Novel Approaches to Treatment of Hepatocellular Carcinoma - Episode 4

Approaching Sequencing in HCC: Second-Line Therapy


Richard Finn, MD: Turning our attention to second-line, where for a decade all we had was sorafenib, and in managing patients on sorafenib, we know that radiographic progression doesn’t necessarily mean the patient isn’t benefitting with that drug. We know that continuing patients on sorafenib regardless of imaging historically delays events, delays decompensation and death, from liver cancer. We now have second-line options, and now that we have those second-line options, we need to manage our frontline population differently. And when I say that, we don’t want to wait until someone has decompensation because it limits what we can do next, just like we talked about in the front line. In order for someone to get a benefit from systemic treatment, they need to have somewhat preserved liver function.

And when we’re talking about that today, we have, as of this week, a few new drugs approved, but, historically, regorafenib was the first drug approved, based on the RESORCE study. Then, in sequence, we learned about nivolumab, the work you led with the CheckMate-040 study, and then we’ve had some positive data. As far as approvals, pembrolizumab is now approved on a similar data set as nivolumab, and then just recently, cabozantinib. So now it gets even a little more complicated. What about this idea of sequencing TKIs [tyrosine kinase inhibitors]? The first was the sorafenib to regorafenib sequence. And I think after so many failures, people were very gratified to see that we could do something in second-line. How do you approach second-line?

Anthony El-Khoueiry, MD: First, I think it’s exciting to have multiple options for patients. It’s wonderful news for patients. What’s challenging right now is all these data sets were in patients who were previously treated with sorafenib. They’re not compared against each other. We don’t have studies showing what sequence is optimal. So the way to look at second-line patients, first, is when do we switch from first-line? I think you alluded to that. I think because we now have multiple subsequent options, we are tending to switch at the first sign of real radiologic progression. But if that is a clear RECIST [response evaluation criteria in solid tumors]-type progression or new lesions that are definitive, even if the patient is clinically doing well, I think at this point it’s accepted that we would switch out of sorafenib to a second-line treatment. It’s important to note that the phase III data currently in second-line are limited to regorafenib and cabozantinib. This is where we have phase III data.

Richard Finn, MD: Approvals.

Anthony El-Khoueiry, MD: Approvals and survival benefit. So, with the sorafenib/regorafenib sequence, that sequence, based on the work you reported and published, has a very nice sequential survival of 26 months, which is something we really never thought about or imagined in patients with advanced disease. Now, that’s a selected population. These were patients who tolerated sorafenib at a minimum dose of 400 mg daily for 20 out of 28 days before going to regorafenib. They were Child-Pugh A patients. So, yes, there are limitations but that’s very exciting data for that sequence.

Now, switching from sorafenib from cabozantinib would also be fair, based on the recent approval and the phase III data showing superiority in overall survival of cabozantinib versus placebo. What’s distinguishing sorafenib is that about a third of the population in that trial was actually a third-line population as well.

Richard Finn, MD: Right, right.

Anthony El-Khoueiry, MD: Now, the nivolumab and pembrolizumab data are also patients who were, for pembrolizumab, really mostly second-line. For nivolumab, there were some third-line patients. But these are phase II data. We don’t have survival in a randomized setting, but we have phase II data.

Richard Finn, MD: When we’re talking about the second-line choices, we can talk about the clinical data, and we have the phase III data with regorafenib and cabozantinib, 2 small molecules that are both FDA approved. Ramucirumab is not approved yet, but has positive phase III data in the high AFP [alpha-fetoprotein] group. Before we get to the clinical differences, how about mechanistically. Is there anything about these molecules that would make you choose one or the other after a frontline TKI?

Anthony El-Khoueiry, MD: So, unfortunately, we do not have established predictive biomarkers to say if the patient has this type of biomarker in their tumor, that would make me choose one drug over the other. Now, we can speak a bit about mechanism in the sense that the regorafenib targets largely overlap with the sorafenib targets, right? The VEGF receptor-2 axis, RAS, RAF, etcetera. But it has some targets that are beyond sorafenib such as Tie2, CSF1R, etcetera, which may explain its activity post-sorafenib. Cabozantinib targets the VEGF receptor axis as well, VEGF receptor-2 but also AXL and MET. And MET is thought to play an important role in hepatocellular carcinoma. Now, this was not a selection trial. Cabozantinib took all-comers, and we will see whether patients with high MET expression had a differential outcome at a subsequent date with more data.

Richard Finn, MD: Yes. I think that will be very exciting.

Anthony El-Khoueiry, MD: Similarly, with the PD-1 [programmed cell death protein 1] inhibitors, both nivolumab and pembrolizumab, it is not established currently whether PD-1 expression or PD-L1 [programmed death-ligand 1] expression in the tumor are really predictive biomarkers for selection. As far with the nivolumab data, it seems that the responses occurred in both the PD-L1—negative and PD-L1–positive patients. But this data set is evolving. I think we have to wait for more biomarker data for the PD-1 inhibitors to see whether we can select patients better. Of course, there are phase III studies also in the second-line, a pembrolizumab study versus best supportive care, and the first-line nivolumab versus sorafenib data. And I think that will enrich our understanding of patient selection, hopefully.

Richard Finn, MD: Certainly, from the scientific standpoint and obviously, those are potentially practice-changing studies. And you alluded to this a little earlier. We have the sequence data of sorafenib to regorafenib. We hope to see sequence data of sorafenib to cabozantinib. We don’t have high level evidence of lenvatinib to a second-line agent, right?

Anthony El-Khoueiry, MD: Correct.

Richard Finn, MD: Just stepping back, does that affect your choice for frontline therapy? And how does that influence your choice for second-line?

Anthony El-Khoueiry, MD: So, as we both mentioned, most of the studies have been done in the post-sorafenib setting. Some of the labels mention post-sorafenib as well. Now, lenvatinib is another TKI. Again, it’s a multitargeted kinase inhibitor. Many of the targets are similar to sorafenib, except that lenvatinib also targets the FGF [fibroblast growth factor] axis in addition, which is important and unique. One can hypothesize that we should not see a differential if we use lenvatinib versus sorafenib first-line as well as sequencing multiple lines later. But we don’t have those data, so I think those data are going to come from hopefully practice and registries to see what happens in the post-lenvatinib setting.

Richard Finn, MD: That’s progress, right?

Anthony El-Khoueiry, MD: Right.

Richard Finn, MD: Fortunately, we’ve had a lot of progress in liver cancer and it’s all happened very fast. And I think in other diseases we don’t necessarily restrict things based on how they were done in the study, per se. For example, if CheckMate-459 is positive where nivolumab is superior to sorafenib, if it meets that endpoint, then that shuffles the whole deck again. I think the point of our studies is to prove that these are active drugs, that they have anticancer activity. And then it comes down to practice patterns and making decisions.

Transcript Edited for Clarity