Richard S. Finn, MD: All of the phase 3 studies in advanced liver cancer are focusing mostly on this Barcelona Clinic Liver Cancer [BCLC] stage C population, which as Dr Llovet described, has extrahepatic spread, metastases, vascular invasion, or tumor-related symptoms. But most of the phase 3 studies have about 15% or 20% of patients who are BCLC stage B, which means they’re intermediate technically by staging, but they’ve either progressed on TACE [transarterial chemoembolization], as Dan mentioned, or they weren’t TACE candidates at presentation. It’s interesting to think about how these patients who are BCLC stage B do in the real world in some of these clinical trials with systemic treatment. Amit, you had some data at ASCO GI [the American Society of Clinical Oncology Gastrointestinal Cancers Symposium], and you’re familiar with the space. Can you comment about what is available for patients with BCLC stage B being treated with systemic treatment?

Amit G. Singal, MD, MS: When we think of that intermediate stage, BCLC stage B, the first thing I typically do is look at those clinical trials. As you said, each of these clinical trials has about 15% to 20% of patients who are in that intermediate stage and still qualify for the clinical trial for any number of reasons, including TACE failure and large multifocal bilobar disease, who we would consider poor candidates for locoregional therapy.

When we look at those registration trials, we typically see that the effect of systemic therapy is fairly consistent in that intermediate stage. The most recent trial, the one with atezolizumab-bevacizumab in terms of the first-line standard of care for most patients in the unresectable stage, there was an interesting signal that people may have “overinterpreted” that subgroup analysis.

When we look of the subgroup analysis of the IMbrave150 trial, and you take a look at progression-free survival [PFS], that effect of atezolizumab-bevacizumab compared with sorafenib is consistent in terms of PFS, although when you look at the overall survival signal, the hazard ratio tends to be closer to 1. Some researchers interpreted those data to mean that atezolizumab-bevacizumab may not work as well in the intermediate stage.

We’re going to see as this plays out with longer data that were presented at ASCO GI, but I don’t think there’s any reason why systemic therapy would work any worse in the intermediate stage in terms of all of the considerations that we would have, all the options what we have.

Of course, the number of patients who were BCLC stage B that we see in those registration trials is limited, a relatively small group of patients, so we do want to see more real-world data. The real-world data that have been presented for systemic therapies, for example, in the frontline setting, sorafenib, lenvatinib, and I’m sure we’ll see atezolizumab-bevacizumab real-world data come out as well, all suggest these agents work equally well in the locoregional space as they do in the systemic space in terms of their efficacy.

As we’ve mentioned, the field is trying to define which patients in BCLC stage B are best or most appropriate for systemic therapy. Laura, you referenced the trial led by [Masatoshi] Kudo, [MD, PhD,] that compared lenvatinib with chemoembolization in patients with tumor burden beyond up-to-7 prognostic score. As a proof of concept, Dr Kudo and his colleagues demonstrated that those patients with larger tumor burden beyond the up-to-7 criteria, in a retrospective manner, did better with systemic therapy than locoregional therapy. The trial included relatively small numbers, 90 patients overall, but as both locoregional therapies and systemic therapies continue to evolve, these data are going to be more and more essential for defining which patients are the best candidates for systemic therapy.

Richard S. Finn, MD: You bring up an important point, because the phase 3 studies are not always powered to look at these smaller subgroups, but we often see a trend where the PFS and overall survival is very long for the intermediate group, as Josep commented earlier, and these patients go on to receive other treatments, which may make it more difficult to show an overall survival benefit for this group.

Daniel, at the Mayo Clinic, do you find yourself offering the BCLB stage B group systemic treatment, and what characteristics does that patient have?

Daniel H. Ahn, DO: As you mentioned, BCLC stage B is a wide range of true Bs and those who are closer to As. As Dr Li mentioned, some of these patients may be better candidates for systemic therapy, where they could be downstaged to get surgery. Some of the conflicting factors are whether these patients would be candidates for a liver transplant because the data are not emerging about the role of downstaging to potentially get these patients to a liver transplant. So there are some confounding factors as well.

Within BCLB stage B, there is a very gray area for who qualifies. And as you spoke at ASCO GI this year, we have to compare the role of systemic therapy with locoregional therapy, especially with a lot of these new emerging options that seem to have high response rates and durable responses that can provide an equal benefit and potentially be more efficacious for these patients classified as stage B.

Richard S. Finn, MD: Our armamentarium for systemic treatment has grown. We had sorafenib for a long time in the frontline setting, and then we had lenvatinib approved in the frontline setting based on noninferiority, but it does have a higher response rate and a higher effect on PFS compared with sorafenib. It has some adverse effect differences to consider, more hypertension, proteinuria with lenvatinib vs more hand-foot skin syndrome with sorafenib.

Then we had a slew of second-line therapies approved, which has markedly changed the landscape for patients. Starting these drugs earlier might offer patients more opportunities to see systemic treatment that can impact their survival. Recently the atezolizumab-bevacizumab data in the frontline setting are giving response rates that are comparable to locoregional treatment—not too far off—as far as objective response rates of 30% in the subset that has CRs [complete responses], and that’s in advanced disease. To your point Amit, maybe it is time to look at these drugs in earlier lines of treatment.

Transcript Edited for Clarity