Novel Combinations to Treat HCC

Richard S. Finn, MD: We now know that durvalumab/tremelimumab has completed its phase 3 HIMALAYA study. And that’s durvalumab/tremelimumab versus sorafenib in the frontline setting. Those data are eagerly awaited. Similarly, nivolumab and ipilimumab are in an ongoing phase 3 study versus sorafenib. Given atezolizumab/bevacizumab’s approval, we hope that study will continue to move along, given that the standard of care is evolving, at least in the United States.There’s also interest, though, in not only combining IO and IO or IO and IV drugs, but also IO and TKIs [tyrosine kinase inhibitors]. COSMIC-312 is looking at atezolizumab with cabozantinib, which is a multi-kinase inhibitor to VEGF, CMET, and AXL, which we know is approved in the second-line setting, and now trying to move that to the frontline. That study is ongoing. I don’t know that we’ve seen much early data with that combination as far as efficacy, but at ASCO 2020, a story that’s been evolving over time has been data with pembrolizumab and lenvatinib. Anthony, why don’t you fill us in with the latest take on that update?

Anthony El-Khoueiry, MD: At ASCO, we saw a slightly bigger data set from a phase 1b with the combination of pembrolizumab and lenvatinib. About 100 patients, and with longer follow-up where we saw that the RECIST 1.1 response rate with the combination was at 36%. So certainly, again, this repeating theme of 30% range of response rates with these combinations. With modified RECIST, there was a 46% response rate. Again, I’m quoting both, because it’s important to be careful when comparing these response rates and make sure that we’re comparing apples to apples. More importantly, if we think about beyond response, the median PFS with this combination was in the n 9-month range. The median OS, again, very promising, hitting the 18- to 20-month range. So again, the toxicity is certainly higher than using the single agents alone, but within a manageable range if one knows what to expect with these combos. The discontinuation rate was about10% to 14%, depending if they looked at both agents or single agents of what was stopped. So again, a manageable profile. This combination is currently being evaluated in a phase 3 setting with LEAP-002 comparing pembrolizumab/lenvatinib with lenvatinib alone

Richard S. Finn, MD: [Interposing] Yes.

Anthony El-Khoueiry, MD:Sorry, Richard. We may end up with multiple combinations in the first-line as different options. And I just wanted to say, the challenge will be, are these combinations going to be used just based on physician preference and a slightly different toxicity profile? Or will there be other markers that help us select among the different combos in the first-line?

Richard S. Finn, MD: it’s incredible, right? If we think back of where we were at the beginning of our discussion about just having sorafenib and then the advance with the approval of Lenvima to where we are now in just 2 or 3 years, it’s a rapidly changing landscape. One of the things I wonder about is, is it the same group of patients; we’re getting 30% plus or minus with all these regimens, I think you could say. Is it the same group of patients who are benefiting from CTLA-4, PD-1, as from VEGF PD-L1 from TKI PD-1. Katie, what are your thoughts on that?

R. Kate Kelley, MD: I think that’s becoming a really interesting question and draws us back to our eternal quest for a good biomarker. I think if we return to the durvalumab/tremelimumab circulating T-cell profile data that we talked about earlier with that phase 2 regimen, it’s a really intriguing idea to use that as a tool—as a marker, at least, for immune activation, if it bears out as a true biomarker response in the HIMALAYA study. So of course, we need to validate with an independent set. It may be it only holds true for a CTLA-4 population. But again, that kind of an assay to tell us, are we on the right track to really prompting an immune response with a given combination is what we need to discern the efficacy in different subpopulations. I think, just to echo again what Anthony was alluding to with the combinations of the TKI plus immunotherapy, I think it is really interesting, as we consider also that not only are the TKIs adding an anti-angiogenic effect that removes the immune suppression of the VEGF pathway in the tumor microenvironment, they also have their own independent targets for both Lenvima as well as cabozantinib, for example. If we take cabozantinib, it inhibits suppressive elements like tumor-associated macrophages and myeloid-derived suppressive cells that through the TAM family kinases, that if we were to identify exactly the mechanism, we might be able to choose which combination works better for certain patients now that we have so many more tools with positive results.

Richard S. Finn, MD:Certainly room for optimization, for sure, as we see this evolving data.

Transcript Edited for Clarity