| |
Ribavirin's antiviral effect aids interferon in hepatitis C therapy
|
| |
| |
NEW YORK (Reuters Health) - In patients with chronic hepatitis C, ribavirin improves the effectiveness of interferon alpha treatment. The mechanism that underlies the combined effect of the two drugs is unclear, however.
In the March issue of Gastroenterology, an international team led by Dr. Jean-Michel Pawlotsky of Hopital Henri Mondor in Paris, France, sheds some light on this mechanism with the results of its analysis of viral kinetics and ribavirin pharmacokinetics in patients with chronic hepatitis C.
Thirty-eight patients received either ribavirin or interferon monotherapy, or interferon and ribavirin in combination. Seven patients served as controls.
Of the 7 patients who received ribavirin alone, 4 had "a significant, moderate, early and transient viral load decrease" at days 2 and 3, according to the researchers. This effect was associated with longer ribavirin clearance half-lives and higher serum concentrations.
In the 8 patients treated with the standard 3-times-per-week interferon injections plus daily ribavirin, the antiviral effect of ribavirin partly reduced the viral load rebound that preceded the second injection. "The magnitude of the rebound was inversely related to ribavirin concentrations," the investigators point out.
After the viral load rebound, patients receiving the standard interferon regimen in combination with ribavirin had "a slow, but significant, second slope of viral decrease and cleared HCV RNA." Patients on ribavirin monotherapy, however, did not experience a second decline in either viral level or HCV RNA clearance.
The authors concede that the mechanisms underlying their observations "remain hypothetical." They conclude that ribavirin's transient antiviral effect "helps explain the better initial and end-of-treatment response when ribavirin is added to interferon alpha."
Gastroenterology 2004;126:703-714.
Text from published article:
Antiviral action of ribavirin in chronic hepatitis C
Jean-Michel Pawlotsky
Gastroenterology, March 2004
ABSTRACT
In the patients with chronic hepatitis C, the addition of ribavirin to interferon (IFN)- significantly increases the virologic responses. Our aim was to assess the antiviral action of ribavirin on hepatitis C virus (HCV) as a function of ribavirin pharmacokinetics and to evaluate the influence of this antiviral effect on IFN- efficacy.
Forty-five patients with chronic hepatitis C (genotype 1b) received various schedules of IFN- and/or ribavirin administration. Frequent blood sampling was performed for HCV RNA kinetics and ribavirin pharmacokinetics assessment.
Ribavirin monotherapy induced a significant, moderate, early, and transient viral load decrease in approximately half of the patients. The occurrence of this effect was associated with longer ribavirin clearance half-lives and higher serum ribavirin concentrations. Ribavirin antiviral effect partly reduced the rebound preceding the second IFN- injection in patients receiving standard IFN- 3 times per week plus ribavirin. The magnitude of the rebound was inversely related to ribavirin concentrations. These patients subsequently experienced a slow, but significant, second slope of viral decrease and cleared HCV RNA. The addition of ribavirin to daily IFN- monotherapy did not have any impact on the second phase of viral decline.
The authors concluded that ribavirin exerts a significant, moderate, and transient antiviral effect in a significant proportion of patients with chronic hepatitis C. The antiviral effect of ribavirin correlates with ribavirin pharmacokinetics and is partly responsible for the improved efficacy of the combination of standard IFN- and ribavirin compared with IFN- monotherapy by increasing the incidence of the initial response.
BACKGROUND
Hepatitis C virus (HCV) infects over 170 million individuals worldwide. Chronic hepatitis C also represents an important public health threat and economic burden because it is responsible for a high incidence of cirrhosis and hepatocellular carcinoma and accounts for approximately 30% of all patients undergoing liver transplantation. Antiviral treatment of chronic hepatitis C is aimed at preventing the development of these complications. Therapy is currently based on the use of interferon (IFN)--based regimens, which may induce a sustained virologic response (SVR), characterized by eradication of serum HCV RNA detection with sensitive assays 24 weeks after treatment withdrawal. Sustained responses were recently shown to correspond with a probable definitive cure of infection in approximately 98% of cases.
Historically, IFN--based therapy in chronic hepatitis C was initially based on the administration of 3 megaunits of standard IFN- 3 times per week for 24 weeks. With this schedule, a SVR was obtained in approximately 2% of the patients infected with HCV genotype 1 and 16% of the patients infected with HCV genotype non-1. A 48-week administration of the same IFN- dose increased SVR rates to 9% in genotype 1 and 30% in genotype non-1 patients, respectively. The enhanced efficacy of longer treatment was later explained by viral and cellular kinetics studies in which the mean half-life of infected liver cells ranged up to 70 days. Ribavirin is a synthetic guanosine analogue that has been used principally in the past for the treatment of severe respiratory syncitial virus infections in infants. Ribavirin has been reported to have an antiviral effect in vitro on a number of RNA and DNA viruses. The drug was shown to inhibit in vitro the replication of bovine viral diarrhea virus and of GB virus B, two members of the Flaviviridae family closely related to HCV. In addition, ribavirin was recently shown to weakly inhibit HCV RNA-dependent RNA polymerases from all 6 HCV types and the subgenomic replicon system in vitro. Ribavirin monotherapy has been administered to patients with chronic hepatitis C and associated with a significant (but transient) effect on serum alanine aminotransferase (ALT) activity and no apparent effect on HCV RNA levels. In spite of the apparent lack of antiviral action of ribavirin in vivo, pilot studies suggested that the combination of IFN- and ribavirin could improve sustained response rates. Large clinical trials subsequently indicated that the two principal effects of the addition of ribavirin to 3 megaunits 3 times per week of IFN- were (1) to significantly increase the initial and end-of-treatment virologic responses (patients who became HCV RNA-negative during treatment) and (2) to increase significantly the SVR as a result of a marked decrease of the relapse rate. SVR rates were enhanced to 17% and 29% in genotype 1-infected patients receiving the combination for 24 weeks and 48 weeks, respectively, and 62% in genotype non-1 infected patients whatever the treatment duration.
However, the mechanisms underlying this beneficial effect of ribavirin on IFN--based therapy remain unknown. This effect was suggested to be principally mediated by the immune system, based on observations that ribavirin could induce a shift toward T-helper 1 (Th1) response and/or suppress HCV-specific interleukin-10 production.
More recently, pegylated IFN- molecules with long-lasting antiviral and immunomodulatory effects allowing weekly administration have been developed. Pegylated IFN- monotherapy is more efficient than 3 times per week standard IFN- monotherapy but less than the combination of standard IFN- 3 times per week and ribavirin. The addition of ribavirin to pegylated IFN- has further improved efficacy, increasing the SVR rates to 42% and 46% in patients infected with genotype 1 and 82% and 76% in those with genotype non-1 infection using pegylated IFN- 2b (PegIntron; Schering-Plough Oncology, Kenilworth, NJ) and pegylated IFN- 2a (Pegasys; Hoffmann-La Roche, Basel, Switzerland), respectively. The combination of pegylated IFN- and ribavirin is now the standard best treatment for chronic hepatitis C.
To understand the potential mechanisms underlying the effect of ribavirin addition to IFN--based treatment in chronic hepatitis C, we used frequent serial sampling during therapy to assess the intrinsic antiviral action of ribavirin on HCV replication in relation with ribavirin pharmacokinetics and to evaluate the influence of this antiviral effect on IFN- efficacy.
PATIENTS & METHODS
Forty-five consecutive treatment-naive patients with chronic hepatitis C infected with HCV genotype 1b (INNO-LiPA HCV II; Innogenetics, Gent, Belgium) were included in this study. There were 30 men and 15 women; their mean age was 46 ± 11 years (range, 26--69 years). The patients were randomized at inclusion to receive 1 of the following schedules for 24 weeks:
--group A: no treatment, control group (n = 7 patients);
--group B: ribavirin (Rebetol; Schering-Plough Oncology), 1.0 to 1.2 g daily according to body weight (patients <75 kg received 1.0 g daily and patients >75 kg received 1.2 g daily) orally (n = 7 patients);
--group C: IFN- 2b (Intron-A; Schering-Plough Oncology), 3 megaunits 3 times per week subcutaneously (n = 8 patients);
--group D: IFN- 2b, 3 megaunits 3 times per week plus ribavirin, 1.0 to 1.2 g daily (n = 8 patients);
--group E: IFN- 2b, 3 megaunits daily (n = 9 patients);
--group F: IFN- 2b, 3 megaunits daily plus ribavirin, 1.0 to 1.2 g daily (n = 6 patients).
At the end of the 24-week study period, all 45 patients then received an additional 24 weeks of treatment with the combination of IFN- 2b, 3 megaunits 3 times per week plus ribavirin, 1.0 to 1.2 g daily. They were then followed for an additional 24 weeks.
Frequent blood sampling was performed for HCV RNA detection and quantification. Blood samples were taken 14 days, 2 hours, and just before the first IFN- injection or ribavirin intake. During the first few days, sampling was performed at the following hours after treatment initiation: 0, 4, 8, 12, 18, 24 (day 1); 30, 36, 42, 48 (day 2); 60, 72 (day 3); and 84 and 96 (day 4). Finally, later samples were taken at week 2, week 4, week 12, and week 24 of treatment; at the end of treatment (week 48); and at the end of follow-up (week 72). Peripheral blood levels of ribavirin were measured in the same samples during the first 24 weeks of therapy. Ribavirin pharmacokinetics were correlated with viral kinetics.
RESULTS
Effect of ribavirin monotherapy
In group B (patients receiving ribavirin monotherapy for 24 weeks), a moderate and early (i.e., occurring at days 2 and 3) mean viral load decrease was observed. This effect was transient and disappeared after 4 days. The mean viral load drop was significant relative to the control group (P < 0.01), showing that ribavirin exerts a transient and moderate but significant antiviral effect on HCV. The individual patient viral load kinetics show that ribavirin exerted this antiviral action in 4 out of 7 patients.
In these patients, the viral load decrease ranged from --0.5 to --1.6 log10 IU/mL, and viral load returned to pretreatment levels after 12 to 96 hours in all 4 patients. None of these patients experienced a second slope of viral decrease. The remaining 3 patients experienced minor viral load fluctuations similar to those in untreated patients. They did not differ from patients B1 to B4 by age, sex, or baseline viral load. Overall, no patient from group B cleared HCV RNA during ribavirin monotherapy. A relationship was found between ribavirin pharmacokinetics and the occurrence of an HCV RNA decline during ribavirin monotherapy. Indeed, the 4 patients with a viral decline had longer ribavirin clearance half-lives (nearly significant, P < 0.07), higher ribavirin levels at 30 hours (P < 0.07), and significantly higher ribavirin levels at 62 hours (P < 0.03) than the 3 patients with no viral decline.
Effect of ribavirin in the patients receiving standard IFN- 3 times per week
Among the 8 patients receiving IFN- monotherapy at the dose of 3 megaunits 3 times per week, 3 did not respond, i.e., viral load did not decrease by more than --0.5 log10 IU/mL after the first IFN- injection.
...... Six out of the 8 patients receiving the combination of IFN- 3 times per week plus ribavirin daily (group D) responded to the first IFN- injection...... At the second day of treatment, the moderate antiviral effect of ribavirin partly reduced the rebound in 5 responders relative to that observed in the patients receiving 3 times per week IFN- monotherapy..... Viral kinetics in these patients were significantly related to ribavirin pharmacokinetics.....
Effect of ribavirin in the patients receiving daily standard IFN-
All 9 patients receiving daily IFN- monotherapy responded to IFN-. As in the patients receiving 3 times per week IFN- administration, a short lag of a few hours after the first IFN- injection preceded the significant, more than --0.5 log10 decrease at day 1. The mean replication blocking effectiveness of IFN- was 93% (range, 72%--100%) in this group, not different from groups C and D receiving IFN- 3 times per week, alone or in combination with ribavirin, respectively. However, in contrast with patients receiving IFN- 3 times per week, none of the patients receiving daily IFN- experienced a viral load rebound at day 2. All of them then exhibited a second slope of viral decrease. Accordingly, the median viral load at day 3 to minimum viral load during the first 48 hours ratio was 0.8 (range, 0.2--9.3), i.e., significantly less than in patients receiving IFN- 3 times per week. The second phase slope of viral decrease was significantly slower than the first phase slope but significantly faster than the second phase slope in the patients receiving IFN- 3 times per week. It led to HCV RNA clearance in all instances at the end of the 24-week treatment period. In group F (patients receiving IFN-, 3 megaunits daily plus ribavirin, 1.0 to 1.2 g daily), 4 patients out of 6 responded to IFN-. HCV viral load decline did not differ from the patients receiving daily IFN- monotherapy. It was indeed characterized by a biphasic viral load decline. At day 1, the mean percentage blocking effectiveness was 93% (range, 85%--99%), not different from the other groups receiving IFN- 3 times per week or daily. Although the IFN- blocking effectiveness at day 1 did not significantly differ between the patients receiving IFN- with (groups D and F) or without ribavirin (groups C and E), we found a significant relationship between the area under curve of ribavirin concentrations between 0 and 12 hours and HCV RNA load reduction after the first 24 hours of therapy in the patients from both groups D and F (r = --0.65, P < 0.015).
In the patients from group F receiving IFN- daily plus ribavirin, as in the patients from group E receiving IFN- daily, there was no rebound after the first IFN- injection. As a result, ribavirin pharmacokinetics had no significant effect on the viral load at day 3/minimum viral load during the first 48 hours ratio, which was not different from that observed in the patients from group E treated with IFN- daily only. A second phase slope of viral decrease starting at day 2 was thus observed in all of the responders from this group. It was faster than in the groups receiving 3 times per week IFN-, but not different from that in group E, receiving daily IFN- without ribavirin. Overall, the addition of ribavirin to daily IFN- therapy did not appear to have any impact on the second phase of HCV viral kinetics. One responder patient was lost to follow-up, and the remaining 3 patients were HCV RNA negative at the end of the initial 24-week period of treatment (Table 2).
DISCUSSION
In this study, we showed that ribavirin exerts a significant antiviral effect in approximately half of the treated chronic hepatitis C patients. This antiviral effect is moderate (--0.5 to --1.6 log10 reduction) and transient, occurring at days 2 and 3. Subsequently, no patient achieved a second slope of viral decrease nor HCV RNA clearance, confirming the lack of efficacy of ribavirin monotherapy as an antiviral treatment in chronic hepatitis C. Ribavirin's early, transient antiviral effect was not seen in a previous study, probably because this work did not include a control group of untreated patients and had less frequent blood samplings. The study of ribavirin pharmacokinetics allowed us to explain that only 4 out of 7 patients experienced this ribavirin early antiviral effect. Our study is indeed the first to correlate early viral kinetics with ribavirin pharmacokinetics in patients receiving ribavirin alone or in combination with IFN-. The Cmax and area under curves of ribavirin concentrations were in keeping with previous findings. However, ribavirin half-lives were smaller in our study than those reported by Tsubota et al., maybe because of the recently reported first-elimination phase of the drug. As already shown, the plateau level was achieved on average 28 days after the first intake. However, high serum levels of ribavirin were achieved a few hours after the first dosing and maintained thereafter, although undergoing fluctuations. This finding makes possible a pharmacologic effect of ribavirin within the first few days of administration, provided that intracellular levels correlate with serum levels, a parameter that could not be analyzed here. Our observation of longer ribavirin half-lives and higher ribavirin levels at 30 and 62 hours in the patients who had a viral decline during ribavirin monotherapy than in those with no decline (Figure 4) shows that the observed antiviral effect is related to the drug and needs that certain blood levels be achieved to occur, regardless of other patient characteristics. Our observation that this effect was transient in spite of constantly high and growing serum ribavirin levels suggests that, whatever the mechanism(s) involved in ribavirin antiviral action, this effect is quickly reversible through down-regulation mechanisms that could not be explored in this study.
The mechanisms underlying the observed transient antiviral effect of ribavirin in vivo are unclear. Ribavirin, a synthetic guanosine analogue, could directly inhibit the HCV RNA-dependent RNA polymerase at the catalytic site level. However, the slow, delayed viral decline during ribavirin monotherapy, together with the lack of a difference in virus production blocking effectiveness between IFN- monotherapy and IFN--ribavirin combination in this study, make this hypothesis unlikely. On the other hand, the broad-spectrum ribavirin antiviral activity was suggested to be related to the inhibition of viral genome synthesis through depletion of intracellular GTP pools caused by ribavirin inhibition of the inosine monophosphate dehydrogenase (IMPDH) enzyme. Nevertheless, other potent IMPDH inhibitors, such as mycophenolic acid or VX-497, do not exert a significant effect on HCV replication in infected patients, questioning the actual influence of IMPDH inhibition on ribavirin antiviral action in vivo. Ribavirin was also recently shown in vitro to be an RNA virus mutagen. As a result, ribavirin could potentially exert its antiviral action by incorporating into newly synthesized genomes and inducing lethal mutations leading to "error catastrophe" (the generation of nonviable viral quasispecies populations).
Interestingly, ribavirin was also recently shown to strongly inhibit replication of GB virus B, the closest member of the Flaviviridae family to HCV, in primary cultures of tamarin hepatocytes, an antiviral effect associated with error-prone replication of GB virus B. In addition, Contreras et al. recently observed that ribavirin accelerates the accumulation of mutations on HCV genome in an in vitro HCV replication system. The demonstration of an accelerated accumulation of HCV mutations during ribavirin therapy remains to be proven. Although such an accumulation might be masked in vivo by the many pressures on HCV genome sequence evolution, we recently failed to show any acceleration of mutation accumulation on HCV genomes in patients receiving ribavirin monotherapy (Pawlotsky JM et al., manuscript in preparation). Understanding why ribavirin's antiviral effect is only transient will require that the mechanisms of ribavirin antiviral effect be better understood.
The transient and moderate antiviral effect of ribavirin at days 2 and 3 that we report helps explain the better initial and end-of-treatment response when ribavirin is added to IFN- 3 times per week. Although ribavirin did not increase IFN- production blocking effectiveness (i.e., did not induce larger first-phase decline), the additive effect of its antiviral action at day 2 partly reduced the rebound preceding the second IFN- injection in a significant number of patients. This allowed viral load to be lower than in the absence of ribavirin at the time of the second IFN- injection in these patients and led to a slow but significant second slope of viral decrease followed by HCV RNA clearance, a necessary, although insufficient, condition to achieve an SVR. Overall, this suggests that the better efficacy of the combination of IFN- 3 times per week and ribavirin is partly related to the moderate antiviral effect of ribavirin. Again, this conclusion is reinforced by our pharmacokinetic observation that ribavirin levels and the areas under curves of ribavirin concentrations were inversely related with the magnitude of the rebound after the initial antiviral effect of IFN- at day 1. Our data cannot show whether this additive antiviral effect continues over a longer period of time, but the simulation of the model using the early data indicates that it may. It is not clear, however, that this effect can completely explain the difference in sustained virologic response rates between IFN- monotherapy and IFN--ribavirin combination, in particular the reduced relapse rate after treatment withdrawal. Later ribavirin-induced and/or IFN--related and ribavirin-enhanced immune mechanisms leading to an accelerated late infected cell clearance (that was not studied here) probably also play an important role.
Daily IFN- monotherapy was associated with a typical biphasic viral load decline, including a first rapid phase at day 1 and a second, slower but significant phase starting thereafter and leading to HCV RNA clearance during therapy, as already reported. Surprisingly, among the patients receiving IFN- plus ribavirin, viral reduction within the first 24 hours was significantly related to ribavirin accumulation between 0 and 12 hours. The data presented here do not support a direct antiviral effect of ribavirin at day 1 of administration, in spite of already high circulating levels. Whether the latter finding was related to individual pharmacologic parameters also influencing IFN- pharmacokinetics remains to be elucidated. In contrast to the patients receiving IFN- 3 times per week, ribavirin had no impact on viral kinetics after the first day of treatment in the patients receiving daily IFN-. This finding was not surprising because the patients receiving daily IFN- are subjected to a constant IFN- pressure, in contrast with those receiving IFN- 3 times per week.
This suggests that the antiviral effect of ribavirin does not play an important role in viral clearance during treatment in patients receiving daily IFN- injections. It cannot be excluded, however, that the antiviral effect of ribavirin plays a role in preventing relapse. Daily IFN- therapy was shown to accelerate viral load decrease and improve end-of-treatment virologic response but did not improve the SVR rate in several trials. This was probably because of the fact that the patients received daily IFN- treatment for only a few weeks in most cases and were then switched to 3 times per week IFN- therapy. Results from studies of prolonged periods of daily IFN- injections with or without ribavirin would be needed to understand the potential role of ribavirin antiviral action on relapse in these patients.
The recent development of pegylated IFN- molecules that can be administered once weekly and maintain a plateau effect until the following injection now allow more continuous IFN- pressure over longer periods. Recent data suggest that the weekly administration of pegylated IFN- results in a viral decline pattern that is not always biphasic but rather shows a partial viral rebound toward the end of the first week, especially in genotype 1-infected patients. Although this rebound is not as drastic as that observed at day 2 with 3 times per week standard IFN- injections, our results with the combination of 3 times per week IFN- and ribavirin suggest that the antiviral effect of ribavirin could also play a role in the improved efficacy of combined pegylated IFN- and ribavirin relative to pegylated IFN- monotherapy. Immune mechanisms accelerating infected cell clearance are also likely to play an important role in enhancing the SVR rates in these patients. A third-phase decay was recently observed in patients treated with pegylated IFN- and was suggested to be faster when ribavirin was added to pegylated IFN- than when it was not. In the present study, we did not observe any acceleration of HCV RNA clearance after day 14 of therapy in the patients receiving ribavirin in combination with standard IFN- 3 times per week or daily. However, the difference could be due to the different forms and pharmacokinetics of the IFN- molecules. The role of ribavirin antiviral effect and pharmacokinetics in this putative third-phase decay would deserve to be studied.
In conclusion, we have shown that ribavirin exerts a significant, moderate, and transient antiviral effect in approximately half of the patients with chronic HCV infection and that this effect correlates with longer ribavirin half-life and higher serum concentration. The mechanisms underlying this antiviral effect and its transient character remain hypothetical. The antiviral effect of ribavirin appears to be partly responsible for the improved efficacy of the combination of 3 times per week standard IFN- and ribavirin compared with 3 times per week IFN- monotherapy by increasing the incidence of the initial response, and this effect correlates with higher ribavirin levels and areas under curves. In contrast, ribavirin antiviral effect does not appear to play a significant role on second-phase viral decline when it is added to daily IFN- therapy. Further analysis of viral kinetics during pegylated IFN- treatment with or without ribavirin is needed to understand the impact of ribavirin antiviral effect in this setting.
|
|
| |
| |
|
|
|
