the session kind of characteristics are, then you're looking at how much time in between, the default amount of time in between historically has been a day, right? If I'm coming in once a day, I'm basically at the 24 hours. How many of you use note cards in college to cram for, nobody crammed for their med school test in here, did they? Whatever, neuroscience PhD test or whatever it was. Yeah, so how'd you do it, right? You wrote out about 60 to 90 note cards, right? And then you looked at one about every minute, and then about an hour, an hour and a half later, you get back to the first one, right? Everybody intuitively did that. Anybody do it differently? Everybody intuitively does it like this, why? Because that's the optimal learning process for the brain. You need to be exposed, and then you need to be re-exposed, right? And I would imagine most of you didn't write out one note card, just look at it over and over again 50 times and set it down and never look at it again. That's what we call in mass stimulation. And this mouse experiment has, you know, shows us they've replicated this multiple times. And if you just stem, stem, stem without a break, you get a change, but then you don't get any more change. You get one level of change, but you don't get any more change. You don't get interval increases in dendritic spine enlargement with more stem. You need the stem, and then you need a break, and then you do it again about an hour later, and you get a break, and you can build these connections in the brain, and that's how memory is formed, right? And so what we're doing is we're stimulating with a memory signal in the brain, same one that has these same principles, and we're saying, turn on, remember to stay on, and then about an hour later, turn on, remember to stay on, right? I don't know if there are any parents in the room. Have you ever had to tell your kid more than once to do something? Right, same idea. You're, about an hour later, did you clean your room? You know, it's the same basic idea that you're using the memory approaches to get folks to remember things. And we can do this in motor cortex too. We can stimulate and use those same motor of potential readouts, and you wait 15 minutes, you don't get anything new. You wait about a half an hour, and you get some. There are newer studies where you wait an hour, and you get quite a bit, right? The second principle is this principle of dose. Any pharmacologists in the room? There's one, a few over here, right, right, right. So when you're looking at a new drug, what's the first, what's the phase one study? Dose finding study, right? You need to know kind of what the dose is, and then the side effect profile of the various doses. So the thing about TMS was that there was a set of assumptions applied to basically the protocol and its complexity, right? And so this protocol in the mid-90s was hard to use, right? You needed to have an ice bath or dry ice to cool the coil off, right? Think about that. You're up there with this patient, you're putting a high-powered electromagnet into a thing of water, and you're cooling this thing off, right, it's not for the faint of heart, and it'd blow up sometimes, and it was a really kind of wild time frame back in the mid-90s, and so the idea that you'd be doing this more than once a day, I wouldn't have wanted to do it more than once a day in that setting, right? And so there was a certain, like, this probably is good enough, and this duration probably is good enough, and that was a good starting point, and it's true, I think it was a good starting point, right? But what they found in the subsequent trials was if you treat folks and you have non-responders and you just keep treating them, in this case out of 16 weeks, you just keep treating them, you pick up more and more responders. You actually accumulate benefit as you increase dose. And that's the second principle. So we talked about reorganization of the signal in time, the second principle is this principle of dose. Not only do you need to look at a note card once and an hour later see it again, but probably more than that. How many, I mean, maybe some really serious, you know, photographic memory people in this room can say once or twice is enough, but for the rest of us, the normal mortals, you know, you probably needed to see it 10 times or 20 times, right? You need to go through that note card stack quite a few times, you needed more dose, right? And that's the idea of dose, this idea that you're sending a memory signal over and over into the brain. So here's another video talking about dose and the timing of the dose. And so it's just, you know, I kind of walked you through this verbally, but this is just a way, kind of another way of seeing it. Right, and so, like I said, 50 hertz triplets every fifth of a second for two seconds with an inner train interval of eight seconds. That whole unit is 10 seconds, the on and the off. And you're giving, in the case of the accelerated protocol, about nine minutes of stimulation, right? And the conventional theta burst approval was for three. And so then the next question, again, is that question of the dose response curve and of space learning. You're giving it, in the case of conventional TMS, once a day. And for the accelerated protocol, we're giving this 10 times a day, right? So it's just kind of a different dose and a different kind of application of that dose. It's a much more aggressive dose. And so if you look at this, that six week application is what's given in a single day, right? So if you reorganize the pulses in time, what you're doing is you're actually applying this over a much shorter period of time, 10 hours, right? So the six week course that we were giving with the 2018 clearance is given dose-wise in a single day. And then we give five times that to cover that second principle, which is the principle of dose response curve, of giving more dose. Right, does that make sense? And so it's just like a visualization of the same way of thinking about it, right? It's a completely different schedule, and so it's a different treatment in that way. Right, so the next principle is this principle of where. Right, if you're gonna stimulate, you need to know kind of where in the brain that you wanna go. Right, so you wanna stimulate in a place that's optimal for that person. And so, any violinists in the room? Yeah, so your hand motor cortex is gonna be probably bigger than everybody else's in the room, right, if you're a significant, if you play a significant amount of violin, right? Any soccer players in the room, martial artists or anything? Maybe some of that, right? So the same thing, your foot representation, right? Our brains are different just both from the environment and from the kind of native makeup of where things are positionally. So using average positions within a named brain region doesn't make sense because there's kind of sub-region specificity of function, right? And that makes sense, right? Like it would be weird if there were like 50, 100 named brain regions and they only did one thing. We can only do 100 things, we do like way more than 100 things, right? So you need local specificity. And the kind of general, and this is definitely simplified for the neuroimagers in the room, we're gonna say there's a lot more brain regions involved and that's true. But for this kind of basic story, the idea is that mood is a regulatory process between the dorsolateral prefrontal cortex and the anterior cingulate. And depression is a problem where the dorsolateral is underactive and the cingulate is overactive. And what TMS is doing in the case of depression is really re-regulating that system, right? It's re-regulating the nature of those two brain regions. And we'll talk about it, the anterior insula is also involved. And so what we felt like you needed to do in order to kind of fix that is, you have to find the right place to go in. So then you can enhance the inhibition of the sadness region, right? So that's the basic idea, is that we're trying to use one brain region, which it's native process is to clamp down and suppress another brain region. So if you take that ruler measurement piece that you saw a little bit earlier, and you put the coil in that same place, but you get brain images of functional connectivity, brain images of that person. You're not gonna use them for targeting, you're just gonna get them. And then you stimulate in the ruler place, but then you go back and take that scan and look. How far away was that ideal spot to the place that we actually stimulated, to the ruler spot? What you find is, the distance from the ruler spot to the ideal spot, the closer it is, the more antidepressant. The further away it is, the less antidepressant. This has been replicated a ton of times, so another aspect of this. So what we're essentially doing, right, is we're taking this brain image. At an individual prospective level, we are going through and finding that individual's target using the analysis pipeline I'll talk about in a second, and zooming in and putting the coil right in that spot. And what that does is that guarantees when you put the coil over that spot, that you're actually engaging that brain network, where you think you are. And that gives you some guidance. It's like, sometimes I put the pill in my mouth. Sometimes I put it in my nose. Sometimes I don't put it in anything, and it falls to the side. That would be a problem, right? It's the same sort of thing here, right? You wanna be able to 100% sure you're getting into the system that you're trying to get into. And so we do this through functional connectivity analysis called hierarchical clustering, where we cluster similarly behaving voxels in the dorsolateral and the subgenual. And then we use a decision algorithm to pick that best spot, and it outputs out to the structural brain. So this is the protocol. I've described it in depth for you. This is a visualization of that protocol. So essentially, doing a dose response curve in target. At a fundamental level, it's a very easy idea. I just wanna know how much, if I wanna put the drug in the mouth of the patient every time, and then I wanna give a pretty significant elevating dose and see how much better I can get as I put the dose in. So reorganization in time, space, and dose. Our first study was published in Brain. It wasn't called Saint-Bac then, we called it STBS. And we treated folks who were on the list to get brain surgery for depressions. They failed ECT, they failed conventional TMS. And we treated them, and we saw this dramatic but unsustained response, right? And everybody said, this must be placebo. You're bringing these guys in, they're spending a week, and then they lose it, right? And we've done subsequent studies where we've treated people with two weeks, and we've been able to right shift the durability of that curve, right? Suggesting it's, again, a dose phenomenon. We've done beta versions of this with implanted devices and seen a similar sort of outcome, although the time scale for this kind of beta approach was much slower because we didn't have the pattern, we didn't have the targeting back then. But suggesting this idea that just like with spinal cord stimulators, we can do a test stem with non-invasive stimulation and then follow it with an implanted device is underway. And what's useful about that is that we can get direct brain information with an implanted device, and we get larger amounts of scope with a non-invasive device. The second finding is in a larger group of open label patients, and these were quite ill people, so about five med failures or more for about half the patients who are in moderate to severe depression range. And what's useful to understand is, again, the day one dose, the day zero to day one dose is the same as the six week approved ITBS daily dose, right? And so what you see is you see the same sort of drop that you would expect. You see it's pretty good, but on average, people are better but not well, right? And that's what you see with conventional TMS. If you average the 30% remission, the 30% response, the 40% non-response, you get about the same average number in the drop. But if you keep going, you pick up more and more remitters, right? So we have, like I said, 5X the dose of the original ITBS approval. And by day two, three, four, you start to see people cross the remission line for the HAM-6. That's a score of five. And then treatment naive folks versus folks that have failed RTMS, you get the same outcome, but it takes longer in the conventional RTMS failing people, the people that got the six week thing and it didn't work for them. And they relapse a little bit quicker. And then when you treat them and retreat them, you can get them back to the same spot, suggesting there's no tachyphylaxis. And when you look at the target distributed over the whole brain, you can see the position of each one of these dots on this average brain relative to the standard ruler position that we normally use. So we did a randomized control trial in moderate to severe severity, moderate to severe treatment resistance. We excluded all the standard stuff, but in particular, people that had passed RTMS because of the blind and non-response to ECT because of the durability. About 32 people initially came through, two were excluded prior to starting. One was excluded from the analysis due to not being transparent about their diagnoses. The mean duration of the current episode was nine years. 50% were on disability, and they came in with moderate to severe severity. And here were the response rates on the bottom. All of this was highly statistically significant effect sizes above one, any way you slice it. And if you look at the single subject data, 78.5% of people transited through the remission line at some point in the active group, and only about 13% in the sham group, which is really dramatic, right? It's a very big striking difference between the two groups. And it suggests that, again, back to that early, early thing that we were talking about at the very beginning. If you think about TMS as a probe and treatment, when you perturb a network node and you move it, then that tells you something about the nature of the problem, right? So this is a way of actually understanding biology, and we'll talk about that in a second. And you get to this question of, can we do this in suicidal inpatients? We treated about 12 people, average of 1.5 suicide attempts prior to admission to two hospitalizations prior to admission. We did retrospective matching with ECTs. This was in a non-inferiority study, it was a pilot study, but it just gives some sense of things. So these were ECT patients that were sex, age, treatment resistance severity matched. And so you can see that this is much quicker than ECT's time scale. The SSI is clinically relevant at a six. So you can see by day one, two, you start transiting people out of that risk zone. When we followed people who responded, they maintained it out for a long time, right? So unlike ketamine, which is pretty quick in a single dose, this is a much longer durability. And then I can end here, if folks are ready to go, or I can tell you guys five more minutes of exciting information, and we can do questions. Is that all right? All right. Great, so this is a paper that came out in the Proceedings of the National Academy of Science on Monday. So pretty exciting stuff, a post-doc that I share with Karl Deisseroth, who is a MD-PhD student with Marcus Rakel in Wash U. So what he did for his whole PhD, again, boring science becomes exciting science at some point. That's a good kind of rule of thumb, right? And so he was really technically looking at this one phenomenon, which is that if you look at the blood fluctuations and network nodes that are supposed to be in a connected neural network, they'll fluctuate in basically the same time scale. And if you average them, you get a average network kind of fluctuation. But if you look at each one independently, you can see that one network may be slightly in front of another one. So can anybody tell me from this graph, which one is slightly in front of the other one? All right, let's take a guess. Only one wrong answer and one right answer. Anybody? Can you see it? Which one's slightly more to the left of the other one? So the blue, the blue slightly precedes, my pointer, slightly precedes the, I think it's iron, maybe red. I'm not sure, I think I'm kind of color blind. You guys have a pointer? There may be a pointer, did you plug in? There it is, yeah, there it is, okay, this should work, right? So can you, I want you to believe this. Can you see this right here? You see the blue? You see it here? You see it's slightly in front here? It's in front here? You see that? Just slightly, it's very slight. So if you do this mathematical maneuver called parabolic interpolation, which is on the panel to the your right, you can do this math where basically you can figure out where the peak of that fluctuation would be if you had continuous data, which MRI is not continuous data, and you can interpolate where one peak is versus another peak. And so in normal healthy controls, what do you think would be, based off of what I've talked about already, what do you think would be temporally in front? The dorsolateral or the cingulate? What would be, if being in front means that one controls another, which would be in front? The dorsolateral, exactly, right? You'd want the dorsolateral prefrontal cortex to be in front of the anterior cingulate, why? Cuz it's in control, right? So that's like the warden and the prisoner, or that's like, whatever, you can get various parent and the kid, I don't know, whatever analogy you want. That one's in control, right? Now, what's interesting is when you do active st. with folks, what you get, and this is two data sets, what you get is, if you look at the pre-post, you see that the after effect is that the dorsolateral is in front of the cingulate, right? If you replicate that, and we did that, so we did a single trial, replicated again, dorsolateral after, if you do an image after the treatment, it's in front of the cingulate. And if you look at the lag in that process, it's correlated with the moderate score, with the depression scores. Now, what do you think is the case with the pre-scan, the depressed pre-scan? Anybody wanna take a guess? The cingulate is in front of the dorsolateral temporally, right? And some of you may be skeptical of neuroimaging having any role in psychiatry, right? It's been a lot of promises for a long time, right? So one of the big tests in this kind of thing is, does it survive going to a new scanner, right? Cuz that's one of the biggest problems, a lot of noise in scanners. So if I go to a different scanner at a different university, with a different brand, am I gonna see anything? So this is our GP scanner, and this is our GP scanner, and this is the Siemens Prisma at Wash U. So if you contrast depression with our scanner, you see that the cingulate and the anterior insula are, I'm sorry, the dorsolateral and the anterior insula are slowed compared to the cingulates, so the cingulates in front. If you look at the articraft patients and compare them to the Siemens Prisma at Wash U, you see the exact same thing. So that contrast between scanners is there, same across both of the samples, suggesting that it's not a scanner specific effect, this isn't an artifact. But what's really cool is if you then take the folks that had kind of a home run from Saint, they got a lot better, and you contrast some of the people that didn't get any better, what do you think the contrast looked like? Looked like the normal healthy control contrast. So the folks that got better all had a very intense cingulate way out front of the dorsolateral and anterior insula. And then the folks that didn't respond, their brain scan didn't change, and they didn't have it, they looked like normal healthy controls. And that contrast is the same difference as the difference that you see with the normal healthy control contrast. So, the last question I'll ask you. You are a general practitioner in an office and you're seeing a patient with polyuria, polydipsia, headache, and blurred vision. Does every single one of those patients have diabetes? No, a lot of them do. And what's the first thing you're gonna do? You're gonna do a blood sugar. And then what's the next thing you're gonna do? You're gonna give insulin. And then what are you gonna do? You're gonna check the blood sugar again, and the blood sugar is gonna be normal, right? But some of those people have migraine headache, and they need new glasses. And they are just drinking a whole lot of drinks at home, and da, da, da. And it's a hard picture, right? It's a very hard picture to discern that, right? And so this idea that there may be two flavors of depression, there may be two biotypes of depression, there may be two kind of neural manifestations of depression, with one of them being this particular finding is interesting. And if it replicates, and we're doing that in IH trial, it's gonna be very important because we have a way to predict treatment, decide on treatments, which I think is pretty cool, if that's true. And the last piece is the severity of the illness, the higher moderate scores are correlated with more of the flipping, more of that signal, suggesting that this actually, and we've seen this. It works better in more severe people. It's also counterintuitive, right? That you really wanna, and it makes sense for those of us who are mood disorder folks, and folks who've worked with a lot of depressed patients, because when somebody's really severely MDD depressed, it's so clear, you know. When it's on the borderline, it could be other things, right? But when they're in that kind of 30 to 40 modulus range, it's pretty clear what their diagnosis is. And it's pretty clear in the brain, it seems as well, which is I think very, very, very interesting. So I will leave it at that. I have a lot, a lot of people to thank, lists of faculty and post-docs and staff that are working on these things tirelessly. And couldn't do it without any of those guys and without the support of our funders. Thank you for spending time with me on, at the end of your conference, after a long week. So we will be taking questions alternating between the live audience. Please come up to the microphone and the questions that are coming in from our remote live stream audience. If I may, I'll kick off with a straightforward clinical question, which is presuming a perfect world, availability, as well as insurance coverage. How do you think about where to put it in the algorithm for treatment resistant depression? Yeah, that's a great question. So, as I suggested earlier, and I can kind of tell you a little bit about the process as it stands now, but I really thought about this as putting this in in inpatient units, into places where there's a real need and PCT isn't taking care of all that problem. And in many ways, it's not taking care of much of that problem if you look at the total numbers, and so having another treatment in that setting makes a lot of sense. From a combination of clinical need, risk reduction, and actually insurance, Medicare proposed rule out on this, considering paying for an inpatient through a new treatment out on payment process. And so there's a real possibility that actually the feds will agree with that and be open to covering something like that. Which I think, that's kind of the first point of entry, right? It's really trying to get, for the reason I just said, right? This most severe people seem to do the best, right? That really getting into that highest severity group and really trying to push the envelope there feels like the right starting point. And in the outpatient world, for someone who's failed, let's say, two or three standard pharmacologic approaches, and how do you think of it vis-a-vis ketamine, ECT? I can imagine this room in this hospital with a patient hooked up to ketamine while undergoing this and ECT ready to go. There you go, yeah, that's right, yeah, that's right. Yeah, it's a complicated thing because we don't, PCORI started to fund these things, but we don't have good head-to-head trials on, even take this out, even the algorithm of do you do TMS first, do you do ketamine first, do you do ECT first? There have been some head-to-head ECT ketamine trials, which it's looking like ketamine really, in older adults is not a good idea, doesn't seem to do as close to what ECT does. When you get into the younger adults, maybe it's a little closer, but at the big level, isn't quite getting to the ECT phase. And there's some open questions about the kind of ketamine abuse liability issues that are ongoing and debated. But I think the beauty of non-invasive neuromodulation, TMS, and all these things, they all, they have the same basic principle that the risk-benefit is just so good, right? Like, you know, folks either, you know, largely either get well, or they don't, and then they're not saying, oh, they're disappointed or something, but they're not saying that, you know, they've got this list of side effects, right? Some folks will have some headache around the stimulation, there's this exceedingly rare risk of seizure with conventional TMS, it's less so with the newer forms of stimulation, but, you know, but outside of that, exceedingly low, kind of well-butrin level risk, you're not really looking at, you're not really looking at any other major risks. So I think that non-invasive neuromodulation makes sense to introduce earlier on, getting, you know, getting conventional TMS paid for in that population is happening, right? Getting these newer stimulation approaches, you know, have time to get it done. Yes, please. Sorry, you're done? Thank you for such a fantastic presentation. I have a clinical question. I'm from New Zealand. I've been using TMS for the last three or four years. Started with theta burst, left-sided theta burst, with difficult-to-treat patients, and just a fantastic response, but a big group of non-responders. Excellent. Switched over to bilateral theta burst, left, one hertz, right, another step up. Fantastic, couldn't be more pleased, but a group of non-responders. There's no insurance in New Zealand. This is all out-of-pocket, private. It's expensive treatment. Yeah. But just to try and make it more and more effective, do you think, my question is, do you think there's a cost-benefit advantage to neuro-navigated TMS over standard beam F3 positioning on a coil? Yeah, that's a great question. So, you know, the economics of this are always interesting because, you know, particularly the international level, right, but what I would say at the like, you know, at the country level, or at a health system level, or at an insurance level, if you have a big insurance company that's providing this, if you can, if you can do both of these things that I described earlier, use the target, use the imaging as a where, but also use the imaging of, you know, typing people, then at the population level, it may be cheaper, right? If you're, if you know that this group isn't gonna really respond in that, we need to replicate this, there's a lot of work that needs to get done to really be able to say that with a lot of, you know, be very definitive about that, but if this plays out and that's real in that hypothetical example, then it's probably gonna be cheaper to image because you're gonna be able to take some people out that would end up costing the system a lot, and then that'll offset the cost of doing the scans for everybody else, and I think that's where the real value add is, and so that's the, I think the dream for all of us doing neuroimaging, you know, research, is this dream of being able to type people and then know where to stimulate, and if we can do that, then I think that the payers will be very happy because they like, from what I understand, you know, they like to pay for things that work, but they don't like to pay for expensive things that don't work, right, at least here in the US, and I think that's gonna probably be true for a private pay market too, right? Awesome, okay, yes, sir. You know, one of the things I really like about this is that it helps us rethink the way we thought about what we do as psychiatrists, right? The paradigm shift is from synapses to networks, which to me is a phenomenal shift, right? And it's not just one network, but it's maybe interrelationship between different networks. Absolutely. The other profound concept is the temporal one that you mentioned, that it's almost like looking at the rhythm, and sometimes, I imagine like looking at an orchestra script, or listening to an orchestra, but I think for most of psychiatry, we've been looking at maybe one pianist playing with one finger at one piano, and we've missed the whole orchestra. So I'm wondering, how do you make that shift, number one, because it's so hard for people to understand, and finally, the thing that I've seen using TMS is you can use other things, like when you do the motor threshold, you may have done this. There is a lag, you have to wait before you repeat the motor threshold for the medial nerve stimulation, but if I ask a person to think about moving their finger, even though I haven't waited longer for the intervention to occur, the finger moves. How do you think about that, and maybe a little bit more about using like therapy with TMS and then de-cyclosthenic? Yeah, no, that's a great question. So I'll answer the second one first, because a little more straightforward on that one, and you're absolutely right. We've learned from motor physiology experiments that the act, the mental act of thinking about your thumb before you move it changes the excitability. So that actually prompted a lot of people to start thinking about exposing people to other things where they're thinking about it while they're stimulating, because it actually drives around cortical excitability, changes the propensity of the brain to fire from TMS. The thing is very cool, right? It's this idea that we can drive up and down systems by exposing people to those systems, essentially bringing those systems online, which is what you're saying. So we've got a lot of work to do there. We don't understand it. De-cyclosthenic actually will improve TMS outcomes, which is kind of interesting. Psychotherapy agent, the people have explored, and now it seems to really augment the TMS outcome, and so that's another example where we don't know a lot. What's the optimal pharmacology? What's the timing of that optimal pharmacology relative to the stimulation? These are huge, unexplored areas that we're just right at the precipice of understanding. I mean, this is very much what I just showed you is very much a blind person, or not literally or figuratively, however you wanna think about it, walking through a room, and you barely know what's in it, right? In that way of trying to learn and find things, we found a few things that seem to do that surrogate that we're looking for, but we don't know what the rest of the room is, right? We know this little piece of the room that we've felt out with our senses, that we've kind of explored with science, and as we keep going, we're gonna map out more of that room, and I think we're gonna have more of a sense of exactly the questions you're asking. Now, the first question I'll ask you, right? We've been working on these animations, trying to find ways of illustrating this information quicker. Did you think that was helpful? Okay, all right, that's great, yeah. Dr. Nolan, we have questions coming in from our live stream audience. One of them was just, I think I know the answer here, but they said, if insurance was not a factor, do you think it's worth skipping med trials for some patients and jumping right into this ITBS? Yeah, that's a great question. So, I don't know if Richard Bermudez is here, but he writes for Psychiatric Times. He's been pushing that button, and I think it's a smart one to push. Yeah, yeah, yeah, no one agrees, yeah. So, here's the deal. You know, our neurology colleagues know this really well, that if you have a mesial temporal epilepsy patient that has sclerosis in their mesial temporal lobe, meaning you do a scan on an epilepsy patient, and you see the kind of midline part of the temporal lobe has a little lesion on it, the chances of that person getting relief from their epilepsy with a drug is somewhere in the like, less than two, less than three percentage range. And so then, they know when they see that, that they go and do surgery, right? And then they're able to then, you know, triage people, right, because it's expensive to do temporal lobe epilepsy surgery, and it's also, you know, in that case, unlike this, it's also pretty invasive, right? And people have some language issues sometimes. And so, you know, I go back to the same thing I was talking about earlier, which is if you can triage people, and you know that, for instance, we don't know this yet, but just as a hypothetical example, you have a really clear flipping of that temporal signal, and you take those people, and you put them through five drug trials, and you can't get them out of it, then we've got that epilepsy story. Then you can, and I think you can do it really effectively. You can go to insurance companies, or whoever you need to convince, and you say, listen, in this subpopulation of patients, where you've got the flip, they should just be treated right away. The thing is, is that there's quite a few people that flip in that population, but we're pre-selecting for very severe people. I have no idea what the incidence of this is in the normal population. It may be 30%, it may be 50%. When we get a sense of that, that's gonna be really helpful, too, because I can answer that question more clearly. Sir. Thank you very much for the kind of landmark work you've been doing. I was telling you earlier how excited I was ever since I saw the paper. I'm Shu Barman. I'm an addiction psychiatrist from Wisconsin. My question to you is, did you see any difference in the side effect profile between the same protocol versus the conventional? My other question is, are you aware of places which are trying to replicate this at this time, or even doing it in clinical practice? I've been hearing from some people, some places have already started doing this clinically. Yeah, there are a couple of those guys at the rim, I think. Yeah, so there are people that are doing it clinically, and that's all getting worked out right now. I'm happy to talk offline with you about that. The other piece is around replication, and that's going on, too. So there are a number of studies that are actively trying to replicate it, including us, right? We're trying to replicate our own work. Side effects. Oh, yeah, side effects, yeah. So the one thing that we observed that we didn't really see with the once a day theta burst is a side effect that happened in both groups, and that's fatigue, right? And you can imagine, I mean, this is an all-day thing, they're not sitting in the chair for 90 minutes, but they're there all day, right? And you're kind of on the whole time. And so in both groups, although not statistically significantly different, so we don't think it's from the stim itself, both groups were kind of fatigued. You tell them to take the weekend off and rest, and then on Monday, if they got well from this, they're still well, they're not fatigued anymore. Euthymic people can go and do a marathon and be knocked out, it's kind of like that. And they'll tell you that, they're like, I'm not tired like I'm depressed, I'm tired like I just ran a marathon. Thank you. Yes, sir. So next online question is, what kind of budget would I need to bring this to my hospital, and what else would I need to be offering this to the patients? Yeah, you know, there's a company called Magnus Medical that's commercializing this now. You know, they'd be the ones to talk about that. I don't totally understand those numbers, to be honest with you, some of us do, and some of us don't. But I can tell you that, you know, there's a whole process, I think, in place for that sort of thing as people start pushing this into the real clinical space, so. Okay, thanks. Yes, please. Hi, I have a question about the parameter space. Someone else was talking about exposure before therapy. I was wondering about if pulse waveforms have been explored, but just generally, what are you most excited about in terms of all the parameters that are maybe currently being looked into? Yeah, you know, I mean, one of the biggest, and thanks for coming to this talk, we met at the last talk, you know, it's actually the conversation we had earlier, right, around the cross-device translation, and I talked a little bit about it in here, but this idea that, you know, TMS is probably not the device we're gonna use for this in 50 years. If we're still using a big magnet that's like the size of, you know, whatever, this can of water or whatever, then we haven't innovated enough, right? And so, you know, how do we do this with smaller footprint devices? How do we build devices that don't require magnets, like ultrasound? How do we do this with implanted devices? How do we do this in new ways that get this into places that, you know, a decent-sized footprint can go, but can't, you know, maybe optimally go, right? So just really trying to innovate further and figure out ways to kind of narrow the footprint and make it a smaller approach, and with that, all the device-related differences that we have to work out, and it's a big, it's an engineering challenge at a fundamental level, but an exciting one, right? So I think it's that. I think that once we have an implanted device in place, you can do all sorts of really interesting parameter evaluations to generate new parameter sets, you know, and I think that's probably where it's gonna go. This is a, just like it has been for the last 40 years, good guesses based off of logic and neuroscience that we know, and, you know, a lot of, like, translational work, but we really need, we're probably gonna need our machine learning, you know, assistance going forward with really figuring this space out, because it's an infinite space, and I don't think the human brain can derive all of it, if that makes sense, make that library, so. Thank you. Yeah. Thank you, the next online question is, are there certain patient populations that may not respond to this, like, for example, traumatic brain injury patients? Yeah, you know, it's a great question. I mean, there's some work in traumatic brain injury that looks pretty good, and, you know, for that data set, there isn't a clear, like, this doesn't work in, say, mild traumatic brain injury. When you get into severing of the cortical, you know, severing of the tracts that underlie the cortex, you know, that's clear that there's some issue with transmitting, right, if I live, if I've got a telegraph, you know, sender, and the wire's cut, you know, I can't send the signal, so that, in those extreme cases where there's real, visualizable lesions, I think that's the population that probably doesn't benefit, and we have to think about other ways of doing this. Yes, sir. Yes. Thank you for your work. It's been a ball to watch it. So I have a clinic in Rapid City, South Dakota, and we serve a catchment area of about 250 miles, and so a lot of people can't come in daily for TMS, so you might imagine, when I first saw the same protocol, I thought, well, whoopee, you know, we can compress the protocol, but I gotta do it right now, because I got people right now who are driving, you know, who would have to drive two hours each way for treatment, so I thought, well, okay, we can at least try doing, you know, three a day, two days a week, so they don't have to quit their jobs and move to Rapid City to do the treatment. Well, insurance won't cover that, but still, people, it's worth it, because they don't have to quit their job for six weeks to do it, and we've had some success, but I guess the question that I've got is, am I bullshitting myself to think that we're getting benefit from that, or does it seem logical that that would do it? I mean, we're giving essentially the same number of treatments. It's not compressed in one week, but, you know, a couple days a week, that sort of thing. Yeah, that was a great question. You know, in the old days of deep brain stimulation, there was a debate about whether or not you needed to map the brain or not, right, and so there were folks who were implanting into the subthalamic nucleus, but they didn't have the microelectrode recording tech to be able to do it, and so some of those Parkinson's patients did get better because of chance, right, and, you know, you're in a situation where, you know, in that case, some of those folks were in third world countries, they were in, you know, places where they didn't have the technology or whatever, and then the decision's made as a doctor to do kind of what you have to do. You know, on the flip side, I think that the danger of doing it and then dismissing it is there, but if you're pretty clear that you, you know, you're like, you know what, at the end of the day, when this all is available or whatever, then you're gonna really do a deep exploration of what you see to be the right move for your patients, then I think you just have to be kind of open to that whole process, if that makes sense, but yeah, my sense is that, you know, blind targeting doesn't always get you to where you need to go, and it doesn't mean from my view, that doesn't mean the person's failed, you know, TMS or SANE or however you wanna think about it, so that they've failed that particular way of doing it, and I think if you can get that into your head as kind of almost a different drug, if you will, you know, it's a different treatment, and you're trying a different treatment with these patients, and then you're willing to try that treatment or whatever treatment after, then I think I'd think about it like that, but as long as I think you're not conflating the two together, it's probably fine. Yeah, well, we do it in the same place that we, and we don't do neuroimaging to determine where it is, but we use the same strategy we've always used to find the place that we're using them, and we've got a good result with that, it's just that we're doing it at a different pace, you know. Yeah, definitely, I mean, it's, you know, I always ask the ketamine aficionados, you know, if you gave, you know, you give a half a milligram per kilogram in a 40-minute, you know, dose for ibuprofen ketamine, if you spread it out and give, you know, 0.1 milligram per kilogram, you know, five times, is that gonna do the same thing? And the answer that they give me is, I don't know, you know, is there something about hitting that level of a dose that gets you there? And that's the part I just don't know. We've been pretty religious because we, we're so focused on doing this really fast for this high-acuity setting that trying to slow it down for anything doesn't make as much sense, and so I only, in my, like, kind of metaphorically blind look at this, I only know what I know, so. Thank you, appreciate it. Dr. Williams, it seems to me that there are certain symptoms that have inherent temporal qualities, such as rage, irritability, impulsivity. Do you think there may be some special application directed at those types of symptoms where different circuits may be operating out of, perhaps out of sequence or in the wrong temporal relationship with each other? Yeah, it's a great question. So I think, you know, and we talked a little bit about this earlier, right, so there's data for one part of the dorsolateral prefrontal to be involved in anxious somatic symptoms, and another part of the dorsolateral that's involved in dysphoric symptoms, right? And this is some of Sean Siddiqui's work. So if you stimulate here, you get more of this. If you stimulate here, you get more of this, all right? That tells us that there's symptom-level clustering that's specific to brain circuitry, and if that's true, and they've got to prospectively replicate that, but if that's true, then what that tells you is that there's probably circuits in the brain that you move around, and you can hit symptom clusters, and if you can do that, then you can go symptom-specific as far as the treatment algorithm goes, and I think that's really where you get a lot of bang for your buck. Now, the problem is we need to spend a whole lot of money and have the whole, and we, one, we have to have the right brain mapping technology to do it. Two, we need to spend a whole lot of money then pushing on all those systems and verifying it trans-diagnostically, but if we can do that, then we turn psychiatry a lot more into what it looks like with epilepsy. There's certain seizures in certain places in the cortex, and in those places, you get these stereotyped symptoms, right, you know, you get this, you know, set of symptoms with temporal lobe epilepsy is a different set of symptoms with occipital epilepsy, and so, and that's, you know, anatomically specific, and so it's just like a totally different way of thinking about the problem. It's about the anatomy and where the system has gone wrong versus, you know, clumping diagnoses together based off of co-occurring symptoms. Well, I'm afraid that we've run out of time. We want to offer our great gratitude to Dr. Williams for telling us his presentation, and thank you all for coming to San Francisco. See you next year. Thank you.

Transcranial Magnetic Stimulation

TMS

ThetaBurst Stimulation

psychiatric conditions

depression treatment

clinical neuroscience

brain stimulation

Stanford Brain Stimulation Lab

neuroimaging

rapid-acting approaches

Dr. Nolan Williams

brain networks