Panelists: Ghassan K. Abou-Alfa, MD, Memorial Sloan Kettering Cancer Center; Richard S. Finn, MD, University of California Los Angeles; R. Kate Kelley, MD, University of California San Francisco; Andrew X. Zhu, MD, PhD, Harvard Medical School; Tony Saab, MD, Mayo Clinic
Published: Wednesday, Mar 13, 2019
Ghassan K. Abou-Alfa, MD: Along that line, and, Andrew, the drug was approved by the FDA literally 5 days ago, on Monday. We are Friday, it was on Monday. So tell me a little bit. Are you using it or whatís your experience?
Andrew X. Zhu, MD, PhD: I have to be very honest. I think with cabozantinib, the experience that I have, which is relatively anecdotal, although I definitely have used that in patients who are going through a clinical trial. I am definitely using it right now with selected patients. I agree. I think itís another TKI [tyrosine kinase inhibitor]. I think you definitely need to be vigilant about all the TKI related [adverse] effects, including hypertension. We need to make sure that the GI [gastrointestinal] [adverse] effects are also managed well. The fatigue is probably also something that we need to be vigilant [about]. And also, nobody has mentioned the dosing. I think we have to really get a better sense, whatís the optimal dosing for cabozantinib in our patient population.
In my opinion, if we follow these patients they actually develop dose-related toxicity. I think by lowering the dose to 40 mg, these patients will tolerate it better and also get longer drug exposure; this may secure them to get the benefit of cabozantinib. So I think thereís a learning curve. Iím pretty sure with the approval of the drug, I think in the next few months to the next few years, weíre going to have an enriched experience in the clinical setting.
Ghassan K. Abou-Alfa, MD: I totally agree. If anything, the drug is relatively well tolerated, definitely fatigue might come into play, and no doubt the dose that was chosen for the trial was the appropriate one that understandably people tolerate rather acceptably well.
And itís fascinating This is amazing that weíre sitting here today. Exactly 1 year ago at GI ASCO [Gastrointestinal Cancers Symposium of the American Society of Clinical Oncology] 2018 we presented the data of the phase III trial, the CELESTIAL study. And then in July 2018 we had the New England Journal of Medicine report on the cabozantinib. And here we are today, exactly 1 year, with an FDA approval for the drug. Thatís really what I would say is a great move and, more importantly, itís a great other choice of therapy for our patients. Thatís what we are definitely thrilled about.
With this said, the story did not really stop, actually it keeps going. And if anything, we have another drug that comes into play, and maybe Katie, you can tell us about ramucirumab and what is it, what do we know about it? Until the REACH-2, which of course weíll let Andy talk about this one.
R. Kate Kelley, MD: Ramucirumab is a VEGFR2 monoclonal antibody humanized thatís approved in other tumor types including colorectal, and gastric, and so we have familiarity with it in GI cancers. And prompting the study called REACH, now we refer to it retrospectively as REACH-1, but the first REACH trial looking at ramucirumab versus placebo in the second-line population was conducted and published a few years back with Andrewís leadership.
But quite interestingly in subgroup analyses there was a trend that was observed showing benefit predominantly in high alpha-fetoprotein [AFP] subgroups, really across a range of elevated alpha-fetoprotein protein tumor markers. And that observation led to, itís not fully understood and Iím eager to hear what Andrew says, but I think one of the suppositions either itís a poorer prognosis group with greater effect size, or potentially a biomarker of susceptibility to antiangiogenic targeting owing to the underlying pathways activated in alpha-fetoprotein signaling in that profile of cancers. But the genesis of REACH-2 is that this subgroup analysis of REACH-1 showed benefit in the high alpha-fetoprotein biomarker selected subgroup.
Ghassan K. Abou-Alfa, MD: Maybe the right thing to do, Andrew, is to ask you, because you led the REACH-2. So tell us about REACH, REACH-2, and also, as we understood from ESMO [European Society for Medical Oncology], there was an evaluation of collection of the data together, the pooling of the 2 studies together. Tell us a little bit about that.
Andrew X. Zhu, MD, PhD: As Katie correctly pointed out, this is the drug specifically targeting VEGFR2. I think angiogenics has been implicated. Itís a very important pathway for HCC [hepatocellular carcinoma]. But having said that, all the TKIs I would mention, theyíre not specifically targeting VEGFR2.
So you always have to say, well, is this really driven by VEGFR? We all believe that. But I think with a drug thatís actually specifically targeting VEGFR2 that shows clinical benefit, weíre really validating that the pathway of angiogenesis is critical for hepatocarcinogenesis. And secondly, in the initial REACH study, we definitely observed a trend in the intend-to-treat population, as Katie pointed out, but on the other hand it did not meet the statistical significance in the whole trial population.
But those with high AFP, a) they actually have very poor prognosis, and they definitely fare very poorly. But also you know when they received the ramucirumab in that study, the magnitude of improvement was very reasonable, with a hazard ratio of 0.69. So with that initial finding, we pursued the REACH-2 study looking at this specific population. And I think, in the competitive landscape that weíre facing for drug development, we took a very practical approach. So instead of running a 700-patient trial, we decided to see whether we could pick up the minimum number of patients to reproduce the benefit.
I think you can argue for that reason, in the end the study was positive but perhaps not as positive as we would like in terms of the absolute magnitude of the difference in survival. However, keep in mind, hazard ratio was 0.71, which is incredibly comparable to all the positive trials we all mentioned today, including REGO [regorafenib], CABO [cabozantinib]. So the hazard ratio was incredibly comparable.
This is another good option for the patients with this disease, and obviously the challenge for all of us sitting here is how to position each drug, and also how to partner the existing drugs with additional agents.
Ghassan K. Abou-Alfa, MD: So, Rich, I would love to hear your thoughts. Is AFP a tumor marker?
Richard S. Finn, MD: Well, yes, it is a tumor marker. That canít be argued, right? I think the question is, is it a prognostic marker? It is a tumor marker because itís increased, as Andy said earlier, in about two-thirds of liver cancer. Normally itís very low. And the question is, what does it mean when itís elevated?
I think weíve seen in several studies that a high AFP tumor doesnít behave as good as, or the natural history of that tumor is not the same as one that has a normal or low AFP. Some of the molecular studies that have been done suggest maybe AFP represents a more progenitor type of liver cancer. Maybe one thatís not as well differentiated, and biologically itís a marker of something else thatís driving the tumor.
I think thatís the prognostic component. The question is, is it a predictive marker? And I think thatís been determined and the answer is, yes. You know you alluded to the tivantinib study that looked at c-MET IHC, immunohistochemistry, that did not pan out, and that might have been the failure of the assay or the failure of the drug. But now we have a prospective study, 2 prospective studies. One taking all comers with a drug that is negative, and then 1 is that prospective selecting on the biomarker of high AFP, and itís positive. So I think we can now say that a high AFP is prognostic, itís a poor marker of a bad outcome, and itís predictive of response to ramucirumab.