Review article: pharmacological therapy for hepatocellular carcinoma with sorafenib and other oral agents


Dr R. Moreno-Otero, Unidad de Hepatología (planta 3), Hospital Universitario de La Princesa, C/Diego de León 62, 28006 Madrid, Spain.


Background  Hepatocellular carcinoma (HCC) is the fifth most common malignancy worldwide. Unresectable disease patients have median survival of few months. There is a substantial need for novel treatments for patients with advanced HCC.

Aim  To provide an update review of mechanism of hepatocarcinigenesis and systemic therapies for HCC and the relevant role of Sorafenib in patients with advanced disease.

Methods  A Medline search was performed to identify pertinent original research and review articles. Selected references in these articles were also evaluated.

Results  Systemic chemotherapy for HCC has been quite ineffective. Preclinical studies demonstrated that Raf/MAPK-ERK kinase (MEK)/Extracellular signal regulated kinase (ERK) pathway has a role in HCC. HCC tumours are highly vascularized and vascular endothelial growth factor (VEGF) augments HCC development and metastasis. Sorafenib blocks tumour cell proliferation by targeting Raf/MEK/ERK signalling and exerts an antiangiogenic effect by targeting VEGF receptors-2/3 and platelet derived growth factor receptor β tyrosine kinases.

Conclusions  Currently available therapies are not effective for patients with advanced HCC. Sorafenib has demonstrated for the first time to prolong survival in patients with advanced HCC, and it is the new reference standard for systemic treatment in these patients.


Hepatocellular carcinoma (HCC) is a major health problem, being the fifth most common cancer worldwide, with 626 000 new cases in 2002.1 The incidence of HCC is increasing in Europe and in the US2 and it is a leading cause of death amongst cirrhotic patients.3 Once diagnosis is established, the prognosis of patients will vary according to the evolutionary stage of the disease and the treatment received. The main prognostic factors are related to the tumour status (defined by number and size of nodules, presence of vascular invasion, and extrahepatic spread), liver function [defined by Child-Pugh (CP) class, bilirubin, albumin and portal hypertension] and general health status [defined by Eastern Cooperative Oncology Group performance status (ECOG PS) classification and presence of symptoms].4

Patients with HCC classified in the early stage of the BCLC score have excellent outcomes in terms of both survival (resection, 5-year survival of 89%; percutaneous treatment, 5-years survival of 71%) and recurrence (only 8% at 3-years).5 Patients with HCC at an intermediate stage (multinodular asymptomatic tumours without an invasive pattern) present a median survival without treatment of about 16 months.6 Chemoembolization expands the median survival of these patients to 19–20 months. A majority of HCC patients present advanced stage (with ECOG 1–2 performance status or vascular invasion/extrahepatic spread) and have a median survival of 6 months. For reasons of the poor track record of systemic therapy in HCC, there has been a sense of nihilism for this disease in the oncology and hepatology community for decades. Sorafenib, a molecularly targeted agent, is the first systemic therapy to prolong survival in patients with advanced HCC and the new reference standard for systemic treatment for these patients.

Molecular basis of hepatocarcinogenesis

Signalling pathways for cell proliferation and survival in HCC

Although heterogeneous, certain pathways have been identified repeatedly in patients and HCC cell lines.

RAF/MEK/ERK.  The extracellular-related kinase (ERK) pathway [also known as the mitogen-activated protein kinase (MAPK) pathway] is a mitotic signal transduction cascade that leads to cell division.7 The small protein RAS and the serine/threonine kinase RAF are important molecular signal transducers. Intermediate signalling is mediated by MAPK/ERK kinase (MEK), also known as MAPK kinase. MEK phosphorylates and activates the final downstream signalling molecules ERK1 and ERK2. Blocking ERK signalling has multiple anticancer effects in human HCC cell lines as inhibition of cellular proliferation and induction of cell-cycle arrest, increase in cellular apoptosis and decrease in tumourigenicity.7

Wnt/β-catenin.  Direct stimulation of Wnt receptors is rarely involved in carcinogenesis, but mutations that mimic Wnt stimulation and its downstream signalling have been identified in HCC.8 A member of this signalling pathway, the transcription factor β-catenin, is translocated from the cytoplasm to the nucleus during Wnt signalling.9 Among the target genes transactivated by β-catenin are the oncogenes c-MYC and c-MYB, as well as cyclins that regulate the cell cycle.9, 10 More than 25% of the HCCs have activating mutations in β-catenin.10 These mutations are twice as common (41%) in HCCs associated with chronic HCV infection.8 Mutations in β-catenin appear to be an early event in carcinogenesis in HCC.9 Several molecules have been shown to interfere with this signalling pathway, like specific or unspecific inhibitors of cyclooxigenase (indomethacin, aspirin, celecoxib, rofecoxib) or imatinib and endostatin. There is currently a phase I/II trial investigating the combination of epirubicin with celecoxib in HCC patients (

PI3K.  The phosphatidylinositol-3-kinase (PI3K) signalling pathway act to block cell apoptosis and increase cell survival.11 PI3K is regulated in part by action of phosphatase and tensin homolog (PTEN), the product of the tumour suppressor gene that is mutated in many human cancers, including HCC. Deletion of the PTEN gene or reduced expression of its protein product is found in nearly half of patients with HCC. Decreased PTEN is an independent prognostic factor for worse overall survival (OS) and is significantly correlated with high grade tumours and advanced disease stage. A clinical trial with endothelial growth factor receptor (EGFR) inhibitor erlotinib in patients with advanced HCC led to disease control [complete response (CR), partial response (PR) or stable disease (SD)] in 59% of patients during 4 months (median).12 In another trial, erlotinib was administered in combination with bevacizumab, a specific antibody against vascular endothelial growth factor (VEGF), leading to disease control in 74% of patients.13

P53.  An essential function of the tumour suppressor protein p53 is to protect cells from accumulating genetic damage.14, 15 When p53 becomes inactivated, as in many tumours and precancerous lesions, cells with compromised genome integrity survive inappropriately. Mutations in p53 have been observed in up to 50% of HCC.15

Angiogenesis in HCC progression

Formation of neovessels (angiogenesis) is a process intimately related to HCC. As a matter of fact, HCC is a highly vascularized tumour and this phenomenon is even used for diagnostic purposes. Angiogenesis in HCC is based on the same fundamental principles of activation, proliferation and migration of endothelial cells (ECs) as in other tumours and diseases featuring enhanced angiogenesis.16 Chronic liver diseases are characterized by fibrosis and inflammation. Fibrotic tissue offers resistance to blood flow and to the delivery of oxygen, thus becoming hypoxic.17 Hypoxia and inflammation are the main stimuli for angiogenesis. Hypoxia promotes angiogenesis by signalling through hypoxia inducible transcription factors.18–21

Endothelial budding is facilitated by vasodilatation, loosening of interendothelial contacts and leakage of pre-existing vessels, which allow extravasation of plasma proteins which, together with extracellular matrix (ECM) components, lay down a provisional scaffold for migrating ECs. Nitric oxide (NO), the angiogenic properties of which have been well characterized,22–25 is the main factor responsible for vasodilatation, whereas VEGF increases vascular permeability. Next, the basement membrane (mainly collagen IV and laminin) and the ECM (collagen I and elastin) must be degraded to allow subsequent EC migration and proliferation. EC proliferate in response to growth factors secreted by EC themselves or surrounding cells, including hepatic stellate cells, leucocytes, hepatocytes and Kupffer cells. The most thoroughly characterized molecule is VEGF,26 with the evidence that the VEGF promoter contains hypoxia-inducible factor-responsive elements. VEGF plays a crucial role in virtually all pathological situations in which angiogenesis occurs and the therapeutic use of strategies aimed at blocking its mechanism of action is currently under evaluation. In addition, EC proliferation may be stimulated by other growth factors such as acidic and basic fibroblast growth factors (aFGF and bFGF), hepatocyte growth factor (HGF) and transforming growth factor (TGF).21, 24, 25, 27

Endothelial cell proliferate in an orderly manner that leads to the formation of a lumen.28, 29 Angiopoietin 1 (Ang-1) stabilizes neovessels by binding to the Tie-2 receptor and thereby affects junctional molecules30 and facilitates communication between EC and mural cells;31 however, an excess of Ang-1 makes vessels too tight and inhibits sprouting. Ang-2 may exert opposing effects: in the absence of VEGF, Ang-2 acts as an antagonist of Ang-1, destabilizes vessels and causes EC death, leading to vessel regression; on the contrary, Ang-2 facilitates vessels sprouting in the presence of VEGF.31, 32

The expression of angiogenic factors in HCC samples and nontumoural liver tissue has been evaluated, finding a high expression of Ang-2 and low expression Ang-1 in HCC samples in comparison with nontumoural tissue;33 thus, both molecules may play a key role in progression of HCC via angiogenesis. Tumour angiogenesis is very different from that associated with physiological one.28 During this process, vascular quiescence and stabilization are mediated by Ang-1, Ang-2 and Tie-2 system; therefore, the pathologic state of imbalanced Ang-2/Ang-1 ratio in the presence of VEGF plays a critical role in the transformation of noncancerous tissue to liver cancer by initiating early neovascularization. Evidence exists that tumour development in their very early stage is initiated by host vessels via VEGF, VEGF receptor-2 and Ang-2.29

Systemic therapy of advanced hepatocellular carcinoma

Hormonal therapy

  • Tamoxifen. Oestrogens are involved in stimulating hepatocyte proliferation in vitro and may promote liver tumour growth in vivo.34 The antioestrogen compound tamoxifen has been shown to reduce the level of oestrogen receptors in the liver.35 Barbare et al.36 reported their results of a randomized phase III study. They concluded that tamoxifen did not improve the OS of patients with advanced HCC.

  • Antiandrogen. Antiandrogen therapies have failed to improve survival in randomized studies in patient with advanced HCC.37, 38

  • Octeotride. No survival benefit has been seen in patients with advanced HCC in a phase III study performed by Barbare et al.39 The median OS time was 6.5 months in the octreotide arm and 7.3 months in the placebo arm.


Although large number of controlled and uncontrolled studies has been performed with most classes of chemotherapeutic agents, no single or combination chemotherapy regimen has been found to be particularly effective in HCC.

  • Doxorubicin. It is perhaps the most widely used agent in HCC. Recent large studies have shown a 4–10.5% response rate in HCC patients treated with doxorubicin.40, 41 Significant grade 3 or above haematological and gastrointestinal toxicities were encountered in patients treated with doxorubicin.

  • Nolatrexed. This is a novel thymidylate synthase inhibitor that is not polyglutamated and does not require facilitated transport for uptake into cells. In a large phase III randomized controlled study that was conducted comparing nolatrexed with doxorubicin,41 the median survival times for patients treated with nolatrexed and doxorubicin were only 20.7 and 31 weeks respectively. Patients treated with nolatrexed had more treatment-related toxicities and withdrawal.

  • Other drugs such as cisplatin, 5-fluorouracil, mitoxantrone, etoposide and fludarabine have failed to demonstrate meaningful activity.42–48 Newer chemotherapy drugs, including paclitaxel, irinotecan, gemcitabine, capecitabine, and pegylated liposomal doxorubicin, have been studied in HCC with disappointing results.49–54

Biologic and biochemical therapy

  • Interferon. It has been tested extensively in HCC, but studies have not shown significant beneficial effects.43, 55, 56

  • Thalidomide. Several phase II studies have examined the use of thalidomide either as a single agent or in combination with epirubicin or interferon and have shown limited activity in HCC.57–60

Molecularly targeted therapy

  • Erlotinib. It is an oral EGFR tyrosine kinase inhibitor that has demonstrated the safety and modest efficacy in advanced HCC.12, 61

  • Bevacizumab. An antibody against VEGF that has been evaluated in a phase II trial in patients with unresectable HCC, showing promising safety and efficacy data.62

  • Sunitinib. It is multiple kinase inhibitor. A phase II trial in patients with unresectable HCC was conducted, showing an OS of 45 weeks (median).63 In a second phase II study with sunitinib, a median OS of 10 months has been recently reported.64 For these reasons, a future use for this agent in HCC may be expected.

Treatment with Sorafenib

Sorafenib (Nexavar BAY 43-9006) is a multikinase inhibitor that has shown efficacy against a wide variety of tumours in preclinical models and clinical studies. It has been shown to block tumour cell proliferation and angiogenesis by inhibiting serine/threonine kinases [c-Raf, and mutant and wild-type BRAF (v-raf murine sarcoma viral oncogene homolog B1)] as well as the receptor tyrosine kinases VEGFR2, VEGFR3, platelet derived growth factor receptor (PDGFR), FLT3, Ret and c-kit.65, 66

Chemistry of Sorafenib

Sorafenib [N-(3-trifluoromehyl-4-chlorophenyl)-N′-(4-[2-methylcarbamoyl pyridine-4-yl]oxyphenyl)urea] was synthesized at Bayer Corporation (West Haven, CT, USA).

Mechanism of action of Sorafenib

The mechanisms of action (Figure 1) were analysed in a preclinical study by Liu et al.67 in two HCC cell lines: PLC/PRF/5 (p53 mutant) and HepG2 (p53 wild type).

Figure 1.

 Sorafenib has been shown to inhibit tumour cell proliferation in vitro by targeting the Raf/MEK/ERK signaling pathway at the level of Raf kinase. Sorafenib demonstrated an antiangiogenic effect in vitro by targeting the receptor tyrosine kinases VEGFR-2 and PDGFR and their associated signaling cascades (Adapted from Ref. 67).

Inhibition of proliferation and induction of apoptosis in HCC cell lines.  Sorafenib dose-dependently inhibits cell proliferation with an IC of 6.3 μmol/L in PLC/PRF/5 and 4.5 μmol/L in HepG2 cell. Furthermore, Sorafenib induces DNA fragmentation in both HCC cell lines.

Sorafenib inhibits RAF/MEK/ERK signalling pathway in HCC cell lines.  Raf kinases are key regulators of the MEK/ERK cascade and up-regulated signalling through the RAF/MEK/ERK pathway has an important role in HCC. Sorafenib inhibited MEK and ERK phosphorylation at concentrations of between 3 and 10 μmol/L in PCL/PRF/5 cells and between 1 and 3 μmol/L in Hep2G2 cells by down-regulating cyclin D1 in both cell lines, consistent with the known inhibitory activity of Sorafenib against Raf kinase isoforms.67

Sorafenib reduces phosphorylation of eIF4E and down regulates Mcl-1 levels in HCC cells, independently of MEK/ERK signalling.  Mcl-1, an antiapoptotic member of the Bcl-2 family, is an important factor for apoptosis resistance in HCC. Sorafenib has been reported to induce apoptosis in human tumour cell lines through the inhibition of translation and down-regulation of Mcl-1. In a recent study, Mcl-1 protein levels and the phosphorylation state of eIF4E were determined to understand if Mcl-1 could play a role in the mechanism of Sorafenib-induced apoptosis in HCC cells; Sorafenib reduced the level of phosphor-eIF4E after 2 h of treatment at concentrations of 1 and 10 μmol/L in both HCC cell lines.67

The in vivo efficacy and mechanism of action of Sorafenib against PLC/PRF/5 human HCC tumour xenografts.  Sorafenib tosylate produces dose-dependent growth inhibition of subcutaneous implanted PCL/PRF/5 tumour xenografts in mice. Dose levels of 10 and 30 mg/kg produced significant and dose-dependent tumour growth inhibitions of 49% and 78%, respectively. The effect at 30 mg/kg of Sorafenib tosylate was confirmed in an independent experiment and the dose response was further evaluated at a dose level of 100 mg/kg. Sorafenib tosylate produced durable partial tumour regressions in 50% of the mice at the 100 mg/kg dose level.67

Sorafenib inhibits tumour angiogenesis.  Sorafenib tosylate at dose levels of 30 and 100 mg/kg reduced the percentage of tumour microvessels area [(area of CD34-positive objects/measured tissue area) × 100] in mouse xenografts tumour. CD34 was used as a specific EC marker. Sorafenib exerts its antiangiogenic effect by targeting VEGFR -2/-3 and PDGFR β tyrosine kinases.67

Pharmacokinetics of Sorafenib

The pharmacokinetics of Sorafenib was evaluated in a phase I trial in patients with advanced refractory solid tumours.68 Pharmacokinetic (PK) parameters were assessed in 23 patients. Three different dosing levels of Sorafenib were tested: cohort 1, 100 mg (50 mg tablets) b.d.; cohort 2, 200 mg (50 mg tablets) b.d.; cohort 3A, 400 mg (50 mg tablets) b.d. and cohort 3B, 400 mg (200 mg tablets) b.d. Maximum concentration (Cmax) and area under the curve (AUC; 0–8) were measured in each group. Cmax and AUC were 1.48 and 7.27, 3.29 and 15.2, 8.39 and 49.6 and 2.92 and 17.2 in cohorts 1, 2, 3A and 3B respectively.

Thereafter, the pharmacokinetics of Sorafenib were evaluated in phase II study in patients with impaired liver function and measurable, histologically proven, inoperable HCC who had not received previous systemic treatments for HCC.66 This study included 147 patients (CP A or B classification) who received Sorafenib 400 mg b.d. continuous in 4-week cycles, but PK parameters were assessed in only 22 patients. Inclusion criteria included ECOG PS of 0 or 1; CP score of A or B; life expectancy of at least 12 weeks; elevated alphafetoprotein (AFP) level, and adequate haematological, hepatic and renal function. Exclusion criteria included patients with tumour of mixed histology or fibrolamellar variant, pregnant or lactating women or those requiring systemic anticancer therapy or those receiving biologic-response modifiers or CYP34A inhibitors or those with medical/psychological/social problems that might affect study participation or evaluation. Blood samples were collected on day 1 of cycle 2 at 0 h (predose) and 0.5, 1, 2, 4, 8 and 12 h postdose for PK analysis of Sorafenib plasma concentrations. Parameters included AUC, Cmax and time to maximum concentration (tmax).There was some variability in AUC and Cmax values, which were slightly greater in CP B than in CP A patient groups but these differences were not considered significant. The AUC, Cmax and tmax were 25.4 mg h/L, 4.9 mg/L and 1 h respectively in the CP A group and 30.3 mg h/L, 6 mg/L and 0.5 h in the CP B group. Importantly, it is unlikely that dose adjustment would be necessary when administering Sorafenib to patients with mild (CP A) or moderate (CP B) hepatic insufficiency. The dose response data from the Phase II trial confirmed that the dose of Sorafenib of 400 mg (200 mg tablets) b.d. maximizes its antitumoural effect.

Clinical efficacy of Sorafenib

Clinical efficacy of Sorafenib administered individually was assessed in the phase II trial in patients with measurable, histologically proven, inoperable HCC who had not received prior systemic treatments for HCC.66 Patients received continuous oral Sorafenib 400 mg b.d. in 4-week cycles. Tumour response was assessed every two cycles using modified World Health Organization criteria by bidimensional tumour measurements every 8 weeks. Throughout the study,66 lesions measured at baseline were evaluated using the same technique. Overall tumour response was scored as CR, PR, or minor response (MR; a reduction in tumour size of <25% but <50% vs. baseline). Dose delays or modifications were required for drug-related toxicities. For grade 3/4 toxicities, patients received lower doses when toxicity improved to grade 2 or better, but therapy was discontinued if recovery time was 3 weeks or longer. A dose delay was introduced for grade 3 nonhaematological toxicities, until toxicity was grade 2 or better; patients were then treated at one dose level lower and therapy was discontinued if recovery time was 3 weeks or longer. Patients with drug-related nonhaematological toxicities were treated at two dose levels lower at the first appearance and withdrawn at the second. A modified scale was used for hand-foot skin reaction (HFSR), to facilitate interpretation and specific dose modifications were implemented.

In this study, 137 patients were included.66 Median study duration was 3.4 months, median number of treatment cycles was four and 72% of patients received six of fewer treatment cycles. Independently assessed responses were as follows: three patients (2%) achieved PR, eight (6%) had MR and 46 (33.6%) had SD (>16 weeks). The time to progression (TTP) was 5.5 months and median OS was 9.2 months.

In a large multicentre, randomized, placebo-controlled phase III trial (SHARP)69 the efficacy of Sorafenib vs. placebo in patients with HCC was evaluated. Patients with advanced measurable HCC, no prior systemic treatment, ECOG PS 0-2 and CP A received Sorafenib 400 mg b.d. Primary efficacy endpoints were OS and time to symptomatic progression (TTSP). TTP and disease control rate (DCR) were secondary endpoints. In this study, 602 patients were randomized. Interestingly, when the second interim analysis of the trial was carried out in October 2006, it was observed that the efficacy results in the Sorafenib arm were much better than those in the placebo arm. The independent committee monitoring the trial recommended that it was interrupted and since February 2007, all placebo patients started treatment with Sorafenib.

Intention-to-treat analysis showed that the median OS was 10.7 vs. 7.9 months (Sorafenib vs. placebo). Hazard ratio was 0.69, with a P-value of 0.00058. Primary TTSP analysis demonstrated no statistically significant difference for Sorafenib vs. placebo. An independent central revision showed that the median TTP was longer for Sorafenib than for placebo (5.5 vs. 2.8 months respectively; hazard ratio, 0.58; = 0.000007) and DCR was higher (43% vs. 32%) with Sorafenib vs. placebo. Evaluation of the response (independent revision with the RECIST system) evidenced a PR in 2.3% of Sorafenib-treated patients, compared with 0.7% of placebo treated patients, as well as a progression-free rate of 62% in the Sorafenib group vs. 42% in the placebo group. The subgroup analysis showed in all cases a benefit of Sorafenib vs. placebo (ECOG PS 0 vs. 1 or 2; with vs. without extrahepatic distribution; with vs. without macroscopic vascular invasion; and combination of the two).69

Safety and tolerability

Sorafenib has been demonstrated to be safe. In a phase II study of Sorafenib in patients with advanced HCC,66 the most common drug related adverse events reported (any grade) were dermatological HFSR in 31%, rash in 7%, and alopecia in 10%; other adverse events were fatigue in 30% and gastrointestinal (diarrhoea 43%, nausea 16% and anorexia 14%). Relatively infrequent dose-limiting toxicities were observed. Notable grade 3/4 adverse events included fatigue 9%, diarrhoea 8% and HFSR (5%). In this study, one death secondary to an intracranial haemorrhage was reported, but it remains unclear whether it was drug-related or not.66

In the SHARP trial (randomized placebo controlled trial in patients with advanced HCC), the incidence of serious adverse events was similar for Sorafenib vs. placebo (52% vs. 54%).69 The most frequent grade 3/4 events were diarrhoea (11% vs. 2%), HFSR (8% vs. 1%), fatigue (10% vs. 15%) and bleeding (6% vs. 9%) for Sorafenib vs. placebo.69


Hepatocellular carcinoma is the third most frequent cause of death from cancer worldwide. Current therapy for early disease using liver transplantation, surgical resection or local ablation is now often curative, but a majority of HCC patients present with advanced HCC, when the currently available therapies do not lead to long-term cure. Sorafenib is a multi tyrosine kinase inhibitor with antiangiogenic and antiproliferative effects that have demonstrated unquestionable survival benefits for advanced cases. The results of the SHARP trial represent a breakthrough in the management of this complex disease. Sorafenib is the first systemic therapy to prolong survival in HCC and the new reference standard for systemic treatment in HCC patients.


Declaration of personal interests: None declared. Declaration of funding interests: The writing of this paper was funded by CIBEREHD, funded by the Instituto de Salud Carlos III, Madrid, Spain.