FDA Approvals Represent New Wave of Progress in Advanced HCC

Benjamin Leon Musher, MD

Benjamin Leon Musher, MD

Until recently, sorafenib (Nexavar) was the only FDA-approved systemic therapy for the treatment of patients with unresectable hepatocellular carcinoma (HCC). Now, lenvatinib (Lenvima), regorafenib (Stivarga), cabozantinib (Cabometyx), nivolumab (Opdivo), pembrolizumab (Keytruda), and ramucirumab (Cyramza) have been added to the paradigm for patients across a multitude of settings, explained Benjamin Leon Musher, MD.

“A few years ago, we had 1 option to treat [patients with] unresectable HCC. Now, we have several,” said Musher. “I encourage clinicians to review the data. Until we have better data to guide us in terms of what to use for first-, second-, and third-line therapy, it’s going to be up to clinicians to determine what their patients can tolerate and what helps the best. Stay tuned because there will be data in the next few years.”

In an interview during the 2019 OncLive^® State of the Science Summit™ on Gastrointestinal Malignancies, Musher, associate professor of medicine—hematology and oncology at Baylor College of Medicine, reflected on the progress that has been made in HCC and highlighted promising combinations that are emerging in the space.

OncLive: How has the field of newly diagnosed HCC changed in the last couple of years?

Musher: As of a few years ago, we only had 1 option for treating unresectable HCC, and that was sorafenib. In the last couple of years, several new drugs have been approved by the FDA, including targeted agents and a couple of immunotherapy agents. We still have a long way to go with this cancer, but we have some more options now than we did a few years ago. We also have to do a little bit more work to figure out what sequence to use them in.

Could you discuss the CELESTIAL trial and the clinical implications of its findings?

The CELESTIAL trial randomized patients who had already received 1 or 2 lines of therapy to either cabozantinib or placebo. Cabozantinib is an oral agent, and it showed a survival benefit of a couple of months. Like other studies that have shown benefit in HCC, we’re not talking about a huge benefit; however, there was a statistically significant benefit that was better than what we have seen with chemotherapy in the past. This is an important study, as it gives patients another option once they have received sorafenib and possibly another line of therapy.

What were some of the data to come from the RESORCE trial?

In the RESORCE trial, patients received either regorafenib or placebo after having progressed on sorafenib. Back then, sorafenib was the only drug that had been approved [by the FDA], so there was nothing else that it could compete against. The trial showed a 2- to 3-month overall survival (OS) benefit. Not surprisingly, toxicities were associated with regorafenib just like we’ve seen in colon cancer, including fatigue and hand—foot syndrome. These are events that we have to take very seriously in patients with liver cancer. Patients were all given the full dose of regorafenib which we, as clinicians, know is not necessarily the easiest to tolerate. Just like we do in colorectal cancer, we would start at a lower dose and titrate up. It is an FDA-approved and evidenced-based option for second-line therapy for patients with unresectable HCC.

Would you like to highlight any other clinically impactful trials?

The CheckMate-040 trial tested nivolumab and the KEYNOTE- 224 trial looked at pembrolizumab. They both had fairly similar results, with a 20% response rate. Responses do appear to be durable, and the drug is very well tolerated as it is with other malignancies. Interestingly, PD-L1 expression did not appear to predict any response or lack of response, so it’s important that all patients who have received 1 line of therapy be considered for checkpoint inhibition.

A big caution area would be patients who have undergone liver transplantation, as there have been some reports of patients who are rejecting their transplanted liver. Patients who have hepatitis B virus should be treated for the infection for fear of reactivation or exacerbation of their liver disease. I would caution that a patient’s hepatitis B virus should be treated, and there’s a risk of more problems with coinfection of the hepatitis B and C viruses.

Either pembrolizumab or nivolumab is an acceptable option.* Those are nonrandomized data, as both studies were single-arm trials. A randomized trial has completed in the first-line setting, but we’re still waiting on results. We’re waiting on some randomized data, but both drugs are approved [by the FDA] for patients who have refractory HCC.

What were the key takeaways from the REACH-2 trial with ramucirumab?

Ramucirumab is a VEGF inhibitor; it’s used most extensively in gastric cancer. The agent has also been tested in liver cancer because these are vascular tumors and it targets the VEGF receptor. In the REACH trial, patients who had received prior therapy were randomized to receive either ramucirumab or placebo. Although there was no OS benefit in the whole population, a subgroup analysis showed that patients who had alpha-fetoprotein (AFP) greater than 400 appeared to benefit.

Therefore, the REACH-2 trial was designed for patients who had AFP greater than 400; that was the inclusion criteria. That study showed about a 2-month OS benefit in favor of ramucirumab. Where do we use that drug? These are pretreated patients. If someone has an AFP above 400, it seems like a reasonable option to use. However, we have to be careful in patients who have hypertension or cardiovascular disease.

The phase III REFLECT trial gave the opportunity to validate and potentially clarify the association between objective response and survival in HCC. What were the key takeaways from that study?

REFLECT was a first-line trial pitting sorafenib, already approved, versus lenvatinib in treatment-naïve patients with unresectable HCC. This was a noninferiority trial, and it, in fact, showed noninferiority. Interestingly, the progression-free survival was a few months better with lenvatinib. The response rate was considerably better, especially when you look at the modified RECIST v1.1 criteria. That being said, we cannot draw any conclusions about OS because it was not the endpoint of that study.

It also bears mentioning that patients with a heavy disease burden were not included in the trial. I don’t know if there’s any reason to believe it would work any less well in those patients. I would say lenvatinib is an acceptable option for first-line treatment. It did appear to cause less hand—foot syndrome than sorafenib, but it does have more hypertension and cardiovascular issues. That might help clinicians decide between sorafenib and lenvatinib for first-line therapy.

What is the phase III HIMALAYA trial investigating?

The HIMALAYA trial is studying combination immune therapy—a PD-L1 inhibitor and a CTLA-4 inhibitor—compared with sorafenib. As opposed to studies that looked at VEGF inhibition and checkpoint inhibition, this is combining 2 checkpoint inhibitors and comparing them with sorafenib. That will be an interesting study to see the results of.

Are you excited about any emerging combinations?

One ongoing trial that is very important is a phase III trial combining the VEGF inhibitor bevacizumab (Avastin) with the PD-L1 inhibitor atezolizumab (Tecentriq). That regimen is being compared with sorafenib in untreated patients with advanced HCC. It’s an interesting approach. It’s called the IMbrave150 study, and we will be anxiously awaiting the results. If that approach proves to be beneficial, then we’ll have a few frontline options and we’ll have to make even more decisions.

Many other types of immunotherapies are being looked at in general in the oncology world—for example, chimeric antigen receptor T-cell therapy. In HCC, we’re looking at targeting certain HCC antigens and oncolytic viruses with or without checkpoint inhibition. Many exciting things are going on. The advantage of immunotherapy is that if you get response, the responses are durable.

HCC is not a simple disease; it will certainly require some type of combination therapy to stabilize or shrink the disease. We’re also going to have to start to work on the etiology of HCC as far as predictors of responses to certain agents. All HCC cases are not the same as far as their etiology and their behavior. We need some biomarkers—tumor biomarkers as well as host biomarkers—to guide us in one direction or the other.

Editor’s Note: This interview was conducted prior to the readout of the phase III KEYNOTE-240 trial with pembrolizumab monotherapy in previously treated patients with advanced HCC.