Optimization of Non-Coplanar Delivery in SGRT for Complex Thoracic Lesions- A Case Study in MapRT use for OAR Dose Reduction

So I have to admit, the title is a little bit different. It’s actually “Using MapRT to Improve the SGRT Efficacy and OAR Reduction.” So it’s a two-pronged approach here. That’s just a picture of where I live, California. It’s beautiful. So just a quick outline of the talk.

To give you a little bit of background of myself, if anybody was paying attention at yesterday’s talk, I was the one that was asking about the faster-than-light travel. That’s because my background is actually in particle physics, so I transitioned into medical physics. And actually did that in a postdoc at San Diego, and I remember sitting in San Diego looking at a patient with a large brain lesion and just taking a step back. And I think that that’s helpful for us because we’re all professionals, and I should get close to the microphone so you can hear me. So take a metaphorical step back. Realizing that we use this technology all the time, and it’s very familiar to us. But maybe it is not for certain people, and I like the two examples that we had yesterday about caretakers that actually had cancer and had went through it, and they were looking at their plans. They were saying, “Okay, let’s push this dose down.” So we can always push these dose down. We can use the surface guidance to make sure that we’re actually targeting where we want to target. So just a little bit of an interlude there, so sorry about that. I will be getting into the actual meat of the talk. So before I start, who here is familiar with SFRT, so spatially fractionated radiation therapy? Okay, so more people than I thought. Okay, that’s great.

So just talking about SFRT. So we put high dose shots inside of a larger volume that’s treated to a uniform dose. And I want to just talk about the rationale of why we do these high dose treatments. So I’m kind of putting on a bunch of different hats, so I apologize for that. So we do know that dose escalation is helpful in a number of different cancer histologies. And this has been apparent with Gamma Knife technology. So I actually started with doing Gamma Knife planning, and we always used dose escalation, but now it’s being more widespread, and SFRT is essentially one of these kind of techniques.

In healthy tissue, we have vasculature that follows regular patterns. So I did a postdoc in imaging, and actually, the vasculature follows a fractal-esque pattern, which is very, very fascinating. But when you get to tumor tissue, it is very different. And we definitely know that tumor tissue is highly metabolic, so it needs to have oxygen. So we can actually think about the vasculature as a target when we are targeting the tumor. Because if we can kill the vasculature, then we can inadvertently kill the tumor. So there’s a bunch of different studies showing that if you go above a certain dose, you’re actually targeting the vasculature, and you have a sustained vascular kill. So here is another graph showing different fractionation schemes. So I think 20Gy in one. So you see a more sustained killing than a lower fractionation. So this has been shown in a bunch of different studies here.

So there are some other factors that have been hypothesized that lead to this increased cell death post-radiation, and I’m not a biologist, so I’m not going to really get into this. But again, I’m showing this picture showing that if you have a higher dose per fraction, you’re having a more sustained kill. So essentially, I’m trying to build up the rationale for this SFRT treatment.

So this is a little bit of a not problematic, but maybe controversial slide. But there’s been a lot of talk about how you can actually prime the immune system with these high doses of radiation. So now I’m not talking about the abscopal effect, where you’re trying to influence something away from where you’re actually targeting. I’m talking about inside the actual target that you are targeting here. So again, so sorry about not understanding this technology here. So here, just showing that if you’re giving a single dose of 15 gray radiation versus a more fractionated approach versus no radiation. So you’re seeing that we’re using the mean thigh diameter as a surrogate for tumor growth, and we’re actually having a sustained non-growth of the tumor compared to a different fractionation scheme and no radiation. So again, not a biologist, but there has been some studies in this paper and other papers showing that these high doses of radiation actually prime the immune system, and you actually have an effect. Okay, so hopefully I gave you enough rationale for SFRT.

Okay, there’s a timer right there, making sure I’m on time. So now I just want to introduce the technique for people that are not as familiar with it. So there are two different forms of SFRT. So we have a Grid, which is here, which is essentially a 2D treatment approach, and then we have Lattice, which is a three-dimensional treatment approach. And you can imagine that in these situations, you have to have a robust planning sequence. This is essentially created with an aperture that can go into the treatment head. So there’s not so much of a planning approach. You can actually do this with MLCs, but it’s a little bit more rigorous. This is an older technique. SFRT, the first studies were introduced in the ’80s, I think it’s like ’85 and ’87. So this is definitely technology that’s been around for a while, but I think in the physics field for sure is that if you actually want to make an advancement in the field, all you have to do is look into the past where people came up with these ideas and didn’t have the technology to be able to do what they wanted to do here.

Okay, so here is another, just showing you some clinical results. So here’s a study where they actually found that using the Grid approach, they had a better partial complete response than a RTOG study. So obviously, most of the time when we’re treating these large lesions, or I should say, treating things with SFRT, they’re very large lesions, so the patients don’t have a very good outcome. But using this Grid approach, they were actually able to get something better than with conventional radiation and chemotherapy. But again, I have to emphasize that these are very large tumors. So I mean, as a physicist, I can always tell who wrote this paper because this is a linear dimension, even though this is three-dimensional, so this is an MD paper.

So if we go to another study here, this was put out by WashU, which definitely has put out a lot of studies. So, very small number of patients, very large lesions that we’re seeing. We’re seeing on the order of several hundreds of cc’s. So I mean, this is very large. Not too much toxicity in this small cohort, and it was just because this patient was treated with two very large lesions. And we’re seeing that using this technique, granted a short follow-up, they had significant shrinkage of your GTV.

So for talking more about the logistics of the treatment, inside this, we have these high dose spheres that are placed in a regular pattern inside of this larger target. So you actually have two targets inside of your GTV or different kind of nomenclature. Sometimes you say the GTV is the low dose target, and then you have the PTV being the high dose target, but just know that there is a low dose target and a high dose target. So there’s different fractionations. Sometimes people do one fraction of 15. WashU uses a five fraction technique where they’re going up to 66.7, and then the base dose is 20 gray, right? And there’s a bunch of clinical trials. These initial results led to some clinical trials, so I definitely encourage people to look at this technology.

So because I did some brain planning in the past, when you have this complicated geometry like this, you have to worry about the dose that you’re having be in between the lesions. So one could say, “Well, we don’t really care because as long as we keep our low dose outside of the non-target, who cares what we’re doing on the inside of the target?” And what I want to show and what I want to communicate to you is that is definitely not necessarily the case because going back to the brain case is that, sure, if we look just at our prescription doses, everything looks just fine and comparable between the two. I think somebody was talking about that yesterday, saying this like, “When people do plan reviews, the only thing you do is about 95% of the prescription, and if that looks okay, then who cares about anything else?” But it really is the case that you have to look beyond just your prescription. And the Single Isocenter Multi-Met has really, really shown this, and there are multiple different things that you have to optimize when you are planning this. And the suite of LinearRT, and then specifically MapRT, allows us to optimize these things. So that’s what I’m going to be talking about in these two case studies that I want to get to. And then hopefully I can end a little bit early and then leave some time for questions.

So here is a—All cases that we treat are very unfortunate, but this is an unfortunate sarcoma of the heart with pulmonary metastatic spread. Very young patient. The GTV was on the order of 1,300. The PTV, the high-dose spheres, you can see them here in the red, and then I wanted to show the three-dimensional view of that, was about 16cc. So you can use different diameter spheres, you can use different spacing of the spheres. That has not been really set yet. So I would encourage people that are wanting to get into this to be able to look into the literature and find out what works best for you. This particular prescription was 50 in five fractions to the PTV, and 20 in five to the GTV. Right? And you can just see by the geometry of the lesion presentation, we had significant overlap. So you don’t want to just say, “Okay, well, let’s just throw two coplanar beams at this, and let’s just use the 10 collimator angle and the 350 collimator angle.” So, again, take a step back and realize that if you have a plan, and if you’re treating a patient, would you be comfortable if this was one of your loved ones? And I think the personal stories that we heard from the speakers yesterday is that they were pushing on their plans. So I think that we, as professionals, we have to take it on ourselves to be able to say, “Can we get this plan better? Can we get the 23.4% dose level better?” And we can. And that’s definitely the case, because if I just did a coplanar beam arrangement, you’re going to be going through the other lung. Even if I do partial arcs, you’re going to get something, and you have this significant heart overlap here. So I should say, too, is that at Children’s Hospital, the SFRT cases, I plan all of them. So I take these personally.

This previous speaker was showing us the interface. I just wanted to show this as well. The two cases that I’m going to show this, this is one that I planned originally using the not coplanar beams, and the second one, it’s that it was planned, and I replanned it and show it was better. So using the technology of MapRT, I was able to get this clearance map, and this case was very, very tricky because the patient had the pulmonary metastatic spread. They were coughing quite a bit, so they had to be propped up. So that essentially limited our range. We couldn’t just do two partial arcs on this case. So you can see that I actually selected beams that were just inside this clearance window in here.

And I should say to the therapists in the room is that, I know that doing dry runs is a pain in the butt, and like a previous speaker was saying is, is that you have no guarantee that the patient—You can’t reproduce the patient on the table when you’re doing the dry run. So you don’t know if you’re going to have clearance issues with this elevated position or elbows, like we have seen in the past. And this technology gives us the confidence that we are going to be able to treat this patient.

Okay. So sorry, I’m meandering here. That’s kind of what I do sometimes. But, looking at the non-coplanar versus coplanar plan. So essentially, when I was doing the coplanar plan, I just condensed all those beam angles that I had, exact same collimator angles, and just re-optimized it. So here we see, as one would expect, the high-dose target and low-dose target were equivalent in coverage. But for the non-coplanar, we are getting a lower mean dose and canal max. So heart mean was a little bit higher, but that was just due to the overlap here. One could say, “Oh, hey, that’s only .9, so that’s a very, very small amount.” But again, and one could say also is that, well, this canal, we’re underneath what the TG 101 says for five fractions. But again, I want to implore that if this was your loved one on the table, would you want a dose of seven, or would you want a dose of eight to the cord or the canal? And yeah, I’m sure that everybody would say this. And it’s especially important for the pediatric cases, because a lot of these patients can reoccur, and we’re dealing with re-irradiation. And the other thing that is amazing about MapRT is if you have these re-irradiation cases, apologize for my language, you can create avoidance structures and be able to use MapRT to specifically avoid those structures. And then the other thing I want to show you here is that look at the dose cloud in between these lesions. So that is something that’s important for SFRT, and then I’ll highlight that in case study two.

So here, another patient with pulmonary disease. This is a case that was already planned using just partial arcs, and then my goal was to, okay, well, let’s see. Granted, we treated this case. It was safe. There was nothing wrong with the case, but is there other way that I could use this MAPRT technology, non-coplanar delivery, to make the plan better? And in this case, what happened was is that it wasn’t so much… And I’ll just skip ahead. It wasn’t so much that the OARs were better performing with the lower doses with the non-coplanar plan. They were, but it was just a very small reduction. So I think in this case, it was going from a mean dose of 1.6 to 1.1. So I could entertain the argument about, is that really better or not? But what I want to show is that using the non-coplanar beam angles, we were able to get this higher peak-to-valley dose ratio, which is essentially what is thought to be what the power of this SFRT is. Is that if you can create these high-dose cores spheres and then a low-dose area in between, increasing this ratio increases the efficacy of this treatment and is supposed to be priming the immune system. So we can see that for our non-coplanar plan, the D5 divided by the D95, which is a metric that we want to look at, is better in the non-coplanar versus coplanar plan. So I think the two speakers ago, they were saying the literature’s open for, is non-coplanar better? I would argue that it is. It just is going to be better than the coplanar. And then just to test around and saying, okay, well, maybe if I just add one more beam and seeing if that makes things better because maybe we can just add more beams and everything’s going to be better. We actually saw that it really didn’t improve it. It really was the angular gantry angle optimization, collimator angle optimization, that we were able to get the optimal plan with this non-coplanar delivery.

So I appreciate you listening to my ramblings. I just hopefully got across to you that SFRT is designed for large lesions with otherwise low control rates. This treatment takes advantage of this extra biological damage to high dose, and then also the high dose/low dose combination to enhance the efficacy of the treatment. And we can use the MAPRT to make sure that we’re delivering beams that can be delivered, that are going to be safe, and are actually going to treat the patient better, in addition to using, obviously, surface guidance to make sure that you’re actually targeting what you’re targeting when you are targeting it.

*This transcript has been AI-generated. Contact us at secretary@sgrt.org if there are any issues.