Hello, my name is Evan Thomas, radiation oncology resident at the University of Alabama at Birmingham. I'm going to be giving you this lecture entitled Frameless MLC-Based Linear Accelerator Radiosurgical Thalamotomies Are Safe and Effective? The Early Results of Our Prospective Phase 1-2 Clinical Trial. So our clinical trial was designed to assess the safety and the efficacy of a new technique of delivering radiosurgical thalamotomies on a linear accelerator without the use of cones and without the use of a stereotactic frame. Secondarily, we assessed the toxicity in our patients, the quality of life changes, and also the usefulness of employing new targeting methods of the ventral intermediate nucleus of the thalamus based on structural and functional connectivity information obtained from 3T and 7T MRIs and also resting state functional MRI images. So our trial was structured as thus. Patients had to have either medically refractory essential tremor or tremor-dominant Parkinson's disease. They could not be candidates for DBS or would have had to have been offered DBS and refused it. They couldn't have had a prior brain radiation and had to be adults with good performance status. These were the metrics that we followed in each of the patients. We collected their toxicity, rated their tremor based on the Fontello-Samarin scale and their quality of life by the SF36 and PROMIS indices and followed MRIs after treatment as well as patient satisfaction. So briefly, how we do functional radio surgery at our institution. We do it non-invasively without a stereotactic head frame. Patients are immobilized in this thermoplastic mask, which has an open face mask system. The exposed area here is actually tracked by an optical surface monitoring system that can detect as small as 0.1 to 2 millimeters of misalignment, which allows us the opportunity to turn off the beam if there is any sort of patient motion. Patients are delivered very quickly on the order of 45 to 60 minutes, with the shortest time for this particular procedure being under half an hour at 25 to 29 minutes. So as everyone listening to this presentation is probably aware, when a radiothoracic thalamotomy is performed, the lesion is placed at the ventral intermediate nucleus, ideally of the contralateral thalamus for the tremor limb that's being desired to be treated. The way we deliver the radiation without using cones or a gamma knife is to actually set the leaves of the collimator of our linear accelerator to a pre-prescribed 5 millimeter by 2.1 millimeter aperture. This aperture is kept the exact same throughout the treatment. And then the gantry is rotated around through all these different angles, a total of 20 for this particular dose. And the intensity of the beam is actually varied with the sign of the gantry angle to allow us to have a nice spherical dose distribution that very closely mirrors that of the gamma knife. On the right here, you can see what the dose distribution looks like for our virtual cone method on the linear accelerator. And then on the left here, you can see what it looks like for a typical gamma knife 4 millimeter shot. And you can see how very similar they appear. So our particular trial on this presentation has finished accrual of all the patients. We had actually anticipated not enrolling the full amount of our patients until late next year, but actually experienced greater than anticipated enrollment by a number of self-referrals of patients discovering this trial on their own. So about half came from our movement disorder group and about half were self-referrals that we screened through our movement disorder group and with our movement disorder neurosurgeon. So how do we do the targeting? For our trial, we started with the classical stereotactic-based reference location of the VIM. And we would denote the anterior commissure and the posterior commissure as well as the midline. And then a script would actually take these markers that we place on the scans and then do an automated shift to where the VIM was expected to be based on those stereotactic coordinates. Then we would identify the posterior limb of the internal capsule and then move the shot slightly medially to make sure that we keep the 20 percent isodose line medial to the internal capsule to avoid unnecessary risk of toxicity. These were the imaging protocols that we used to give us this information. On our 3T Siemens PRISMA scanner, we would obtain a high-resolution MPH in F-gator sequence, which is a white matter null sequence that allows us to see the posterior limb of the capsule very nicely, as well as DTI imaging, resting state fMRI imaging. And then if patients were willing, we provided funding for them to go down to Auburn University and get these same sequences on a 7TESLA scanner. So here you can see what the dosimetry of one of our plans looked like, along with the F-gator sequence that we used to do most of the targeting. You can see the internal capsule very nicely outlined here, and how we're able to keep the 20 percent isodose line just medial of it to spare this patient risk of toxicity. So here you can see what some of the results of our probabilistic tractography experience are. On the left here, the target that was determined by the stereotactic ATLAS coordinates right here. And then you can see the SMA and PMC zones, as were predicted by the probabilistic DTI, and where our treatment would have overlaid with respect to those coordinates. On the right, this is actually the functional motor network representation that's predicted based on resting state fMRI obtained on that 7T Siemens Magnetom scanner. So here's a look at the post-contrast MRI that was obtained on a patient 10 weeks post-procedure. You can see the lesion, and overlaying on it are the isodose lines associated with our treatment. So it's a very nice alignment of the lesion and where we expect it to be with where the dose was designed to be delivered. Here you can see an MRI from a patient a little bit farther out from treatment, and we end up with this nice ring-enhancing lesion where it's supposed to be right in the VIM, and you'll notice this central opacity here associated with the high-dose region of the treatment. So these are the results of our patients and how they did in terms of tremor reduction. On the top left, you hear this is the reduction in the Fontalis & Marin score by patient and over time. You can see our patients in general have most of their response in the first six months, but some of the patients continue to respond even farther out. This is the pre- and post-tremor score. We noted that there's a mean reduction of about 60% in their tremor score at six months, and here you can see the percent change in the tremor score depicted on our waterfall plot here. We have one patient here, you can see this one was at six months and he hadn't really demonstrated a response yet, but the rest of our patients had all demonstrated pretty good response, including one patient who had a near complete total reduction in their tremor. So how quickly do our patients respond? Well, we noticed that there's a wide-range median delay of three to five months, but it could be much shorter. One patient reported a one- to two-week response with 60% to 70% tremor reduction, and another reported tremor response in the 12- to 18-month range. So very briefly, I'd like to show the results for one of our later responders. You can see the evolution of the lesion from the radiosurgery on the T1 post-contrast MRI. Small lesion here at six months, larger at 12, even larger at 18. And here you can see the evolution of the edema associated with the lesion on the T2. Notice we have about maximal edema starting to emerge at 12 months, doesn't change a whole lot between 18 months. However, this particular patient really didn't have their response till 18 months, and you can see this is what her Archimedes spirals and her handwriting looked like at the beginning, and here's what they looked like at the end. This is a nice artwork representation that the patient drew us and brought us in, first time she'd been able to do art in about 20 years, so she was very pleased with the response. So to wrap it up, the virtual cone MLC-based technique for delivering radiosurgeries on a LINEQ is safe, effective, very efficient, majority of our patients treated in under 45 minutes, for delivering radiosurgical thalamotomies to tremor patients that either can't get or just don't want to get DBS. Integration of functional imaging directly into the treatment planning system, as well as a script, streamlines the process for treatment planning. Another benefit of this approach is that because it's the same exact modulation sequence for every patient, there's no patient-specific quality assurance or dosimetry needed for these patients. And based on our very encouraging results, we were able to secure additional funding for 20 more patients, so that we'll end up with a total of 40 patients prospectively assessed in this trial. Thank you very much for your time and attention. I'd like to acknowledge the other CNS radiosurgery faculty at our institution.

radiation oncology

MLC-based linear accelerator

thalamotomies

essential tremor

Parkinson's disease