Skip to Content
SIO
  • About
    • Leadership
      • Board of Directors
      • Committees
      • Staff
    • SIO Story
      • Founding Members
    • What is IO?
    • FAQS
    • News
      • IO Insights
      • SIO Corner
    • Contact Us
  • Membership
    • Member Center
      • Join SIO
      • Renew My Membership
    • Member Benefits
      • Online Directory
    • Resident and Medical Student Section
    • International Partnerships & Chapters
    • Get Involved
  • Events
    • Annual Scientific Meeting
      • 2026 Program
      • Abstracts
      • CME Information
      • Sponsors
      • 2026 Photos
    • NEXT Summit
      • Registration
      • Program
      • Faculty
      • Hotel & Travel
      • Scholarship
    • MSK Master Class
      • 2025 Photos
    • Breast Cryoablation Master Class
      • 2025 Photos
    • IO Awareness Week
    • Future Events
    • IO Clinical Trials Methodology Workshop
      • Workshop Schedule
      • Faculty
      • Invited Scholars
    • POWER Workshop
      • Workshop Schedule
      • Faculty
      • Application Requirements
  • Education
    • SIO Learning Center
    • Talk Tuesdays
    • IO Awareness Week
    • RMSS Immersive Virtual Training Course
    • Upcoming Webinars
    • On-Demand Webinars
    • Podcasts
  • Research & Resources
    • Recommendations to NCCN Guidelines
    • Society Publications
    • Society Priorities
    • Research & Grants
      • ACCLAIM Trial
      • TRIBUTE Trial
      • SIO Grant Program
        • Application Procedures
      • Past Grant Recipients
    • IO Insider
      • Archives
      • Case Spotlight
        • Sponsored Case Spotlight
    • Clinical Trial Collaborative
  • Industry
    • Industry Membership
    • Support the Society
Login     Patient Resources
Research & Resources IO Insider Case Spotlight Sponsored Case Spotlight

Sponsored Case Spotlight

The following Case Spotlights are supported by Replimune. All content is the property of, and has been developed by, SIO and includes opinions of the participating physicians.



A Conversation Between Muneed Ahmed, MD, FSIR, Chief, Division of Vascular and Interventional Radiology; Vice Chair for Interventional Services, Beth Israel Deaconess Medical Center and Harvard Medical School; Beau Toskich, MD, Director of Interventional Oncology, Mayo Clinic Florida; and Roxana Dronca, MD, Site Director, Mayo Clinic Comprehensive Cancer Center, Mayo Clinic Florida

Read More »

Dr. Ahmed: I'd like to introduce two very experienced individuals within oncology. I'll start with Dr. Roxana Dronca. She's a medical oncologist and Site Director of the Mayo Clinic Comprehensive Cancer Center at Mayo Clinic Florida.

We're also joined by Dr. Beau Toskich, Director of Interventional Oncology at Mayo Clinic Florida. I am Dr. Muneed Ahmed from Interventional Radiology at Beth Israel Deaconess Medical Center and Harvard Medical School in Boston.

Dr. Ahmed: As a disclaimer: This non-CME program, presented with support from Replimune, is intended only for US healthcare professionals involved in the treatment of adult patients with advanced melanoma. Information contained herein is intended for general information purposes only. It is not a substitute for your professional medical advice and judgement. This webinar includes a discussion of investigational therapies whose safety and effectiveness have not been evaluated by any regulatory agency. The experience of any patient during the clinical trials discussed during this segment may not represent the experience of all patients who received this investigational agent as part of the clinical trial.

And important: RP1 (vusolimogene oderparepvec) is an investigational therapy, and it has not been approved by the US Food and Drug Administration or any other regulatory agency outside of the US. BLA has been submitted for RP1 with a PDUFA action date of July 22, 2025.

Dr. Ahmed: Alright, I think we can get started. Thank you both for joining me. We'll kick it off by getting an overview of how we think about intratumoral therapy. What that class is, how that's starting to fit into oncology, and maybe specific to melanoma as well, where you see that coming?

Dr. Dronca: Yes, happy to get the conversation started. In melanoma, it's a tumor that is very immune-responsive, right? We have had experience with immunotherapy for many decades. As a fellow, we were in that era of interferon and GM-CSF. And a lot of the treatments did not necessarily work.

But I think, as the treatment paradigm evolved with the immune checkpoint inhibitors in early ~2012 or 2014, we started seeing significant responses to immune checkpoint inhibitor therapy in melanoma. The intratumoral therapy also had its beginning in melanoma in the form of TVEC, which was the very first intratumoral oncolytic viral therapy. A genetically modified herpes virus that was actually studied and approved by the FDA for treatment of patients with stage 3BC as well as stage 4 melanoma.

Dr. Dronca: Now, the experience that we have had with TVEC in the clinic is that intratumoral injection is feasible. It's fairly safe. It is tolerated by patients and doesn't really lead to any concerns about disseminated herpetic infections. It is, though, a treatment that mostly works in the lesions that you are injecting. In the clinical trials with TVEC, and in the practice, we really have not seen significant activity in distant lesions. So, if you have a patient with in-transit melanoma or skin or subcutaneous lesions but also lung or liver disease, it is extremely rare, if at all, that you will see responses in the lungs or in the lesions that are not injected.

We have also not ventured into injecting visceral lesions. At the most, with TVEC you inject skin metastasis, subcutaneous metastasis, potentially lymph nodes. The treatment with TVEC—the experience has really evolved in the management of patients with locoregional and in-transit disease.

It took a little while for practitioners to get comfortable with this and for the logistics and workflow to be worked out in the clinic. Every institution does it a little bit differently. In some institutions, the oncologist injected or the nurse practitioners in oncology. In other institutions, it's the dermatology colleagues. Other institutions it's the surgeons. And I think Mayo has evolved through all of that.

At Mayo, Florida, for instance, my colleague previously who was treating melanoma patients would inject himself. He would have scheduled one of the rooms and most of the TVEC patients around that particular day and then performed the injections himself for the subcutaneous and the skin lesions. Then it kind of went to our colleagues in surgery. Interventional radiology occasionally would be consulted to inject ultrasound-guided lymph nodes, etc.

Now we have a workflow where a lot of these lesions, the patients we're referring are the in-transit disease patients for injection of cutaneous and subcutaneous lesions. It's actually now worked out pretty seamlessly. We put a consult in for dermatology. We say, "This is a patient that we want to have considered for TVEC injection." Dermatology then sees the patient, assesses whether this is a good candidate, and then they schedule the treatments themselves after that. It kind of goes on in parallel with their oncology care.

TVEC was designed to be a treatment that is given by itself, but we have started combining it a lot with immune checkpoint inhibitor therapy in the clinical practice, kind of "off-label." There was a clinical trial that explored the combination that did not show any benefit over either therapy or over immune checkpoint inhibitor alone.

But I think, in practice, a lot of folks do both. Patients get treated in dermatology with TVEC, in oncology with immune checkpoint inhibitors, and then we consult with each other or we do the imaging and then we decide whether the treatment should continue based on response. So, I think that is, at this point, most of the experience that we have with oncolytic viral therapy in cutaneous melanoma, at least.

As we probably will discuss a little bit later, RP1 is different, based on the clinical trial data suggesting that you can inject visceral lesions. Not only can you inject; it's also recommended, because patients with deeper lesions injected seem to respond better and deeper. Also, you can inject some lesions and other lesions respond. So now you get to consider what is the best lesion combination to inject in such a way that you get to the most lesions. And you see responses systemically, plus you know this is a treatment that likely will be approved with an immune checkpoint inhibitor, because that's how it was studied.

So, we will have to continue the treatment in oncology, and then a treatment in IR should be scheduled in parallel. It's a limited number of cycles with RP1, rather than TVEC that keeps going. There are a few differences, but I think we can learn from the pathways and the workflows that we've created for TVEC.

Dr. Ahmed: I appreciate that. It's a great summary and description of the progression of this and where we are now with potentially visceral agents.

Beau, can you talk a little bit and build on this, because you were one of the early sites for this trial in particular as well. So, you have experience using this agent [RP1] for visceral injections. What do you think or how does interventional radiology think about best practices around injecting these tumors, selecting some of the sites or patients?

And then, really building on what Dr. Dronca said, how do we think about integrating interventional radiology in the care pathway? Is it a referral? How do things get scheduled? How do you approach some of these questions?

Dr. Toskich: Yes, I think it starts off with the interventional oncologist understanding the potential impact of the therapy. Because the more discretionary enthusiasm you get helps push an early program through its early challenges—which all of them will have—but they're definitely solvable. When we had our first experiences of viral oncolytic therapies, this was in the mid-2010s for HCC (or at least that was my first involvement), a lot of these were really unclear as to what clearly was the oncolytic benefit that was at hand. What task were we solving?

If I can advance a needle into a lesion, and let's say it's a 2-cm lesion. Well, if my needle is there, we have ablation technologies that can eliminate that lesion. So, what we start realizing is it wasn't that kind of patient that was being sent to us. It was a patient that was clearly not a candidate for something like thermal ablation or even certain transarterial therapies where it was more challenging disease. But if we were placing a needle and infusing something that had a response rate of <10% in only localized response, and it wasn't doing anything for a more challenging presentation for the patient, the wind in the sail wasn't quite there for us to handle the logistic issues or learn more about an alternate mechanism of action within a therapy, etc.

I do think that the general concept of my simplistic view of this biological warfare against tumors is still very exciting for the IRs. Being able to use the tumor's inability to protect itself from a pathogen like a virus and using host mechanisms that are highly capable, because through years of evolutionary biology, that's all something that's been maintained. It just makes plenty of sense.

And I think that once you put in smart elements, like something that makes the target make end products that work against itself. Suddenly, even though that's Star Wars, conceptually, people are very excited to get involved, even if it is "just a needle being advanced in something," because we do think that there is the potential for this obviously to have a greater impact. So, once the team is aware of that potential, and certainly now, with some of the data we've seen upcoming from oncolytic therapies, particularly for immune-responsive tumors like melanoma, people are excited. And I think that is sort of a critical mass to generate momentum. People do the necessary work.

So, what does that work look like? And where do we see it? To where we've learned lessons and where we've been able to create a stable foundation for not just melanoma therapies, but we actually do viral oncolytic injections for multiple different therapies.

I think it starts off with recognizing that interventional radiology is your friend in cancer care as a disease specialist and not your vending machine, not a commodity, not sort of, "I dial in and get an IR procedure like you would get a candy bar." But it is hard to change that narrative or perception, because we also have a lot of citizenship work that we do within the hospital, like placing a central line, which doesn't involve having to have someone be an expert in cancer to do so. IR has this broad responsibility.

But that's where I think things like interventional oncology are so important, because it defines this as really its own specialty within a broader way of treating disease. I have friends who are amazing vascular disease specialists and can recanalize tibial arteries and keep them open, and they don't necessarily have those skillsets translate into understanding biology of disease when it comes to someone with MSI-high colorectal cancer. I think that it's really understanding that disease expertise, understanding that there's an element to this that's outside of a commodity. We make it feel like a commodity because of convenience, being able to do this outpatient, and low risk, etc.

But I do think that vision of the interventional oncologist being a key player, just like dermatology was mentioned, just like surgery was mentioned. IR is the exact same level of disease expertise. And what that means is involving the IR early in your program.

What does that look like? Some of the lessons we learned in early viral oncolytic studies was a protocol is already matured, signed off, budgeted, and stamped, and you would open the letter, and it wouldn't fit with what we were able to do from a workflow standpoint. Our nursing when it comes to recovering patients and to things like terminal cleans for certain pathogens. They weren't even involved, and they had to backtrack, and that was a loss for activation times. But it was also a loss of opportunities to really improve and refine these things, because this is an early technology, as far as if you look at its general use. And there's always room for improvement. So, you're missing out on that aspect of the MDT.

I do think involving your IR first is important. Second thing is, a lot of these patients are going to require repeat injections, and then you build a relationship with them. So, they're going to trust their doctor that's treating their cancer. And I do think it's important to have folks that are impassioned to do this. It may not be someone's favorite thing. Maybe they don't like using ultrasound. Maybe they don't like following research protocols. There's different IRs out there, but you find the one that is going to be appropriately geeked out for that opportunity to help a patient with something, particularly with these kinds of numbers we're seeing nowadays. And then, you build that core team of folks.

That also means that sometimes you need more than one IR. Because if you're doing something on research protocol and you have scheduled injections, if there's a vacation or an illness or whatever, you don't want to have someone who really knows nothing about it. Because that's how you get deviations, and that's how you have errors. So, I would say, find at least 2 IRs who are really into this and involve them early. And they're going to involve their teams early, whether it's nursing, whether it's the technologists setting up the room. That's all important to be polished, so that the patient has a very smooth experience, so that there are no errors. And again, this is something that, once this commercializes, you can translate easily into practice, and it's not felt to be disruptive. Sometimes people can say, "Oh, these new therapies! It's not as smooth as Y90 or it's not as smooth as ablation or whatever." And it's because, well, you haven't done the iterations.

So, I do think getting involved in particularly promising therapies at the phase 3 level is very important for IRs to do if they are in an academic setting, or at least being aware of them and learning how to not reinvent the wheel once this does commercialize and they're offering it locally to their patients outside of research settings is very important.

When you talk about the IR's responsibility on the assessment level and intake, it's usually, "Can you get to this?" and "How do you want to inject this?" and "How do you do this safely to where you don't have this pinnacle of technology that's in 5 ccs end up in a vein and just go right to the lungs or spill out into the retroperitoneal fat?" That is where there is a lot of nuances that really cannot be protocolized.

When we first started doing these injections, I made a spreadsheet of all these injection endpoints, and I gave up because there's so many permutations. And I think there's obviously opportunity even for developing these technologies to improve their injectability and feedback during the procedure. But nonetheless, that planning stage is important, because you may think you'll be able to get to it with ultrasound, but the patient shows up on that day and you don't have CT blocked, or you don't have a hybrid room, or whatever it may be. And you can't see with ultrasound, so what are you going to do? Are you going to scramble? Try to find room in between diagnostic schedule? So, definitely pre-procedure planning is key here.

Now, certain lesions you can have a high degree of certainty to be able to inject. You know you have a patient who has very little sub-cutaneous tissue with the lesions right underneath it. It's in an inguinal nodal station or they have a liver where it's a very large lesion, and in your experience, you think you'll see that well with ultrasound. You can go ahead and book it, but some of the deeper, smaller lesions can be challenging sometimes to make sure that you're in it. Your endpoint is not an ablation zone that you can see with contrast afterwards. You really don't have an endpoint outside of maybe some minor change you can see with an ultrasound or a little bit of fluid signal.

So, I think that your planning is probably the most important element that's going to require your personal expertise that only an IR can really provide. And it's really the intervention of deep visceral tissue. Whatever you're comfortable with, and it's sometimes patient-specific, just make sure you have a plan, and if you're relying on something like ultrasound, make sure you have a backup plan in case you cannot see it. Or if you're going to use contrast ultrasound, make sure that's available, whatever it may be. But definitely the planning stage makes it to where, when the patient shows up, you dim the lights, you have a very smooth experience. You go straight to the target, and the therapy is actually incredibly well-tolerated. The patients that have been through multiple surgeries are like, "What are you talking about? That's it?" And I'm like, "Yep, that's it."

And so, again, that brings us a lot of joy to be able to provide a therapy that has a 1 out of 3 response rate, where they barely feel the needle. That's going back to Star Trek. I think that's really a very strong step forward.

But plan, plan, plan, coordinate. Make sure your techs are comfortable. Make sure your nurses are comfortable. Make sure recovery understands whatever the protocol may be. And that also involves the understanding of this virus. So, if you have a herpes virus that has been modified, people are like, "Well, what's my risk? I'm pregnant. Can I be in the room with this virus?" "What if I had spilled this on the floor, what do we do?" It involves, and this may be different per vector, education and making people very comfortable so that there isn't nervousness in the room. And then people finally understand it. Why? Because IR is commonly going to be a place where new ones are constantly coming in, and we have to sort of have an education, pre-procedural briefing, and a debrief afterwards until people are comfortable. And that's it.

After that point, it is an extremely straightforward procedure I'm excited to offer them, because again, patients usually come in, and I can have a conversation with them. Some folks like sedation just because of experience, and we offer it. But I had one patient that every time he'd come and visit, he was telling me about his travels, etc. So again, it's a very enjoyable procedure to perform for patients, particularly when there's promising results. That's kind of a global overview.

Dr. Ahmed: That's fantastic. I want to double down on something that you said to highlight that this is a group of patients that, within the practice of IR, interventional oncology, we don't always really get to see. It's a new set of patients that we have the opportunity to really have a substantial impact on. And I think that's something that, as a community, we should be really excited about as well. I think that's really important.

Can you talk a little bit about how your multidisciplinary program works? How do you guys collaborate? Is there a workflow? Is there a specific review of cases? How do they get identified, like how does one make or replicate the seamlessness you are describing in your program? Dr. Dronca, maybe you can talk about how you think about referring patients in and sharing or getting review of cases from IR. How does that work?

Dr. Dronca: I can start. And then Beau can also add to how we collaborate.

I think it starts, for instance, with identifying the patients, because, as I mentioned, a lot of these patients are being treated with systemic therapies, such as immune checkpoint, combination and or single agent. And at some point, this treatment is considered when patients are, at least now—it may move to an earlier line—PD1-refractory patients, whether it's primary resistance, they never responded, or secondary resistance where they responded but now, you're losing that response and you see progression.

How I usually start is with a phone call. I know my group of interventional radiologists, and it could be either a phone call or a presentation at the tumor board if the phone call is not available right that minute. But a lot of the times, if I can, I call, and then Beau, or one of his colleagues, picks up.

I usually say, "Do you have a minute to look at this scan?" And on the phone, we look together, and we review some of the lesions. And I don't need a comprehensive consultation there. But I need a "Is this a patient that we should see in consultation and evaluate? Seems like there could be areas we could inject." We may need to image further, but yes, this is a patient that at least this therapy should be considered. We would want to see them in a consultation, and then I place a consult.

Then, usually IR sees the patients and then the communication comes once they see the patient. Yes, we're setting this up. We're going to do this alternative route. If it's a patient that's more complex, we chat about the patient and we say we may need additional expertise from the collective wisdom of all the specialties and colleagues that we have present at the tumor board, and then we decide that may involve sequencing. When should we consider this? Or, what type of visit, a risky therapy? What should you know? Do we just need some additional input from colleagues in the room?

So, it usually starts by communicating, really evaluating the case together, and then proceeding with a formal consultation. I want to also pick up on something that both of you said about the biology. The importance of recognizing that this changes the biology of some tumors. You're now able to make them more hot in terms of getting more immune cells in the tumor. Or tumors that initially had responses to immune checkpoint inhibitor therapy, does additional inflammation now make them respond better to immune checkpoint inhibitors? But also you see a systemic response, a systemic immune activation with this virus.

For a lot of patients in whom right now we don't have good second-line therapies or that are being considered right now for TIL. In melanoma, we recently had another treatment introduced for refractory patients, which is tumor infiltrating lymphocyte (TIL) therapy. I think it's going to be an evaluation right now, should this patient go for TIL, or should this patient go for RP1. You're at the crossroads of immune checkpoint doesn't work, our main tool right now, and you've already probably tried targeted therapy for patients with the BRAF mutation. What do we do next? Do we do TIL, or will we do RP1, if this gets approved?

I think here we will have to look very carefully at patient factors: age, tumor burden, ECOG performance score, comorbidities, and then logistics of therapy. Do we do tumor harvest, 3 weeks of chemotherapy, followed by one TIL infusion and IL-2? Or do we do repeated injections for patients in combination with immune checkpoint? It's going to be two different categories of patients that navigate through these, or you may do both treatments successively. But it's definitely an exciting time for us, because now we have more treatments.

Dr. Ahmed: Yeah, that it does. There's a couple of things. First off, I'll say that many of us have Beau on speed dial. You're not the only one!

Dr. Dronca: Thanks for confirming that.

Dr. Ahmed: I think this sort of dovetails into an interesting question, which is, when we think about the landscape of melanoma, and where this fits in? If you have two competing therapies, assuming this, could one use this first and then another one as you think about TIL versus oncolytic viruses? Is there an order that needs to happen? Are they mutually exclusive? Could one incorporate the other elements?

Dr. Dronca: I don't think they're mutually exclusive, but there is an urgency in terms of time, for instance, with TIL therapy. I think there is a window where patients, if you get the patients to TIL too late, you may be losing the opportunity to treat. The reason I'm saying that is therapy involves a number of logistical steps.

You have to first make sure that the patient qualifies from an organ perspective, right? So, pulmonary function test, echocardiogram, potential consultation with cardiology and pulmonary, etc. to make sure the patient is good that way. Then, you need a surgical consultation to evaluate the best place to do the tumor harvest, and then you need to coordinate the OR date with the appointment date, for the company to receive and process the tumor. That sometimes can take weeks. From the moment the company receives the tumor, you have a number of weeks to get the product manufactured, usually about 5 to 6 weeks, and then you have to admit the patient and do chemotherapy. So, all the steps are a number of weeks to months.

So, if you have a patient with a rapidly progressive disease, you need to get on this right away. If one thinks about okay, maybe we do RP1 first and the patient then progresses on RP1, how much time do I have to do TIL therapy, or vice versa? If I wait for TIL therapy, can I do RP1 if the patient progresses to TIL? Or even as a bridging therapy? I am sure that now that we have both, we're going to get very creative on how to get the patients to potentially benefit from both if they can.

There are some patients that are not candidates for TIL: patients who are above a certain age, patients with comorbidities that will never get to TIL. So immediately, we'll be considering RP1. There may be patients who could not get to RP1, because RP1 is injected or is used with an immune checkpoint inhibitor therapy. I have a number of patients in my practice where they saw myocarditis with a history of neurological side effects. With myasthenia, I would not be able to reuse an immune checkpoint inhibitor therapy. So those patients, would we consider them automatically to go to TIL? Or will we venture in using RP1 alone, similar to our experience in solid organ transplant patients? Which I think is a consideration as well, given the response that we see in solid organ transplant patients, which is around 30%. So, like I said, I'm sure we'll get very creative.

Dr. Ahmed: It's going to be a really interesting landscape. I think that you raise this question of timing and also use of one to bridge the gap for the other. I think this is really an interesting intersection, because I think there are logistical issues with pursuing an oncolytic virus injection, but they're more logistical in the moment, not as the setup of therapy.

Dr. Dronca: Correct. They are more logistical in the moment, but once you figure them out, I think it goes a lot smoother, yes.

Dr. Ahmed: As I understand, using these types of therapies, superficial lesions can also be injected. So, if it's superficial, do they go in one route? Does the interventional oncologist do all of the potential things, superficial and deep? Is there some back and forth over which lesion should be injected, and where they go in terms of workflow? Have you guys come across that at all or does it all sort of get done in one place or one process?

Dr. Dronca: Beau, I'll let you start and then I'll venture my opinion as well after.

Dr. Toskich: Sure. Maybe if I can backtrack just for a second on your previous question, I do want to emphasize Dr. Dronca's answer is very granular. And look at how much she's thought about the ins and outs of all of this, and I think that comes from something that she has taught all of us in the cancer centers. No matter your experience, your involvement in research, you're a patient advocate first and foremost, and you have to see it from the view of the patient.

And I think that level of granularity is difficult to capture on trial, but it's how these therapies ultimately manifest themselves as being impactful for patients. And so, I'm always excited when there are multiple therapies, because I rarely see them as having a purely overlapping Boolean. They almost always have a nuance that can be used. She mentioned solid organ. And again, when it comes to treatment failures, particularly when you're leveraging the immune system, there are immunological compartments all over the body. That's why we see what uveal melanoma does to the liver. That's why we see people that have distal KRAS wildtype colorectal primaries that show up in the lungs, and you ablate them, and they'll go 5 years without having their lesion.

So, these immunological compartments mean that there isn't going to be a one-size-fits-all. And I think obviously, the prospective data is necessary for FDA approval and for clarity. But how we pick and choose this is going to be very tailored to the sites, and I think what should be driving that is patient advocacy, not anything that's any more rigid than that. So, I wish I had a great "this is our computer formula as to how we decide on these patients" answer. But I am proud to say that it is a phone call between medical oncology and IR about a shared passion to help this person. And we look at all the instruments.

I hope that it remains that way in the future. Because again, it's what I enjoy the most about how this program works is, it's totally a phone call. Actually, when we were doing a treatment, I got a phone call. I believe Roxana called me. It was like 10 o'clock at night, but she'd just thought of a very important point that she just discovered in the evening. And of course, we can always talk to each other about patients, but that's something that is organic that cannot be put in a spreadsheet. And I think it's probably the most critical element, communication with a shared passion for advocacy. So again, from the higher standpoint, I think it's so well-received when we have colleagues like that.

You asked me a question, and I think I drifted here. I want to make sure I answer it, so I want to give you the opportunity to ask it one more time.

Dr. Ahmed: Yeah, absolutely. And I'll just say that I think it's really fantastic that you guys have such a great collaborative relationship, and again, it may not be as present between melanoma experts and IRs at other places, but it's definitely something that we should all be aspiring to. You framed it very well that it isn't all about spreadsheets and tumor boards, necessarily, but it's really about the shared passion. And I think that's what we, as representatives to communities, need to continue to foster.

The question I had asked was around, how one thinks about superficial and deep lesions? What that serialization injection plan looks like? Does it move back and forth? Things like that?

I'm going to ask one other part to this while you think about it. Which is the other piece is that it occurs to me that it is a bit of a dynamic plan, which is that you're injecting something. And then you're moving through and injecting other lesions. And so, is that an ongoing discussion between the both of you for shared patients? How does that intersection look?

Dr. Toskich: I would say that it's probably going to be different in different places. And you know, Roxana had given you multiple examples of how you can manage without image guidance. You know a good number of lesions when you can identify them and do serial injections. The best resource utilization, I believe, is once you've gone through enough cases that you build a neural network in your MDT.

What I mean by that is you review a couple of cases. Say, all right, I think this is probably going to be okay to be injected. And then you develop a normative experience. And that normative experience, then people say, "Okay, this is something we've handled. And we're okay with image guidance if available. If we tried and weren't able to, we can be quick and provide that without much scheduling delay."

But then there's ones that are clear. This is going to require image guidance, and it's usually visceral. So, I don't know if there's necessarily a benefit to say that all injections go to IR. Just from the massive demand that is actually increasing exponentially with how many things are done in IR within the hospital nowadays, I think that there's nothing wrong with it, but I do think that you will find a stride that works well for the interests and capabilities within your MDT.

And particularly in your early experience, you will communicate as to what is needed, and then you'll just sort of find a comfort zone. But as a rough rule of thumb, if it's within the abdominal compartment, or if it's something that is deep and is not palpable, and image guidance is going to be necessary for a high degree of certainty—not I can kind of inject it, but like this is a very adequate, thorough injection—then that's typically when I would say: call your friendly, local IR.

Dr. Dronca: And I would just add, from a melanoma standpoint, there are two categories of patients, right? There are patients who are going to be locoregional and are going to stay locoregional for quite a while. What I mean is recurrence at the scar, in-transit disease over an extremity or scalp, or skin-based, in-transit disease and/or lymph nodes and local recurrences. A lot of the patients with in-transit disease today are handled, for instance, with TVEC by dermatology.

So, if patients are purely skin and/or subcutaneous, it would make sense for dermatology to do both points. There is not a resource in IR that is infinite, so we want to make sure that we know what to make use of and we call in IR help when we absolutely have to. Dermatology, I think, is happy to see the locoregional patients.

Now, if you have patients who have both locoregional disease and visceral disease, it probably would make sense that patients are seen by IR and injected for all those lesions, because you kind of have to rotate those 10 mLs. You have to try and inject as many lesions as you can. We are learning from the data with RP1 that it's good to actually do not only superficial, but venture into some of the deep lesions again for response.

That doesn't mean that you should inject only deep lesions. If some are at-risk or more complex than others, I think you can inject a few deep lesions and do a few subcutaneous. It also depends on the pattern of disease and progression but recognizing that with melanoma there are some patients who stay locoregional, or local, or in-transit for years. Those patients could potentially be handled outside of IR.

Dr. Ahmed: I want to dig into some of what you've just said now, which is: how should the interventionalist be thinking about picking optimal lesions? Are larger lesions better biologically? Are ones in the liver versus the lung? Once you get out of a locoregional distribution, should we be picking different base organs? What do we know and what are things that we're going to need to learn about?

Dr. Dronca: I think we know a little bit from the data that was presented. I think more data and more in-depth analysis is coming. I think there were patients who had lung lesions injected or deeper lymph nodes injected that responded very well. Some liver lesions as well. I think we need to be careful with the really big lesions to inject, and I think there is a guidance in terms of how many mLs we inject per size of the lesion. So, I think if you only can inject 10 mL total, which is the current dose, you may not want to inject only one lesion that's really, really big.

We have had one patient that we used viral oncolytic therapy in a compassionate use fashion that had so much liver disease that actually went into tumor lysis. So, I think that is something that we need to be very careful with, as well monitoring the patient. Melanoma is kind of a tumor in between the solid and the liquid; sometimes it can behave like a Burkitt's lymphoma. If a melanoma grows really fast, I think we would want to be very careful in monitoring tumor lysis parameters, because as fast as it grows, it can respond just as fast. So, being very careful in how you pick those lesions and the size, and how you alternate, or you divide the total dose into the lesions is going to be important.

I think we have some preliminary data that lung actually responds quite well. Lung always responds well in melanoma to immunotherapy. So, it's a very important site to consider, but I wouldn't necessarily say that other deeper, muscular lesions, etc., would be very good to consider as well. Beau, what do you think?

Dr. Toskich: One thing that we've seen when we inject tumors is that their stromal signature is very different. I remember one of the very first viral oncolytic trial patients I had had a mucinous tumor. And I remember I put the needle in the tumor, and I injected it, and I could literally see the therapy swirling around, and I'm just like, I'm not sure, really what this is. Am I doing the right thing, or am I not? Am I going to try to go into a tangent and inject the rind of this thing of heavily necrotic tumors? Is there a lot of the correct immune landscape within that microenvironment?

I mean, I don't know, but certainly we do see a difference in the interstitial hypertension of some tumors. Let's take a cholangiocarcinoma where you inject it, and it's almost like trying to inject leather. It's not accepting your therapy, and when it does, there's a separation in the tumor, and you get a depot as opposed to an infusion. These are so apparent when you inject that you've got to ask yourself if the effector is being distributed differently.

These are not points anyone can answer at this time, but they are questions that are very valuable to prospectively collect and follow. The injectability of a tumor may be a biomarker, just the ability for it to receive the therapy as opposed to being very cicatricial and not willing to accept virus, may predict something. So, I think right now, we do our best with just common, basic knowledge. Looking at this like play school, Sesame Street level. Like, I'm going to take and distribute this amongst here, and I'm hoping it has the same infusion.

And I think that, if this therapy works the way that I'm hoping that it does, this is more of an inoculation, and it doesn't have to be as precise. Like, for example, for those who are still familiar with ethanol therapy with HCC. You try to get a needle into the tumor and coat the whole thing with ethanol, and if they had a pseudo-capsule, it would get a nice injection, and you'd have some curative outcomes. But if it had an infiltrative border, you couldn't really get the ethanol to do the job. My hope is that those days are over when it comes to a liquid in therapy, because it's an inoculation. And then I think what's more important is, what can we do to optimize injection parameters for inoculation? Not just for things like uniform coverage.

Questions definitely to be answered. I think SIO has a huge role in asking some of those people that have interest in this to dive into those but that's sort of the injection interface. When it comes to dividing all this and trying to make the experience good for the patient as well, I think what's important is to let the patient know this is what the plan is. So again, going back to planning. I think that the injection is super smooth if you lay things out like an architect. Okay, first, we're going to put you in this position here, you're going to feel a needle sting. We're going to inject this. It's going to be over before you know it. Hold a little pressure. Then we're going to get you comfortable. We're going to put a pillow over here, and you can go through these multi-injection plans, and the patient then has a sort of a sequence in their brain that they can play out, and then they're comfortable.

What you don't want to do is to make this feel like you're doing multiple injections, like, when is it going to end, because again, it's not fun to be to be injected multiple times. But I suspect that this is all part of what will be the experience refinement for patients as these therapies become adopted. You'll have a plan. You'll inject things out. You'll know what imaging modality is best for each site. And, more importantly, while you're injecting, you're going to have good confirmation and feedback that there's been a good delivery at the site, because what you don't want to do is just infuse and not have that kind of feedback. I think that that's where IR has a large responsibility in bringing this therapy to the next step.

Yes, tumor lysis, I think, as things become more effective, and you start treating everything that's present in the body, it is 100% one of the pipers to be paid for efficacy. And again, I think going back to planning, just having a plan for that's important.

Dr. Dronca: And I just want to add about the planning. I think it's important to decide what lesions you inject and what size they are. Because I think pharmacy, then, has to make them in pre-filled syringes that they transport from the pharmacy or to the hospital or to the IR suite to be injected. So planning is going to be absolutely key.

Dr. Toskich: Agreed.

Dr. Ahmed: It does seem like one of the key lessons here is about really setting up everyone for success, and I think that's going to be an important kind of continued message. I know we were going to be wrapping up soon, so I just want to build on something that Beau had mentioned about really putting the patient at the center and understanding the patient experience.

Really a question for Dr. Dronca around how do you see what the patient's perspective is here? Are they going to be excited about this therapy? Should they be? What is it? What is the impact or meaning of this really for the patients themselves, who, I think, are already two or three lines deep in some of these cases?

Dr. Dronca: 100%. I'm already talking to patients about treatments. Every single patient that I see wants to know what's next. You know, what do I do? A lot of times we talk about this, even early on with the first-line. Okay, we're going to go through this. And patients say, "What if this doesn't work?" or "What if this stops working? What do I do?" or "What if you have a BRAF mutation?" We go to this.

Then we talk about TIL. We already talk about clinical trials or data that is showing promise. So, a lot of patients actually know a lot more about RP1 and about TIL than we think they do. And right now, because RP1 is not approved yet, a lot of patients are researching. I see a lot of enthusiasm.

Because a lot of the patients who are in this crossroads are patients who are no longer responding adequately to immunotherapy. There is a huge acceptance of immunotherapy in the melanoma community, at least, because immunotherapy has shifted melanoma from a cancer that had 1-year survival on average and a survival rate at 1 year of 20% at best to 50% of patients now are alive and well beyond 5 to 7 years. And 20% of patients are in CR and potentially cured from stage 4 melanoma, including patients with brain mets. So, patients have seen what immunotherapy can do in melanoma.

And I think they do grasp onto the idea that this is an immunotherapy option. That it's different. Initially, when you say it's a virus, you can see, you know, a little bit of uncertainty. And then you say it's a virus that is engineered to only attack tumors and stimulate your immune system to fight this tumor better. There is no risk, or the data suggests there is very little risk, if any, to be infected outside of the tumor cells. I think the beauty of RP1, or the data that was just shown at ASCO, is that it doesn't cause disseminated herpes reactions. It still is very susceptible to acyclovir, is killed by regular disinfectants, doesn't need terminal cleaning, it doesn't ooze. So, you see, the biosafety is there. And I would say from a patient perspective, knowing they have another powerful tool that can have a response in 30 to 40% of patients is huge.

Dr. Ahmed: Well, it's going to be an exciting development in the space, and I think, for our mutual collaboration. I think we can wrap up. We've probably gone a little bit over. I appreciate everyone's time.

 

 

 

The following Case Spotlights are supported by Genentech. All content is the property of, and has been developed by, SIO and includes opinions of the participating physicians.



A Conversation Between Muneeb Ahmed, MD, Chief, Division of Vascular and Interventional Radiology; Vice Chair for Interventional Services, Beth Israel Deaconess Medical Center and Beau Toskich, MD, Director of Interventional Oncology, Mayo Clinic Florida 

Read More »

Dr. Ahmed: It’s great to be here with you today. Both of us have probably been practicing long enough to have lived in an era where for HCC, interventional therapies were the main treatments offered almost all the time with very few systemic therapy options available. In recent years, that has definitely changed. Can you reflect on how that changing landscape is starting to influence how we practice clinically and where newer data fits into your practice?

Dr. Toskich: Yes, so how has changing systemic therapy changed interventional practice?

Dr. Ahmed: Exactly.

Dr. Toskich: When I first started, the thought was treat as much HCC as you can within IR, and when there were no options left, patients would transition to Sorafenib. The perception of the interventional radiologist at that time for systemic therapy, was – well, “let's just delay progression as long as we can.”

Ever since the advent of immunotherapy, systemic therapy now means something different to the interventional radiologist. The notion that current systemic therapy efficacy is being largely driven by the immune system is of great interest to the interventional radiologists, because it is the patient's body as opposed to a synthesized agent that is attacking the tumor. Many interventionalists I speak with are open to the addition of immunotherapy with local therapy earlier in the disease process because it is generally well tolerated and not mutually exclusive. In fact, combination therapy may provide additive and even potentially synergistic benefits to patients, and - and when this connects - we provide something far better than "let's keep this from progressing."

These changes have brought cooperation, inclusivity, and a combining of forces amongst disciplines that was organic – no one ever told us to do this – but people believe in the concept. It has certainly raised a ton of new questions as well. It's going to be a long time before we answer them, but I think they're good questions and represent progress.

Dr. Ahmed: I think that's a great answer and a great overview of many of the things in the changing perspective of interventional radiologists. You highlighted one in particular to touch on which is that in an earlier generation, there was a definitely a push to do as much locoregional treatment as possible, with less emphasis on patient characteristics and / or liver function, because the alternatives really had very little potential improvement on outcome. Therefore, the need to think about that point at which locoregional therapy starts to do maybe more harm even than good was much farther out than it is now, potentially. There is a shift and there will continue to need to be a shift to think about how we are administering interventional therapies and saying, “What is that threshold where should we be layering in these other therapies?” 

The other point you made well is because they are immunotherapies there is a difference in tolerabilities, risk profile, and time to efficacy, effectiveness, and other things that are just different than some of the conventional systemic therapies, or even TKI and other types of molecular therapies. We don't necessarily understand all of that, as we interweave all these treatments together, but I definitely think it is changing the perspective, and will continue to change the perspective, of the interventionalist to say, “How do we bring these systemic therapies in sooner? How can we use them potentially together or who are the right patients for that?”

The other thing I would note is that the staging systems are continuing to evolve. The one that is used for liver cancer, the Barcelona Clinic Liver Cancer staging system (BCLC) is a great example of that. They updated this year their guidelines and recommendations. One of the things that they highlighted is this concept of tumor stage migration and untreatable progression where conventional therapies assigned to specific tumor stages might need to be used at differently (i.e., in earlier or later stages) based on how tumors and patients' characteristics are on the ground, so to speak, and in the tumor board.

What are your thoughts on how that is incorporated into your multidisciplinary tumor board and clinical practice?

Dr. Toskich: There are two points here to comment on. Firstly, I believe that liver preservation should be paramount to any interventional oncologist. There's no point in making all the tumor stop enhancing if the liver stops working, or if you box out future options for a cancer that has up to a 50% recurrence rate at two years after resection. HCC is a disease that, outside the of a liver transplantation, many patients will face on a recurring basis and liver function is their lifeline.

When considering the new BCLC recommendation for systemic therapy in patients with more advanced presentations of intermediate stage disease, my sense is that most providers have been intrinsically accepting of the guideline, despite the absence of positive phase 3 randomized data against the gold standard of TACE, if decisions are made with multidisciplinary consensus. Why? Because as IRs, we have all been there, when we see patients with intermediate disease that are probably not going to respond well to transarterial therapy. Whether it's extensive or infiltrative disease, whether it's a high AFP, we've been accepting of the addition or migration to systemic treatments because we see the potential benefit, particularly with immunotherapy. The acceptance of systemic therapy for intermediate stage disease by the greater oncology community without trial supported evidence does beg why the same consideration does not exist for the addition of local therapy in select patients with advanced disease, where there has certainly been positive signal.

As interventional radiology matures as a separate medical specialty, so does our clinical prowess and patient selection. Local therapy is no longer and order, it is a consultation in clinic. BCLC is a very useful foundation, but our hepatobiliary team is phenotype focused and considers more granular patterns of presentation – again, patient selection is paramount. For example, do you offer ablation to every single 2 cm HCC? Do you apply only systemic therapy to every single patient with vascular invasion? There are a lot of overlapping good options, along with healthy debates, and that's a positive for patients.

The bigger issue at hand is that more personalized we get in oncology, the less likely we'll be able to power studies with 500 patients per arm to meet an endpoint. We will have to become increasingly reliant on well-designed, smaller, study outcomes that move oncology forward incrementally. In the meantime, most decisions will to come to local multidisciplinary expertise until we get more comprehensive recommendations. It's the most challenging portion of our practice - to compartmentalize, to formulate, to boil down to an algorithm. However, I do think tumor boards work when disciplines are well educated with each other's data and have the same mission, the needs of the patient. Irrespective of new drugs or devices, this level of collaboration is probably what generates the biggest impact in our practice.

Dr. Ahmed: I could not agree more, and I think that really is where multidisciplinary collaboration both needs to go and where it is going, because you can't really interweave these different therapies without that discussion.

In our tumor board I'm probably one of the biggest advocates for use of first line systemic therapy and atezolizumab and bevacizumab. In particular in cases that might have traditionally come without question to interventional radiology, but based on, as you noted, the phenotype and the experience of knowing what would be the likelihood of success, starting systemic therapies earlier in situations less suited for interventional treatments.   The strategy you're talking about it also points to situations where interventional therapies are being introduced potentially later or they're being interleaved in a way where a patient gets systemic therapy first and then gets interventional therapies later to treat bulk disease or tumors that don't seem to respond.

There's an ability to mix and match that comes from that discussion. One of the other points you made as well is what the impact of combination effects are and how there is a lot of unknown in the biology of immunotherapy combined with interventional therapies and how they influence the immune system. One of my own points of passion is that it's unclear as to whether one should get immunotherapy first or after a neoadjuvant therapy.

In reality, immunotherapy given first might actually be advantageous in terms of priming the tumor that then may also be responsive to local effects from various local therapies, of which there are many choices, all of which have different immune effects as well. There is so much unknown in all of that space, which is a great opportunity to continue to study as well.

Can you talk a little bit about specific scenarios where you see the most change in your practice? You mentioned phenotype a little bit if we were to talk through scenarios where that plays out most commonly. In my mind, maybe it'll be the intersection of different stages, or certain phenotypes definitely where we see that kind of influence more often.

Dr. Toskich: Sure. If a patient comes with metastatic or extensive bilobar disease, the backbone of treatment is clear – systemic therapy. But, what may be nice us to discuss are the instances where interventional radiology raises the notion of systemic therapy for their patient. It seems to come down to three basic categories.

The first one is the patient with "early advanced disease." They have macrovascular invasion, but it's limited to a lobe or a segment and we can treat their disease locally without compromising liver function. Many IRs, surgeons, and radiation oncologists offer local for high-risk disease and adjuvant studies are underway, but most see no reason why systemic therapy should exclude other effective treatments. There are numerous other GI malignancy precedents which offer local therapy and systemic therapy as guideline supported protocols – this is not a novel concept. Our group has published on successful downstaging to transplant with a similar combination approach resulting with complete tumor pathologic necrosis and the potential warrants exploration.

The second one, as mentioned earlier, is the patient who you think is not going to respond well to your treatment, or local therapy is going to be unsafe. A classic example is the patient with intermediate stage disease that has milliary subcentimeter lesions and nearly the whole liver is restricting diffusion. Yes, technically speaking there is no vascular invasion, there is no disease outside of the liver, but you cannot provide selective therapy to that patient. Our group support those as first line systemic therapy scenarios, and it does not mean that IR has not future role. We've performed local therapy as a salvage to failed systemic treatment in rare cases which ignited systemic response. It's case reportable, but that doesn't mean that we stop investigating how it happened, or stop trying just because we cannot predict who will benefit at this time.

Finally, I refer to this one as the “discordant presentation”: this is a patient who has a three and a half centimeter tumor, but the AFP is 4,000. This is someone who we know is going to have bad biology, is high risk for local recurrence and metastases. While we are essentially using a serologic marker as a surrogate for microvascular invasion, this could be applied to any concerning phenotype, such as patients with disrupted pseudocapsules, large tumors, or in the presence of satellites. Ultimately, this disease presentation is where we are interested bringing bringing systemic therapy to BCLC A, because as good as the BCLC staging system is at unifying and trying to get the world to practice with some commonality, it's very difficult to have all those patterns in an algorithm. To help standardize this more granular approach, we are currently developing AI to identify high risk phenotypes as part of a transforming liver transplant initiative that has the potential to reproducibly identify risk outside of tumor size and number alone.

I think those three are some of the strongest current reasons for IRs to knock on systemic therapy's door and say "come on in and help out!"

Dr. Ahmed: Yes, we practice in a very similar fashion. In all of those instances, we practice almost identically in where we think about interweaving systemic therapy early or perhaps outside of the standard study-based parameters.

I’d say that the other cohort are people who have liver-dominant disease and a few sites of oligometastatic tumor outside of the liver. For example, the patient that has a a few metastatic lymph nodes at one or two sites, but has an eight centimeter tumor in the liver, where penetration of systemic therapy and the effect of that may not be as robust for the dominant liver tumor.  This is where we might combine therapies even though they fall conventionally into in a later stage grouping or category, particularly if we're concerned about the rate of growth or intrahepatic progression, which is its own predictor of mortality and outcome in these patients.

Then you have the last group, which is interesting too, where the medical oncologist brings a patient back to us as there has been oligoprogression. This where the patient is on a systemic therapy regimen and has two small sites or three small sites that are growing. By the conventional medical oncology review, one would say that they have disease progression, and need to be switched to something else, but all the other sites seems to be well-controlled. So we have those instances as well, where we are treating them on a case by case basis to allow the systemic therapy agents to continue to be used for a longer period of time.

That is another hot area of interest because there is now data across other cancer types as well around this concept of immune escape in lesions and the value of potentially treating some of those with local therapies as well.

Dr. Toskich: Yes, the concept of immunologic compartments being different in your nodal basin, as compared to your liver, lungs, or bone, is a great additional point.

The notion of trying to separate all hands on deck sequentially may not necessarily be the best way to fight cancer. As mentioned, there are so many examples in GI oncology where you have “chemorads” protocols. I see similarly in the liver, particularly when using radioembolization for more advanced disease, I see no reason why you can't explore “immunorads” in the liver.

It all comes to addressing the individual aspects of any given patient's condition that presents a hazard to life – and doing it with agility. Is it their liver function over time? Is it the systemic taxation of chronic disease burden? Is it local hazard such as biliary or portal obstruction? Do we have an opportunity for transplant? I think these are all worth engaging in real time and may uncover benefits of not treating everything with a solitary hammer, whether it's infused through a catheter or through an IV.

I honestly think we're getting there. I am impressed by how many people can quote other disciplines data in depth at my tumor board these days. We're co-educating each other and it’s really an exciting time to help patients with HCC. As matter of fact, I recently had to unexpectedly cancel a scheduled interventional oncology talk at the last minute due to an emergency. I reached out to my medical oncologist at the time and asked if he wanted to give the talk instead. He not only agreed, but knocked it out of the park. That was a great litmus test for multidisciplinary communication. 

I believe this is the way forward. Ridiculous turfs are set aside, and we start thinking about how we generate durable responses with low adverse events in parallel. Overall survival data is great, but it's going be really hard to get when we have a million permutations of therapy. Despite the amazing advancements of systemic treatment, tumor response is still not greater than 50%. If patients are interested in rolling the dice for immune activation in addition to concurrent local therapy, with the potential for a favorable DOR/AE ratio, why not try? That is why we are seeing interest in local and systemic therapy trials for even early stage disease, and patients are equally invested.

Dr. Ahmed: You made a lot of excellent points. The nuance that is now in place in making actual treatment decisions is really both quite interesting and exciting. It really does speak to all the different treatment options and choices that people have. You outlined a really great example of the engagement of the different oncology treatment communities. On the ground, on a daily basis, and with each other is really a point of collaboration that is just going to be a great example of where medicine should go in a multidisciplinary way. HCC is an evolving entity itself, in part because of advances in liver disease. The overall development of fatty liver disease and its implication in HCC development is a different biology so there's a lot of different sort of changing pieces as well makes the space very interesting.

Do you have any advice for people coming into practice? You and I both have very robust training programs. When I have fellows and trainees, the complexity of some of the discussion at multidisciplinary conferences can be overwhelming sometimes because of the rising amount of data and treatment options and patient considerations.

Is there any perspective or advice that we should offer people who are earlier in their careers on how to think about engaging with their medical oncologists to understanding key elements to data, practicing in real time, as you alluded to?

Dr. Toskich: The best way I would approach that question would be what would I have wanted to know when I was finishing my fellowship. The first thing to realize is that you're not going to be successful alone.  You need to team up with good hepatologists, oncologists, surgeons, and a transplant team with whom you can foster a mutually respectful relationship. That's so important because it’s the human element that cannot be protocolized.

People who get along, who respect each other...that benefit translates to better patient care. I believe in that. So start early, start from day one, be humble. Learn from these people, they can teach you amazing things. My passion was HCC since I was a resident, and yet the knowledge and experience blind spots I had in my first years as an attending were massive. Be aware of those blind spots and rely on people you trust to help you eliminate them.

That brings me to my next point which is, a lot of people within interventional radiology, (although I do think this is changing) are overly focused on the technical aspects of therapy. What is a harder and more fruitful question is “should I intervene?” and “when?” Patient selection is key. That is what people should be asking when they call their mentors, not to find out what catheter they use or some awesome tricks to perform with it. You will develop your own tricks one day, that's going to come naturally.

Collecting and communicating your outcomes is next most important step. Most people have adopted and the advances of systemic therapy, but I think that it's human nature experience a pendulum swing and forget established treatments that are effective. Think about the patients who are coming back transarterial therapy for neuroendocrine disease who were poor responders to theranostics. Embolization never went away, it was merely a pendulum swing and having your local data will help keep the best options for a patient at the forefront. This is human nature and it is normal. But it is up to the interventional radiologists to champion what we know does well for patients, what is it that is the crown jewel of our practice that needs to be protected. The best way to do that, if you're in academics or private practice, is to continuously know your outcomes and hold others to that standard. This will also allow you to identify which aspects of your practice are worth preserving and aspects that maybe aren’t working as well and consider moving on to the next opportunity.

Now I have a question for you. If you incorporate an aspect of your practice into a nonexistent guideline right now that works well for your multidisciplinary team, what recommendation would you make?

Dr. Ahmed: I would say, not something that we necessarily do well in our practice, but what I would like to see more broadly is the incorporation of biologic data more aggressively in both studies and ultimately in treatment guidelines. I do believe that in HCC because the imaging has become, and has been for a while, so accurate in the diagnosis of HCC against non-malignant entities, that the perception of not needing biopsy data and/or biologic sampling is very heavily embedded as a culture within the treatment of HCC.

Actually, what we're appreciating now, whether it's in these aspects of HCC, and also in liver disease in general, is that our stratification needs to be much more nuanced. We are not going to be able to do that, particularly on the immunotherapy front and on the locoregional therapy front, without some additional sampling of tumors, acknowledging the limitations of what that sampling entails. We are starting to see some of these studies for locoregional and interventional therapies, where the peripheral blood is sample to look for secondary signals of response after local treatment, or to help stratify patients who to predict response or progression.  There is quite a lack of that in the HCC space compared to other GI malignancies, lung cancers, and other sort of cancer types where correlated genotype or phenotype by sampling to radiological imaging and response is farther along.

I'd love to see that culture expand in HCC, and we actually can be strong proponents of that because we are the people doing the biologic sampling. To your point, if one thinks about the patient, then the treatment biopsies have added value and we can be strong advocates of incorporating that into studies and clinical practice.

Dr. Toskich: I agree, and that is what I was hoping you would say, because biomarkers exist in so many other malignancies, but HCC biomarkers have shown potential but have not panned out. That doesn't mean we should stop trying.

I think being smarter in our approach – whether combining systemic and local treatments – for a particular patient’s pattern of diseases earlier in their diagnosis is going to provide oncologic leverage. And that's really where I am hopeful of our upcoming biggest strides.

Dr. Ahmed: Great. This has been a great discussion, thank you for speaking with me today and sharing your insights.

Dr. Toskich: Likewise, thank you.



A Conversation Between Isabel Newton, MD, PhD, Chief of Interventional Radiology at the Veterans Affairs San Diego Healthcare System and Associate Professor of Radiology at the University of California, San Diego, and Dr. William Rilling, MD, FSIR, Professor of Radiology and Surgery and Vice-chair of Clinical Affairs at the Medical College of Wisconsin.

Read More »

Dr. Rilling: Hello, Isabel. It is great to be here with you today.

Dr. Newton: Great to be here with you, as well. Today we are discussing some of the changes to how we are taking care of our patients with liver cancer, in light of the new BCLC guidelines and some of the new therapies that are available.  I wanted to run some patients by you and see, in your opinion, how you would treat them, what is happening at Medical College of Wisconsin (MCW), and how it differs to what is happening down here in San Diego, especially at the VA.

Dr. Rilling: Sure, that sounds great. It is an exciting time right now for HCC treatment, as you know. A lot of advances in systemic therapy for HCC and in liver-directed therapy, and a lot of shifting of the guidelines are going on, so it is a great time to have this conversation. Let's talk about some of your potential patients.

Dr. Newton: I have this one patient: he is a 67-year-old Hispanic male, and he has NASH cirrhosis and bilobar, multifocal HCC. His performance status is 1. We can talk about what that means. His total bilirubin is 1.8, INR is 1.2, albumin is 3.8, platelets are 120, and creatinine is 1. According to the new BCLC guidelines, he falls between an intermediate stage B and an advanced stage C. I am curious how you would approach him.

Dr. Rilling: A lot of interesting points about this patient: first, as you alluded to, the issue of performance status as a subjective assessment in BCLC is something that is pretty controversial. Remember that patients who have anything other than an ECOG of zero are automatically put into the advanced group, which, for an objective measure, is sort of over-staging. Certainly, some of these patients you and I have talked previously about, such as people with cirrhosis without liver cancer, often have some degree of fatigue, if you really drill down on it. So, do you count that if they have chronic fatigue? Because, if they have cirrhosis or other things going on with them, such as diabetes (which this patient probably has), it is hard to tease out the symptoms from these chronic diseases versus true cancer-related performance status changes. That whole conversation is challenging, and where you put these people, based on performance is obviously really subjective when compared to other measures in the BCLC, which are quite a bit more objective. I assume that this person is outside Milan criteria. He is not a transplant candidate right now but because he is a young man, if there is any chance to get into a curative-intent therapy, we would want to try to do that.

Assuming he is outside of Milan criteria, his liver function is not awesome, but it is not terrible with total bilirubin 1.8, and his albumin is down a little bit. What we do for this patient would depend purely on their tumor burden and the size of distribution of their tumor. With a bilirubin of 1.8, we certainly do not want to treat his entire liver. We want to be careful to try to preserve his liver function as much as possible. It has been suggested that regarding immunotherapy in a patient like this with NASH cirrhosis, his response to immunotherapy might be less robust than patients with hepatitis-related cirrhosis, so that plays into the decision-making process.

With this patient, we would probably try to do liver-directed therapy, and, if we could treat most of the tumor burden with only a couple of segments, then we would probably do Y90.  If it was more widespread, then, in the interest of trying to make sure we do not hurt his liver function, we might do TACE as a starting point.

Then we would have follow-up discussions based on his response and tolerance to that sort of treatment around where to go next. The best goal would be to downstage him to be within Milan if there is any way to do that.

What are your thoughts?

Dr. Newton: All that sounds concordant with what we would do.

You mentioned that NASH as a driver of cirrhosis might lead to a poorer response to immunotherapy. Let’s say he has hepatitis C driven cirrhosis instead, or if his disease was not located in discrete segments that you could just take off, are those the factors that would push you towards atezolizumab and bevacizumab?

Dr. Rilling: Those are great points. Certainly, if it is hepatitis C cirrhosis, that pushes us a little bit more towards immunotherapy. If there is tumor in every segment where you are not going to be able to treat him without treating the whole liver, that is another factor in favor of systemic therapy.

The biology of these BCLC-B patients is all over the map. BCLC-C by tumor burden, not by performance status - the biology is also quite variable.  We often first meet these patients with just a single data point of their scans and their alpha-fetoprotein, and then you're going to be learning about their tumor biology as you go forward. Some of these patients have a fair amount of tumor, but it is not rapidly progressive. So you treat them, and you see the next scan, and you either get a response or not, but you are not seeing a bunch of new tumors.

Then there are other patients, where every time you scan, there is disease progression in various segments. Those patients are on a completely different path than the patients where you are treating the same lesions without progression elsewhere in the liver.

If I see a patient who has a couple of new tumors, even if they are really small on every scan, that is a patient I want to get on systemic therapy sooner rather than later. You have more of a field effect going on, and I think it is important to try to get other therapies on board.

Dr. Newton: That is such a good point. Whenever I am establishing care with an HCC patient, I tell them first that we're in it together for the long haul and second, I tell them that our first treatment is the "getting to know you" treatment. I see how you tolerate things, how your tumor responds, how you respond and we'll go from there. The decision tree is wide in the beginning and then you will tell me how we narrow it at the next step.

The point you made before about performance status, a lot of times when we have these patients in our office we get a gestalt kind of a feeling of what they could tolerate. You can assess if are they pretty strong or are they feeble, and all that kind of stuff.

That was really helpful. I appreciate your parsing out of how you would make that decision.

Dr. Rilling: The point you just made about seeing these patients in-person is really critical. With patients, the way they look on their scans or on paper and based on their labs can be completely different from the way they look when they're sitting in front of you.

Literally, you can’t tell. You obviously know their labs are their labs and their scans are their scans, but when that person is in front of you in the clinic, there is so much more information.

We are in the age of virtual visits, and, I don't know about you, but I don't get as much from a virtual visit as I do from having the patient right in front of me. Being able to look them in the eye and ask them questions about what their daily life is like, what is limiting them, and how things are going is invaluable.

Dr. Newton: And you hear and notice the other things they have too, right?

Dr. Rilling: Right, because they usually have comorbidities too. So, I think that point that you made is critical.

Dr. Newton: Alright, I have another patient for you. He is a 55 year old, Caucasian man, and he has alcoholic cirrhosis, so not NASH, and he has HCC that is pretty advanced with portal invasion. His performance status is zero. This patient passes the eyeball test. He does not look sick. He looks great; in fact, he is a muscular guy. He came in, he drove himself and he is fine.

But, his total bilirubin is 2.5. His INR is creeping up, it is 1.8, and his albumin is 3.3. He had some reserves to be able to now have an albumin of 3.3 and look as good as he does. His platelets are 85 and his creatinine is 1.5.

In your experience, how would you approach this patient?

Dr. Rilling: Boy, this is a challenging patient as well. It is tough to see these young patients with a terrible problem like this. These patients who present with vascular invasion are in a tough situation.

First of all, this patient is obviously in the BCLC-C category, by definition, with the portal invasion.

I think it is worth discussing a little bit that one portal invasion is not exactly like another portal invasion.  The branch PVT are different than a main right or all the way into the main portal vein. In our practice, how we approach these patients and definitely the degree of portal invasion is taken into consideration along with, obviously, hepatic reserve here, which is a big problem with this patient. What you would really like to do differs from what you will end up having to do because of concerns of his hepatic reserve. The portal invasion is really rough and, by definition, a bad marker for tumor biology for this patient.

Even though he is a young person and, by looking at him, you could be aggressive, we have to be concerned. When we see an elevated bilirubin, we always make sure we get direct and indirect bilirubin to make sure it is not Gilbert syndrome that is artificially making the bilirubin look worse and it is actually not a synthetic problem. This ensures that the synthetic function is fine, even though the bilirubin is so elevated. This gentleman, by the guidelines, would get systemic therapy first line, probably atezolizumab and bevacizumab.

The thing that you are worried about is if he does not respond, you have a narrow window in these patients. While his liver function is not great, if he progresses, you may lose the window to be able to try to do some locoregional therapy. The discussion we would have at tumor board would be, if we start systemic therapy, he would get a couple of cycles, and we would check very early to get a sense for whether this is working or not working. We would be very quick to switch gears in a young healthy patient like this if it is not working right away.

Dr. Newton: That is an interesting approach to be very vigilant. Other institutions would just go straight to Y90 with a patient like this and leave the systemic for in case he progresses, which is not entirely in line with these new guidelines, but it is kind of the local practice.

What are your thoughts on that given his age? If he were 85 years old or 70 years old it would be different, but he is 55 years old.

Dr. Rilling: Yes, and again I think the degree of portal invasion is important.

We struggle with this at our tumor board honestly, which way to go in the order of therapies. At the end of the day, a lot of these patients end up getting both locoregional therapy and systemic therapy. I don't think we know at this point in time, from a data standpoint, which one of those is better and which to start with and exactly how to integrate them in an optimal way. That is obviously going to be the subject of a lot of research over the next couple of years.

The other thing, as you are well aware, is looking at the Y90 dosimetry in a patient like this. If you have one of the personalized dosimetry platforms, to determine based on Technetium MAA scan if you can get over 200 Gy into the tumor and not hit much of his background liver, then, by DOSISPHERE data, you would be on good standing to treat this patient with Y90 with a very high response rate to start off. If you find out in your Y90 mapping and dosimetry that you're not going to be able to get 200 Gy or you can't isolate the tumor, then that's another story for this patient in terms of which direction to go.

Dr. Newton: Right. With the data, I think everybody's wondering about the order in which to do it, as you discussed. When you treat with Y90, you are releasing so much tumor antigen that could potentiate or help the systemic therapy. It is going to depend on how the data plays out, but that is certainly the hypothesis that's emerging. Hopefully, we can have more guidance in terms of how to combine these things or the order to do it in.

Dr. Rilling: Yes, and I think if we are going to combine we need to first show safety. I mean there's not a lot of reasons to suspect that there would be significant cumulative toxicity from, say, atezolizumab and bevacizumab and Y90, but it has not been really well studied up to this point in time, to my knowledge.

Just showing that it can be combined, that it's safe to put them together, is an important thing for us to do. Right now, we are not, in our practice, doing them right on top of each other. Especially for bevacizumab: if they have that on board, we're waiting four to six weeks from last dose to do Y90, so that's important. I think we still have a lot to learn about combining these treatments.

Dr. Newton: We have the same practice exactly. At the VA, we don't have Y90 yet, so we send our patients to the university or up to one of the other VAs that has it, and so there is this inherent delay that comes with the bureaucracy of sending a patient.

That's something else I wanted to bring up. Each place locally has its own challenges. You mentioned that with your gastroenterologists being very backed up and how that is impacting how you are practicing. Can you talk about that a little bit?  

Dr. Rilling: Sure.  For the patients that are going to start atezolizumab and bevacizumab, if they have portal hypertension, they generally have to get an endoscopy to look for high risk varices. Right now, our GI service is scheduling three or so weeks out for an outpatient upper endoscopy.  

It's interesting, some of the knock on Y90 has been the delay from when you see a patient in clinic until you are treating them. We actually can get a patient seen in clinic, mapped and treated oftentimes within a week.  We can do it very fast, and we prioritize these patients in our clinic and in our scheduling. We have a coordinator who does a fantastic job expediting the process. We can treat oftentimes less than a week from the time we're doing the mapping.

I think if a new patient comes in, who is going to be waiting a month until they can legitimately get on atezolizumab and bevacizumab, that is a great opportunity if they are eligible for locoregional therapy to do that up front. Then, you're doing something to the tumor while you're waiting for the patient to be able to get started on their systemic therapy. This would be a great patient to do that on.

Dr. Newton: Yes, I agree. Alright, I have my last patient I would like to pick your brain on. He is a 72-year old, Filipino male who has HCV cirrhosis, and his HCC is infiltrative. He is one of those people who, when you look at his MRI, it looks awful. Then you look at his labs and you look at him, and he looks fine. He has preserved liver function and total bilirubin is 1.3.  INR is 1, albumin is 4, platelets are 145, creatinine is 1.1. His performance status is zero. He is still gardening and volunteering at his church.

What would you recommend for this patient?

Dr. Rilling: This is an interesting patient, as well.

This guy, by BCLC, would be sort of an advanced B because of the infiltrative of tumor morphology. As you and I were discussing earlier, one of the big changes in the revised BCLC guidelines is subdividing the Bs. This patient would be more of a high-risk B patient.

There's a lot of discussion out there now about whether the high risk B patient should be getting on systemic therapy as their first line of therapy or after liver-directed therapy, because traditionally the BCLC-B patients have had TACE as the standard of care.

So this patient has many options because of his good performance status and good liver function. I would just say that I get really, really worried about the infiltrative tumors. I think it is almost a worse prognostic sign than PVT. I haven't seen good data to compare those two, but it is a really bad biomarker of how this tumor will act. We try very hard, if we can, to treat these infiltrative tumors with Y90.

Again, this would be a great patient to see what the dosimetry looks like. Some of these infiltrative tumors are sort of hypovascular sometimes.

If you see that you're not going to get a good dose into the tumor, this helps guide our decisions. Obtaining this additional information to triage therapy is extremely helpful. These infiltrative tumors are really tough. You want to be as aggressive as possible with them right away.

Dr. Newton: That is really, really helpful, thank you. You have so much more experience than me, and I respect you so much, I appreciate your insights.

Dr: Rilling: Yeah, I’m just older than you.

Dr. Newton: Well, we do have something in common, which I just found out. We both started our liver tumor boards, and we both run our liver tumor boards, which is not typical that it is IR- run. It is a great advantage. For anybody who's considering starting a liver tumor board, it is some work, but there is a lot of advantage to being at the helm. You see all the patients who go through, you have an opportunity to refine these algorithms to understand the nuances, and that's what we talked about today. I really appreciate your input. Thank you so much.

Dr. Rilling: Thank you as well, this is a great conversation and always fun talking to you.

Dr. Newton: Likewise.

2001 K Street NW, 3rd Floor North

Washington, DC 20006 USA

+1-202-367-1164

info@sio-central.org

SIO Prospectus

IO Awareness Week

SIO Annual Meeting

SIO Learning Center

SIO
Terms & Conditions | Privacy Policy | © Copyright

2001 K Street NW, 3rd Floor North

Washington, DC 20006 USA

+1-202-367-1164

info@sio-central.org
Terms & Conditions | Privacy Policy | © Copyright
Login