Targeting mitogen-activated protein kinase kinase with the inhibitor PD0325901 decreases hepatocellular carcinoma growth in vitro and in mouse model systems


  • Matthew Hennig,

    1. Department of Surgery, Indiana University School of Medicine, Indianapolis, IN
    Search for more papers by this author
  • Michele T. Yip-Schneider,

    1. Department of Surgery, Indiana University School of Medicine, Indianapolis, IN
    Search for more papers by this author
  • Sabrina Wentz,

    1. Department of Surgery, Indiana University School of Medicine, Indianapolis, IN
    Search for more papers by this author
  • Huangbing Wu,

    1. Department of Surgery, Indiana University School of Medicine, Indianapolis, IN
    2. Walther Oncology Center, Indianapolis, IN
    Search for more papers by this author
  • S. K. Hekmatyar,

    1. Biomedical NMR Research Center, Department of Radiology, Dartmouth Medical School, Hanover, NH
    Search for more papers by this author
  • Patrick Klein,

    1. Department of Surgery, Indiana University School of Medicine, Indianapolis, IN
    Search for more papers by this author
  • Navin Bansal,

    1. Department of Radiology, Indiana University Purdue University at Indianapolis, IN
    2. Indiana University Cancer Center, Indiana University School of Medicine, Indianapolis, IN
    Search for more papers by this author
  • C. Max Schmidt

    Corresponding author
    1. Department of Surgery, Indiana University School of Medicine, Indianapolis, IN
    2. Walther Oncology Center, Indianapolis, IN
    3. Indiana University Cancer Center, Indiana University School of Medicine, Indianapolis, IN
    4. Richard L. Roudebush Veterans Affairs Medical Center, Indianapolis, IN
    5. Department of Biochemistry and Molecular Biology, Indiana University School of Medicine, Indianapolis, IN
    • Departments of Surgery and Biochemistry Molecular Biology, Cancer Research Institute, 980 West Walnut Street, Room R3-C522, Indianapolis, IN 46202
    Search for more papers by this author
    • fax: 317-278-4325

  • Potential conflict of interest: Nothing to report.


Hepatocellular carcinoma (HCC) is a common cause of death from solid organ malignancy worldwide. Extracellular signal-regulated/mitogen-activated protein kinase kinase (MEK) signaling is a critical growth regulatory pathway in HCC. Targeting MEK with a novel small molecule inhibitor, PD0325901, may inhibit HCC tumorigenesis. PD0325901 (0.01-100 nM) inhibited growth and MEK activity in vitro in immortalized murine transforming growth factor alpha (TGF-α) transgenic hepatocyte (TAMH) cells, derived from the livers of TGF-α transgenic mice. Treatment of athymic mice bearing TAMH flank tumors with vehicle or PD0325901 (20 mg/kg) revealed a significant reduction of MEK activity ex vivo 24 hours after a single PD0325901 dose. The growth rate of TAMH flank tumors over 16 days was reduced threefold in the treatment arm (1113 ± 269% versus 3077 ± 483%, P < 0.01). PD0325901 exhibited similar inhibitory effects in HepG2 and Hep3B human HCC cells in vitro and in Hep3B flank tumors in vivo. To confirm this in a developmental model, MT-42 (CD-1) TGF-α mice were treated with vehicle or PD0325901 (20 mg/kg) for 5 weeks. Gross HCC was detected in 47% and 13.3% of the control and treatment mice, respectively. Tumor growth suppression by PD0325901 relative to vehicle was also shown by magnetic resonance imaging. These studies provide compelling preclinical evidence that targeting MEK in human clinical trials may be promising for the treatment of HCC. (HEPATOLOGY 2010.)

Hepatocellular carcinoma is the most common primary liver malignancy worldwide, and its incidence has been rising over the last 20 years.1 Surgical resection or liver transplantation is the best hope for improving survival in patients with HCC; however, only a minority of patients are candidates for these procedures.2 Surgical resection for cure is limited to those patients without distant metastases or local invasion of adjacent tissues.3 Most patients are diagnosed with HCC at stages too advanced for curative therapy, with poor prognosis even with disease spread only to regional lymph nodes.4 In selected patients, however, surgical resection and transplantation can achieve 5-year survival rates of approximately 60%.5–7

Because many patients are ineligible for surgical therapies, several chemotherapies have been evaluated for treatment of this disease. As a single agent, doxorubicin has no effect on prolonged survival and demonstrates increased mortality caused by cardiac toxicity.8 Currently chemotherapy regimens consist of doxorubicin/5-fluorouracil combinations; however, these drugs show a response rate of only 20%-30%.9 Doxorubicin and 5-fluorouracil target broad cellular processes by blocking DNA topoisomerase II or acting as a pyrimidine analog, respectively, leading to cell cycle arrest. Meta-analysis of more than 21 chemotherapy studies shows no improved survival or decrease in recurrence after resection.10 Newer chemotherapies target specific signaling pathways that are unique or up-regulated in various carcinomas and therefore may be more effective. For example, sorafenib (BAY 43-9006, Nexavar) is an oral multikinase inhibitor of Raf kinase, which functions upstream of extracellular signal-regulated/mitogen-activated protein kinase kinase (MEK), as well as receptor tyrosine kinases, including vascular endothelial growth factor receptor and platelet-derived growth factor receptor. Sorafenib has recently been shown to provide a survival benefit in select hepatocellular carcinoma (HCC) patients.11 A randomized phase III double-blind placebo-controlled trial including 602 patients with advanced HCC showed a 3-month survival improvement in patients treated with sorafenib. The median overall survival was 10.7 months with sorafenib compared with 7.9 months with placebo.12 The clinical efficacy of sorafenib suggests that targeting such kinase pathways may hold promise for the treatment of HCC.

The p42/p44 extracellular signal-regulated/mitogen-activated protein kinase (ERK/MAPK) pathway is up-regulated in most human HCC.13–16 This pathway is responsible for regulating cell growth by its effects on growth factor receptors, transcriptional factors, cytoskeletal proteins, phospholipases, and other protein kinases.17 Further studies indicate that up-regulation of the MAPK pathway occurs specifically in hepatic carcinoma, not in normal hepatic tissue, suggesting a mechanism for proliferation in HCC.13, 16 Transforming growth factor alpha (TGF-α) is also up-regulated in most HCCs.18–22 Like the MAPK pathway, TGF-α is specifically up-regulated in tumor cells as compared with normal liver cells.23 TGF-α signals through the epidermal growth factor receptor, which in turn signals via the MAPK pathway, making TGF-α a potent stimulator of this pathway.24–26 Prior in vitro studies suggest that blocking the MEK-ERK signaling pathway induces the death of certain human HCC cell lines.27 In the current study, we use a novel MEK inhibitor, PD0325901, that blocks the conversion of ERK to its activated, phosphorylated form by inhibiting activated MEK1 and MEK2.28 The effect of PD0325901 in HCC is evaluated in vitro and in vivo, using an athymic mouse model and a MT42 (CD-1) TGF-α transgenic mouse model. In vivo studies on mice with orthotopic HCC are performed using magnetic resonance imaging (MRI) for tumor volume determination.


ERK, extracellular signal-regulated protein kinase; HCC, hepatocellular carcinoma; HPMT, 0.5% hydroxypropyl methyl cellulose, 0.2% Tween 80; MAPK, mitogen-activated protein kinase; MEK, mitogen-activated protein kinase kinase; MRI, magnetic resonance imaging; P-ERK, phosphorylated extracellular signal-regulated protein kinase; TAMH, transgenic hepatocyte cell line; TGF-α, transforming growth factor alpha.

Materials and Methods

Cell Culture and Treatments.

The immortalized murine TGF-α transgenic hepatocyte cell line (TAMH, provided by Nelson Fausto) was derived from freshly isolated hepatocytes from the livers of TGF-α transgenic mice. These cells have the characteristic appearance of well-differentiated human HCC. TAMH cells were cultured in Dulbecco's modified Eagle medium/F12 (Mediatech, Herndon, VA) supplemented with nicotinamide (10 mM), gentamicin (50 μg/mL), dexamethasone (10−7 M) (Sigma, St Louis, MO), insulin-transferrin-selenium supplement (ITS-X) (1 mL/L; Gibco, Grand Island, NY), and epidermal growth factor (20 ng/mL; Invitrogen, Carlsbad, CA). Cells were treated with various doses of PD0325901 (Pfizer, Ann Arbor, MI). The human hepatocellular carcinoma cell lines HepG2 and Hep3B were obtained from American Type Culture Collection (Manassas, VA). These cells were cultured in modified Eagle medium-alpha containing 10% fetal bovine serum, 100 units/mL penicillin, and 100 μg/mL streptomycin at 37°C (5% CO2, 95% O2).

Western Immunoblotting.

Cells were lysed or tumor tissue was homogenized in radioimmune precipitation buffer (phosphate-buffered saline, 1% Nonidet P-40, 0.5% sodium deoxycholate, 0.1% sodium dodecyl sulfate, 1 mM phenylmethylsulfonyl fluoride, 10 μg/mL aprotinin, 1 mM Na3VO4). Lysates were resolved by sodium dodecyl sulfate polyacrylamide gel electrophoresis on 4% to 20% gradient gels (Pierce/Thermo Fisher Scientific, Waltham, MA). Separated proteins were electrophoretically transferred to Immobilon P membranes (Millipore, Bedford, MA) and probed with specific phospho-p44/42 mitogen-activated protein kinase (MAPK; Thr202/Tyr204; Cell Signaling, Beverly, MA) and total extracellular signal-regulated kinase (ERK)1/2 (Santa Cruz Biotech, Santa Cruz, CA) antibodies.

Cell Proliferation Assay.

Cell proliferation was determined using a colorimetric assay, CellTiter 96 AQueousOne Solution Cell Proliferation Assay (Promega, Madison, WI). Cells were plated in triplicate in 96-well plates (2.5 × 103 cells/well). Twenty-four hours later, PD0325901 or vehicle control (dimethylsulfoxide) was administered. After treatment for the indicated time, cells were incubated with 20 μL CellTiter 96 AQueous One Solution Reagent, and absorbance was recorded at 490 nm. Percent cell growth was determined from a ratio of average absorbances of the treatment wells to the control wells.


HepG2 or Hep3B cells were plated in 96-well plates. The next day, PD0325901 or vehicle was added for 48 hours. Relative apoptosis was determined using the Cell Death Detection enzyme-linked immunosorbent assay (Roche, Indianapolis, IN) by comparing the absorbance of drug-treated with that of vehicle-treated cells.

TAMH and Hep3B Athymic Mouse Model of HCC.

Six-week old athymic mice (Harlan, Indianapolis, IN) were obtained, and 1 × 107 TAMH cells were injected into the right flank. The tumors were allowed to grow for 5 weeks before the start of treatment. The first arm was treated for 24 hours, and the second arm was treated for 16 days. The first arm of this trial consisted of 10 mice divided into two groups, receiving either a single dose of PD0325901 (20 mg/kg) or an equivalent volume of vehicle (0.5% hydroxypropyl methyl cellulose, 0.2% Tween 80 [HPMT]) via orogastric gavage. After 24 hours, the animals were sacrificed, and the flank tumors were excised, frozen in liquid nitrogen, and stored at −80°C for ex vivo tumor analysis. The second arm of this trial consisted of 17 athymic mice receiving either PD0325901 or HPMT vehicle daily by orogastric gavage. Tumor volume was measured twice per week via vernier caliper (Scienceware, Pequannock, NJ). Volume estimations were determined by the following formula:

equation image

Hep3B cells (1 × 106 cells) were injected into the flanks of athymic mice. When tumors were visible, treatment with either PD0325901 (10 mg/kg) or vehicle was initiated and tumor volume monitored as described for 4 weeks. All animals were housed and fed in American Association for Accreditation for Laboratory Animal Care–approved facilities, and animal research and handling were in strict conformance with federal Institutional Animal Care and Use Committee guidelines.

TGF-α Transgenic Mouse Model of HCC.

MT42 (CD-1) TGF-α transgenic mice (provided by Glen Moreno) were injected with diethylnitrosamine (5 mg/kg) at 14 days of age. The mice were started with daily treatments of vehicle or PD0325901 (20 mg/kg) at 30 weeks of age. After 35 days of treatment, the animals were sacrificed and examined for gross tumor formation. A second group of mice underwent screening MRI to detect HCC at older than 6 months of life. Sixteen mice with index lesions (HCC) underwent determination of baseline tumor growth and were randomized to receive daily orogavage of either HPMT vehicle or PD0325901 (20 mg/kg). Serial MRIs were performed biweekly, and the tumor volume was determined. At the time of sacrifice, representative liver sections from different lobes were fixed in 10% formalin, paraffin embedded, and serially cut (5 μm).

MRI Acquisition.

In vivo proton magnetic resonance imaging ([1H]-MRI) of the mice with orthotopic tumors were performed biweekly using a 9.4-Tesla, 31-cm horizontal bore system (Varian Inc, Palo Alto, CA) equipped with a 12-cm-diameter shielded gradient set capable of up to 38 gauss/cm gradient strength in three directions. The mice were anesthetized with 0.75% isofluorane delivered in medical air at 1 L/minute using a nose mask connected to a gas anesthesia machine (Vetland, Louisville, KY). The animal was positioned inside a 30-mm-diameter and 25-mm-high loop-gap volume coil tuned to 400 MHz. Warm air was blown through the magnet bore to maintain the animal core body temperature at 35°C to 37°C, which was monitored with a fiber optic rectal probe (FISO Technologies, Quebec, Canada). Transaxial proton-density weighted images with fat suppression were obtained using a multi-slice spin-echo sequence and the following imaging parameters: field of view, 3 × 3 cm, slice thickness = 1 mm, number of slices = 20, matrix size = 256 × 128, signal averages = 2, repetition time (TR) = 1500 ms, and echo time (TE) = 15 msec. The total image data collection time was approximately 9 minutes. Tumor area was calculated by drawing a region of interest on the liver tumor slices using the Varian Browser software. The area of tumor in each slice was multiplied by the slice thickness plus slice gap to calculate tumor volume per slice. The total tumor volume was obtained by adding the total area of all slices.


Immunohistochemistry to detect DNA fragmentation was performed on formalin-fixed, paraffin-embedded liver sections using ApopTag Peroxidase In Situ Apoptosis Detection Kit (Millipore, Billerica, MA). Positively stained cells were counted in four fields (400×) with the highest density of staining per slide and expressed as a percentage relative to the total number of cells.


The Fisher's exact test was used for comparison of two groups with one observed variable. The Student t test was used for comparison of two groups, with P < 0.05 considered significant.


Effect of the MEK Inhibitor PD0325901 on MEK Activity, Cell Growth, and Apoptosis In Vitro.

The TAMH cell line, derived from TGF-α transgenic mouse hepatocytes, was exposed to PD0325901 (0-100 nM) for 1 or 24 hours. MEK activity reflected by the level of active, phosphorylated ERK (P-ERK) was determined by immunoblot (Fig. 1A). At both time points, P-ERK levels decreased with increasing concentrations of PD0325901; total ERK levels were unchanged with treatment. PD0325901 also dose dependently inhibited the growth of TAMH cells in vitro (Fig. 1A). Inhibition of MEK activity and cell growth were also observed in PD0325901-treated HepG2 and Hep3B human HCC cells in vitro, thus demonstrating MEK-dependent growth (Figs. 1B, C). Furthermore, apoptosis was dose dependently induced in HepG2 and Hep3B cells in vitro after treatment with PD0325901 for 48 hours (Fig. 1D). The higher dose of PD0325901 required to induce apoptosis in Hep3B cells correlates with the greater sensitivity of HepG2 cells to PD0325901 compared with Hep3B cells.

Figure 1.

Effect of PD0325901 on MEK activity (P-ERK), cell growth, and apoptosis in vitro. (A) After the indicated treatments for either 1 or 24 hours, TAMH cells were lysed, and western immunoblots were performed using antibodies to detect phosphorylated ERK1,2 and total ERK1,2 as shown in the representative experiment. Effect of PD0325901 on TAMH cell growth is shown below. Cells were treated for 72 hours with the indicated concentrations of PD0325901. Cell growth was determined by colorimetric proliferation assay and expressed as a percentage relative to control-treated cells (100%). (B) Effect of PD0325901 on MEK activity and cell growth in HepG2 cells. (C) Effect of PD0325901 on MEK activity and cell growth in Hep3B cells. (D) PD0325901-induced apoptosis in HepG2 and Hep3B cells. Results from at least two independent experiments are shown. Error bars represent standard error of the mean (SEM). *P < 0.05 compared with vehicle control.

Effect of PD0325901 on MEK Activity (P-ERK) in Athymic Mouse TAMH Flank Tumors.

Athymic mice bearing TAMH flank tumors were given the MEK inhibitor PD0325901 (20 mg/kg) or HPMT vehicle daily by orogastric gavage. After a one-time 24-hour treatment, ex vivo tissue analysis was performed, and P-ERK levels were determined by immunoblot (Fig. 2A). After 24 hours of treatment, intratumoral P-ERK levels decreased in animals receiving a single dose of PD0325901 compared with those animals receiving vehicle.

Figure 2.

Effect of PD0325901 on MEK activity (P-ERK) and growth of athymic mouse TAMH flank tumors. Athymic mice bearing TAMH flank tumors were given the MEK inhibitor PD0325901 (20 mg/kg) or HPMT vehicle by orogastric gavage. (A) After a one-time 24-hour treatment, ex vivo tumor analysis was performed. The tumors were homogenized, and western immunoblots were performed to detect phosphorylated (P-ERK) and total ERK1,2 in representative animals. (B) TAMH flank tumors (n = 8 for vehicle; n = 9 for PD0325901, daily treatment for 16 days) were measured twice per week for volumetric estimation of size. Data are expressed as percent change in tumor size from the start of treatment on day 0. Error bars represent SEM. *P < 0.05 compared with vehicle control.

Effect of PD0325901 on TAMH and Hep3B Flank Tumor Growth.

Serial volumetric measurements of the athymic mouse TAMH flank tumors treated for 16 days were obtained. The growth rate of these tumors was expressed as the percentage increase in tumor size from the start of treatment (Fig. 2B). After 16 days of treatment, the growth rate of the flank tumors in the PD0325901 treatment arm (n = 9) was reduced threefold compared with the vehicle-treated arm (n = 8) (1113% ± 269% versus 3077 ± 483%, P < 0.01). This demonstrates that PD0325901 can effectively reduce TAMH tumor growth in vivo. Similarly, PD0325901 (10 mg/kg) significantly inhibited human HCC tumor growth over a 4-week period in the Hep3B xenograft model (Fig. 3). No drug-induced toxicity indicated by weight loss or treatment related mortality was observed in either of these studies.

Figure 3.

Effect of PD0325901 on Hep3B flank tumor growth in athymic mice. Athymic mice bearing Hep3B flank tumors were treated with either PD0325901 (10 mg/kg, n = 4) or vehicle (n = 4) daily by orogastric gavage for 4 weeks. Tumor volume was monitored weekly as shown. Error bars represent SEM. *P < 0.05 compared with vehicle control.

Effect of PD0325901 on HCC Tumor Incidence in TGF-α Transgenic Mice, a Developmental Model of HCC.

MT42 (CD-1) TGF-α transgenic mice were injected with diethylnitrosamine (5 mg/kg) at 14 days of age. At 30 weeks of age, TGF-α transgenic mice were treated with either PD0325901 (20 mg/kg) or vehicle by daily orogastric gavage for 35 days. No evidence of weight loss was observed in either group. On sacrifice, the animals underwent necropsy, and gross tumor incidence was noted. Whereas vehicle-treated animals had a 47.1% (8 of 17 total) incidence of HCC, animals treated with the MEK inhibitor PD0325901 demonstrated a significant decrease in HCC incidence, down to 13.3% (2 of 15 total; P < 0.05). This represents an approximately 3.5-fold decrease in gross tumor incidence.

Effect of PD0325901 on TGF-α HCC Tumor Growth: Magnetic Resonance Imaging.

Because what percentage of transgenic mice had tumors at the start of the prior randomized trial was unknown, a second study was initiated in which the mice were screened by MRI to identify tumor-bearing animals (Fig. 4). Once tumors were noted on MRI (week 0), a follow-up MRI was performed 2 weeks later (week 2) to determine baseline growth of the index lesions. At this point, the animals were divided into two arms that received either PD0325901 (20 mg/kg; n = 12) or vehicle (n = 4). Biweekly MRI examinations followed to determine volumetric changes in tumor size between the two arms compared with the initial rate of tumor growth (Fig. 5A). After only 2 weeks of treatment, the MRI at week 4 showed a significant difference in volume between the two arms, with the MEK inhibitor arm regressing in volume (vehicle = 108.5% ± 5.3%, PD0325901 = 53.9% ± 9.3%, P < 0.02). The next MRI at week 6 continued to show a significant difference in tumor volume between the two arms, with the PD0325901 arm demonstrating further regression in tumor volume (vehicle = 136.3% ± 10.5%, PD0325901 = 51.4% ± 10.2%, P < 0.001). The next series of MRI images at week 8 demonstrated tumor growth with vehicle treatments (141.7 %) and continued regression in tumor volume with PD0325901 (55.9% ± 19.5%). Apoptosis was significantly induced in the PD0325901 arm compared with the vehicle arm (Fig. 5B). Some mortality was observed in both arms of this study, most likely because of the stress of undergoing the MRI procedure in combination with drug treatment. Similar tumor regression was detected by MRI after treatment with a lower dose of PD0325901 (10 mg/kg; data not shown),

Figure 4.

[1H]-MRI of PD0325901-treated TGF-α transgenic mice. Pretreatment and posttreatment MRI examinations for a representative control (top) and a PD0325901-treated (bottom) mouse are shown at weeks 2 (pre) and 4 (post). Daily treatment with the vehicle or PD0325901 (20 mg/kg) by orogastic gavage was begun following the second MRI at week 2. Biweekly MRI examinations were performed. Index lesions are indicated by the arrows.

Figure 5.

The effects of PD0325901 on TGF-α HCC tumor growth as measured by 1H MRI. (A) Mice with index lesions were identified by MRI at week 0 (n = 16). MRI scans were performed biweekly. Daily treatment was begun after the second MRI scan at week 2 (n = 4 for vehicle; n = 12 for PD0325901). Manual tracing was performed on the tumor to calculate tumor area. The area in each slice was multiplied with slice thickness plus slice gap and co-added to calculate tumor volume, and the percent change in tumor volume from the start of treatment was determined. *P < 0.02. (B) Immunohistochemistry was performed to detect apoptosis in vehicle and PD0325901 arms as shown in the representative Apoptag-stained liver sections (original magnification ×400). Percent apoptosis was determined for each arm (vehicle, n = 3; PD0325901, n = 5; four fields counted per slide) as shown graphically. *P < 0.05 compared with vehicle control.


Current chemotherapy for HCC has had little success in treating this disease. The future direction of chemotherapy is to target specific pathways that are known to be involved in the particular cancer. The ERK/MAPK pathway is up-regulated in most human HCC tumors; thus, targeting this pathway could suppress tumor growth and in turn increase the life span of HCC patients. Prior attempts at targeting the MEK-ERK signaling cascade have not proved successful in human trials and have led to the development of newer, more bioavailable MEK inhibitors. PD0325901, a derivative of CI-1040, is potent at nanomolar concentrations and has greater duration, potency, and solubility, resulting in improved bioavailability and increased metabolic stability over CI-1040.28 The inhibitor binds to MEK1/2 at a non–adenosine triphosphate binding site, causing conformational changes that prevent it from phosphorylating ERK, making it a highly selective inhibitor.28

The current study employed TAMH cells, an immortalized line obtained from the MT42 (CD-1) TGF-α transgenic mouse, as well as HepG2 and Hep3B human HCC cells. In all three cell lines, we demonstrated that PD0325901 effectively reduced P-ERK levels and cell growth in vitro, with effects seen in the nanomolar range. Growth inhibition was associated with the induction of apoptosis in HepG2 and Hep3B cells in vitro. PD0325901 also inhibited TAMH and Hep3B tumor growth in an athymic mouse model in vivo. TAMH flank tumors showed decreases in P-ERK levels 24 hours after a single dose of PD0325901 compared with vehicle control, confirming inhibition of the MEK-ERK pathway. In addition, TAMH tumor growth rate of the inhibitor-treated arm was threefold less than the vehicle-treated arm, demonstrating therapeutic efficacy of PD0325901.

We have previously demonstrated that other MEK inhibitors (PD098059, U0126, PD184161) reduce ERK phosphorylation (MEK activity) and growth in human HCC cells.27, 29 PD0325901 is much more potent than these MEK inhibitors in HCC cells in vitro based on its median inhibitory concentration, which lies in the nanomolar range. In a recent study, Raf-1 small interfering RNA (100 nM) caused a 50% decrease in phosphorylated ERK levels that was associated with a 50% decrease in HCC growth in vitro30 (and unpublished results). Similar results were obtained with ERK1,2 olignonucleotide anti-sense (300 nM) that decreased total ERK levels with a corresponding decrease in HCC cell growth.27 The effective dose and inhibitory effects of the small interfering RNA and anti-sense are comparable to that of PD0325901 in HepG2 and Hep3B cells in vitro. Taken together, these results suggest that PD0325901 is a MEK inhibitor with absolute specificity.

To investigate the efficacy of MEK inhibition in a more clinically relevant model, TGF-α transgenic mice from which the TAMH line was derived were employed. These animals have a human TGF-α transgene that is specifically up-regulated in HCC tumors within the liver.31 The TGF-α transgenic mice are known to develop well-differentiated HCC in 70% of animals by 15 months of age.32 Indeed, studies of rat HCC show that preneoplastic regions in the liver grow at a threefold faster rate with up-regulation of TGF-α.33 Furthermore, because most human HCC tumors have an increased amount of TGF-α present, the TGF-α transgenic mouse is believed to be a valid model of HCC for the current study.34, 35

Because TGF-α is a potent activator of the MEK-ERK pathway, these animals are ideal for treatment with a MEK inhibitor.26 In TGF-α transgenic mice at 30 weeks of age, we previously demonstrated an eightfold increase in P-ERK expression within HCC hepatocytes compared with normal hepatocytes. In addition, the ability of PD0325901 to decrease P-ERK within normal hepatocytes in the treatment arm correlated with its ability to prevent HCC formation in this model.36 In the current randomized study, the incidence of carcinoma in the diethylnitrosamine accelerated transgenic model was determined. A 3.5-fold decrease in tumor incidence was seen in animals given the MEK inhibitor. We wanted to further examine the mechanism of PD0325901 and determine whether it was preventing formation, halting progression, or causing regression of established tumors in this developmental model. To achieve this, MRI confirmation of the presence of tumors was performed, and then treatment was initiated. After serial examinations, a dramatic regression in tumor volume was observed (Fig. 4). Taken together, the animals that were examined by MRI showed a difference in tumor volume approaching a threefold decrease in the MEK-inhibitor treated mice compared with the vehicle-treated mice (Fig. 5). Although PD0325901 reduced tumor growth by inducing apoptosis, thereby resulting in regression, tumor persistence as well as proliferation were still observed after treatment.

Preliminary studies from our laboratory suggest that the anti-tumor effects of PD0325901 are reversible. MRI was performed to identify index lesions and determine baseline tumor growth. PD0325901 (20 mg/kg) was then administered for a period of 4 weeks, resulting in tumor regression; placebo instead of PD0325901 was subsequently administered for an additional 6 weeks. With placebo treatment, the index lesions began increasing in size, reaching the starting tumor volume or greater by the end of the 6-week period (data not shown). This suggests that multiple cycles of drug treatment will be needed for therapeutic efficacy.

In conclusion, a novel MEK inhibitor, PD0325901, shows efficacy in the regression of HCC in a TGF-α transgenic mouse model. Because this model recapitulates the environment that is present in human hepatocellular carcinoma, PD0325901 may have similar efficacy in the treatment of human disease. Preclinical studies in other cancers suggest that continuous therapy is needed to maintain tumor regression; however, this has not been examined in HCC.28 Because the MAPK pathway is also important in cellular growth and hepatic regeneration, studies on the effects of inhibiting MEK in the presence of partial hepatectomy would be important in determining when chemotherapy should be administered in the adjuvant setting. Taken together, these findings provide preclinical support for the use of PD0325901 or other small molecule MEK inhibitors as chemoprevention as well as chemotherapy of HCC. In addition, our results reveal that MRI can play an important role in the preclinical evaluation of drugs using a transgenic tumor model.