“Inside Access” provides interviews and background on open access articles from the Journal of Interventional Radiology.
Sandow, Tyler et al. Using Voxel-Based Dosimetry to Evaluate Sphere Concentration and Tumor Dose in Hepatocellular Carcinoma Treated with Yttrium-90 Radiation Segmentectomy with Glass Microspheres. JVIR. 2024;35(11):1602-1612.e1.
Listen to Dr. Sandow discuss his research on the Kinked Wire.
Tell us about you, your team and your institution.
Tyler A. Sandow, MD: I am at Ochsner Health in New Orleans, Louisiana, and I'm just a small cog in a big wheel, and I've been fortunate to be a part of that since I was a resident. I've been at Ochsner for 12 years, and we have an incredibly collaborative team in IR. The majority of our oncology work is shared amongst the group.
Ochsner's transplant success obviously led to a lot of referrals for bridging and downstaging to transplant for hepatocellular carcinoma (HCC). When I was a resident, I built a database of all the patients we had treated with HCC. This caught the interest of the head of our transplant surgery, Ari Cohen, MD, as well as his research partner, Paul Thevenot, PhD, who is really the brains behind this whole science. We decided to combine forces and build a rolling database to study our outcomes, trends and opportunities for improvement. Several years later, the transplant allocation scheme started changing, which affected our (Ochsner’s) access to livers. There were differences in the way HCC exception points were changing, and wait times were getting longer for patients with HCC. Our IR section chief Juan Gimenez, MD, had a brilliant idea, and he really does deserve all the credit for us being where we are today.
Following the results of PREMIERE, which was that randomized trial comparing transarterial chemoembolization (TACE) to Y90, he made a push for us to shift from our local regional treatment strategy from TACE to Y90. Now 5–6 years out, we’re on pace to treat over 300 patients with Y90 for the second consecutive year.
Why did you pursue this topic?
TS: We treat a lot of patients with Y90 and the majority of these receive radiation segmentectomy. Since we were in a position to track outcomes, we looked at our trends and noticed some interesting things.
One of the most notable trends was that our objective response rates took a dramatic jump in the positive direction in July of 2021. In June of 2021, LEGACY was published; even though some of us were treating with higher doses, some of the holdouts in our group were waiting for actual literature to be published to support that idea.
Once everyone jumped on board, we started treating with higher doses, which was driving better outcomes and higher rates of complete pathologic necrosis (CPN). The next year, two papers were published in JVIR and the European Journal of Nuclear Medicine and Molecular Imaging which showcased the importance of high-sphere activity.
At this time, there was a question of tumor space limitations, and how that impacts radiation dose within the tumor. There's just not enough space for spheres to fit. Dr. Gimenez had been pushing for us to get Simplicit90Y. I was talking with him and said, “I want to call this the sphere conundrum. If we know sphere activity at the time of delivery, we know the dose to tumor and we know the dose that we gave, we could figure out what the sphere concentrations are. But we need to figure out what the dose to tumor is.”
And he said, “just do it in Simplicit90Y.” And at that point we just used the dosimetry tool we had to solve the problem. So that's where we are.
What are the key takeaways from this paper?
TS: The first is that sphere distribution in tumors is heterogeneous, and it remains heterogeneous regardless of how many spheres you use. I think once you understand that concept, it starts to affect how you think about treating tumors with Y90 and particularly radiation segmentectomy. We know that adding more spheres doesn’t significantly improve homogeneity. That's something that's been shown in the higher particle-load papers that have recently been published that talk about radiation segmentectomy as well.
What it means is that there's going to be colder areas of the tumor that receive fewer spheres and thus a lower dose compared to hotter areas of the tumor that receive more spheres and more dose. What our data shows is that achieving a higher dose or getting more activity to the coldest area 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 to the coldest areas of the tumor, the more effective way to increase dose to the coldest areas of the tumor is to make your spheres hotter. Because you can only get so many spheres in, we have to get our spheres hotter.
Another takeaway is that we were treating solitary tumors, which is similar to what they were treating in LEGACY. But if you look at our cohort and compare it to LEGACY, we're treating a higher volume of larger tumors (>3cm). We were closely replicating the durable results that we were seeing from LEGACY. And that was what got it (Y-90) on the Barcelona Clinic Liver Cancer (BCLC) algorithm—the durability of the radiation segmentectomy. We were able to show that even though we're treating larger tumors, we're getting very close to the same results, if not the same results of what they're seeing in LEGACY.
The last takeaway is that we're getting to an area of smaller volume radiation segmentectomies or sub-segmentectomies. That limits the amount of normal tissue that gets treated with high doses of radiation. However, it's important to recognize what happens to the tumor and to that perfused territory because you're not getting the same hypervascular tumor uptake benefit that you get with a larger perfused volume, because the tumor tends to occupy more of a relative volume to the perfused territory. As you make everything smaller, the percentage of tumor that occupies that perfused volume is greater, so it's going to affect the dose to tumor. In one of the figures in the paper, we show that as you see the perfuse volumes drift smaller and smaller towards the left, the sphere concentration in tumor and normal tissue starts to converge. So, 400Gy to the perfused territory might imply different radiation doses to tumor when the tumor occupies 2% of the perfused territory versus 25% of the perfused territory.
What do you think would be the long- or short-term impact of what you and your team found?
TS: I think this helps validate all the things that we've seen in published literature from the IR legends regarding Y90. Dose is critical. It's the most important thing. But high-sphere activity plays a pivotal role in getting dose up.
From those that I've talked to, I think we're starting to see people striving to use higher sphere activity and higher doses. I don't think it has anything to do with the paper. I think it's just people recognizing that this makes a lot of sense now. Once you start to pose this concept to people and they start looking at their own outcomes, they begin realizing that this makes a lot of sense.
As people get more comfortable with using and understanding dosimetry—and also look at their post treatment follow-ups, higher doses, high sphere activity and post-treatment imaging—it will increase confidence and comfort with treating with high sphere activity and higher doses.
How might this research influence treatment practice or clinical processes?
TS: If you look at the clinical dosimetry guidelines for radiation segmentectomy with glass microspheres, you'll realize this is what IRs have been saying for years. It also helps support the conclusion that had been identified several years ago. And it reinforces what we know: if you treat within the appropriate radiation segmentectomy treatment window with the way the guidelines advise, that's a week one Wednesday to a week two Tuesday for glass microspheres, then you’re treating with relatively high sphere activity. Assuming that the lungs and the liver can tolerate it, 400Gy is the new floor for radiation segmentectomy and is the minimum dose to get things the way you want.
Are there any next steps or plans for follow-up?
TS: Nothing to the same degree as this research, but we have some other plans to explore dosing and outcomes in the advanced HCC patients. Obviously, the paper that we just published is on early HCC patients, but we’ll look at advanced patients who are receiving combination therapy and how the dose to the tumor affects them. Also, because we have that huge database, we're also looking at our Y90 outcomes versus other modalities across the BCLC spectrum. That includes microwave ablation and TACE. We have some interesting findings in those datasets that we are writing up right now and hope to publish those in the coming months.