Edward Kim, MD

 Dr Kim, do you notice anything with your medical oncologists pattern wise in terms of how they approach these patients?

 Our medical oncologists see them when the patients progress on the systemic therapy. We have inside discussions that for individuals who may have issues coming in, either they live far away and certainly not in New York, but elsewhere, rural areas, maybe an oral agent is easier than coming in for infusions every couple of weeks, 3 to 4 weeks, or whatever the regimen would be. That may be a potential advantage in that type of situation. The majority of our patients now, with the IMbrave150 data, tend to get atezo-bev [atezolizumab plus bevacizumab] first line as long as all those issues aren’t in play for the hypertension, varices, etc. We will combine them with locoregional therapy, usually Y-90 [yttrium-90 radioembolization]. And there are no data to really support this, but we’ve started with the atezolizumab component and held the bevacizumab due to issues with transarterial therapies in bevacizumab, and then once they receive the Y-90 or a chemoembolization, start the bevacizumab component right afterward. But again, this is all in the setting of a multidisciplinary conference.

 You’re right. The future is going to be, how are all these combination therapies going to come into play? As we bring the session to a conclusion, my thought to both of you would be what do you think the future looks like? I personally feel like there’s going to be a struggle in terms of what I do in the second line, particularly if there are combinations that come out, because I don’t know how to look at that space. I’m curious to know, Dr Kim, what do you think from an interventional radiology perspective, and then maybe Dr George, from a medical oncology perspective?

 The future is quite exciting to me because we’re going to start looking at combinations. What has been published to date, a 35%, 40% objective response with the systemic therapies alone, I don’t think we’re satisfied with that. Combining it with some type of other therapy, whether it’s locoregional therapy to incite the immune response, the possibility of getting much higher responses for our patients is very exciting. Also the combination of checkpoint inhibitors, and as discussed previously, the CTLA-4, will hopefully open a whole new bag of tools for oncologists to use. It was quite frustrating in the past when everyone would shout out, “Let’s do Y-90 for this case.” And I’d sit there and say, “I don’t know if this is appropriate for this patient, they have 70% of their liver overtaken with disease.” And then, “We can’t just sit here and not do anything, the sorafenib isn’t going to do much for this individual.” But now it’s different because those patients can truly benefit from checkpoint inhibitors, and if they’re one of the lucky few who are part of that response, we’ve seen incredible and dramatic responses that have altered patients’ lives who were previously told, “Hey, you’ve got 3 to 4 months to live.” I’m seeing them 3, 4 years out now. It’s quite an exciting time for these patients.

 The future is going to be very confusing, as it should be, in a sense that there’s going to be a plethora of options to choose from. We are already seeing therapeutics in the entire spectrum of the disease. We are already seeing the data from neoadjuvant cabo-nivo [cabozantinib plus nivolumab] that were presented. We will see the readouts from the adjuvant studies. We will see the readout from the EMERALD and the LEAP studies that are combining locoregional treatments with systemic therapies. Then we will have readouts from a bevy of metastatic studies with all kinds of combinations, immunotherapy with TKIs [tyrosine kinase inhibitors], immunotherapy with anti-VEGF agents, and by the time those readouts are out, I hope there will be new exciting agents that will be in the phase 1 and phase 2 setting that will be raising the bar even higher. Overall, as Dr Kalyan and Dr Kim mentioned, there will be a plethora of options, and it will be much better for our patients. Ultimately, how do you sequence it? And that may be something where we will all have to think very carefully through, and identifying subsets. Ultimately that may be what is key, identifying the subsets that respond to specific therapeutics. How do you maximize efficacy, minimize toxicity, and put people in various groups or buckets based on therapeutic options? But we’ll see what the future brings.

 It’s an exciting time from a medical oncology perspective. For so long, it’s been so depressing that it’s nice to have all these options. The one thing I would say is patients ultimately see the benefit from all of this. They are the ones who we’re able to have some of these nice conversations with. I just want to end on a nice note. I had a patient who was having ascites on a frequent basic. He finally got a little better and we put him on a clinical trial, and he was just getting single-agent nivo [nivolumab]. He’s now 3 years out of treatment and I just monitor him. I usually give him a call when I’m having a bad day and I’m like, “I just want to see you in clinic just because I feel better when I see you.” It’s nice when we’re able to do that, and I don’t think we’ve had that very often, and this is a nice space for us to think about moving forward.

I want to take this time now to thank our panel. It’s been such a thoughtful and lively and informative discussion. I’m always surprised at how much I learn from all these discussions, no matter how many times you see the same thing, you learn so much. To our viewing audience, thank you for joining us for this 

 presentation. We hope today’s discussion was a valuable use of your time and that you acquired some practical knowledge that you can take back to your clinic. Thank you.

Videos

Closing out their discussion on the management of hepatocellular carcinoma, panelists discuss future novel therapies that may impact the treatment landscape. 

Future Treatment Approaches for HCC

 The REFLECT trial was a phase 3 trial that looked at lenvatinib and comparing it to sorafenib in frontline patients with unresectable HCC [hepatocellular carcinoma]. And we should point out that this study was designed as a noninferiority trial, essentially not trying to beat sorafenib but just to see if we could be at the same benchmark. Patients could not have received previous treatment for advanced HCC, and they were randomly assigned in a 1:1 fashion to either lenvatinib or sorafenib, the primary end point being overall survival. Lenvatinib demonstrated a median overall survival of about 13.6 months, and sorafenib had a median survival of about 12.3 months. And essentially the hazard ratio demonstrated that it was noninferior to sorafenib. Frequent adverse events seen in both drugs were hypertension, diarrhea, and some appetite changes. Perhaps the biggest one that we always see with sorafenib is the palmar-plantar erythrodysesthesia, which was significantly less in the lenvatinib arm. Dr George, do you use lenvatinib a lot given the realm that we’re in with IO [immune-oncology] therapy? I know we talked about the 15%, but are there other patients for whom you consider the TKI [tyrosine kinase inhibitor] for in the front line?

 Maybe 2 or 3 different subgroups, the one that you mentioned, Dr Kalyan, where they’re not candidates for T-cell therapy for whatever reason, autoimmune conditions, or bleeding risk, or what have you. The second, like you said, people who could potentially be a candidate for transplant for whatever reason, if that small subgroup exists, and third, of course, the people who progress on frontline bevacizumab plus atezolizumab, what is the optimal second-line treatment in that patient population? In the absence of data, I have defaulted somewhat to lenvatinib in that population as well. First of all, lenvatinib, even though the data look very nice in terms of the progression-free survival [PFS] compared to sorafenib, that did not translate into an overall survival benefit. And secondly, even with that nice response rate and the PFS benefit, the drug is not super well tolerated either. I struggle with that, patients have a fair amount of asthenia and dysphonia, and overall the hypertension issues. But we can get them through it, it takes several dose modifications and adjustments and hand holding, all of the above. But certainly, it is a drug that can be tolerated, a drug that is effective, and I tend to go to that in the second line after [bevacizumab], T-cell failure. What is your pattern of use of lenvatinib, Dr Kalyan?

 I did jump on the bandwagon of lenvatinib when it first came out, but then I really struggled. I struggled with patients, with their appetite being a bigger issue than I anticipated, and as you said, the asthenia. Those adverse events were bigger than I thought they were, even though on paper, the numbers looked good. I did shy away from using lenvatinib as frontline therapy a lot. The population of patients that we talked about who are not eligible, we don’t have a choice for, we have to go down that route. I’ve leaned toward the immunotherapy option if I’m able to, and certainly if there are clinical trials, that’s a different question. But I’ve been more cautious about lenvatinib because it’s not as exciting as I thought it was going to be. The results are certainly nice that it’s noninferior, but from a tolerance perspective, what I saw on the trial is not what my patients have complained about.

Shifting their focus to the REFLECT trial of lenvatinib in unresectable hepatocellular carcinoma, panelists consider use of tyrosine kinase inhibition in this setting.

Evaluating the REFLECT Trial of Lenvatinib in Unresectable HCC

 The IMbrave150, as many of us know, is a phase 3 trial looking at the combination of atezolizumab plus bevacizumab in unresectable HCC [hepatocellular carcinoma]. Patients could not have received prior systemic therapy, and they were randomized in a 2:1 fashion to either the combination arm or sorafenib, which at that time was the standard of care for these patients. The study was designed with a co-primary end point of overall survival and PFS [progression-free survival] with an intention-to-treat arm. And patients were continued on study until either toxicity or loss of clinical benefit with progression seen on imaging. This is a look at the baseline demographics, and again, if we look here, we can see that this is really primarily geared with Child-Pugh A patients, with the majority of them having an A5 score. Then if we look at the BCLC [Barcelona Clinic Liver Cancer] staging, a lot of them did have BCLC-C, and I would point out at this point that BCLC-C doesn’t necessarily mean that it’s automatically metastatic because it’s variable based on whether there is vascular involvement. And patients did have extrahepatic spread in over 60% of cases.

When we look at the results of this, the confirmed overall response rate for the combination arm was roughly 29.8%, and then when we look at the modified RECIST, it was 35.4% vs 13% for the sorafenib arm. Then we look at the duration of response, it was roughly 18 months versus 14.9 months. Now the median overall survival was updated earlier this year with 29.2 months for the atezo-bev [atezolizumab plus bevacizumab] arm vs 13.4 months for the sorafenib arm, and the PFS was roughly 6.8 with a combination compared to sorafenib at 4.3 months. Based on these results, it met the preset statistical end point, leading to the NCCN [National Comprehensive Cancer Network] giving it frontline approval. Now, when we look at the safety outcomes, again, none of this was unexpected. A lot of us have been using bevacizumab in multiple other cancer types, and we’re used to seeing the expected adverse effects of hypertension and bleeding. And very briefly, I want to also talk about the REFLECT trial, but before I do, I want to get thoughts from the panel about this study and what they think about it.

 One of the questions that frequently comes up is that when you look at that sorafenib median overall survival arm, it’s certainly much higher than what was published in SHARP, the Asia-Pacific trial, at almost 11 months. Maybe patient selection here is not as sick of a population. But regardless, I do think that the 19 month median overall survival in the atezolizumab and bevacizumab arm is quite impressive.

 Some of it is post-progression therapy in the sorafenib arm too. At the time that the IMbrave study was done, there were probably more options for the control arm, sorafenib patients as well, as opposed to what was available at the time of the SHARP trial. Regardless, I can’t argue with a hazard ratio of 0.58, on that study for overall survival. Again, I’d like to hear Dr Kalyan’s thoughts as well.

 We were all very excited by this. There’s finally something that pushed sorafenib off the pedestal that it was on. This is a population that clearly is immunogenically responsive. This has been shown time and time again, and for the first time, we understand that perhaps the combination of immunotherapy with some sort of a VEGF inhibition because of the vascularity of this disease is important, and you’re absolutely right, Dr George, that with a hazard ratio of 0.58, nobody can argue with that. The caveat is that there are about 15% to 20% of patients who we’re not able to give this combination, primarily because of the Avastin [bevacizumab], but certainly people who have underlying autoimmune hepatitis, for example, who get HCC because of that. And Dr George had brought this up, about using the TKI [tyrosine kinase inhibitor] route as an option. I know this is not pertinent to the case we just discussed, but if there’s any utility or any role of transplantation down the line, the data in that space are very limited, in terms of using immunotherapy and then coming back to transplantation. So if there’s any possibility the patient could be considered for transplant, we should keep the door open by using a TKI, at least for now. That maychange as we go through.

A panel of experts review data from the IMbrave150 trial of atezolizumab plus bevacizumab in unresectable hepatocellular carcinoma and consider implications for treatment.

Unresectable HCC: Interpreting Data From the IMbrave150 Trial

 I’m curious to know, I’ve seen a few more of these cases recently. The space of adjuvant therapy is unknown, we’ve had a lot of missed hits in that space for multiple reasons. Dr Kim, I’m curious, we see this a lot. I have a patient who has a Child-Pugh A score, has a large lesion, around 4 or 5 cm, for other reasons not necessarily related to the tumor is not a surgical or a transplant candidate, and we treat them with Y-90 [yttrium-90 radioembolization], which is the standard in this space. Are there situations where your team discusses adjuvant therapy for those patients? I know there are trials ongoing, but I’m just curious to know how you approach it.

 Yes, absolutely, let’s say this is a nice, clean, Child-Pugh A patient with a 5-cm lesion, well-encapsulated HCC [hepatocellular carcinoma] without elevated AFP [alpha-fetoprotein] or any extrahepatic disease, vascular invasion, and who is not a transplant or resection candidate. We would treat them with radioembolization, and that would be it for us. We wouldn’t add any adjuvant therapy to that because the assumption here is that it’s probably a well-differentiated HCC, not really showing any signs of aggressive tumor biology, and we would just follow those patients at 3-month intervals for the first year, and then every 4 months. But if there was any reason for doubt of that type of well-differentiated HCC, elevated AFP, maybe above 200 or 400 [ng/mL], or if there is any hint of vascular invasion, or let’s say that 5 cm is not well encapsulated but it has more of an infiltrative appearance to it, those types of presentations then give us pause. Even if they have a complete response according to modified RECIST on follow-up imaging to the Y-90, we would either closely monitor them or start them on a systemic regimen for fear that it has an aggressive tumor biology that could manifest itself later on, either extrahepatic or elsewhere in the liver.

 You’re absolutely right, I’ve seen more of those questionable infiltrative-type pattern or questionable vascular invasion, questions about what do we do because they’re high risk. Dr George, what are your thoughts on this?

 It’s a tough problem. Like you said, maybe because of the selection bias, but at tertiary centers, you tend to see more of these cases where the answer is not very clear. There is one dominant lesion, a couple of other lesions that look small and indeterminate, the appearance and the biology seem somewhat aggressive. Do we stop with just locoregional treatment to the dominant tumor, or do we do something else on the back end, particularly if the patient is not deemed to be a transplant candidate? We struggle with this just like others. In general, what I have favored, if there is clearly not a lesion that can be followed to assess response after locoregional therapy, is to do a short interval follow-up to get a sense of what the biology is, and then incorporate systemic therapy at that point. For the patient for whom transplant is not an option, the options become easier, where if the AFP did not normalize within a certain time after liver-directed therapy, should we just throw in the systemic therapy and watch them longitudinally over time? Then every so often, you run into a severe…toxicity, and then you’re kicking yourself and thinking why did I do this? I don’t know that there is a right answer.

 No, I think you’re right. We all struggle with that. But I’m hoping that’s the area where the new data will help us understand more.

A brief discussion on the potential role for adjuvant therapy in patients with hepatocellular carcinoma.

Defining the Role of Adjuvant Therapy for HCC

 We know that this is a patient with metastatic cancer. How would you approach this patient in terms of picking treatment options, or coming up with a treatment plan for her?

For most of us, the frontline treatment for metastatic disease—unless there are counterindications—is the combination of bevacizumab and atezolizumab. This seems like a Child-Pugh A patient. What was not mentioned in the history, and so we need some additional information, is when was the most recent endoscopy? If there wasn’t a recent endoscopy, maybe we get an endoscopy to make sure there aren’t any viruses. I’d typically go with bevacizumab and atezolizumab in the absence of counterindications for this patient. We could talk more about the special circumstances where you would go away from bevacizumab-atezolizumab and think about other options. But I’ll stop here. I’d like to hear your thoughts, Dr Kalyan.

 I echo your thoughts, Dr George. The data that have come out from IMbrave150 point to using the combination of atezolizumab plus bevacizumab as frontline therapy. About 15% or so of patients, for a multitude of reasons, are not able to get Avastin [bevacizumab]. Whether it’s variceal bleeding or hypertension, there are multiple reasons why they’re not able to get it. Those patients benefit from a TKI [tyrosine kinase inhibitor] approach. In that setting, I tend to use lenvatinib based on the REFLECT trial, mainly because it tends to be better tolerated than sorafenib, based on our experience with it. Most patients are treated with an I/O [immuno-oncology] combination.

It’s interesting to think about what may change, given the recent results of tremelimumab plus durvalumab. That’s an interesting area, and we don’t know how that’s going to change. We must wait and see. Dr Kim, what are your thoughts are on this? Sometimes we have this discussion at our tumor board: if patients have metastatic disease, is there a utility in targeting the largest region, or larger lesion, with some form of a lytic therapy and then continuing systemic therapy? It’s entirely in a palliative regimen, not with the intent of changing the course of the treatment but more the phenomena of generating neoantigens and getting a better response from that standpoint. What your thoughts are on that, Dr Kim?

We’re intrigued. We also have trials running here. Looking at the potential immunomodulatory effect of locoregional therapy, particularly with radiation, to incite T-cell activation and potentiate the effects of CheckPoint inhibitors. We’re investigating that. There are also trials ongoing across the country that are investigating that. When it comes to this patient, I know the patient is Child-Pugh A, but the total bilirubin is borderline at 2.2 mg/dL. I’d be more in favor of this situation because the data have been strong with Child-Pugh A patients who are in the advanced stage in terms of systemic therapy. My only concern, which was cited before, is that these patients are on atezolizumab-bevacizumab and need the endoscopy, and that could sometimes delay treatment and have a risk of variceal rupture.

The patient also had hypertension, which is controlled on ACE inhibitors, but maybe that gives pause. I’m not an oncologist, but lenvatinib caused higher rates of hypertension in that study. You also threw out the recent results of HIMALAYA. Maybe the PD-L1 and CTLA4 combined don’t have any effect on hypertension and all that stuff theoretically, but that could also be another tool for oncologists to use. I’m OK with that because as long as patients derive the benefits from all that, we can always add locoregional therapy later if needed. But in this patient population with metastatic disease and good underlying liver function, I say give them a shot at a systemic therapy. The objective response ranges anywhere from 30% to 40%. But if they’re in that group, they benefit. We’ve seen some dramatic responses in checkpoint inhibitors. Give them a shot. If they’re not in that group, then certainly we can supplement with local neoadjuvant therapy.

 Dr George, until now, assuming we didn’t have a clinical trial, this patient probably would have gone on atezolizumab plus bevacizumab, barring varices or bleeding issues. Do the data from HIMALAYA make you think about considering that? What do you think the future is going to look like with those data coming out?

It’s an exciting time for medical oncologists, isn’t it? I want to go back to the previous issue of local approaches for patients with metastatic disease. One situation that we look at in our tumor board is if the tumor is anatomically large enough to cause mass effect on the bile duct, or vascular structures, then you ask the question while administering systemic therapy. Is there any value in doing some local therapy to be able to obviate that mass effect for faster response? Coming back to your question about HIMALAYA, that’s fascinating because first and foremost, there’s the idea that CTLA4 antagonists, along with checkpoint blockade, clearly seem to be giving a strong signal with excellent outcome. We see that in the second-line setting with the CheckMate data. We see that in the frontline setting. We’ve seen that with the higher dose of triamine. We also see that with the higher dose of ipilimumab in the second-line setting. It seems as if higher-dose CTLA4 inhibition and priming may give a longer tail to the curve, either on the frontline or second-line setting.

The question then becomes, how do we sequence our therapies? What’s the trade-off in terms of adverse effects with the combinatorial therapies, the response being somewhat equal in a sense? If you look at the second-line CheckMate data—ipilimumab 3, nivolumab 1, response of about 32%—we’re looking right at about 30% with our bevacizumab-tisotumab data. We know the long term data are coming out, and maybe we’ll talk about it down the road. Those data are preliminarily looking good. The COSMIC-312 data preliminarily don’t look too bad. We’re still waiting for the final readout. There are all kinds of combinations, whether it’s a dual immune modulation vs a TKI plus immunotherapy or anti-VEGF plus immunotherapy, that merits a long discussion.

Taking recent clinical trial data into account, panelists consider the best treatment option for a patient with metastatic hepatocellular carcinoma.

Case 3: Selecting Therapy for Metastatic HCC

Before we talk about treatment, I want to get an idea from each of our panelists about your multidisciplinary clinic is run. We’re all different in terms of how we’re set up, and it’s a key point to keep in mind when we’re treating this cancer.

I’ll start because I had my office hours today. After office hours, I came here. It’s 1 of my favorite days of the week. We have 31 exam rooms on the 12th floor of our building. We have a large room where all the clinicians gather—it’s almost like a lounge—where we have our work stations and pack stations. Interventional radiology, hepatology, how to bill every surgery, and our transplant coordinator are all in that room. We see patients all in 1 day, and patients appreciate that. We formulate plans together, then we present all those patients because we don’t have 1 day where we see these patients on Thursday mornings in our multidisciplinary tumor board. We formulated a true multidisciplinary fashion. We navigate how many mile points a patient is. If they need transplant, they see transplant evaluation that day. For new patients either of us see, we’ll request that other physicians see them. I’ll see patients for whom I don’t think locoregional therapy is appropriate, and a surgical resection is probably better. Our surgeons and hepatologist will see patients who may require locoregional therapy, and it’s a great time. We all really enjoy that day.

Just to clarify, Dr Kim. You’re saying that you guys don’t have a clinic together. You run through your cases, decide who’s going to be seen, and review the plan through.

 We have 31 exam rooms. Usually I’ll see between 15 and 20 patients on a given day, and they’ll all be seen by another clinician on the same day. It’s a true clinic.

We start the week with neuro conference: Monday morning, 7 to 8 AM. That’s fairly heavily attended by the usual suspects. The interventional radiologists [IRs], the hepatologists, medical oncologists, radiologic oncologists, surgeons, transplant, etc. It’s a nice-looking meeting, rather than a meeting that runs through fascinomas. The downside is we look at 10 to 15 cases. They’re not all HCCs [hepatocellular carcinomas], by the way. There are cholangiocarcinomas, all kinds of liver stuff. We don’t have an integrated clinic where everyone sees the patient. With all the concerned specialties seeing the patient on 1 day. We have about 5 of us in GI [gastrointestinal] oncology. On any given day, there are at least 2 GI oncologists in the clinic. Our interventional radiologists are in the clinic maybe 3 of 5 days. Same with the surgeons. It’s not 5 days a week. We try to get them to see as many of them of the multidisciplinary providers in 1 day if possible. But often that’s not feasible. Ultimately, we have a very nice working tumor board. The discussions are extremely collaborative and collegial. We come out well from those discussions.

 Dr Mantry, how does your tumor board run?

 I set up our multidisciplinary tumor board in 2008 when I moved from New York to Dallas [Texas]. It’s a special pride for me and our system because it has produced such amazing results and does such good validation by scientific studies. This is the 1 way you can improve outcomes within the treatment of HCC. We have 2 IRs. We have 2 cross-sectional imaging radiologists. We have 6 hepatologists—all our cases are presented by hepatology. We have a couple of surgeons. We have 2 medical oncologists and tumor navigators, who will arrange everything around the system. We have a big ACU [ambulatory care unit] of all 7 hospitals in our system. We present 10 to 15 cases every week and prepare dispositions on them. Then our navigator runs with the plan to the patients because we cover a very big geographical area. We have patients come from 200 to 300 miles away. Then we teach them how to do their own locoregional therapies and bring them to our center for second-line therapies or transplant.

 We’re very similar. Interestingly, our tumor board is run by interventional radiology. Dr Kim, I’m sure you know this. We recently started a multidisciplinary HCC clinic. It’s helpful because we run through the cases on a Thursday and then select patients who can be seen together on a Monday. We sort who receives new imaging, they get it done on the same day, and then get IR, hepatology, and medical oncology all in the same day. In an ideal world—I know everybody is stretched—we could have palliative care. It’s a huge component of a lot of these patients. We all struggle with this, particularly in the population that we have. But that’s the unicorn in the room, so to speak. But it’s our vision. We’ll see when we get there.

In the context of metastatic hepatocellular carcinoma, panelists reflect on the value of multidisciplinary care at their institutions.

Edward Kim, MD | Authors

Articles

Future Treatment Approaches for HCC

Evaluating the REFLECT Trial of Lenvatinib in Unresectable HCC

Unresectable HCC: Interpreting Data From the IMbrave150 Trial

Defining the Role of Adjuvant Therapy for HCC

Case 3: Selecting Therapy for Metastatic HCC

Case 3: Multidisciplinary Care in HCC Management