Phase 3 REFLECT Trial in Patients with Unresectable Hepatocellular Carcinoma - Episode 3
A key opinion leader in liver cancer reviews the study design of the REFLECT trial of lenvatinib in patients with HCC and shares key insights into its safety and efficacy data.
Richard S. Finn, MD: The REFLECT study, was open-label randomized study of the multikinase inhibitor lenvatinib, versus sorafenib in the frontline treatment of liver cancer. And this study was powered to not only show superiority but also non-inferiority. And that was in terms of overall survival. Now, lenvatinib is a little different than sorafenib. Again, they're both multikinase inhibitors, against the VEGF receptor. But if we look at the kinase profile of these molecules, we see that lenvatinib is a much more potent inhibitor of the VEGF receptors. And interestingly, it also has very potent activity against the Fibroblast growth factor receptor family. The FGFR family is a receptor tyrosine kinase family, on normal and tumor cells. And in liver cancers, there's been data that the FGF family can be important driver of proliferation and survival. Also, there's been data evolving over the years that the FGF family, is a way to circumvent resistance to VEGF-directed therapies. In this randomized open-label study, we saw that lenvatinib did meet its non-inferiority endpoint versus sorafenib. In that case, it was the first positive phase three study, after the approval of sorafenib. It took about 10 years. And we saw that survival with lenvatinib was about 13.6 months, versus sorafenib of 12.3 months. That was a hazard ratio of 0.92. But the confidence interval Prost1, but the upper limit of that confidence interval was 1.06, which was within the non-inferiority margin. And therefore, we could claim that lenvatinib was non-inferior. Currently, that survival of 13.6 months. Nivolumab was the longest survival we had seen in a phase three study, in advanced liver cancer. And that data was supported by improvements, statistical improvements, and secondary endpoints. Specifically, if we look at progression-free survival assessed using the modified resist criteria, we saw that PFS went from 3.7 months to 7.4 months, with lenvatinib. And this was a hazard ratio of 0.66. And we see that, unlike sorafenib, we actually had double-digit responses within that by independent review, using conventional resist criteria. The objective response rate with lenvatinib was just under 19% as compared to about six and a half percent using resist criteria with sorafenib. Similarly, if we look at modified resist criteria by independent review, it was 41% with lenvatinib versus 12.4% with sorafenib. So, while the overall survival end point was non-inferior, we did see that other end points favored lenvatinib. And if we look at the side effect profile, there is a difference there as well. Both drugs cause hypertension, but this is more frequent and more potent or higher grade with lenvatinib. Also, hypothyroidism and proteinuria, are two events that are seen more frequently and higher grade with lenvatinib, versus sorafenib. However, if we look at hand-foot skin syndrome or Palmer-planter, or erythrodysesthesia, this happens at higher grade and higher frequency with sorafenib. So, a side effect that can be problematic with sorafenib, occurs less frequently with lenvantanib. If we look at GI toxicity and anorexia and weight loss, those two things tended to be similar. Now, we tend to think now that objective response is important in liver cancer. And we say that because now we have drugs that are significantly affecting response. And specifically, the checkpoint inhibitors. But in an interesting analysis that came out of the REFLECT study, when we looked at survival for patients who had an objective response in that study, we saw that survival in the responders regardless of drug was 22 months, just over 22 months. Whereas in the non-responders, it was 11.4 months. Now, we don't have a way to predict who will respond upfront, but it is reassuring to know that if patients do have an objective response, their prognosis will be better than if they do not.
This transcript has been edited for clarity.