Efficacy and Tolerability of Peginterferon α-2a and Peginterferon α-2b, Both plus Ribavirin, for Chronic Hepatitis C: A Meta-Analysis of Randomized Controlled Trials

Background. The efficacy and tolerability of peginterferon α-2a and peginterferon α-2b in chronic hepatitis C (CHC) patients remain controversial. Methods. PubMed, Ovid, and Cochrane libraries were electronically searched until August 30, 2012. Studies that met the inclusion criteria were systematically evaluated by two reviewers independently. Results. The overall sustained virologic response (SVR) rate of the peginterferon α-2a group was significantly higher than that of the peginterferon α-2b group (46.7% versus 42.4%, P value = 0.01). The same tendency was observed for naïve, genotype 1/4, and genotype 2/3 patients. The early virologic response (EVR) and end-of-treatment response (ETR) rates were significantly higher in the peginterferon α-2a group than in the peginterferon α-2b group (56.1% versus 49.8%, P < 0.0001; 67.9% versus 56.6%, P < 0.00001, resp.). Peginterferon α-2a had a significantly lower discontinuation rate than peginterferon α-2b (27.9% versus 33.9%, P < 0.0001) in naïve patients. In both naïve CHC and hepatitis C virus genotype 1 patients, peginterferon α-2a had a higher relapse rate than peginterferon α-2b. Conclusions. Peginterferon α-2a has superior efficacy with higher EVR, ETR, and SVR than peginterferon α-2b for CHC patients, both plus ribavirin. Peginterferon α-2a might obtain a similar or even lower discontinuation rate than peginterferon α-2b. However, peginterferon α-2a had a higher relapse rate than peginterferon α-2b.


Introduction
The World Health Organization has estimated that up to 170 million people (approximately 3% of the world population) worldwide might be infected with hepatitis C virus (HCV). This virus is responsible for approximately 350,000 deaths every year. HCV is cleared spontaneously in only approximately 20% of individuals. Chronic infection frequently progresses to cirrhosis, end-stage liver disease, hepatocellular carcinoma, and death [1][2][3][4].
Currently, in many countries, the recommended therapy for chronic hepatitis C (CHC) is still the combination of peginterferon and ribavirin [1,2]. Two licensed products of peginterferon are available: peginterferon -2a (Pegasys, Hoffmann-La Roche, Nutley, NJ, USA) and peginterferon -2b (Peg-Intron, Schering Plough Corp., Kenilworth, NJ, USA). However, differences in structural modifications and dosing (weight-adjusted versus fixed) between the two peginterferons may lead to various clinical outcomes. In addition, a recommendation about the two regimens has not been proposed in the current guidelines [5][6][7][8][9][10][11]. Although recent studies have compared the response rates obtained using the two peginterferons in CHC, they have failed to reach a consensus as to which treatment options are the most effective.
Some systematic reviews [12][13][14][15], which include meeting abstracts or HCV/HIV coinfected patients, concluded that peginterferon -2a has higher sustained virologic response (SVR) than peginterferon -2b in CHC but revealed that both have similar safety. The virologic responses and tolerability of peginterferon plus ribavirin in HCV/HIV coinfected patients are substantially different from those in chronic HCV monoinfected patient. In addition, some reported meeting abstracts were found to be inadequate. Thus, we performed a meta-analysis of randomized controlled trials (RCTs) with critical inclusion and exclusion criteria to evaluate the efficacy and tolerability of the two regimens.

Search Strategy.
We searched PubMed, Ovid, and Cochrane libraries until August 30, 2012. The following medical subject headings were used: "Hepatitis C, Chronic, " "interferons, " "peginterferon alfa/alpha/ -2a, " "peginterferon alfa/alpha/ -2b, " and "ribavirin. " Electronic searches were supplemented with manual searches of reference lists of all retrieved review articles, primary studies, and abstracts from meetings to identify other studies not found in the electronic searches. The literature was searched by two authors (Z. Yang and L. Zhuang) independently.

Study Selection.
Two authors independently selected trials and discussed them with each other when inconsistencies were found. Articles that meet the following criteria were included: (1) study types, randomized controlled trials; (2) participants, chronic HCV virus monoinfection patients either naïve or retreatment were randomly divided into two groups; (3) interventions, peginterferon -2a and peginterferon -2b, both plus ribavirin; (4) outcome measures, studies that used one or more of the following measurements were eligible: rapid virologic response (RVR), early virologic response (EVR), end-of-treatment virologic response (ETR), SVR, relapse rate, and discontinuation rate; and (5) full texts available.
Studies with the following situations were excluded: (1) followup period less than 6 months and (2) studies that included patients with other liver diseases (e.g., HBV infection, human immunodeficiency virus infection, and hepatocellular carcinoma) aside from HCV.

Quality Assessment.
The methodological qualities of the included RCTs were assessed according to Cochrane Collaboration's tool described in Handbook version 5.1.0 [16]. Two authors (Z. G. Yang and L. Yang) assessed the quality independently, and inconsistency was discussed with a third review author (X. R. Chen) who acted as an arbiter.

Data Extraction.
Two researchers read the full texts independently and extracted the following contents: publication data (first author's name, year of publication, and country of population studied), study design, sample size, patient characteristics (age, gender, body weight, distribution of genotype, and liver histology), treatment protocol (peginterferon type and dose, ribavirin dose), outcome measures (RVR, EVR, ETR, SVR, relapse rate, and discontinuation rate), and reasons for discontinuing combination therapy.
Authors were contacted by e-mail for additional information if data were unavailable.

Definitions.
Chronic hepatitis C is defined by anti-HCV positive, HCV RNA positive as determined by a qualitative polymerase chain reaction (PCR) assay for more than 6 months. The primary outcome measure of efficacy of SVR was defined by a sensitive PCR assay as the absence of HCV RNA from serum at 24 weeks after completion of therapy. Secondary outcome measures of tolerability, including discontinuation rate, RVR, EVR, and ETR, were also determined. RVR was defined using a sensitive PCR assay as undetectable HCV RNA at 4 weeks after treatment. EVR was defined as ≥2 log reduction or complete absence of HCV RNA at 12 weeks after therapy compared with the baseline level. Undetectable virus at the end of either a 24-week or 48-week course of therapy was referred to as ETR. Virologic relapse refers to the reappearance of HCV RNA in serum after treatment was discontinued and ETR was documented.

Statistical
Methods. Data were processed in accordance with the Cochrane Handbook [16]. Intervention effects were expressed with odds ratios (ORs) and associated 95% confidence intervals (CIs) for dichotomous data. By contrast, the effects were expressed with mean differences and 95% CIs for continuous data. Heterogeneity among studies was informally assessed by visual inspection of forest plots and formally estimated using 2 and 2 tests (both > 0.05; 2 < 50% indicates no evidence of heterogeneity between the pooled studies) [17]. The fixed-effects model was first used for meta-analyses. The random-effects model was used in the presence of heterogeneity. Description analysis was performed when the quantitative data could not be pooled. Intention-to-treat (ITT) principle was used. Review Manage (v. 5.1; The Cochrane Collaboration) was used for data analysis.
Totally 1845 and 1823 patients were randomly treated with peginterferon -2a and peginterferon -2b, respectively, both plus ribavirin. The baseline characteristics of each study included in this meta-analysis are described in Table 2.

Methodological Quality Assessment.
All studies included in this meta-analysis were described as randomized. Three  studies [5,6,11] did not report the method of randomization, but randomization was adequate in other studies [7][8][9][10]. Among these studies, two were randomized by a computergenerated randomization list [9,10], one was randomized by an interactive voice system [8], and the study by Scotto et al. was randomized by a table of random numbers [7]. One study revealed that the randomization list was not available to the treating physicians. Double blinding was described in one trial by McHutchison et al. [8]. And, Ascione et al. [10] designed a study where the physician received the report on the allocation of each patient from an independent researcher who knew nothing about the patient except for the genotype.
The statistical analyses in one study by Yenice et al. [5] were not based on ITT, and more than 20% of the participants in the study by McHutchison et al. were lost to followup, both of which were considered as high risk in the item of incomplete outcome data. No descriptions of lost to followup were found in the two studies by Di Bisceglie et al. [6] and Scotto et al. [7], thus accounting for the ambiguity in the item of incomplete outcome. No patient was lost to followup in the study by Ascione et al., and the other studies described the lost to followup participants, which were balanced between the two groups and considered low risk. Selective reporting was found in the study by Di Bisceglie et al. because it failed to include the expected results (e.g., SVR rate) for such a study. The other potential biases were unclear in these trials ( Figure 1).  Figure 2(a)).

Discussion
Most previous meta-analyses concluded that peginterferon -2a has higher SVR rate than peginterferon -2b in CHC patients, but no difference in the safety profile was noted [12][13][14][15]. However, a recent meta-analysis has revealed that these two types of peginterferons have similar effects on RVR, SVR, and tolerability [29]. Moreover, the above analyses included either meeting abstracts or coinfected patients of HIV/HCV, Gastroenterology Research and Practice 9 which may have an impact on the conclusions. In the present meta-analysis, we included more RCTs and restricted our trial analyses to full papers. We excluded abstracts because they did not contain adequate details of patients and outcomes.
Interferon-based therapy could lower the risk of cirrhosis and hepatocellular carcinoma and improve the survival of CHC patients who have an SVR with a large possibility through eradicating HCV and cutting liver fibrosis procession. Our analysis showed that peginterferon -2a might achieve a higher SVR rate than peginterferon -2b, including nonresponders. Subgroup analysis revealed that peginterferon -2a was also more effective than peginterferon -2b for HCV genotype 1 or 4 patients or treatment-naïve patients. However, these two types of peginterferons had similar SVR effects on HCV genotype 2 or 3 patients. These analyses indicated a difference in antiviral activity between the two therapeutic regimens. A previous study [30] proved that combination therapy with peginterferon -2a is an independent pretreatment predictor of SVR (OR = 1.88, 95% CI = 1.20-2.96). Peginterferon -2a achieves higher SVR rates than peginterferon -2b in patients infected with HCV-1 and HCV-2; however, the two therapeutic regimens obtain similar SVR rates in patients infected with HCV-3 and HCV-4 [9]. Our results indicated that patients with genotype 2 or 3 had similar SVR rates in both groups. Given that the patients included in this meta-analysis mostly had HCV genotype 1 or 4, only less than 200 patients in each group were infected with HCV genotype 2 or 3; high-quality trials with a large sample size are needed to estimate the efficacy of the two regimens for genotype 2 or 3 CHC patients, especially for the comparison of the therapeutic efficacy in each genotype stratum.
Further analysis showed that no significant difference in RVR rate was found in the patients treated with the two peginterferon--based regimens. However, peginterferon -2a could achieve higher EVR and ETR rates in CHC patients than peginterferon -2b. Early eradication of HCV is important to the therapeutic resolution of CHC, and RVR remains the most notable on-treatment response predictor of SVR. Moreover, the present guidelines concluded that the absence of EVR is the most robust means of identifying nonresponders. Approximately 97%-100% of the treatmentnaïve patients with HCV genotype 1 infection who did not reach EVR failed to elicit SVR. Thus, patients without EVR can discontinue therapy early without compromising their chance to elicit SVR [1,2]. This finding might be associated with the potentially higher SVR rate of patients treated with peginterferon -2a. ETR does not accurately predict the occurrence of SVR; however, ETR is necessary for SVR to take place [1,2,31].
Our meta-analysis of RCTs [5][6][7][8][9][10] suggests that the two peginterferons may be comparable with regard to any reasons leading to treatment discontinuation, including naïve and retreatment patients with any HCV genotype. However, for naïve CHC patients, peginterferon -2a had a significantly lower discontinuation rate than peginterferon -2b. Previous meta-analyses [12][13][14][15] concluded that peginterferon -2a has a similar safety profile as peginterferon -2b. Given that our results were based on ITT analysis, all patients who withdrew therapy were considered as treatment discontinuation, either for adverse events or nonsafety reasons. The reason above may explain why our analysis of discontinuation rate in naïve CHC patients conflicted with those of the previous studies.
Although peginterferon -2a should achieve higher virologic responses and gain lower discontinuation rate, peginterferon -2a had a higher relapse rate than peginterferon -2b. The high relapse rate with peginterferon -2a was a novelty, as in previous studies. Relapse rates ranged from 17% to 25% for peginterferon -2a in patients with HCV genotype 1 [32,33], which is significantly lower than the 31.5% reported in the IDEAL study [8]. These findings were not supported by two randomized studies that reported no difference in relapse rate between the two regimens [9,10]. Many factors might have contributed to the difference in the findings above. Some of these factors include differences in epidemiological and genetic characteristics, mean body weight, distribution of genotype CC in the IL28B polymorphism, and ribavirin dose reduction schemes applied to the two regimens [34]. Maintaining a high ribavirin dose (≥12 mg/kg/day) during the full treatment period can lead to suppression of relapse in HCV-1 patients responding to peginterferon -2b plus ribavirin. Ribavirin dosing seems to be instrumental in preventing posttreatment relapse [35], and ribavirin concentration in the later stages of treatment is an important marker for discriminating relapse [34,36]. In the present meta-analysis, no significant difference in peginterferon and/or ribavirin dose reduction was found between the two groups. However, in the IDEAL study by McHutchison et al. [8], the dose reduction for the peginterferon -2b arm occurred in two steps. The first step was a reduction of either 200 mg (in patients receiving 800 mg/day-1,200 mg/day of ribavirin) or 400 mg (in patients receiving 1,400 mg/day). The second step was reduction by another 200 mg, if required for resolution of the adverse event. For the peginterferon -2a arm, the dose was reduced to 600 mg/day. The abrupt reduction of ribavirin dose to 600 mg/day might have played a crucial role in the high relapse rates observed in patients receiving the peginterferon -2a regimen [8][9][10]34].
Therefore, the peginterferon -2a regimen holds a slight advantage in terms of virologic responses and discontinuation rates compared with the peginterferon -2b regimen. This advantage may be considered as a direct consequence of the better pharmacokinetic profile of peginterferon -2a than peginterferon -2b. The pharmacodynamic properties of peginterferon -2a allow slower absorption and elimination than peginterferon -2b. Therefore, maximum concentrations occur later with peginterferon -2a than with peginterferon -2b. Peginterferon -2b is associated with fluctuating blood levels and rapid rise and fall in the blood level because of the relatively rapid release of interferon -2b molecule [37][38][39]. Previous studies [38,40] showed that the concentration of peginterferon -2b did not remain stable over the week as a whole. At the end of the week, serum interferon could not be detected in most patients treated with peginterferon -2b. When interferon was no longer detectable in the serum, the viral load increased until the next interferon injection. This phenomenon increases the potential for more side effects and reduces the efficacy of the drug. Peginterferon -2b is distributed widely in the body fluids and tissues [14,39]. By contrast, peginterferon -2a is distributed predominantly to the blood and interstitial fluid, resulting in high drug concentrations in the liver. The reduced clearance of peginterferon -2a, as a consequence of metabolism via nonspecific proteases, provides significant, consistent, and measurable therapeutic plasma levels even at the end of the weekly dosing period [41]. These differences between the two types of peginterferons should lead to better compliance and superior safety of peginterferon -2a [14].
In conclusion, current evidence suggests that peginterferon -2a has superior efficacy with higher EVR, ETR, and SVR than peginterferon -2b for CHC patients, both plus ribavirin. Peginterferon -2a might obtain similar or even lower discontinuation rate than peginterferon -2b. However, peginterferon -2a had a higher relapse rate than peginterferon -2b. Further trials must focus on the comparison of the two types of peginterferons in terms of achieving SVR and clinically relevant outcomes, such as liver-related cirrhosis, hepatocellular carcinoma, mortality, and morbidity.