Hepatocellular carcinoma: recent advances and emerging medical therapies

Hepatocellular carcinoma remains a deadly disease with poor prognosis in patients with unresectable cancer. Trans-arterial chemoembolization is the primary locoregional therapy for intermediate-stage hepatocellular carcinoma, with an estimated median overall survival of less than two years. For almost a decade, sorafenib has been the only standard systemic treatment for metastatic disease or tumors which progress or are considered unsuitable for locoregional therapy. Major breakthroughs have been made over the past few years in the management of hepatocellular carcinoma, especially in medical therapies for advanced disease. In this article, recent advances in intra-arterial therapy, multi-kinase inhibitors, and immunotherapy will be reviewed.


Introduction
Among all cancers globally, liver cancer ranked the sixth in incidence and the fourth in mortality, accounting for 841,080 new cases and 781,631 deaths worldwide in 2018 1 . Hepatocellular carcinoma (HCC) is the predominant form of primary liver cancer, with risk factors including hepatitis B virus (HBV) and hepatitis C virus (HCV) infection as well as cirrhosis of any etiology such as alcoholic liver disease, non-alcoholic steatohepatitis, and primary biliary cholangitis. Diagnosis of HCC is usually made by contrast enhanced imaging. Barcelona Clinic Liver Cancer (BCLC) staging remains the most commonly used staging system 2,3 . BCLC staging incorporates tumor burden, liver reserve, and performance status and classifies HCC into very early stage (0), early stage (A), intermediate stage (B), advanced stage (C), and terminal stage (D) 4 . The management strategy can then be recommended according to different stages of disease 2 . Recently, a popular staging system, the Hong Kong liver cancer staging system (HKLC), which is derived from a group of Hong Kong experts 5 , was introduced. According to this study, it has a better discriminating power than BCLC 6 . Nevertheless, it is noteworthy that BCLC is derived from mostly HCV-related HCC cases, while HKLC is derived from an Asian cohort where HBV is the predominant risk factor for HCC. Surgical resection, tumor ablation, and liver transplantation are curative treatments indicated for very early and early stage HCC.
Owing to a lack of widely implemented HCC surveillance programs in many parts of the world, most patients are unfortunately diagnosed at an advanced stage, precluding curative therapy 7 .
Non-surgical management for HCC includes locoregional intra-arterial therapy, systemic treatment with multi-kinase inhibitors, and immunotherapy. Recent advances and emerging strategies in these modalities to improve the outcome of HCC will be reviewed.

Intra-arterial therapy
Trans-arterial chemoembolization Trans-arterial chemoembolization (TACE) is the most popular form of intra-arterial therapy and is the established first-line treatment for intermediate-stage HCC 2,3 . It involves arterial catheterization, usually via the femoral artery, to cannulate the hepatic arterial branches. Cytotoxic agent, usually doxorubicin or cisplatin, is then delivered to the highly vascularized liver tumor. It is followed by blocking of the feeding vessel with an embolization agent, commonly lipiodol, to induce ischemic necrosis of the tumor. The rest of the liver parenchyma is relatively preserved as a result of dual blood supply from both the portal vein and the hepatic artery 8 . Significant improvement of survival was shown in patients with locally advanced HCC and Child-Pugh A or B cirrhosis treated with TACE as compared with best supportive care, with relative risk of death reduction more than 50% 9-11 (  43.5% for TACE, P = 0.110 for superiority), but a significant reduction in liver toxicity and doxorubicin-related side effects was seen in the DEB-TACE arm 19 . TARE is a form of selective internal radiation therapy (SIRT), in which radioactive Yttrium-90 (Y90) microspheres are introduced into the tumor vasculature. The main anti-tumor effect in TARE is achieved by radiation instead of embolization.
As the patency of the hepatic artery is maintained, TARE can be used in HCC with main portal vein invasion or thrombosis, which is considered a contraindication for conventional TACE.
A meta-analysis of eight studies involving 1,500 patients demonstrated superiority of TARE over TACE in overall survival (OS), 3-year OS, time to progression (TTP), and hospitalization days 20 . A subsequent phase II randomized trial also showed significantly longer TTP in the TARE group compared with conventional TACE (>26 months versus 6.8 months, P <0.01) 21 . However, improvement of OS was not shown. A randomized controlled trial comparing DEB-TACE and TARE is currently underway (NCT01381211).

Multi-kinase inhibitors
Tumor cell proliferation, differentiation, and angiogenesis are postulated to be mediated by multiple intracellular and cell surface protein kinases with their downstream pathways, as depicted in Figure 1 24 . Sorafenib, an inhibitor of platelet-derived growth factor receptor (PDGFR), vascular endothelial growth factor receptor (VEGFR), rearrange during transfection (RET), and C-kit, is the first multi-kinase inhibitor proved to be beneficial in unresectable, advanced-stage HCC. The SHARP trial is the first phase III, placebo-controlled trial of the use of sorafenib in HCC.
In this landmark study involving 602 patients with advanced disease naïve to systemic treatment, median OS was significantly improved in the sorafenib group compared to the placebo group (10.7 versus 7.9 months, P <0.001) 25 . Another multi-national randomized controlled trial in the Asia-Pacific region, where chronic hepatitis B is the major risk factor for HCC, also confirmed the findings in the SHARP study 26 . Further analysis of the two randomized controlled trials identified HCV-related HCC, absence of extrahepatic spread, and low neutrophil-tolymphocyte ratio as predictors of greater survival benefit with sorafenib 27 . The remarkable result with sorafenib is considered a major breakthrough in more than 30 years of pursuing a systemic treatment for HCC. However, the therapeutic window is narrow with sorafenib, with restrictions to patients with good performance status and compensated cirrhosis. Also, doselimiting side effects including hand-foot-skin reaction, diarrhea, and weight loss are not uncommon. However, recent studies have shown an association between dermatological adverse effects with sorafenib with better treatment outcomes 28,29 . Over subsequent years, other agents have been studied as alternative first-line treatments against sorafenib or as second-line treatments against placebo in patients intolerant to or who have progressed while on sorafenib. Sunitinib, brivanib, linifanib, everolimus, and tivantinib were unable to meet their respective study end points as either non-inferior or superior to sorafenib or show survival benefit in those who failed sorafenib 30-34 .
Lenvatinib, an inhibitor of epidermal growth factor receptor (EGFR), fibroblast growth factor receptor (FGFR), VEGFR, PDGFR, RET, and C-kit, emerged almost a decade after sorafenib as an alternative first-line treatment for advanced HCC. Non-inferiority to sorafenib in terms of OS (13.6 versus 12.3 months, 95% confidence interval [CI] 0.79-1.06) was demonstrated in the REFLECT trial 35 . In patients who progressed while on sorafenib, regorafenib and cabozantinib prolonged survival in their respective phase III randomized controlled trials (    The recombinant IgG1 monoclonal antibody ramucirumab, which inhibits type 2 VEGFR-mediated angiogenesis, is the latest approved agent for advanced HCC. The initial REACH trial using ramucirumab failed to demonstrate benefit over placebo in patients with BCLC stage B and C disease not amenable to locoregional therapy and treated with first-line sorafenib 38 . However, the effect of ramucirumab was noted to correlate with baseline AFP level, and benefit in OS was seen in the subgroup of patients with baseline AFP concentration above 400 ng/ml. The finding formed the basis of the follow-up REACH-2 trial, in which only sorafenib-treated patients with baseline serum AFP concentration over 400 ng/ml were recruited. Median OS (8.5 versus 7.3 months; HR 0.71, P = 0.020) and progression-free survival (PFS) (2.8 versus 1.6 months; HR 0.45, P <0.001) were significantly improved in the ramucirumab group compared with the placebo group 39 . The study also highlighted that HCC is a heterogeneous disease and that stratification with biomarkers such as AFP is important in the future development of cancer therapy.

Immunotherapy
The past decade has witnessed major breakthroughs in cancer immunotherapy, which has demonstrated benefit in various solid organ and hematological malignancies 40 . HCC develops in an inflammatory milieu, and immune tolerance is reported to play an important role in tumor pathogenesis 41,42 . Immune checkpoint inhibitors (ICPIs) targeting the programmed cell death protein 1/programmed death-ligand 1 (PD-1/PD-L1) pathway and cytotoxic T-lymphocyte-associated antigen 4 (CTLA-4) pathway are the two most studied immunotherapy mechanisms in HCC (Figure 1). The two pathways are believed to act at different stages of the immune response. The PD-1/PD-L1 pathway promotes immune tolerance by suppressing the activity of T cells and mediating the differentiation of regulatory T cells, while the CTLA-4 pathway prevents autoimmunity by inhibiting the proliferation of potential autoreactive T cells 43,44 . By inhibiting these immune checkpoints, self-tolerance to malignant cells is lost, resulting in immune-mediated clearance of tumor tissue.

CTLA-4 inhibitors
CTLA-4 blockage with tremelimumab is the first checkpoint inhibitor being evaluated as a potential treatment for hepatitis C-related HCC. A phase II study found that tremelimumab exhibits both anti-tumor and antiviral activities, with an acceptable safety profile 45 .

PD-1/PD-L1 inhibitors
Nivolumab is a monoclonal antibody inhibiting PD-1 receptor and demonstrated a favorable response rate of 20% in the dose expansion and escalation study (CheckMate 040) 46 . The median OS in sorafenib-experienced patients in this phase II study was 16.7 months with an overall response rate of 14.5%. A similar response rate and improvements in OS, objective response rate, and complete response rate were confirmed in the subsequent head-to-head comparison against sorafenib as first-line therapy in advanced HCC (CheckMate 459), but the primary endpoint of OS did not reach statistical significance 47 .
Pembrolizumab is another anti-PD-1 monoclonal antibody being investigated as a potential treatment for advanced HCC. In the phase II Keynote-224 trial, an objective response was observed in 18 of 104 sorafenib-experienced patients (17%), including one complete and 17 partial responses 48 . A consistent result was seen in the Keynote-240 study, a follow-up phase III randomized controlled trial comparing pembrolizumab with best supportive care as second-line therapy after sorafenib, but the co-primary endpoint of improvement in OS and PFS was not met 49 . A similar phase III study using pembrolizumab in Asian subjects is being conducted (Keynote-394, NCT03062358).

CTLA-4 and PD-1/PD-L1 combination
Studies have also looked into a potential synergistic effect with combined inhibition of PD-1 and CTLA-4 pathways. Tremelimumab together with the anti-PD-L1 monoclonal antibody durvalumab have demonstrated favorable anti-tumor activity with acceptable tolerability in a small pilot study involving patients with advanced HCC or biliary tract carcinomas 50 . A phase III multicenter randomized study (HIMALAYA, NCT03298451) is currently underway testing the dual therapy against sorafenib as first-line treatment in patients with unresectable HCC. The combination of nivolumab and the CTLA-4-inhibiting antibody ipilimumab also showed efficacy with a median OS of 23 months in its dose-finding study 51 .
The side effect profile of ICPIs is diverse and distinctive from conventional cytotoxic chemotherapy. The immunerelated adverse reactions can affect virtually any organ but most commonly involve the endocrine system, skin, gastrointestinal tract, and liver. For grade 1 toxicity, such as mild endocrinopathy, ICPIs can be continued with close monitoring and hormone replacement. Grade 2 or above toxicity usually mandates interruption of ICPIs with or without the need for steroids and immunosuppressants 52 . Compared with patients with melanoma and non-small-cell lung cancer, HCC patients tend to have more elevation of AST/alanine aminotransferase when given ICPIs 53 . It can be difficult to differentiate hepatitis associated with ICPIs from other causes of deranged liver function such as disease progression, reactivation of concurrent viral hepatitis, or decompensation of underlying cirrhosis.

Combination of therapies of different mechanisms
Investigators have brought together locoregional therapy with systemic therapy or systemic therapy of different classes with the hope of augmenting tumor response. The initial experience with adding sorafenib or other molecular targeted therapy to TACE was disappointing 54 . Inadequate dose and duration of sorafenib in these trials were identified as key reasons for failure. The Japanese TACTICS trial eventually demonstrated superiority of TACE plus sorafenib over TACE alone in terms of longer PFS (25.2 versus 13.5 months, P = 0.006) and longer time to TACE untreatable progression (26.7 versus 20.6 months, P = 0.020) 22 . The SORAMIC study looked into the combination of sorafenib and SIRT with Y90 compared with sorafenib alone for advanced HCC. The primary endpoint of OS was not met, but subgroup analysis of non-cirrhotic patients or patients aged under 65 years did show a survival benefit with combination therapy 23 . Future trials should address issues such as timing of sorafenib administration in relation to intra-arterial therapy, patient selection for such combination therapy by balancing potential benefits with incremental adverse effects, and more precise definition of both treatment efficacy and failure.
The combination of atezolizumab, an anti-PD-L1 antibody, and bevacizumab, an anti-VEGF antibody, has been investigated as a first-line systemic therapy in unresectable or metastatic HCC. It is postulated that the anti-angiogenic and immunomodulatory effect of bevacizumab can augment the anti-tumor immune activity of atezolizumab, as shown by a partial response rate of 62% in the phase I trial 55 .
Encouraging results were recently reported from the phase III IMbrave 150 trial comparing the dual therapy with sorafenib, in which statistically significant improvement in both OS and PFS was shown (HR of OS 0.58, P <0.001 and HR of PFS 0.59, P <0.001) 56 .

Conclusion
HCC remains a deadly disease, and unresectable HCC is associated with limited survival. Considerable advancement in local and systemic therapies has been achieved in recent years owing to better understanding of the tumor microenvironment and the interplay between the tumor and the host immune system. With the latest breakthrough achieved with ICPIs, the armamentarium of treatment for HCC is rapidly expanding. However, heterogeneity in patient selection and definition of treatment response in the existing literature make the application of trial results to real-world practice difficult. With the standardization of trial design, prognostic markers of treatment response can be identified, allowing the selection of patients for the most appropriate anti-tumor therapy. Future studies should also continue to utilize therapeutic agents of different classes to achieve synergistic activity while minimizing toxicity.