Overview of the Yttrium-90 Glass Microspheres Tumor Dose Analysis Using Voxel-Based Dosimetry

Transcript:

Tyler Sandow, MD: Historically, radioembolization was considered a standardized treatment to a lobe, often around 120 gray (gy). If you think about it, that involved delivering radiation throughout an entire lobe, even if a tumor only occupied a smaller territory of the liver. The Northwestern Group, with Drs Riad Salem and Robert Lewandowski, came up with a concept of concentrating radiation and activity, the dose, on a smaller part of the liver. That, in turn, leads to a greater dose of the tumor, and it limits the amount of radiation that the rest of the liver receives, so it's essentially a win-win situation. That concept has taken off in our understanding of radiation segmentectomy and dosimetry continue to evolve, though we've seen the floor or the minimum dose that we consider giving for radiation segmentectomy increase from 200 gy to 400 gy to 600 gy and beyond.

TheraSphere, and really all Y-90 products, emit beta particles, which I like to joke are little micro bullets. By shooting these micro bullets, we're hoping to create DNA damage that results in cell death or apoptosis. So, it would make a lot of sense that the more micro bullets that you get in and around the tumor, the higher you are, or the higher likelihood you have of killing the tumor. Radiation can travel only so far, especially beta particles, most frequently around 2.5 mm, and at max no more than 11 mm in tissue. Obviously, there has to be a balance of making sure you get enough particles to get everywhere in and around the tumor, but, and this is also key, similar to what we saw with Beau Toskich’s group, having those spheres be hot enough to deliver enough micro bullets to kill tumor cells, and that's what gives you CPN. Now, tumors don't behave normally, just like there is preferential flow to areas of tumor vs normal liver.  

When you look at the actual flow within a tumor, there are areas of the tumor that get more blood supply or perfusion compared to areas of the tumor that receive less blood supply. So, when Y-90 is delivered, this is going to result in areas that get more radiation because they get more flow and, thus, a hotter dose compared to areas of the tumor that receive less radiation or colder, colder areas of the tumor. We don't necessarily want to focus on the areas that get the most flow or the most radiation because those are the ones that are going to get the most micro bullets. Those are going to be the ones that have the higher likelihood of cell death. The focus needs to be on the areas that receive less flow and less radiation because we know that with fewer micro bullets, we have a lower likelihood, statistically, of getting cell death and, thus, a lower likelihood of CPN.

In the realm of Y-90 and radiation segmentectomy, especially glass Y-90, LEGACY is probably the pinnacle paper. It was a retrospective study. They looked at three of the largest institutions to do Y-90 over the course of 3 years. They looked at treating single tumors up to 8 cm in size, and the key takeaway from that manuscript was both the durability of Y-90, in terms of time to progression, as well as the durability of overall survival. What they saw is that 100% of the tumors that were treated at 3 years continued to show a durable response, and the overall survival in that patient population was 86%. This is what ultimately got TheraSphere its FDA approval and its listing on the BCLC algorithm. Now, there's a subset paper that's frequently cited as well, talking about complete pathologic necrosis or CPN in patients who underwent resection or transplant. The key finding of that paper is that all those patients who received a perfused dose of 400 grade or greater had complete pathologic necrosis at explant, meaning the tumor was entirely killed.  

RASER is a similar study. It was a single-arm prospective study. It was looking at how effective radiation segmentectomy can be as an ablative therapy for patients who are not suitable for resection or thermal ablation. What they did is they looked at 29 patients with tumor sizes up to 3 cm. 90% of those patients had a sustained complete response, and the overall survival for that group was 96% at 2 years. Eight patients within that cohort went to transplant, and all of those patients had CPN. So, the highlights of these studies start to show that CPN or at least another presumed variable, somewhat comparable to CPN, which is the duration of response or time to progression, are the key focuses. Now, Beau Toskich’s group at Mayo Jacksonville sought to identify the variables associated with Y-90 radiation segmentectomy that affect CPN, and what he found is that both dose, achieving a high enough dose, and specific activity, or the amount of radiation that's emitted per sphere, are critical variables in achieving CPN. Interestingly, sphere concentration to the perfused territory didn't matter. So, in other words, packing more spheres in didn't help, and of all the variables that were important towards CPN, specific activity was the most important. This led to another question. If hotter spheres lead to higher rates of CPN, is there limited space inside a tumor to achieve a threshold dose of radiation?  

If you look at all the landmark papers in Y-90—LEGACY, RASER, Toskich papers, DOSISPHERE, TARGET—they had similar things in common. They all use relatively high sphere activity, and we don't really see many studies, if any, that showcase similar or effective results using lower activity spheres or a lower radiation dose. At Ochsner, we've been tracking the local and regional outcomes of our HCC population for over a decade. One of the most notable trends in our dataset was a dramatic jump in objective response rates in the patients that we were treating with Y-90 for HCC. And this occurred in July of 2021. It's funny if you go back and look in June of 2021 LEGACY was published, and that's really when everybody got on board, and we started treating with hotter doses. Juan Gimenez, one of my partners, and I had been talking about space limitations in tumors for a while.  

We knew it mattered, but we didn't really know how to explain it until some of these newer software products came around, like Simplicit90Y™. With Simplicit90Y™, we're able to use voxel-based dosimetry to evaluate tumor dose to the perfuse territory, both inside the tumor as well as the normal tissue. If you think about it, if we know what the average specific activity per sphere is at the time of delivery, and we know how much activity we gave, we can actually figure out the sphere concentrations in tumor and normal tissue. But not only just in the tumor itself, we can take it a step further and we can actually look at the coldest areas of the tumor that received the least amount of radiation.  

The goal of our study was really to evaluate radiation distribution at the voxel level, which is really the smallest clinically applicable level. Our aim was to explain why specific activity mattered. We looked at 56 patients with solitary tumors with HCC, who were treated with single session Y-90 and had SPECT imaging following that Y-90 delivery over a 3-year period. We used voxel-based dosimetry to explain the relationship of tumor sphere concentration as well as D70, D90, and D99, which are really the areas that received the least amount of radiation. We looked at all these in regard to treatment outcomes. The primary outcome studied was the durability of response or time to progression. In our cohort at 2 years, a complete treatment response was maintained in 97% of the patients. Fifteen patients within our cohort went to transplant, and the median tumor necrosis rate in all patients was 99%. Similar to RASER, all patients with tumors less than 3 cm in size had 100% CPN.  

One of the first conclusions is that sphere distribution in tumors is heterogeneous, and it remains heterogeneous regardless of how many spheres you use. Once you understand that concept, it starts to affect how you think about treating tumors with Y-90, particularly when it comes to radiation segmentectomy. We know that adding more spheres does not improve homogeneity. What it means is that there will be areas of the tumor that are colder and receive fewer spheres in a lower dose compared to areas of the tumor that will receive more spheres in a higher dose and they'll be hotter. Our data shows that achieving a higher dose or getting more activity to the coldest areas of the tumor is what drives better outcomes. Not only does it drive better outcomes, but it gives you durable results and higher rates of CPN. Once we understand that there are limitations that affect how many spheres can get into the coldest areas of the tumor, we start to realize that the more effective way to increase the dose to the coldest areas of the tumors is to make your spheres hotter because you can only get so many spheres in, ie, there are only so many seats on the bus or so many seats at the table. You have to make your spheres hotter to get an appropriate dose into those areas. Another takeaway from our study is that we were treating solitary tumors similar to what they were treating in the LEGACY study. If you look at our cohort and compare it to LEGACY, you would see that we're treating a higher volume of larger tumors or tumors greater than 3 cm in size, whereas, in the LEGACY study, the majority of the patients that were treated in the LEGACY study had tumors less than 3 cm in size. Even though we're treating tumors that are slightly larger than what they were treating in LEGACY, we're able to show that these results are reproducible and reproducible at institutions outside the mecca of Y-90, that being Northwestern and Mount Sinai. This is a very good treatment for patients who are looking for CPN and durable results.