Iglesia Spotlights the Current Complexities of Treatment Sequencing in Advanced HCC

Despite the replacement of TKI monotherapy with dual immune checkpoint inhibition in the frontline setting for patients with advanced hepatocellular carcinoma (HCC), not much has changed regarding second-line preferences, according to Michael Iglesia, MD, PhD. He acknowledged that although that more data are needed, the pivotal trial leading to cabozantinib (Cabometyx)’s approval in the second line is the most representative of a modern-day population.

“[Selecting second-line treatment options] is an unmet need and an unanswered question. After progression on either HIMALAYA [NCT03298451] or IMbrave150 trial [NCT03434379] based treatment, would it be reasonable to use lenvatinib [Lenvima] in the second line? Or should we use agents already approved in the second line? Or is there even a role for switching to the other first-line treatment? For someone who has received only a checkpoint inhibitor, could they benefit from bevacizumab [Avastin] plus atezolizumab [Tecentriq]? Trials designed to answer these questions are going to be important moving forward,” Iglesia explained.

In an interview with OncLive^®, Iglesia underscored the evolving landscape of HCC treatment, emphasizing the need for tailored approaches and ongoing research to address remaining questions and optimize patient outcomes. Iglesia is an instructor in the John T. Milliken Department of Medicine within the Division of Oncology at Washington University School of Medicine in St. Louis, Missouri.

OncLive: Could you elaborate on the key findings from the CELESTIAL trial?

Iglesia: CELESTIAL was one of a number of similar trials that evaluated second-line TKIs in HCC. In terms of findings, cabozantinib met its primary end point of overall survival [OS] versus placebo, which is similar to other trials in this space. That poses a little bit of a dilemma regarding how to use cabozantinib or other TKIs in the second line. CELESTIAL has a number of perhaps advantages in this space. Number one is that many of the patients who received cabozantinib had previously received immune checkpoint inhibition. The trial was late enough that many of these patients had been exposed to immune checkpoint inhibitors, which is now true for most [patients in the second line]. Additionally, there was a healthy amount of patients who had received lenvatinib which was at the time the standard of care. And then of course there was no requirement for alpha-fetoprotein and up to two prior lines of therapy were permitted. Now that the phase 3 HIMALAYA and IMbrave150 trials are out, people are considering using cabozantinib not just in the second but the third line.

What are the safety considerations for cabozantinib?

The adverse effects [AEs] expected with cabozantinib are in line with other TKIs. So not only fatigue, diarrhea, and palmer plantar erythrodysesthesia or hand-foot syndrome, but also VEGF-related effects like hypertension, proteinuria, and increased risk of bleeding and clotting. In CELESTIAL 68% of patients experienced a grade 3 or 4 AE and 62% of patients reduced the dose of cabozantinib. [Patients received] a relatively robust dose of cabozantinib at 60 mg, and I know many people in practice will start at a lower dose, perhaps 40 mg, because they expect patients to have difficulty with AEs.

How do you determine the selection of second-line treatment options?

This is an area where we need more data. There’s not really a lot to guide us in this setting. Many patients [find themselves] in this setting after progression on first-line therapy with either durvalumab [Imfinzi] plus tremelimumab [Imjudo] or bevacizumab plus atezolizumab. I typically do use the second-line regimens in the second line including cabozantinib and other TKIs, almost bypassing sorafenib [Nexavar] and lenvatinib, but that’s not based on solid data. The best data there would be that from CELESTIAL––a lot of patients had received a VEGF-targeting TKI and many patients had also received an immune checkpoint inhibitor. So perhaps that’s similar to the group of patients who had received what we think of as modern first-line therapy.

Transitioning to the frontline setting, could you elaborate on the objectives of the COSMIC-312 trial and highlight some of its key efficacy findings?

COSMIC-312 [NCT03755791] was an open-label, randomized phase 3 trial comparing cabozantinib plus atezolizumab with sorafenib for the first-line treatment of [patients with] HCC. The primary end points [were progression-free survival (PFS) and OS] in the intention-to-treat population. Unfortunately, the trial did not meet the primary end point. There was no improvement in median OS at the first interim analysis, which is a little surprising based on the strikingly positive results of IMbrave150, which seems to be a similar combination of a VEGF-targeting agent plus an immune checkpoint inhibitor. It really speaks to the variability among these VEGF-targeting TKIs. It’s not well understood which targets are really the critical ones or which combination of targets are the critical ones to get that therapeutic benefit.

In the COSMIC-312 trial did the addition of atezolizumab to cabozantinib lead to increased toxicity?

There was added toxicity by adding cabozantinib to atezolizumab. Comparing treatment-related AEs, the rate of grade 3 or worse AEs was around 60% in CELESTIAL. COSMIC-312 had a 57% rate of grade 3 events; 7% of patients had grade 4 AEs and 12% had grade 5 events with the [investigational] treatment group. Additionally, 60% of patients decreased the dose of at least one drug, and about 14% discontinued at least one drug. By looking at the pattern of AEs, you can see the expected toxicities of cabozantinib as well as this sort of smattering of immune-related AEs and the expected pattern.

Considering the future, how might disease etiology influence the selection of frontline treatment for HCC?

This is a question that’s been looked at a couple of times, mainly comparing viral and nonviral [etiologies], so hepatitis B or C compared with other etiologies of HCC. This is something we get mainly from global studies. Outside the US we have a higher prevalence of viral etiologies compared with primarily non-viral etiologies in the US. But it does seem like among patients with a viral cause of HCC there is some additional benefit from immune checkpoint inhibition. So, if anything, you may give preference to immune checkpoint-based treatments in that population. Right now, that decision doesn’t have a huge impact for me clinically in practice, but perhaps moving forward, there will be a way to stratify treatment based on the underlying etiology of the disease.

Are you or your colleagues currently engaged in any ongoing HCC research that you would like to emphasize?

A number of fascinating trials are moving forward, and really a couple of areas in HCC are advancing rapidly. We know the combination of a TKI or some sort of VEGF-targeting treatment plus immune checkpoint inhibition is beneficial. A number of trials are investigating modifying one of those [components], either the VEGF-targeting agent or the immune checkpoint inhibition piece. The phase 3 IMbrave152 trial [NCT05904886] is open here at Washington University, which is investigating essentially taking the IMbrave150 regimen and adding tiragolumab, an anti-TIGIT agent to enhance the immune checkpoint inhibitor.

There are a number of other exciting areas [seeking to] enhance first-line treatment. There are also a number of trials moving forward for patients who are eligible for transarterial chemoembolization, or other liver directed therapies and combining that with systemic therapy to perhaps get a longer PFS or disease-free survival benefit. Mostly the OS results from these trials like EMERALD-1 [NCT03778957] are pending, EMERALD-1 and LEAP-012 [NCT04246177] perhaps being the most interesting right now.