Potential conflict of interest: Dr. McGovern advises Vertex and is on the speakers' bureau of Roche. Dr. Chung is a consultant for and received grants from Roche, Schering-Plough, and Gilead. He is also a consultant for Vertex, Merck, and Pfizer.
The development of specifically targeted antiviral agents against hepatitis C is a major therapeutic advance that promises to markedly improve treatment response rates in patients with chronic infection. However, rapid emergence of drug resistance has already been described, the consequences of which are not yet understood. Although there are important differences between hepatitis C (HCV) and human immunodeficiency virus (HIV) infection, the judicious use of candidate agents against HCV should be guided by principles that have been established in the HIV therapeutic arena. In this review, we attempt to draw useful parallels between the development of antiretroviral therapy for HIV and preliminary data on antiviral agents for hepatitis C virus infection. Applying concepts learned in HIV therapeutics will hopefully lead to a prudent and cautious path in HCV treatment paradigms, particularly with respect to drug resistance. (HEPATOLOGY 2008;48:1700–1712.)
Hepatitis C virus (HCV) is a major cause of chronic liver disease and the leading indication for liver transplantation worldwide. The current best therapy, a combination of long-acting pegylated interferon and ribavirin, leads to sustained virologic response (SVR) rates >50%.1, 2 Attainment of a rapid virologic response (RVR), defined as an undetectable HCV RNA after 4 weeks of therapy, has emerged recently as the most powerful positive predictor of SVR.3–5 Unfortunately, a RVR is achieved by <25% of patients with genotype 1 infection, the most common and difficult to treat genotype in the United States.3, 4 Thus, there are still a substantial proportion of patients who do not achieve SVR, despite excellent adherence.
If the limiting factor for achieving SVR is rapid virologic suppression, can the emerging new therapies for HCV infection change the landscape of HCV treatment outcomes? Proof of principle has been illustrated in preliminary data with some of the specific targeted antiviral therapy against HCV (STAT-C) agents, which have led to remarkably prompt declines of viremia. However, the documented rapid emergence of resistance within days of initiation of therapy is reminiscent of early HIV monotherapy trials.6 The parallels between these two prolific viruses strongly suggest the need for combination therapy with drugs that have nonoverlapping resistance profiles. However, it should be noted that HCV-infected patients would need only a finite treatment period, instead of the life-long therapy that is required by patients infected with HIV.
This review will address STAT-C agents in advanced stages of clinical development, and their resistance profiles, with frequent reference to the lessons learned from 2 decades of HIV drug development.
HCV, hepatitis C virus; HIV, human immunodeficiency virus; NI, nucleoside inhibitors; NNI, non-nucleoside inhibitor; NS, nonstructural protein; PEG-IFN, pegylated interferon; RVR, rapid virologic response; SOC, standard of care; STAT-C, specifically targeted antiviral agents against hepatitis C; SVR, sustained virologic response.
HCV Targets and Drug Development
The development of antiviral therapy for HCV infection was initially hampered by the inability to propagate virus in culture. A major breakthrough was the development of the HCV replicon model.7 Replicons are HCV RNAs harboring viral nonstructural proteins and a selectable marker, G418. Under G418 selection, cells transfected with replicons are capable of supporting autonomous RNA replication that can be readily quantified, thereby enabling screening of agents that inhibit viral RNA replication. Replicon systems are also useful for selection and characterization of resistance mutations to specific HCV inhibitors and assessment of replication fitness.8 More recently, two groups have also reported infectious virus production in tissue culture,9, 10 which now permits study of agents that block all steps in the viral life cycle (Table 1).
Table 1. Agents in Clinical Development
Phase of Study
Failed (renal toxicity)
Failed (cardiac toxicity)
Failed (gastrointestinal effects)
HCV is an enveloped, single-stranded RNA virus whose 9.6-kb genome encodes a polyprotein of approximately 3,000 amino acids flanked by a highly conserved internal ribosome entry site (Fig. 1). Proteolytic enzymes mediate cleavage of the HCV polyprotein into four structural and six nonstructural (NS) proteins.11 The catalytic core of the replication machinery, the viral RNA-dependent RNA polymerase, is encoded by NS5B. The N-terminal domain of the NS3 protein encodes the serine protease that, together with its cofactor NS4A, carries out post-translational processing of the NS proteins at the NS3/NS4A, NS4A/NS4B, NS4B/NS5A, and NS5A/NS5B junctions.12 The current major targets for drug development include highly selective inhibitors of NS5B and NS3/NS4A.13 One of the challenges of HCV drug development is that the physical nature of the binding sites differs across six genotypes and more than 100 subtypes14; a single residue change in a viral isolate can lead to loss of inhibitor binding.15
NS3/NS4A is a heterodimeric serine protease comprising the N-terminal domain of the NS3 protein and the small hydrophobic NS4A protein that cleaves the viral polyprotein and is important for viral replication.14 Small peptides derived from the N-terminal cleavage products of its substrates have been shown to competitively inhibit this enzyme.16 This has led to the discovery of several peptidomimetics with potent and specific activities against the NS3/NS4 protease.
The three-dimensional structure of the NS5B polymerase can be thought of as a “right hand” motif, with the palm containing the active site of the enzyme and the fingers and thumb encircling the viral RNA template (Fig. 2).17 NS5B inhibitors can be classified into active site and allosteric inhibitors.8 Nucleoside inhibitors (NIs), which resemble HIV nucleosides, act by serving as alternative substrates of NS5B at its active catalytic site (Fig. 2).8, 15, 18 They function as chain terminators through their incorporation into the growing RNA molecule.8 In contrast, non-nucleoside inhibitors (NNIs) are structurally diverse and inhibit the initiation and/or elongation steps of viral replication by binding to one of several allosteric binding sites (e.g., NNI-A, B, C, D, or E) (Fig. 2).8, 19 This is in marked contrast to NNIs of HIV reverse transcriptase, which bind to the same site of the viral enzyme. These binding site characteristics may translate into varying propensities for drug resistance.20
HCV NS3-4A Protease Inhibitors
The first protease inhibitor in this class was BILN 2061, which produced a 2 to 3 log10 IU/mL decline in HCV RNA after 2 days of therapy.21, 22 Although development of this drug was halted due to cardiac toxicity in animals, these results provided “proof-of-concept” for the efficacy of targeting the NS3/NS4 protease.
In vitro23 and clinical data24 support synergy between telaprevir and interferon. In a study of 20 treatment-naïve patients, 14 days of telaprevir and pegylated interferon α2a (PEG-IFN α2a) led to greater median declines of HCV RNA (−5.49 log10 IU/mL) than either drug alone (−3.99 log10 IU/mL with telaprevir versus −1.09 log10 IU/mL with PEG-IFN-α-2a).24
The PROVE 1 (n = 250) and PROVE 2 (n = 323) trials were similarly designed to determine if a shorter duration of treatment was feasible for genotype 1–infected patients using telaprevir in combination with PEG-IFN/ribavirin therapy compared to standard-of-care (SOC) or telaprevir plus PEG-IFN alone (Fig. 3).25, 26 Both trials showed much higher RVR and SVR rates in the triple therapy arms compared to the “no ribavirin” and SOC arms. For example, in the PROVE 2 trial, patients in the triple therapy arm had the highest proportion of patients with a RVR (74%) compared to dual therapy without ribavirin (53%) and the SOC arm (14%).25 Preliminary data on the 24-week and 12-week triple therapy arms showed SVR rates of 65% and 59% compared to the much lower SVR rates of 29% in the “no ribavirin” arm. Surprisingly, interim data also showed that 83% of the patients in the two PROVE studies, who had been defined as nonresponders or relapsers in the SOC arms (n = 72), subsequently attained RVR when retreated with triple therapy (telaprevir + SOC).27
These data are enlightening in several respects: (1) the addition of a potent STAT-C agent achieves greater overall efficacy and may permit a shorter duration of therapy for those who achieve RVR; (2) the role of ribavirin is reaffirmed because omission of this drug had a significant negative impact on RVR; (3) the utility of the week 4 viral load clearance in predicting outcome remains strong; and (4) telaprevir retains potency in prior nonresponders.
It will be important to determine whether telaprevir exposure can be shortened in patients who achieve a RVR, in light of its side effect profile. Adverse events include rash; 4% to 6% had a severe rash and 10% to 15% of patients discontinued drug by week 12 in the PROVE 2 trial. Other side effects include pruritus, asthenia, nausea, and anemia.25
Boceprevir (SCH-503034) binds reversibly to the NS3 protease active site and has additive potency when combined with pegylated interferon α2b (PEG-IFN α2b).28 In a study of 357 prior null or nonresponders with genotype 1 infection, SVR rates were generally low (<15%); however, this was an early phase-II dose-finding study and due to the safety profile of the chosen doses, later studies have used higher doses of 800 mg three times a day.29
In treatment-naïve patients, a strategy of using a lead-in phase of PEG-IFN α2b plus ribavirin followed by the addition of boceprevir was compared to triple therapy from initiation (Fig. 4).30 In a 12-week interim analysis after treatment discontinuation, both boceprevir-containing arms achieved similar rates of SVR (57% versus 55%) despite higher rates of RVR in the induction arm (60% versus 39%).
As with the telaprevir studies, preliminary data from the SPRINT trial at 12 weeks of follow-up have confirmed the robust predictive value of RVR for achieving an SVR. Furthermore, among the patients who completed 28 weeks of therapy, lack of viral suppression at 4 weeks was predictive of virologic relapse. This was particularly true of those patients in the “lead-in arm” where virologic relapse occurred in 57% in those who did not achieve an RVR compared to 8% of those who did. Further understanding of this group of patients, including their initial response to PEG-IFN/RBV alone and the possible evolution of drug resistance after the addition of boceprevir, will be important in guiding future clinical trial design. Encouraging preliminary data were also presented regarding improved SVR rates in African-Americans, compared to historic controls.
Discontinuation rates were approximately two times higher in the boceprevir-containing arms than SOC; dysgeusia, fatigue, and anemia occurred more frequently with boceprevir.
(3) Other Protease Inhibitors.
ITMN-191 (R7227) is a potent HCV NS3/4A protease inhibitor, which binds to proteases of genotypes 1 to 6 through a two-step mechanism.31 ITMN-191 dissociates slowly from this complex with all tested genotypes but genotype 3.31 In vitro systems show synergy between ITMN-191 and peginterferon.32 TMC435350 is a potent PI that led to HCV RNA declines of 3.9 log10 IU/mL at day 6 with pharmacokinetics suggesting feasibility of once-daily dosing.33
Nucleoside Polymerase Inhibitors
(1) Valopicitabine (NM283).
Valopicitabine, a nucleoside analog that inhibited viral RNA polymerase, was associated with a greater HCV RNA decline in combination with PEG-IFN compared to standard therapy.34, 35 However, because of dose-limiting gastrointestinal adverse events, drug development was halted.34
R-1626 is the prodrug of R-1479, a potent inhibitor HCV RNA polymerase.36 A phase 2a study evaluated the safety and efficacy of R-1626 administered for 4 weeks in combination with PEG-IFN α-2a with or without weight-based ribavirin in HCV genotype 1 treatment-naïve patients.37 In this four-arm study of 104 patients, the highest rates of RVR were seen in the triple therapy arm (81%) with a mean reduction of 5.2 log IU/mL (Fig. 4). Although the sample size was small, these preliminary data suggest an in vivo synergistic effect of R-1626, PEG-IFN α2a, and ribavirin. However, grade 4 neutropenia occurred with much higher incidence in the R1626-containing arms and was the most common reason for dose reductions. Whether lower doses R-1626 can retain effectiveness with less hematologic toxicity will be determined in future trials.
R7128 is a prodrug of PSI-6130, an oral cytidine nucleoside analog polymerase inhibitor. A multiple, 14-day ascending dose study in 40 genotype 1-infected patients with a history of prior treatment failure showed that R7128 was associated with a dose-dependent HCV RNA decline (mean 2.7 log) in the 1,500 mg twice daily arm.38 The results of this study are encouraging because no virologic rebound was noted. The maximum tolerated dose of R7128 has not yet been identified; the most common side effect was headache.38 A 28-day phase 1 multicenter, randomized placebo-controlled clinical trial to evaluate R7128 in combination with PEG-IFN plus ribavirin in treatment-naïve patients is currently enrolling.
Non-Nucleoside Polymerase Inhibitors
GS-9190 is a novel non-nucleoside HCV NS5b polymerase inhibitor with potent antiviral activity.39–42 In a dose-finding study HCV RNA declined by 1.4 and 1.7 logs in the 40 mg and 120 mg twice daily dosing arms over an 8-day period and the plasma half-life was approximately 10 to 13 hours. These preliminary data suggest the possibility of daily or twice daily dosing of GS-9190. The symptom reported most commonly was headache. One patient developed asymptomatic QT prolongation of uncertain significance. A subsequent study focused on electrocardiographic findings is planned before further clinical trials. In vitro, the combination of GS-9190 with interferon had additive activity than either drug alone.39 It is much less active against HCV genotype 2.40
In a 14-day dose-ranging study of NNI VCH-759 in treatment-naïve genotype-1 infected patients, a mean maximal HCV RNA decline of 2.5 log was seen in the 800 mg TID group.43 Genotypic sequencing of the NS5B polymerase is being investigated to determine whether some of the observed virologic breakthroughs were associated with drug resistance. Diarrhea was seen in the active drug and placebo arms, suggesting that the polyethylene glycol formulation may have contributed to this adverse event.
HCV-796, the first NNI of the HCV polymerase, showed significant HCV antiviral activity against multiple genotypes.44 However, the development of this drug was discontinued due to hepatotoxicity.
Inhibitors of Host Cofactors: Cyclophilin Inhibitors
Cyclophilin B promotes RNA binding activity of the HCV RNA-dependent RNA polymerase (NS5B) in vitro.45 Early observations with cyclosporine, an immunosuppressive drug with CYP-inhibiting activity, showed anti-HCV activity in vitro.46, 47 Subsequent drug discovery led to the development of nonimmunosuppressive cyclophilin inhibitors, including DEBIO-025.48, 49
DEBIO-025 has shown activity against both HIV and HCV in vitro and in vivo49–51 and synergy with PEG-IFN against HCV in vivo.52 In a dose-ranging study of 90 treatment-naïve patients with varying genotypes, the combination of DEBIO-025 (1,000 mg/day) plus PEG-IFN α-2a led to greater declines in HCV RNA (>4 log) compared to DEBIO-025 alone (>2 log).53 Future studies will be carried out with 600 mg daily because higher dosages have been associated with isolated conjugated hyperbilirubinemia.51, 53 The proposed mechanism of this effect is competitive inhibition of biliary canalicular transporters.54 NIM811, another nonimmunosuppressive cyclophilin antagonist, has also entered early-phase clinical studies.
Emergence of HCV Drug Resistance
The emergence of resistance is a common theme shared by HIV and HCV due to the error-prone nature of reverse transcriptase of HIV and the RNA-dependent RNA polymerase of HCV.55–57 Drug resistance has already been recognized as a major obstacle in the treatment of HCV, particularly in light of the high rates of viral production (up to 1013 particles/day)58 and its circulation as a quasispecies within infected persons.57, 59 By applying population-based models developed previously for HIV,60 it is conceivable that every single nucleotide base substitution in HCV can be predicted to occur early on, setting the stage for the development of viral variants with mutations that would predict drug resistance, even before onset of drug exposure (Table 2).61
Table 2. Characterization of Antiviral Agents for Chronic HCV in Clinical Studies
Indeed, the HCV protease inhibitors have been associated with rapid emergence of viral variants with low to high levels of drug resistance. In the replicon system, mutations in the NS3 serine protease domain (A156S/T, D168A/V, T54A, and V170A) have been identified, which confer cross-resistance to telaprevir and boceprevir.62 In vivo data for active site polymerase inhibitors suggest a high barrier to the development of resistance, although distinct patterns of mutations have been identified in in vitro systems.8, 63, 64 In contrast, NNIs have been associated with a high frequency of resistance, likely attributable to increased tolerability of the virus to conformational changes at allosteric sites that reduce binding. Preliminary data indicate no or limited cross-resistance between NI and NNI inhibitors.8
Over the past 2 decades, a thorough understanding of resistance mutations,65 patterns of treatment response,66 replication capacity,67 and drug sequencing of antiretroviral agents68 has led to marked advancements in HIV therapy. These shared concepts, as they apply to STAT-C agents, are discussed below.
Protease Inhibitor Resistance
(1) Telaprevir Resistance.
Using a highly sensitive sequencing method to detect minority populations of viral variants, mutations were identified in patients taking telaprevir or telaprevir plus PEG-IFN2a during a 14-day study.69 Although all patients had a rapid first phase HCV RNA decline of approximately 3 log10 IU/mL, only those taking combination therapy had a continuous decline during the second phase. In contrast, those taking telaprevir alone had heterogeneous patterns of virologic response: continuous decline, plateau, or viral rebound, suggesting the emergence of resistance. Several important points arose from these early observations of the potent antiviral activity of telaprevir, which are akin to what has been seen in HIV drug development.70, 71 First, drug exposure and adequate trough concentrations were necessary for the rapid reduction of HCV RNA and minimizing the risk of drug resistance. Second, the rapid emergence of viral variants suggested that these viral strains pre-existed in the quasispecies population and were selected through drug pressure. Finally, the genetic barrier to resistance to telaprevir is low.
A detailed analysis of the patient samples included genotypic/phenotypic characterization, phylogenetic analysis and measurements of quasispecies complexity/diversity and replication capacity.72 The mutations observed most frequently were V36A/M, T54A, R155K/T, and A156V/T/S. These changes occurred as either single or double mutations, some of which were not predicted by in vitro testing.73 Low-level resistance was conferred by most single mutations whereas high-level resistance was mainly related to the presence of double mutants (36 + 155 or 36 + 156). However, the single mutation of A156T/V was also associated with high level resistance, in contrast to the low level resistance seen with A156S.74 The development of the A156S/T mutant confers cross-resistance to boceprevir,72 which may limit the ability to sequence these two agents for HCV therapy.
Although most of the mutant viruses were shown in the “breakthrough” and “plateau groups,” low-level resistant mutants were also identified in the “continuous decline” group 7 to 10 days after dosing.69 This observation supports the need for combination therapy, even in patients with the best virologic response. Also of interest, V36M and R155K/T mutations were detected only in genotype 1a isolates. This observation was attributed to the need for two nucleotide substitutions in genotype 1b viral isolates compared to only one substitution in genotype 1a.72, 75 These findings suggest for the first time clinically relevant differences between genotypes 1a and 1b.
It is also noteworthy that virologic breakthrough was uncommon (3%) in prior null/nonresponders or relapsers from the two PROVE trials who were treated subsequently with triple therapy.27 These data support the potency of telaprevir and the protective efficacy of interferon/ribavirin in preventing resistance (see “Role of Peginterferon and Ribavirin in Combination with STAT-C Agents”)
(2) Boceprevir Resistance.
Three mutations, T54A, V170A, and A156S confer low to moderate levels of boceprevir resistance in vitro.75, 76 However, longer drug exposure leads to the selection of the A156T mutation, which confers a marked decline in drug sensitivity (>100-fold). The combination of boceprevir with interferon-α or valopicitabine, markedly reduced the emergence of these resistant variants in vitro.77
In vivo, combination PEG-IFN α2b plus boceprevir led to fluctuating HCV RNA curves, which increased toward baseline at the end of the weekly dosing interval.29 In a subsequent dose-finding study of 357 prior pegIFN/RBV null and nonresponders, the emergence of resistance mutations was as high as 82% in initial responders who developed virologic breakthrough.78 Furthermore, 58% of those who had virologic relapse after treatment cessation had evidence of drug resistance. Although most of these difficult-to-treat patients had received less than optimal dosing of boceprevir, the patterns of virologic response and the risk of drug resistance are still instructive for future research. Resistance data from the clinical trial of boceprevir plus SOC in treatment-naïve patients is not yet available.
Nucleoside Inhibitor Resistance
(1) R1626 Resistance.
In vitro data show that amino acid substitutions S96T or S96T/N142T in NS5B polymerase confer a four- to five-fold reduction in sensitivity to R1479.42 These mutations also lead to a 95% reduction in replication capacity compared to wild-type virus. In vivo, these mutations have not been isolated in patients with viral rebound, suggesting a high barrier to resistance.37, 63 This observation may be related to low viral fitness and/or low prevalence of these viral strains in the circulating pretreatment quasispecies.79 One study of clinical isolates from treatment-naïve patients found that resistance mutations to NI were not observed.80 In contrast, natural polymorphisms at amino acid residues associated with NNI resistance were found in low frequency although rapid selection of these minor variants occurred with drug pressure (i.e., HCV-796).
Non-Nucleoside Inhibitor Resistance
The C316Y mutation, which was seen during the development of HCV-796, was also documented with passage of GS-9190 in vitro40; however, no mutations were documented in the multiple-dosing patient cohort over 8 days of exposure.42
Combination STAT-C Therapy and Drug Resistance
In the studies discussed above, the rapid emergence of resistance in as little as 8 days makes clear the need for combination therapy with drugs with nonoverlapping resistance profiles.77, 81 One in vitro study evaluated cross-resistance between NIs (i.e., R1479 and R7128), NNIs (i.e., HCV-796, whose development was halted because of hepatotoxicity), and PIs (i.e., telaprevir).82 R1479 and R7128 maintained activity against telaprevir and HCV-796 resistant mutants, whereas telaprevir and HCV-796 were active against NI-drug resistant mutants. The lack of cross-resistance between PIs, NIs, and NNIs, support the potential use of combination therapy.77, 83, 84 One model showed synergistic interactions of one protease inhibitor and two polymerase inhibitors, giving support to the notion that interferon-sparing regimens may be possible eventually.83
Role of Peginterferon and Ribavirin in Combination with STAT-C Agents
Although the scenario of combination STAT-C therapy described above is appealing, it will undoubtedly be years before this treatment paradigm is reached. In the meantime, it is critical to assess the individual roles of PEG-IFN and ribavirin therapies in combination with specific HCV agents. In vitro and in vivo studies suggest these standard agents play a major role in improving efficacy (by way of additive or synergistic activity) and in reducing drug resistance of protease and polymerase inhibitors.23, 28, 39
Furthermore, the protease inhibitors may have great therapeutic potential not only in specifically blocking HCV viral replication, but also in restoring endogenous interferon activation.84, 85 HCV NS3-4A protease decreases phosphorylation of interferon regulatory factor-3 (IRF-3), a key antiviral signaling molecule, thereby blocking its translocation into the nucleus, where it functions as a transcriptional activator of IFN-β.84 Thus, the NS3/4A protease may represent a dually advantageous therapeutic target, the inhibition of which may both block viral replication and restore IRF-3 control of HCV infection.
Another role for interferon therapy is to “shield” STAT-C drugs from the development of drug resistance, a concept that has been shown already in the treatment of hepatitis B.86 Clinical studies of telaprevir and PEG-IFN α2a (with or without ribavirin) have shown that combination therapy inhibited both wild-type and drug-resistant variants from emerging, even in prior interferon/ribavirin-nonresponders.27, 69 Furthermore, PEG-IFN α2a and ribavirin were shown to be effective against drug-resistant viral variants after they were detected.24, 74, 87, 88
It is unknown as to whether one particular formulation of pegylated interferon may be better suited to “protect” HCV protease inhibitors, like telaprevir, from the development of resistance. Although the SVR rates associated with PEG-IFN α 2b and α 2a are comparable in HCV monoinfected patients, the pharmacologic properties of the drugs differ considerably; specifically, PEG-IFN α2b is associated with low therapeutic levels at the end of dosing, which could potentially lead to a window of “monotherapy”.89 It is conceivable, therefore, that the risk of STAT-C drug resistance could be influenced by the specific pharmacologic properties of the paired interferon formulation. Thus, it will be important for prospective STAT-C agents to be tested in combination with each interferon formulation to test these hypothetical concerns.
Role of Ribavirin with STAT-C
Ribavirin is a guanosine nucleoside inhibitor whose contributory role in viral suppression has also been confirmed, whether as a companion to protease inhibitors or polymerase inhibitors.74 Proposed mechanisms explaining the efficacy of ribavirin have included its role in driving mutational frequencies, decreasing viral synthesis, or as an immunologic enhancer.90–93
The importance of ribavirin in delaying the emergence of drug resistance has also been well illustrated. In PROVE 2, viral breakthrough (defined as >1 log above HCV RNA nadir or increase to >100 IU/mL in patients with previous viral suppression) was much higher in the “no ribavirin” arm compared to the SOC and triple therapy arms (24% compared to 1% and 2%, respectively).25 Furthermore, patients with virologic breakthrough developed additional mutations while continuing on telaprevir monotherapy. As noted previously for HIV, these observations emphasize the importance of rapid viral suppression, not only in achieving a sustained virologic response, but also in preventing resistance.94 These concepts were well illustrated in the clinical trial results of boceprevir in PEG-IFN null responders.78
One advantageous outcome of drug resistance can be a decline in replication capacity (RC),64 which has been well described with HIV.95 For example, in vitro selection of replicon variants dually resistant to two different NNIs has been associated with severe limitations of RC, compared to wild-type.64 The RC of telaprevir-resistant variants is reduced or significantly impaired with R155K/T or A156T/V mutations, respectively,88, 96 whereas V36A/M or T54A mutations have much less effect on viral fitness. This is akin to HIV, which is constrained to become both highly resistant and highly fit.95 Consistent with these observations, A156T/V HCV viral variants decline much more rapidly compared to other less resistant mutants after cessation of telaprevir dosing.72
However, the development of V36A/M mutations in a background of R155K/T or A156V/T result in compensatory effects on replication efficiency97 and have also been associated with increased HCV replication in vivo.69 These data support similar principles seen in HIV therapy, whereby ongoing replication in the face of drug resistance favors the development of compensatory mutations that will eventually benefit the virus.98
Potential Role for Ritonavir
With few exceptions,99 most agents in development require multiple daily dosing. In studies done to date, telaprevir has been administered on a strict “q8 hour-dosing” to maintain adequate trough concentrations, due to rapid rates of clearance. As was seen in the past with HIV therapies, such rigid dosing schedules are difficult to maintain and lead to poor adherence and drug resistance.100 However, pharmacologic boosting of HIV protease inhibitors with ritonavir markedly changed the landscape of HIV treatment. Through its potent inhibition of CYP3A, use of this pharmacologic enhancer led to much more favorable drug inhibitory quotients, potency, and in some cases, more “forgiving” dosing schedules.101–103 Preliminary animal data show that the addition of ritonavir, to telaprevir or boceprevir markedly inhibit their metabolism resulting in a 50-fold increase in trough concentrations and a three-fold increase in peak concentrations.104 Whether “boosted PIs” will enter the lexicon of HCV therapeutics will depend on results of studies in humans.
HIV drug resistance is archived as latently integrated provirus in resting memory T-cells105; re-exposure to a drug given in the past will lead to re-emergence of previously documented resistance mutations.106 Most experts believe that HCV drug resistance will not be archived in the same way because HCV is not integrated within the genome and no long-lived reservoir has been identified.
Whether HCV-related drug resistance will be any less formidable than that related to HIV or hepatitis B virus is unclear, however, because telaprevir-associated low level resistant variants were maintained as long as eight months after drug exposure.97 It will also be important to determine how well the STAT-C drugs reach other replication compartments where HCV RNA has been identified (e.g., seminal fluid, central nervous system, plasma, and peripheral blood mononuclear cells).107
Another issue is how acquired resistance to one agent can increase the risk of subsequent drug resistance to new agents. For example, in the field of hepatitis B resistance, <1% of treatment-naïve patients developed entecavir resistance at 4 years compared to >15% in lamivudine-experienced patients.108 In contrast, resistance to one drug can confer hypersusceptibility to another, as shown for HIV.109 These types of data will need to be developed fully for STAT-C agents to understand optimal sequencing of drug therapy.110
Looking to the Future
STAT-C agents hold the promise of enhancing RVR, increasing overall SVR rates, shortening treatment duration and improving tolerability compared to current therapeutic agents. Inevitably, genotypic and phenotypic resistance tests will also enter the therapeutic arena. Once several STAT-C agents are available, treatment strategies may evolve that are analogous to HIV with a two-drug polymerase “backbone” with a PI.83 Whether induction/maintenance therapeutic strategies may be effective for high viral load patients should also be explored. Drug interaction studies will be a high priority, particularly in HIV-HCV coinfected patients who are already subject to polypharmacy.
The rapid development of multiple classes of drugs shows that the HCV research field has clearly benefited from the arduous path of HIV therapeutics. Novel modeling approaches to HIV drug development have included enhancement of substrate-inhibitor binding to decrease the risk of resistance.111 Such approaches are already being applied to HCV drug discovery.112 Once safe and efficacious drugs are available, their ongoing efficacy will depend on their judicious use, which should be firmly based on principles learned during the last two decades of HIV therapy.