The Current State of Focused Ultrasound Indications: Essential Tremor, Parkinson's & Bone Metastases

[Dr. Jacob Fleming]
You mentioned a couple of different clinical scenarios, essential tremor, Alzheimer's, are those the primary clinical issues we're seeing? What kind of clinical issues are you seeing in the clinic in treating these patients?

[Dr. Bhavya Shah]
The focused ultrasound technology or high-intensity focused ultrasound is currently FDA-approved for a couple of things, essential tremor, tremor-dominant Parkinson's disease. Those are patients who have a very specific type of Parkinson's disease in which their main symptom is tremor. Number three is also other motor symptoms in the setting of Parkinson's disease such as bradykinesia, rigidity, dyskinesias, things like that. Depending on what you're doing, the target is different. For essential tremor and tremor-dominant Parkinson's disease, the target has historically been considered the ventral intermediate nucleus of the thalamus, or some people, for the tremor-dominant Parkinson's disease, would consider even the ventral oralis posterior nucleus.

For Parkinson's disease, the target in the United States as FDA-approved is the globus pallidus internus, which is in contrast to some of the clinical trials that are coming out of Europe, in which they're targeting the subthalamic nucleus. Just, I think, last week there was a paper published in the New England Journal of Medicine looking at HIFU pallidotomy in the United States, multi-site clinical evaluation of how well it worked. Similarly, in the Journal of Neurology, maybe two or three weeks ago, the Europeans published their experience with the STN. I think without really getting into the nitty gritty, it's pretty obvious that the reductions in these motor symptoms are not as great as people had initially anticipated, although there is some effect and there is still a pretty significant adverse effect profile.

[Dr. Jacob Fleming]
Interesting. I understand that part of the reason that speaking to one of our colleagues, Fabricio Feltrin, shout-out to Fabricio. He's a great fellow, research fellow, and now attending at UT Southwestern. He's been very involved with this as well. He says part of the reason that your patients have such good improvement in a lot of cases has to do with the workup and using the MR tractography. Can you talk about that whole part of the pre-procedure aspect?

[Dr. Bhavya Shah]
Yes, absolutely. A shout-out to Fabricio as well. He's probably one of the best fellows I've ever worked with and he's a fantastic attending.

There's so many different ways to target the part of the brain that you're talking about. Let's talk about essential tremor because I think that's going to be the most straightforward way of thinking about this. The ventral intermediate nucleus is the target, but people can't see it on any kind of high-resolution imaging so they're relying on landmark or indirect-based methods. They identify key structures in their brain, like the anterior commissure and posterior commissure, and they make measurements from this and say, approximately, this is where they think the target is.

The great thing about focused ultrasound is before you treat, meaning, that before you ablate a part of the brain, you can use low-intensity focused ultrasound to stimulate what you think is the target. Then if you go and examine the patient in real-time, you can see that their tremor is getting better and you can also ask if they're having any side effects. If their tremor is getting better and they're not having side effects, it means you're in a great spot. You may not be in a great spot and patient's tremor might not have gotten better or they might have some side effects, in which case, you're going to have to move and search around looking for the target.

To get around this, we use this method called four-tract tractography that I developed here. Basically, we started in cadavers, maybe when I was recruited, so four or five years ago, that we obtained through the World Body Program. These people had passed, but they had donated their bodies to science. We were able to acquire them right after they'd passed so no fixation, no formaldehyde, nothing like that. We would get them on the scanner in the middle of the night and we'd run really high-resolution, 32, 64, 128 direction DTI, advanced structural imaging, such as this Fast Gray Acquisition T1 Inversion Recovery, and then we would process these different tractography things.

Meanwhile, we would take the patient's brain and we would put it in a 3D-printed slicing guide. The slicing guide, the whole purpose of it was to hold the brain like it is held anatomically, but in a specific plane known as the anterior commissure, posterior commissure plane. We would then slice through the brain at two-millimeter sections from the front to the back and then register these images of the MRI back to the path using block-face photography. Then we would dissect down in micron sections and identify white matter tracts so we're entering and leaving the thalamus.

We were looking at a couple of things. Number one, anytime you say DTI out loud, someone says distortion, huge gradients, lots of distortion. We're in the central part of the brain so we wanted to make sure that that wasn't going to be the case. Number two, people talk about biomarkers all the time. Imaging biomarkers are one of those things where just because you can identify it, is it really there? We wanted to show that, "Yes, it's really an anatomic structure that you can identify and it's pretty high likely what you're looking at.

The third thing we want to do, the reason we ran so many DTI scans, was because we wanted to use something that's clinically relevant. Radiology, sometimes we get really excited that we can make pretty pictures and then we have 256 directions, but no clinical protocol can support that. We wanted to drive it down to as little as possible so we were using 32 directions. At the same time this was going on, people in functional neurosurgery, neurosurgeons were reporting that the improvement of the patient's tremor, and essential tremor, didn't really matter how close you were to the VIM. What really mattered was how close you were to the white matter tract, the dentato-rubro-thalamic tract.

What tract they were looking at and what tract you should prospectively target, nobody really seemed to understand that. That's what we did. We came back and said, "Look, there are two components to the DRTT, a decassating and a non-decassating component. From histology and pathology and all the anatomic studies that have ever been done, we know that the decassating DRTT goes into two nuclei, the VOP and the VIM, and the non-decassating only goes into the VIM. If you're targeting at the posterior margin of those two, you're pretty much hitting VIM.

We did a lot of this preliminary work in cadavers. It really proved that what we were looking at was what we could see, and then it became translating into FDA-approved software for treatment. We started using Brainlab and processing this. Ever since then, we've been using this approach four-tract tractography to identify the targets, the decassating and the non-decassating DRTT, and also, identify tracts that we want to avoid, the medial lemniscus, which is primary sensory for the whole body, and also, the corticospinal tract, which is your primary motor tract. You want to stay away from those. That's how we've developed our protocols.

A lot of people tell me, "we've done tractography and it doesn't work." I think that that is a fair criticism because it depends on who's doing your tractography. One of my friends, he's a neurosurgeon, and he's like, "We gave the radiologist the protocol and they did it and they made the tracts." I would say to that, that's probably not going to work even without seeing the tracts because if you're the treatment physician, you should be making your tracts and looking at them yourself and making sure they make sense.

[Dr. Jacob Fleming]
It's definitely one of the advantages of you being the diagnostician and the treater.

[Dr. Bhavya Shah]
Yes, absolutely, and I think that's important. The other thing is, I think that protocols vary greatly and having a really good understanding of the anatomy is really essential. That being said, we've expanded. We've shared our approach with UCSF, Mayo Clinic in Rochester, University of Colorado, and Duke, and they seem to be getting the same tracts that we're getting, and the same responses so I think it's working out.