HCV796: A selective nonstructural protein 5B polymerase inhibitor with potent anti-hepatitis C virus activity In Vitro, in mice with chimeric human livers, and in humans infected with hepatitis C virus


  • Potential conflict of interest: Drs. Kneteman, Mercer, and Lund own stock in KMT Hepatech, Inc. Dr. Immermann owns stock in Wyeth Research. Dr. Villano owns stock in Viropharma, Inc. Dr. Collett owns stock in Wyeth Research and Viropharma.


Anti-hepatitis C virus (HCV) drug development has been challenged by a lack of experience with inhibitors inclusive of in vitro, animal model, and clinical study. This manuscript outlines activity and correlation across such a spectrum of models and into clinical trials with a novel selective nonstructural protein 5B (NS5B) polymerase inhibitor, HCV796. Enzyme assays yielded median inhibitory concentration (IC50) values of 0.01 to 0.14 μM for genotype 1, with half maximal effective concentration (EC50s) of 5 nM and 9 nM against genotype 1a and 1b replicons. In the chimeric mouse model, a 2.02 ± 0.55 log reduction in HCV titer was seen with monotherapy, whereas a suboptimal dose of 30 mg/kg three times per day in combination with interferon demonstrated a 2.44 log reduction (P = 0.001 versus interferon alone) Clinical outcomes in combination with pegylated interferon and ribavirin have revealed additive efficacy in treatment naïve patients. Abnormal liver function test results were observed in 8% of HCV-796 patients treated for over 8 weeks, resulting in suspension of further trial activity. Conclusion: The RNA-dependent RNA polymerase inhibitor HCV796 demonstrated potent anti-HCV activity consistently through enzyme inhibition assays, subgenomic replicon, and chimeric mouse studies. Strong correlations of outcomes in the mouse model were seen with subsequent clinical trials, including a plateau in dose-related antiviral activity and additive impact from combination therapy with interferon. These outcomes demonstrate the utility of the range of in vitro and in vivo models now available for anti-HCV drug development and support the potential utility of polymerase inhibitors in future combination therapies for HCV treatment. (HEPATOLOGY 2009.)

Hepatitis C virus (HCV) infects an estimated 170 million people across the world,1 with substantial risk of severe chronic liver disease, cirrhosis, and hepatocellular cancer.2, 3 Standard therapy with pegylated interferon and ribavirin clears virus in only half of those treated, is associated with significant treatment-related morbidity, and is cost-prohibitive in much of the world.4, 5 HCV is the leading indication for liver transplantation across the developed world. Despite intensive effort, no new drugs have been approved for therapy in the last decade. More effective and less toxic anti-HCV therapeutics as well as preventive vaccines are greatly needed. A major challenge with development of anti-HCV drugs has been the lack of in vitro, animal model, and clinical experience with inhibitors. Although correlation exists for interferon, there are limited data for molecules with novel mechanisms. The goal of the current work was to delineate the activity of a selective nonstructural protein 5B (NS5B) polymerase inhibitor in vitro by enzyme assay and in the replicon, in vivo in the chimeric mouse model, and in patients infected with HCV.

Since the discovery of the HCV genome in 1989,6 extensive studies have been conducted to understand the structures and functions of individual components involved in HCV replication. Among them, the nonstructural protein 5B (NS5B) is a viral-encoded RNA-dependent RNA polymerase (RdRp) essential for HCV replication. Because there is no counterpart in mammalian cells, this protein has been a prime target of intensive antiviral research.

In vitro enzymatic activities of the NS5B protein have been characterized extensively.7–9 Early drug discovery for HCV polymerase inhibitors typically use either full-length or carboxyl-terminally truncated forms of NS5B as the target for high-throughput compound screening. The availability of high-order crystal structures of NS5B from two genotype 1b strains10, 11 and one 2a strain12 has accelerated our understanding of structure–activity relationships, and consequently many potent compounds have been discovered through structure-based drug design. Although HCV replicates poorly in tissue culture, viral replication can be evaluated in the HCV subgenomic replicon system.13, 14 The use of the replicon system has helped to confirm intracellular antiviral activities of potential drug hits and eliminate many potential nonspecific cytotoxic compounds. The in vitro enzyme and cell-based systems have selected many anti-HCV inhibitors; however, development of a bona fide drug candidate must have acceptable physicochemical and biopharmaceutical properties as well as safety standards in vivo. Currently, the chimpanzee and the “chimeric” mouse transplanted with human hepatocytes are the two infectious HCV animal models used to evaluate the correlation between in vitro screening results and in vivo outcomes.15, 16 In particular, the utility of the chimeric mouse model in predicting clinical antiviral outcomes in humans has been validated with interferon alpha 2b, and a protease inhibitor, BILN2061.17

HCV-796, a benzofuran-C3-carboxamide inhibitor of HCV RdRp, is the second clinical candidate derived from this chemical series as part of collaborative discovery efforts by Wyeth Research and ViroPharma Incorporated (Fig. 1). The original molecule identified from screening an internal compound collection was well tolerated but exhibited only modest clinical activity. Subsequent to further chemical optimization to improve the potency, metabolic stability, solubility, and genotype spectrum, HCV-796 displayed a remarkable preclinical profile and also has demonstrated clinical antiviral activity in HCV-infected patients. This report describes the in vitroin vivo correlation of HCV-796 by comparing the enzyme and replicon assays with the chimeric mouse model. The biological relevance of the in vitro replicon system and animal model in comparison with clinical activity also are discussed.

Figure 1.

Structure of HCV-796.


EC50, half maximal effective concentration; hAAT, Human α-1 antitrypsin; HCV, hepatitis C virus; IC50, median inhibitory concentration; IFN, interferon; NS5B, nonstructural protein 5B; PEG-IFN, pegylated interferon; RdRp, RNA-dependent RNA polymerase; SCID, severe combined immunodeficiency; uPA, urokinase plasminogen activator.

Materials and Methods

Replicon Assay

Clone A cells (derived from Huh-7 cells), plasmid pBB7 containing HCV genotype 1b, and BB7 replicon complementary DNA were licensed from APATH, LLC. Characteristics of cells and their propagation and drug susceptibility of the replicon-containing Huh7 cells to compounds was evaluated as described previously18 and as outlined in the supporting information.

NS5B Enzyme Assay

NS5B enzymes were prepared by isolating the gene from patient sera, cloning and expressing in Escherichia coli, then purifying the enzyme using column chromatography. Details of enzyme purification have been previously published.18 The RdRp assay is described in the supporting information.

Animal Care

Recipient animals [homozygous albumin (Alb)-urokinase plasminogen activator (uPA)/severe combined immunodeficient (SCID) mice] were housed in a virus-free/antigen-free environment, and cared for in accordance with Canadian Council on Animal Care (1993) guidelines. Experimental protocols were reviewed and approved by the University of Alberta Health Sciences Animal Welfare Committee.

Isolation and Transplantation of Human Hepatocytes

Ethical approval for human tissue use was obtained from the University of Alberta Faculty of Medicine Research Ethics Board, and informed consent was obtained from all donors. Hepatocytes were collagenase isolated from phosphate-buffered saline flushed normal human liver segments and Percoll purified using previously described techniques then injected into the inferior pole of the spleen of 5-day-old to 14-day-old uPA/SCID mice.15

Human α-1 Antitrypsin Analysis

Human α-1 antitrypsin (hAAT) analysis was used to confirm stable ongoing function of the human hepatocyte grafts and to determine whether any change in HCV titer was attributable to hepatocyte death or injury. Mouse serum was analyzed by sandwich enzyme-linked immunosorbent assay as previously described (supporting information).

HCV RNA Quantitation

Murine serum analysis was performed in blinded fashion by the Alberta Provincial Laboratory of Public Health (Edmonton, Canada) using the Cobas Amplicor HCV Monitor system (Roche Diagnostics, Laval, Canada). Lower limit of quantification was 600 IU/mL. HCV titers are reported in log10.

Chimeric Mouse Studies

Chimeric mouse studies were performed for assessment of antiviral activity. As such, no formal observational or laboratory studies or other safety assessments were performed in this model.

Experimental Conduct

Six weeks after hepatocyte transplantation, mice were screened for serum hAAT, and animals above a 100-μg/mL cutoff were inoculated by intraperitioneal injection with 100 μg genotype 1a HCV-laden human serum (approximately 2 × 105 copies/mL). Baseline HCV levels were obtained at 1 and 2 weeks after inoculation, and mice with titers above 2 × 104 copies/mL were allocated to experimental groups. Allocation sought to balance groups for HCV titers, hAAT levels, sex, and weight with decreasing priority.

Pharmacokinetics/Toxicity Study of HCV-796

Eight mice were used in brief pharmacokinetics and preliminary toxicity evaluation to assure protocols would achieve target serum levels of HCV-796 severalfold greater than maximal effective concentration (EC50) in replicon at trough without undue morbidity. HCV-796 was administered three times daily by oral gavage for 5 days with blood drawn for trough levels immediately before a second dose of drug and 8 hours after the last dose of drug.

Experiment 1

An initial proof of concept study used two groups of six mice receiving 50 mg/kg HCV-796 or vehicle (0.75% methylcellulose, 1% Tween-80) by oral gavage three times daily by mouth for 5 days. Sera for HCV titers were collected on days 0, 3, 5 and 12 (7 day post dosing) to demonstrate efficacy. Sera from the same draws were also evaluated for hAAT to confirm stable ongoing function of the human hepatocyte grafts, and as such to document that any changes in HCV titer were not attributable to hepatocyte death or injury.

Experiment 2

A confirmation study added as comparison a positive control treatment previously evaluated in the chimeric mouse model. Three groups of five mice received 50 mg/kg HCV-796 or vehicle three times daily by mouth or interferon alpha 2b (IFN) 1350 IU/g body weight daily by intramuscular injection for 10 days. Sera were collected for HCV titer and hAAT analysis at baseline and at days 3, 6, and 11

Experiment 3

Given that introduction of new anti-HCV therapies into clinical practice will likely occur in combination with interferon, the next study sought to evaluate the impact of such combined therapy in the mouse model. Four groups of mice were studied. Seven mice received HCV-796 30 mg/kg three times daily by mouth, seven mice received IFN alpha 2b, 1350 IU/g daily intramuscularly, and eight mice received a combination of HCV796 30mg/kg three times daily by mouth and IFN alpha 2b, 1350 IU/g intramuscularly. Five mice received vehicle three times daily by mouth as controls. Serum was collected on days 0, 3, 11, and 18 for HCV RNA and hAAT analyses.

Clinical Studies.

A phase 1a randomized, double-blinded, placebo-controlled, ascending, single-dose study of oral HCV-96 was conducted with administration of 25-mg, 50-mg, 100-mg, 250-mg, 500-mg, 1000-mg, and 2000-mg doses to groups of six healthy patients, with comparison with two placebo-treated patients per dose group.

A phase 1b clinical trial was conducted as an ascending multiple-dose study in treatment-naïve patients with chronic HCV. An initial group of 12 patients received HCV-796 monotherapy, with four receiving placebo. Subsequent groups of 12 to 14 patients received pegylated interferon alpha-2b (PEG-IFN; 1.5 μg/kg/dose) on days 1 and 7 combined with placebo or HCV-796 (100, 250, 500, or 1000 mg) every 12 hours from days 1 to 14. Genotype 1 infected 64% of combination therapy patients. Mean viral titers were more than 6 log10 IU/mL HCV RNA.

A phase 2 randomized, open-label study evaluating safety, tolerability, antiviral activity, and pharmacokinetics of HCV796 in combination with pegylated interferon plus ribavirin versus IFN/ribavirin alone in treatment-naïve HCV genotype 1 patients will be the subject of a subsequent report.


In Vitro Studies

Biochemical assays of polymerase inhibition yielded median inhibitory concentration (IC50) values of 0.01 to 0.14 μM for genotype 1, and 0.6 to 1.7 for genotypes 2, 3, and 4 (Table 1). In the in vitro subgenomic replicon assay, HCV-796 demonstrated reduction of steady-state levels of HCV RNA with an EC50 of 5 nM against genotype 1a and 9 nM against genotype 1b (Table 1). Treatment was associated with a concomitant decrease in the level of HCV protein (EC50 for genotype 1a = 19 nM, and for genotype 1b = 14 nM, Table 1). Multiple treatments with HCV-796 in replicon-containing cells resulted in a 3 to 4 log10 reduction in HCV RNA levels; as compared with 2 to 3 log10 reductions with interferon (Fig. 2). No adverse impact was observed for cellular functions at effective antiviral concentrations; a therapeutic Index of greater than 1100 was demonstrated (Table 1). These findings showed that HCV-796 is a potent and specific inhibitor of the HCV polymerase in vitro. Replicon resistance patterns associated with treatment with HCV796 have been previously published.19

Table 1. In Vitro Biochemical and Intracellular Activities of HCV-796
NS5B EnzymeIC50 (μM)
  • *

    N = number of isolates.

  • Therapeutic Index is determined by comparing IC50 values of HCV-796 in GAPDH RNA versus EC50 for HCV replicon RNA.

Genotype 1a0.02–0.14 (n = 3)*
Genotype 1b0.01– 0.04 (n = 6)*
Genotype 21.7 (n = 1)*
Genotype 30.22 (n = 1)*
Genotype 40.57 (n = 1)*
Replicon RNAEC50 (nM)
Genotype 1a, H775 ± 2
Genotype 1b, BB79 ± 4
Replicon ProteinEC50 (nM)
Genotype 1a19 ± 14
Genotype 1b14 ± 15
Therapeutic Index>1100
Figure 2.

Effect of HCV-796 and interferon on HCV RNA. Clone A cells were treated with HCV-796 every 2 to 3 days (a total of 8 times) over a 22-day period. In this experiment, cells were treated with either HCV-796 or IFN-alpha 2b at the concentrations indicated. As a control, cells were passaged in parallel with the same medium without compound. The HCV RNA from the replicon-containing cells was prepared and quantified as described in Materials and Methods. Each point represents an average of three cell samples; each was analyzed in duplicate, with the associated standard error bars. Values are presented as HCV RNA (copy number) normalized to ribosomal RNA (μg total RNA) (Y-axis).

In Vivo Studies

HCV-796 had a half-life in plasma of 2 hours in the chimeric SCID mice containing a uPA transgene linked to an albumin promoter that targets the liver. Mean plasma concentration at 8 hours post-dosing for a single 50-mg/kg dose was 40 ng/mL, whereas 8 hours post-dosing of the last dose after 5 days of 50 mg/kg every 8 hours was 100 ng/mL (Fig. 3). These plasma levels after acute and chronic dosing were 18-fold and 45-fold, respectively, above the EC50 in the replicon system (Fig. 3), and 5-fold and 12.5- fold above the human serum adjusted EC50 of 8 ng/mL relative to the 1a replicon.

Figure 3.

Plasma HCV-796 (ng/mL) in mice treated orally with either a single dose or multiple doses of HCV-796 (50 mg/kg, every 8 hours by mouth for 5 days).

The pilot study with HCV-796 at 50 mg/kg three times daily for 5 days resulted in a 2.02 ± 0.55 log10 decrease in HCV titer with one mouse below the level of detection, whereas levels in the control mice were relatively stable (0.26 ± 0.16 log10 decline, Table 2). Seven days after stopping treatments, control mouse HCV titers were unchanged, whereas the titers in the HCV-796 group rebounded to within 1.16 log of baseline, supporting a direct antiviral effect of HCV-796 for the decline during treatment (data not shown). Human alpha-1 antitrypsin levels were unchanged in both groups over the duration of treatment (data not shown).

Table 2. Summary of Anti-HCV Results with HCV-796 Treatment in Chimeric Mice
GroupNo. of AnimalsGeometric Mean HCV Titer Reduction from Baseline±SE (log10IU/mL)*PValue
  • SE, standard error.

  • *

    End of treatment measurements: day 6 (5 days' treatment) for experiment 1, day 11 (10 days' treatment) for experiment 2.

  • P values comparing with vehicle control group were determined using two-sided (unpaired) t tests, with the error term for the t statistics based on the error term from a one-way analysis of variance.

  • For calculation purposes, HCV titer values in animals that have HCV titers below the limit of quantification (600 IU/mL, or 2.78 log10 IU/mL) were assigned a value of 2.78 log10 IU/mL.

  • §

    One animal was sick and euthanized on day, day 6 data carried forward for calculation of mean and standard error.

  • P < 0.05 when comparing HCV-796 and IFNα-2b.

Experiment 1   
Vehicle60.26 ± 0.16
HCV-796 (50 mg/kg, every 8 hours, by mouth, 5 days)62.02 ± 0.550.01
Experiment 2   
Vehicle§40.35 ± 0.13
HCV-796 (50 mg/kg, every 8 hours, intramuscularly, 10 days)51.78 ± 0.270.009
IFNα-2b (1350 IU/g, daily for 10 days)51.11 ± 0.20.04

The confirmatory study included both negative and positive controls. Figure 4 and Table 2 demonstrate confirmation of the pilot study data with a 1.78 ± 0.27 log10 decline in HCV titer after 10 days' treatment with HCV-796 as compared with a 1.11 ± 0.2 log10 reduction with interferon (P < 0.05 HCV-796 versus IFN). Vehicle-treated control mice again demonstrated relative stability of HCV titers with a mean of 0.35 ± 0.13 log10 drop at day 11 (P = 0.009 HCV-796 versus control). One vehicle-treated mouse was withdrawn from the study before day 10; day 6 data were carried forward for this animal. Four of five mice in both HCV-796–treated and IFN-treated groups demonstrated ongoing decline in viral titers between days 5 and 10 of treatment (Fig. 4). One mouse in the HCV796 group showed a minor rise in HCV titer between days 5 and 10 from log 5.32 to log 5.48. A subsequent dose–response study with four groups of five mice each receiving 0, 30, 75, or 125 mg/kg of HCV-796 by gavage three times daily confirmed earlier studies showing that maximal impact occurred at approximately 50 mg/kg three times daily with a 0.9 to 1.0 log10 drop in HCV titer after 3 days' therapy and 1.4-1.6 log after 10 days (data not shown).

Figure 4.

Antiviral activities of HCV-796, interferon α-2b, and vehicle control. Three groups of five mice received 50 mg/kg HCV796 or vehicle three times daily by oral gavage or a positive control treatment with interferon α2b (IFNα-2b) 1350 IU/g body weight daily by intramuscular injection for 10 days.

Given the potential that new small molecule HCV inhibitors will be introduced to clinical practice as combination therapies with interferon, such a combination study was carried out in the chimeric mouse model. Mice receiving a suboptimal dose of HCV-796 at 30 mg/kg three times daily, IFN at 1350 IU/g daily, or both, were compared with vehicle-treated controls (Fig. 5). Consistent with earlier studies, treatment with a reduced dose of HCV-796 was associated with a reduction in HCV titers of 0.66 log10 by day 11; IFN yielded a 1.35 log decline by day 11; vehicle-treated mice rose 0.12 log10. The group receiving the combination of interferon and a suboptimal dose of HCV-796 demonstrated at least additive activity with a decline in HCV titer of 1.17 log10 after 3 days' therapy and continuing declines to a 2.44 log10 drop after 10 days' therapy (P = 0.0001 HCV796 /IFN versus vehicle and P = 0.001 HCV-796/IFN versus IFN alone).

Figure 5.

A suboptimal dose (30 mg/kg) of HCV-796 was delivered three times daily by oral gavage to six HCV-infected chimeric mice for 10 days alone or in combination with subcutaneous interferon alpha 2b at 1350 IU/g body weight and compared with interferon alone or vehicle control. Additive (at minimum) impact of the small molecule therapy and interferon combination was observed.

Serum levels of human alpha-1 antitrypsin remained relatively stable throughout dosing periods in all experimental series, demonstrating that the human hepatocyte grafts remained viable and functional, supporting the antiviral effects of HCV-796 and IFN as responsible for the decreases in viral titers (data not shown).

Clinical Studies

Preliminary phase 1 clinical data have been reported for dose-ranging trials in which HCV-796 was administered as single doses in healthy volunteers20 and as multiple doses for up to 14 days in HCV-infected patients.21, 22

Phase 1a Studies.

HCV-796 was generally well tolerated with no serious treatment-emergent adverse events, and no patient discontinuation because of adverse events. The area under the curve reached a plateau of 18,800 ± 7754 ng/hour/mL at the 1000-mg dose.

Phase 1b Studies.

In genotype 1 patients, mean reduction from baseline at day 7 ranged from 1.5 to 2.3 log10 in the combination group versus 0.9 log10 for PEG-IFN alone. At day 14, the comparison was 2.6 to 3.2 log10 (combination therapy) versus 1.3 log10. Nongenotype 1 patients demonstrated reductions from baseline of 2.8 to 3.5 log10 and 3.5 to 4.8 log10 on days 7 and 14, respectively, in comparison with 1.5 and 2.6 log10 for PEG-IFN alone. One third of combination patients dosed with HCV-796 at 250 mg twice daily or over achieved undetectable viral levels at day 14 (limit of detection, 50 IU/mL HCV RNA). Combination therapy was well tolerated; adverse events were mild to moderate in severity and were characteristic of the known side effects of interferons. No dose-limiting toxicities were identified.


Whereas drug discovery for HCV has been challenging since elucidation of the RNA sequence 17 years ago,6 advances in HCV research systems have contributed to the recent acceleration of antiviral developments for HCV infection. The replicon system has become an integral part of development of anti-HCV drugs that target the viral replication machinery. Although therapeutics that target other aspects of the HCV life cycle such as cell entry, packaging of the RNA strand and envelope, as well as cellular egress are not amenable to study in the replicon, recent developments such as the pseudoparticle system23 and the cell culture systems using the Japanese fulminant hepatitis virus24–27 promise in vitro approaches to such study. Although it is not yet clear how closely results with the unique genotype 2a Japanese fulminant hepatitis virus may follow the typical chronic HCV disease, a modified full-length genome has produced virus particles that are infectious in cell culture, neutralizable by anti-E2 antibodies, dependent on CD 81 for cell entry, and inhibited by interferon-alpha and HCV-specific antiviral compounds.

The chimpanzee, the most established animal model of HCV infection,28 is far from ideal because of the extreme cost, limited access, and significant ethical challenges. Development of alternative small animal models of HCV disease has proved challenging. The trimera mouse29 and tree shrew30 produced relatively brief infection with modest viral titers. The immunodeficient transgenic SCID-uPA mouse model with a chimeric human/mouse liver as used here supports long-term HCV infection at clinically relevant titers.15 These SCID mice carry a tandem array of four copies of a uPA transgene linked to an albumin promoter that targets the transgene to liver. Transgene expression in the liver leads to urokinase overproduction associated with hypofibrinogenemia, accelerated hepatocyte death, and a phenotype of subacute liver failure that is associated with a strong stimulus to hepatocyte proliferation.31–33 The SCID phenotype abrogates rejection so that human hepatocytes transplanted to these mice by intrasplenic inoculation transmigrate to the liver and undergo dramatic expansion resulting in substantial repopulation of the mouse liver with human hepatocytes. Inoculation of these mice with HCV-positive human serum results in mice that are capable of supporting HCV at titers of 104 to 107 copies/mL for periods beyond 1 year.

One of the early lead molecules developed in the Wyeth/Viropharma collaboration of polymerase inhibitors was the pyrranoindole series molecule, HCV-371. HCV371 had an IC50 of 0.5 μM against a genotype 1b NS5B enzyme, and an EC50 of 2.2 μM against the HCV subgenomic replicon.34In vivo studies in the chimeric mouse model failed to demonstrate significant decline in HCV titers after 2 weeks' therapy.17 The compound failed to achieve efficacy targets in clinical trials and was dropped from further clinical evaluation (Viropharma press release, July 15, 2003, Viropharma web site).

A subsequent molecule, HCV-086, was the first benzofuran series compound that exhibited potent replicon activities (EC50 = 200 nM, unpublished data); The earlier compounds from this series had poor metabolic stability (t1/2 < 20 minutes in human and rat S9 liver extracts) and minimal oral bioavailability (<5% in rats). The clinical predecessor of HCV-796, HCV-086, demonstrated improved replicon potency and metabolic stability (in vitro t1/2 > 200 minutes in rat S9 and microsomes). Pharmacokinetics was challenging in the mouse model because of rapid elimination. In a clinical phase I trial, HCV-086 was found to be well tolerated and to possess a good safety profile, but modest clinical efficacy halted further development (ViroPharma press release, March 10, 2005, ViroPharma web site).

The current report documents the in vitro potency of HCV796. Activity against diverse HCV genotypes was demonstrated in enzyme inhibition assays (Table 1). In tissue culture, submicromolar IC50 activities were observed in genotypes 1a (H77), 1b (BB7) (Table 1), and Con 1 (unpublished data). Lower enzyme activity was demonstrated in vitro against an enzyme isolate from a genotype 2a source. Although not an integral part of these development studies, HCV796 has been studied in a genotype 2 infectious culture system and has demonstrated an EC50 of approximately 45 and 50 nM in independent evaluations (personal communication, Dr. T-I Lin, Tibotec Pharmaceuticals)

With improved in vitro and pharmaceutical properties, HCV-796 next demonstrated a 2-log10 reduction of HCV RNA level in the chimeric mouse after 5 to 10 days of dosing, and further reductions in viral titer when combined with interferon. Because of the favorable outcomes from enzyme inhibition assays, the replicon system, and the SCID/uPA chimeric mouse model of HCV infection, HCV-796 was advanced to clinical trial.

Preliminary Phase 1 clinical data have been reported for single ascending dose studies in healthy volunteers20 and multiple ascending dose (14-day) trials in HCV-infected patients.21, 22 These studies demonstrated favorable pharmacokinetics and tolerability up to 14 days of dosing, with no dose-limiting toxicity. At the highest doses, treatment with HCV-796 led to a mean reduction in HCV titer of 1.4 log10 at day 4.21 Co-administration with PEG-IFN reduced HCV titer up to 3.5 log10 at day 14, compared with 1.7 log10 with PEG-IFN alone.22 These studies demonstrated additive effect of HCV-796 with PEG-IFN across multiple genotypes of HCV in treatment-naïve subjects with no dose-limiting toxicities over 14 days.

Triple therapy with HCV-796/PEG-IFN/Ribavirin for longer than 8 weeks in the phase 2 studies resulted in identification of a safety issue: elevated liver enzymes in a subset of 8% of patients (Viropharma press release, August 10, 2007, Viropharma web site). Similar changes were identified in only 1% of control patients. Further administration of HCV-796 was suspended, and patients were given the opportunity to continue dosing of PEG-IFN and ribavirin, to allow evaluation of long-term outcome. Further studies have been suspended (Viropharma press release, April 16, 2008, Viropharma web site). Reports of detailed outcomes of the phase 2 studies are pending.

The in vitro, in vivo, and clinical results from the polymerase inhibitors described above clearly demonstrate the predictive value of the chimeric mouse model that bridges the in vitro activities (enzyme inhibition and intracellular antiviral activities in replicons) with the clinical outcomes. The uPA chimeric mouse model of HCV infection combines an ability to look at all aspects of the HCV life cycle as targets for therapy, and the advantages of an in vivo system. Although complex and demanding, the chimeric mouse model has proven to correlate directly with clinical outcomes of anti-HCV therapy. A series of validation studies have been conducted that demonstrated unequivocal correlation of outcomes in the mouse model and clinical trials with interferon alpha, and small molecule inhibitors BILN206135 and HCV-796, among others.17

HCV has an extremely high rate of replication and an error-prone replication mechanism. With an estimated 1012 viral copies produced daily in an infected patient, and with an error rate in the HCV polymerase of 1 in 104 to 105 replications, as many as 108 mutant viruses may be produced daily in those infected, potentially representing virtually all possible variants.36, 37 As such, a wide range of potentially resistant mutants exists in most infected patients, and specific mutants may easily emerge under the selective pressure of drug therapy. Combinations of drugs with different mechanisms of action could offer a promising strategy in minimizing the risk of drug resistance. In addition to HCV-796, several novel therapeutic compounds have now yielded reductions in HCV titers in clinical trials. Among them, NS3/4A protease inhibitors, telaprevir and boceprevir, both demonstrated clinical efficacy with greater than 2 log10 HCV RNA reduction.38, 39 The combination of HCV-796 with IFN in the chimeric mouse model demonstrated important advantages to such a strategy (Fig. 5), and consistent findings were observed in subsequent clinical studies. Animal studies using a combinations of small molecules may yield invaluable insights to the clinical development of antiviral cocktails for HCV infection.

Unlike other chronic viral infections such as human immunodeficiency virus and hepatitis B virus, HCV is not known to establish a sequestered reservoir of viruses. With intensive therapy of HCV using a combination of effective small molecule anti-virals with or without interferon, a cure for HCV infection may be an eminently reachable goal. The efficacy outcomes in animal models, and initial clinical trials outlined in this report, suggest that NS5B polymerase inhibitors such as HCV-796 or related compounds may play a significant role in future strategy for HCV treatment.


The authors thank Jackie Boudreau for support with animal care, Lin-Fu Zhu and Chan-Zhi Yao for microsurgical support, and Dr. Anne Deatly for editorial review.