Anticipating Results of Checkpoint Inhibitors in HCC

Transcript:

Ghassan K. Abou-Alfa, MD: So Rich, we heard about nivolumab, and we heard about pembrolizumab, and both were already approved by the FDA based on the phase II data, and it seems that, yes, similarly the response and duration of response were kind of like the appealing components for the FDA to understand that this is definitely valuable for the patients. But we do understand that both of those drugs are still pending the phase III trials, and the one for the nivolumab is in first line, nivolumab versus sorafenib. And the other one is the pembrolizumab versus placebo in second line.

Now 1 thing, and at least this is [similar to what] we’ve been talking about here and there. You know, everybody is expecting and anticipating that nivolumab versus sorafenib study—[a] no-brainer, it’s going to be positive. And it was ironically even simple to say, “Oh, what if it’s negative?” But now we’re talking about this because we’ve been waiting for the event; it’s not happening yet. So tell us a little bit about, like, what could have went, if nivolumab, and of course, we hope it’s positive, but if we say nivo versus sorafenib, why would this have happened?

Richard S. Finn, MD: Well, I think if we look at the approval, that explains it, right? It was approved on 15% response rate roughly with these drugs. And if you’re 1 of those responders, you do very well. Right? One of the main questions for, “Will you beat sorafenib” is the benefit in that 15% to 20% of patients enough to separate survival curves? Sorafenib we’ve had for a long time. It improves survival with a response rate close to 0. And that was a paradigm shift for the way we thought about novel therapeutics, the way we thought about liver cancer at the time. But it’s been very consistent in that, and even when we look at the recent data with lenvatinib versus sorafenib, the survival now is about 12 months in a global study. So we’ve moved beyond that 10.5 months, 10.7 months. So 1 thing is, is the benefit a large benefit in a small subgroup? Which we don’t have a biomarker for. Is that better than some effect in a general population treated with sorafenib?

The other thing, this was an open-label study, these drugs are approved certainly in the US and maybe other parts of the world, and if a patient gets sorafenib, it is very easy to see that they could get crossover to either pembrolizumab or nivolumab, off-label.

Andrew X. Zhu, MD, PhD: Were you taking all of them for pembrolizumab?

Richard S. Finn, MD: You know, a PD-1 [programmed cell death protein 1] inhibitor, right? They’re available. And so overall survival in any study is the progression-free survival [PFS] on your frontline. It’s a factor of PFS1 plus PFS2, 3, 4 after subsequent therapies. So it’s not only crossover, but everything we’ve talked about, other drugs, right? The benefit of regorafenib, of maybe cabozantinib, of lenvatinib after progression on your first-line therapy. So that is something else that could affect an overall survival because these drugs weren’t available when it was designed.

And the other thing that I think might be very relevant is the changing natural history of advanced liver cancer. And we’ve talked about this, and I think it’s worth reiterating, is who is the appropriate patient for advanced disease? And these are patients who progress on TACE [transarterial chemoembolization], or present with de novo advanced disease, but it’s that advanced TACE population that is probably changing in that there was so much excitement about PD-1 inhibition, maybe the doctors in the community, or at centers that participate in the study globally said, “Well you know what, this patient, maybe I won’t give them that extra TACE, but I’ll put them on the phase III study to give them access to this great group of drugs, PD-1 inhibitors, that look like they’re really game changers.” And maybe in doing that we might see the natural history of sorafenib, the control arm, being longer than 12 months. Not because sorafenib is more active, but it’s just been started earlier in their disease course, right? Instead of doing an extra 2 cases, they might get started earlier.

And the example I would give at that is if you look at sorafenib in the Asia-Pacific study versus the SHARP [Sorafenib Hepatocellular Carcinoma Assessment Randomized Protocol] study. In the control arms, they behave differently. In North America [and] Europe, the control arm, [the] placebo arm, did 8 months, and in Asia it was about, I think, 3 months, 3 to 4 months. So why is that? The drug still works in both populations. But in the control arm in Asia it was shorter because it reflects a heavier reliance on local regional therapies even when patients are advanced.

Ghassan K. Abou-Alfa, MD: So to go back to the point about the nivo-sorafenib—and again, we’re using this as an example to kind of illustrate that further—I agree with you on the 2 points that you mentioned: the change in the whole outcomes of the disease to begin with. I like very much what you said. That, yes, and I can imagine in the clinic telling a patient, “You got randomized to sorafenib,” what the feel is going to be like. And, of course, because of the ability over the checkpoint inhibitors [it] could be.

But there’s a third interesting point that could have happened. Would it have been [of] any value for the prior exposure to the TKI [tyrosine kinase inhibitor] that kind of preset the patients for a better outcome? TKR [tyrosine kinase receptor] followed by checkpoint inhibitor probably, ironically, better than checkpoint inhibitor followed by TKR. And interestingly, and by the way, we have seen data already, and we know about studies going on. But Katie,…it’s 1 of the examples like cabo-nivo for example. Your thoughts?

R. Kate Kelley, MD: Right. Well, I think there [are] a variety of ways that one can think about these combinations. Of course, there’s the goal that patients get active [drugs], whether they’re going to respond to a TKI or to a PD-1 inhibitor as soon as possible, and if they can be given safely, concurrently, there can be additive benefits theoretically there. But I think more [important] than that is the potential that’s been at least suggested in some translational studies and preclinical work that TKI therapy…may modulate the immune microenvironment around the tumor to be favorable to respond to PD-1 or other checkpoint inhibition.

For example, some of the TKIs might reduce the proportion of regulatory T cells and increase the activated T cell, effector T cell or CD8 population, or likewise reduce the myeloid-derived suppressor cells. And so while these affects absolutely need to be tested prospectively or in human samples treated with these drugs, that’s the hypothesis underpinning the current efficacy we’re seeing in early studies, which absolutely need to be tested in a randomized or prospective way.

Transcript Edited for Clarity