The current landscape of therapies for hepatocellular carcinoma

Abstract

Globally, primary liver cancer is the third leading cause of cancer-related deaths, with approximately 830 000 deaths worldwide in 2020, accounting for 8.3% of total deaths from all cancer types (1). This disease disproportionately affects those in countries with low or medium Human Development Index scores in Eastern Asia, South-Eastern Asia, and Northern and Western Africa (2). Hepatocellular carcinoma (HCC), the most common type of primary liver cancer, often develops in the background of chronic liver disease, caused by hepatitis B or C virus, non-alcoholic steatohepatitis (NASH), or other diseases that cause cirrhosis. Prognosis can vary dramatically based on number, size, and location of tumors. Hepatic synthetic dysfunction and performance status (PS) also impact survival. The Barcelona Clinic Liver Cancer (BCLC) staging system best accounts for these variations, providing a reliable prognostic stratification. Therapeutic considerations of this complex disease necessitate a multidisciplinary approach and can range from curative-intent surgical resection, liver transplantation or image-guided ablation to more complex liver-directed therapies like transarterial chemoembolization (TACE) and systemic therapy. Recent advances in the understanding of the tumor biology and microenvironment have brought new advances and approvals for systemic therapeutic agents, often utilizing immunotherapy or VEGF-targeted agents to modulate the immune response. This review will discuss the current landscape in the treatments available for early, intermediate, and advanced stage HCC.

This review highlights the treatments for hepatocellular carcinoma, from early to advanced disease, and focuses on recent advances in systemic treatments and liver-directed therapies.

Graphical Abstract

Graphical Abstract.

This review highlights the treatments for hepatocellular carcinoma (HCC), from early to advanced disease, and focuses on recent advances in systemic treatments and liver-directed therapies. Staging, treatment, and prognostication is based on the Barcelona Clinic Liver Cancer (BCLC) staging system with important consideration to number, size, and location of tumors in combination with performance status (PS) and liver function. Adapted from: Villaneuva A. 2019. Hepatocellular carcinoma. N Engl J Med 380: 1450–1462; Reig M, et al. 2022. BCLC strategy for prognosis prediction and treatment recommendation: The 2022 update. J Hepatol 76:681–93. Figure created with BioRender.com.

Introduction

Liver cancer continues to pose a global health challenge with estimates showing that more than one million cases of hepatocellular carcinoma (HCC) will be diagnosed annually by 2025 (3). It is most commonly a consequence of chronic infections with hepatitis B and C viruses, although non-alcoholic steatohepatitis (NASH) is quickly becoming a common culprit; often, it develops in the background of cirrhosis of various etiologies (1,4–6). The disease is quite complex, and prognosis is influenced by more than the extent of disease with recurrence rates and risk of secondary HCC tumors affected by underlying etiology. While many diagnostic stratification systems have been developed, the Barcelona Clinic Liver Cancer (BCLC) clinical staging system is commonly used due to its reliable prognostication. Table 1 summarizes the BCLC staging classification, emphasizing how survival is impacted by the number and size of tumor, extravascular invasion, liver function, and PS (7–9). Survival has substantially improved in recent years due to the progress seen in the field; for example, the natural history of advanced stage HCC is usually estimated around 8 months, which has been markedly extended with the advances that will be discussed in this review (5).

Table 1.

The Barcelona Clinic Liver Cancer (BCLC) staging classification and associated prognosis (7–9)

BCLC stage	ECOG performance status	Tumor characteristics	Liver function status	Median overall survival
Stage 0 (very early)	Excellent (0)	Single, ≤2 cm	Intact (no portal hypertension and normal bilirubin)	>6 years
Stage A (early)	Excellent (0)	Single, ≤5 cm or Up to 3 nodules, ≤3 cm	Intact (may have portal hypertension and/or abnormal bilirubin)	>6 years
Stage B (intermediate)	Excellent (0)	Large multinodular	Intact	20–26 months
Stage C (advanced)	Adequate (1–2)*	Vascular invasion or extrahepatic spread*	Intact	12–17 months
Stage D (end-stage)	Poor (3–4)**	Any	End-stage**	3–4 months

All criteria should be fulfilled in stages 0-B. At least one criterion in stage C (*) and stage D (**). Liver function is characterized as being intact or end-stage (i.e. decompensated liver disease as evident by jaundice, ascites, or encephalopathy). Intact liver function may be better stratified by albumin-bilirubin (ALBI) score (10). Abnormal portal hypertension is defined as hepatic vein pressure gradient ≥10 mmHg; abnormal bilirubin level is defined as ≥1 mg/dL.

There are multiple modalities for the treatment of early and intermediate HCC, including surgical resection, liver transplantation, liver-directed therapies such as transarterial ablation or embolization, radiotherapy, and systemic therapy. Downstaging to be evaluated in the future for surgery or transplant can sometimes be offered. Therefore, it is best to consider multidisciplinary discussions with surgery, radiation oncology, interventional radiology, and medical oncology at high-volume specialty centers when determining treatment strategies.

For advanced or metastatic disease, sorafenib was the only approved systemic therapy in the first-line setting for HCC by the U.S. Food and Drug Administration (FDA) until 2017. Now, the landscape has dramatically changed with 11 FDA approved therapeutic options, for differing lines of treatment. Figure 1 demonstrates how prognosis has improved with these expanded options, according to SEER17 database of patients diagnosed with liver cancer (excluding intrahepatic cholangiocarcinoma) from 2000 to 2019. In 2020, the first-line standard of care broadened to atezolizumab and bevacizumab, with many clinicians favoring this approach over sorafenib or lenvatinib if patient eligible, after the results of IMbrave150 showed extended survival in advanced disease from 13.4 months with sorafenib alone to 19.2 months with atezolizumab–bevacizumab (11).

Figure 1.

1-year relative survival in liver cancer by stage and year of diagnosis. A total of 79 344 cases of liver cancer (excluding intrahepatic cholangiocarcinoma and other biliary tract cancers) diagnosed in 2000–2019 were identified in the Surveillance, Epidemiology, and End Results (SEER) 17 database and analyzed for 1-year relative age-standardized survival rates according to stage, as shown by the solid line. Stage was defined as the following: localized—confined to primary site, regional—spread to regional lymph nodes, and distant—cancer has metastasized. Prognosis improved between 2017 and 2019 when the FDA approved several treatments (regorafenib nivolumab, lenvatinib, pembrolizumab, cabozantinib, and ramucirumab), compared to 2000–2016 when only sorafenib was approved in 2007. Source: Surveillance Research Program, National Cancer Institute SEER*Stat software (www.seer.cancer.gov/seerstat) version 8.4.0.1. Surveillance, Epidemiology, and End Results (SEER) Program (www.seer.cancer.gov) SEER*Stat Database: Incidence—SEER Research Data, 17 Registries, Nov 2021 Sub (2000–2019)—Linked To County Attributes—Time Dependent (1990–2019), Income/Rurality, 1969–2020 Counties, National Cancer Institute, DCCPS, Surveillance Research Program, released April 2022, based on the November 2021 submission.

This review intends to provide a broad overview of the treatment advances in HCC, from early to advanced disease.

Treatment of early stage HCC

Liver-directed therapies, whether surgical resection or image-guided ablations, remain the only curative treatment in HCC. The outcome of these therapies is highly influenced by the size and number of liver tumors. Very early stage disease (BCLC stage 0) is defined as a single tumor less than 2 cm in size with preserved liver function and excellent PS, whereas slightly larger single tumors (2–5 cm) or several small tumors (2 or 3 tumors ≤3 cm each) constitutes early stage disease (BCLC stage A), which can represent a heterogeneous spectrum reflected by worsening 5-year survival rates with the development of portal hypertension or jaundice regardless of size. The cumulative overall 5-year survival ranges from 65.8 to 72.1% with single tumors <5 cm in size and preserved liver function; these rates decrease to 33.2% as liver impairment develops and is associated with a serum albumin level ≤3.5 g/dL (hazard ratio [HR] 1.459) and alpha-fetoprotein (AFP) level <20 ng/mL (HR 1.863) in a multi-variate analysis (12).

Surgical resection and liver transplantation

Selected patients may be considered for surgical resection and liver transplantation, which are curative options for patients with early stage HCC. Outcome varies due to presence of underlying liver disease. In the patient with HCC in a cirrhotic liver, liver resection, and transplant result in comparable survival rates (5-year overall survival of 74% for partial hepatectomy and 69% for liver transplant); whereas, liver resection in the non-cirrhotic patient has a 5-year overall survival (OS) of nearly 50% (13). Although the non-cirrhotic liver may be able to tolerate larger, more complex resections, the lower survival rate may be due to a late diagnosis with more advanced disease, impaired liver regeneration, and increased post-operative complications seen in the absence of cirrhosis with viral-related chronic liver disease or steatosis in non-alcoholic fatty liver disease (NAFLD) (14,15).

Resection is generally considered in patients with solitary lesions without major vascular invasion and well-preserved liver function with an expected liver remnant of more than 20% (16). Solitary lesions in patients with cirrhosis may also be considered if adequate liver function (i.e. Child-Pugh A) with an expected liver remanent of at least 30–40%; if volume of liver remnant is borderline, portal vein embolization of the segments to be removed should be considered to induce compensating hypertrophy of the future liver remnant (17). Locoregional control with other liver-directed techniques may allow for downstaging to meet above criteria; patients that respond may be re-considered for surgical resection in consultation with multidisciplinary specialists.

Liver transplantation is often considered for patients with cirrhosis as it provides an excellent therapy to address both malignant and non-malignant diseases (18). Transplant may also be considered in non-cirrhotic patients with larger solitary lesions (2–5 cm) or small multinodular disease (up to three lesions that are ≤3 cm each) based on the Milan criteria (19). Transplantation with living or cadaveric donors may be considered; improved survival, better organ quality, and less technically challenging procedures are more often seen with living donors (20). Allocation is often prioritized by the Model for End-Stage Liver Disease (MELD) score (21).

Peri-operative systemic therapy

There is currently no approved therapy in the neoadjuvant or adjuvant setting, although promising research is ongoing. The results of IMbrave050, a global phase III study of adjuvant atezolizumab and bevacizumab following surgical resection or ablation, were presented at the annual American Association for Cancer Research (AACR) meeting in April 2023. This practice changing clinical trial showed clinical benefit in recurrence-free survival (RFS) with adjuvant atezolizumab and bevacizumab, in comparison to active surveillance, after curative-intent resection or ablation in patients with high-risk features (22). High-risk features were defined as large tumor size (>5 cm), multiple tumors (>3), poor tumor differentiation, or vascular invasion. At the pre-specified interim analysis, the trial met its primary endpoint of improving RFS (HR 0.72, 95% CI 0.56–0.93, P = 0.012); after median follow-up of 17.4 months, the 12-month RFS event-free rate was 78% (95% CI 73–82) in the 334 patients randomized to adjuvant atezolizumab 1200 mg flat dose and bevacizumab 15 mg/kg given every 3 weeks for up to 12 months, in comparison to 65% (95 CI 60–71) in the 334 patients randomized to active surveillance. Of note, crossover was permitted. Safety profile was as expected for the study drugs; two treatment-related deaths occurred.

Methods to downstage unresectable disease remain an area of active investigation.

Image-guided percutaneous thermal ablation

Percutaneous thermal ablation with radiofrequency or microwave energy is an important curative option for patients with early stage HCC (generally tumors <3 cm). Thermal ablation may be an attractive option for patients who are poor surgical candidates or refuse surgery. Heat-based ablative modalities generate energy that heats tissue up to at least 60°C, leading to thermal necrosis and cell death. These procedures are generally considered to have low complication rates and short hospital stays; the reduced blood loss may have broader implications for patients with cirrhosis and associated coagulopathies (23). Tumors larger than 3 cm or located near a major vessel may not derive as much benefit or present too much risk for this procedure (24). It should be noted that a margin of normal tissue may also be treated with these ablative techniques. Compared to surgical resection, thermal ablative modalities have similar rates of complete tumor elimination (100% with surgery and 94.7% with ablation) and survival outcomes (3-year OS: 86.4% with surgery and 87.1% with ablation, P = 0.808) (25). Moreover, this therapy is less invasive and has cost-saving potential with less personnel and length of post-procedural hospital stay.

Microwave ablation (MWA) offers some advantages over radiofrequency ablation (RFA) that should be considered. With improvement in technology, newer generations of MWA probes allow for larger ablations (up to 5 cm) and near major vessels (26). Importantly, MWA techniques do not appear to succumb to the heat-sink effect like RFA. The heat-sink effect with RFA is a phenomenon that occurs when the flowing blood from nearby intrahepatic vessels cause a cooling effect on the thermal ablation zone, dissipating thermal energy and leading to unpredictable, potentially inadequate ablation. While similar therapeutic effects as RFA, MWA provides more predictable, larger ablation zones, improved technical feasibility by allowing multiple probes to be used simultaneously and decreased ablation times through faster heating time of probes, and lower complication rates during percutaneous procedures related to less susceptibility to the heat-sink effect by local perfusion (27).

External beam radiation therapy

Historically, external beam radiation therapy (EBRT) has had a limited role in the management of early stage HCC; however, according to NCCN guidelines, it may be considered as an alternative if aforementioned therapies have failed or are contra-indicated such as in medically inoperable patients due to co-morbidities. It may also be an effective bridging therapy for patients awaiting a liver transplant. Preferred modalities include stereotactic body radiotherapy (30–50 Gy typically in 3–5 fractions if <3 tumors) or intensity-modulated radiotherapy (37.5–72 Gy in 10–15 fractions) (28). Radiation may play a better role in the palliative setting for symptomatic lesions, in preventing complications from metastatic lesions such as osseous or central nervous system metastasis, or for providing local control to extensive disease burden in liver (alone or in combination with other locoregional techniques) (29). There may also be a role for it in extensive liver tumor burden. Importantly, safety data is limited in patients with Child-Pugh B or more severe liver impairment.

Treatment of intermediate stage HCC

Approximately 30% of HCC patients present with intermediate stage disease (BCLC stage B), which includes multinodular disease, not limited to a single segment, without extrahepatic or vascular invasion (30,31). Patients should have preserved liver function and good PS. This stage is very heterogeneous with disease encompassing innumerable small tumors to a few large (>10 cm) tumors. Median OS without treatment is 10 months (32,33). This has improved due to symptom management and treatment advances in locoregional therapies but unfortunately remains incurable. Therefore, the preservation of liver function deserves equal attention as obtaining a high objective response as clinicians select therapies to prolong survival (34).

Transarterial chemoembolization

Conventional transarterial chemoembolization (TACE) involves intra-arterial infusion of a highly concentrated dose of cytotoxic agent into the feeding hepatic arteries of the tumor, followed by embolization of those vessels with gelatin sponges or polyvinyl alcohol particles (35–38). Cisplatin, doxorubicin, or these in combination with other agents are commonly used cytotoxic agents, but to date, no specific cytotoxic agent or dose has shown superiority, although lower doses may be associated with less toxicity (36–38).

Multiple randomized clinical trials have demonstrated that TACE prolongs median OS in patients with intermediate stage HCC to 20–26 months in specific subgroups of patients, but due to the heterogeneity of this disease, not all patients with intermediate stage HCC will benefit from a TACE procedure (34,39,40). Several scoring systems (i.e. six- and-twelve score, ALBI-TAE, STATE score) have been proposed to predict outcomes, and thus, help to better select patients (41–43). Though TACE remains the first-line standard treatment option recommended in international guidelines (44,45), TACE is only suitable for patients with preserved portal flow, defined tumor burden, and selective access to feeding tumor arteries, according to the BCLC 2022 version (7,44,45).

Attempting to refine the efficacy of TACE and limit toxicity from systemic exposure of chemotherapy, microsphere beads loaded with chemotherapy were added to the procedure to deliver targeted and sustained release of cytotoxic chemotherapy directly to the feeding artery of the tumor, called drug-eluting beads TACE (DEB-TACE) (46). In comparison to conventional TACE, DEB-TACE has demonstrated similar objective response rates (52% versus 44%, P = 0.11), time to progression (approximately 9 months), and 2-year survival rates (56.8% versus 55.4%, P = 0.949) (47–49). DEB-TACE has generally not shown superiority over conventional TACE, although in the PRECISION V study more advanced patients with Child-Pugh B, ECOG 1, bilobar disease, or recurrent disease showed an increase in objective response with DEB-TACE over cTACE (P = 0.038) with improved tolerability and less serious hepatotoxicity (47).

Systemic therapy combinations with TACE have had limited success in intermediate stage HCC. There are several ongoing clinical trials with the combination of immunotherapy and TACE (NCT02821754 and NCT037789); the results of these trials should be informative for clinical practice.

Prior to acceptance of TACE as standard practice for intermediate stage HCC, transarterial embolization techniques without cytotoxic therapies were successful in palliation of unresectable HCC tumors. This bland embolization, or transarterial embolization (TAE), inhibited tumor growth through ischemia induced by embolization tumor blood flow. Historically, controversy has rose regarding merits of TACE over TAE and the value added by chemotherapy due to limited quality studies with head-to-head comparison. With that said, a recent retrospective study in 2021 attempted to elucidate the differences in efficacy and tolerance of TAE compared to TACE in 265 patients with intermediate stage HCC at two centers in France. The study demonstrated improvement in radiological responses with TACE, but there lacked a difference in progressive disease or overall survival between the two treatment groups when weighted by a propensity score (50). Consistent with prior reports, higher rate of toxicity was associated with TACE, but no difference in severe adverse events was detected. Therefore, the goal of therapy, PS, liver function, and patient co-morbidities may help guide selection of therapy as some symptomatic patients or downstaging candidates for liver transplant may benefit more from tumor control with a larger objective response.

Transarterial radioembolization

Transarterial radioembolization (TARE), also known as Selective Internal Radiation Therapy (SIRT), has emerged as an alternative therapy to TACE after the multicenter LEGACY study showed clinical meaningful response rates (88.3% best ORR by localized mRECIST, 95% CI 82.4–92.4%) and prolonged duration of response (11.8 months, median) in unresectable, solitary HCC tumors ≤8 cm (51). In contrast to TACE, radioembolization involves injecting micro-embolic radioactive particles into the hepatic artery that supplies the tumor, which become trapped in the tumor due to arteriolar capillary blockade, permitting a local radiotherapeutic effect to the tumor (and minimal surrounding normal liver tissue) through internally emitted beta radiation that usually dissipates after 2 weeks (52). Yttrium-90 (Y-90) is the most studied radioisotope in this minimally invasive procedure and was approved in March 2021 for early disease but is widely used in more advanced disease. TARE may be preferred in some cancer centers in locally advanced, liver-limited disease who are poor candidates for TACE such as large tumors involving more than two segments, portal vein invasion or occlusion, and/or progressive disease post-TACE (53). The reported OS with TARE in patients with intermediate stage HCC is 16.4–18.0 months (54,55).

When compared head-to-head with conventional TACE in a phase II randomized prospective trial evaluating time to progression, TARE with Y-90 improved time to progression significantly compared to conventional TACE (>26 months versus 6.8 months; HR 0.122, P = 0.007), whereas median survival time, censored to liver transplantation, were similar between the two groups (18.6 months versus 17.7 months, respectively; P = 0.99) (54). A meta-analysis comparing TARE and TACE included 10 studies with only 2 prospective randomized trials and found that tumor progression at one year is delayed more with TARE (1-year PFS OR = 1.67, P = 0.02) and a survival benefit is seen at 2 and 3 years after treatment with TARE but not after 1 year (OR at 1 year: 1.01, P = 0.93; OR at 2 years: 1.43, P = 0.01; OR at 3 years: 1.48, P = 0.04) (56). Prospective head-to-head comparison trials with the primary endpoint of overall survival are warranted.

Systemic therapy

While TACE is often considered first-line therapy for intermediate stage HCC, there is some evidence that patients with larger tumor burden or multinodular intermediate stage HCC may benefit more from lenvatinib than cTACE with overall survival significantly longer with the VEGF TKI (37.9 versus 21.3 months; HR 0.48, P < 0.01) (57). Thus, certain clinical scenarios require consideration of upfront systemic therapy in intermediate stage HCC such as large bilobar or infiltrative tumors, residual disease after arterial-directed therapies not amenable to additional local therapies, or as part of a strategy to downstage for other curative therapies (7,34,44).

Downstaging for other curative therapies

Neoadjuvant systemic therapy is an attractive approach for downstaging intermediate stage HCC. There has been recent success in early phase clinical trials with neoadjuvant checkpoint inhibitors showing major pathological responses at time of curative-intent surgical resection without delaying time of surgery or causing unacceptable toxicities (58–60). However, to date, there are no approved therapies in this setting.

Alternatively, unresectable disease should be considered for downstaging with liver-directed therapies and if response is seen, be re-evaluated for curative-intent surgical resection, thermal ablative procedure, or liver transplant. Long-term follow-up of patients with HCC after downstaging and undergoing liver transplant showed a 4-year post-transplantation survival rate of 92.1% (61).

Treatment of advanced stage HCC

Before 2007, there was no effective systemic treatment for advanced HCC. Improved understanding of the immunosuppressive tumor microenvironment and immune escape mechanisms led to improved drug developments for advanced HCC. Sorafenib received the first FDA-approval in 2007, becoming the standard of care for many years. The following nine single-agent or combination therapies have since been approved since 2017: regorafenib, nivolumab, lenvatinib, pembrolizumab, cabozantinib, ramucirumab, nivolumab–ipilimumab, atezolizumab–bevacizumab, and durvalumab–tremelimumab, most recently.

These advances in systemic therapies have changed the landscape of advanced stage HCC, leading to improved survival and better quality of life. This review will exclusively focus on the clinical trials that led to the FDA-approvals for these 10 single-agent or combination therapies for advanced disease; however, clinical research has accelerated in the last decade, adding to our understanding of the biology of HCC, mechanisms of action, and factors associated with drug susceptibility or resistance. The clinical outcomes of completed phase III randomized controlled trials of systemic therapies in advanced HCC in the USA since 2005 are described in Table 2, and a more detailed account of the studies that met their primary endpoint, leading to FDA-approval, are discussed below. While an in-depth discussion of ongoing or unreported clinical trials is beyond the scope of this review, Table 3 describes ongoing phase III clinical studies that are not yet finalized and remain active and/or still recruiting.

Table 2.

Outcomes of completed phase III randomized controlled trials of systemic therapies in advanced HCC in the United States, 2005–2023

Randomized controlled trial (clinicaltrial.gov identifier)	Pts (n)	Intervention	Primary endpoint	Outcomes	References
Met primary endpoint
CELESTIAL: A Study of Cabozantinib vs Placebo in Subjects with HCC Who Have Received Prior Sorafenib (NCT01908426)	707	Cabozantinib vs placebo (2L)	OS	10.2 vs 8 mo (HR 0.44, 95% CI 0.36–0.52, P < 0.001)	(62)
REFLECT: A Multicenter, Randomized, Open-Label, Phase 3 Trial to Compare the Efficacy and Safety of Lenvatinib vs Sorafenib in First-Line Treatment of Subjects with Unresectable HCC (NCT01761266)	954	Lenvatinib vs Sorafenib (1L)	OS (non-inferiority)	13.6 vs 12.3 mo (HR 0.92, 95% CI 0.79–1.06)	(63)
REACH-2: A Study of Ramucirumab vs Placebo as Second-Line Treatment in Pts with HCC and Elevated Baseline AFP after Sorafenib (NCT02435433)	399	Ramucirumab vs placebo (2L)	OS	8.5 vs 7.3 mo (HR 0.710, 95% CI 0.531–0.949, P = 0.0199)	(64)
RESORCE: A Study of Regorafenib in Pts with HCC After Sorafenib (NCT01774344)	573	Regorafenib vs placebo (2L)	OS	10.6 vs 7.8 mo (HR 0.63, 95% CI 0.5–0.79, P < 0.0001)	(65)
IMbrave150: A Study of Atezolizumab in Combination with Bevacizumab vs Sorafenib in Pts with Untreated Locally Advanced or Metastatic HCC (NCT03434379)	558	Atezolizumab + Bevacizumab vs Sorafenib (1L)	OS	19.2 vs 13.4 mo (HR 0.58, 95% CI 0.42–0.79, P < 0.001)	(66)
			PFS-IRF	6.8 vs 4.3 mo (HR 0.59, 95% CI 0.47–0.76, P < 0.001)
SHARP: A Phase III Study of Sorafenib in Pts with Advanced HCC (NCT00105443)	602	Sorafenib vs placebo (1L)	OS	10.7 vs 7.9 mo (HR 0.69, 95% CI 0.55–0.87, P < 0.001)	(67)
			TTSP	4.1 vs 4.9 mo, P = 0.77
Did not meet primary endpoint
A Phase III Study of ADI-PEG 20 vs Placebo in Subjects with Advanced HCC Who Have Failed Prior Systemic Therapy (NCT01287585)	636	ADI-PEG 20 vs placebo (2L)	OS	7.8 vs 7.4 mo (HR 1.02, P = 0.88)	(68)
METIV-HCC: A Study of Tivantinib in Subjects with MET Diagnostic-High Inoperable HCC Treated with One Prior Systemic Therapy (NCT01755767)	383	Tivantinib vs placebo (2L)	OS	8.4 vs 9.1 mo (HR 0.97, 95% CI 0.75–1.25, P = 0.81)	(69)
PHOCUS: A Study Comparing PexaVec (Vaccinia GM CSF/ Thymidine Kinase-Deactivated Virus) plus Sorafenib vs Sorafenib alone in Pts with Advanced HCC without Prior Systemic Therapy (NCT02562755)	459	PexaVec + Sorafenib vs Sorafenib (1L)	ORR per mRECIST	Failed interim futility analysis	(70)
BRISK FL: A Study of Brivanib vs Sorafenib as First-line Treatment in Pts with Advanced HCC (NCT00858871)	1714	Brivanib vs Sorafenib (1L)	OS (non-inferiority)	9.5 vs 9.9 mo (HR 1.06, 95.8% CI 0.93–1.22)	(71)
EVOLVE-1: A Global Study to Evaluate the Efficacy and Safety of Everolimus in Adult Pts with Advanced HCC After Failure of Sorafenib Treatment (NCT01035229)	546	Everolimus vs placebo (2L)	OS	7.6 vs 7.3 mo (HR 1.05, 95% CI 0.86–1.27, P = 0.68)	(72)
SEARCH: Sorafenib-Erlotinib Combination Therapy for First Line Treatment for HCC (NCT00901901)	732	Erlotinib + Sorafenib vs Sorafenib (1L)	OS	9.5 vs 8.5 mo (HR 0.929, P = 0.408)	(73)
CALGB 80802: Sorafenib with or without Doxorubicin in Pts with Advanced HCC (NCT01015833)	356	Doxorubicin + Sorafenib vs Sorafenib (1L)	OS	Failed interim futility analysis 9.3 vs 9.4 mo (HR 1.05, 95% CI 0.83–1.31)	(74)
Nolatrexed Dihydrochloride compared with Doxorubicin in Treating Pts with Recurrent or Unresectable Liver Cancer (NCT00012324)	445	Nolatrexed vs Doxorubicin (1L)	OS	22.4 vs 32.3 weeks (HR 0.753 in favour of doxorubicin, P = 0.0068)	(75)
BRISK-PS: A Study of Brivanib vs Placebo in Subjects with Advanced HCC Who Have Failed Sorafenib (NCT00825955)	587	Brivanib vs placebo (2L)	OS	9.4 vs 8.2 mo (HR 0.89, 95.8% CI 0.69–1.15, P = 0.33)	(76)

AFP, alpha-feta protein; BSC, best supportive care; HCC, hepatocellular carcinoma; Pts, patients; L, line of therapy; mo, months; vs, versus; OS, overall survival; ORR, overall response rate; RECIST, response evaluation criteria in solid tumors; TTSP, time to symptomatic progression; PFS-IRF, progression-free survival as assessed by independent review facility.

Table 3.

Selected ongoing prospective phase III randomized controlled trials of systemic therapies in advanced HCC in the United States

Randomized controlled trial (clinicaltrial.gov identifier)	Pts (n)	Intervention	Primary endpoint	Outcomes (if reported)	References
ICI monotherapy or combined anti-PD-(L)1 + anti-CTLA-4
CheckMate 459: A Study of Nivolumab vs Sorafenib as First-Line Treatment in Pts with Advanced HCC (NCT02576509)	743	Nivolumab vs Sorafenib (1L)	OS	16.4 vs 14.7 mo (HR 0.85, 95% CI 0.72–1.02, P = 0.075)	(77)
				Negative study
RATIONALE-301: A Phase 3 Study of Tislelizumab vs Sorafenib as First-Line Treatment in Pts with Unresectable HCC (NCT03412773)	674	Tislelizumab vs Sorafenib (1L)	OS	15.9 vs 14.1 mo (stratified HR 0.85, 95% CI 0.712–1.019)	(78)
				Positive study as met non-inferior OS
CheckMate 9DW: A Study of Nivolumab in Combination with Ipilimumab Compared to Sorafenib or Lenvatinib as First-Line Treatment in Pts with Advanced HCC (NCT04039607)	732	Nivolumab + Ipilimumab vs Sorafenib or Lenvatinib (1L)	OS	Not yet completed
HIMALAYA: A Study of Durvalumab and Tremelimumab as First-line Treatment in Pts with Advanced HCC (NCT03298451)	1504	STRIDE vs Sorafenib (1L) Durvalumab monotherapy vs Sorafenib (1L)	OS (superiority)	16.43 vs 13.77 mo (HR 0.78, 96.02% CI 0.65–0.93, P = 0.0035)	(79)
			OS (non-inferiority)	16.56 vs 13.77 mo (HR 0.86, 95.67% CI 0.73–1.03)
				Positive study as met superior OS with STRIDE and non-inferiority OS with Durvalumab monotherapy
ICI + VEGF inhibitor
A Study of PD-1 Antibody Camrelizumab Plus Apatinib Mesylate vs Sorafenib as First-Line Therapy in Pts with Advanced HCC (NCT03764293)	543	Camrelizumab + Apatinib vs Sorafenib (1L)	OS, PFS	Not yet reported
LEAP-002: A Study to Evaluate the Safety and Efficacy of Lenvatinib in Combination with Pembrolizumab vs Lenvatinib as First-line Therapy of Pts with Advanced HCC (NCT03713593)	794	Pembrolizumab + Lenvatinib vs Lenvatinib (1L)	OS	21.2 vs 19.0 mo (HR 0.84, 95% CI 0.708–0.997, P = 0.0227)	(80)
			PFS-BICR	8.2 vs 8.0 mo (HR 0.867, 95% CI 0.734–1.024, P = 0.0466)
				Negative study as did not meet pre-specified statistical significance
A Study of Nofazinlimab in Combination with Lenvatinib vs Lenvatinib Alone as First-Line Therapy in Pts with Advanced HCC (NCT04194775)	534	Nofazinlimab + Lenvatinib vs Lenvatinib (1L)	OS, PFS-BICR	Not yet completed
A Phase III Study of Toripalimab Combined with Lenvatinib vs Lenvatinib as the 1st-line Therapy for Advanced HCC (NCT04523493)	519	Toripalimb + Lenvatinib vs Lenvatinib (1L)	OS, PFS-BICR	Not yet completed
COSMIC-312: A Study of Cabozantinib in Combination with Atezolizumab vs Sorafenib in Pts with Advanced HCC Who Have Not Received Previous Systemic Anticancer Therapy (NCT03755791)	740	Cabozantinib ± Atezolizumab vs Sorafenib (1L)	PFS-BICR	6.8 vs 4.2 mo (HR 0.63, 99% CI 0.44–0.91, P = 0.0012)	(81)
			OS	15.4 vs 15.5 mo (HR 0.90, 96% CI 0.69–1.18, P = 0.44)
				Interim analysis did not meet dual endpoints
Other
NRG/RTOG 1112: Sorafenib with or without SBRT in HCC (NCT01730937)	193	SBRT + Sorafenib vs Sorafenib	OS	15.8 vs 12.3 mo (HR 0.77, one-sided P = 0.0554) Positive study	(82)
LIVERATION: Namodenoson in the Treatment of Advanced HCC in Pts with Child-Pugh Class B7 Cirrhosis (NCT05201404)	471	Namodenoson vs placebo (2L)	OS	Not yet completed
Evaluating Length of Treatment with PD-1/PD-L1 Inhibitor in Advanced Solid Tumors (NCT04157985)	578	Continue vs discontinue anti-PD(L)1 after 12 mo (any line)	TTNT, PFS	Not yet completed

HCC, hepatocellular carcinoma; ICI, immune checkpoint inhibitor; Pts, patients; L, line of therapy; mo, months; vs, versus; OS, overall survival; ORR, overall response rate; TTNT, time to next treatment; PFS-BICR, progression-free survival as assessed by blinded independent central review; SBRT, stereotactic body radiotherapy; TACE, transarterial chemoembolization.

The vascular endothelial growth factor (VEGF) and Raf-1 pathways have been implicated in the molecular pathogenesis of HCC; thus, inhibition of the VEGF pathway became an early target for drug therapies for advanced HCC with its anti-angiogenesis properties playing an important role in these hypervascular tumors. However, in more recent years, it appears that the VEGF pathway may have wider implications than inhibiting tumor angiogenesis. The VEGF pathway appears to contribute to the immune suppression of the tumor microenvironment by inhibiting antigen-presenting cells and effector T-cells and also activating immunosuppressive elements like T_reg-cells and myeloid-derived suppressors cells (83). VEGF further exerts immunosuppressive effects by creating T-cell exhaustion through increasing PD-L1, CTLA-4, TIM3, and LAG3 expression on T-cells (84). Early hematopoietic progenitor cells cannot differentiate into CD4⁺ and CD8⁺ lymphocytes, decreasing T-cell proliferation and cytotoxic effects (85). In summary, VEGF plays a key role in tumor immunomodulation as well as angiogenesis, and the inhibition of this pathway, with potential synergy by immune checkpoint inhibitors (ICIs), provides the rationale for most of the approved therapies and ongoing clinical trials, turning a ‘cold’ tumor into a ‘hot’ tumor, as reviewed below.

Early on, the rationale for investigation of sorafenib was evident due to its targeted activity of inhibition of multiple receptor tyrosine kinases, including VEGF receptors 1/2/3, platelet-derived growth factor receptor β, Raf-1, and B-Raf. The small molecule multikinase inhibitor can facilitate apoptosis, suppressing tumor cell proliferation, and mitigating tumor angiogenesis. In 2007, the SHARP trial demonstrated superiority of sorafenib over placebo based on a median OS of 10.7 months versus 7.9 months (HR 0.69, 95% CI 0.55–0.87; P < 0.001); without differences in time to symptomatic progression (4.1 months versus 4.9 months, P = 0.77) (67). Toxicities occurred in 80% of patients, most commonly diarrhea, weight loss, hand-foot skin reaction, and hypophosphatemia. The rate of discontinuation was 38%, most often due to toxicities.

In 2018, lenvatinib, another small molecule oral tyrosine kinase inhibitor (TKI) that targets VEGFR-1-3, PDGFR-α, FGFR-1-4, KIT, and RET, challenged sorafenib in the first-line setting in the phase III, open-label, randomized REFLECT trial. While powered for superiority and non-inferiority, it only met criteria for non-inferiority, mOS was 13.6 months for lenvatinib versus 12.3 months for sorafenib (HR 0.92, 95% CI 0.79–1.06) (63). Approximately 50% of both groups experienced G3–4 treatment-related adverse events (trAE), most commonly hypertension, hand-foot skin reaction, anorexia, diarrhea, and weight loss, with a 15% withdrawal rate.

The IMbrave150 trial with atezolizumab (anti-PDL-1 antibody) plus bevacizumab (IgG1 antibody that targets VEGF-A ligand) demonstrated improved survival over sorafenib in 2020. 501 Patients with systemic treatment-naive unresectable HCC were randomized in a 2:1 ratio to either atezolizumab–bevacizumab or sorafenib in this multinational, open-label, phase III randomized trial (11). Eligibility criteria allowed for advanced disease with macrovascular invasion of the main portal vein and >50% of hepatic involvement—a population commonly excluded in clinical trials. Patients were required to undergo upper endoscopy within 6 months prior to enrollment and complete treatment of esophageal varices, if needed, to mitigate bleeding risks with bevacizumab. The trial met its co-primary end points of median OS and PFS. At time of primary analysis, mOS was 19.2 months versus 13.4 months (HR 0.58, 95% CI 0.42–0.79; P < 0.001) and mPFS was 6.8 months versus 4.3 months (HR 0.59, 95% CI 0.47–0.76; P < 0.001) in the treatment and control arms, respectively. ORR was 33.2% with a durable response of 18.1 months by mRECIST. Patient-reported outcomes were excellent with improved quality of life for a longer duration in the treatment arm (median time until symptomatic duration, 11.2 months versus 3.6 months). Higher rates of G3–4 hypertension were seen in the atezolizumab–bevacizumab group (15.2% versus 12.2%).

Both multikinase TKIs, regorafenib and cabozantinib, showed survival advantage in the second-line setting after progression on sorafenib in the randomized, placebo-controlled, phase III RESORCE and CELESTIAL trials, respectively. Regorafenib improved mOS to 10.6 months versus 7.8 months (HR 0.63, 95% CI 0.5–0.79; P < 0.0001) with an ORR 11% versus 4% (65). Cabozantinib improved mOS to 10.2 months versus 8 months (HR 0.44, 95% CI 0.36–0.52; P < 0.001) with single digit ORR (62).

Ramucirumab, an IgG1 antibody targeting the extracellular domain of VEGFR-2, is the only biomarker-guided therapy approved for HCC based upon the results of the REACH-2 randomized, double-blind, placebo-controlled phase III trial that enrolled patients with a baseline AFP ≥400 ng/dL after disease progression on first-line sorafenib. Median OS was 8.5 months versus 7.3 months (HR 0.710, 95% CI 0.531–0.949; P = 0.0199) (64). Importantly, three patients in the ramucirumab group died from treatment-related AEs.

The FDA granted accelerated approval to pembrolizumab monotherapy (an anti-PD1 antibody) after progression on first-line sorafenib based on the KEYNOTE-224 phase II study showing a durable response with 89% of responses lasting longer than 6 months and 56% longer than 12 months (ORR 17%, 95% CI 11–26) (86,87). The placebo-controlled confirmatory phase III KEYNOTE-240 trial could not meet its co-primary end points of OS and PFS as it did not reach statistical significance per specified criteria (mOS 13.9 months with pembrolizumab monotherapy versus 10.6 months with placebo; HR 0.781, 95% CI 0.611–0.998; P = 0.0238) (88). As of April 2021, the FDA upheld the accelerated approval in anticipation of the ongoing KEYNOTE-394 phase III trial. KEYNOTE-394 similarly evaluated pembrolizumab monotherapy in the second-line setting post-sorafenib for patients in Asia with advanced HCC. The randomized, double-blind phase III study randomized 453 patients to pembrolizumab monotherapy or placebo with best supportive care and demonstrated statistically significant and clinically meaningful improvement in survival (89). The study met its primary endpoint of overall survival with a median OS of 14.6 months (95% CI 12.6–18.0) in the patients treated with pembrolizumab alone, compared to 13.0 months (95% CI 10.5–15.1) in patients treated with best supportive care only (HR 0.79, 95% CI 0.63–0.99; P = 0.0180). The survival benefit appears durable at 24 months with 34.3% of patients alive who were treated with pembrolizumab, compared to 24.9% for those treated with best supportive care alone. These results support the trends seen in KEYNOTE-224 and KEYNOTE-240 that pembrolizumab monotherapy may provide a durable survival benefit in the second-line setting.

The phase II CheckMate 040 trial similarly led to accelerated approval of both nivolumab monotherapy (anti-PD1 antibody) and nivolumab plus ipilimumab (anti-CTLA-4 antibody) in the second-line setting. Bristol-Myers Squibb later withdrew the U.S. indication voluntarily for nivolumab monotherapy due to lack of superiority in CheckMate 459 (90). However, long-term follow-up (33.6 months) of survival outcomes shows clinically meaningful improvement in OS with nivolumab monotherapy over sorafenib in first-line setting with mOS 16.4 months versus 14.8 months (HR 0.85, 95% CI 0.72–1.00; P = 0.0522) (91). Importantly, over time, nivolumab was associated with greater preservation of liver function as evident by improvement in albumin-bilirubin levels and Child-Pugh scores. The combination of nivolumab–ipilimumab demonstrated an ORR 32% (95% CI 20–47%) with an unreached duration of response (8.3–33.7+ months) (92). One treatment-related death occurred (pneumonitis). The confirmatory phase III CheckMate 9DW trial is underway with comparators either sorafenib or lenvatinib in the first-line setting.

The HIMALAYA multinational, open-label, randomized phase III trial utilized a single-priming dose of tremelimumab (anti-CTLA-4) combined with durvalumab (anti-PDL1 antibody) in 1171 untreated patients. This STRIDE regimen improved anti-PDL1 inhibition without adding significant toxicity in a phase II safety study. The STRIDE regimen significantly improved mOS over sorafenib (16.43 months versus 13.77 months) (79). Associated with a 22% reduction in risk of death over sorafenib (HR 0.78, 95% CI 0.65–0.93, P = 0.0035), the combination regimen, durvalumab–tremelimumab, was the most recent FDA approved therapy for advanced HCC in October 2022 based on the successful results of this trial. Additionally, the trial showed that durvalumab monotherapy was non-inferior to sorafenib with mOS 16.6 months versus 13.8 months (HR 0.86, 96% CI 0.73–1.03) and is now included in the NCCN guidelines for consideration in first-line setting (28).

Summary and future directions

Treatment of HCC is complex, highly dependent on stage at diagnosis, and can range across multiple disciplines. Curative strategies are currently limited to surgical resection, liver transplant, and ablative techniques, which require early detection; while survival rates are generally good, outcome is highly influenced by any associated underlying liver parenchymal damage such as cirrhosis, chronic liver injury from viral hepatitis, or steatosis due to NAFLD. Clinical research is ongoing to improve upon methods for downstaging unresectable disease to broaden the population eligible for these curative-intent therapies. Multiple locoregional techniques (transarterial chemo- or radioembolization and external beam radiation) allow for good disease control and ­potentially downstaging unresectable disease. For non-surgical patients with intermediate disease, conventional TACE is most recommended upfront by many clinical practice guidelines, and clinicians may even consider upfront systemic therapy based upon patient specific factors such as more diffuse infiltrative, large tumors (7,34,44).

Significant advances in the understanding of HCC tumor biology have led to multiple new approvals for advanced stage disease, which have led to prolonged survival, improved symptom control, and better quality of life. Current therapies primarily focus on targeting angiogenesis with multikinase VEGF TKIs and tumor microenvironment with ICIs. While many clinicians opt for atezolizumab–bevacizumab as first-line standard of care, patient specific factors may influence decision towards another first-line options such as sorafenib, lenvatinib, or durvalumab–tremelimumab. With many options in the first and second-line setting including TKIs or immunotherapy, the optimal sequencing of these therapies is not clear, and further research is needed to better understand this challenging decision.

Looking forward, new combination regimens of TKIs plus ICIs and novel immunotherapies are promising therapies being investigated in advanced stage disease. The IMbrave251 trial is exploring management after disease progression on atezolizumab–bevacizumab by adding either sorafenib or lenvatinib to atezolizumab in the second-line setting (5). As immune escape pathways and mechanisms of ICI resistance become clearer, the potential for novel immunotherapeutic approaches expands: oncolytic viral vaccines; genetically engineered T-cells targeting AFP, GCP3 and MUC1; immune-checkpoint bispecific antibodies; and targeting other ­co-inhibitory checkpoints such as LAG3 or TIM3 (93,94).

The importance of biomarkers in the future is clear. While other solid tumors have many biomarker-driven effective therapies, HCC only has one; ramucirumab is associated with better outcomes if elevated serum AFP. AFP has limited utility though since it may not be detectable in early stage disease and can be associated with other non-HCC malignancies and in pregnancy (95,96). To date, targeting molecular mutations has not been very successful. However, the landscape may change as more correlative studies are analyzed from the many recently completed and ongoing prospective randomized clinical trials. Molecular profiling and genetic sequencing of tumor biopsies will provide a better understanding of potential therapeutic targets, determining mechanisms of drug resistance, and developing diagnostic tumor markers. The inclusion in clinical trials of DNA-sequencing technology, such as circulating tumor DNA, has implications for monitoring treatment response or developing an early marker of recurrence. We hope further biomarker-driven therapies in HCC improve the treatment possibilities and prognosis for these patients in the future.

Glossary

Abbreviations:

AFP

alpha-fetoprotein

BCLC

Barcelona Clinic Liver Cancer

EBRT

external beam radiation therapy

HCC

hepatocellular carcinoma

MWA

microwave ablation

NASH

non-alcoholic steatohepatitis

OS

overall survival

PS

performance status

RFA

radiofrequency ablation

TACE

transarterial chemoembolization

TAE

transarterial embolization

TARE

transarterial radioembolization

Y-90

Yttrium-90

Funding

CM is a recipient of the Robert A. Winn Career Development Award Program. TFG was supported by the Intramural Research Program of the NIH, NCI (ZIA BC 011343 and ZIA BC 011870).

Conflict of Interest Statement: None declared.

Data availability

The data underlying this article are available in the article.