The treatment of hepatitis C virus (HCV) has undergone a dramatic evolution over the past 2 decades. Currently, the sustained virological response (SVR) rates with peginterferon (PEG-IFN)and ribavirin (RBV) are approximately 40% to 46% for genotype I and 80% to 82% for genotypes II and III.1–3 The most problematic side effect of RBV remains reversible hemolytic anemia, which significantly affects patients' quality of life.4, 5 While the etiology of anemia is multifactorial, the RBV dose–dependent hemolytic anemia is due to RBV accumulation in erythrocytes, where it is phosphorylated; depleting adenosine triphosphate (ATP) reserves, which ultimately leads to senescence and erythrophagocytic removal.6 The addition of interferon also suppresses red blood cell (RBC) production, and the anemia seen with the combination of PEG-IFN and RBV therapy can lead to RBV dose reductions and a lower chance of SVR.
A common strategy for treating this anemia is the use of recombinant human erythropoietin-stimulating agents (ESAs). Although this therapy is effective at ameliorating RBV-induced anemia, it adds treatment costs and is associated with rare but life-threatening thromboembolic, cardiovascular, and hematologic events (red cell aplasia).7, 8 Studies have suggested that ESAs raise hemoglobin levels and improve the quality of life, but a significantly improved SVR rate in patients who receive ESAs has been difficult to demonstrate.9, 10 However, a recent study has also suggested that ESA administration in the setting of RBV-related anemia is associated with reduced dropout and higher SVR rates.11 Therapies that decrease the risk of RBV-induced anemia without compromising SVR rates or raising costs are desirable.
Recently, genome-wide association studies have demonstrated 2 important genetic polymorphisms that may allow the refinement of HCV therapy and help with the prediction of those at a higher risk for RBV-induced hemolytic anemia. Although the data require validation by prospective studies, the recent identification of interleukin-28 (IL-28) polymorphisms appears to be a powerful tool for helping us to predict the response to therapy12 The recently reported polymorphisms (rs1127354 and rs7270101) in the inosine triphosphatase (ITPA) gene causing inosine triphosphatase deficiency have been shown to protect against RBV-induced hemolytic anemia during the early stages of treatment.13 The identification of these ITPA variants, similar to the identification of the IL-28 polymorphisms, may provide a valuable pharmacogenetic diagnostic tool and the opportunity for pharmacological intervention that can ameliorate RBV-induced anemia in higher risk individuals.
Reducing anemia rates will become more important as we enter the era of direct-acting antiviral (DAA) agents, which have been shown to improve response rates and shorten the treatment duration in HCV genotype I–infected individuals when they are added to PEG-IFN/RBV. Studies with the nonstructural protein 3/4A protease inhibitors telaprevir and boceprevir have demonstrated that the elimination or reduction of RBV is associated with lower efficacy rates and a greater likelihood of drug-resistant mutations developing.14, 15 In addition, both telaprevir and boceprevir are themselves associated with additional anemia over and above that of RBV. Because PEG-IFN and RBV will remain the backbone of therapy in combination with the first generation of DAA agents for HCV infection, the management of anemia will be of the utmost importance
Taribavirin (TBV), which was previously known as viramidine, is a nucleoside analogue and prodrug of RBV that is converted from TBV to RBV by adenosine deaminase. RBV is concentrated primarily in hepatocytes and less in RBCs, so RBV is concentrated in the liver, and RBV exposure is reduced in erythrocytes.16 An initial phase 2 study compared PEG-IFN alfa-2a (180 μg) with TBV (800, 1200, or 1600 mg) to RBV (1000/1200 mg).17 TBV use was associated with fewer episodes of anemia (hemoglobin level <10 g/dL), and the SVR rate of 37% in a mixed genotype population using 1200 mg of TBV was comparable (not inferior) to the SVR rate of 44% seen in the RBV group. Thus, two phase 3 clinical trials, Viramidine Safety and Efficacy Versus Ribavirin 1 (ViSER1) and ViSER2, were undertaken.18, 19 ViSER1 randomized 972 treatment-naive patients to flat-dose TBV (600 mg twice a day) or weight-based RBV (1000 or 1200 mg/day); each was given with PEG-IFN alfa-2b. The patients were given flat dose of TBV despite evidence that weight-based dosing of RBV is superior to fixed-dose RBV in patients receiving either PEG-IFN alfa-2a or PEG-IFN alfa-2b. Although significantly fewer individuals developed anemia (hemoglobin level <10 g/dL) with TBV versus RBV according to an intention-to-treat analysis, the TBV arm failed to demonstrate noninferiority to weight-based RBV with a reduced SVR rate of 37.7% in the TBV arm versus 52.3% in the RBV arm. However, a post hoc retrospective analysis of TBV exposure by body weight showed a beneficial effect on patients who received TBV doses > 18 mg/kg, and this underscored the need for weight-based dosing. In the ViSER2 study, a similar pattern was seen in 962 patients with an SVR rate of 55% in the weight-based RBV–treated groups versus 40% in the flat-dose TBV–treated groups. Again, a post hoc analysis noted improved efficacy with higher TBV exposure. Lighter patients fared better than heavier ones, and patients who received TBV doses > 15 mg/kg achieved SVR rates close to 50%, whereas only 25% of those with TBV exposure levels ≤ 13 mg/kg achieved an SVR. Overall, patients treated with fixed-dose TBV did not achieve adequate drug exposure and had lower SVR rates. These trials suggest that flat-dose TBV can reduce anemia but at the expense of lower SVR rates. In addition, RBV was associated with greater rates of fatigue, neutropenia, and pyrexia in comparison with TBV, whereas TBV was associated with a greater incidence of diarrhea. TBV also necessitated fewer dose reductions or interruptions due to adverse effects in comparison with RBV in the ViSER2 study.
In this issue of Hepatology, Poordad and colleagues20 report the SVR rates of naive HCV genotype I–infected patients receiving weight-based TBV or weight-based RBV. In this US phase 2b, randomized, open-label, controlled, parallel-group study, 278 naive genotype I subjects were randomized to TBV (20, 25, or 30 mg/kg/day) or RBV (800-1400 mg) and PEG-IFN alfa-2b for 48 weeks. The early virological response, which was defined as undetectable HCV RNA (<39 IU at week 12) or a 2-log reduction in the baseline HCV RNA level (the primary endpoint of the study), was comparable across all treatment arms. The SVR rate was also preserved across all treatment arms and ranged from 27% to 28%. The overall response rates in this trial were low, although the high percentage of African Americans (20%) and patients with advanced fibrosis may explain the lower SVR rates. It would be interesting to know the IL-28 composition of the treatment population because there may have been a high prevalence of patients with the unfavorable IL-28 CT or TT genotype, and this could also explain in part the low SVR rates.
Although the SVR rates were not different between the treatment arms, a lower relapse rate was seen with an incremental increase in the dose of TBV, and this was similar to that observed with RBV. In addition, the per protocol SVR rates were substantially higher, and this again demonstrated the importance of adherence to therapy for optimal SVR rates in the genotype I population. Lower anemia rates (defined as a hemoglobin level <10 g/dL) were seen in the TBV arms throughout the entire period of 48 weeks, and the cumulative anemia rates were significantly lower across the two lower TBV arms (20 and 25 mg/kg) versus the weight-based RBV arms. However, because the dose of TBV was increased to 30 mg/kg, the anemia rate was numerically lower than the rate with RBV, but this was not significant except in week 4; this suggests that higher doses of TBV may lead to similar rates of anemia and other side effects observed with RBV. The pharmacokinetic analysis showed that this effect correlated with RBV plasma exposure. Furthermore, within the first 12 weeks of treatment, the period in which maintenance of the dose of RBV has been shown to be most critical, significantly lower rates of anemia were observed with TBV versus RBV (7%-15% versus 24%, respectively), although this translated clinically into comparable but not superior SVR rates in the TBV arms. Even though fewer patients treated with TBV required a dose reduction (13%-28% versus 32% of the patients treated with RBV), it should also be noted that the dropout rates for anemia were not different between the TBV arms and the RBV arms in this study; however, this may have been due to the relatively small sample size. There does appear to have been an increased rate of diarrhea in the TBV arms versus the RBV arms. This may be significant because some DAA agents are also associated with increased gastrointestinal side effects, and as we enter an era in which DAA agents and other drugs are combined, side effects could limit the efficacy of multiple-drug combinations. Finally, although it was not statistically significant, insomnia occurred more often in the TBV arms and should be a side effect of some concern in future trials. Thus, because significantly fewer dose reductions were noted only in the 20 mg/kg TBV arm versus the arms with higher doses of TBV and RBV with similar SVR rates, the dose of 20 mg/kg may also require study in the future with DAA agents.
So what does the future hold for TBV? Phase 2 and ongoing phase 3 trials strongly suggest that DAA agents will be added to PEG-IFN and RBV to obtain higher SVR rates, albeit at the expense of higher rates of anemia and other side effects. Currently, the role of ESAs in the treatment of HCV with DAA agents is not yet precisely defined, although we await the results of ongoing trials. The inclusion of TBV in the HCV armamentarium may serve as an opportunity to combine it with PEG-IFN and DAA agents to reduce the rates of anemia and prevent RBV dose reduction or the introduction of ESAs. Because RBV reduction or removal is associated with increased rates of breakthrough and development of resistance to DAA agents, TBV may have a role in populations particularly sensitive to RBV-related anemia, including those with advanced liver disease, older patients, patients who have undergone liver transplantation, human immunodeficiency virus/HCV–coinfected individuals, and patients with hemoglobulinopathies and chronic renal failure.21 However, with the commencement of several trials using multiple combinations of DAA agents with and without PEG-IFN/RBV and with the development of newer protease inhibitors with potentially lower rates of anemia, the role of TBV remains less precisely defined and could potentially have a finite life cycle. Studies using combinations of DAA agents with PEG-IFN/RBV have been initiated, and these studies will multiply. Whether or not RBV (and TBV) can be eliminated altogether remains to be determined. Particularly for those patients with unfavorable treatment characteristics, RBV may remain a part of our therapeutic armamentarium for years to come; if so, TBV could be an option with the potential to limit toxicity and potentially reduce costs. The ideal study may combine TBV with a DAA agent and PEG-IFN and compare this to RBV to determine if SVR rates can be preserved or improved by the minimization of dose reductions and the reduction of the emergence of resistance. Because of the long wait between the approval of PEG-IFN and RBV and the yet-to-come approval of DAA agents, we should not discount the potential contribution of TBV. Many promising agents have already been stopped in development because of a lack of efficacy or toxicity.22, 23 Thus, if TBV can be shown to preserve or improve efficacy rates in combination with DAAs and PEG-IFN and bring lower rates of anemia, the use of TBV in these clinical settings would be a welcome addition to the HCV armamentarium as we begin to expand the HCV populations that we treat.1