“True” weight-based dosing versus “flat” dosing of ribavirin: Will the WIN-R please come forward?


  • Jason Smith Pharm.D.,

    1. Greater Los Angeles Veterans Healthcare Administration Hospital, Los Angeles, CA
    Search for more papers by this author
  • Steven-Huy B. Han M.D., A.G.A.F.

    Corresponding author
    1. Associate Clinical Professor, David Geffen School of Medicine at University of California Los Angeles, Los Angeles, CA
    • Pfleger Liver Institute, 200 UCLA Medical Plaza, Suite 214, Los Angeles, CA 90095
    Search for more papers by this author
    • fax: 310-206-4197

  • Potential conflict of interest: Nothing to report.

  • See Articles on Pages 971 and 982

Since the introduction of ribavirin in combination with interferon for the treatment of chronic infection with hepatitis C virus (HCV), the optimal dose of ribavirin has become the Holy Grail for investigators interested in slaying HCV. Prior to 2001, flat doses (FDs) of ribavirin (800 mg daily) were used in combination with interferon alfa-2b or interferon alfa-2a. The concept of weight-based dosing (WBD) of ribavirin originated from a retrospective finding in the pivotal trial of Manns et al.1 The authors found that patients receiving 10.6 mg/kg body weight of ribavirin in combination with peginterferon alfa-2b were more likely to respond to therapy. They postulated that dosing ribavirin according to the patient's total body weight would improve sustained virological response (SVR). Indeed, subsequent studies2, 3 revealed improved SVR in genotype 1 patients receiving ribavirin according to a “standard” WBD scheme (that is, 1000 mg daily for patients weighing <75 kg and 1200 mg daily for patients weighing >75 kg). Further studies examining WBD of ribavirin have used a similar “standard” WBD scheme.4–6 However, the concept of “true” WBD of ribavirin over multiple or continuous weight ranges remained unproven. Accordingly, currently approved therapies recommend “standard” WBD of ribavirin in combination with peginterferon alfa-2a or alfa-2b.


FD, flat dose; HCC, hepatocellular carcinoma; HCV, hepatitis C virus; NIH, National Institutes of Health; SVR, sustained virologic response; WBD, weight-based dosing.

The idea of increasing the dosage of ribavirin has been studied in other ways since the registration trials. A recent pilot study of 10 chronic HCV patients investigated ribavirin dosing determined by measuring serum ribavirin concentrations rather than dosing according to body weight.7 Very high doses of ribavirin were administered to achieve a steady-state plasma concentration of 15 μmol/L of ribavirin (equivalent to an average 2540 mg ribavirin/day, with a range of 1600-3600 mg/day). This concentration was chosen empirically based on the knowledge that a standard daily dose of 800-1200 mg ribavirin would result in a plasma concentration of 8.2 μmol/L. An SVR was achieved in an impressive 9 of 10 patients, although 2 patients required red blood cell transfusions and all required growth factors. Another alternative to dosing ribavirin explored adjusting ribavirin dose to creatinine clearance. This approach is based on the observation that ribavirin is predominantly excreted renally. Separate investigators studied population-based ribavirin pharmacokinetics in relation to renal function. Bruchfeld et al.8 found that ribavirin clearance was linearly dependent on renal function. They proposed a ribavirin dosing scheme based mainly on renal function and body weight. However, conflicting data was reported by Wade et al.9, who could not find an association between ribavirin concentration and creatinine clearance, except possibly at creatinine clearance rates < 34 mL/minute. Therefore, further research will be necessary before creatinine clearance–based dosing of ribavirin can be recommended.

During the time that WBD of ribavirin was evolving, a tandem quandary was challenging investigators in regards to the observation that interferon therapy for chronic HCV infection was not as effective in African American patients as compared to Caucasian patients. In their seminal article from 1999, Reddy et al.10 retrospectively analyzed data from a large consensus interferon trial. When compared to Caucasian patients, African American patients treated for the same duration with consensus interferon monotherapy achieved a lower end-of-treatment response and SVR (end-of-treatment 5% versus 33%, P = 0.04 and SVR 2% versus 12%, P = 0.07). They hypothesized that the lower response rates seen in African American patients were attributable to a significantly greater number of genotype 1 African American versus Caucasian patients in their study (88% versus 66%, P = 0.004). Shortly thereafter, the National Institutes of Health (NIH) responded to the growing body of evidence indicating lower SVRs in African American patients with HCV infection11 by organizing a workshop designed to enumerate the discrepancies between Caucasian and African American patients in order to facilitate the development of strategies directed toward a solution.

Results from 5 trials (1 NIH trial and 4 published trials) were reviewed with respect to their African American representation (Table 1).11–14 The understanding gained from this workshop cast a light on the disparity between HCV infections in African American versus Caucasian patients. Evidence presented at the workshop showed that chronic HCV was 2-3 times more prevalent in African Americans than in Caucasians, and that African Americans had the highest prevalence of HCV among individuals 30-39 and 40-49 years of age.15 Moreover, African Americans were more likely to develop chronic HCV after acquisition compared to Caucasians (90%-95% versus 65%-75%).15 Therefore, disproportionately more African Americans develop cirrhosis, hepatocellular carcinoma (HCC), and death as compared to Caucasians. Finally, the workgroup presented findings indicating that African Americans have a higher prevalence of genotype 1 HCV compared to Caucasians (91% versus 67%),11 predicting poorer treatment outcomes for African Americans.

Table 1. Trials Comparing Sustained Viral Response Rates in African Americans and Caucasian Americans with Chronic Hepatitis C Treated with Interferon Plus Ribavirin Combination Therapy
StudyTreatmentEthnicityNumber of patients% HCV Genotype 1SVR
Interferon doseRibavirin dose
  1. Standard WBD is equivalent to 1000 mg for Total Body Weight <75 kg and 1200 mg >75 kg; “true” WBD is equivalent to 800 mg (<65 kg), 1000 mg (65 kg-85 kg), 1200 mg (85 kg-105 kg), 1400 mg (105 kg-125 kg). Abbreviations: AA, African American; CA, Caucasian American; FD, flat dose; NA, not applicable; PegIFN, pegylated interferon; RBV, ribavirin; SVR, sustained virologic response.

NIH11IFN α-2b 1.5 mcg/kg in escalating doses AA71000
Reddy et al.10IFN α-2b 1.5 mcg/kg vs. Consensus IFN (9 mcg)NAAA40882%
Shiffman et al.12IFN α-2a vs. 4 doses of PegIFN α-2a 180 mcgNAAA13928%
Heathcote et al.13IFN α-2a vs. 2 doses of PegIFN α-2a 180 mcgNAAA11Na9%
McHutchinson et al.14IFN α-2b vs. IFN α-2b plus RBV AA539611%
Jeffers et al.5PegIFN α-2a 180 mcgStandard WBDAA78 26%
Muir et al.6PegIFN α-2b 180 mcgStandard WBDAA100 19%
Conjeevaram et al.4PegIFN α-2a 180 mcgStandard WBDAA196 28%
Jacobson et al. (this issue)PegIFN α-2b 1.5 mcg/kgFDAA387 WBD: 21%
  WBD   FD: 10%

Basic differences in viral kinetics between African American and Caucasian patients on interferon therapy16, 17 have been observed. HCV RNA follows a biphasic decline in the presence of interferon alfa. The first phase decline is rapid and represents interferon-induced inhibition of HCV RNA replication. The second phase is slower and represents declining HCV RNA production related to loss of infected hepatocytes. Both the first and second phase of viral decline is crucial for attainment of SVR. It has been observed that African American patients have an impaired first and second phase response compared to Caucasian patients,16 possibly explaining the ultimate disparity in response rates between the 2 groups. However, the second phase is more predictive of nonresponse when the observed rate of decline is slow. Ribavirin has been postulated to exert an enhanced effect on the second phase of viral decline.18 If this hypothesis is validated, the observed differences in viral kinetics between African American and Caucasian patients could be reconciled by increasing ribavirin dosages.

More recently, 3 prospective studies4–6 unequivocally concluded that genotype alone does not explain the difference in SVR seen between African Americans and Caucasians (Table 1). All 3 studies consisted predominantly of genotype 1 patients receiving combination therapy with pegylated interferon alfa-2a with “standard” WBD of ribavirin. The response rates from these trials suggest that Caucasians are twice as likely as African Americans to respond to combination therapy. Multivariate analysis did not reveal a significant predictor of response among African American patients. Mathematical modeling in the Virahep-C trial4 suggested that higher baseline viral load may be a predictor of response to therapy between African Americans and Caucasians, but this finding needs further evaluation.

In this issue of HEPATOLOGY, Jacobson et al.19 present 2 articles reporting the results of the main WIN-R (weight-based dosing of peginterferon alfa-2b and ribavirin) trial and a post-hoc analysis of a subset of data from the main WIN-R trial that gives us some insight into the efficacy of “true” WBD of ribavirin and how African American patients respond to “true” WBD.

The WIN-R trial was a prospective, multicenter, open-label, investigator-initiated study conducted at 236 community-based and academic-based sites in the United States. Treatment-naïve patients with chronic HCV who had <125 kg body weight were randomized to receive combination therapy for 48 weeks with peginterferon alfa-2b at 1.5 μg/kg and either a daily FD (800 mg) of ribavirin or a “true” WBD (13 mg/kg; patients weighing <65 kg receive 800 mg, 65-85 kg receive 1000 mg, 85-105 kg receive 1200 mg, and 105-125 kg receive 1400 mg) of ribavirin. The primary objective of the study was to evaluate the safety and efficacy of peginterferon alfa-2b in combination with either FD or “true” WBD ribavirin. A secondary analysis was performed to compare the efficacy of 24 versus 48 weeks of therapy for chronic HCV patients with genotype 2/3. The large number of African American patients enrolled in the trial allowed for a sub-hoc analysis to be performed on this important group.

The study found that, based on an intention-to-treat analysis, the overall SVR obtained with peginterferon alfa-2b and “true” WBD of ribavirin was superior to peginterferon alfa-2b and a FD of ribavirin (44.3% versus 40.6%, P = 0.007). The WIN-R trial also found that in combination with peginterferon alfa-2b, “true” WBD of ribavirin is as safe as FD of ribavirin. Discontinuation rates due to adverse events in WIN-R were reported to be comparable to those reported by Manns et al. (14.7% versus 14%, respectively). However, any comparisons in this regard should be viewed with caution as per protocol ribavirin dose reductions differed between studies. The WIN-R study allowed erythropoiesis-stimulating agents for patients who became anemic; whereas in the Manns study, anemia was treated with ribavirin dose reduction until hemoglobin of 8.5 g/dL was reached, whereupon patients were discontinued.

In regards to the secondary efficacy analysis, the WIN-R study found no advantage to extending treatment from 24 to 48 weeks in genotype 2/3 patients (SVR = 66.3% versus 57.5%, respectively). Additionally, no difference was seen between the “true” WBD and FD ribavirin groups for these easier to treat genotypes, confirming the approved combination regimen for genotypes 2 and 3.3

The WIN-R study offers a significant contribution to the literature with its findings from the subanalysis of African American patients. This subanalysis reports the results of 362 African American patients with genotype 1 HCV treated for 48 weeks with peginterferon alfa-2b 1.5 μg/kg and either a daily FD or “true” WBD of ribavirin per protocol. In this cohort of patients, “true” WBD of ribavirin in combination with peginterferon alfa-2b was significantly better than FD of ribavirin (21% versus 10%, P = 0.04). This trial enrolled the largest number of African American patients to date, and has allowed more meaningful assessments of antiviral therapy in this population. Despite the large number of African American patients enrolled, the investigators admit that their conclusions were a result of a subanalysis that was not powered for statistical significance. Still, many of their insights deserve a thoughtful discussion.

The authors of the WIN-R study compare the landmark study of Hadziyannis et al.3 to their own and note that the greater relative difference in SVR seen between the FD and “true” WBD groups in their study suggests that “true” WBD is superior to the “standard” WBD that patients received in the Hadziyannis trial. The Hadziyannis trial randomized patients to receive peginterferon alfa-2a at 180 μg/week and either a FD of ribavirin or a “standard” WBD of ribavirin. SVR was seen in 52% of genotype 1 patients receiving “standard” WBD ribavirin versus 41% of patients receiving a FD of ribavirin (odds ratio of 1.55 [1.14-2.10]; P = 0.005). In this subanalysis of WIN-R, SVR was reported in 34% of genotype 1 patients receiving “true” WBD ribavirin versus 29% receiving a FD of ribavirin (P = 0.004). The comparison of these 2 trials should be viewed with caution because the Hadziyannis trial was statistically powered to detect the difference seen, and this WIN-R subanalysis was not. Furthermore, criteria for dose reductions and growth factor use were more restrictive in the Hadziyannis trial than in the WIN-R. Several limitations of the current study are disclosed, including a high patient dropout rate and missing data (only 56.3% of African American patients had 24-week follow-up data), making a comparative analysis difficult.

One of the main concerns addressed by the investigators was the low SVRs achieved, because the SVR seen in this subanalysis is not as robust as that seen in other similar trials.4–6 The Vira-hep C trial4 similarly enrolled 196 African American patients and 202 Caucasian patients to determine if there was a difference in response to peginterferon alfa-2a and “standard” WBD of ribavirin. Accordingly, these patients never received a dosage of ribavirin greater than 1200 mg/day. Despite the lower dosage, African American patients in the Vira-hep C trial treated for 48 weeks achieved an SVR of 28% as compared to 21% in the “true” WBD arm of the WIN-R subanalysis (P < 0.0001). Also, growth factors were not permitted in the Vira-hep study. It is plausible that the high drop-out rate and missing data in WIN-R contributed to the suboptimal response rates.

The most puzzling finding in this WIN-R African American subanalysis is the independent finding that SVR increased as body weight increased in the “true” WBD arm. This is counterintuitive given that the dosing protocol was designed to insure that patients in the “true” WBD arm receive equivalent amounts of ribavirin adjusted for weight. Indeed, the authors are unable to explain this finding. Further study is required to investigate this finding in order to determine if this is indeed a clinically significant observation or a finding attributable to chance.

The WIN-R study represents the largest trial to date designed to assess virologic response to combination therapy for chronic HCV and is distinctive because it was a U.S.-based trial that sought to alleviate geography and body weight as confounders. WIN-R adds significantly to our understanding of interferon therapy in African American patients: (1) This study enrolled the largest number of African American patients into an interferon study to date, and (2) it is the first large study to investigate “true” WBD of ribavirin in African American patients with chronic HCV. Despite our lack of confidence in secondary analysis, important information can be derived from this African American substudy and future studies that are inspired by this trial. The WIN-R subanalysis of African American patients may not have the power to detect its statistical goal, but it does have the power to change the way we think about ribavirin dosing in African Americans, who will benefit from the awareness derived herein and its translation into more vigilant care of this difficult-to-treat population. Ultimately, however, the larger question still looms: Is “true” WBD of ribavirin superior to the currently approved “standard” WBD of ribavirin? Though the definitive answer still requires a prospective, randomized, controlled trial, the results of WIN-R are compelling. At least the traditional notion that ribavirin dosage should be fixed has now been sidelined by the idea that we should tailor ribavirin dosing to our patients.