REFLECT Trial in BCLC Stage B HCC

Amit Singal, MD, MS: The REFLECT trial is the registration trial that led to the approval of lenvatinib in the frontline setting. It was a large, noninferiority, open-label phase 3 trial that compared lenvatinib versus sorafenib in the frontline setting, with over 950 patients randomized to 1 of these 2 agents. In short, what the REFLECT trial showed was the overall survival for lenvatinib was similar to sorafenib, with a hazard ratio that showed a numerical benefit, but that was not statistically significant. When you look at the 95% confidence interval, it crossed 1 (ie, was not statistically significant), but fell within the predefined noninferiority margin. Overall, this trial was large enough to prove that lenvatinib was noninferior to sorafenib in terms of overall survival. This was the primary outcome for the study, and that noninferiority is what led to the approval of lenvatinib in the frontline setting.

Now, the interesting thing about the REFLECT trial is that it had many secondary outcomes, and this is where lenvatinib showed improvements over sorafenib. The secondary outcomes included objective responses as well as progression-free survival [PFS], and you see nearly 3 times higher objective responses with lenvatinib compared to sorafenib, and you see nearly a doubling of the progression-free survival. These are very exciting secondary outcomes that make many providers consider lenvatinib in the frontline setting when debating between sorafenib and lenvatinib.

When we look at the adverse event [AE] profile from lenvatinib versus sorafenib, these are both TKI [tyrosine kinase inhibitor] agents, so overall, the AE percentages are similar between the two, and the classes of AEs that we see are similar, although the percentages of specific AEs do differ. Lenvatinib has lower risk of hand-foot skin reaction, although we see higher rates of other AEs such as anorexia, weight loss, fatigue, hypertension, and proteinuria. When we consider lenvatinib versus sorafenib, at least in my opinion and the way that we practice, I think of these 2 agents providing similar benefit when we take a look at overall survival, the most important end point for a provider as well as a patient. But I think the small differences in the AE profile can help you when deciding between lenvatinib versus sorafenib in the frontline setting.

Masatoshi Kudo, MD, PhD: The response rate in REFLECT in the lenvatinib arm is 40% compared with around 10% in sorafenib by mRECIST [modified RECIST]. So a very high response rate is obtained, and PFS is 7.4 months in the lenvatinib arm and 3.7 months in the sorafenib arm. That means direct tumor efficacy is much better in lenvatinib. It can be introduced in intermediate stage over up-to-7 criteria [sum of the size of the largest tumor and total number of tumors], so high tumor burden. In the intermediate stage by RECIST, lenvatinib achieved a 62% response rate. In the intermediate stage, a 62% objective response rate is very important.

When we see bilobar multifocal disease, if we perform TACE [transarterial chemoembolization], the tumor response is poor and the liver function is deteriorated. But if we introduce lenvatinib first, tumor shrinkage is obtained and some of the small nodules disappear. Then for the remaining tumor we perform super-selective TACE. That way the hypoxia-inducible cytokine, such as VEGF or angiopoietin-2, is not released because of pretreatment with lenvatinib. Then our base cell is normalized by lenvatinib. Response by subsequent TACE is maximized, and we can achieve a complete response even with bilobar multifocal disease or a large tumor. Complete response and drug free can be obtained in maybe half of the patients. It’s a very good strategy because of lenvatinib's high response rate. In that study, we showed a 77% response rate in Child-Pugh A disease, up-to-7 outpatient.

Amit Singal, MD, MS: In addition to the differences that came out of the REFLECT trial, in terms of those secondary outcomes as well as the specific AEs, we have to remember that sorafenib does have the real-world data to use this in expanded patient populations, most notably patients with Child-Pugh B disease. Given lenvatinib has been on the market for a shorter period, we don’t have that same degree of real-world data to use this in those expanded patient populations, once again most notably that Child-Pugh B patient population. The other thing in terms of real-world data for lenvatinib, is that we have to consider that the REFLECT trial excluded certain patient populations from its inclusion criteria. It excluded patients with greater than 50% liver involvement, patients with Vp4 or main portal vein invasion, and patients with bile duct invasion. Even though we didn’t have data for those patients, it didn’t mean that lenvatinib wasn’t effective, we just needed real-world data to come out to show that lenvatinib had similar effectiveness and safety in those patient populations.

Now, while we don’t have the same prospective data for lenvatinib in a real-world setting, there have been several retrospective analyses looking at lenvatinib in these extended patient populations. Several of these real-world data have come out of Asia, whether that’s Japan or other countries, showing safety and effectiveness in these populations. And presented this year at ASCO GI [American Society of Clinical Oncology Gastrointestinal Cancers Symposium], we now have more real-world data coming out of North America, both a United States study as well as a Canadian study, showing once again effectiveness and safety of lenvatinib in these patient populations.

Overall, I think these data show that lenvatinib shows similar effectiveness in terms of responses as well as expected survival in these patient populations. The other thing that we see is that there’s reasonable tolerability as one would expect in patients with Child-Pugh B disease, not only in terms of specific AEs that you see, but also dose discontinuations and dose reductions. When I look at the totality of the real-world data coming out of both Asia as well as North America, I think that we have to watch these patients carefully and dose reduce when needed, but I think we’re having more and more data that at least suggest that lenvatinib can be considered in well-selected patients with Child-Pugh B and likely used safely.

Transcript Edited for Clarity