Evaluation of VCH-759 monotherapy in hepatitis C infection

Evaluation of VCH-759 monotherapy in hepatitis C infection

Jnl of Hepatology Articles in Press

Curtis Cooper1, Eric J. Lawitz2, Peter Ghali3, Maribel Rodriguez-Torres4, Frank H. Anderson5, Samuel S. Lee6, Jean Bédard7, Nathalie Chauret7, Roch Thibert7, Isabel Boivin7, Olivier Nicolas7, Louise Proulx7

Received 16 October 2008; received in revised form 17 February 2009; accepted 19 March 2009. published online 23 April 2009.

Corrected Proof

ABSTRACT
Background

VCH-759 is a non-nucleoside inhibitor of HCV RNA-dependent polymerase with sub-micromolar IC50 values versus genotype 1a/1b replicons.

Methods

The antiviral activity, pharmacokinetics and tolerability of VCH-759 administered as monotherapy for 10 days with a 14 day follow-up period were evaluated in 31 treatment-naı¨ve genotype 1 participants. Three cohorts received: 400mg thrice (t.i.d.), 800mg twice (b.i.d.), 800mg t.i.d or placebo.

Results

VCH-759 was well tolerated with the most frequent adverse event being gastrointestinal upset in both the active and placebo groups attributable, in part, to the dosing vehicle. VCH-759 was rapidly absorbed and trough plasma levels were at or above the IC90 (non protein-adjusted) for all dosing regimens. The mean maximal decrease in HCV RNA log10 (IU/mL) was 1.97, 2.30 and 2.46 for 400mg t.i.d., 800mg b.i.d. and 800mg t.i.d. doses. Viral polymerase genotypic sequencing revealed emergence of HCV variants in a majority of participants that coincided with on-treatment viral rebound.

Conclusions

VCH-759 was well tolerated and achieved a2 log10 decline in HCV RNA with 800mg b.i.d. and t.i.d doses. In a subset of participants, viral rebound was observed and associated with resistant variants. This data supports further evaluation of VCH-759 in combination with interferon–ribavirin treatment.

Introduction

The hepatitis C virus (HCV) is a 9.6kb positive strand ribonucleic acid (RNA) virus of the flaviviridae family, genus hepacivirus. The genome comprises a single open-reading frame coding for a ∼3000 amino acid polypeptide which is further processed into individual structural (core, E1 and E2) and non-structural (NS2, NS3, NS4A, NS4B, NS5A and NS5B) proteins by host and viral NS2 (zinc-dependent) and NS3 (chymotrypsin-like) proteases [1]. The non-structural proteins function as enzymes or accessory factors involved in genomic replication. The viral replication strategy is similar to that of other positive strand viruses, with the initial synthesis of a replicative intermediate negative strand RNA by the NS5B RNA-dependent RNA polymerase. This negative strand RNA then serves as a template for genomic RNA production. The NS3 and the NS5B enzymes are key targets for anti-HCV therapy, as they are essential for HCV replication and infectivity [2], [3], [4], [5]. The NS5B enzyme has the characteristic right-handed “fingers-palm-thumb” domain of polymerases [4]. The active site, which resides in the palm region, contains the conserved GDD motif of polymerases and is partially enclosed by the finger and thumb domains.

VCH-759 is a novel substituted thiophene-2-carboxylic acid derivative non-nucleoside inhibitor of HCV NS5B polymerase genotype 1a and 1b. This compound inhibits the NS5B (IC50 1a=0.41μM and IC50 1b=0.38μM) by binding to an allosteric site in the ‘thumb’ domain situated ∼35Å from the active site. X-ray crystallography studies suggest that inhibition of RNA synthesis initiation may result from enzyme conformational changes induced by occupancy of the compound binding site [6]. VCH-759 is active against HCV sub-genomic replicon in Huh-7 cells (IC50≈0.3μM for both 1a and 1b genotypes). The compound is also selective for the HCV NS5B polymerase relative to human DNA polymerases α, β and γ (IC50>100μM). VCH-759 has a good in vitro therapeutic index (TI) (CC50/IC50)>600 and non-clinical safety profile.

Given an estimated 170 million people worldwide infected by HCV and limitations of currently available interferon-based therapies, there is an important unmet need for novel, more effective, more convenient, and better tolerated anti-HCV treatments [7]. Hence, we evaluated the antiviral activity, safety, tolerability and HCV variant selection of VCH-759 administered as monotherapy for 10 days in HCV genotype 1a or 1b-infected treatment-naı¨ve participants. Pharmacokinetic (PK) profile, plasma HCV RNA kinetics and correlation between VCH-759 plasma trough levels and HCV RNA reduction were also assessed.

Discussion

Rapid HCV virologic clearance is well recognized to predict high rates of sustained virological response (SVR) in recipients of pegylated interferon and ribavirin [11], [12], [13]. However, only approximately 15% of genotype 1-infected individuals achieve rapid virological response (RVR) (i.e., undetectable HCV RNA at week 4 of treatment) [14]. A rapid, dose-related, clinically significant antiviral response was observed with VCH-759 monotherapy which appeared similar for both genotypes 1a and 1b. As observed with other specifically targeted antiviral therapy for Hepatitis C (STAT-C), this potent antiviral activity may be utilized in combination with current interferon-based HCV therapies to potentially improve treatment outcomes of HCV therapy.

Repeated oral administration of VCH-759 at three different dosing regimens for 10 days was well tolerated without severe adverse events. This may represent an advantage over several other compounds in development or previously abandoned due to toxicities [15], [16], [17], [18]. Gastrointestinal side effects, although frequent, did not compromise activities of daily living. Formulation development is on-going to identify a dosage form that will minimize or eliminate polyethylene glycol and/or Solutol, the suspected causes. No liver-related biochemical abnormalities were identified with the lower doses (e.g., 400mg t.i.d. and 800mg b.i.d.) of VCH-759 evaluated. Once the alternative formation tolerability and PK has been established additional controlled studies with larger subject populations for longer durations of dosing will help evaluate the effect of VCH-759 on liver chemistries.

Viral genotypic sequencing of the NS5B polymerase revealed HCV resistant variants in most participants [19]. In all cohorts, clinical breakthrough associated with resistant variants was observed in a subset of participants despite VCH-759 plasma trough concentrations that were at or above the replicon IC90 (non protein-adjusted) for the duration of treatment with each dosing regimen. These variants most likely originate from pre-existing minor populations and the in vitro fitness of these variants, in absence of VCH-759, was significantly reduced compared to the wild-type. In addition, the in vitro drug resistance studies have demonstrated that these variants remain sensitive to interferon-α 2a and ribavirin [20]. Therefore, the rapid evolution of resistance and lack of drug cross-resistance are relevant to the study design of future trials with pegylated interferon and ribavirin.

Although interferon-based therapeutic efficacy differs by genotype, successful outcomes are possible for all. This may not be the case for HCV polymerase and protease inhibitors alone. The lack of any virological response in the genotype 6-infected individual dosed in the study raises important issues regarding cross-genotype efficacy of small molecules to treat HCV. The current clinical research approach that emphasizes the treatment of genotype 1-infected participants may require re-thinking. Not only may results from these studies not be applicable to other genotypes but compounds without effect in genotype 1 may be abandoned prior to investigating the therapeutic potential in non-genotype 1 infection.

We have shown that VCH-759 was well tolerated, well-absorbed and virologically potent over a short period of observation. Additional, longer-term evaluation in combination with interferon-based therapy is warranted.

3. Results Of 34 randomized participants, 32 completed the study and were evaluated for compound safety. Two participants dropped out before receiving any treatment: one participant no longer met the inclusion/exclusion criteria and one subject withdrew consent. As a single genotype 6-infected participant was erroneously enrolled, 31 were ‘efficacy-evaluable’.Table 1 presents the principal key demographic characteristics of the participants.

Table 1.Characteristics of participants at Day 1 (or Baseline) with evaluable efficacy data.

Placebo (n=9) 400mg t.i.d. (n=8) 800mg b.i.d. (n=5) 800mg t.i.d. (n=9)
Average Age (±SD) (range) (yr) 45.2±1.92 (41–52) 46.5±5.35 (39–54) 35.2±9.85 (25–47) 45.0±13.10 (20–61)
Average BMI (±SD) (kg/m2) 26.7±3.48 23.2±1.67 25.1±4.91 26.6±4.99
Male/Female (n) 7/2 6/2 2/3 6/3
Genotype 1a/1b (n) 8/1 7/1 4/1 6/3
Race (n)
White 5 6 2 5
Black 1 – – –
Hispanic 3 – 3 2
Other – 2 – 2
Average viral RNA concentration (log10 IU/mL) 6.3 6.4 5.8 6.4
Median ALT (U/L) (range) 73 (20–186) 48 (27–203) 71 (43–192) 62 (31–177)

3.1. Safety and tolerability
The compound was well tolerated without severe adverse events. The overall incidence of adverse events (AE) was similar across treatment groups (Table 2). There were no discontinuations due to AEs or death. Adverse events were experienced by 30 (88%) participants. Mild and moderate AEs occurred in 27 (79%) and 12 (35%) participants, respectively. The most frequent treatment-emergent AEs in active and placebo recipients, respectively, included diarrhea (79% vs 56%), headache (30% vs 22%), flatulence (24% vs 11%), dyspepsia (19% vs 11%) and fatigue (22% vs 0%). We believe that the formulation vehicle (an aqueous mixture of polyethylene glycol 400 and Solutol® HS15) likely contributed to the gastrointestinal (GI) adverse events observed in this study. Of note, GI adverse events did not negatively impede daily activities. All AEs were resolved by the end of follow-up. No QT prolongation or other electrocardiographic abnormalities were observed.

Table 2.Summary of safety data for all cohorts and placebo.

Placebo n=9 400mg t.i.d. n=9 800mg b.i.d. n=5 800mg t.i.d. n=9
Any adverse events (number of subjects) 6 7 4 9
Gastrointestinal disorders 6 7 4 9
Diarrhea 5 5 4 9
Flatulence 1 2 2 1
Dyspepsia 1 1 0 4
Nausea 1 1 2 0
Abdominal pain 0 2 0 1
Vomiting 0 0 1 3
General disorders 0 4 1 5
Fatigue 0 2 1 2
Chills 0 1 0 0
Nervous system disorders 2 4 0 5
Headache 2 3 0 5
Somnolence 1 0 0 1
Vascular disorders 0 0 0 2
Flushing 0 0 0 2

Clinical laboratory evaluations revealed no clinically significant effect of VCH-759 treatment on complete blood count, blood chemistry or urinalysis. There were no clinical signs or symptoms of liver toxicity in any participants during dosing or in the post-dosing period. On average, ALT levels decreased with VCH-759 administration in all cohorts. Increased total blood bilirubin occurring after VCH-759 dosing was observed in one participant (3%) in the 800mg t.i.d. cohort. In this case, the total bilirubin was slightly above the upper limit of normal at baseline (i.e., before dosing). The bilirubin doubled from Day 1 (1.2mg/dL) to Day 2 (2.4mg/dL) of dosing but subsequently returned to pretreatment levels by Day 7 despite continued dosing of VCH-759. The increased bilirubin was predominantly unconjugated. This increase in bilirubin was accompanied by an increase in ALT from 113 U/L at screening and Day 1 to a maximum of 231 U/L on Day 6. The ALT value returned to baseline levels by Day 17. In another recipient of 800mg t.i.d. dosing, a single ALT elevation (154 U/L) on Day 4 was observed, however, ALT on Days 3 and 5 were 80 and 81, respectively. All ALT values for this patient were in the normal range (range of 60–90 U/L), both at baseline and during the dosing period. No change in bilirubin was observed in this patient.

3.2. Pharmacokinetics
VCH-759 was rapidly absorbed with a time at which plasma concentration was maximal of 1.62, 1.90 and 2.77h and peak plasma levels of 1857, 4627 and 4857ng/mL at Day 1 for the 400mg t.i.d., 800mg b.i.d. and 800mg t.i.d. doses, respectively (Table 3). VCH-759 did not accumulate significantly following 10 days of twice- or thrice-daily administrations as shown by the ratio of approximately 1 when comparing AUCDay10 to AUCDay1. Rate (Cmax) and extent of exposure (AUC) were variable and appeared to be more than proportional between 400 and 800mg administered thrice-daily. Tmax was delayed by approximately 1h with the 800mg dose compared to the 400mg dose. Elimination half-life, clearance and volume of distribution appeared to be similar between the three cohorts.

Table 3.Pharmacokinetic parameters (±SD) of VCH-759 after administration of different dosing regimens on Days 1 and 10 of treatment.

Day 1 Day 10 AUCDay10/AUCDay1
Cmax(ng/mL) Tmax(h) AUC0−6ha (ng·h/mL) t1/2(h) Cmax(ng/mL) Tmax(h) AUC0−6ha (ng·h/mL) t1/2(h)
400mg t.i.d. 1857±773 1.62±0.65 6553±2518 2.65±0.99 2130±1185 1.16±0.37 7181±3799 2.87±0.59 1.10
800mg b.i.d. 4627±1689 1.90±1.02 20706±13395 2.76±0.75 4351±2468 2.20±1.79 21806±19926 2.95±0.40 1.05
800mg t.i.d. 4857±4107 2.77±2.06 19286±17173 3.02±1.29 5740±5917 2.31±1.66 22706±26035 2.80±0.66 1.18
a
Corresponds to AUC0−12h for b.i.d. cohort.

VCH-759 plasma trough concentrations between Days 2 and 11 were at or above the replicon IC90 (420ng/mL, 0.85μM non protein-adjusted) for the duration of treatment for each dosing regimen. There was no clear correlation between baseline HCV RNA levels and viral load decline during the dosing period in active compound recipients. However, this analysis was limited by the small sample size.

3.3. Antiviral activity
All participants who received VCH-759 achieved a rapid initial decline of approximately 1.5log10 in plasma HCV RNA level during the first two days of dosing as shown in Fig. 1B–D. The change from baseline in plasma HCV RNA to nadir was 2log10 or greater in the three active treatment groups. The 800mg t.i.d dose produced the greatest change in viral load (mean: −2.46 log10, range: −1.2 to −3.4) followed by the 800mg b.i.d. (mean: −2.30log10, range −1.5 to −2.9) and 400mg t.i.d. (mean: −1.97 log10, range: −1.4 to −2.6) dosed cohorts (Fig. 1B–D). The change from baseline to nadir was statistically significant in all groups (400mg t.i.d., p=0.0002; 800mg b.i.d., p=0.0021; 800mg t.i.d., p=0.0004). Viral rebound, defined as >0.75log10 increase from nadir before Day 11, was observed in 5 of 8 (63%), 5 of 5 (100%) and 6 of 9 (67%) participants in the 400mg t.i.d., 800mg b.i.d. and 800mg t.i.d. cohorts, respectively. There was no apparent difference in HCV RNA decline between subtypes 1a and 1b. However, low numbers of 1b (n=6) limited our ability to fully investigate this issue.

Fig. 1. Individual plasma HCV RNA reduction profiles following the administration of placebo (A) or VCH-759 400mg t.i.d. (B), 800mg b.i.d. (C) and 800mg t.i.d. (D). (note: shaded area represents treatment period; solid lines are patients with HCV genotype 1a; dotted lines are patients with HCV genotype 1b).

3.4. Genotypic analysis of HCV variants
Initially, the entire NS5B gene was sequenced (population sequencing of overlapping PCR fragments) at baseline (Day 1), end of dosing (Day 11), and follow-up days (Days 17 and 24) of the study for each participant treated with VCH-759. Amino acids 419, 423, 482 and 494 within the NS5B gene were consistently found to present more amino acid variations since mixed nucleotide populations were found at these positions. Amino acid changes outside the VCH-759 binding pocket were sporadic and not considered potential resistance variations. In order to obtain a quantitative evaluation of the HCV variants selected during VCH-759 dosing, a clonal sequencing analysis method of the VCH-759 binding pocket (amino acids 340–539) was developed and validated using mixes of known amounts of recombinant RNAs. An average of 30 clones per post-treatment timepoint, per participant, were analyzed which theoretically allows a 75% probability of detecting a minority clone present at a frequency of 5% within a population [10]. Amino acid changes at specific positions within the VCH-759 binding pocket, other than L419, M423, I482 and V494, with an increase of less than 5% frequency from baseline (Day 1) to end of study (Day 24) were regarded as non significant (data not shown). Genotypic analysis of all samples from patients treated with VCH-759 collected at Days 1, 11, 17 and 24 revealed the most prevalent changes at position 423 (M423T/V/I) (Table 4). The number of sequenced clones with any variation at position L419, M423, I482 or V494 was found to be 1.4% at the baseline, 67.5% at the end of treatment, 60.4% at Day 17 and 53.1% at the end of the study (Day 24). Increase of amino acid changes up to 6.4%, 50.4% and 6.2% from baseline was observed at positions 419, 423 and 482, respectively, at the end of treatment (Table 4).

Table 4.Clonal analysis of VCH-759 binding pocket (% of each variation or dual variations from all the treated patients).

Variant Day 1 Day 11 Day 17 Day 24
419V 0 0 0.2 0
419M 0 6.4 6.0 6.0
423T 1.4 32.1 27.4 19.6
423V 0 15.9 15.7 17.7
423I 0 2.4 6.9 8.2
482L 0 5.8 1.5 1.6
482V 0 0.4 0 0
482T 0 0 0.3 0
423T/482L 0 0.4 0.5 0
423I/482L 0 0.2 0.2 0
419M/482T 0 0 0.2 0
419M/494I 0 3.2 1.4 0
494A 0 0.2 0.2 0
494I 0 0.4 0 0
TOTAL 1.4 67.5 60.4 53.1
Total number of clones 145 520 598 616

The M423T variant was the predominant variant found at Days 11, 17 and 24 (Table 4) but decreased over time (e.g., Day 11 (32.1%) vs Day 24 (19.6%)). The frequency of M423V and L419M variations was relatively stable during the study: 15.9% and 17.7% for M423V and 6.4% and 6.0% for L419M at Days 11 and 24, respectively (Table 4). Some double variants were also detected at very low frequency. M423T/I482L, M423I/I482L, L419M/I482T and L419M/V494I variants were detected at 0.5% and 0.6%, 0.2% and 0.2%, 0% and 0.2% and 3.2% and 1.4% frequency at Days 11 and 17, respectively (Table 4). No double variants were detected at Day 24.

3.5. Phenotypic analysis of HCV variants
Since amino acids L419, M423, I482 and V494 were found to be enriched in about two-thirds of all sequenced clones at the end of treatment, their actual role in conferring resistance to VCH-759 was investigated. Recombinant replicons with single variation at positions L419, M423, I482 and V494 as well as various combinations of the two simultaneous variations were generated and stable replicon cell lines were obtained for each one of these variations.

The activity of VCH-759 against the wild-type replicon cells (genotype 1a or 1b) was compared to the variant replicon. A 1.6- to 54-fold increase of the compound’s IC50 was recorded for the replicons harboring single variations at positions L419, M423, I482 or V494 whereas up to 462-fold reduction in susceptibility (IC50) to the compound was noticed for the replicon harboring the M423I/I482L double variation. In the majority of participants who experienced a viral rebound, variants were found to remain present up to 13 days despite an in vivo high HCV replication rate. Some of the variations found in participants treated with VCH-759 were not previously selected in vitro using the replicon system (e.g., V494A). However, it should be noted that the 1a genotype was over-represented in treated patients (17 treated patients) compared to genotype 1b (5 treated patients) and all the in vitro studies were performed with the 1b sub-genomic replicon.

A transient replicon assay using variants (L419M, M423I, M423T, M423V, I482L, I482T, I482V, V494A, L419M/I482T, M423I/I482L, M423T/I482L) engineered into the wild-type sub-genomic replicon 1b was chosen for characterization of their replication capacities. All variants were associated with in vitro reduced replication fitness when compared to the parental wild-type replicon in the absence of VCH-759. The degree of reduced fitness ranged broadly depending on the specific mutation. The alanine substitution at position 494 engineered in the 1b replicon backbone was found to confer the lowest replicative fitness (0.98%). The L419M and I482L were associated with higher replication fitness with, respectively, 66.3% and 65.6% when compared to the wild-type.

Overall, these data suggest that resistant HCV variants contributed to the viral rebound observed in participants administered VCH-759 with each dosing schedule.

There was no identified relationship between the frequency of emergence of the variants and their nature and the virus subtypes. However, there was a relatively low number of subjects infected with subtype 1b limiting this analysis.

3.6. Genotype 6 antiviral activity
A single genotype 6-infected participant received 10 days of VCH-759 400mg t.i.d. The compound was well tolerated without adverse event or laboratory abnormalities. There was no significant change in HCV RNA level from baseline during or after dosing.

2. Methods 2.1. Study design
This was a randomized, double-blinded, placebo-controlled study conducted following research ethics review/institutional review board and approval at all participating sites. Consenting participants were assigned to VCH-759 doses (400mg t.i.d., 800mg b.i.d. and 800mg t.i.d.) or corresponding placebo in a 3:1 ratio (n=12 for 400 and 800mg t.i.d. cohorts; n=8 for 800mg b.i.d cohort). Dosing occurred daily under direct supervision for 10 days at 7h:00, 13h:00 and 21h:00 for t.i.d. dosing, and at 7h:00 and 19h:00 for b.i.d. dosing. Before dosing, participants were required to consume a light meal. VCH-759 was supplied as an oral solution formulation in individual 120mL clear glass bottles. The oral solution was reconstituted by combining the appropriate VCH-759 powder-in-bottle (PIB) dose in a 30% polyethylene glycol 400/15% Solutol® HS15 aqueous reconstitution vehicle (RV) (20 and 40mL for the 400 and 800mg doses, respectively).

2.2. Participants
Treatment-naı¨ve, genotype 1-infected male or female participants between 18 and 60 years of age with a body mass index (BMI) ⩽33kg/m2 were recruited. Baseline plasma HCV RNA greater than 100,000IU/mL, alanine aminotransaminase (ALT) values less than five times the upper limit of normal and a Metavir liver fibrosis stage between 0 and 3 were required.

2.3. Endpoint measures
The primary endpoint was defined as the absolute change in plasma HCV RNA levels between baseline to nadir measured with the COBAS Amplicor® HCV Monitor v.2 kit (Roche Diagnostics, Laval, QC). Blood samples for evaluation of the plasma HCV RNA viral load were collected at screening, before the first dose on Days 1–10 and at follow-up visits (Days 11, 17 and 24). Blood samples for NS5B polymerase sequencing were collected before the first dose on Day 1 and on Days 11, 17 and 24. Sequence analyses were performed at ViroChem Pharma.

The complete PK profile was obtained on Days 1 and 10 for the first daily dose. Approximately 6mL of blood were collected at the following time points: for the t.i.d. cohorts, at 10min before dosing (nominal time 0) and at 0.5, 1, 1.5, 2, 3, 4, 5 and 6h after administration of study medication and for the b.i.d. cohort, 10min before the morning dose (nominal time 0), at 0.5, 1, 1.5, 2, 3, 4, 5, 6, 8 and 10h after administration of the study medication, and at 5min before the evening dose (nominal time 12h). For determination of pre-dose VCH-759 trough levels, blood was drawn within 10min before the first dose on Days 2, 3, 4, 5, 6, 7, 8 and 9. On Day 11, blood sampling for VCH-759 plasma trough level quantification was collected at the same time as the Day 9 blood draw. A validated liquid chromatography/mass spectrometer/mass spectrometer bio-analytical method for the evaluation of VCH-759 plasma concentration was used [unpublished internal report; Anapharm, Quebec, Canada].

2.4. Statistical methods
The data were summarized using descriptive statistics. Change in plasma HCV RNA from baseline to subsequent time points was evaluated by a two-sample t-test. Comparisons between dosing groups and their respective placebo group were performed. Analyses of antiviral activity were based on per protocol (PP) participant population. A series of PK parameters (e.g., Cmax, Tmax, AUC, t1/2) were determined from blood samples collected at pre-specified time points. Correlations between VCH-759 trough plasma levels and plasma HCV RNA from Day 2 through end of treatment were analyzed by Pearson’s correlation coefficient.

2.5. Genotypic analysis of HCV variants
The HCV RNA was isolated from patients’ sera using the QIAamp® Viral Minikit (Qiagen, Mississauga, ON). The cDNAs were synthesized using random hexamers (Invitrogen, Burlington, ON) and the SuperScript® III reverse transcriptase (RT) (Invitrogen) for 1h at 50°C. For population sequencing, a series of specific primers for genotype 1a and genotype 1b were designed based on the comparison of 92 HCV NS5A and NS5B sequences retrieved from the Hepatitis C Database (www.hcvdb.org). Primers designed by Yao et al. were also utilized [8]. Briefly, aliquots of the cDNA reactions were subjected to amplification by PCR using combinations of upstream primers and downstream primers in order to obtain overlapping products of approximately 700–1200bp to cover the entire NS5B region. The PCR reactions of 50μl consisted of the high fidelity buffer, 2mM MgCl2, 3% DMSO, 200μM dNTP, 0.5μM each primer and 1 unit of the Phusion Hot start DNA polymerase (Finnzymes Oy, Epsoo, Finland). The following conditions were used for all PCRs: initial denaturation of 2min at 98°C, followed by 42 cycles with 10s at 98°C, 15s at 55°C, 70s at 72°C, and a final elongation step of 10min at 72°C. For the clonal analysis, PCR products, encompassing the VCH-759-binding pocket (amino acids 340–539), were produced using two generated primers (5′ ACG GAG GCY ATG ACY AGR TAY TC and 5′ AT YGG AGT GAG YTT RAG YTT KGT) for both genotypes 1a and 1b. The appropriate sized product (599bp) was then gel-purified (QIAXII® gel purification kit, Qiagen) and cloned using the Zero Blunt® TOPO® PCR cloning kit (Invitrogen) and transformed TOP10 E. coli cells were plated on kanamycin LB agar plates. Alternatively, PCR fragments were treated with the Taq DNA polymerase and cloned using the TA cloning® kit (Invitrogen). Colonies were randomly picked and a variable number (6–30) of clones per participant and per time point were sent for sequencing at Génome Québec (Montreal, QC).

2.6. Phenotypic analysis of HCV variants
2.6.1. Generation of stable HCV replicon cells
With the exception of V494A, all variations were introduced into the pFKI389/NS3-3′/adapt vector [2], [9] obtained from Dr. Ralf Bartenschlager (ReBLikon GmbH, Gau-Odernheim, Germany) or into a genotype 1a sub-genomic replicon that was shown to be sensitive to VCH-759 by PCR-based site-directed mutagenesis and standard recombinant DNA technologies. In vitro produced transcripts (Ambion, Austin, TX) were electroporated into Huh-7 derived cells as described by Lohmann et al. [2] and selected colonies were expanded and sequence of the progeny RNA was confirmed by direct sequencing of the RT-PCR products.

2.6.2. Drug sensitivity assay using real-time PCR
The 9–13 replicon cells and the stable recombinant replicon cells were seeded in complete DMEM media plus 10% FBS. After an incubation period of 3–4h, VCH-759 was added at various concentrations. Cells were then further incubated for 4 days at 37°C with 5% CO2. Thereafter, total RNA was extracted using the RNeasy® kit (Qiagen) according to the manufacturer’s protocol and the cDNA synthesis was performed using the MMLV reverse transcriptase and random hexamer primer (Invitrogen). The inhibitory effect of drugs against the wild-type replicon cells, resistant colonies or recombinant replicons cells was determined by monitoring the levels of HCV RNA normalized to the 18S ribosomal RNA by real-time PCR using the ABI PRISM® 7700 Sequence Detection System (Applied Biosystems, Foster City, CA). To investigate the potential effects of VCH-759 on 9–13 replicon cells proliferation, a [3H] thymidine incorporation assay was performed. Briefly, replicon cells were seeded at sub-confluence density for 96-h incubation in the presence of VCH-759, then 10μCi/mL solution of [3H] methyl thymidine was added in culture medium. The plates were incubated for an additional 18h at 37°C and the accumulation of [3H] methyl thymidine, representing viable replicating cells, was measured using a liquid scintillation counter.

2.6.3. Replication capacity of HCV recombinant replicon clones
A transient replicon system was used for this study. Briefly, transcripts (10–20μg) derived from recombinant pFK I389luc/NS3-3′/5.1 constructs were electroporated in Huh-7 derived cells [9]. Cells were immediately transferred to 40mL of complete DMEM and seeded in 96-well plates, with one plate per each time point (4 and 96h). Cells were then further incubated for 4h up to 96h at 37°C in a 5% CO2 incubator. For each time point post-transfection, the culture media from each corresponding plate was removed and cells were lysed by the addition of a luciferase buffer (luciferin substrate in buffered detergent). Plates were then read for luciferase counts using a luminometer (Wallac MicroBeta Trilux™, Perkin-Elmer, MA). Replication capacity of all recombinant variants were determined as the ratio of the luciferase signal at day 4 post-electroporation divided by the luciferase signal at 4h post-transfection to normalize for the transfection efficiency.