Ghassan K. Abou-Alfa, MD, MBA: Pierre, Ahmed brings up a very important point, the issue of the safety and adverse events. In the ATEZO-BEV [atezolizumab-bevacizumab] data—thankfully we also have the data—it’s great data, and thankfully they did not really have anything of serious concern. But historically based on prior exposure to the information on bevacizumab in a patient with HCC [hepatocellular carcinoma], people had esophageal varices. In that study, they specifically prohibited patients with varices from getting on the study. Tell us a little about what varices are, what does it mean, and why we’re concerned.

Pierre Gholam, MD: Sure. I think this is a critically important issue if we are going to venture into VEGF therapy with patients with cirrhosis and advanced liver disease. Esophageal varices are a by-product of portal hypertension that occurs in the vast majority of patients with advanced liver disease over their lifetime. They are more likely to occur in patients with more advanced liver disease. If you remember Child-Pugh score, Child-Pugh score C patient will have a greater than 50% chance of having varices, but Child-Pugh score A patients will have a roughly 7% to 10% chance of having varices. Varices grow in size over time, and the bigger a varix is, the more likely it is to rupture and bleed. The annual rate of transition from small to large is about 10%, which is why we survey patients over time for variceal or risk of variceal bleeding.

The variceal bleeding is a catastrophic event. It used to carry a mortality rate of about 20%. There were medical advances. There is still a substantial mortality of about 7% every time you have a variceal bleed.

We do have effective therapy for variceal bleeding endoscopically through banding to eradication. This is a process where a gastroenterologist would endoscope the patient, apply rubber bands to restrict the flow of blood in the varix, and usually after 3 to 5 procedures, on average, obliterate the flow of blood into the varix. This needs to be repeated over time, but it’s pretty effective. We also use non-cardio-selective beta-blockers, which are medications that lower portal pressures, lower the hepatic venous pressure gradient, and can also reduce the lifetime risk of bleeding.

All of these are things that medical oncologists, hepatologists, and all stakeholders should become pretty familiar with when engaging in treatment of patients with this regimen.

Ghassan K. Abou-Alfa, MD, MBA: That’s a very important introduction to the concept. Catherine, I’ll follow up with you. In the study, we’re talking about the IMbrave150, where ATEZO/BEV [atezolizumab-bevacizumab] versus sorafenib had great, positive outcomes. I agree with Anthony that very likely it will be a standard of care. They did not allow or permit patients with varices to be on the study. However, we had to say it’s going to be that we have to screen patients on the study. But my simple perspective is that if you put the scope in a patient, it’s not going to be like a yes-no: yes varix, no varix. There will be a big spectrum. What do you think of this common gray zone? How can you handle that?

Catherine T. Frenette, MD: The study required an endoscopy within 6 months. I think that’s a reasonable time period if we’re going to start thinking about starting ATEZO-BEV [atezolizumab-bevacizumab], except with the patients who have vascular invasion, because they can develop varices very quickly. Those patients I would re-endoscope before starting.

Varices go anywhere from very small varices to very large varices, and they also can have what we call high-risk stigmata, which means they have red wale signs or white nipple signs, all of which are an increased risk of bleeding. It’s going to be very difficult in some patients to say, “This patient is a high-risk of bleeding or a low-risk of bleeding. What can we do to decrease that chance?”

The problem with banding is that it does work very well, but it takes time. As Dr Gholam said, it takes 3 to 5 procedures, and those are usually 2 to 4 weeks apart. It could be months before the varices are controlled with banding to be able to get them on ATEZO-BEV [atezolizumab-bevacizumab] therapy.

Ghassan K. Abou-Alfa, MD, MBA: That’s very important. And again, for all of us who are keen on catching up with the data, because evidently even patients now are asking about that new information that came out of Singapore, ie, ESMO [European Society for Medical Oncology Congress] Asia, in regard to ATEZO-BEV [atezolizumab-bevacizumab].

We do anticipate and of course we recommend that patients are screened for varices. In a discussion with the hepatologist, as our 2 fine hepatologists, Dr Gholam and Dr Frenette, are talking, we have to make sure we truly understand what things are exactly and what does it mean.

Interestingly, I was asked a challenging question yesterday. If you have a varix that was treated, can you put the patient now on ATEZO-BEV [atezolizumab-bevacizumab]? Admittedly we don’t know the data because these patients were not in the study to begin with. Again, I’d just wait for all the information as we go, but that all means if this is to be of interest, we’ll need to make sure that the patient is actually having an endoscopy and evaluated for that purpose.

Anthony B. El-Khoueiry, MD: Sure. I’d like to just make 1 comment about the ATEZO-BEV [atezolizumab-bevacizumab] combination. I believe the study did not, as you said, allow untreated varices. However, for patients who had an endoscopy within 6 months, if the varices were treated, the local institutional standard was subsequently allowed.

Ghassan K. Abou-Alfa, MD, MBA: That’s good. I stand corrected on that.

Anthony B. El-Khoueiry, MD: What we don’t know is what were these treatments and how effective they were. I think there needs to be a careful look at that, so I agree with the word of caution.

Transcript Edited for Clarity