By continuing to browse this site you agree to us using cookies as described in About Cookies
Notice: Wiley Online Library will be unavailable on Saturday 7th Oct from 03.00 EDT / 08:00 BST / 12:30 IST / 15.00 SGT to 08.00 EDT / 13.00 BST / 17:30 IST / 20.00 SGT and Sunday 8th Oct from 03.00 EDT / 08:00 BST / 12:30 IST / 15.00 SGT to 06.00 EDT / 11.00 BST / 15:30 IST / 18.00 SGT for essential maintenance. Apologies for the inconvenience.
Dr M. Simin, The Cochrane Hepato-Biliary Group, Copenhagen Trial Unit, Centre for Clinical Intervention Research, Rigshospitalet, Copenhagen University Hospital, Department 33.44, Blegdamesvej 9, 2100 Copenhagen, Denmark. E-mail: email@example.com
Background About 170 million patients worldwide have chronic hepatitis C. Pegylated interferon plus ribavirin is currently the recommended therapy.
Aim To evaluate the beneficial and harmful effects of pegylated interferon plus ribavirin vs. interferon plus ribavirin for chronic hepatitis C infection.
Methods We searched The Cochrane Library, MEDLINE, EMBASE, LILACS, Science Citation Index Expanded and contacted pharmaceutical companies and authors of trials (to March 2005).
Results We included 18 randomized clinical trials with 4811 patients. Eleven trials (61%) had allocation bias risks and all had assessment bias risk because of lack of blinding. Compared with interferon plus ribavirin, pegylated interferon plus ribavirin had significant beneficial effects on sustained virological response [risk ratio (RR): 0.80; 95% CI: 0.74–0.88]. Data were insufficient to determine impact on long-term outcomes. Pegylated interferon plus ribavirin significantly increased dose reductions (RR: 1.44; 95% CI: 1.14–1.82) and adverse events including neutropenia (RR: 2.25; 95% CI: 1.58–3.21), thrombocytopenia (RR: 2.28; 95% CI: 1.14–4.54), arthralgia (RR: 1.19; 95% CI: 1.05–1.35), and injection-site reaction (RR: 2.56; 95% CI: 1.06–6.22).
Conclusions Pegylated interferon plus ribavirin compared with interferon plus ribavirin increased the proportion of patients with sustained virological response, but at the cost of more adverse events.
Hepatitis C virus is a blood-born RNA virus with six genotypes and belonging to the Flaviviridae family.1–3 Up to 85% of acutely infected patients develop chronic hepatitis C.4 Globally, an estimated 170 million people are chronically infected with hepatitis C.5 The large majority of them will never develop end-stage liver disease.6 Among the small subgroup that does develop this complication, it has been estimated that, on average, it takes about 30 years for the cirrhosis to occur.7, 8 Annually, 2–4% of patients with cirrhosis develop hepatocellular carcinoma.9, 10 Chronic hepatitis C is the most common indication for liver transplantation.11
Sustained virological response, i.e. viral clearance six months after treatment, is the standard measure of treatment response.12 However, it is an intermediate outcome measure and it is unclear if its response to treatment actually predicts the treatments effect on liver-related morbidity and mortality. Interferon was the main therapy for chronic hepatitis C infection for years, but only 20% of patients showed sustained virological response.13–16 The efficacy was doubled by adding ribavirin to interferon.17, 18 Pegylated interferon has recently been developed to further improve response rates.19 Its advantage is a prolonged half-life by covalent bonding of polyethylene glycol to interferon. Randomized trials have found that pegylated interferon increased the number of patients with sustained virological response, without increasing the risk of adverse events.20–22 Pegylated interferon plus ribavirin was superior to pegylated interferon monotherapy23 and is considered the standard treatment for chronic hepatitis C.24 We conducted a systematic Cochrane review and meta-analysis of randomized trials on pegylated interferon plus ribavirin vs. interferon plus ribavirin for chronic hepatitis C infection.
We conducted this review according to our predefined, peer-reviewed Cochrane Hepato-Biliary Group protocol.25 Two of us (MS and JB) selected eligible trials using electronic searches of The Cochrane Hepato-Biliary Group Controlled Trials Register, The Cochrane Central Register of Controlled Trials in The Cochrane Library (Issue 1, 2005), MEDLINE (1950–March 2005), EMBASE (1980–March 2005), LILACS (1982–March 2005), and Science Citation Index Expanded (1945–March 2005).25 We scanned reference lists, contacted pharmaceutical companies that produce pegylated interferon, interferon, or ribavirin, and wrote to authors for additional trials or information.
We included randomized trials on pegylated interferon plus ribavirin vs. interferon plus ribavirin for chronic hepatitis C infection regardless of regimens.25 Co-interventions were permitted if received equally by both arms.25 We excluded trials on patients with hepatitis B virus, human immunodeficiency virus, cancer, or liver-transplant.25
Four authors (MS, JB, DS and LLG) independently gathered data and assessed the methodological quality of the included trials. CG arbitrated any disagreements. We assessed the methodological quality of trials according to the Cochrane Handbook for Systematic Reviews of Interventions 4.2.4 26 by evaluating reported data on generation of the allocation sequence, allocation concealment, and blinding.27 The extracted patient characteristics included the mean age, proportion of men, proportion with cirrhosis, proportion with genotype 1, proportion with pre-treatment high viral load, and status regarding previous anti-viral treatment. The intervention regimen and duration of follow-up were registered. We contacted primary investigators if data were not provided in the published trial reports.
The two primary outcome measures were failure of sustained virological response (number of patients with detectable hepatitis C virus RNA at least 6 months after treatment) and the composite outcome of liver-related morbidity plus all-cause mortality (clinical manifestations of cirrhosis, i.e. ascites, variceal bleeding, hepatic encephalopathy; or hepatocellular carcinoma; or death). The secondary outcomes included failure of histological response, impaired quality of life, and adverse events.25
Analyses were performed with the statistical software review manager (version 4.2.7; The Nordic Cochrane Centre, Copenhagen, Denmark) and stata (version 8.0; Stata Corp, College Station, TX, USA). Outcomes were summarized as risk ratios (RR) and weighted mean differences (WMD) with 95% CI. Rare events (morbidity plus mortality) were estimated by Peto odds ratio (OR).28
For all analyses, we used random-effects29 and fixed-effect models.30 In case of concurrence between models, we only reported the results from the random-effects model which tends to give the most conservative estimate. Regression analysis was performed to estimate funnel plot asymmetry and examine for potential bias.31 Heterogeneity was explored by χ2 test and the quantity of heterogeneity was measured by I2 statistic.32 The following sources of heterogeneity in assessment of sustained virological response were explored in meta-regression analyses: sex, age, cirrhosis, viral genotype, viral load, dose, type, and duration of intervention. We performed the following subgroup analyses regarding sustained virological response: methodological quality (trials with or without adequate generation of the allocation sequence and allocation concealment), patients (treatment-naive patients, relapsers, or non-responders), genotypes (different hepatitis C virus genotypes), pre-treatment viral load (more than or less than 800 000 IU/mL or 2 000 000 copies/mL33, 34), co-intervention (trials with or without co-intervention), and type of pegylated interferon (trials with pegylated interferon alpha-2a or -2b). We used test of interaction to compare the treatment effect in subgroups.35
All outcomes, except histological response, were analysed according to the intention-to-treat principle. For histological response, we used available case analysis.
We identified 1605 references by our electronic searches using pegylated interferon, ribavirin, and hepatitis C as the keywords. After we read titles and abstracts, we excluded 1541 duplicates and clearly irrelevant references (Figure 1). After reading full publications, we further excluded 14 references because they did not meet our inclusion criteria. Accordingly, we included 50 references on 18 randomized trials, one of which was still on-going. The trials were published from 2002 to 2005; eight were published in abstract form and 10 as full articles.
Characteristics of included trials and quality assessment
Eleven trials (61%) were considered to have allocation bias risks based on the unclear or inadequate reporting of generation of the allocation sequence and allocation concealment (Table 1). None of the trials was double blind. However, two trials used blinded outcome assessment for histological response.
Table 1. Methodological quality and bias risk of randomized trials comparing pegylated interferon plus ribavirin vs. interferon plus ribavirin
Generation of the allocation sequence
* Bias risk as assessed by the adequacy of reporting on generation of the allocation sequence and allocation concealment.
All trials had high bias risk because of the lack of blinded outcome assessment.
A total of 4811 patients were randomized to pegylated interferon plus ribavirin vs. interferon plus ribavirin in 18 trials (Table 2). The number of patients in each trial ranged from 48 to 1530. The median age was 46 years (range: 20–68). The proportion of men was reported in 17 trials (median 66%; range 46–79%). The proportion of patients with cirrhosis was reported in 13 trials (median 10%; range: 0–38%).
Table 2. Characteristics of randomized trials comparing pegylated interferon plus ribavirin vs. interferon plus ribavirin
* The dose of pegylated interferon alpha-2a was 180 μg weekly in all trials. The dose of pegylated interferon alpha-2b was 100 μg,36, 38, 41 50 μg,37, 49 or 1.5 μg/kg weekly. The dose of interferon alpha-2a, alpha-2b, or leucocyte interferon alpha was 3 MU, 4.5 MU,51 and 6 MU37, 42, 45, 47, 50, 53 t.d.s. The dose of consensus interferon was 15 μg t.d.s.52 Four trials used high-dose interferon induction periods.39, 42, 48, 53 The dose of ribavirin varied from 600 mg to 1200 mg daily based on the body weight and was identical in both arms for all but two trials.41, 46 In these trials, all patients41 or half of the patients46 in the pegylated interferon plus ribavirin arm received 800 mg of ribavirin vs. 1000 to 1200 mg in the interferon plus ribavirin arm. Three trials administered amantadine hydrochloride 200 mg daily in both arms.39, 44, 45 Duration of treatment was 48 weeks, 24 weeks,41, 43 and 52 weeks.50 Duration of follow-up was 24 weeks in all trials except one where it was 48 weeks.49
Pegylated interferon plus ribavirin vs. interferon plus ribavirin significantly increased the proportion of patients with sustained virological response (50% vs. 38%; RR: 0.80; 95% CI: 0.74–0.88; n = 16 trials) (Figure 2). Heterogeneity was considerable (I2 = 56%). The available-case sensitivity analysis on sustained virological response (RR: 0.79; 95% CI: 0.70–0.90; n = 16) supported our intention-to-treat result. Among all the variables analysed by univariate meta-regression (sex, age, cirrhosis, viral genotype, viral load, dose, type and duration of intervention), only low viral load was significantly associated with sustained virological response (P = 0.006). In multivariate analysis, none of the variables seemed to be significantly associated with sustained virological response. Funnel plot visual inspection and statistical tests (Begg’s test z = 0.32, P = 0.75 and Egger’s test coefficient = 0.79, P = 0.29) did not indicate significant evidence of bias.
There was no significant difference regarding sustained virological response in subgroup analyses of methodological quality, viral genotype, viral load, co-interventions, or type of pegylated interferon (tests for interaction, P > 0.05) (Table 3). In subgroups of treatment-naïve patients, pegylated interferon plus ribavirin significantly increased sustained virological response (RR: 0.77; 95% CI: 0.69–0.87; n = 12), but not in relapsers (RR: 0.77; 95% CI: 0.57–1.04; n = 1) or non-responders (RR: 0.93; 95% CI: 0.84–1.03; n = 2). Tests of interaction demonstrated no significant difference between treatment-naïve and relapsers (z = 0.00, P = 1.00), but a significant difference between treatment-naïve and non-responding patients (z = 2.39, P = 0.02).
Table 3. Subgroup analyses regarding intervention effect of pegylated interferon plus ribavirin vs. interferon plus ribavirin on sustained virological response. Figures are relative risks with 95% CI (random-effects models)
Type of pegylated interferon
* Not significantly different from intervention effect in comparison group in a test of interaction.; # Significantly different from intervention effect in treatment-naïve patients (z = 2.39; P = 0.02).
Low 0.78 (0.67, 0.92)
Treatment-naive 0.77 (0.69, 0.87)
Genotype 1 0.86 (0.79, 0.94)
High 0.82 (0.71, 0.95)
With 0.78 (0.57, 1.08)
Pegylated interferon alpha-2a 0.76 (0.64, 0.89)
High* 0.84 (0.77, 0.92)
Relapsers* 0.77 (0.57, 1.04)
Genotype 2 or 3* 0.73 (0.57, 0.95)
Low* 0.71 (0.59, 0.85)
Without* 0.80 (0.74, 0.87)
Pegylated interferon alpha-2b* 0.87 (0.81, 0.94)
Non-responders# 0.93 (0.84, 1.03)
Genotype 4, 5 or 6* 0.73 (0.56, 0.94)
There was no significant difference in liver-related morbidity plus all-cause mortality in the five trials reporting this outcome (OR: 1.22, 95% CI: 0.33–4.51; I2 = 41%) or in all trials (risk difference: 0.00, 95% CI: 0.00–0.00; 99% CI: −0.01–0.01).
There were no significant differences between the two interventions regarding histological inflammation score (RR: 0.54; 95% CI: 0.24–1.23; n = 2) or fibrosis score (RR: 1.05; 95% CI: 0.87–1.27; n = 2) according to the random-effects model. According to the fixed-effect model, pegylated interferon plus ribavirin significantly increased the number of patients with improvement of histological inflammation score (RR: 0.40; 95% CI: 0.35–0.45; n = 2).
One trial with 897 treatment-naïve patients reported quality-of-life data.40 During treatment, pegylated interferon plus ribavirin was associated with significantly better quality of life on Fatigue Severity Scale (WMD: 1.50; 95% CI: 0.32–2.68) and in two domains of SF 36 Health Survey: bodily pain (WMD: 3.70; 95% CI: 1.35–6.05) and vitality (WMD: 2.60; 95%CI: 0.44–4.76). There was no significant difference between the two arms after treatment.
Pegylated interferon plus ribavirin significantly increased the risks of neutropenia (15% vs. 6%; RR: 2.25; 95% CI: 1.58–3.21; n = 10 trials), thrombocytopenia (6% vs. 2%; RR: 2.28; 95% CI: 1.14–4.54; n = 8), arthralgia (30% vs. 24%; RR: 1.19; 95% CI: 1.05–1.35; n = 4), injection-site reaction (55% vs. 30%; RR: 2.56; 95% CI: 1.06–6.22; n = 3), injection site inflammation (55% vs. 18%; RR 1.45; 95% CI 1.17–1.79; n = 1) and dermatological symptoms (36% vs. 20%; RR 1.78; 95% CI 1.15–2.73; n = 1). According to fixed-effect model analyses, the risk of nausea and coughing was also increased.
Subgroup analysis comparing pegylated interferon alpha-2a and -2b revealed that the risk of neutropenia and thrombocytopenia was significantly higher with alpha-2a (tests for interaction P = 0.005 and P = 0.022, respectively). We were not able to perform subgroup analyses on injection–site reactions because they were reported only in trials using alpha-2b.
Dose reductions were significantly more common in the pegylated interferon plus ribavirin arm (38% vs. 29%; RR: 1.44; 95% CI: 1.14–1.82; n = 8), but according to only the fixed-effect model significantly less patients on pegylated interferon plus ribavirin discontinued treatment (17% vs. 21%; RR: 0.83; 95% CI: 0.74–0.95; n = 12). There was no significant difference between alpha-2a and -2b regarding dose reductions and discontinuation of treatment (test of interaction, P = 0.638 and P = 0.053, respectively). Sensitivity analysis including only patients who discontinued treatment because of early stopping rules showed no significant difference between pegylated interferon plus ribavirin vs. interferon plus ribavirin (16% vs. 20%; RR: 0.80; 95% CI: 0.59 to 1.09; n = 3).
Our systematic review of 18 randomized trials included 4811 patients. We found that pegylated interferon plus ribavirin had beneficial effects on the sustained virological response and, presumably, on liver inflammation and quality of life during treatment compared with non-pegylated interferon plus ribavirin. There was no significant difference regarding liver-related morbidity plus all-cause mortality, liver fibrosis, or post-treatment quality of life. Pegylated interferon plus ribavirin increased the risk of dose reductions and adverse events including neutropenia, thrombocytopenia, arthralgia, and injection-site reaction, but was associated with less treatment discontinuations.
According to subgroup analyses, pegylated interferon plus ribavirin had a significantly beneficial effect on sustained virological response for treatment-naïve patients, regardless of viral genotype, viral load, or interferon. Our findings are in accordance with previous reviews comparing pegylated and non-pegylated interferon regimens54, 55 as well as the American Association for the Study of Liver Diseases guideline.24Results from only three trials were available for relapsers and non-responders, which may explain the non-significant results in these groups. Although, test of interaction indicates that the benefit of pegylated interferon is similar in relapsers compared to treatment-naïve patients. The potential benefit (if any) in non-responders is significantly smaller.
Although pegylated interferon plus ribavirin improved clearance of hepatitis C virus, we found no significant difference between the two interventions regarding liver-related morbidity plus all-cause mortality. The follow-up of 24 weeks is very short to assess any long-time benefits considering the protracted course of chronic hepatitis C.7, 8 Cohort studies suggest that viral clearance is associated with reduced risk of liver morbidity and mortality.56, 57 However, randomized trials with longer follow-up are necessary to determine possible long-term clinical benefits of pegylated interferon plus ribavirin.
Only two trials reported histological outcome and assessed it blindly. We found no significant difference between the two interventions regarding fibrosis. Regarding histological inflammation, we found some evidence of beneficial effect of pegylated interferon plus ribavirin, but this was not robust to choice of meta-analysis model. A study comparing the impact of 10 different regimens combining interferon, pegylated interferon and ribavirin on liver fibrosis has been published.58 This analysis pooled individual data from 3010 treatment-naïve patients from four randomized clinical trials. The analysis revealed that fibrosis progression rate per year was lower after treatment than before for all regimens without significant difference between treatments. This supports our finding of no significant difference regarding fibrosis response after treatment between the two compared interventions.
In general, adverse events were reported insufficiently and only a single trial reported quality of life.40 According to this trial, the quality of life during treatment with pegylated interferon plus ribavirin was significantly better. Significantly fewer patients discontinued treatment with pegylated interferon plus ribavirin. Better quality of life and fewer patients discontinuing pegylated interferon contradict our findings of increased adverse events. A likely explanation is that these trials were unblinded and that investigators and patients adhered to the experimental intervention by performing dose reductions of pegylated interferon.
Reported adverse events were statistically significantly increased. They included mainly haematological laboratory abnormalities. Although the increase in adverse events may not seem dramatic, the fact that significantly more pegylated-interferon patients had dose reductions may signal their clinical importance. To check the possibility that the adverse events and dose reductions were predominantly caused by a single type of pegylated interferon, we performed additional subgroup analyses. This revealed no significant difference between pegylated interferon alpha-2a and -2b. Acknowledging the limitations of subgroup analyses, head-to-head trials comparing the two pegylated interferons are needed to evaluate this thoroughly.
To avoid publication and ascertainment bias, we included trials regardless of publication status and language. Authors independently extracted data to avoid information bias. However, our review has potential limitations. None of the trials had long-term follow-up. Secondly, only seven of 18 trials (39%) reported adequate randomization methods. Thirdly, none of the trials was double blind. These aspects may affect the risk of allocation, performance, and ascertainment bias.27, 59, 60 We found no significant association between methodological quality and trial results. These findings support the robustness of our results, but do not exclude the possibility of bias. Thirdly, in several of our meta-analyses, heterogeneity was significant. If there is considerable variation in trial results, it may be misleading to quote an average value for the treatment effect. Viral load was identified as a possible source of heterogeneity. However, as no major difference employing fixed- and random-effects model analyses was found, it is not likely that heterogeneity affected our overall conclusions.
In conclusion, we found evidence that pegylated interferon plus ribavirin increases the sustained virological response, but at the cost of more adverse events. There is insufficient evidence on impact of pegylated interferon plus ribavirin on mortality and liver-related morbidity as well as on liver histology. Longer follow-up assessment on existing trials and research on the long-term association between sustained virological response and liver-related morbidity and all-cause mortality seems warranted.
The study was funded by the Copenhagen Trial Unit, Centre for Clinical Intervention Research, Rigshospitalet, Denmark. This review will be published as a Cochrane Review in The Cochrane Library. Cochrane Reviews are regularly updated as new evidence emerges and in response to comments and criticisms, and The Cochrane Library should be consulted for the most recent version of the Review.