|Efficacy of telaprevir and boceprevir in treatment-naive and treatment-experienced genotype 1 chronic hepatitis C patients: an indirect comparison using Bayesian network meta-analysis
|Cure S, Diels J, Gavart S, Bianic F, Jones E
This indirect Bayesian comparison of telaprevir versus boceprevir, combined with pegylated interferon alpha 2a or 2b and ribavirin, for genotype one chronic hepatitis C, suggested better efficacy for telaprevir, with a probability of 0.93 for patients new to treatment, and 0.98 for patients who had received treatment. These conclusions were cautious, and are likely to be reliable.
To indirectly compare the efficacy of telaprevir versus boceprevir, combined with pegylated interferon alpha 2a or 2b and ribavirin, for patients, with genotype one chronic hepatitis C, who had or had not previously been treated.
MEDLINE, EMBASE, The Cochrane Library, and CRD databases, were searched, for studies published in English, from January 2000 to July 2011; search terms were reported. The bibliographies of relevant systematic reviews and meta-analyses, and recent conference abstracts, were searched.
Randomised controlled trials (RCTs) comparing the efficacy of pegylated interferon alpha 2a or 2b and ribavirin versus another pegylated interferon and ribavirin, or telaprevir- or boceprevir-based therapy, were eligible for inclusion. Patients had to be adults infected with genotype one chronic hepatitis C, who had failed to respond to standard treatment or had not received treatment before. The primary outcome was the sustained viral response, at 24 weeks after the end of treatment (defined as an undetectable hepatitis C virus RNA level). Trials of patients with co-infections, such as HIV or hepatitis B, were excluded.
In the included trials, the interventions were pegylated interferon alpha 2a or 2b and ribavirin; pegylated interferon and ribavirin followed by triple therapy plus boceprevir; pegylated interferon and ribavirin with boceprevir; pegylated interferon and ribavirin with telaprevir; and pegylated interferon and ribavirin with telaprevir, followed by pegylated interferon and ribavirin alone. Treatment lasted from four to 48 weeks; sometimes guided by the response. The mean or median age of patients ranged from 44 to 54.8 years; the percentage who were male ranged from 27 to 73. Where reported, the percentage who were Black ranged from one to 19; and the percentage with a viral load of over 800,000 international units of RNA ranged from 53 to 92.
Two reviewers selected the trials, with differences reconciled by a third reviewer.
Assessment of study quality
Cochrane criteria were used to assess trial quality for adequacy of randomisation (performance and detection bias); concealment of treatment allocation (selection bias); the comparability of groups at baseline; differences in drop-out between groups; and the appropriate use of an intention-to-treat analysis (attrition bias).
It appears that two reviewers assessed quality, with differences reconciled by a third reviewer.
Two independent reviewers extracted the data, with differences reconciled by a third reviewer. The numbers of events were used to calculate odds ratios, with 95% credible intervals.
Methods of synthesis
A Bayesian indirect network meta-analysis was used to pool the results, using both fixed-effect and random-effects models, to give odds ratios, with 95% credible intervals. There were too few trials to calculate Cochran Q or Ι², to estimate heterogeneity between them. Τ² was used for treatment-experienced patients.
Separate analyses were performed for treatment-naive and treatment-experienced patients. For treatment-experienced patients, data from patients who had relapsed and those who had partially responded were included as no clinical results for non-responders were available for boceprevir.
Sensitivity analyses were performed excluding outlier trials, and to generate separate estimates for partial responders and those who had relapsed, for the treatment-experienced patients. With the Bayesian approach, significance was indicated by the probability of the result being above one.
Results of the review
Eleven RCTs were identified, with 6,927 patients (range 74 to 2,054). Eight RCTs were of treatment-naive patients (5,584; range 74 to 2,054) and three were of treatment-experienced patients (1,343; range 77 to 863). Seven trials met all quality criteria and the other four were unclear for one criterion or more. The fixed-effect model sustained virologic response results are reported.
Treatment-naive patients: With pegylated interferon alpha 2a and 2b plus ribavirin, there were probably positive benefits, in sustained virologic response, with telaprevir (12 weeks of response-guided treatment) versus pegylated interferon and ribavirin; and with boceprevir (24 weeks of response-guided treatment) versus pegylated interferon and ribavirin. The results for three other comparisons were not probable. The overall result was a probability of 0.93 of telaprevir being more effective than boceprevir (OR 1.42, 95% CrI 0.89 to 2.25).
Treatment-experienced patients: There were probably positive benefits with telaprevir (12 weeks) versus pegylated interferon and ribavirin; and with boceprevir (32 weeks of response-guided treatment) versus pegylated interferon and ribavirin. There were probable benefits with telaprevir versus boceprevir. The overall result had a probability of 0.98 of telaprevir being more effective than boceprevir (OR 2.45, 95 CrI 1.02 to 5.80). This result was governed by the stronger effect for patients who had relapsed than for partial responders.
The results of the sensitivity analyses were reported. The random-effects model produced lower probabilities for some analyses.
In the absence of trials directly comparing telaprevir and boceprevir, combined with pegylated interferon alpha 2a or 2b and ribavirin, for patients with genotype one chronic hepatitis C, the indirect Bayesian network comparison suggested better efficacy for telaprevir than for boceprevir, with a probability of 0.93 for treatment-naive patients, and 0.98 for treatment-experienced patients.
This review addressed a well-defined question for study design, participants, interventions and the relevant outcome. Due to the language restriction, relevant studies may have been missed. Trial quality was assessed, using suitable criteria, and was generally adequate. Efforts were made to reduce error and bias, within the review processes. Relevant trial details were provided, but the interventions were not described consistently; descriptions from the tables were used. The indirect Bayesian synthesis was appropriate, since head-to-head trials were not available. Suitable sensitivity analyses were performed. It was not possible to explore heterogeneity effectively, since too few trials were identified and the interventions varied widely.
The reliability of the results is limited by the indirect comparisons, variability that could not be fully explored, and the fact that credible intervals for one comparison included harm as well as benefit. The authors' conclusions were cautious, stating that greater efficacy was suggested, and this is likely to be reliable.
Implications of the review for practice and research
Practice: The authors noted that the treatment regimens used in the phase III trials corresponded most to the product label descriptions and were the main focus of their comparisons, but there were some minor differences.
Research: The authors did not make any recommendations for research.
Funded by Janssen, manufacturers of telaprevir.
Cure S, Diels J, Gavart S, Bianic F, Jones E. Efficacy of telaprevir and boceprevir in treatment-naive and treatment-experienced genotype 1 chronic hepatitis C patients: an indirect comparison using Bayesian network meta-analysis. Current Medical Research and Opinion 2012; 28(11): 1841-1856
Subject indexing assigned by NLM
Antiviral Agents /administration & Bayes Theorem; Drug Therapy, Combination; Genotype; Hepacivirus /genetics; Hepatitis C, Chronic /drug therapy /virology; Humans; Oligopeptides /administration & Proline /administration & Randomized Controlled Trials as Topic; derivatives /therapeutic use; dosage /analogs & dosage /therapeutic use; dosage /therapeutic use
Date bibliographic record published
Date abstract record published
This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.