I think that the talks that you've had so far on spine radiosurgery have all focused on metastatic disease. And so I'm gonna talk about a couple other of the types of pathology. You've seen the spinal cord, if you're gonna do spinal cord radiosurgery. So let's start with spinal AVMs. I think these are probably the most difficult of the spine lesions that we do radiosurgery for. Much of what you've seen so far when you treat bony metastasis of the spine is blocking out the spinal cord from receiving significant spinal cord dose. That obviously is gonna be a lot more challenging or perhaps even impossible in some cases when you're talking about spinal cord AVMs. So which ones are we talking about? We're talking about unresectable spinal cord AVMs. Those that can be resected should. We're not talking about dural fistulas. So I'll show you the classification system for these spinal lesions. And the type one dural fistulas should be treated either with surgery or endovascular procedures. Small to medium size vascular malformations, lesions with difficult operative approaches and patients with significant comorbidities are also good candidates for consideration of spine radiosurgery. So why are we even talking about spine radiosurgery for AVMs in the first place? The rationale is that these are challenging lesions. And if you look at what's in the medical literature, the published data, you can see it here. From a microsurgical standpoint, if we look at a very experienced group, you can see a complication rate of 17 to 30%. With a very experienced endovascular group, the complication rate's also about 22%. So with this type of data and with the fact that we have decades of experience with radiosurgery for brain AVMs, we decided back in the early to mid-1990s to take a look at that at our institution. What is a safe way to deliver radiosurgery for spine AVMs? So classification system, the type, there's several different spinal cord AVM classification systems. This is perhaps the most common one. So the type one spinal cord vascular malformations, the AV fistula, it's the most common type. It's a fistula right at the nerve root sleeve, and I'll show you a picture of that. It's typically treated with surgery or invascular therapy, very high cure rates and typical ages in the 40 to 60 range. The type two are the intramedullary spine AVMs. This is the type that we typically do radiosurgery for. These are compact niduses. They usually present as a young adult. Hemorrhage is the most common presentation causing neurologic deficits with many patients presenting with subsequent multiple hemorrhages. Type three, we typically don't use radiosurgery for, and I'll show you a picture of why. These are juvenile vascular malformations. They extend well beyond the spinal canal into the bone and to the paraspinous tissue. It would be a large target if you're trying to treat the whole lesion. We typically don't do that. And number four is a very, very rare vascular malformation to perimedullary or peel fistula, and I'll show you what one of those looks like. So here's the pictures. This is a type four. This is a direct peel fistula from artery to vein right at the surface there. It's not gonna be a radiosurgery target. There's not really a nidus to target. This is the type one spinal cord vascular malformation right at the nerve root sleeve. This is very easily treated either with surgery right in this location or endovascular technique. And both of these are not typically radiosurgery targets. So this is the type two, the compact intermedullary AVM here. And this is what we would consider a good radiosurgical target. Again, there are a lot of people that'll say, well, should you try to resect this? And I'll show you in our data at our institution, we've had of the AVMs that we did do radiosurgery on, 50% had prior treatments, either surgery or embolization. So it's not that we're saying don't resect this if you can, but we're just saying if you cannot resect this, or you can do only a partial resection and there's residual, consider radiosurgery. This is the type three, the juvenile AVMs. So you can see they extend well beyond the spinal canal. And they're not really something that you're gonna be able to target very well. So when we started doing this spine radiosurgery in the mid 1990s, we were very fearful of the dose of radiation to the spinal cord. We are talking about spinal cords that are not normal. If you've had a hemorrhage from your AVM, oftentimes there's myelomalacia in the cord. So you've heard other speakers talk about what's the dose tolerance to the normal spinal cord. In this case, the spinal cord's not even normal. So even if you can come up with a reasonable dose tolerance to a normal cord, how much are you gonna reduce that tolerance in a cord that's not normal? So we started using fractionated regimens. In this case, four times five and a quarter gray for 21 gray. And then you can see over the subsequent years, we started escalating the doses. More recently, we've been in either 16 to 18 gray in a single shot or 20 gray delivered in two 10 gray sessions. And you can see the equivalent radiobiologic dose equivalent for conventional fractionated regimens. So a couple of examples. So here's a AVM right there. And the cord is pushed off to the side as a crescent. And this was treated with 16 gray in a single session. So again, the parameters that you saw when you're dealing with metastatic diseases are completely different here. And you can see in some of the AVMs, basically you're targeting the whole cord. You cannot separate cord from AVM. So if you're gonna deliver 20 gray to the AVM, you're gonna deliver 20 gray to the cord. Here's a couple of examples of the radiographic results. This is the AVM on the MRI scan. Here's an angiogram of the AVM. Here's how we target the AVM. The volume of the AVM was 2.3 CCs. And this was treated with 20 gray in two sessions. And again, you're pretty much targeting the cord. There's no separate normal cord that you're gonna get in there. Initial AVM here in the cervical spine. After two years, you see marked reduction. And after three years, the AVM's completely gone. So the time course is similar to brain AVMs. Here's one that occurred over about two years. This is down at the CONUS. This was treated with 16 gray in a single shot. Pretty small volume. And two years later, the vascular malformations obliterated. So this is an interesting slide. This is the first 29 patients that we treated. The bars show the length of time the patient had their AVM diagnosed prior to the zero point time, which is their date of radiosurgical treatment. Each one of the red triangles is a hemorrhage. And you can see some of these patients have hemorrhaged multiple times. Some multiple times very close together. And then we treated the patient with radiosurgery. Subsequent to the treatment, we've had actually no bleeds. In 30 patients. And we're up now to about 158 patient years. I'll show you the obliteration rates. They're not high compared to brain AVMs. But we do seem to be changing some of the natural history when it comes to the AVM hemorrhage rates. So prior to the treatment of the AVM in the spine, hemorrhage rate is 3% a year. That's remarkably similar to what we quote patients for brain AVMs. Here's our clinical results. We had 30 patients treated since 1997 up through a few years ago. The average follow-up was about five years. 57% of the patients had prior hemorrhage. The rest presented with progressive neurologic symptoms, but did not appear to have an active hemorrhage or evidence of prior hemorrhage. That could be from steel or from just other types of compression syndromes in the cord. We had 50 patients had prior surgery or embolization and there was still residual vascular malformation, so we weren't able to treat the lesion completely with those modalities. Of the patients that had greater than three years follow-up, only 27% had complete obliteration. But another 64% had a significant decrease in the size of their AVM. And you might say, well, this is low compared to what you're achieving in the brain. But two things. One is, remember, we started out with low doses because we're fearful of injuring a cord that's not normal. But secondly, interestingly enough, for the 158 patient years so far since the treatment, we've had no hemorrhages despite the 3% per year hemorrhage rate that we were noticing beforehand and no patient mortality. In terms of symptom improvement shown here, patients either were improved or showed no change. There were very few patients that actually worsened as a result of this treatment. When you looked at the modified Rankin scores or the Karnafsky performance scores, no significant change pre- and post-treatment. So in terms of complications, no mortality, no new hemorrhages. We did have three cases of worsening symptoms in that those are patients typically that had edema in the spinal cord. We had one case of radiation-induced myelopathy. So overall complication rate in our series was about 10%, so three divided by 30, and then 3% rate of radiation-induced myelopathy. So I think those are pretty acceptable numbers. So our current dose regimen is 18 to 20 gray to the nidus, I'm sorry, 20 gray in two sessions, or 16 to 18 gray if it's a small AVM nidus. And we're still evolving the dosing schemes, but we think we're up to reasonable numbers. And so hopefully going forward, our obliteration rate will go up. I wanna talk a little bit about spinal cord tumors for the second half of this talk. So these are gonna be schwannomas and meningiomas for the large part. And surgery is going to be the primary treatment for these tumors, particularly if there's solitary tumors showing up for the first time, no pathology. Surgery's also gonna be the treatment if there's spinal cord compression, you don't know the pathology, or if there's significant neurologic deficits. But there are gonna be a subset of patients in which radiosurgery could be useful, and these are, the reason why these benign tumors are often good radiosurgical targets is they're so well-defined. You know, when you give contrast for a meningioma or a schwannoma, you can see it clearly. You're not looking at some of those bony metastases where you're trying to identify the GTV. I mean, you're gonna see the edge of the tumor. So if they're relatively small, there's no cord compression, or they're in tough locations, or the patient has comorbidities, they can be good targets for radiosurgery. So this is a typical example. This patient has a thoracic schwannoma. She's got NF2. There's multiple lesions. She's had prior surgery. She didn't want to have another surgery, so if you're considering treating that target, that might be a reasonable option. We've published our data several times on this. This is one of our more recent publications from several years ago, talking about long-term good results with spine radiosurgery for these benign tumors. We had over 100 patients here. So here's the breakdown here. More meningiomas and schwannomas and neurofibromas. We'll talk a little bit about that. The mean tumor volume typically is not too large, unlike some of these bony metastases that can be over multiple levels. In these cases, we're talking about relatively small tumors. The doses are comparable to doses used in the head, and in some cases, it may look a little bit higher, but some of these are fractionated regimens, and I'll show you a table that breaks down the number of fractions. A couple of slides on just patient characteristics. NF patients were 25% of the population there, equal breakdown between male and female. Lesions, more than half were located in the cervical spine. The rest, as shown here, histology, we talked a little bit about, mostly meningiomas and schwannomas. So in terms of dosimetry characteristics, you can see it's shown here. We broke it down by meningiomas, neurofibromas, and schwannomas. The prescribed dose in kind of the 18 to 20 grade, but again, that's over two sessions, so if you convert that to a single session dose, you're gonna be probably closer to 15 or so. Meningiomas tended to be a little bit smaller than the neurofibromas and schwannomas. This table shows kind of the dose fractionated regimen that we use for the various lesions. When possible, we try to treat in a single session, so look for the schwannomas, most of them pretty small going out the nerve root. A single session of 18 grade was used in many of them, but if the lesion was expanding over several levels, then we would start to fractionate treatment. Presenting symptoms, this is interesting. So as with metastatic disease, you would expect pain, sensory deficits, and motor deficits, and that's what we saw with these benign spine tumors. Pain was the most predominant symptom. There were 25% of patients that had asymptomatic tumors, and you might say, well, why are we treating those tumors? And these are patients that typically had tumors that were increasing in size radiographically, and we were following them, perhaps in an NF patient, and you have several scans in a row that show progression of tumors, so you know that if the tumor keeps growing, they're gonna become symptomatic. Unlike the metastatic disease, which you get very rapid response within two weeks, for example, for pain, when you're dealing with a benign spine tumor, these tumors take several years to die, to respond to the treatment, so you cannot wait till the patient's having a lot of symptoms from a tumor and then run in and treat it with radiosurgery and expect the same rapid response that you're gonna get with metastatic disease. So that's why in these patients, you need to have a discussion with the patient that if they want to keep following the tumor that's growing and wait till they have significant symptoms, they may be looking at a surgery rather than a radiosurgery. So we had, average follow-up is about three years. The vast majority of the tumors actually were either stable or decreased in size, so you're getting very good control rates. We had one patient that had tumor progression, and interestingly, that was quite far out from treatment. Stable or improved clinical symptoms in the majority of patients here, and we had one patient which developed a transient myelopathy at nine months from radiation treatment, and I'll show you that picture. So here's one patient before I show you the myelopathy picture. This patient had prior treatment. You can see loss of the posterior elements here from surgery, had a recurrent spinal schwannoma at T7, and then over radiosurgical treatment, done, you can see the follow-up period shown there, so good response over time. Here's another patient. This patient has a C7T1 schwannoma anterior to the cord and was treated with radiosurgery, and you can see a good response over time. So the pain relief after radiosurgery for these benign tumors is not quite known. It seems to be pretty good for meningiomas and schwannomas, but does not seem to be in the same ballpark with neurofibromas, so even though with neurofibromas you're getting control, the tumor's not growing, you're not getting a clinical response, whereas with meningiomas and schwannomas, we seem to be getting better clinical responses. This pain relief occurs in absence of tumor shrinkage, so just because your tumor decreases in size does not mean you're gonna have a better pain response than a patient whose tumor's just controlled. There was no correlation between pain relief and the number of doses or stages of radiation treatments used. The time to onset of pain relief was a few weeks to months. We have some patients presenting with pain relief at six months, which is longer than you would for metastatic disease. Here's the one case of radiation-induced myelopathy after radiosurgery. It's a residual meningioma. There's prior evidence of surgery done there, and here's the target. You can see here's the various beams coming in, and again, the target there prior to treatment, and post-treatment, you can see a T2 signal change in the cord in this patient. This was about nine months after treatment. Just gonna conclude with two examples where we try to potentially combine surgery and radiation. We've talked about that for metastatic disease, so let's talk about that in a couple of cases with a benign tumor. This patient was starting to get some symptoms from cord compression, but really wanted a relatively limited surgery if possible. She was an older patient, so what we did is we decompressed the spinal cord with surgery, and then we did radiosurgery to the residual, and she did quite well. This is another patient where the exact opposite happened. It's a 42-year-old female. She has slight right leg pain. She's an Asian woman, and you can see this schwannoma sitting right there on the nerve root at the L1 level, and I recommended surgery for her because of the pain that she was experiencing, but the more I recommended surgery, the less her pain became, so I think she was kind of under-describing her pain, and I think she was very fearful of surgery, so by the time we had multiple clinic visits and discussions, her pain was down to one or two out of 10 when talking about the surgical option, so we ended up doing radiosurgery on her, and then about a month and a half later, she now comes in. She now says, well, actually, the pain was worse than I admitted prior to surgery, and after the radiosurgery, now it's even worse, so I went from eight out of 10 to Dr. Wu's 50 out of 10 patient, so we ended up having to go in there and debulk the tumor. Now, the tumor had already been treated with radiosurgery, so there wasn't necessarily a rationale to try to get every little bit of the tumor out. It should respond very well to the prior treatment, but we wanted to decompress the nerve roots that were the cause of the pain, so that's what we did. It was a pretty simple surgery, took out the part of the tumor we could easily visualize, did not have to chase out the neuroframe, and the patient woke up with a dramatic improvement in her pain, and it's done well, and now she's about two years out. So, radiosurgery for these benign tumors is feasible and safe, very high control rates, similar to that, what we see in the brain radiosurgery for these benign tumors. The optimal dose fraction and regimen is evolving, but seems to be very consistent with those used for schwannomas and meningiomas in the brain. Pain relief is achievable, particularly if the patient has meningiomas and schwannomas, not so much so with neurofibromas, and with these benign tumors, the follow-up period is gonna be important, and when you're talking about three, four years, that's probably still intermediate follow-up. That's not long-term follow-up when we're dealing with these benign tumors, so unlike the metastatic disease where you can see good results because the patient's survival is not that long, you can see good results in a short period of time. For these, you really gotta follow them up for longer periods of time. Thank you for your attention. Thank you.

spine radiosurgery

spinal arteriovenous malformations

radiosurgery for AVMs

spinal AVM classification

radiosurgery for spinal cord tumors

spine radiosurgery for benign tumors