Multimodality Management of Advanced Liver Cancer - Episode 6

I/O Combination Strategies in Hepatocellular Carcinoma


Ghassan K. Abou-Alfa, MD, MBA: We’ll move on to the next point, and we’ll go back to the checkpoint inhibitors. We all theorize that maybe an anti—PD-1 [anti–programmed cell death-ligand 1] is not enough. And interestingly, the data that were reported or are to be reported are not really agreeing with this. I’ll start again with Tim. Tell us about the pembrolizumab data that were reported in KEYNOTE-240, not that long ago here in Chicago actually. Tell us about pembrolizumab versus placebo in second-line therapy.

Tim F. Greten, MD: Well, basically after the initial attempts and demonstration of efficacy of immune checkpoint inhibitors using both pembrolizumab and nivolumab in a smaller phase I study, there were phase III studies conducted to look at this in a randomized controlled fashion against placebo. And basically with what the study shows, it turned out for statistical reasons to be a negative study. If you look at the individual endpoints of survival and PFS [progression-free survival], it was actually positive, but we have to mention that the way the study was designed, it is a negative study. Now if you talk to the patient and show them the data, it’s very hard to argue against the drug.

Ghassan K. Abou-Alfa, MD, MBA: Fair. Mike, we’re hearing this, and this is public. Already there was a press release that the CheckMate-459 trial with nivolumab versus sorafenib is also negative. It’s going to be reported hopefully in a week. But again, do you see this point Tim’s bringing up with maybe the design, or does it really not work?

Michael A. Morse, MD, MHS, FACP: I think clearly these are design questions. As the field has evolved there have been attempts to create combined endpoints and certainly attempts to decrease the size of some of the studies. And I think that’s really led to the results we’ve seen. I think we all have seen remarkable results with immune therapy, patients who have complete responses and so on. So they clearly work. The response rates are very similar; well, we don’t know about CheckMate-59 yet, but at least in KEYNOTE-240, they are very similar to those response rates that were reported in prior studies like KEYNOTE-224. The drug clearly has consistent activity.

Ghassan K. Abou-Alfa, MD, MBA: I hear your point, clearly it could be a design question. But I think there’s more to that because remember, they aspired for an improvement in outcome that was really going to be considered overwhelming, and it didn’t really cut it. And especially in first or in second line it’s telling us maybe what we spoke about at the beginning of the program today, that maybe anti—PD-1 is not enough by itself, and maybe an anti–CTLA-4 [anti–cytotoxic T-lymphocyte–associated protein 4] combination therapy might actually be the one to break it. And interestingly, it brings in of course the potential for other combinations, which include the antiangiogenics as well. With this said, we’re hearing about several combinations at the moment. We already spoke about 1, which is durvalumab plus tremelimumab, the HIMALAYA study, and we’re waiting for the data on that 1. But besides that, tell us about the other combinations. Tim, you can tell us about the pembrolizumab-lenvatinib trial.

Tim F. Greten, MD: Yes. With this concept of combining different mechanisms, it may just be that the vascular-targeting component of lenvatinib or others—we mentioned AXL or other mechanisms, some of them we probably don’t even understand—seems to be very attractive. Because as you just said, we have to be realistic. A response rate somewhere between 15% and 20% that we see with the checkpoint inhibitors as a single agent is nice. We treat these patients, it’s long lasting, it’s good, but obviously it is way too low because what we see is that 80% actually do not show this response.

So we combine this and that’s very interesting, and we see now in these very early data that we have so far a significant increase in the response rate. The problem we have at this point is, we don’t know necessarily that response rate translates into overall survival. Obviously from small studies we usually look at this. And that is basically what time will tell us, but the approach of combining TKIs [tyrosine kinase inhibitors] or mainly using a vascular-targeting agent together with checkpoint inhibitors is very attractive.

Ghassan K. Abou-Alfa, MD, MBA: Catherine, back to you. There is also another combination that we are hearing quite a bit about, and I’d like to bring up 2 trials over here and try to at least dissect them a little bit further. The atezolizumab plus bevacizumab, which is an antiangiogenic, and also atezolizumab plus cabozantinib. Your thoughts?

Catherine Frenette, MD, FAST, AGAF: Right. The combination trials I think are what we’re all very excited about. And I think we will end up with better response rates, which hopefully will translate into the better overall survival. Combining the atezolizumab with a VEGF inhibitor, or atezolizumab with cabozantinib—which as we said, does have an immune system affect as well—will be very interesting to see if we can improve that survival as we’ve seen. We have to remember of course the toxicity of the combinations, which is something that I always worry about and discuss with my patients. The more drugs you add, the more toxicity you may be at risk for.

It’s also very interesting to see the combinations that were used neoadjuvantly and then adjuvantly with resection, and the high response rates that we’re seeing pathologically. And it would be great if we could figure out how to tell which patients are going to have that response, and perhaps not have to go through a major surgery because they’ve had such a great response.

Transcript Edited for Clarity