Prof. M. S. Khuroo, Post Box 3354, MBC 46, Riyadh 11211, Saudi Arabia. E-mail: email@example.com
Background : Guidelines for treatment of patients infected with hepatitis C virus genotype 4 are not available.
Aim : To perform a meta-analysis of randomized controlled trials comparing peginterferon plus ribavirin with interferon plus ribavirin treatment in treatment-naïve patients infected with chronic hepatitis C virus genotype 4.
Methods : The outcome measure was sustained virologic response. The measure of association employed was relative risk calculated by the random-effect model, with heterogeneity, sensitivity and subgroup analyses.
Results : Of the 565 studies screened, six randomized controlled trials including 424 patients (peginterferon plus ribavirin 219, interferon plus ribavirin 205) were analysed. Duration of therapy was 1 year in all trials. Sustained virological response obtained with peginterferon plus ribavirin (55%) was significantly higher than with interferon plus ribavirin (30%) [relative risk, 1.71 (95% confidence interval, 1.15–2.56); P = 0.0088]. In the subgroup analyses, sustained virological response in trials using standard-dose ribavirin (1000 or 1200 mg/day) was 72% as against 45.8% in trials using low-dose ribavirin (800 mg/day) (P = 0.01). Further sub-group analyses for treatment duration, body weight, viral load and cirrhosis could not be performed because of lack of relevant data.
Conclusion : Treatment-naïve patients infected with hepatitis C virus genotype 4 should be treated with peginterferon plus standard-dose ribavirin for 1 year, with an expected sustained virological response rate of 72%.
Chronic infection with hepatitis C virus (HCV) affects approximately 300 million people worldwide and is the most frequent cause for liver transplantation in the US and Europe.1 Natural history studies suggest that 5% and perhaps as many as 20% of patients develop cirrhosis after 20 years of HCV infection. Moreover, the incidence of chronic liver failure is expected to increase over the next 10 years as a result of the ‘silent epidemic’ of HCV infection.2
Recently, major advances have occurred in the treatment of chronic hepatitis C (CHC) and so sustained virologic response (SVR) rates of 42–82% can be achieved.3 A number of viral and host factors, which influence the SVR, have been identified.4 Of the viral factors, HCV genotype is the most important factor influencing the SVR. HCV genotypes and subtypes are intrinsic characteristics of the transmitted viral strain and do not change during the course of the infection.5 On the basis of variations in the nucleotide sequence of HCV, six genotypes (numbered 1 to 6) and more than 50 subtypes (indicated by lowercase letters, for example, 1a and 1b) have been identified. HCV genotypes show a geographically defined distribution. In North America and Europe, 70% of patients are infected with HCV genotype 1, while 25% of patients have genotype 2 or 3 and <5% of patients are infected with genotype 4, 5, 6 or mixed genotypes.6 Genotype 4 is the most prevalent in Egypt, Saudi Arabia, and other North African countries, constituting up to 64% of HCV infections. Other genotypes prevalent in these countries include genotype 1 (25%), 3 (5%) and 6 (0.3%).7
The most noteworthy therapeutic advance over the past few years has been the introduction of peginterferons.8 Peginterferon alpha-2a (40 kDa) and alpha-2b (12 kDa) have been approved for treatment of CHC. Based on studies performed in Europe and North America, peginterferon (alpha-2a or alpha-2b) plus ribavirin therapy is established as the standard therapy for CHC.9, 10 However, these studies predominantly included patients with genotype 1, 2 and 3. Treatment guidelines for antiviral therapy for these genotypes have been adopted.11 Only a small percentage of patients were infected with genotype 4 in these studies. Thus, treatment guidelines for patients infected with HCV genotype 4 are not known.
To address this issue, we performed a meta-analysis of randomized controlled trials comparing peginterferon plus ribavirin with interferon plus ribavirin treatment in naïve patients infected with chronic HCV genotype 4.
Materials and methods
We conducted a meta-analysis that identified the treatment effects of peginterferon plus ribavirin therapy on specific end points for treatment-naïve patients with HCV genotype 4. A written protocol specified several aspects of the meta-analysis as per the defined guidelines.12 A checklist was developed for the purpose of data entry.
Acquisition of data
Medline (National Library of Medicine, Bethesda, MD, USA) and EMBASE (Elsevier, New York, NY, USA) were searched from 1966 onwards to locate published research in the area of CHC and antiviral therapy. MeSH terms used for key and text word searching included ‘chronic hepatitis C’, ‘genotype’, ‘interferon-α’, ‘peginterferon’ and ‘ribavirin’. We also searched the Cochrane Registry of controlled trials database. Published abstracts from international gastroenterology and hepatology conferences (previous 5 years) were reviewed. Manufacturers of interferons and known contacts in the area of this research were approached. Finally, a manual search was performed using bibliographies from each full-published report.
Criteria for selection. The following selection criteria were applied: (i) study design – prospective randomized controlled trial; (ii) language of publication – both English and other languages; (iii) study population – treatment-naïve patients with chronic HCV genotype 4; in trials including patients with HCV with any genotype, data on HCV genotype 4 were extracted for analysis; (iv) intervention (treatment group) – peginterferon plus ribavirin; (v) comparison intervention (control group) – interferon alpha plus ribavirin; and (vi) outcome measure – SVR. Reporting of duplicated studies was excluded by examining the author list, parent institution, sample size and results.
Outcome measure. We employed the SVR as the outcome measure to assess the effects of peginterferon plus ribavirin treatment in the selected trials. SVR was defined as absence of detectable HCV RNA performed with a sensitive technique (lower detection limit of ≤50 IU/mL) 24 weeks after end-of-treatment. All analyses were performed according to the intention-to-treat method. Two investigators independently evaluated trials for inclusion and for outcome measure and subsequently resolved any disagreement by discussion.
Assessment of study quality
Two reviewers (M.S.K., Saudi Arabia and S.T.D.) independently assessed trial quality by examining the allocation sequence generation and allocation concealment.13, 14 The allocation sequence was classified as adequate if based on computer-generated random numbers, table of random numbers or similar. The allocation concealment was classified as adequate if the allocation sequence was concealed until the moment of randomization by a central independent unit, sealed envelopes or similar devices. Any differences were resolved by consensus.
Each reference was read independently by two authors (M.S.K., Saudi Arabia and M.S.K., India) for quantitative analysis and a checklist of questions was completed. The quantitative data abstracted included: sample size; pre-treatment patient characteristics including mean age, sex, mean body weight, mean body mass index (BMI), mean HCV RNA load, percentage of patients with high viral load (>800 000 IU/mL), percentage of patients with cirrhosis; type of peginterferon (alpha-2a or alpha-2b) and interferon (alpha-2a or alpha-2b); dose of peginterferons and interferons; dose of ribavirin (body weight-based high dose vs. fixed low dose); SVR rate; per cent dropouts; per cent with adverse reactions; and per cent needing drug dose reduction or withdrawal. Attempts were made to get additional information on missing or ambiguous data from the investigators. Discrepancies in analysis were resolved by consensus.
The measure of association used in this meta-analysis was relative risk (RR) with 95% confidence interval (95% CI). Summary RR with 95% CI was calculated using random effect model.15 A statistically significant result was assumed when the 95% CI did not include one. Statistical heterogeneity of the trials was evaluated; however, random effect model incorporated the heterogeneity of the studies. When the RR between the treatment and the control group differed significantly in the outcome measure, the number needed to treat to either benefit (NNTB) or harm (NNTH) one additional patient with 95% CI was calculated for that measure. Publication bias was investigated through visual inspection of funnel plots whereby odds ratios were plotted against sample size. Because graphic evaluation can be subjective, a rank correlation test for publication bias was employed.16
We performed sensitivity analysis to assess the stability of conclusions to assumptions about the probabilities used in the analysis and were done as an assessment of the methodology. For this purpose, we assessed whether trial quality could influence the results of our meta-analysis. Next, we performed subgroup analysis with the purpose of delineating differences in the effects of intervention that were biologically based. These included the influence of the following factors on the outcome measure: type of peginterferon (alpha-2a vs. alpha-2b) and dose of ribavirin (body weight-based high dose vs. fixed low dose). Heterogeneity between corresponding subgroups was calculated and when this was significant, it was concluded that the effect developed in one subgroup differed significantly from that in the other subgroup.
All analyses and calculations were performed using a program (EasyMA) developed by the Department of Clinical Pharmacology of the University Hospital in Lyon, France.17
We identified 565 articles for review. Six trials were selected for analysis.9, 10, 18–21 There was unanimity between the two authors MSK (Saudi Arabia) and STD about selection of relevant articles.
Study design and patient characteristics
Of the six trials included, two are published as peer-reviewed articles,9, 10 while four have been published as conference abstracts.18–21 Two trials were international multicentre studies9, 10 and another two involved multiple centres at the national level in Saudi Arabia (Table 1).19, 21 All trials were open-labelled randomized trials. The allocation sequence and concealment was assessed to be adequate in four trials9, 10, 19, 21 and not known in two trials.18, 20 The number of patients included in six trials was 424 (treatment group 219, control group 205). The mean number of patients per trial was 71, with 32, the size of the smallest trial and 120 that of the biggest trial. Two trials were performed in a western population9, 10 and had included few number of patients with HCV genotype 4 and the remaining four trials were performed in patients from the Middle East.18–21 Duration of therapy was 1 year in all trials. Three trials used peginterferon alpha-2a 180 μg/week,10, 19, 20 while three trials used peginterferon alpha-2b.9, 18, 21 In two of the latter three trials, a fixed dose of peginterferon alpha-2b of 100 μg/week was used18, 21 and in one trial, a body weight-based dose of alpha-2b of 1.5 μg/kg body weight/week was used.9 Interferon alpha-2a was used in the control group in two trials19, 20 and interferon alpha-2b in four trials.9, 10, 18, 21 The dose of interferon alpha in all trials was 3 million units subcutaneously three times per week. Body weight-based high-dose ribavirin (1000 mg/day for patients with body weight <75 kg or 1200 mg/day for patients with body weight ≥75 kg) was used in two trials10, 20 and fixed low dose (800 mg/day) in another four trials.9, 18, 19, 21 SVR was reported in all the six trials. Dropouts were reported in three trials.18, 19, 21
Table 1. Characteristics of the studies included for meta-analysis
–, Results not reported or available for genotype 4 patients.
* SVR, sustained virologic response in treatment vs. control group.
† Included some patients with genotype 5 and 6.
‡ 1000 mg/day for <75 kg and 1200 mg/day for ≥75 kg.
Total trial size*
Study size – (genotype 4)*
alpha-2b (1.5 μg/ kg body weight/week)
alpha-2a (180 μg/week)
alpha-2b (100 μg/week)
alpha-2a (180 μg/week)
alpha-2a (180 μg/week)
alpha-2b (100 μg/week)
1000 or 1200 mg‡
1000 or 1200 mg‡
Mean age (years)
% of males
% of cirrhosis
% of high viral load
Table 2 gives the data of primary, sensitivity and subgroup analyses with number of trials included, sample size in the treatment and control group, relative risk with 95% CI, test of association with the P-value of chi-square statistic, number needed to treat to benefit one additional patient with 95% CI as calculated for that measure, statistical heterogeneity between trials as evaluated by the Cochran Q-test and rank correlation test for publication bias. The trials in all the analyses were homogeneous (P-value for heterogeneity ≥0.1) and none of the analyses showed publication bias as assessed by visual inspection of funnel plots and rank correlation test for publication bias.
Table 2. Results of primary, sensitivity and subgroup analyses
Type of analysis
Included trials by reference numbers
Sample size (treatment/ control)
SVR % treatment/ control
Relative risk (95% CI)
P for association*
NNTB (95% CI)†
P for heterogeneity‡/ P for publication bias§
*Test of association with the value of chi-square statistic, when P is <0.05 virological responses between treatment vs. control groups were significant.
† Number needed to treat to benefit one additional patient with 95% CI as calculated for that measure.
‡ Statistical heterogeneity between trials was evaluated by the Cochran Q-test and was considered to exist when P < 0.1.
§ Rank correlation test for publication bias, publication bias was presumed to exist when P < 0.05.
Primary meta-analysis. In the treatment group, 120 (54.8%) of the 219 patients had attained SVR (Figure 1). The SVR in individual trials varied from 40% (95% CI, 24.8–55.1) to 79.1% (95% CI, 62.9–95.4). In the control group, 61 (29.8%) of the 205 patients had attained SVR. The SVR in individual trials varied from 16.3% (95% CI, 5.9–26.6) to 39.4% (95% CI, 23.9–55.0). Of the six trials, only two had shown a significantly higher SVR in the treatment group. Summary RR for all the six trials was 1.71 (95% CI, 1.15–2.56; P = 0.0088); a significant increase of 25% (95% CI, 11.6–36.1) in SVR in the treatment over control group.
Sensitivity analysis. The study of the sensitivity analysis revealed that the summary RR for the outcome measure remained statistically stable (similar to summary RR obtained from six trials) when meta-analysis was limited to four trials with adequate allocation sequence and concealment. In these four trials, the SVR in the treatment group (53.9%; 69/128) was higher than in the control group (32.2%; 38/118) (summary RR, 1.6; 95% CI, 1.16–2.19; P = 0.003). These data vindicated the stability of the results of the meta-analysis.
Type of peginterferon. In the three trials using peginterferon alpha-2a, SVR in the treatment group (62.2%; 84/135) was higher than that in the control group (25%; 30/120) (summary RR, 2.41; 95% CI, 1.35–4.13; P = 0.003) (Figure 2). In the three trials using peginterferon alpha-2b, SVR in the treatment group (42.9%; 36/84) was not significantly higher than that in the control group (36.5%; 31/85) (summary RR, 1.17; 95% CI, 0.80–1.70; P = 41). The heterogeneity between treatment effects obtained with peginterferon alpha-2a vs. alpha-2b was significant (P = 0.04). On this basis, it was concluded that the SVR attained with alpha-2a (62.2%) was significantly higher than that attained with alpha-2b (42.9%).
Dose of ribavirin. In the two trials using body weight-based high-dose ribavirin, SVR in the treatment group (72%; 54/75) was higher than that in the control group (20%; 12/60) (summary RR, 3.16; 95% CI, 1.67–5.98; P ≤ 0.001) (Figure 3). In the four trials using fixed low-dose ribavirin, SVR in the treatment group (45.8%; 66/144) was higher than that in the control group (33.8%; 49/145) (summary RR, 1.35; 95% CI, 1.0–1.80; P = 0.04). The heterogeneity between treatment effects attained in trials with body weight-based high-dose ribavirin vs. fixed low-dose ribavirin was significant (P = 0.01). On this basis, it was concluded that the SVR attained with peginterferon plus body weight-based high-dose ribavirin (SVR 72.0%) was significantly higher than that attained with peginterferon plus fixed low-dose ribavirin (45.8%).
We could not perform a subgroup analysis comparing 24 with 48-week treatment as the duration of treatment in all the six trials included in the meta-analysis for 1 year. Moreover, factors like mean age, per cent males, mean body weight, mean BMI, per cent with high or low viral load and per cent with cirrhosis were not available in treatment and control group separately for us to perform subgroup meta-analyses. We believe that further trials involving adequate sample size need to be carried out in order to find out the influence of these factors on the SVR in patients with HCV genotype 4.
Patients with HCV genotype 4 have traditionally been described as ‘difficult-to-treat’, with SVR of 0–8%, with interferon alone and 14–42% with interferon plus ribavirin therapy.4, 22 The SVR in the control group of the trials included in the meta-analysis was 30.1% (range 16.3–39.4%). Based on trials from North America and Europe, the SVR obtained in patients infected with CHC genotype 1 is 33–36% and in those infected with genotype 2 or 3 is 61–79%.3 Furthermore, viral replication kinetics between various genotypes of HCV on interferon therapy have a distinct pattern.23, 24 During peginterferon plus ribavirin treatment, the early HCV genotype 4 kinetics are similar to genotype 1 kinetics, and significantly different from genotype 3 kinetics. These observations suggest that the response rate of chronic HCV genotype 4 should resemble that of genotype 1 and not of those with chronic HCV genotype 2 or 3. Consequent to these observations, it has been recommended that patients infected with chronic HCV genotype 4 should be treated similar to patients infected with genotype 1.
Four of the six randomized trials included in this study failed to show significantly higher SVR with peginterferon plus ribavirin on patients infected with HCV genotype 4 when compared with interferon plus ribavirin therapy.8, 9, 18, 21 This is possibly related to a small number of patients included in these studies. In the meta-analysis, the summary SVR obtained with peginterferon plus ribavirin treatment (54.8%) showed a significant increase (25.0%; 95% CI, 11.6–36.1) when compared with interferon plus ribavirin treatment (39.8%). This SVR rate (54.8%) is higher than that reported in patients infected with genotype 1 (42–46%) and much lower than that in genotype 2 or 3 (76–82%).3 However, the SVR rate calculated from this meta-analysis is based on trials which were performed with fixed low-dose ribavirin (four trials)8, 18, 19, 21 and fixed doses of peginterferon alpha-2b 100 μg/week (two trials).18, 21 Subgroup analysis identified these issues and further defined the SVR based on different dose schedules of ribavirin and peginterferon.
Trials using peginterferon alpha-2a had higher response rates than those using peginterferon alpha-2b (SVR 62.2% and 42.9%, respectively; P = 0.04). We believe that this may not be because of difference in efficacies of the two types of peginterferon for two reasons. First, two of the three trials with peginterferon alpha-2a were performed with body weight-based high-dose ribavirin;10, 20 while all the three trials with peginterferon alpha-2b used fixed low-dose ribavirin.8, 18, 21 Moreover, in two of the three trials, peginterferon alpha-2b dose was 100 μg/week18, 21 and only one trial used the body weight-based dose (1.5 μg/kg body weight/week) of the drug.8 Peginterferon alpha-2a has a restricted volume of distribution and so weight-based dosing is not needed. In contrast, weight-based dosing (1.5 μg/kg/week) of peginterferon alpha-2b is recommended, as this drug has a wider volume of distribution and faster clearance.25 Our argument was supported by the study of Hassan et al.26 from Kuwait. In this study, 66 patients with HCV genotype 4 were treated with peginterferon alpha-2b 1.5 μg/kg body weight/week plus ribavirin (1000 or 1200 mg/day) for 48 weeks and an SVR rate of 68% was obtained.
The impact of ribavirin dose in achieving SVR is related to the HCV genotype.27 In patients infected with genotype 2 or 3, the response rate is similar with low- or high-dose ribavirin (79% vs. 80%). However, in patients infected with genotype 1, the SVR with peginterferon plus body weight-based high-dose ribavirin is significantly higher than that with fixed low-dose ribavirin (52% vs. 41%; P = 0.005). Subgroup analysis from the meta-analysis revealed that SVR attained with peginterferon plus body weight-based high-dose ribavirin (SVR 72%) was significantly higher than that with peginterferon plus fixed low-dose ribavirin (45.8%) (P = 0.01). This observation is important and on the basis of these data, we recommend that body weight-based high-dose ribavirin must be used in patients with HCV genotype 4.
The duration of therapy for all trials included in the meta-analysis was 1 year. Thus, we could not assess the efficacy of antiviral therapy given for 24 or 48 weeks. Hadziyannis et al.27 included 36 patients infected with genotype 4 and obtained SVR in 67% (eight of 12 patients) with peginterferon plus high-dose ribavirin given for 24 weeks and in 82% (nine of 11 patients) with peginterferon plus high-dose ribavirin given for 48 weeks. Analysing these 36 patients plus 13 other patients included in another international trial,10 Diago et al.28 reported on virologic response in 49 patients with chronic HCV genotype 4. The SVR in patients treated for 24 weeks was 67%, while in those treated for 48 weeks was 79%. The authors concluded that the optimal treatment regimen for patients with HCV genotype 4 appears to be peginterferon plus body weight-based high-dose ribavirin for 48 weeks. These data, although important, are not sufficient to recommend the duration of therapy in patients with HCV genotype 4. We recommend randomized controlled trials with peginterferon plus ribavirin therapy for 24 or 48 weeks in patients with HCV genotype 4 to define the optimal duration of therapy in such patients.
On the basis of the data available, treatment-naïve patients infected with chronic HCV genotype 4 should be treated with peginterferon plus body weight-based high-dose ribavirin for 1 year. With this regimen, around two-thirds of patients are expected to show SVR.
We identified two problems during acquisition of data for this meta-analysis. First, the two studies published as full papers included patients with all genotypes and the study groups were not matched according to the genotype. Genotype 4 patients accounted for only 3.1% and 3.9% of the included patients, respectively. However, genotype 4 patients from these two studies have been analysed separately and thus, inclusion of these patients in this meta-analysis was valid.28 The remaining four studies were available as abstracts only. However, we were able to assess the study quality in all these trials. This made the inclusion of genotype 4 patients from these four trials appropriate for the meta-analysis.
Concept and Protocol: M.S.K., Saudi Arabia; Acquisition of data: M.S.K., Saudi Arabia, S.T.D.; Qualitative analysis of trials: M.S.K., Saudi Arabia, S.T.D.; Quantitative analysis of trials: M.S.K., Saudi Arabia, M.S.K., India; Data analysis: M.S.K.,Saudi Arabia, M.S.K., India; Statistical analysis: M.S.K., Saudi Arabia, M.S.K., India; and Manuscript preparation: M.S.K., Saudi Arabia, M.S.K., India.