The characterization of the viral life cycle facilitated the development of directly acting antiviral drugs. Among those, several inhibitors of the viral RNA-dependent RNA polymerase have proven effectiveness in clinical trials. The characteristics of different nucleos(t)ide and non-nucleoside polymerase inhibitors, as well as their clinical applications and combinations with other classes of directly acting antiviral drugs are reviewed herein.
According to the World Health Organization (WHO), about 3% of the world's population has been infected with the hepatitis C virus (HCV). Of those, approximately 170 million are chronic HCV carriers at risk of developing cirrhosis and hepatocellular carcinoma (HCC) contributing to a large percentage of liver transplantations in Europe and the United States .
For almost a decade, the combination of pegylated interferon (PEG-IFN) and ribavirin (RBV) has been the standard of care (SOC) producing sustained virological response (SVR) rates of ~50% in treatment-naive patients infected with HCV genotype 1 and 70–90% in patients infected with HCV genotypes 2 and 3 . Recently, this therapeutic backbone has been supplemented by the addition of first- generation protease inhibitors (Telaprevir, Boceprevir) directly targeting the viral NS3/4A protease. This ‘triple therapy’ improved SVR rates to 70–80% in treatment-naive patients infected with HCV genotype 1 [3, 4].
The development of directly acting antiviral (DAA) HCV therapeutics was facilitated by the adaptation of HCV to a cell culture system allowing the description of the HCV lifecycle and identification of potential novel drug targets. In addition to protease inhibitors (PI), these targets include inhibitors of the non-structural NS5A enzyme, a protein possibly involved in HCV replication, and nucleoside/nucleotide analogue and non-nucleoside inhibitors of the HCV RNA-dependent RNA polymerase (RdRp/NS5B) [5-7].
The following sections provide an overview of novel therapeutic strategies involving nucleoside/nucleotide analogue and non-nucleoside inhibitors of the HCV RNA-dependent RNA polymerase (RdRp/NS5B).
Function and biological role of the Hepatitis C Virus (HCV) polymerase inhibitor
A highly structured association of RNA and viral proteins, of cellular proteins and cofactors, and of rearranged intracellular lipid membranes derived from the endoplasmic reticulum are essential for replication of the viral RNA genome . The key enzyme in this process is NS5B, a RNA-dependent RNA polymerase, catalysing the synthesis of a complementary negative-strand RNA by using the positive-strand RNA genome as a template. Numerous RNA strands of positive polarity are produced by NS5B activity from this negative-strand RNA and serve as templates for further replication and polyprotein translation .
Schematically, the NS5B protein has the shape of a right hand with an active site located within the palm domain and encircled of a thumb and a finger domain (Fig. 1). The thumb domains act as regulator of nucleic acid binding and the catalytic efficiency of the enzyme's active site, while the palm domains coordinate the function of the active site and carry out the nucleotidyl transfer reaction .
Two classes of NS5B inhibitors have been developed: nucleoside/nucleotide and non-nucleoside polymerase inhibitors.
Nucleoside/nucleotide analogues act as natural polymerase substrates leading to termination of RNA chain elongation by inhibition of the active site of the HCV RdRp.
Mostly synthetic prodrugs of nucleotides are administered to facilitate resorption, and additional steps of intracellular phosphorylation are required to gain full functional activity as a nucleosid triphosphate.
Because of high conservation of the active site of the HCV RdRp across all HCV genotypes, these drugs have pan-genotype equivalent antiviral activity in vitro. In vivo data have not yet been completed. In vitro, nucleosidic inhibitors display a low ‘genetic barrier to resistance’ as drug resistance can be observed with single amino acid substitutions. However, because of the poor fitness of resistant variants in the presence of nucleosidic inhibitors, these agents are considered to have a high overall ‘barrier’ to resistance.
Non-nucleoside polymerase inhibitors
Non-nucleoside inhibitors bind outside the active site and target allosteric sites on the surface of the enzyme, downregulating the RdRp activity through induction of conformational changes (Fig.1).
Non-nucleoside inhibitors are so far specific for HCV genotype 1. The efficacy against genotype 1 subtypes, however, may differ. The barrier to resistance of non-nucleoside inhibitors is considered to be low.
Four groups of non-nucleoside inhibitors have entered clinical development (summarized in Table 1).
Table 1. Polymerase inhibitors in clinical development
Efficiency / log decline (log10 IU/ml)
Phase of development
0.87 (15 d)
Gastrointestinal side effects
2.6-3.7 (14 d)
Neutropenia, lymphopenia, neurotoxicity
0.6-4.25 (7 d)
0.75-3.79 (R/7 d)
Liver and severe cardiac & renal toxicity
4.8-5.8 (14 d)
4.54 (7 d)
2.7 (14 d)
4.7 (7 d)
Thumb I inhibitors
BILB-1941 (Boehringer Ingelheim)
> 1 (5 d)
gastro-intestinal side effects
3.75 (7 d)
gastro-intestinal side effects
BI-207127 (Boehringer Ingelheim)
0.6-3.1 (5 d)
Thumb II inhibitors
2.5 (10 d)
low anti-viral efficacy
1.5 (3 d)
low anti-viral efficacy
3.5 (3 d)
0.68-2.13 (8 d)
3.1-3.4 (3 d)
Palm I inhibitors
0.7-1.5 (2 d)
1.19-2-3 (2 d)
2.4-2.9 (3 d)
Palm II inhibitors
1.4 (14 d)
0.5-1.1 (1 d)
1.4-1.7 (8 d)
Thumb I inhibitors (benzimidazole site): MK-3281 (Merck, Rahway, New Jersey, USA), B ILB1941 and BI 207127 (Boehringer-Ingelheim, Ingelheim am Rhein, Germany) are thumb I site inhibitors. Pre-existing NS5B substitutions known to reduce sensitivity were identified e.g. for BILB1941 at positions V494I/A, I424V and P496A .
Thumb II inhibitors (thiophene site): filibuvir (PF-00868554/Pfizer, New York, USA), VX-759, VX-916 and VX-222 (Vertex, Cambridge, Massachusetts, USA) bind to the thumb II site. Variants at position M423 known to confer resistance to filibuvir from in vitro studies were also selected in most patients in vivo. For VCH-759 as well as VCH-916, viral breakthroughs with selection of resistant variants conferring high (M423T/V/I) and medium (L419V/M, I482L/V/T, V494A/I) levels of resistance were described .
Palm I inhibitors (benzothiadiazine site): ABT-333, ABT-072 (both Abbott, Abbott Park, Illinois, USA) and Setrobuvir (ANA598; Anadys/Roche, Basel, Switzerland) are inhibitors of the palm I site. Several variants (M414T, G554D, D559G) were described from in vitro replicon studies to confer resistance to ANA598. Mutations associated with resistance in patients treated with ABT-333 were observed at positions C316Y and Y448C .
Palm II inhibitors (benzofuran site): nesbuvir (HCV-796/ViroPharma/Wyeth, Exton, Pennsylvania, USA) and IDX-375 (Idenix, Cambridge, Massachusetts, USA) act at the palm II site. The C316Y mutation in NS5B is associated with resistance to HCV-796 in vitro and in vivo .
Polymerase inhibitors in clinical trials
Nucleoside and non-nucleoside inhibitors have been investigated in combination with PEG-IFN and RBV and in PEG-IFN-free therapeutic regimens.
Several drugs in clinical development were discontinued because of clinically significant side effects including valopicitabine (NM-283), R-1626, PSI-938 and BMS-986094 (formerly known as INX-189) (Table 1).
Promising clinical data have been presented for the following HCV nucleos(t)ide analogues that have entered phase 2/3 clinical testing: mericitabine (RG-7128; prodrug of the pyrimidine (cytosine) nucleoside analogue PSI-6130) and sofosbuvir (PSI-7977/GS-7977; chirally pure isomer form of PSI-7851; prodrug of the nucleotide pyrimidine (uridine) analogue GS-7411). Early results were recently reported for ALS-2200, a novel nucleotide pyrimidine analogue polymerase inhibitor with a median reduction in HCV-RNA of 4.54 log10 after 7 days of dosing .
Nucleos(t)ide inhibitors in clinical trials with interferon
Mericitabine (RG 7128)
The PROPEL study investigated safety, tolerability and efficacy of 8 and 12 weeks of mericitabine (1000 mg BID) in combination with PEG-IFN/RBV in 408 treatment-naive patients with chronic HCV genotype 1 or 4 infection. The rapid viral response (RVR) rates were up to 62% in the treatment arms with mericitabine compared with 18% in the control arm with PEG-IFN/RBV alone. SVR rates, however, did not differ between patients receiving mericitabine in combination with PEG-IFN/RBV and patients who received placebo plus PEG-IFN/RBV for 48 weeks. In the response-guided arms, fewer patients achieved SVR (33–49%) because of high virological relapse rates .
The JUMP-C trial (phase 2) investigated the safety and efficacy of 24 weeks of response-guided therapy with mericitabine (1000 mg BID), PEG-IFN/RBV in 168 treatment-naive patients with HCV genotype 1 or 4 infection. Patients were randomized (1:1) to response-guided therapy with mericitabine plus PEG-IFN and RBV for 24 or 48 weeks or to placebo, PEG-IFN/RBV for 48 weeks. In the mericitabine treatment arm, therapy was stopped at week 24 in patients with an extended rapid virological response [eRVR], defined as undetectable HCV RNA (<15 IU/ml) from week 4 to 22. Patients without eRVR received PEG-IFN/RBV for another 24 weeks. Overall SVR rates were higher in patients treated with mericitabine plus PEG-IFN/RBV than in patients treated with PEG-IFN/RBV alone (58% vs. 36%). Among 49 patients (60%) who achieved an eRVR with mericitabine plus PEG-IFN/RBV and discontinued therapy at week 24, the SVR rate was 78%. Combination therapy with mericitabine was safe, well-tolerated, demonstrated a high resistance barrier and a low potential for pharmacokinetic drug–drug interactions .
The MATTERHORN study, another phase 2 trial, is an ongoing study evaluating the efficacy and safety of various combinations of mericitabine and ritonavir-boosted danoprevir and PEG-IFN/RBV in patients with HCV genotype 1 infection.
The ongoing phase 2 DYNAMO I and II studies investigate quadruple therapy with mericitabine in combination with the protease inhibitors boceprevir or telaprevir and PEG-IFN/RBV in HCV genotype 1 non-responders to PEG-IFN/RBV.
Sofosbuvir (GS-7977, formerly PSI-7977)
The combination of sofosbuvir (400 mg QD), PEG-IFN and RBV was assessed for 12 weeks in 25 treatment-naive patients infected with HCV genotypes 2/3 in an open-label, uncontrolled pilot study (PROTON). Of the patients, who completed therapy (n = 24), 100% achieved an SVR. Based on these favourable results, the PROTON study was expanded to treatment-naive patients infected with HCV genotype 1. Patients were randomized 2:2:1 to different dose groups of sofosbuvir (200 or 400 mg QD) for 12 weeks plus PEG-IFN/RBV for 24 weeks. Patients without RVR were continued on PEG-IFN/RBV through week 48. Reported SVR rates were 88, 91% and <50% for sofosbuvir 200, 400 mg and the control group respectively [15, 16].
The ELECTRON study investigated sofosbuvir (400 mg QD) and RBV for 12 weeks in combination with 0, 4, 8 or 12 weeks PEG-IFN. All patients (n = 10 per group) achieved an SVR after 12 weeks of therapy .
The ATOMIC study, a Phase 2 randomized open-label trial investigated sofosbuvir in combination with PEG-IFN/RBV in 316 non-cirrhotic HCV genotype 1, 4 and 6 patients. Patients infected with HCV genotype 1 were randomized to 12 or 24 weeks sofosbuvir, PEG-IFN/RBV, or 12 weeks of the triple combination followed by re-randomization (1:1) to receive additional 12 weeks of either sofosbuvir alone or sofosbuvir plus RBV. Also, 16 patients infected with HCV genotypes 4 and 6 were randomized to the 24-week regimen of sofosbuvir plus PEG-IFN/RBV. Results of an interim analysis showed SVR rates of 90% in patients who received 12 weeks of the triple combination .
The phase III trial NEUTRINO, a single-arm study, evaluating a 12-week course of sofosbuvir plus PEG-IFN/RBV in 300 patients infected with HCV genotypes 1, 4, 5 and 6 is currently ongoing.
Nucleos(t)ide inhibitors in clinical trials without interferon
Nucleos(t)ide analogues were also investigated in combination with RBV, NS3/4A protease inhibitors, NS5A inhibitors and/or other nucleotide inhibitors in IFN-free treatment regimens.
Mericitabine (RG 7128)
The INFORM-1 study provided first proof of principle that suppression of HCV RNA with an interferon-free combination of mericitabine and danoprevir (NS3/4 protease inhibitor) is effective. Different doses of mericitabine in combination with danoprevir were administered for 14 days to patients infected with HCV genotype 1. At the highest combination doses tested (1000 mg mericitabine and 900 mg danoprevir BID), the median change in HCV RNA concentration from baseline to day 14 was −5.1 log(10) IU/ml in treatment-naive patients and −4.9 log(10) IU/ml in previous null-responders to PEG-IFN/RBV therapy.
INFORM-SVR, a phase 2b trial, subsequently investigated a 12 or 24 week interferon-free regimen of ritonavir-boosted danoprevir (DNV/r, 100/100 mg), and mericitabine (1000 mg BID) with or without RBV in treatment-naive HCV genotype 1 infected patients. Patients with undetectable HCV RNA at week 2 and week 10 were re-randomized at week 12 to stop therapy or to continue until week 24. Rapid viral response was similar and comparable in both arms (91 and 93% respectively). Owing to high relapse rates, however, the 12-week treatment arm was prematurely stopped. Furthermore, patients randomized to the RBV-free group were offered to continue on PEG-IFN/RBV therapy because of insufficient SVR rates. The data show SVR in 71% of HCV genotype 1b patients, but only in 26% of the genotype 1a-infected patients who received 24 weeks of DNV/r, mericitabine and RBV treatment. Higher SVR rates were reported among patients who were rapid virological responders. Among patients with undetected HCV-RNA at week 2, 80% with genotype 1b and 31% with genotype 1a achieved SVR. IL28B genotype appeared to have less impact on SVR rates. Breakthrough rates were higher among patients not receiving RBV and showed in all patients, the selection of danoprevir-resistant variants, while only one patient showed the NS5B S282T polymerase mutation associated with resistance to mericitabine. Neutropenia or treatment emergent liver toxicity was not observed .
The currently ongoing phase II study ANNAPURNA evaluates the safety, tolerability, and efficacy of combination treatment with mericitabine (1000 mg BID), ritonavir-boosted danoprevir (DNV/r 100/100 mg), setrobuvir (ANA-598) and RBV administered for 12–14 or 24–26 weeks to treatment-naive patients or for 24–26 weeks to non-responders to PEG-IFN/RBV therapy with chronic hepatitis C genotype 1 infection.
In the ELECTRON trial, the interferon-free combination of sofosbuvir and RBV achieved 100% SVR rates in treatment-naive patients infected with HCV genotype 2/3. In subsequent cohorts, patients with HCV genotype 1 infection and prior non-response to PEG-IFN/RBV received sofosbuvir and RBV for 12 weeks, but showed high relapse rates after the end of 12 weeks of treatment (89%). Reported SVR4 rates in treatment-naive patients infected with genotype 1 or treatment-experienced patients infected with genotype 2/3 receiving sofosbuvir and RBV for 12 weeks were 88 and 80% respectively .
The QUANTUM study, a phase 2b study, was planned as the first interferon-free, pan-genotypic, all-nucleotide study, combining a pyrimidine and a purine analogue, sofosbuvir and PSI-938 with or without RBV for 12 weeks. However, PSI-938 had to be discontinued because of liver toxicity. An SVR rate of 59% was reported for 17 HCV genotype 1 patients .
Three ongoing US studies for patients with HCV genotype 2/3 infection are FISSION, POSITRON, and FUSION. FISSION will compare 12 weeks of treatment with sofosbuvir and RBV to 24 weeks of PEG-IFN/RBV therapy. POSITRON evaluates the same combination for 12 weeks in interferon ineligible or intolerant patients. FUSION will examine treatment-experienced patients exploring 12 or 16 weeks of treatment with sofosbuvir and RBV.
Furthermore, a phase 3 study (VALENCE) was recently started to investigate the combination of sofosbuvir and RBV for 12 weeks in treatment-naive and treatment-experienced patients with HCV genotype 2 and 3 infection also in Europe.
Different oral combinations of sofosbuvir (400 mg QD) plus daclatasvir (BMS-790052, 60 mg QD) with/without RBV were tested in an open-label Phase 2a pharmaceutical cross-collaboration in treatment-naive non-cirrhotic HCV genotype 1-, 2- and 3-infected patients. SVR rates were 88, 100, and 79% respectively . Furthermore, a combination study of simeprevir (TMC435) plus sofosbuvir is ongoing and a phase III study, investigating the combination of sofosbuvir and GS-5885 (NS5A inhibitor), is currently in preparation.
Several non-nucleoside inhibitors were discontinued from clinical development for different reasons. BILB1941 and MK-3281 were discontinued because of gastrointestinal side events [23-25], VX-759 and VX-916 showed only low-to-medium antiviral activity [26, 27], nesbuvir (HCV-796) and IDX-375 showed elevation of liver enzymes .
Currently, BI 207127, VX-222, ABT-072, ABT-333, setrobuvir (ANA-598), filibuvir and tegobuvir (GS-9190) are still investigated in clinical trials.
Non-nucleoside inhibitors in clinical trials with interferon
The SOUND C1 trial evaluated the combination of the non-nucleoside polymerase inhibitor BI 207127 with the protease inhibitor BI 201335 and RBV for 4 weeks followed by BI 201335 and PEG-IFN/RBV in treatment-naive patients with HCV genotype 1 infection. The RVR rates ranged from 73% (genotype 1a) to 100% (genotype 1b). An SVR rate of 94% was achieved in the group receiving a higher dose of BI 207127 (600 mg; TID) .
The Phase 2 ZENITH study was designed to evaluate the antiviral activity of the protease inhibitor telaprevir and two dose levels of VX-222 administered with or without PEG-IFN/RBV for 12 or 24 weeks in treatment-naive patients with genotype 1 infection. This quad regimen led to SVR rates of 82–93% .
Setrobuvir was studied at different doses in combination with PEG-IFN and RBV in treatment-naive and treatment-experienced patients with HCV genotype 1. Both doses of setrobuvir demonstrated a favourable safety and tolerability profile through 12 weeks. SVR was achieved in 8 of 11 patients (73%) who received 24 weeks of treatment .
The antiviral activity of tegobuvir and the NS3/4A protease inhibitor GS-9256 alone or in combination with RBV or PEG-IFN/RBV in treatment-naive patients with HCV genotype 1 was assessed in a phase II trial. Patients were randomized to 28 days of tegobuvir/GS-9256, tegobuvir/GS-9256/RBV or tegobuvir/GS-9256/Peg-IFN/RBV, followed by PEG-IFN/RBV for further 44 weeks. Rapid viral response was observed in 100% (14 of 14) of patients receiving tegobuvir/GS-9256/PEG-IFN/RBV . Response-guided quadruple therapy with tegobuvir and GS-9256 plus PEG-IFN/RBV based on a virological response at week 2 enabled to shorten antiviral therapy to 16 weeks and yielded an SVR rate of 95% .
Because of pancytopenia in quadruple therapy combinations including tegobuvir, the polymerase inhibitors GS-9256 or GS-9451, Peg-IFN and RBV, dosing of tegobuvir was discontinued .
Non-nucleoside inhibitors in clinical trials without interferon
SOUND-C2 (n = 362), an open-label, randomized, phase IIb trial investigated safety and efficacy of the combination faldaprevir (BI 201335; protease inhibitor), BI 207127 with or without RBV for 16, 28, 40 weeks in genotype 1-infected treatment-naive patients, including 10% of patients with compensated cirrhosis. Patients were randomized to one of five treatment arms: (A) 120mg QD BI 201335 (1335QD) plus 600mg TID BI 207127 (7127TID) and RBV for 16 weeks, (B) 28 weeks, or (C) 40 weeks; (D) 1335QD plus 600mg BID BI 207127 (7127BID) and RBV for 28 weeks; (E) 1335 plus 7127TID (no RBV) for 28 weeks. Randomization to Arm E was stopped prematurely according to FDA feedback on study design prior to data analysis. The RBV-sparing arm showed substantial but lower response rates than other arms of the trial.
This combination therapy achieved a SVR rate of 60% in treatment-naive patients with genotype 1 after 16 weeks of treatment, confirming the potent antiviral activity of this treatment regimen. In patients with liver cirrhosis, an SVR rate of 60% in HCV genotype 1a and up to 83% in HCV genotype 1b-infected patients was reported. The BID (for BI 207127) regimen demonstrated the most favourable safety and tolerability profile with a low rate of adverse event-related study discontinuations. Post-hoc subanalyses by IL28B genotype showed clinically important differences in SVR rates: HCV 1a (IL28B: non-CC) 32% vs. HCV 1a (IL28B: CC) 75% and HCV 1b (IL28B: CC and non-CC) 82–84% .
ABT-072 and ABT-333
In the PILOT trial, 11 treatment-naive, non-cirrhotic HCV genotype 1 infected patients expressing the IL28B CC genotype were treated with ABT-450/r (150/100 mg QD), ABT-072 (400 mg QD) and RBV for 12 weeks. The SVR rate was 91%.
Another study (Co-Pilot) investigated ABT-450/r in combination with ABT-333 and RBV in 33 treatment-naive HCV genotype-1 infected patients and 17 non-responders to previous PEG-IFN/RBV therapy. Therapy-naive patients were treated with two different doses of ABT-450/r (250/100 mg and 150/100 mg, Arm 1 and 2 respectively) while prior non-responders received open-label ABT-450/r (150/100 mg, Arm 3). ABT-333 was administered in a fixed dose of 400 mg QD across groups. SVR rates in treatment-naïve patients were independent of ABT-450/r dose and IL28B genotype (95 and 93% for arm 1 and 2 respectively. In previous non-responders (Arm 3), the SVR rate was 47% .
The phase 2 ZENITH study was designed to evaluate the antiviral activity of the protease inhibitor telaprevir and VX-222 in a dual regimen, in addition with RBV as triple or with PEG-IFN/RBV as quadruple combination therapy in treatment-naive patients with genotype 1 infection. A proportion of 38% of patients receiving 100 mg VX-222 plus telaprevir and 50% of patients receiving 400 mg VX-222 plus telaprevir were able to undergo only 12 weeks of therapy because of RVR (undetectable HCV-RNA at week 2 and 8) and achieved SVR rates of 83 and 90% respectively .
Another interferon-free study, investigating the combination of tegobuvir, GS-5885 (NS5A inhibitor), GS-9451 (NS3/4A protease inhibitor) and RBV for 12 and 24 weeks in treatment-naive and experienced HCV genotype 1 patients demonstrated SVR rates up to 89% in patients treated for 12 weeks and up to 100% in patients treated for 24 weeks. Viral breakthrough occurred only in GT1a-infected patients .
The optimal treatment regimens for patients with chronic hepatitis C must still be defined. In many trials, nucleos(t)ide and non-nucleoside polymerase inhibitors are key components. The high resistance barrier of the nucleoside analogue inhibitors provides a potential backbone in various interferon-free treatment combinations. The non-nucleoside inhibitors have complementary resistance profiles to protease inhibitors and NS5A inhibitors and are also positioned to become relevant components in combination trials. In the near future, polymerase inhibitors as key components could be clinically available if ongoing clinical trials continue to show promising results on efficacy, safety and tolerability.