A randomized, double-blind, placebo-controlled clinical trial of high-dose interferon-alpha induction treatment combined with ribavirin for chronic hepatitis C in hemophilia


Karina Meijer, University Hospital Groningen, Division of Hemostasis, Thrombosis and Rheology, Hanzeplein 1, 9713 GZ Groningen, the Netherlands.
Tel.: +31 50 361 2791; fax: +31 50 361 1790; e-mail: karinameijer@hetnet.nl

Many patients with hemophilia and other congenital coagulation disorders have been infected by hepatitis C virus (HCV), and chronic hepatitis C has become a significant cause of mortality in this patient group [1,2]. Although major improvements in therapy for chronic hepatitis C have been made in recent years, the rate of sustained response to combination therapy of interferon-alpha (IFN-α) (also in pegylated form) and ribavirin is still not optimal and different strategies for improvement are tested. Studies of HCV-RNA kinetics have suggested that the rate of response to IFN-α therapy could be increased by starting with a high induction dose, especially in genotype 1 [3,4] One clinical study confirmed an increased response rate with induction dosing in genotype 1, while another trial found no difference [5,6]. Hemophilia patients are excluded from most large trials in hepatitis C [7–9], and are not likely to enter trials that require liver biopsies. Therefore, specific trials for hemophilia patients must be designed. Such trials have the advantage over other hepatitis C trials that patients form a homogeneous group with equal mode, time and duration of infection, without referrral bias.

We initiated a double-blind placebo-controlled trial to compare induction-dosed with standard-dosed IFN-α combined with ribavirin in patients with congenital coagulation disorders (mainly hemophilia). At the time of planning, the use of ribavirin was still experimental and its optimal duration unknown. Pegylated IFN-α was not yet available.

Patients were eligible if they had a congenital coagulation defect, had chronic hepatitis C (anti-HCV antibodies for at least 6 months and HCV-RNA-positive), were previously untreated for HCV and provided written informed consent. Patients were recruited from six Dutch hemophilia centers. The study protocol was reviewed and approved by the Ethics Committees of all participating hospitals.

Patients were randomly assigned to one of two treatment groups. In the first group [induction therapy (IT)], patients received 5 MU of IFN-α2b (Schering-Plough, Kenilworth, NJ, USA) twice daily during 4 weeks, followed by IFN-α 6 MU every second day for another 48 weeks. In the second group [standard therapy (ST)], patients received 5 MU of IFN-α once every second day during 4 weeks, followed by IFN-α 6 MU every second day for another 48 weeks. All patients received ribavirin for 24 weeks. Randomization was performed in strata of participating hospitals and genotype (genotype 1 vs. other genotypes). Therapy was stopped if qualitative HCV-RNA polymerase chain reaction (PCR) remained positive at week 16. Sample size calculations were based on end-of-treatment response. We assumed a 40% response in the standard group, vs. 70% in the induction group. To achieve a 80% power to detect a difference with an α error of 0.05, 42 patients were needed in each of the two groups. According to prior inventory of the Dutch hemophilia centers, this seemed realistic. However, recruitment was slower than anticipated and the trial was closed after the inclusion of 66 patients. Primary endpoints were end-of-treatment and sustained response. They were defined as negative qualitative HCV-RNA PCR at weeks 52 and 78, respectively. Secondary endpoint was the occurrence of side-effects, evaluated as reported adverse events and the need for dose adjustments or cessation of therapy.

Qualitative HCV-RNA PCR was performed locally at weeks 16, 52 and 78. Blood for quantitive HCV-RNA was collected for central testing (Cobas Amplicor HCV Monitor™ Test, version 2.0; Roche Diagnostics, Branchburg, NJ, USA).

Sixty-six patients were included between June 1998 and December 1999. The randomized treatment assignment in strata of participating hospitals led to an unequal number of patients in the two treatment groups (37 in IT vs. 29 in ST). Genotype was 1 in 51, 2 in 13 and 3 in two. No differences in baseline characteristics were seen between the two groups (data not shown).

Dose adjustments were made in 35/66 patients, 23/37 (62%) in IT and 12/29 (41%; P = 0.09) in ST. Grade 4 (life-threatening) adverse events were not seen. One patient with a long-standing inhibitor to factor VIII developed a large hematoma in the abdominal wall after subcutaneous injection; no other coagulation disorder-related complications were seen.

Response rates are shown in Table 1. The initial response was similar in IT and ST. At weeks 52 and 78, more patients in ST than in IT had relapsed, but the difference was not significant. Sustained response was achieved in 50% of patients, 57% in IT vs. 41% in ST (P = 0.22). In genotype 1, sustained response was seen in 43% in IT vs. 35% in ST (P = 0.56). In genotypes non-1, sustained response was achieved in 100% in IT vs. 67% in ST (P = 0.06). In univariate analysis, genotype non-1 and low baseline levels of γ-glutamyl transferase and α-fetoprotein were associated with sustained response (Table 2). In multivariate analysis, only genotype non-1 and low baseline α-fetoprotein were independently associated with sustained response (P < 0.01 and P = 0.01, respectively).

Table 1.  Virological response, defined as negative hepatitis C virus-RNA polymerase chain reaction, in all patients and compared by treatment allocation
 All patientsITSTP*
n = 66 n = 37 n = 29 
  1. IT, Induction therapy; ST, standard therapy. *IT vs. ST, compared by χ2.

Week 1649/6674%28/3776%21/2972%0.76
Week 5240/6661%24/3767%16/2955%0.42
Week 7833/6050%21/3757%12/2941%0.22
Table 2.  Baseline characteristics of sustained responders to treatment with interferon-alpha plus ribavirin compared with patients who did not respond or who relapsed
 Sustained responseNon-response/relapseP
n = 33 n = 33 
  1. All values in medians with range or in proportions; medians compared by Mann–Whitney U-test, proportions by χ2. ALT, alanine aminotransferase; GGT, gamma-glutamyl transferase; AFP, α-fetoprotein.

Age (years)4318–583920–680.26
HCV genotype 161% 94% < 0.01
HCV RNA <2.106 copies mL−159% 47% 0.36
Hemoglobin (mmol L−1)9.36.8––11.00.15
Platelets (109 L−1)21270–34520671–3780.66
ALT (U L−1)6122–3436322–3130.75
ALT ≤ 40 U L−118% 31% 0.25
GGT (U L−1)3812–1097717–280<0.01
Bilirubin (µmol L−1)114–33135–550.27
Albumin (g L−1)4640–524639–490.37
Antithrombin (%)9666–1269647–1170.84
AFP (µg L−1)31–1142–440.02

No double-blind comparisons of different doses of IFN-α in chronic hepatitis C have been published before. In this study, we showed that such a comparison can be made, and that patients—after explanation of the purpose—accepted placebo injections. Frequent subcutaneous injections were feasible in these patients with coagulation disorders. Although there was a trend towards a higher rate of dose reductions in the induction group, patients in the induction group were not more likely to discontinue therapy prematurely.

This study turned out to have insufficient power to demonstrate a difference in the primary endpoints, end-of-treatment and sustained response. This would have remained so if we could have included the planned number of patients, because the pattern of response we observed was different from the one expected. Unexpectedly, induction treatment did not increase initial response but it did seem to result in a lower rate of relapse (Table 1). In addition, patients in ST had a higher than expected rate of end-of-treatment response.

We conclude that induction therapy is safe and feasible, even in hemophilia patients, although intensive monitoring and frequent dose adjustments are necessary. Double-blind comparisons of IFN-α doses and the use of placebo injections are realistic.

The overall rate of sustained response in this trial was high and, although no statistically significant difference between induction and standard therapy could be demonstrated, all comparisons favored induction therapy. We would recommend conducting a larger double-blind induction therapy trial, in combination with pegylated IFN-α.


The CHARIBDIS Study Group consists of K. Meijer,* M. C. Kappers-Klunne,† I. R. O. Nováková,‡ E. P. Mauser-Bunschoten,§ M. M. W. Koopman,¶ P. J. Voogt,** F. W. G. Leebeek,† E. B. Haagsma,* W. M. Smid,* J. Schirm,†† N. J. G. M. Veeger* and J. van der Meer.* (*University Hospital Groningen, †Erasmus University Medical Center Rotterdam, ‡University Hospital Nijmegen, §Van Creveldkliniek, University Medical Center Utrecht, ¶Academic Medical Center Amsterdam, **Atrium Hospital Heerlen, ††Regional Public Health Laboratory Groningen; all in the Netherlands).


The authors thank Maria A. J. Voskuilen for assisting in study co-ordination. This study was funded in part by a research grant from Schering-Plough.