Dr. Martijn Figee on the Psychiatry-Neurosurgery Collaboration that is Changing Everything
Deep Brain Stimulation comes of age.
Yesterday, I attended a talk by Dr. Martijn Figee, MD, PhD, a psychiatrist and Associate Professor of Psychiatry, Neurosurgery, Neurology, and Neuroscience at Mount Sinai, and director of the Mount Sinai Interventional Psychiatry Program.
For those not in academia, that is a lot of departments in which to be a professor.
Regular readers will recognize his name. I’ve written about him before, first in my overview of DBS for OCD, where I wrote about heading to a consultation with one of our patients. Dr. Figee’s talk yesterday was to an audience of other psychiatrists…welcome to the inside of our conversations, outside world!
In the recent post featuring Helen Mayberg, when she came to Mount Sinai, she asked for specific things — a lab embedded in neurosurgery, adjacent offices, automatic collisions — and one of the people she had the pleasure of colliding with was Martijn. Figee’s talk, it turned out, filled in a lot of the science behind what I watched happen.
He works at the Nash Family Center for Advanced Circuit Therapeutics, where his team has built one of the largest DBS-for-psychiatry programs in the world. He trained in Amsterdam, where he helped pioneer the first application of fMRI in DBS-implanted individuals, before moving to New York eight years ago. He has 130 published papers. He is also, it turned out, a genuinely compelling speaker. What follows is my attempt to capture the talk, mostly in his own words.
Dr. Figee opened with a frame that set the tone for everything that followed:
“I think it’s a special, very interesting collaboration. And I also feel that it opens up a more brain-oriented psychiatry — where if you want to know something, if you want to do surgery, you’ve got to think about the brain very precisely. Most of medicine emphasizes voucher diagnostics, but I think it’s an interesting lesson for all of us, trying to be a little more precise in the way we think about it.”
He described his team at Mount Sinai — psychiatrists, neurologists, neurosurgeons, a psychologist, a movement disorders neurologist, and biomedical engineers. What does that composition do for how they think?:
“We didn’t see the non-disciplinarity as a problem. I think most of these conditions, they rely on circuits. And I think that’s just the importance — from the outside, you also see that you need a whole village of people for this.”
This is the culture Mayberg described when she came to Sinai. Collaboration is structural, both in the brain and in the center that studies it.
The Circuit Logic
The theme of his talk was to map psychiatric symptoms onto the topography of the basal ganglia-thalamocortical loops. I’ve written before about how neurology and neurosurgery moved into circuit-based understanding of their disorders while psychiatry continued to futz around with brain chemistry. Figee’s talk was a call for psychiatry to get with the program.
He walked us through it:
“You can conceptualize these as integrating actions with emotions and cognitions. You can separate those in a hierarchical way. You have the motor layers. The motor basal ganglia area, the motor cortex, the more cognitive ones, and the more emotional ones. They’re separated topographically.”
And then the clinical implication:
“You can actually target specific parts of these circuits. So, for example, Parkinson’s has all of the above, and you can basically choose the target based on whether your patient has more limbic involvement or more motor involvement.”
He used a slide overlaying functional resting-state networks onto this structural model, and explained why that distinction matters for choosing your tool:
“This is not typically used for neurosurgical targeting, because that’s more variable. When you want to place a lead into a specific structure, you want to go after the hardwiring (Ed: the white matter tracts); you don’t necessarily go after functional fluctuation. But for TMS, this is something we do use more often.”
TMS: Choosing Your Target
On TMS targeting, he encouraged the audience to think in terms of symptom profiles rather than diagnoses:
“You can dig through the lens of the brain and think about which brain surface you think your patients, regardless of which specific illness they have, will be best served by. Maybe a combination of those. That’s how we conceptualize it.”
He described the motor-cognitive-limbic gradient along the cortex:
“TMS to the motor cortex, the pre-motor cortex, the supplemental motor area, often used in neurology. The more cognitive regions, such as the dorsolateral prefrontal cortex, are our main target for depression. We also use it for some cases with PTSD. And then the more limbic targets, such as the medial prefrontal cortex, the orbitofrontal cortex, those are typical targets for OCD, depression, and anxiety.”
When TMS Works But Doesn’t Last
One of the most practically useful moments in the talk was his description of using TMS response as a predictor for DBS candidacy:
“If you have a patient that responds very well to TMS in one of these nodes but keeps relapsing — I think that’s a good indication that they will also respond, in a more durable way, to DBS of the same circuit.”
He gave a concrete example:
“I had a patient who was responding to TMS. She needed it in every month, basically. So we transitioned her to deep brain stimulation of the same circuit, and now she doesn’t need the TMS anymore. Of course it’s invasive, but it’s a good option if TMS and other options haven’t held.”
This is the logic I referred to in my OCD piece when I ended with “stay tuned.” The question isn’t TMS or DBS. It’s what circuit, and which tool best reaches it durably for this specific person.
The Intervention Menu
He pushed back on the tendency to present treatment-resistant patients with a sense of exhausted options:
“I think it’s very hopeful for patients to come to me even with refractory depression, because you can see there are so many options still left. I think there’s still a gap between general psychiatrists and interventional psychiatrists, and that’s unnecessary.”
And he made the case that this menu should be presented explicitly:
“It’s important to let patients know that there’s a whole menu of options, which is quite extensive. I think it’s very rare to have no one responding at the end of this option list, if you follow all the steps.”
What Happens in the OR
I have been in that very Operating Room. I watched Dr. Brian Kopell place a lead while Figee’s team were a swarm of busy bees in a smaller room than would have been ideal. This is involved clinical care and science happening all at once!
So when Figee described what it means for a psychiatrist to be there, it landed differently for me than it might have for others in the room:
“You have to be in the surgery room because you’re going to need to be familiar with your patient’s symptoms in real time. You have to be there, you have to participate, just to wait for part of the procedure. If you want to check in real time whether the lead is placed in the right position, you need to talk to your person. You also, of course, need to reassure your patients.”
On what draws neurosurgeons to functional work:
“Neurosurgeons in this business are very enthusiastic. They’re used to seeing tremors go away immediately. That’s kind of why neurosurgeons who are in this business chose functional surgery. For them, it’s amazing. It’s, in some way, so much more tangible. You really see, I think, in real time, the symptoms go away once they hit the right spot. This is a very gratifying experience for the surgeon.”
And for psychiatrists:
“It can also be very fast and moving in your psychiatric position — where you hear the patient report an immediate improvement of mood or anxiety, or sometimes obsessive-compulsive symptoms, although those tend to be a little slower.”
Watching it happen is not something one forgets.
Individualized Targeting: Why Group Coordinates Aren’t Enough
This was where the science got specific and compelling. He showed data from the Amsterdam cohort — nearly 100 OCD patients — illustrating why standard coordinates fail:
“When we started in 2005, we didn’t really know where to stimulate exactly. On our way there, we had all sorts of side effects. If the placement is too limbic, your patient will become hypomanic, have difficulty sleeping, and be restless. Or too motor. That’s actually good news, right? You change settings. But it makes it a difficult experience for psychiatrists. You need to be a psychiatric specialist to know how to manage it. Neurologists also don’t know how to do it because they’re used to motor symptoms. It took us a long time to get there.”
The key insight came from looking at why placement varied so dramatically between patients:
“If you place your lead based on traditional anatomical landmarks — the way neurosurgeons started doing this — you can see that it’s a very variable way to end up in these white matter bundles. And those white matter tracts are actually your targets, not so much the gray matter of a typical MRI scan. Each patient has their own specific highway structure. That’s why we need individual targeting.”
The solution his group developed:
“We now plan based on the patient’s diffusion imaging scan — their individual white matter — and see how much it overlaps with our responder map. That’s where we place our lead, so that the neurosurgeon can get there precisely. And you can even manipulate how many bundles you’re hitting with a different spot. You can navigate the lead around on the model before we actually implant, and decide if a patient benefits from more limbic connections based on their phenotype — you can ramp that up and change our target accordingly.”
The results from the last 25 patients using this approach: an 80% response rate. In a population that had failed eight to ten prior treatments.
The Depression DBS Story
This is the section where Figee and Mayberg’s accounts overlap almost perfectly — and where I found myself connecting dots between what I’d heard at the BRaSSS meeting and what I was hearing now.
He was candid about why DBS for depression has not yet achieved the clinical traction the science warrants:
“The recent DBS for depression is still experimental, unfortunately. I think it works really well — but in the original pivotal trial, more than a decade ago, they didn’t push hard enough for surgeons to go after network precision, which is so important for psychiatric diseases. Our diseases tend to be more network diseases, which are a little more complicated than, say, a movement disorder. You have to make sure that you hit the network.”
“In retrospect, they were actually missing two of the four bundles. They were actually incomplete entirely. And then they did a randomized controlled trial comparing apples and beer, because it’s not like a drug where it works or it doesn’t. You have to target it correctly first. And they were comparing it with a sham condition. Which is why the trial was halted.”
Mayberg said something nearly identical at BRaSSS. She watched that trial get halted, watched the field publicly attribute the failure to the treatment, and spent years following the data to prove that a failed trial is not the same as a failed treatment.
The problem was never the biology. The problem was that no one verified that the lead was in the right place. You wouldn’t run a drug trial where you weren’t sure the patient actually received the drug…except we kind of also do that all the time, too.
Figee described the current 22-site U.S. trial with characteristic dry humor:
“The targeting is all being administered by our center. And if our neurosurgery team disagrees with a site’s targeting, because our imaging team can see they’re missing one or two of the four bundles, the patient won’t be randomized. They’ll be sent back, and the surgeon has got to do a better job.
You’ve never lived peak awkward till you’ve tried to tell a neurosurgeon how to do their job, I suspect:
This is not a very comfortable interaction. I’m certain they have their own way of doing this. So it’s been a little tricky, but it’s working so far.”
Schizophrenia: The Experimental Frontier
The final section described something I genuinely did not expect: DBS for schizophrenia1, targeting the substantia nigra pars reticulata, a region just below the subthalamic nucleus that connects to the prefrontal and auditory cortices.
“If you put the lead in the right position, you can influence hallucinations through the auditory cortex and basal ganglia. You can influence negative symptoms through prefrontal-limbic stimulation.”
Five patients were implanted. Good response on positive symptoms and negative symptoms at five years. And then this:
“One of these patients is completely free of hallucinations — and this happens two seconds after we turn on the device. Again, it’s like something dramatic — you can literally see the hallucinations stop. And then they return when we turn it off. I have never seen such a thing in real time. It’s really dramatic.
And with a moment of characteristic humility, he continued, after a moment of pause:
“I wish we understood why.”
He also noted a finding in Parkinson’s patients that he seemed genuinely excited about:
“If you’re a psychiatrist working with neurologists treating Parkinson’s patients, if you make sure the neurosurgeon knows that you want stimulation to also reach these more limbic projections, you may actually also be able to improve their apathy. This is a common, continued problem in Parkinson’s. It’s usually not addressed.”
Q+A: On Access
Someone in the audience asked the question that was probably on everyone’s mind:
How do we scale this?
“DBS for psychiatry didn’t take off the way it did for movement disorders for a few reasons. The movies are one reason. There’s a lot of negative imagery around brain surgery and psychiatry. ECT suffered from the same thing. Another issue is that in psychiatry, psychopharmacology, and psychotherapy are a little far away from thinking about the brain. Your medications don’t really require you to know that much about the brain. You can just use them empirically. It’s not a very specific framework.”
“But I think we’re nearing a point where DBS is becoming more precise, and as soon as it’s there, it’s going to be much easier for all of you to program. You’re just going to have to look at the patient and their symptom-specific or brain-symptom-specific profile. And then it’s defined: the right dose, the right target. You don’t have to endlessly cycle through different settings. And that’s going to meaningfully improve access.”
Q+A: The Psychotherapy Perspective
I also spoke up to highlight issues of concern to my psychodynamic colleagues:
“What DBS has done to the work we’re doing in psychodynamic psychotherapy (referring to a favorite long-term patient), and separately, group psychotherapy, is fascinating. I’d strongly consider group therapy for patients you’re referring to brain circuit interventions, because you’re going to get to do more therapeutic work with them about what their life can now include.”
Dr. Figee echoed my sentiment:
“If you send patients to a specialty center for surgery, you’re going to be part of that, always. More importantly, you’re going to work with a different person. At least in terms of that specific study of who they are, and how you can work with other treatments better. This is true whether that’s psychedelic-assisted therapy, cognitive therapy, or psychology. You’re going to like working with them.”
When I wrote the OCD piece in February, I closed by saying I was headed to a consultation with one of our mutual patients. That was the beginning of something I’ve now been following for over a year. Through that consultation, through the OR, through Mayberg’s talk at BRaSSS, and now through Figee’s talk, thanks to the NY Country Psychiatric Society? The thread running through all of it is that precision matters, circuits are real, and the menu for those suffering is longer than most of us learned.
The collaboration has already yielded one accepted publication, but you will have to join Radial at CTMSS to read more.
Dr. Figee’s research and clinical program at Mount Sinai has more info available on their website.
Cascella N, Butala AA, Mills K, Kim MJ, Salimpour Y, Wojtasievicz T, Hwang B, Cullen B, Figee M, Moran L, Lenz F, Sawa A, Schretlen DJ, Anderson W. Deep Brain Stimulation of the Substantia Nigra Pars Reticulata for Treatment-Resistant Schizophrenia: A Case Report. Biol Psychiatry. 2021 Nov 15;90(10):e57-e59. n