Molecular Targeted Therapies in Hepatocellular Carcinoma

Genome-wide alterations

Several candidate genes in hepatocarcinogenesis have emerged: c-myc (8q), Cyclin A2 (4q), Cyclin D1 (11q), Rb1 (13q), AXIN1 (16p), p53 (17p), IGFR-II/M6PR (6q), p16 (9p), E-Cadherin (16q), SOCS (16p) and PTEN (10q). Chromosomal amplification or deletions are identified in almost all tumors, being the most prevalent amplifications of 1q (58-78%), 6p, 8q, 17q, and 20q, and deletions in 4q, 8p, 13q, 16q and 17p^12,20. High-level amplifications have been detected in 11q13 (5% of cases, regions encoding Cyclin D1) and 6p21 (4-6%; regions encoding VEGFA)²⁰. Further research is needed to identify key oncogenes in these areas.

Mutations

Few somatic mutations have been described in HCC patients, an area that is expected to advance in the near future with the introduction of high-throughput sequencing technology¹⁴. TP53, a tumor suppressor gene involved in cell cycle control, DNA repair, apoptosis and differentiation, is mutated in 27% of cases (range 0-67%). Aflatoxin B exposure in Africa and Asia is associated with p53 G-to-T mutation at the third position of codon 249. Mutations in beta-catenin in exon 3 are present in 17% of cases of HCC, and in 60% of cases of hepatoblastoma. Less frequent somatic mutations have been described in AXIN1, PI3KA, and K-Ras. Conversely, prevalent mutations in other cancers, such as, EGFR, Her2/nu PTEN or H-Ras are marginal in HCC, while germline mutations have not been described.

Genomic instability

This has been described associated with telomere shortening, aberrant methylation and aberrations in mismatch repair genes. Telomerase activity is increased in nearly 90% of human HCCs and results from HBV integration in the TERT locus, amplification of the gene encoding the telomerase RNA component (TERC), or allelic loss of chromosome 10p, a region encoding a telomerase repressor²⁵. High genomic instability has been associated with the proliferation subclass of HCC, and is more prevalent in HBV-related HCCs.

EGFR-Ras-MAPKK pathway

The ligands EGF, HGF PDGF, VEGF among others activate the RAS/MAPK signaling pathway and induce transcription of genes of the AP1 family, such as c-fos and c-jun, which are key elements for cell proliferation²⁷. The epidermal growth factor receptor (EGFR) is a member of a family of four related receptors (Her2/Neu, ErbB3 and ErbB4) that upon ligand binding trigger tyrosine kinase activity and consequently initiate signal transduction. Gain of function of EGFR classically occurs as a result of point mutations, amplification or increase in ligand-receptor interaction²⁸.

Most of the agents under investigation block membranous tyrosine kinase receptors (TKRs) (Figure 2). Effective blockade of the EGFR signaling pathway can be achieved by the use of monoclonal antibodies against EGFR (cetuximab) or ErbB2/Her2/neu (trastuzumab). Cetuximab is FDA-approved for the treatment of head and neck²⁹ and colo-rectal cancer³⁰, and trastuzumab for overexpressing-Her2 metastatic breast cancer⁸. Alternatively, pathway activation can also be successfully inhibited by small molecules against the catalytic domain of EGFR like erlotinib, active in advanced stages of non-small cell lung cancer¹⁰, or against EGFR and Her2, such as lapatinib, which is effective in the treatment of breast cancer³¹.

In HCC, Ras/MAPK pathway activation might result from aberrant upstream signals (EGFR signaling, IGF signaling) or inactivation of tumor suppressor genes by aberrant methylation, as NORE1A³². Mutations of Raf and Ras are rare findings in HCC. Potent drugs blocking Ras/MAPK signaling are still at exploratory phase, except for sorafenib that has activity inhibiting B-Raf at nanomolar concentrations³³.

c-MET signaling

Hepatocyte growth factor (HGF) is a critical molecule for hepatocyte regeneration after injury³⁴. It is secreted by stellate cells and binds to the c-MET receptor. Aberrant activity of MET has been described in human cancers as a result of MET amplification, germline or somatic mutations, transcriptional up-regulation or HGF-dependent autocrine loops. Dysregulation of c-MET and HGF are common in HCC^14,35, although the exact role of this pathway in the pathogenesis of HCC is not established. Several compounds have been developed targeting MET pathway, including antibodies against HGF or MET receptor, or selective small molecule MET inhibitors³⁶, but none of them are yet in advanced stage of research in HCC.

IGF signaling

The insulin-like growth factor (IGF) family has a major role in fetal development, proliferation, differentiation, cell growth, apoptosis and has been implicated in multiple malignancies³⁷. IGF I has been associated with high risk of neoplasm development, but not in HCC. In fact, experimental studies suggest the potential role of this molecule in improving cirrhosis. IGF-II is overexpressed in around 30% of human HCCs and the tumor suppressors IGFBP-1, -3, -4 are downregulated³⁸. More important, the tumor suppressor IGFR-II, which main action consists in binding and degrading IGF-II, is downregulated in a subgroup of HCCs as a result of chromosomal detections in IGFR-II locus (6q) or point mutations. Several monoclonal antibodies and small molecules blocking IGF-1R are under early clinical investigations (A12, XL228) ³⁹. The interaction between IGF and EGF signaling may act as a mechanism of resistance for some tumors. In fact the blockade of IGF-1R confers sensitivity to breast tumors resistant to trastuzumab.

PI3K/Akt/mTOR pathway

The PI3K/Akt/mTOR pathway plays a critical role in carcinogenesis⁴⁰. Akt can be activated through a Tyrosine kinase receptor (EGF or IGF signaling) or through constitutive activation of PI3K or loss of function of the tumor suppressor gene PTEN by epigenetic silencing or somatic mutations. Mutations of PTEN might confer resistance to small molecules against EGFR. Although the role of pAkt in HCC needs further investigation, recent studies have suggested a worse prognosis for tumors with activated Akt⁴¹.

An important mediator of the PI3K-Akt pathway is mTOR, which acts as a central regulator of cell growth and proliferation, sensing nutritional status and allowing progression from G1 to S phase⁴⁰. mTOR pathway is activated in a subset of HCCs, and its blockade with rapamycin or everolimus inhibits growth in HCC cell lines, and in experimental models⁴². Everolimus – a rapamycin analog- and termsirolimus are approved for the treatment of renal cancer. Novel compounds are currently being tested in early clinical trials. These molecules (rapamycin and analogues) are already approved as immunosuppressive treatments after liver transplantation.

Wnt-βcatenin pathway

The Wnt-βcatenin pathway is implicated in colon cancer and HCC. Unfortunately it constitutes the most undruggable pathway⁴³. The signaling cascade is initiated extracellularly, when Wnt ligands stimulate the Frizzled receptors, which signals β-catenin to uncouple from E-cadherin. The activation of Wnt pathway induces translocation of β-catenin into the nucleus, where it regulates specific oncogenes, including c-myc, cyclin D, and survivin.

Activation of Wnt cascade has been shown in one third of HCCs, particularly in HCV-induced HCCs. Mutations in β-catenin occurs in around 17% of cases (ranging from 0% to 44%)^14,44 (see Table 1); and association between Wnt pathway activation and the nuclear sequestration of β-Catenin has been described by immunohistochemistry²⁰. Wnt canonical pathways can also be activated by aberrant methylation of the tumor supressors as APC and E-cadherin or by increase of autocrine/paracrine secretion of Wnt ligands. Non-canonical activation of Wnt in HCC deserves further research. New drugs targeting this pathway are under early clinical development.

Anti-EGFR agents

The types of molecules targeting EGFR and Her2/nu are detailed in Table 2. Three tyrosine kinase inhibitors targeting EGFR have been tested in HCC: erlotinib and gefitinib targeting EGFR and lapatinib targeting both EGFR and Her2/nu. Cetuximab, a monoclonal antibody against EGFR has also been assessed.

Erlotinib has shown activity both in pre-clinical and clinical studies. The first report testing erlotinib at 150 mg daily in HCC included 38 patients with intermediate/advanced HCC (39% of which with extrahepatic metastases) described a low response rate (9%) and 6-month progression-free-survival in 32% of patients (12 out of 38) ⁶⁴. The median survival was of 13 months, which can be explained by the action of the drug, but also because the target population was different than the conventional advanced HCC population (42% of patients without underlying liver disease). In a second study including 40 patients, the median overall survival was of 25 weeks (95%CI, 18-42 weeks) ⁶⁵. Gefitinib prevented HCC development in experimental models, but a study in 31 patients reported a median survival of 6.5 months and the authors concluded that the drug had no activity⁶⁶. Although Her2/neu overexpression and EGFR mutations are uncommon events in HCC a dual receptor blockade lapatinib (EGFR and Her2) is being tested in experimental models of HCC and early clinical trials.

Pre-clinical studies reported antiproliferative effects and some pro-apoptotic activity of cetuximab, a chimeric monoclonal antibody against EGFR, in cell lines. Two studies have tested cetuximab as a single agent. None of them showed any objective responses^67,68 and while reported a median TTP of 2 months in 32 patients, the second described a median survival of 9.6 months. This conflicting data indicate that the true efficacy of this compound remains to be elucidated.

Inhibitors of angiogenesis

HCC is a notoriously hypervascular malignancy, even at very early stages of the disease. Overexpression of pro-angiogenic factors like VEGF, PDGF and angiopoietin 2 has been demonstrated in HCC^11-14. Experimental studies demonstrate that targeting angiogenesis is of major relevance in HCC. The main compounds tested in clinical trials are monoclonal antibodies (bevacizumab) and of small molecules (sunitinib, sorafenib, bivranib, vatalanib, cediranib).

Bevacizumab is a humanized monoclonal antibody against VEGF approved for the treatment of breast cancer and liver metastasis of colorectal cancer. The mechanism of action of this compound is controversial, since it might act neutralizing VEGF but also normalizing the vasculature to recover the normal perfusion of tumors with aberrant vascularization. A phase I/II trial in 33 selected patients without portal vein invasion revealed modest antitumoral activity (10% objective response) with a median time to progression of 6.5 months⁶⁹. Of note is that 2 patients developed major gastrointestinal bleeding, one of them leading to death. In a French study including 30 patients, 3 patients also presented bleeding complications⁷⁰. Combinations of bevacizumab with chemotherapy (gemcitabine, oxaliplatin, capecitabine), obtain objective responses of 10-20% with median survivals of 9-10 months^71,72. A study in 27 patients assessing combination of bevacizumab with erlotinib reported a surprisingly high median survival of 19 months⁷³. Six patients showed objective responses (20%). These interesting data needs to be confirmed within extensive randomized phase II studies, because it might reflect the synergistic activity of the combination, but also a selection bias due to the recruitment of patients in the ill-defined category of unresectable HCC, or short follow-up.

Two phase II studies have examined the safety and efficacy of the multikinase inhibitor sunitinib in advanced HCC^74,75, a drug already approved for renal cell carcinoma and gastrointestinal stromal tumors (GIST). Sunitinib inhibits VEGFR1 and 2, PDGFR, c-Kit and FLt3, among other kinases. The first European/Asian study treated 37 patients with a daily dose of 50mg/day and reported 1 partial response and 13 showed stable disease⁷⁴. Of note, however, was that four patients suffered from treatment-related adverse events leading to death due to gastrointestinal bleeding, encephalopathy and hepatorenal syndrome. The US study treated 26 patients at 37.5 mg/day and showed a median progression-free survival of 4.5 months, overall survival of 11.6 months, and more manageable side effects, although 1 death due to bleeding was reported⁷⁵. The worse safety profile identified with this drug in HCC compared with other tumors –renal, GIST- points to the need of being cautious in the management of very potent anti-angiogenic agents in cirrhotic patients. Three anti-angiogenic and multikinase agents, brivanib, cediranib and vatalanib are being tested currently and data are eagerly awaited.

PI3K/Akt/mTOR inhibitors

Approximately 50% of patients with HCC have activation of the mTOR pathway as assessed by immunohistochemical analysis of phosphorylated S6⁴². This activation may be the result of increase signaling due to overexpression of ligands (i.e. EGF, IGF1, IGF2) or may be due to mutations in oncogenes (PI3KCA) or tumor suppressor genes (PTEN). Some of these patients have a simultaneous activation of Akt. Rapamycin is an inhibitor of mTOR activity approved as immunosuppressant in liver transplantation. A tempting strategy is to use it as first line anti-rejection therapy in the setting of liver transplantation for HCC, but results of randomized studies are needed to support this strategy. Preclinical studies with other analogues of rapamycin (i.e. everolimus, temsirolimus) have shown activity as single agents in xenograft models ⁴². and phase II studies are currently testing these compounds alone or in combination with sorafenib.

Other molecular agents

IGF signaling activation has been implicated in the pathogenesis of HCC³⁸. Drugs blocking the receptor IGFR1, such as the monoclonal antibody A12, decrease tumor growth and improve survival of HCC xenografts³⁹, providing the rationale to test them in clinical trials. Studies with small molecules and monoclonal antibodies against IGFR1 are ongoing. Similarly, there is rationale to test c-MET inhibitors.

Wnt pathway is activated in at least 30% of HCC^20,22 but, unfortunately, there are not yet drugs available that effectively block its activation without significant side effects. Molecular targets of this pathway include Wnt ligands, Frizzled receptors and the oncogene beta-catenin. Preclinical studies have shown activity with different compounds that are currently tested. Similarly, there are active drugs inhibiting the proteasome activation, but the results of bortezomib, which is approved for multiple myeloma, are not encouraging. Telomerase, thought to be essential in cancer cell immortality, is a potential target in HCC²⁵; there are some ongoing studies applying TERT immunization. Finally, trials with drugs blocking Hedgehog pathway and HDAC inhibitors are expected to be tested in the near future.

Rationale for combination therapies

Combination of molecular therapies is expected to improve the outcome benefits already obtained with sorafenib, but this is highly complex matter (see review Dancey⁷⁶). There is rationale for blocking complementary pathways activated in HCC. This is the case with anti-antiangiogenic agents and blockers of cell proliferation, such as EGFR, MET and IGFR inhibitors. An alternative strategy is to combine therapies abrogating complementary intracellular signaling, such as RAS or MTOR inhibitors with cell proliferation inhibitors. Similarly, pro-apoptotic agents might synergize with cell proliferation inhibitors.

Biomarkers of response and resistance to molecular therapies are needed to provide further rationale for combination therapies. Several mechanism of resistance described in other solid tumors might also apply to HCC. For instance, resistance to EGFR inhibitors is mediated by mutations in downstream oncogenes (RAS) or tumor suppressors (PTEN) or focal amplifications of MET. Similarly, resistance to MTOR inhibitors might be mediated by a negative loop activating IGFR signaling. Understanding these mechanisms will guide drug combinations. Despite some combinations are scientifically sound, toxicity remains as the main practical limitation and safety data from phase I/II studies is mandatory prior launching phase III initiatives⁵³.