Approaching Second-Line Therapy in HCC

Amit Singal, MD, MS: With the advances after the IMbrave150 study, we have some degree of certainty in the frontline setting. However, this has created potential confusion and disarray in the second-line setting, and you see multiple approaches that can be used. Some people use a T-minus-1 approach that essentially uses atezolizumab-bevacizumab, and then after atezolizumab-bevacizumab, they would go to sorafenib or lenvatinib as their next line of therapy. Then they would use what’s currently or what was previously approved in the second-line setting and then move all those agents to the third-line setting. They just shift down the current sequence to be after atezolizumab-bevacizumab.

The other approach is to take all these agents that are approved, whether they’re in the first- or second-line setting, and put them in a big box or a grab bag and then consider them after atezolizumab-bevacizumab. Therefore, what you could do in that setting is use atezolizumab-bevacizumab and then consider any of these agents after failure of atezolizumab-bevacizumab. Both those approaches are reasonable because we don’t have true data to shape or inform which strategy is better.

Personally, take a look at each of those agents, and there are only a handful I would consider in the second-line setting after atezolizumab-bevacizumab. The reason I say this is because with atezolizumab-bevacizumab, we have acted on the PD-1/PD-L1 axis, and we’ve acted on the VEGF axis. When I take a look at the available agents that are approved in the first- or second-line setting, I don’t see a reason why I would go from atezolizumab-bevacizumab to consider single agent PD-1 inhibitor, nor would I see a rationale to consider ramucirumab, once again acting upon that VEGF pathway. I want to use more of a broad-acting agent that would be able to act on pathways that I have not already hit using atezolizumab-bevacizumab in the frontline setting.

The other agent that I would have reluctance using up front is regorafenib. Regorafenib was evaluated in the RESORCE trial, which used a very careful patient selection of requiring sorafenib tolerability for 21 of 28 days. I continue to use sorafenib tolerability to consider regorafenib in clinical practice.

When I cross off or put those agents on the back burner, what I’m left with after atezolizumab-bevacizumab is sorafenib, lenvatinib, and cabozantinib, in terms of agents I would consider using in the second-line setting in a patient who failed atezolizumab-bevacizumab or progressed on atezolizumab-bevacizumab.

Even though atezolizumab-bevacizumab is a huge advance to the field, we have to recognize that not every patient is eligible to receive this therapy. There may be patients who have autoimmune conditions that may not be eligible to receive atezolizumab. There may be patients who are at high risk of GI [gastrointestinal] bleeding—whether that’s large varices, notable portal hypertensive gastropathy, concomitant use of NSAIDs [nonsteroidal anti-inflammatory drugs], or Coumadin, which would increase your risk of bleeding—for whom you may have reluctance to use bevacizumab. In those patients who are ineligible for atezolizumab-bevacizumab up front, I continue to use sorafenib and lenvatinib as my alternative frontline setting agents.

The other group we should remember is that large group of patients who have Child-Pugh B disease or any degree of liver dysfunction. IMbrave150 was very careful in its patient selection, including only patients with Child-Pugh A disease. When we consider those patients with Child-Pugh B disease, not only are they at higher risk of complications or intolerance in general, but I would have specific reluctance in terms of bevacizumab. We know that patients with Child-Pugh B disease are at higher risk of having varices and higher risk of having variceal bleeding. There may be notable intolerance of bevacizumab in these patients with Child-Pugh B disease. For those patients, I continue to use agents where we have real-world data or trial data showing tolerability of agents in a Child-Pugh B setting. We have good, real-world data showing tolerability of sorafenib in Child-Pugh B patients, we have some data from the CheckMate040 study showing that nivolumab can be tolerable and acceptable in well-selected Child-Pugh B patients, and now we have increasing real-world data for lenvatinib in Child-Pugh B patients. Those are the agents I would consider in a Child-Pugh B patient, a population that would be ineligible for atezolizumab-bevacizumab.

When we consider lenvatinib in the frontline setting, whether that’s in patients with larger, bulky, multifocal intermediate-stage tumors or in the advanced-stage setting, we don’t have trial data that show agents that have been effective after lenvatinib. But most of us apply the second-line agents that have been approved after sorafenib and use those, likely after sorafenib in those patients who have failed lenvatinib. In those patients, I continue to use approved second-line agents, whether that’s cabozantinib, regorafenib, or ramucirumab, in patients with an elevated AFP [alpha-fetoprotein], and then also the PD-1 inhibitors nivolumab and pembrolizumab. I would consider those in the second-line setting for patients who failed lenvatinib.

When most of us approach patients with an advanced-stage HCC [hepatocellular carcinoma] at this time, given IMbrave150 data, we would use atezolizumab-bevacizumab up front if somebody was eligible, and you’re then left with the TKIs [tyrosine kinase inhibitors] in the second-line setting. Of course, there may be some “historic patients” who were on TKI therapy prior to publication of the IMbrave150 study, and so were on sorafenib or lenvatinib as first-line agents already and now may have failure, and you’re then considering second-line therapy agents in those patients.

It’s interesting in light of us not having any data, it’s hard to say any 1 strategy is wrong or not. But I do think it’s reasonable, although arguably not evidence based, to ask yourself, given the benefit of atezolizumab-bevacizumab, if this patient is eligible for atezolizumab-bevacizumab. And if so, use that in a patient who was on lenvatinib or sorafenib and progressed, and you have to then invent a second-line therapy. If that patient is eligible for atezolizumab-bevacizumab, it’s reasonable to consider it at that point.

Transcript Edited for Clarity