Navigating First-Line Treatment Options in Advanced, Unresectable HCC

Pierre Gholam, MD

The combination of atezolizumab (Tecentriq) plus bevacizumab (Avastin) altered the standard of care for the frontline treatment of patients with unresectable or metastatic hepatocellular carcinoma (HCC), which now includes another combination option with durvalumab (Imfinzi) plus tremelimumab-actl (Imjudo), according to Pierre Gholam, MD, who added that further data on patient and disease characteristics, plus trials investigating multimodality approaches, could help further inform treatment decisions in this setting.

“We are entering the era where treatment for unresectable advanced HCC is becoming more and more complicated,” Gholam stated in an interview with OncLive^®. “I always find it somewhat unfair to relay the message to community oncology colleagues that they need to almost exclusively present and care for these patients in a multidisciplinary setting, simply because I realize that this is not available in most parts of the country.”

In the interview, Gholam discussed primary considerations for the first-line treatment of patients with advanced, unresectable HCC, emphasized the utilization of systemic therapies in patients ineligible for curative treatments, and highlighted the significance of the phase 3 IMbrave150 trial (NCT03434379), which supported the approval of atezolizumab plus bevacizumab in the first-line setting. Gholam is an associate professor in the Department of Medicine, School of Medicine, Digestive Health Research Institute at Case Western Reserve University, in Cleveland, Ohio.

OncLive: What are some of your primary considerations when selecting frontline treatment for patients with advanced, unresectable HCC?

Gholam: Patients with advanced, unresectable HCC typically present at a stage where options are limited to non-curative treatment. These include treatment with locoregional therapy, which in some patients, can afford them a significant overall survival [OS] benefit. However, more often, particularly if the patient has microvascular invasion or extrahepatic spread, [treatment] options are specifically limited to systemic therapy. We have seen some data in the past 3 months that suggested a multimodality therapy, which includes the combination of a locoregional therapy plus systemic therapy, may be of use.

However, generally speaking, patients who do not benefit from locoregional therapy—those who have metastatic disease—will be candidates for a variety of well-established first-line options. [Frontline options] include: the combination of atezolizumab and bevacizumab, which was approved [by the FDA in May 2020], based on data from the Imbrave150 trial;1 the combination of durvalumab and tremelimumab, which was approved [by the FDA in October 2022] based on data from the phase 3 HIMALAYA trial [NCT03298451],2 and we now have 4-year data from that trial; and [single-agent] TKIs, which were the first agents approved in the [frontline setting for HCC, including] sorafenib [Nexavar] and lenvatinib [Lenvima]. [Single-agent TKIs] are useful agents in situations where immunotherapy is contraindicated for a variety of reasons.

How has the IMbrave150 trial set the bar for future combination regimens in the frontline setting of unresectable HCC?

IMbrave150 looked at the combination of atezolizumab and bevacizumab compared with sorafenib. The trial met its OS end point and its progression-free survival [PFS] end point, and it also showed an impact on time to progression.

This study is important in that it established this combination as a standard of care in the first line for [patients] who have unresectable HCC. [The study] certainly outlines the need to be more inclusive [and] include as many patients in clinical trials as possible. One notable positive from IMbrave150 was that it deliberately included patients with main trunk portal vein tumor thrombosis or tumor thrombosis that extended to the contralateral side of the liver [Vp4]. This makes the study [population] more reflective of practice because many of our patients do have Vp4 tumor thrombosis. [This] also creates an additional liability from a treatment standpoint because patients [with Vp4 tumor thrombosis] are more likely to have bleeding complications. IMbrave150 did show that in those patients, there was a greater risk of potential bleeding. However, it was very much a positive to include these patients because you're reflecting clinical practice as best you can.

How does bevacizumab fit into the frontline treatment paradigm for unresectable HCC considering the bleeding risk that you mentioned in this population? What additional research is looking at the use of bevacizumab for these patients?

Bevacizumab is an agent that by itself does not seem to have any meaningful efficacy [for patients with] unresectable HCC; those studies were done more than a decade ago. However, when combined with immunotherapy, there appears to be a significant compounding, synergistic, or additive effect whereby both drugs work better and seem to give a strong signal for efficacy.

Bevacizumab has now been tried in various combinations. The aforementioned IMbrave150 trial would be one example. More recently, there was an example of combining durvalumab, bevacizumab, and transarterial chemoembolization [TACE] in the [phase 3] EMERALD-1 study [(NCT03778957) in patients with locoregional HCC]. Clearly, the addition of bevacizumab, in both instances–perhaps even more notably in [EMERALD-1]–seems to augment therapy.

Of course, bevacizumab has liabilities. Bevacizumab is a monoclonal antibody that inhibits VEGF, and as such, [could] cause a significant increase in VEGF-driven complications. These [adverse effects (AEs)] notably include hypertension, which is something to contend with in these patients; epistaxis, which is a fairly common problem with bevacizumab; and maybe more serious or potentially life-threatening bleeding problems, including esophageal variceal bleeding, [and] bleeding from all causes would potentially be increased.

Although there is established evidence that bevacizumab is a good add-on—especially to immunotherapy—one also has to contend with the potential liabilities that bevacizumab brings.

Regarding the use of durvalumab and tremelimumab, could you elaborate on data for this combination and situations where you may consider that regimen vs atezolizumab plus bevacizumab?

The combination of durvalumab and tremelimumab was studied in the HIMALAYA trial. This [study] included 3 arms: [the first] arm was a durvalumab monotherapy arm; the second was the sorafenib comparator arm; and the third was the combination of durvalumab tremelimumab, with tremelimumab only being given as a single priming dose. The [combination] is now referred to as the STRIDE regimen, and that indeed is the regimen that we currently use to treat patients today.

[HIMALAYA] showed that when compared with sorafenib, the combination of durvalumab and tremelimumab as administered via the STRIDE regimen clearly had [an overall] survival benefit. [The STRIDE regimen did not show a PFS benefit compared to sorafenib.] The study also showed that durvalumab by itself was noninferior to sorafenib, but there was no signal for superiority.

The HIMALAYA trial excluded patients who had main portal vein thrombosis. In [the frontline] setting [for advanced HCC], the durvalumab and tremelimumab data from HIMALAYA stand on their own as a good first-line standard of care. It is potentially useful in patients who have a high risk of bleeding without an obvious mitigation strategy, [such as] someone who has [a] fixed bleeding lesion driven by portal hypertension where the clinician is unable to somehow mitigate that risk and make it lower; or in someone whose had bled repeatedly, despite attempts at mitigation. A patient [who has bled repeatedly] may want to be exposed to an alternative regimen in the first line [rather than] atezolizumab plus bevacizumab because of the bevacizumab component. It's difficult to assess how often patients [with repeated bleeding] come along. However, in that situation, clearly, [the STRIDE] regimen that would be useful for treatment purposes.

How would you say that Child-Pugh categorization affects treatment planning and sequencing?

One of the major data gaps we have in our understanding of first-line therapy for unresectable HCC is the fact that we lack data [for patients with] more advanced liver disease. [Investigators] have tried to bridge that gap to some extent by potentially looking at patients in various first-line clinical trials who have progressed to Child-Pugh B over the course of treatment.

Perhaps the 2 most salient [examples include first-line] data from the phase 3 REFLECT trial [NCT01761266] of lenvatinib vs sorafenib. This was an open-label, non-inferiority trial of both drugs; and second-line [data] from the [phase 3] CELESTIAL study [NCT01908426] of cabozantinib [Cabometyx] vs placebo in [patients who previously received sorafenib]. Generically, the studies showed that the safety of the drugs seems to be reasonable [in patients who progressed to Child-Pugh B]. These subset analyses are underpowered to estimate a good sense of efficacy, but there appeared to be some signal of efficacy, especially in the second-line setting with CELESTIAL.

We lack data [in patients who are Child-Pugh B] for atezolizumab plus bevacizumab in the first line, and we certainly don't have data for the STRIDE regimen. We have strong data for sorafenib; [however,] this does not get as much traction these days because there are other treatments that have supplanted it. The data from various sources, including a large [observational study] called the GIDEON study, which showed that sorafenib was quite safe and well tolerated in [patients who were] Child-Pugh B and, shockingly, in the small subset of patients who were Child-Pugh C.

There clearly is a decline in OS benefit [in patients with] more advanced liver disease because there is a competing risk for death. Overall, at least in the case of TKIs in general and particularly sorafenib, the data seem reassuring [for single-agent use in patients who are Child-Pugh B]. This data gap [for the newer combinations] hopefully won't be with us for too long because there are planned trials [for patients who are] Child-Pugh B that will hopefully give us answers in the coming years.

When does it become appropriate to consider locoregional therapies for surgical resection or liver transplant?

Curative treatments for persons who present with HCC, unfortunately, are not applicable to the majority of patients. If you take 100 patients with a new diagnosis of HCC in North America and Western Europe, at the most, 30 out of 100 will be candidates for curative therapy. Our Asian colleagues have mastered surveillance and screening much better than we have, and [they will] pick up a greater number of early patients who they can cure.

The only 3 established ways to cure HCC are: resection with the intent to cure, called R0 resection; ablation of lesions less than 3 cm in greatest diameter; and liver transplantation in patients who meet the Milan criteria, [which include] 1 lesion no greater than 5 cm in diameter, or no more than 3 lesions with none greater than 3 cm in diameter, without microvascular invasion or extrahepatic spread.

For the remainder of these patients, we fall outside of the realm of curative therapy and have to consider, in the intermediate stage, locoregional therapy, and in more advanced stages, systemic therapy. Perhaps in some situations, [we can consider] the combination of both, so this is where the EMERALD-1 study comes along.

EMERALD-1 was a phase 3, randomized, placebo-controlled study of TACE combined with durvalumab with or without bevacizumab in patients who had unresectable HCC. The patients had to be eligible for embolization. The data were presented at the 2024 Gastrointestinal Cancers Symposium.³

By way of background, the 3 arms were durvalumab plus TACE and bevacizumab; durvalumab plus TACE and placebo [for bevacizumab], and TACE plus placebos for durvalumab and bevacizumab. If you compare [the triplet vs TACE alone], at the 12-month mark, the PFS [rate] was 55.5% for the durvalumab/bevacizumab/TACE arm vs 39.8% for the TACE alone arm. If you look at 18 months, it was 43.1% vs 28.3%. Interestingly, when you remove bevacizumab and just gave durvalumab [plus TACE and placebo], there was not a statistically significant difference [vs TACE plus placebos], which brings us back to the point that there's something about giving immunotherapy with bevacizumab that seems to seem to make this [regimen] work better.

These are early data [from EMERALD-1]; we await other end points, maybe even OS. [Regardless, these findings] certainly revived the interest in potentially having multimodality therapy as a starting strategy for treating patients in this intermediate-care setting.

References

1. FDA approves atezolizumab plus bevacizumab for unresectable hepatocellular carcinoma. FDA. May 29, 2020. Accessed March 15, 2024. https://www.fda.gov/drugs/resources-information-approved-drugs/fda-approves-atezolizumab-plus-bevacizumab-unresectable-hepatocellular-carcinoma
2. FDA approves tremelimumab in combination with durvalumab for unresectable hepatocellular carcinoma. FDA. October 21, 2022. Accessed March 15, 2024. https://www.fda.gov/drugs/resources-information-approved-drugs/fda-approves-tremelimumab-combination-durvalumab-unresectable-hepatocellular-carcinoma
3. Lencioni R, Kudo M, Erinjeri J, et al. EMERALD-1: A phase 3, randomized, placebo-controlled study of transarterial chemoembolization combined with durvalumab with or without bevacizumab in participants with unresectable hepatocellular carcinoma eligible for embolization. J Clin Oncol. 2024;42(suppl 3):LBA432. doi:10.1200/JCO.2024.42.3_suppl.LBA432