Skip to main content

The LEGACY Study Confirms TheraSphere™ as a Neoadjuvant or Standalone Therapy in Treating HCC

The LEGACY Study Confirms TheraSphere™ as a Neoadjuvant or Standalone Therapy in Treating HCC

Dr Tabori and Dr Salem

In this episode of IOL Radio, we discuss the results of the LEGACY Study, a robust multicenter study confirming Boston Scientific’s TheraSphere™ as a neoadjuvant or standalone therapy in treating hepatocellular carcinoma. Our guest host is Dr. Nora Tabori, an Interventional Radiologist at MedStar Washington Hospital Center in Washington D.C., and a member of the Society of Interventional Radiology. She is joined by Dr. Riad Salem, the principal investigator of the LEGACY Study. Dr. Salem is a Professor of Radiology, Medicine, and Surgery, Chief of the Section of Vascular and Interventional Radiology, and Vice-Chairman of Image-Guided Therapy at the Department of Radiology at Northwestern University in Chicago, Illinois.


Ami Peltier [00:00:00] Hi, and welcome to IOL Radio. I’m Ami Peltier, managing editor of IO Learning, a digital publication geared toward interventional oncologists, and the news source for the symposium on clinical interventional oncology.  In this episode of IOL Radio, we will discuss the results of the LEGACY study, a robust study confirming Y-90 with Boston Scientific’s TheraSphere glass microspheres as a neoadjuvant or standalone therapy in treating hepatocellular carcinoma. Our guest host today is Dr. Nora Tabori, an interventional radiologist at MedStar Washington Hospital Center in Washington D.C., and a member of the Society of Interventional Radiology. She will be speaking with Dr. Riad Salem, the principal investigator of the LEGACY study. Dr. Salem is a Professor of Radiology, Medicine and Surgery; Chief of the Section of Vascular and Interventional Radiology, And Vice-Chairman of Image-Guided Therapy at the Department of Radiology at Northwestern University in Chicago, Illinois.

Nora Tabori [01:13:13] Riad, thank you so much for letting me interview you about the study. I'm really excited about the results that I've seen so far and the opportunity to talk to you about them. Just to start off, generally, how have you been, Riad? Are things in Chicago good these days?

Riad Salem [01:29:14] I've been well, thank you very much, things are good. We're getting through this thing, but things are going well. Thank you. 

Nora Tabori [01:35:08] That’s great to hear! So just to get right down into it, the LEGACY data is pretty amazing, so I wanted to start out just by asking you a little bit about the design of the study and sort of how that offers an enormous amount of strength behind the data that you guys have been able to collect. 

Riad Salem [01:54:02] Thanks, Nora. The LEGACY is a concept we thought up years ago, which was designed to address contemporary discussion points and narratives that I think are becoming much more interesting in the oncology space, particularly in HCC [hepatocellular carcinoma] and particularly in the immuno-oncology landscape response rates and duration of response. And, as you know, there have been several agents that have studied this, either under proof-of-concept and have even got FDA approval based on phase 1 and phase 2 trials because of clearly the regulatory (and probably medical) interest in response and duration of response. And so what we sought to do is, while you and I have been doing this for a long time and understand the ability of local-regional therapy — and particularly, Y-90 and TheraSphere — to get high response rates and a long duration of response, it’s never really been quantified at a high level. So what we decided to do, is to design a clinical trial that would minimize confounding factors and allow us to explicitly study those two variables. And so it was a multicenter retrospective analysis, and we chose solitary tumors. Why did we choose solitary tumors? Because you don't have multifocal bi-lobar disease that confounds survival and confounds response analysis. We didn't want people with portal vein thrombosis or extrahepatic metastases, because that will confound overall survival and it's really not the patient population, so what we did is, if you have a solitary tumor, what sort of effect will you get if you perform high-dose local radiation segmentectomy, and in some cases radiation lobectomy, and what will the outcome be, and what will the path of necrosis look like, and what will the duration of response be? And so that's what the rationale was, and I think one of the most powerful things, Nora, to mention about this study was the fact that we used a blinded, independent central review. We, as investigators, while we are reading our own scans and looking at these things, there's an inherent natural bias that comes in, saying yeah, I think it's reached the response, I think this is the measurement. But if you have an independent party looking at these, doing the measurements, that has no interest in the outcome, this is very powerful data. That's what we did, and it's one of the first studies where we report that using BICR [blinded independent central review] — 2 independent radiologists for each patient and a third one to adjudicate. So that was sort of the rationale: to answer what clinically is going on, to show that also our therapies have a long duration of response. But that's how it was initially crafted.

Nora Tabori [04:48:11] So, you know, to the point of the study design by having an independent review, in a lot of ways do you feel that this makes this data as robust as a double-blinded prospective trial or as close to as we would be able to get with that without actually creating a randomized prospective trial? 

Riad Salem [05:08:18] Yeah, Nora, that's a great question. It was a retrospective, brief multicenter study where the data was reproducible. What would a prospective study look like? So it would be basically that same patient population entered. There may be some patients that would drop out because of technical reasons for example, but we know with rad-segmentectomy that dropout rate is 1%, maybe 2%, so it's very low. So it really is approaching almost the exact same thing you would get had you done this prospectively, and so that's a very important nuance, but at some point a very high-quality retrospective study with some imaging endpoints approaches the quality of data that you get with a prospective, and even in some cases, a randomized trial.

Nora Tabori [05:56:17] Right, and I do, I really agree with you that to highlight the point that this was 3 different centers across the country and that the results were reproducible in all 3 places and, you know, just from your perspective, how do you feel that that allows us to translate this result into the general population of IR across the country, across the world. 

Riad Salem [06:20:22] So one of my early missions early on when I was learning how to use Y-90 and implement that in patients just to make sure that whatever techniques we came up with were imminently translatable, reproducible, and to keep the level of complexity at a pace where interventional radiologists nationwide, worldwide would be able to replicate this. And so, as you know being an expert yourself, when you perform radiation segmentectomy in 1 vessel, this is a pretty standardizable method, particularly when you use MR-dosimetry, it’s pretty straightforward, there’s nothing complicated about it. We paid specific attention to this as we developed the rad-segmentectomy concept. So now, what this study confirms and validates is that indeed you can get the same results in multiple centers. The data are effectively the same, and we solidify those findings with the path-necrosis data that maybe we can talk about about a little bit later on, but it readily then brings into the discussion the curative option when you’ve got that path data.

Nora Tabori [07:27:18]  So let's dive into the actual demographics of the patients that we were treating here. Obviously most of us in the interventional oncology space use the BCLC [Barcelona Clinic Liver Cancer] guidelines or at least classification system when we're stratifying our patients. Where did the patients fall in?

Riad Salem [07:46:26] So you're right, we all like to use the BCLC guideline. I really like the BCLC guideline, because at least on the diagnostic portion, on the top end of it, you really know what BCLC-A means, you know what a B means, you know what a C and a D mean — they’re clearly defined and I'm able to compartmentalize who these patients are. But it was also important, as we designed this study, to have a clinically relevant patient population that we all see in our practice, and so the idea of picking the solitary tumor to homogenize the patient population was clearly very important and and if you look at your patient population and most HCC centers, about half of the patients they treat have a solitary tumor. It's not multifocal disease, HCC very often presents with 1 lesion, and as a result it makes this patient population quite relevant. So, solitary tumors was the first criteria. Child-Pugh A status was very important. Why? Because you want patients to not have the confounding effect of having liver dysfunction. So, why would I treat a solitary tumor in a Child-Pugh C if his overall survival is 4, 5 or 6 months, right? I want to see the effect of the treatment without the confounding effect of liver dysfunction. So we chose Child-Pugh A, which happens to be what the guidelines for research in HCC happens to talk about as well, and importantly, ECOG [European Cooperative Oncology Group] 0 and 1, because there are patients that may have some larger tumors that are solitary, that exhibit cancer-related symptoms, and maybe some weight loss, but are ECOG 1. That now allows you to include advanced HCC patients. And is this a relevant patient population? It sure is, because if you look at most systemic therapy clinical trials, about 20%-30% of the patients do not have PVT [portal vein thrombosis] and they do not have metastases, which means they have liver-only disease and they are in the advanced setting because of ECOG alone. So we are just doing exactly what everybody else does with research and we therefore included Child-Pugh A solitary tumors, which are BCLC A, with or without ECOG 0 or 1, and when you are a 1, you are now technically advanced. So I think it's a very relevant — it’s a patient population that we see in our practices all the time. 

Nora Tabori [10:08:06] Yeah, I completely agree with that statement and it's really nice to have some data that applies to such a large cohort of the disease that we're seeing and really does remove those confounding factors of what underlying cirrhosis does to your prognosis as opposed to just your cancer itself, and I think that's a really important point. So, the results themselves — this is obviously your study, so I would love it if you could tell us a little bit about what we found because I'm pretty excited about it.

Riad Salem [10:38:09] So, let's talk about response rate first. There are two response rates we are reporting — one in the mid-70s one of the mid-80s if I remember correctly, and I want to tell you about why there are two response rates. One of them has to do with confirmed response rate and one of them has to do with best overall response. And in the oncology space, when you achieve a response, in order to count as a response, you have to confirm it. In other words, one of the nice things of the BICR that I talked about Nora early on was that you were assessing response, and when you declare response, you have to make sure that you confirm it. In other words, if the next scan 2 or 3 months later confirms what you have just found, that truly counts as an imaging response. However, there are some patients that show a response and then go to transplant, or in some cases can be lost to follow-up, or get a resection. So, they responded, it’s just by the letter of the law, I'm unable to take credit for that response because it's not really confirmed. And so, the best response was in the mid-80s (mid-to-high 80s) and the textbook response was in the mid-70s or so. So this is very very compelling data, as we reported. The other thing that was important — and I hope I'm remembering most of my numbers, correct me if I'm wrong — is about a 45% or 46% RECIST response rate. So, one of the things that is always challenged in the locoregional space is that, well, lesions don't shrink when you embolize them, or when you do Y-90, and that's not true. About 30%-50% of the lesions will exhibit a RECIST response, and I believe that in fact, that's what we found, so that's on the response side.

Riad Salem [12:19:16] More interestingly was the duration of response — like when you did respond, how long did that last? And about three-quarters of the patients exhibited a response over 6 months and much of it was censored because many patients were treated and never progressed. So, we have a lot of very compelling data that has very high response rates — much higher than you're seeing obviously with some of these systemic agents. And when you do respond, you respond for a very long time, and one of the nice things about this observation is that it really mimics and parallels what I see clinically. When I do a rad-seg, almost invariably I will never have to retreat that area. It is possible that new tumors develop. It is possible that metastases develop as part of normal biology, but that segment of disease of tumor that I've treated — that liver — will almost never progress. In terms of the the adverse events — and again, correct my numbers if I'm not remembering them correctly — but we had a 5.6% adverse event rate. Very, very low. It was a very safe treatment, and so here you have an outpatient treatment doing high-dose (and we’ll probably talk about dose a little bit later on) but a very safe treatment with very few adverse events, with a very high response rate, many of which, the localized M-RECIST rates were 100%. So very, very compelling data and I'm excited when we start looking at the publication strategy for this manuscript.

Nora Tabori [13:51:02] I wanted to talk a little bit about overall survival as well.

Riad Salem [13:55:39] Of course. 

Nora Tabori [13:56:39] And so one of the things is that in this study, you guys found that there was a 93% overall survival rate at 3 years in patients who were transplanted or resected, and you found an 84% overall survival rate at 3 years in patients where it was stand-alone therapy. So, in terms of what we have traditionally seen when we are bridging are downstaging to transplant with either TACE or ablation or even when we were doing it with Y-90 prior to using higher doses. How do you think that these results compare?

Riad Salem [14:33:00] I think you're absolutely right. I think what you're alluding to is the fact that with experience, with improvements in imaging, with improvements in catheter technology, we are better able to inject and deliver higher doses of radiation, and we'll address the doses a little bit later on. But you're right, I think you're right, and so now as a result, if I can give my therapy more effectively, I can give a higher dose and better downstage or bridge my patient, then that should translate to a higher rate of liver transplantation, which as you know, as opposed to non-transplantation, is going to be better. And I think that's what explains that difference, but context matters, right? There are patients that are elderly, that are in their 70s and 80s, that are technically probably higher risk for transplantation. So while yes, transplantation is what I would recommend, for example, in a 50 or 60 year old as we bridge or downstage that would be reasonable, but in the elderly patient population transplantation is not feasible. So as you allude to, Nora, stand-alone therapy is something that you're going to want to be thinking about, because worldwide, yes — the guidelines talk about liver transplantation as one of the treatment options. Unfortunately, worldwide, liver transplants aren't available in many countries afflicted by HCC. There are no organs to go around. Liver transplantation is not performed. So it's no use to have that as one of your guideline recommendations when it doesn't even exist. So, I do believe that stand-alone therapy — the idea that this is going to be your definitive or destination care with Y-90 — is extremely, extremely interesting.

Nora Tabori [16:14:10]  Yeah, and just to that point of, you know, patient age and demographics, one of the things that I really like about this study is that the patient population, about 18% of people treated were actually 75 or older, which in many areas they will have aged out of transplant. So we're really giving a durable result to a patient population that doesn't have necessarily another option or what we would call the gold standard or most definitive option of treatment. So these results are very interesting in that context as well. 

Riad Salem [16:45:27] Yeah, I mean, you're absolutely right. I think, yeah, like 17.9% or something like that, if I remember, were aged over 75 and so what do you have now? Now you have a treatment with high response rate, that's outpatient, that is very well tolerated, with very few adverse events, with a high duration of response, that you can apply across early and advanced patients in all age groups. So clearly, we've moved the scientific field forward to make sure that we, as interventional radiologists, working in collaboration in a multidisciplinary manner with our hepatology, oncology, surgery, nuc-med, and rad-onc colleagues, are able to move the field forward and really make a significant change and impact in treating these patients.

Nora Tabori [17:29:14] So to that extent, one of the interesting things that I found is in 2018, your group — Lewandowski et al’s paper — demonstrated that Y-90 segmentectomy could be used as a destination therapy and what they found is that the 3-year survival probability for Y-90 alone was 66%, and we’re seeing a much higher percentage here, right? We're seeing 84%?

Riad Salem [17:58:20]  Yep. 

Nora Tabori [17:59:00] What do you attribute that difference to? 

Riad Salem [18:01:20]  So, I think I could attribute that to a couple of things. The first one is dose. Dose — the ability to deliver the dose, improvement in technique. So if you go back and look at our 15 to 20 year data, our outcomes — our contemporary data are much better than our historical data. That's just based on better dose, better dosimetry, better technique, better imaging, all of the above. I think that's very important. But the other thing that explains that lower survival is what I was alluding to before, Nora, which had to do with liver function. So, you know, that paper included some more advanced liver disease patients and talking about curative therapy. And hence, you will survive I'm able to kill the tumor, but if your Child-Pugh B, for example, that liver dysfunction is going to be a life-limiting situation as well. And so, here in LEGACY, that was part of the rationale to only take Child-Pugh A patients, just like systemic therapy trials. They take that Child-Pugh A patient, so will we. And why? Because now you're able to assess the long-term effect, the long-term outcome of this therapy without the confounding effect of all of these parameters that have an effect on overall survival. So now we have Child-Pugh A as part of LEGACY — again, no metastases, no PVT. So we're really able to dig in and say this is most likely an effect of the treatment, as opposed to other variables that we’re unable to control.

Nora Tabori [19:30:00] So since we keep saying the word dose, let's talk about what the dose that we found, or you found, was optimal through this trial. 

Riad Salem [19:37:26] Yeah.

Nora Tabori [19:38:26] So, 410. The landmark paper that was put out in 2014 had 190. So we are more than doubled. 

Riad Salem [19:47:09] Yeah. 

Nora Tabori [19:48:26] How did we get to that number? And how much do you attribute the difference in what we’re seeing going forward to that hyper dosing?

Riad Salem [19:56:08] So, as I’ve said many times and I think it's probably my statement of the year, context matters, and understanding that is really relevant. So when we first put out the 190 Gray, what were we doing? We were doing 2 x 2 analyses to identify, at what threshold are we more likely to achieve CPN [complete pathological necrosis] than not? Because if you look at the 190 paper, still, only two-thirds or maybe 70% of the patients got CPN, but one-third of them did not, even though they got more than 190. So that initial 190 was the threshold, and what that evolved to, is with more experience, investigators and guideline documents came out to talk about 250 or 300 as an important improvement in that dose. What do we learn from LEGACY? What we learned with LEGACY is the following: what we decided to do — and we published this in the European Journal of Nuclear Medicine as a sister manuscript that talked about the rad-path component. One of the things, Nora, that I think we do insufficiently in IR is rely sometimes too much on the imaging and not enough on pathologic tissue if it's readily available. So in LEGACY, we had 45 patients — I believe 34 transplants and 11 resections — that we had the tissue. So we correlated this with the dose that we administered using MR dosimetry, and what we found explant. And when we first look at the manuscript and look at the data, there was a clearly a dose-necrosis relationship. Clearly, you get more necrosis as you increase the dose. But interestingly, when we were looking for a cut-off, we asked the question, at what cut-off do we achieve 100% CPN? And that number was, I believe, just over 390, so 400 Gray then became, as part of this study, the new gold standard, so you're right. So 190 is now 400 — 190 is very likely to give you CPN; 400 will always give you CPN, so context and interpretation matters, but here we are, and you and I have done this quite a bit, where we give higher doses anyway in the 4-, 5-, or 600 Gray and you get CPN that way. So this really also parallels and mimics what you see clinically as you manage your patients. The higher dose you're giving, the better results you're getting.

Nora Tabori [22:26:28] And so, Riad, you know when I'm speaking around the country — when we used to actually travel and do these things — a lot of people would ask me questions about how I feel about these super high doses and whether I was seeing more toxicities, and obviously in this paper, in these results, we see very low toxicity. How do you sort of address that questions when physicians that are moving into the field of Y-90 and trying to increase their utilization of it are somewhat nervous about giving this high of a dose?

Riad Salem [22:58:28] Well, I say that I share their nervousness, and I used to share, uh that nervousness, you know, when I first started doing this, but the only thing you can do is to have a hypothesis, investigate it, see what you find, look at your adverse events, look at your rad-path data, validate it with other colleagues and other centers to see if they're seeing the same thing, and then act on that. And so, sure, you might be nervous about 500 Gray in a small segment of liver, but it's been investigated by us, by your group, by Sinai, by Washington, by UCLA, by groups in Germany. So it's been done, and now reproduced internationally, and so it's fair — it's fair to be a little bit nervous about it. If they really are, what I would do is suggest they escalate it to 500 Gray but don't go below 200 Gray when you're going to do rad-seg just because you need to feel comfortable. And so, it's fair to say look, I need to be comfortable myself, and our center, we need to be comfortable with that as well. I think that's reasonable. But what I would recommend in that setting is, at minimum starting at 200 and 250 Gray, because here's multiple manuscripts that show you it can be done safely and then ultimately get the 400 or 500 Gray as you build your confidence, as you see the limited adverse event profile, as you see what those Eovist scans look like, and follow up where you've just radiated only that small segment and nothing else happened. So it's okay to do that. And I think people have that experience, but build and do your own dose escalation until you get to 100% CPN, but don't start too low. I have seen some people make that mistake. They're a little nervous about it. They don't do it selectively, and they go very low, and then they don't get an effect, and they say the treatment did not work. Well, that was just done suboptimally. The other thing I tell them is, again, as you get your experience, take your Child-Pugh A patient, get that experience, get the good performance status people, just because you know, again, any adverse event you see in a good performance status person Child-Pugh A will be attributable to what you've done. But if somebody has got multifocal disease everywhere, ECOG 1-2 is not doing that well, and AE will be confounded by everything else and you won't know if it's your therapy or not. So pick your patients properly, gain that experience, and escalate until you get to that point.

Nora Tabori [25:26:25] And then going back to the point of transplant and bridging to transplant or downstaging to transplant, just to delve a little further into that. So, as you and I know, the priority points for HCC have changed quite a bit in the UNOS-T2 criteria in the last, you know, 12 to 18 months, and there is sort of essentially a mandatory 6-month waiting list in order to get your priority points, and the escalation of priority points is no longer the same as well. So when we talk about that stand-alone therapy achieving, you know in the 80s for a 3-year overall survival, how do you think that plays into transplant when we're trying to make sure that these patients stay inside of criteria and get the opportunity to get a transplant if they are a candidate?

Riad Salem [26:11:18] So, I was very interested to see the 6-month mandatory wait because as you know, one of our earlier research works that have to do with randomization is where we randomize conventional TACE to Y-90 TheraSphere effectively in a patient population being bridged or downstaged to transplantation. And this is where we found the time to progression with TheraSphere was over 26 months. We don't even reach the median. So the interesting thing here is that this now segues into how we are practicing now and is a great sister analysis to what we see in LEGACY. You're now able to apply TheraSphere in a segmental manner and achieve very long time to progression, and that is our job as interventional radiologists working in a multidisciplinary manner with a transplant surgeon. I need to try to keep that person within UNOS-T2 criteria within transplant, and I want to use the agent that is most likely to give me the longest time to progression and the highest ability to maintain within-transplant criteria, and what we found with the PREMIERE trial was a very long TTP with Y-90. As a result of that, that is now our standard of care. That is how we bridge patients who are within criteria and may have a long wait, because we want to maximize the likelihood of doing that to make sure that the patient has a chance to enjoy the long-term survival benefits of a liver transplant.

Nora Tabori [27:40:23] Yeah, and I think that's one of the most important points that I often talk to people about. You know, when I was in New York, our average time to transplant was 24 months. So we really needed quite a long window. In the D.C. Washington area, the time to transplant is shorter, but now with that mandatory bridge of 6 months, we really are trying to make sure that we're getting that durability, and given that even when they do finally get priority points, if they start out as a Child-Pugh A, they may get their priority points, but that never really bumps them high enough on the list to actually get a transplant, and so this takes them out in a much longer progression inside of their underlying disease pathology to optimize their time on the list, and that's been really helpful at least for me with my transplant surgeons, and certainly I hope that would be the same for other people 

Riad Salem [28:33:07] Agreed. 

Nora Tabori [28:35:00] So, you know, I think with all things in life, Riad, we all sort of like personalized stories of different patients we may have bonded with during this period. Is there a patient story or a result that you can remember of one of these LEGACY patients that you really found that like, this proved to you or this spoke to you in a way that made you know that this was the future of where we should be going with interventional oncology? 

Riad Salem [29:01:26] Well, if I may, I would like to share just a couple that some some maybe in LEGACY, some maybe outside, but the theme is really there. So, one of them is a patient that was a Child-Pugh B, but had 2 lesions that were outside that made them UNOS C3 — a very nice gentleman restaurateur in town, a little bit of ascites, and bilirubin of 3-5-4-5, fluctuating like that ,and was presented at tumor board that needed to be downstaged, otherwise, he would never get a liver. He had no live donor. So, we needed to downstage that person. One of the lesions was 8 centimeters, the other one was 4 centimeters, and we treated that person. The lesion shrank to within Milan criteria, the AFP normalized from 5000 to 2000. The patient was maintained within that because of the durability of response that we’re describing in LEGACY, and as a result was able to receive a liver transplant and is alive and well. So this was 4 or 5 years ago, and this is a person that we see in clinic and we have, you know, the clinic visit now is a friendly clinic visit as to how are you doing as opposed to any complex medical issues, because he's doing quite well, so that was, you know, we bonded quite a bit with that, it’s really a very good story. The other one I would tell you is of a gentleman that was given obviously limited long-term prospects given the size of his lesion, and it was in the 20 centimeter type lesion, and the person underwent one right lobe radioembolization that I did on him, and it shrank to almost 4 or 3 centimeters, from 22 to 4, and the person survived 5 years without needing systemic therapy or anything else. And so, these kinds of stories where you're able to really personalize, as you mentioned Nora, you really personalize how you're going to treat that person, break the guideline rule — that's what we're always trying to do. That's what personalized medicine is, designing the therapy explicitly for that patient and then breaking all the statistical rules that you're supposed to exhibit and observe in these patients is really just what makes it all worthwhile. So these two stories, I will remember forever. They influenced me as a physician, influenced me as a person, and the visits with them are just always a pleasure because they're so grateful for what you've done for them, for their quality of life, for what they're able to do. It's really something that is moving on all fronts.

Nora Tabori [31:32:26] I like that you touched on the concept of quality of life. I mean, I think one of the most rewarding parts of Y-90 for me is the incredibly low side effect profile that preserves people's ability to go on with their lives in a way that's meaningful to them. You know, we're not just treating their cancer. We're treating their whole life and we're making sure, in my humble opinion, that they have the ability to enjoy however much time they have left, and I think that's a really important point. 

Riad Salem [32:04:08] Yeah. I can't agree with you more, Nora, I mean I've long been a non-believer in the idea that overall survival is the only relevant endpoint when we think about new studies or designing clinical trials. Maybe it's my Canadian upbringing, but quality of life is 10 times more important in most cases, and when you talk to patients, if you have a certain amount of time, would you rather it be from adverse events or do you want lower adverse event rate and better quality of life? And many times, obviously, as you can imagine, patients want to survive that amount of time that is possible with the fewest adverse events, which is why I'm not always in favor of clinical trials that purely focus on overall survival. To me, if you do a clinical trial and you demonstrate equivalent survival, but a much better quality of life, fewer treatments, fewer adverse events, I would argue that's more important than extending overall survival by a month or 2 months and that's really some of the work that has been done historically, with quality of life, with FACT-Hep or with PROs, and so it's really, really very important and that's almost always the selling point of Y-90 over the more traditional macro-embolic therapies.

Nora Tabori [33:29:16] I completely agree. So Riad, where do you plan to go from here? Is there anything that you have in the works, that you've started, that this is the direction that you see us taking research forward?

Riad Salem [33:44:28] I sure do. Let me share just a little bit with you. So, we are currently in the midst of several projects. I'll share a couple with you. One of them is Y-90 radioembolization for thrombocytopenia in patients with cirrhosis. So, as you know, one of the complicating factors in resection and transplantation and dealing with cirrhotic patients is their low platelet count, and we, as interventional radiologists, have developed partial splenic embolization to increase platelet count. And I do believe that from a technical standpoint, when you embolize a portion of someone’s spleen, the platelet count will go up, so it does work. But it's my opinion that that procedure provides an unacceptable side effect profile. It's very painful. The imaging of the spleen is very complex. It becomes liquefied. Some people end up with tubes unnecessarily and infections. And so, to me, it has an unacceptable quality of life profile and side effect profile. So, what we are studying, is to do whole spleen radioembolization to shrink the spleen and use the fibrotic effect and hopefully translate into an increase platelet count. It's a 20 patient study. We've done the first 4 patients already. We're dose escalating, so we've seen no effect on platelet count yet, but it is safe and there is no lung shunting, so we've eliminated the lung shunting step. And now as we enroll patient number 5, it’s measure the spleen volume and treat, so imagine if I can get that to work — it's really a game-changer for thrombocytopenia. The second thing I want to discuss briefly is our preclinical lab. My colleague, Sam Mouli, is doing some work on Y-90 for prostate cancer. He has a canine model. He's done 12 dogs already. He's got the results, and we're looking at publishing this already, but all the things you might worry about when you place Y-90 in the prostate — urethral injury, perigland fibrosis and scarring — none of these things have been observed and he's got all the explant data because he's done all that stuff. My colleague Bob Lewandowski here is doing an RO1 study on radiation lobectomy, because while I do agree that liver transplant is a great treatment option, there are not enough organs to go around. So we need to figure out ways to make more patients operable, and part of that is to make their platelet count higher with spleen, and the other way is to hypertrophy their FLR by treating, most commonly, the right lobe. And the last thing that is very interesting that we're working on is brain tumors — brain gliomas and meningiomas. We are currently writing a protocol now and collaborating with leadership at Boston Scientific to look at brain tumors in the glioma family and in the meningioma family because it's my contention, Nora, that in organs that are vascularized, the arterial route is the most effective way to deliver radiotherapy because you minimize non-target irradiation, you confirm exactly where the Spheres have gone; in contradistinction, external beam radiotherapy is proton, etc. This is all modeling and you cannot confirm where you've deposited the radiotherapy, but we can by using PET-CT. And if I may just postulate one other thing about Y-90 outside the liver, is I do think that we should study Y-90 for renal cell cancer. You know, I agree that cryoablation is the standard for 1, 2, 3, maybe 4 centimeter RCCs, but why is cryoablation still utilized if you got a 5, 6, 7, 8 centimeter RCC that's inoperable? Why do we just increase the probes and why not think about an internal radiotherapy given the vascularity, given its distance from other structures, so you can do it safely? There's a lot of exciting things that we can do with this technology and I think we've only just begun now that hopefully the liver story is complete.

Nora Tabori [37:45:03] Riad, I just wanted to say thank you so much for your time. Thank you so much to you and all of the LEGACY investigators. Just to sort of recap the incredible overall survival rates at 3 years in the 80% range, the time to progression, the 100% response rate with a maximum of two treatments, but a 96% response rate with just 1 treatment by localized RECIST, and the fact that we've really drilled down a specific number that gives us 100% complete pathologic necrosis with that dose every time is, I just think, a huge movement forward in our field, so thank you for all of the work that you guys have been doing and it's really been a pleasure speaking with you. 

Riad Salem [38:38:26] Thank you, Nora, it's been a pleasure.

Ami Peltier [38:40:03] Thank you again to Dr Nora Tabori and Dr. Riad Salem for being our guests today here at IOL Radio. We would also like to thank Boston Scientific, manufacturers of Therasphere glass microspheres and sponsors of this podcast episode. For more information about Y-90 with Therasphere and other products, please visit

To view a transcript of this podcast, or to listen to more episodes of IOL Radio, please visit our multimedia page at Thank you for listening!

Back to Top