Dr. Laura Goff:
Welcome to the ASCO Post Roundtable Series on the First-Line Systemic Treatment of Advanced Hepatobiliary Cancers. I'm Dr. Laura Goff, a medical oncologist and Professor of Medicine at Vanderbilt University Medical Center in Nashville, Tennessee. Joining me today are two of my colleagues, Dr. Spencer and Dr. Yarchoan.
Dr. Spencer, would you introduce yourself?
Dr. Kristen Spencer:
Thank you. Dr. Goff. I'm Dr. Kristen Spencer. I'm a medical oncologist specializing in hepatobiliary cancers and Director of the Phase I unit at NYU here in New York City.
Dr. Laura Goff:
Dr. Yarchoan?
Dr, Mark Yarchoan:
Mark Yarchoan. I'm a hepatobiliary medical oncologist at Johns Hopkins in Baltimore, Maryland.
Dr. Laura Goff:
Today we'll be discussing the treatment and management of advanced hepatobiliary cancer with three patient case studies, and our second installment will focus on the first-line treatment of advanced hepatocellular carcinoma.
S.O. is a 72-year-old man with hepatitis C cirrhosis who presented with abdominal pain 1 year prior. He had two hypervascular lesions in the right liver that were treated with TACE. A follow-up CT scan of the chest showed a new 2-cm right upper lung nodule, which was subsequently biopsied and consistent with metastatic hepatocellular carcinoma. AFP at that time was 122.
What additional clinical or historical data are needed to direct his treatment plan, and what additional studies would you pursue for this patient?
Dr. Yarchoan, we'll start with you.
Dr, Mark Yarchoan:
Yeah, this seems like a relatively, I think, straightforward case. This is something we see every day in our clinical practices. This certainly looks like a case of advanced HCC in the setting of cirrhosis. From the report that you gave, we don't have a formal LI-RADS score here, but just from the description, this sounds like the sort of case that you could make a diagnosis from the scan alone. But obviously I'd want just some confirmation of that from a radiologist.
Again, the LI-RADS score can diagnose HCC in the right host, the right host being a patient with known cirrhosis, which this patient has. In this case, we have a biopsy that confirms the diagnosis. In that case, we're done. We talked a lot about molecular testing in the prior case of cholangiocarcinoma. HCC is a tumor type where molecular testing is generally less important. The chance of finding an actionable molecular alteration in this tumor type is pretty low, but certainly not wrong to do.
But overall, this looks like a case of advanced HCC. I think this is a patient who would be a candidate for a contemporary immunotherapy doublet, which we can talk about in more detail.
Dr. Laura Goff:
Sure.
Tell me, Dr. Spencer, on investigating the cirrhosis. Tell me a little bit more about what you'd want to know in terms of making a therapy selection about the cirrhosis history.
Dr. Kristen Spencer:
Yeah, I think I would just add when I meet my patients with HCC, I talk to them about needing to know about three different things. I need to know about the tumor, so I think that's some of what we were talking about. Less so molecular profiling, but things like macrovascular invasion, extrahepatic sites are helpful. But I also need to know about the patients, so their comorbidities. Do they have hepatitis? Have they decompensated at anytime soon? What are their other medical problems? Do they have autoimmune conditions, risk for bleeding, cardiovascular conditions?
But then also I need to know about their liver specifically, which I think is maybe what you're getting at. Typically, at least in the US, we commonly use the Child-Pugh score. We can argue the merits of that score at another time. But for that information, we need their albumin, their bilirubin, their INR. We also need to know a little bit about them in terms of ascites or history of encephalopathy. Really, that's just trying to get at the underlying liver function. I look at parameters more related to their synthetic function, which is their albumin and their INR specifically, than I do for obstructive things, which can be related to disease.
But I think those are really the factors that we need to have a great understanding of in order to make a decision about this patient. I will also add being at a site where we use a lot of liver-directed therapy, particularly on trial in combinations in this setting, we need to know if they've had any liver-directed therapy or if that's planned.
Dr. Laura Goff:
Thank you both.
All right, so we'll get into some of his medical history. Medical history includes atrial fibrillation on apixaban, coronary artery disease status post a stent placement, and a history of a myocardial infarction, as well as heart failure. He undergoes upper GI endoscopy that just demonstrated a normal esophagus, portal hypertensive gastropathy, and no varices.
Just I wonder if one of you wants to jump in and tell me a little bit about your thoughts about EGD and banding and how you use that in terms of your therapy selection.
Dr. Kristen Spencer:
I think a lot of this depends on how accessible EGD is at your site. I happen to be at a center where we can pretty readily obtain EGDs, but I understand that that's not the case in many places in this country and certainly not abroad.
I tend to get it on anyone that I can physically get it on unless there's a contraindication for them just because that's the practice. But I think this is evolving in the hepatology world a bit too, right? I think that the actual risk of GI bleed tends to be more related to this hypertensive gastropathy or portal tensive gastropathy that patients have, which tends to be just a friable gastric lining with bleeding. It's not even the esophageal varices that we worry about that we're addressing in EGD, so many radiographically can look and whether there's significant varices that would need to be addressed.
Many will make the argument that there's not a great role for EGD. But I think that's evolving a bit. Maybe that's just out of hesitance on our part, but would love to hear what other sites are doing.
Dr. Laura Goff:
Hopkins practice?
Dr, Mark Yarchoan:
Very much the same.
Dr. Laura Goff:
Okay, good. Same.
I think the main thing that I'm running into more and more probably as our CT scans are getting better is our varices being called on CT scan. Those are patients that I feel like probably I don't have to put them through an EGD or wait for an EGD. If we can see it on a CT scan, it's big. For me, it's big enough to be a contraindication in my world at least, at least early on in the front-line setting. That's-
Dr, Mark Yarchoan:
Maybe those are patients who really need an EGD for banding, right?
Dr. Laura Goff:
They may need an EGD for banding or at least a hepatologist for initiation of beta blockers.
Great, so let's get into the actual therapy. You alluded to this, but we'll talk about what patients do you avoid bevacizumab, for example.
Dr, Mark Yarchoan:
I think we have three contemporary doublet therapies that actually in many cases all three are reasonable and are on the table. One of the challenges is these were all developed in parallel. We have no comparative real data. They were all compared to TKI therapy. They're all improving survival versus TKI therapy. And so we're left trying to parse through the data and figure out who to give which therapy to without really clear data to guide our decisions.
The first of the therapies was atezolizumab/bevacizumab. It remains a very reasonable front-line choice along with durvalumab/tremelimumab and ipilimumab/nivolumab most recently. Atezolizumab-bevacizumab, the bevacizumab can be contraindicated in many patients. And so when I think through my algorithm, one of the key first questions is, is there contraindication to bevacizumab? These are patients with large varices, recent bleeding, but also a high risk of macrovascular complications. We don't have the granular details here, but there's a history of an MI and coronary artery disease. I don't know how recent the MI was, but that would probably give me some pause for using a bevacizumab-based regimen.
This case in particular might potentially be swayed towards using a non-bevacizumab regimen, so either durvalumab/tremelimumab or ipilimumab/nivolumab.
Dr. Laura Goff:
Dr. Spencer, do you still think through atezolizumab/bevacizumab as your first-line standard and reasons for not, or has anything changed over the last couple of years for you?
Dr. Kristen Spencer:
Yeah, it's a decent backbone for me. I think what I'm encountering more that's evolving is what I alluded to earlier is using a lot of liver-directed therapy in combination. Whether that's right or wrong, we'll find out. I think there's some data in combination with TACE. We do a lot more Y-90 at our center, and I know a lot of centers in the country do.
I think it's hard in those situations to start bevacizumab. You're starting, you're holding, you're restarting later. It becomes a little messy, so in those patients, it's another sort of decision point that's easy for me if the patient could otherwise have either regimen. I think I tend to probably use a little bit more atezolizumab/bevacizumab in patients that could get either. My reasoning for that is that although... Now we've gotten to the point where we say this before everything where we shouldn't do cross-trial comparisons and we do them anyway. Now we shouldn't even say that anymore. But I think the IMbrave trial enrolled patients with a little bit more poor prognostic disease. They had a little bit more macrovascular invasion, they had a little more extrahepatic spread, higher AFP, so those patients. And then of course, the Vp4 disease, which we see a lot of, there's data for with atezolizumab/bevacizumab. I tend to lean on that a little bit more, but my use of durvalumab/tremelimumab is growing.
I have not used ipilimumab/nivolumab yet. I am a little worried about the side effects with that 3 mg/kg ipilimumab dosing, so that's my hesitation, though.
Dr. Laura Goff:
Just piggybacking back on that a little bit, Dr. Yarchoan, if you are not using atezolizumab/bevacizumab regimen, do you have a framework of how you decide between IO versus single-agent IO, IO doublet versus single-agent? Does that factor in for you at all?
Dr, Mark Yarchoan:
I have to tell you, I don't use a lot of single-agent IO. I think the CTLA-4 generally adds not that much toxicity.
One of the challenges is that half the patients we see in clinic fall outside of the standard patients that are on the front-line ETC trial, but more and more I think we feel comfortable using IO agents and Child-Pugh B. I think it extends beyond just anti-PD-1.
For me, again, between ipilimumab/nivolumab and durvalumab/tremelimumab, these are both CTLA-4, PD-1 doublets. They're both effective. They both prolong survival. They've never been compared head-to-head, but just looking at the top-line numbers, ipilimumab/nivolumab does appear to have a higher response rate. With that higher response rate does come a bit more toxicity. Again, this is all cross-trial comparisons, so all sorts of caveats to that statement.
This particular patient, I'm struck by a relatively low tumor burden, and so this is a patient who's likely to get multiple lines of therapy. I feel less strongly that this patient needs a response right now and is only going to have one line of therapy. This is a patient where durvalumab/tremelimumab would be very reasonable. I agree with Dr. Spencer.
Dr. Laura Goff:
Thanks to you both.
The key clinical takeaways for this case are that several immunotherapy combinations have demonstrated promising activity in the first-line treatment of hepatocellular carcinoma. Choosing between IO-IO doublets, IO-VEGF doublets, single-agent IO, or even VEGF-TKI should involve a careful review of medical comorbidities.
This brings us to the end of this case. Please see the other segments for further discussion about the latest research in advanced hepatobiliary cancer or visit ascopost.com.
