Hepatocellular Carcinoma: Practical Implications of Emerging Data - Episode 7

HCC: Treating Patients With Atezolizumab/Bevacizumab

, , , , ,


Richard S. Finn, MD: Like most studies in advanced liver cancer, front-line, second-line, these are always done in patients who have Child-Pugh A liver disease. We can discuss the applicability of these data sets to the general population. One thing with bevacizumab, which we evaluated many years ago, actually, even in the sorafenib days, because we’ve known antiangiogenics might play a role in liver cancer because it’s a very hypervascular tumor. There’s been interest in targeting VEGF for some time, one of the things that slowed down bevacizumab, I think, a little bit back a decade ago was the concern that anti-VEGF agents can cause bleeding.Now, in IMbrave, as in some of the newer studies with other agents in liver cancer, endoscopies were done within 6 months of coming on study if you had not had one. Amit, as a hepatologist/GI [gastroenterologist], I think probably that group of doctors would be a little more engaged moving forward as this regimen rolls out.

Amit Singal, MD:Yes, I think, Rich, this really, once again, goes back to that same concept of multidisciplinary care, and people working as teams. I think that this will require oncology working closely with gastroenterology to help facilitate that upper endoscopy. Although we have noninvasive means of identifying patients who are at low risk of varices in patients with cirrhosis, we don’t know if those same noninvasive means apply to patients with HCC And so at this time, I really do think, as you’re saying, that it’s important for all patients who are going to be undergoing atezolizumab/bevacizumab to undergo an EGD [esophagogastroduodenoscopy] and to make sure that those varices either don’t exist, or if they exist, that they’re adequately controlled before doing this regimen. To Tony’s point of this regimen being tolerated so well, with a low risk of bleeding, I do think that part of that was because of this requirement of undergoing the EGD and having the varices under control from their local provider.

Richard S. Finn, MD: Yes, I think that was an important point I was going to add. Bevacizumab is not heparin. It’s not an anticoagulant. I don’t think we need to look at patients so fragile. But again, patients are selected, right? Katie, you had a very thoughtful editorial in The New England Journal of Medicine about this study. But you also had some reservations and some thoughts. Could you build on that?

R. Kate Kelley, MD: I think this echoes what Amit also said, that we know from the bevacizumab monotherapy data and from the randomized data in combination with erlotinib that it does have a meaningful risk of upper GI bleed in particular, but also events like perforation and, to some degree, arterial thromboses, particularly in patients with Child-Pugh stage B or worse liver function who have higher risk for portal hypertension and GI bleeding. I think that while I absolutely embrace the regimen as our new first-line standard in patients who are Child-Pugh A who have had an adequate endoscopic evaluation, I don’t think yet we have the data to support extending this to the Child-Pugh B population without careful examination and discussion of risk, because we just don’t have the data, and we do have reason to worry about bleeding risk.

Richard S. Finn, MD: Reena, what about viral hepatitis and the use of drugs like atezolizumab, a PD-L1 antibody? Has that come up for the gastroenterology group, the hepatologists?

Reena Salgia, MD: I think it’s definitely an area where we have to impart the education to the community-treating physician, as well as at the academic centers, to make sure that our partners are aware that for patients with hepatitis B in particular, we have to make sure that they’re at least virally suppressed or at reasonable levels and on antiviral therapy to control their hepatitis B. Hepatitis C, as we saw in the trial data for patients, both with CheckMate, as well as here in IMbrave, we can successfully treat patients who have viral hepatitis with these regimens. The key is that we’re co-managing the multidisciplinary care, as has been said by others, and in particular, for the hepatitis B patients that they’re being treated with antiviral therapy.

Richard S. Finn, MD: Yes, I think this regimen was approved 2 weeks ago or so. And I think we are all very excited to get this out into clinical practice, because as Tony commented, we filled those voids that we always wanted in front line. We have improved survival, improved response rates, durable responses with a very favorable patient-reported outcome data set and safety profile. Amit, did you have any final thoughts on that?

Amit Singal, MD: Yes, I was going to add one thing. I think Katie brought up the idea of the patients with Child-Pugh B, and we talked about the EGD. But the other requirement of the trial was the platelet count. And so the other thing that we do see are these patients who are Child A but have more significant portal hypertension with lower platelet counts. And I think it’s also worth remembering that even if you do an EGD in those patients, we don’t have safety data, and those patients also may be at higher risk of bleeding. I think the nice thing about the TKIs [tyrosine kinase inhibitors] is that we had a lot of real-world data and these extended patient populations. So we have safety data of sorafenib in patients who are Child-Pugh B and even some selected patients who are Child-Pugh C, patients with significant portal hypertension. I think that we are going to need these data in atezolizumab/bevacizumab in these extended patient populations who aren’t included in this trial.

Richard S. Finn, MD: There’s thrombocytopenia, and there’s thrombocytopenia, right? And in the IMbrave150 study, they allowed patients with platelet countsdown to 75,000. So that should capture a lot of our patients who are Child-Pugh A.

Tanios S. Bekaii-Saab, MD, FACP: Yes. I think, Rich, it’s important also not to forget that other than the PD-1 and PD-L1 inhibitors, every agent that’s available to our armamentarium is primarily a VEGF inhibitor. And actually, my worst case of bleeding was in the first days of sorafenib, with a patient who went on sorafenib, had a wound that was closed for more than a year in the belly, and after a month on sorafenib, came to the hospital with the wound dehisced, and the guy was bleeding profusely. The point is, bevacizumab’s issue is, of course, that it’s a long-acting VEGF, and it’s difficult to control per se. But all agents carry a risk, all the oral agents. With Child-Pugh B, we have very low data with any of the TKIs. In fact, even though sorafenib has some real-world data, it’s really not impressive. We haven’t really made a dent in the patients who are Child-Pugh B yet. We make a lot of assumptions. But that group of patients does incredibly bad, regardless of what we do to them.

Richard S. Finn, MD: Yes, and that’s the challenge of having 2 diseases in 1.

Transcript Edited for Clarity