Sorafenib suppresses postsurgical recurrence and metastasis of hepatocellular carcinoma in an orthotopic mouse model

Authors

  • Yu-Xiong Feng,

    1. Center for Cancer Research, Institute for Nutritional Sciences, Shanghai Institutes for Biological Sciences, Chinese Academy of Sciences, Shanghai, China
    Search for more papers by this author
    • These authors contributed equally to this work.

  • Tao Wang,

    1. Eastern Hepatobiliary Surgery Hospital, Second Military Medical University, Shanghai, China
    Search for more papers by this author
    • Potential conflict of interest: Nothing to report.

  • Yue-Zhen Deng,

    1. Center for Cancer Research, Institute for Nutritional Sciences, Shanghai Institutes for Biological Sciences, Chinese Academy of Sciences, Shanghai, China
    Search for more papers by this author
    • Potential conflict of interest: Nothing to report.

  • Pengyuan Yang,

    1. Department of Pharmacology and Cancer Biology, Duke University Medical Center, Durham, NC
    Search for more papers by this author
  • Jing-Jing Li,

    1. Center for Cancer Research, Institute for Nutritional Sciences, Shanghai Institutes for Biological Sciences, Chinese Academy of Sciences, Shanghai, China
    Search for more papers by this author
  • Dong-Xian Guan,

    1. Center for Cancer Research, Institute for Nutritional Sciences, Shanghai Institutes for Biological Sciences, Chinese Academy of Sciences, Shanghai, China
    Search for more papers by this author
  • Fan Yao,

    1. Center for Cancer Research, Institute for Nutritional Sciences, Shanghai Institutes for Biological Sciences, Chinese Academy of Sciences, Shanghai, China
    Search for more papers by this author
  • Yin-Qiu Zhu,

    1. Center for Cancer Research, Institute for Nutritional Sciences, Shanghai Institutes for Biological Sciences, Chinese Academy of Sciences, Shanghai, China
    Search for more papers by this author
  • Ying Qin,

    1. Center for Cancer Research, Institute for Nutritional Sciences, Shanghai Institutes for Biological Sciences, Chinese Academy of Sciences, Shanghai, China
    Search for more papers by this author
  • Hui Wang,

    1. Center for Cancer Research, Institute for Nutritional Sciences, Shanghai Institutes for Biological Sciences, Chinese Academy of Sciences, Shanghai, China
    Search for more papers by this author
  • Nan Li,

    1. Eastern Hepatobiliary Surgery Hospital, Second Military Medical University, Shanghai, China
    Search for more papers by this author
  • Meng-Chao Wu,

    1. Eastern Hepatobiliary Surgery Hospital, Second Military Medical University, Shanghai, China
    Search for more papers by this author
  • Hong-Yang Wang,

    1. Eastern Hepatobiliary Surgery Hospital, Second Military Medical University, Shanghai, China
    Search for more papers by this author
  • Xiao-Fan Wang,

    1. Department of Pharmacology and Cancer Biology, Duke University Medical Center, Durham, NC
    Search for more papers by this author
  • Shu-Qun Cheng,

    Corresponding author
    1. Eastern Hepatobiliary Surgery Hospital, Second Military Medical University, Shanghai, China
    • Eastern Hepatobiliary Surgery Hospital, Second Military Medical University, 225 Changhai Road, Shanghai 200438, China
    Search for more papers by this author
    • fax: 86-21-65562400

  • Dong Xie

    Corresponding author
    1. Center for Cancer Research, Institute for Nutritional Sciences, Shanghai Institutes for Biological Sciences, Chinese Academy of Sciences, Shanghai, China
    2. Key Laboratory of Nutrition and Metabolism,Institute for Nutritional Science, Shanghai Institutes for Biological Sciences,Chinese Academy of Sciences, Shanghai, China
    • Laboratory of Molecular Oncology, Institute for Nutritional Sciences, Shanghai Institutes for Biological Sciences, Chinese Academy of Sciences, 294 Tai-Yuan Road, Shanghai 200031, China
    Search for more papers by this author
    • fax: (86)-21-54920291


Abstract

Surgical resection is the first-line treatment for hepatocellular carcinoma (HCC) patients with well-preserved liver function. Nevertheless, the rate of postoperative recurrence at 5 years is as high as 70%, and this gravely jeopardizes the therapeutic outcome. Clearly, new approaches are needed for preventing the relapse of this deadly disease. Taking advantage of a luciferase-labeled orthotopic xenograft model of HCC, we examined the role of sorafenib, the first systemic drug approved for advanced HCC patients, in the prevention of HCC recurrence. We found that sorafenib suppressed the development of postsurgical intrahepatic recurrence and abdominal metastasis and consequently led to prolonged postoperative survival of mice in this model. Furthermore, hyperactivity of extracellular signal-regulated kinase signaling caused by elevated levels of growth factors associated with postoperative liver regeneration enhanced the sensitivity of HCC cells to sorafenib; this provides a plausible explanation for the observation that recurrent tumors are more responsive to growth inhibition by sorafenib. Conclusion: Our results strongly suggest that by effectively reducing postoperative recurrence, sorafenib has a potential application in early-stage HCC patients who have undergone hepatectomy with curative intention. (HEPATOLOGY 2011;53:483-492)

Hepatocellular carcinoma (HCC) is one of the major health problems worldwide and ranks as the third-leading cause of neoplasm-related death in the world.1 The incidence of HCC is highest in Asia and sub-Saharan Africa, where chronic hepatitis B virus infection is the predominant risk factor. The frequency has also been increasing in the Western world, and this is especially associated with hepatitis C virus infection.2 The incidence of HCC in the United States and Europe has been projected to equal the present frequency in Japan in 2 decades.3

At present, the percentage of patients suitable for surgical treatment alone is low, even in developed countries.2 However, with the implementation of better surveillance programs and diagnostic tests, the proportion of patients diagnosed with early-stage HCC is increasing, and this is making surgical intervention increasingly important in HCC treatment.4, 5 Unfortunately, postsurgical recurrence of HCC often occurs, and most relapses happen fairly soon after hepatectomy and severely threaten patients' long-term overall survival.6-8 Worse still, most of the adjuvant randomized controlled trials of therapies to prevent HCC relapse have failed, and the few successful ones were not sufficiently robust to inspire their clinical use.5 Therefore, it is tremendously important to explore effectual approaches to preventing the recurrence of HCC after surgical resection.

As a chemotherapeutic agent, sorafenib is the first and only systemic therapy to significantly prolong the survival of HCC patients with advanced-stage disease. Thus, it is currently used as a standard treatment for patients with advanced HCC.9 By targeting multiple cellular tyrosine kinases and serine/threonine kinases, such as vascular endothelial growth factor receptors and B-Raf (v-raf murine sarcoma viral oncogene homolog B1), sorafenib strongly inhibits cancer cell proliferation and angiogenesis and induces apoptosis.10 Apart from its successful application in patients with advanced, unresectable HCC, however, the use of sorafenib in patients with early-stage HCC is largely untested; this is especially true for patients who are considered appropriate candidates for curative intervention. Using a luciferase-labeled orthotopic model of HCC, we found in this study that sorafenib was able to significantly suppress the recurrence of xenograft tumor growth after surgical resection (including intrahepatic relapse) as well as abdominal metastasis. Mechanistically, this effect of sorafenib may be explained at least in part by the targeted inhibition of the hyperactivated extracellular signal-regulated kinase (ERK) pathway, which is associated with postoperative liver regeneration. Taken together, our results strongly suggest that the use of sorafenib after surgical resection for early-stage HCC is a promising approach for preventing recurrence and improving postoperative outcomes.

Abbreviations

bFGF, basic fibroblast growth factor; EGF, epidermal growth factor; ELISA, enzyme-linked immunosorbent assay; ERK, extracellular signal-regulated kinase; HCC, hepatocellular carcinoma; HGF, hepatocyte growth factor; MMP9, matrix metalloproteinase 9; MTT, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide; pERK, phosphorylated extracellular signal-regulated kinase; PVTT, portal vein tumor thrombus; TGF-α, transforming growth factor α.

Patients and Methods

Reagents and Cell Lines.

Sorafenib tablets used in animal experiments were acquired from Bayer Corp. and were prepared according to a previous report.10 The sorafenib that was used in cell culture treatment was obtained from LC Laboratories and was dissolved in dimethyl sulfoxide before use. Luciferin powder was acquired from Xenogen Corp. Mouse anti-human phosphorylated extracellular signal-regulated kinase 1/2 (pERK1/2), ERK1, and tubulin antibodies were procured from Cell Signal Technology. Epidermal growth factor (EGF), basic fibroblast growth factor (bFGF), and hepatocyte growth factor (HGF) were obtained from BD Biotechnology. Enzyme-linked immunosorbent assay (ELISA) kits for the detection of EGF and HGF were acquired from R&D systems. Hep3B cells were a gift from Dr. Hui Wang (Shanghai Institutes for Biological Sciences, Chinese Academy of Sciences, Shanghai, China); the 7404 cells were obtained from the Cell Bank of the Shanghai Institutes for Biological Sciences (Chinese Academy of Sciences, Shanghai, China). Both Hep3B and 7404 cells were derived from the cultures of liver cancer specimens from hepatitis B surface antigen–positive patients. The portal vein tumor thrombus 1 (PVTT-1) cell line was prepared from an in vitro culture of a PVTT-1 xenograft that was established in our laboratory.11

Animals.

BALB/c male nude mice were housed under standard conditions. The animal protocols were performed in agreement with SIBS Guide for the Care and Use of Laboratory Animals and were approved by the animal care and use committee of the Shanghai Institutes for Biological Sciences.

Patients.

Serum samples of 42 HCC patients were obtained from Eastern Hepatobiliary Surgery Hospital during 2007-2008 after their written, informed consent was obtained. All the patients had early HCCs with tumor nodules less than 5 cm and underwent hepatectomy after their HCC diagnosis. After their discharge from the hospital, all patients were closely followed up. When an intrahepatic recurrence was suspected because of ultrasound or computed tomography, percutaneous liver biopsy with fine-needle aspiration was performed under ultrasound guidance for histological confirmation of the diagnosis.

Establishment of a Luciferase-Labeled Orthotopic HCC Model.

Fresh HCC samples were used in the establishment of our orthotopic HCC model. We obtained PVTT specimens from clinic hepatectomies and orthotopically implanted them into the livers of nude mice. Several samples were successfully grown in the nude mice and were used as successive transplantable tumor lines. One of the xenograft lines, named PVTT-1, was able to grow in cultures, and the cells were infected with a lentiviral luciferase reporter gene. One day after infection, the primary cells were harvested and injected into a pre-implanted subcutaneous PVTT-1 tumor in nude mice. Three weeks after the injection, we isolated luciferase-positive cells of the subcutaneous tumor and orthotopically grafted them into the livers of nude mice to establish a luciferase-labeled, orthotopic HCC model. The histology of the established tumor model is shown in Supporting Information Fig. 1.

Surgical Resection of Xenografted Tumors (Partial Hepatectomy).

Ten days after the orthotopic implantation of 2-mm-diameter tumors, mice bearing luciferase-labeled PVTT-1 tumors were anesthetized for tumor resection. An incision was made in the middle upper abdomen, and the portion of the liver bearing the tumor was clearly visualized. The lobe implanted with the tumor was excised, and the distance between the excisional margin and the tumor edge was always greater than 3 mm. Mice with severe peritoneal adhesion of tumors to other organs were removed from the experimental group without undergoing surgical resection of the tumors. The successful removal of primary tumors was documented by postoperative luciferase analysis with the Xenogen IVIS Lumina system according to the manufacturer's protocol.

Detection of Tumor Growth and Recurrence.

In vivo tumor growth, both primary and recurrent, was measured twice a week with the Xenogen IVIS Lumina system. We divided the recurrent tumors into two types, intrahepatic recurrence and abdominal metastasis, according to the location of the luciferase signal in the mice. As anticipated, signals emanating from the liver area indicated intrahepatic recurrence, whereas signals outside the liver area indicated abdominal metastasis. Because signals from the liver area might in fact have been generated by a tumor adhering to the abdominal wall, we verified the luciferase results by euthanizing the mice and measuring the size, location, and frequency of the intrahepatic and extrahepatic tumors.

Treatment of Mice With Sorafenib.

According to our experimental design (Fig. 1A), mice in groups A and B orally received either the solvent control or 40 mg/kg sorafenib every day for 3 weeks; this started on day 4 after tumor implantation. Mice in groups C and D underwent tumor resection on day 10 after implantation; on the day after resection, they began to receive either the oral solvent control or 40 mg/kg sorafenib every day for 3 weeks.

Figure 1.

Experimental design and suppressive effect of sorafenib on xenograft growth. (A) Schematic showing the experimental design. Luciferase-labeled HCC xenografts were orthotopically implanted into the livers of nude mice and then randomly separated into four groups as shown. Mice in groups A and B started to receive either the diluent control or 40 mg/kg sorafenib daily on day 3 after tumor implantation, and the treatment was continued for 3 weeks. Mice in groups C and D underwent HCC resection on day 10 after implantation and then received either the diluent control or 40 mg/kg sorafenib for 3 weeks. (B) Tumor growth of mice in groups A and B was measured twice a week after tumor implantation with the IVIS system. The relative tumor growth was measured according to the luciferase intensity. (C) Tumor recurrences and the growth of recurrent tumors in group C were detected and measured after tumor resection with the IVIS system. Days 0 and 1 indicate the day before resection and the day of resection, respectively. Five representative mice are shown. (D) Repeat of the experiment shown in panel C, except that five mice per group from group D are shown.

ELISA.

ELISA was performed to measure the serum levels of growth factors according to the manufacturer's guide.

3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium Bromide (MTT) Assay.

The MTT assay was used to quantify relative cell growth as previously described.12

Boyden Chamber Assay.

Boyden chambers (polycarbonate membranes with a 8-μm pore size) were obtained from Neuroprobe Corp. (Bethesda, MD). Cells (2 × 105) in 0.05 mL of media with different growth factors were placed in the upper chamber, and the lower chamber was loaded with 0.152 mL of a medium containing 10% fetal bovine serum. After 10 hours of incubation at 37°C with 5% CO2, cellular migration to the lower surface of the filters was detected with traditional hematoxylin and eosin staining, and five fields of each well were counted. Three wells were examined for each condition and cell type, and the experiments were repeated thrice.

Western Blotting.

Western blotting was performed as previously described.12

Statistical Analysis.

All data are presented as means and standard errors of the mean. The Student t test was used for the comparison of measurable variants of two groups. The chi-square test was used for the comparison of HCC recurrence rates of two groups. P < 0.05 was defined as statistically significant.

Results

Sorafenib Suppresses Postoperative Intrahepatic Recurrence and Abdominal Metastasis by Orthotopically Implanted HCC Xenografts.

The experimental design is shown in Fig. 1A. Briefly, the mice bearing orthotopically implanted tumors formed from luciferase-labeled HCC cells were assigned to one of two treatment cohorts: nonresection and resection. The nonresected cohort was divided into two subgroups: A and B. Group B received sorafenib treatment, and group A received the diluent control. To ensure that the PVTT-1 cells used in this study were responsive to sorafenib, we measured the impact of the drug on tumor growth and found a significant reduction in the tumor mass in mice treated with sorafenib (Fig. 1B). The resected cohort was divided into groups C and D, with both groups undergoing tumor resection 10 days after PVTT-1 tumor implantation. One day after surgery, mice in group D received sorafenib treatment, and those in group C received the diluent control. In most of the mice that underwent surgical resection, the primary tumor was totally removed according to the clear absence of a luciferase signal around the liver area (Fig. 1C,D). Strikingly, signals from recurrent tumors emerged within 4 days after surgical resection in the control group. These mice had 100% intrahepatic relapse within 2 weeks after surgical removal of the primary tumor (Figs. 1C and 2A). In marked contrast, the intrahepatic recurrence events were postponed when the mice were treated with sorafenib (Figs. 1D and 2A). More importantly, the overall intrahepatic recurrent rate was reduced almost 40% by sorafenib, and this strongly suggests its potent effect in preventing the regrowth of HCC xenograft tumors. Abdominal metastasis has been used as another marker for HCC malignancy in mouse models. Similarly to the postsurgical formation of intrahepatic tumors, we found that the overall incidence of abdominal metastasis was also significantly suppressed by sorafenib treatment (Fig. 2B).

Figure 2.

Sorafenib suppresses postsurgical recurrence and metastasis and prolongs survival in a mouse model. (A) The postsurgical intrahepatic recurrence rate of mice in groups C and D (n = 9 for both groups) was quantified from the results displayed in Fig. 1C,D, respectively. The appearance of the luciferase signal around the liver area was indicative of intrahepatic recurrence. The chi-square test was applied to analyze the difference between the two groups. *P < 0.05. (B) Postsurgical abdominal metastasis events of mice in groups C and D (n = 9 for both groups) were quantified. *P < 0.05. (C) The relative growth of orthotopic tumors in groups A and B (n = 7 for both groups) and intrahepatic recurrent tumors in groups C and D (n = 9 for group C and n = 6 for group D) was quantified with the IVIS system, and growth curves were constructed. The luciferase activity values for each tumor at the first time point served as a standard for calculating the relative tumor growth for all other time points. The first time point for recurrent tumors refers to the first time when the tumor was detected. (D) The overall survival of mice in all four groups (n = 7 for each group) was measured and analyzed with SPSS software.

Sensitivity to Sorafenib Is Enhanced in Recurrent HCC Tumors, and Postoperative Survival Is Prolonged.

We next compared the inhibitory effects of sorafenib on the growth of HCC xenografts before and after surgery. As previously observed both in murine models and in the clinic with HCC patients,13 postoperative tumor growth was accelerated by tumor resection (Fig. 2C). Remarkably, the extensive postsurgical tumor growth was dramatically suppressed by sorafenib. A careful comparison showed that the antitumor effects of sorafenib were enhanced after partial hepatectomy, and this suggests that recurrent tumors are more vulnerable to sorafenib-induced growth suppression (Fig. 2C). In addition to a considerable reduction of tumor recurrence and robust inhibition of recurrent tumor growth, sorafenib significantly prolonged the postoperative survival of mice originally bearing orthotopic HCC xenografts (Fig. 2D).

Hyperactivation of ERK Signaling Caused by Postoperative Liver Regeneration Sensitizes HCC to Sorafenib-Induced Cell Death.

Liver regeneration and tissue wound healing are initiated after HCC resection, during which an array of growth factors, such as EGF, transforming growth factor α (TGF-α), HGF, and bFGF, are produced and result in the activation of ERK signaling, a crucial pathway for the proliferation of hepatocytes.14-17 We found that the expression of both TGF-α and HGF in the regenerating liver was dramatically elevated after partial hepatectomy in our model (Fig. 3A). In addition, elevated serum levels of the growth factors were tightly correlated to patients' clinical outcomes. As shown in Fig. 3B,C, higher levels of HGF and EGF in the sera of HCC patients predicted shorter tumor-free survival in a statistically significant manner (P = 0.023 for EGF and P = 0.039 for HGF). Thus, we hypothesized that such an alteration in the postoperative liver microenvironment was responsible for the hypersensitivity of recurrent HCC to sorafenib. First, we found that as a response to the increased secretion of growth factors, phosphorylation of ERK was remarkably enhanced in the recurrent tumor, whereas the treatment of sorafenib could effectively block the activation of this signaling (Fig. 4). We proved that such a remarkable alteration in ERK activity would change the sensitivity of HCC cells to sorafenib treatment. It was found that growth factors, including EGF, TGF-α, and HGF, were able to activate ERK signaling and promote cell growth in both 7404 and PVTT-1 cells (Fig. 5A,B). Strikingly, the sensitivities of these cells to sorafenib-induced cell growth suppression were also dramatically enhanced in the presence of growth factors (Fig. 5B). However, when HCC cells did not respond to growth factors, the sensitivity of HCC cells did not change accordingly (Supporting Information Fig. 2). Altogether, hyperactivation of ERK signaling by growth factors can sensitize HCC cells to sorafenib-induced suppression of cell growth.

Figure 3.

Expression of growth factors HGF and TGF-α in the postoperative liver and the relation between growth factors and tumor-free survival for patients. (A) Livers were debrided from the animals in groups C and D. All the tissues were fixed and sent for immunohistochemistry analysis against mouse HGF and TGF-α. (B) Serum levels of EGF in 42 HCC patients were measured by ELISA. The relation between the level of EGF and the time to recurrence of each patient was analyzed by survival analysis with SPSS software. (C) Serum levels of HGF in 42 HCC patients were measured by ELISA. The relation between the level of HGF and the time to recurrence of each patient was analyzed by survival analysis with SPSS software.

Figure 4.

Suppression of ERK activation in recurrent tumors by sorafenib. Both mouse liver tissue and xenografted tissue were debrided from animals in groups A, B, C, and D. All samples were fixed and then sent for immunohistochemistry analysis against pERK.

Figure 5.

The activation of ERK signaling by growth factors elicits hypersensitivity of HCC cells to growth suppression by sorafenib. (A) HCC cell lines 7404 and PVTT-1 were serum-starved overnight, and this was followed by a 2-hour exposure to various cytokines (20 ng/mL EGF, 10 ng/mL TGF-α, or 40 ng/mL HGF) with or without 4 μM sorafenib; cell lysates were subjected to western blotting and probed for expression of pERK1/2/ERK1/2 and tubulin as the loading control. (B) Both 7404 and PVTT-1 cells were placed onto 96-well plates and serum-starved overnight. On the following day, the cells were treated with the diluent control or agents as indicated. After 48 hours, the MTT assay was used to measure the relative growth of the cells. Three independent experiments were performed for each type of treatment, and the results are shown as means and standard errors of the mean.

Sorafenib Blocked the Cytokine-Stimulated Migration of HCC Cells.

Stimulation of postoperative metastasis is another concern for the surgical treatment of cancer.18 The accumulation of several growth factors released by recruited fibroblasts and macrophages during the wound-healing postsurgical period may foster the development of metastases from residual tumors.15 Testing this in our system, we found that bFGF was able to stimulate cell migration by 7404 and Hep3B cells, which were associated with an increased level of activated intracellular ERK, but sorafenib dramatically decreased the bFGF-mediated enhancement of cell migration (Fig. 6A). Likewise, EGF and HGF promoted cell migration of PVTT-1 cells in association with the activation of ERK, and sorafenib blocked the stimulatory effect on cell migration (Fig. 6B). Mechanically, we found that sorafenib could strongly suppress the expression of matrix metalloproteinase 9 (MMP9), a downstream target of ERK signaling that may be responsible for the migratory inhibition effect of sorafenib (Fig. 6C). Conclusively, the migration of HCC cells triggered by growth factors was effectively inhibited by sorafenib, and this likely led to inhibition of the development of metastatic disease.

Figure 6.

Sorafenib suppresses growth factor–induced migration of HCC cells. (A) Hep3B and 7404 cells were tested with the Boyden chamber assay with the diluent control, 5 ng/mL bFGF, or 5 ng/mL bFGF with or without 2 μM sorafenib. (B) PVTT-1 cells were tested with the Boyden chamber assay with the diluent control, 80 ng/mL EGF, 40 ng/mL HGF, 80 ng/mL EGF, or 40 ng/mL HGF with or without 3 μM sorafenib. (A,B) Cells that migrated through the filter membrane of the Boyden chamber were stained, and representative results are shown in the left panels. Quantification of migratory cells was analyzed with Image Pro software and is shown in the right panels. (C) Hep3B, 7404, and PVTT-1 cells were serum-starved overnight, and this was followed by 6 hours of treatment as indicated. After that, the cells were harvested for both reverse-transcription polymer chain reaction analysis (for the measurement of the expression of MMP9) and western blotting (for the detection of pERK).

Discussion

Surgical resection is the mainstay of HCC treatment; nevertheless, the high incidence of postoperative recurrence remains a big challenge for sustained remission. In this study, we showed that sorafenib significantly suppressed postsurgical intrahepatic recurrences as well as abdominal metastasis in our orthotopic HCC xenograft model. This finding may have a beneficial impact on the clinical practice of HCC therapy and especially on the radical intervention used for early-stage patients. It is reasonable to anticipate that the application of sorafenib to patients who have undergone HCC resection could improve the surgical outcome via the effective inhibition of tumor relapse and metastasis. Recently, an increasing number of studies have focused on the accurate prediction of recurrence in early-stage HCC patients, mainly through a state-of-the-art technique generating predictive gene expression signatures.19-21 According to those reports, a certain portion of early HCC patients can be subcategorized as prorecurrent patients with poor prognosis, and these patient may benefit from sorafenib therapy to reduce postoperative recurrence.

Because dysregulation of the Ras-Raf-ERK pathway is involved in a spectrum of oncogenic processes including transformation, cell proliferation, and invasion,22-24 the level of activation of ERK1/2 associated with tumor development is expected to serve as a strong predictor of poor prognosis in HCC patients.25 In addition, some pathological factors contributing to the activation of ERK, such as chronic infection with hepatitis C virus, are also risk factors for HCC recurrence.8 ERK signaling is likely triggered by stress-induced growth factors in the process of postoperative liver regeneration, which initiate and promote the proliferation of normal hepatocytes.26 However, these same growth factors could activate ERK signaling even more vigorously in the residual recurrence-causing HCC cells, in which EGFR and c-Met are frequently overexpressed.27, 28 Thus, sorafenib, a potent Raf kinase inhibitor, might effectively arrest or eliminate the activated residual HCC cells. Our in vitro experiments suggest that, as a result of growth factor stimulation, the sensitivity of HCC cells to sorafenib is apparently enhanced in the presence of ERK activation, and this is consistent with our in vivo finding that sorafenib markedly decreases tumor recurrence by efficient blunting of activated oncogenic kinase pathways (Fig. 7). Our findings highlight the functional relationship between the ERK status and the response to sorafenib.29

Figure 7.

Model for the hypersensitivity of HCC cells to sorafenib caused by postsurgical liver regeneration associated with the release of cytokines.

In addition to growth inhibition, we also unveiled the role of sorafenib in the suppression of growth factor–stimulated cell migration. Dysregulation of cell migration is indispensable for intravenous invasion as well as extravenous colonization by metastasized HCC cells. Our results demonstrate the efficacy of sorafenib in suppressing HCC metastasis, especially in the postresection period, during which the wound-healing process is highly active. Moreover, neovascularization is another vital physiological event occurring during the postoperative wound-healing process that also promotes metastasis.15, 30, 31 Because sorafenib is a potent inhibitor of angiogenesis through the targeting of vascular endothelial growth factor receptor 2, its dramatic suppression of postoperative recurrence and metastasis of HCC may occur through the suppression of hyperactivated endothelial cells (Supporting Information Fig. 4). In summary, this study shows that sorafenib effectively suppresses postsurgical HCC recurrence and metastasis in the HCC xenograft model. These findings strongly suggest that clinical trails should be conducted to accurately evaluate the efficacy of sorafenib in preventing postsurgical HCC recurrence.

Acknowledgements

The authors thank Dr. H. Phillip Koeffler (Cedars-Sinai Medical Center) for his helpful discussions, suggestions, and critical reading of the manuscript.

Ancillary