A 'Revolutionary' Treatment for Suicidal Depression: What's Next?

Stephen M. Strakowski, MD; Scott T. Aaronson, MD; Nolan Williams, MD; Dr. Linda L. Carpenter, MD


March 15, 2023

This transcript has been edited for clarity.

Stephen M. Strakowski, MD: Hello. Thank you for tuning in today. I'm very pleased to have a program to talk about some advances in treatment-resistant depression with a wonderful group of experts. I'm Dr Stephen Strakowski, professor of psychiatry at the Indiana University School of Medicine and at the University of Texas at Austin.

Welcome to the three of you. I would like to introduce each of you. Dr Nolan Williams is a recently promoted associate professor of psychiatry at Stanford and the director of the Stanford Brain Stimulation Lab.

Dr Linda Carpenter is a professor of psychiatry at Brown and the director of the Butler Hospital Clinic and the Neuromodulation Research Facility. Finally, Dr Scott Aaronson is chief science officer at the Institute for Advanced Diagnostics and Therapeutics at Sheppard Brant and a professor of psychiatry at Maryland.

Welcome, everyone. To start out, Nolan, it's been about a year ago now that you published your landmark study outlining a new approach to transcranial magnetic stimulation for treatment-resistant depression in the American Journal of Psychiatry.

Could you tell us a little about it? I know it's called the Stanford neuromodulation therapy (SNT) protocol, but tell us what you did and what you learned.

Nolan Williams, MD: Absolutely. The old way of thinking about transcranial magnetic stimulation (TMS) is that it was kind of one protocol baked into one device, and then that protocol plus the device is the treatment. The treatment is kind of fixed, if you will.

Thinking about it in a different way and thinking about the TMS device as kind of a conduit for a whole host of treatments, the folks on this call have been pioneers in getting other indications explored and stimulating other brain regions outside of the dorsolateral prefrontal cortex and trying to treat obsessive compulsive disorder, PTSD, or whatever the condition may be.

In this case, we thought about it, and we said, "Okay, we would like to take what has been done, kind of pioneered as far as depression treatment and really try to speed it up and make it a treatment for high-acuity psychiatric emergencies." The current approved TMS treatment, the conventional repetitive TMS for depression, really doesn't work in the time frame that you need for a psychiatric emergency.

We thought a lot about this. The average length of stay for an inpatient with suicidal depression is around 7 to 10 days, so a 4- to 6-week course really wasn't going to work. We went in and reengineered TMS in the domains of where, how, and how much. A reorganization in space, time, and dose is another way of thinking about it.

The idea there was that it was a new form of stimulation called theta-burst stimulation, which makes TMS much more efficient.

We figured out a way to take the entire FDA-approved theta-burst course and compress it from 6 weeks into a single day. So using the 18,000 pulses that were approved by the FDA over 6 weeks, we figured out a way to compress that into a single day and use a certain type of MRI to optimize, at an individual level, the best spot within the dorsolateral prefrontal cortex to be able to deliver that treatment course.

We didn't just decide to give it for a day, but we went back to the drawing board based off of some of Linda and Mark George's work, where they found that continuing to treat over an extended period of time with conventional TMS, essentially going up on the dose over months in that case, led to greater improvement in depressive symptoms.

In this case, we said, "Okay, if we're giving an entire course in a day in an optimized spot in the brain, maybe we can do that over multiple days, the equivalent of, in this case, 7.5 months' worth of TMS over 5 days, and give a very potent kind of full dose of TMS and treat a larger range of individuals."

Over three separate studies, we were able to observe that even people who failed electroconvulsive therapy in our first study had dramatic and rapid improvements with the treatment. In that last study that you mentioned, the 2021 Cole paper, we were able to observe that those rapid, dramatic shifts in mood were only in the active group and not in the sham group, with the blind completely intact, which is really nice.

It's really kind of emulating the same sort of thinking that the ketamine pioneers have had for 20 years — that depression can be treated not over the course of months but over the course of days. We can get people well in a short period of time if we do it in a personalized way, and we do it in a way that's aligned with brain physiology and with the amount of dose that individuals need.

Strakowski: Several of those are advances over our traditional approach to psychiatric research. I appreciate that. It seems like a little thing, but I know you asked people if they knew they were getting placebo or not, and they were not able to tell. I don't think we do that enough.

Linda, Nolan alluded to your prior work as kind of setting him up. We all stand on the shoulders of giants. How does some of your prior work contribute to this? How do you think about what they've done that's innovative or novel?

Linda L. Carpenter, MD: We've learned a lot about TMS, but up until this really recent development, we haven't been able to get more than about 50% of people better. That's what's making this particular treatment approach really noteworthy. Regardless of how we've done it. We can get more people better if we, as Nolan said, give more treatment and keep going beyond 4 or 6 weeks.

By and large, across all naturalistic studies, you see about 50% of people respond and about one third of people remit, so we've all been looking for, and thinking about ways that we could make those numbers even better. Nolan's work really combined, as he mentioned, a couple of different elements and taking some time to tease those apart is something that the field's going to continue to work on.

Not only do you get this huge dose each day for 5 days, but you also get a different type of targeting. Not only that but also you have the dose and kind of a unique theta-burst protocol for each of the treatments that you have 10 times a day. There are many elements to the secret sauce, if you will, and all those things come together to make a really promising outcome in, at least, what now is a relatively small sample size, but a compelling and rigorous design for a clinical trial.

Strakowski: That's right. That's a good point. It's still being developed. We still have a relatively small group of people, so there's more to learn.

With that in mind, Scott, we dosed higher, we used theta bursts instead of other approaches, and we localized. Do we know yet which — or if all three — are necessary to gain improvement over traditional approaches?

Scott T. Aaronson, MD: I'd like to take a step back and give a little bit of color commentary about just how revolutionary what Nolan and his group has done. I absolutely agree with Linda. All of us deal with this 50% response rate and 33% remission rate, but that's in folks with moderate to marked depression.

Nolan is treating the worst of the worst. He's really looking at the folks who are generally hospitalized with severe suicidal ideation. These are folks who are excluded from almost any other TMS study that's been done, so he's getting pretty extraordinary results with an even sicker population.

The other thing that I think is really important to underscore is that I go to these scientific meetings multiple times a year and I walk around and see everybody's pretty MRI pictures on all these posters, and rarely do you see MRI results actually changing outcomes for patients. This is actually one of the first clinical trials in which you're using targeted TMS based on a particular MRI profile that appears to be improving outcomes.

Finally, we're getting to the point that our understanding of neuroanatomy and neurophysiology may, in fact, be improving outcomes for some incredibly sick folks. I think we're early in this paradigm as to who this is applicable for.

One of the things I love about Nolan is he tends to think like I do. There are always people who want to cherry-pick patients and take the easiest to treat, but Nolan has taken the hardest people to treat. We didn't mention the exact outcomes within Nolan's studies, but he's looking at outcomes that are dramatically better than we've seen in standard TMS trials.

Strakowski: Do we know if all three elements are necessary or is there still more work to be done?

Aaronson: One of the problems with all of TMS research is that we really don't know what the ideal protocol is. We don't really know how many treatments you need. We don't really know how many pulses you need during a particular session. Many of these parameters, we still don't know.

At this point, we're looking at billions of dollars' worth of research to improve our outcomes another 10% or 15%. I'm not sure what's going to be the most important part of the secret sauce that Nolan has developed. There has certainly been some argument about whether the targeting is essential, and there have been some arguments as to exactly how many sessions are required.

The point is that we have to start from, okay, Nolan got these results using these parameters, now where do we go from here? The problem has been that there really is not tremendously good funding to take this to the next level to see whether we can improve the treatment outcomes or are we exceeding what we need to do in terms of how much stimulation we're providing?

Carpenter: That's sad because these are incredibly important clinical questions at the level of patient care.

Strakowski: They are. Depression is heterogeneous, so it's likely in the end that not all depression runs through the same neural pathways. We're going to have to think about that too.

Has there been additional work longitudinally yet to know… So you get better with SNT, then what? That's the old ECT question that's haunted us now for the history of electroconvulsive therapy (ECT). Do we have findings yet on intermittent treatments like antidepressants?

Williams: I can comment on some of the points that were made by Scott, Linda, and then that as well. We have a couple of really interesting neuroimaging papers that seem to discern SNT response of individuals pretreatment, so that's the tail end and to be determined.

We agree that not everyone's going to be the same. There's clearly a signal in the people that seem to be SNT-responsive on our predictive imaging work. There isn't enough funding generally to cover this sort of work, and I totally agree with what Linda and Scott said.

I will say that NIMH has been supportive of this work and we actually, to your question around dosing, have an ongoing study right now recruiting participants where we are scanning people every single day to try to capture effects and clinical scores every day so we can try to get a sense of this, which I think is great. It's a very involved study, as you can imagine.

You're spot on about the observation around ECT. As you know, if you look at Harold Sackeim's work, 25% of people will relapse within the first week after an acute ECT course with treatment-resistant depression if not provided with maintenance. Most people will relapse at the 6-month mark without maintenance. It seems to probably be a phenomenon of depression, from my perspective, more than actually a device or a treatment-specific thing. There are just people with different relapse trajectories.

That seems to be true with SNT as well. If you look at our single-subject data, there are a couple of people who have already relapsed by the 1-month mark. I can't give you details, but I can tell you as a general statement that it appears to be totally dose responsive. If you give people enough, they will maintain it out as long as you go, at least to the 1-year mark.

It's really a dosing phenomenon, which is great. In the acutely responsive individuals, you can kind of keep it going. That's encouraging. It tells us that if that system is responsive and that target is the right target for that individual, then it's really a dosing phenomenon, just like with ECT. There is a large amount of work to do in that domain, in particular. Hopefully, over the next couple of years, we'll have better answers to that.

Carpenter: Nolan brings up a really good point. I tell everybody we have no permanent cure for this kind of depression. It's highly variable when people will relapse again after they have a successful course of TMS. One of the things I know Nolan's group is working on is trying to jump in there and intervene.

This is something that I've been interested in my own research, which is how to do that maintenance. It's probably a very different schedule for different people. Some people have better plasticity, they stay better longer, and others, as you said, will have a relapse trajectory where they're getting sick shortly after that final treatment session. Nolan's work is another opportunity for us to tackle this really important question about how to keep people better with TMS after we get them better initially.

Strakowski: It's not a unique disease, right? We have experience with most psychiatric treatments when we get early responses and then have problems over time.

The final thing we can talk about and then we'll wind it up is, for the people watching this, when can we expect that this will become available enough that it would be practically part of our treatment armamentarium? I know that's not true now. Do we have any sense of that or a prediction?

Williams: Definitely. Some of my students are now in a company called Magnus Medical — most of the first authors on that last paper. They're working with industry folks, trying to figure out a way to get this out. From what I understand, over the next year, there will be a SNT-specific device that's really optimized to deliver this approach.

The game plan is to get into these high-acuity settings where long length of stay is tricky, such as where having folks stay suicidal in the hospital for a long period of time is really tough on the patient and tough on the healthcare system, and then trying to find ways like this to optimize it. I think that's what the next year will look like. Hopefully, the ball continues rolling for more areas to be affected.

Strakowski: From a safety profile so far, things look pretty good.

Williams: Yes. To date, knock on wood, we've never had a seizure from SNT. There's never been anyone receiving this approach in my domain and, as far as I know, in the company's domain, where there has been a seizure, which is the one risk that you think about with conventional TMS although it's really rare. It's a 1 in 30,000 session time frame. The real side effect that people talk about is headache, so Tylenol-responsive headache.

I would say, as a generality — I'd love to hear what Scott and Linda think about this — my view of TMS just generally — conventional TMS, SNT, whatever it is — is that it's one of the safest devices in medicine. If you really look at the safety profile of TMS, there are very few things in all of medicine that have a safer safety profile than TMS. I think that's really great.

With conventional TMS, I've talked to thousands of people at this point. They'll either say, "Doc, I feel back to my old self," or "I feel halfway there," or "I don't feel any different." You never hear anybody say, "I feel better, but I feel weird," or "I feel better, but I don't feel myself." That's not really something I've heard anybody say with SNT, conventional TMS, or whatever it is, which is really nice to be able to tell patients.

Strakowski: Any other comments, Linda or Scott, you'd like to add as we wind this up? It's been a very interesting conversation. I know these are brief, but anything else you'd like to add?

Aaronson: I just want to underscore what Nolan is saying. Part of the magic of neurostimulation other than ECT — most of the side effect profile for ECT is the fact that you've had a seizure and you're getting anesthetic agents — is with TMS or with VNS, there really is not. We're not doing any systemic type changes, and largely, the effect of medications and the side effect of medications is hitting serotonergic neurons that are in the brain, so we're able to target our treatment. In fact, people tolerate TMS incredibly well. As somebody who specialized my whole career in treatment-resistant mood disorders, TMS is about the most benign thing I do.

Carpenter: Yes, if only if we could get it to be a little bit more convenient so more people could access it. There are so many people who could benefit that don't know about it, aren't referred, or don't have access to it. Nolan's work and other work that people are doing to make it more convenient and more accessible is really important.

Strakowski: I think all of you have been pioneers in this, and I appreciate the work you're doing. It just strikes me, once again, the stigma around this condition. Nolan, you hinted at this but didn't really say it. If another branch of medicine had a very safe technology that would dramatically improve outcomes or even just improve some people, it would move along faster.

Part of the things we can all do together is continue to advocate for better care for our patients. We're hopeful that as this continues to roll out and move TMS forward into new and innovative approaches, that's exactly what will happen.

Well, thank you all. This was very interesting. I've learned a lot and I hope that people who are tuned into or read this find it helpful.

The take-home message is that it's still evolving, we're still learning, and we're still trying to get it available. There's promise in these neuromodulatory techniques that we hope continue to advance how we benefit the lives of people with mental illness.

Thank you.

Aaronson: Thanks for having us.

Williams: Thank you.

Follow Medscape on Facebook, Twitter, Instagram, and YouTube


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.