Multidisciplinary Treatment Strategies for Hepatocellular Carcinoma - Episode 11
Transcript: Ghassan K. Abou-Alfa, MD: We come with another drug that’s based on alpha-fetoprotein [AFP], ramucirumab, and specifically for AFP of more than 400. The data showed that patients with an AFP of more than 400 will have a better outcome on ramucirumab versus placebo. This was in second-line. The median survival [MS] was in the range of 8 months or so, and not necessarily as robust as for example, cabozantinib second-line. Tell us about that data. What’s your reflection on it? And number 2, teach us about AFP. What is it?
Amit Singal, MD: I think that ramucirumab is an interesting drug. We’ve been talking about the need for biomarker-driven treatment decisions. You can say that ramucirumab has at least moved the field to start considering biomarker-driven therapy choices. As you pointed out, we started with the REACH trial where we took ramucirumab and treated all-comers in terms of the second-line therapy, and that trial was negative. Then in a post hoc analysis, they took a look at the subgroup of patients who had an elevated AFP of greater than 400, and they saw a benefit. And in their defense, they then didn’t just try to sell the post hoc analyses, they actually did a full randomized controlled trial in this subgroup of patients, and as you mentioned, did see a significant survival benefit. Albeit, as much as you could compare across trials, it looks like a smaller potential absolute benefit.
I think that the nice thing about ramucirumab, as I said, is it moves the field to push us to find biomarkers. However, in my humble opinion, I don’t think this is what I wanted in terms of a biomarker-driven approach. When I think of a biomarker-driven approach, I want something that says this drug is going to work better than other drugs in this subgroup of patients. And when you take a look at all of the other trials, when you take a look at RESORCE, you take a look at CELESTIAL, regorafenib and cabozantinib also have a benefit in those high AFP patients. So I think the one thing that you take away from this is you can’t say or you shouldn’t say if I have a high AFP patient, I should only use ramucirumab. You still have all of these treatment choices in that subgroup of patients.
Ghassan K. Abou-Alfa, MD: Fair enough. Katie, along that line, what do you think, is AFP really a tumor marker or prognostic, diagnostic? What do you think it is?
Katie Kelley, MD: I think it’s a prognostic marker. We have learned from early stage disease, we alluded to it earlier, that patients with high AFP have a much higher rate of recurrence after transplant. In fact, it’s used as criteria for patients being ineligible for transplant in some centers. And we know across studies that patients with high AFP have unfavorable biology. When we look at genomic profile studies, we do see that certain transcriptome patterns are associated with high AFP; around 70% of patients who innately have an AFP-producing tumor probably have different genetic underpinnings. Though, to be fair, we don’t fully understand exactly the type of subclass that is responsible for AFP production and how it fits in the grand scheme of HCC [hepatocellular carcinoma] across subtypes.
But beyond prognostic value, I think this is the first chance for us to use AFP as a predictive biomarker for a ramucirumab response. Whether this study was positive in high AFP because of a greater effect size, because of a poor prognosis and more events versus a biologic susceptibility to the particular mechanism, it’s hard to know. It may be that the antiangiogenic therapy of VEGFR2 monoclonal antibody inhibition by ramucirumab was particularly effective in the AFP genetic underpinnings. But I don’t think we have the data to tell that yet.
Ghassan K. Abou-Alfa, MD: Fair enough. As we can see here, we’re not done yet, but we’ve seen already 5 treatments. We moved from 1 to 5 treatments in really barely 2 years. Sorafenib, as we started, now we have lenvatinib, with very robust outcome, including impressive response rates and a very acceptable adverse events profile. Then we moved to cabozantinib, again quite impressive data in the second-, third-line setting and pretty tolerable. We also spoke about regorafenib but with the conditional thing that it has to be prior to sorafenib, ie, patients starting sorafenib probably would do well on the regorafenib. We spoke about or we heard about the combined 26.1-month improvement in survival in a retrospective fashion.
And of course, a fifth item that we added is ramucirumab. It’s quite the impressive drug. I would say, however, with certain particularities to it, that admittedly maybe they are valid but we are still yet to understand them better. That really will depend only on being given to the high AFP patients, patients with an AFP of more than 400. And again, with an improvement in survival compared to placebo but not to the same magnitude as we saw with other drugs, as Amit said
Transcript Edited for Clarity