SEARCH

SEARCH BY CITATION

Keywords:

  • boceprevir;
  • hepatitis C;
  • meta-analysis;
  • protease inhibitors;
  • telaprevir

Summary

  1. Top of page
  2. Summary
  3. What is known and objective
  4. Methods
  5. Results
  6. Discussion
  7. What is new and conclusions
  8. Acknowledgements
  9. References

What is known and objective

Two NS3/4A protease inhibitors (PIs), telaprevir and boceprevir, were recently approved in the United States. The primary objective was to compare the efficacy and safety of triple therapies including either PI to dual therapy in patients with chronic hepatitis C genotype 1; the secondary objective was to conduct subgroup analyses to make comparisons based on patients' race.

Methods

Published and unpublished RCTs were selected if they: (i) had patients with chronic hepatitis C genotype 1, (ii) compared triple therapies (telaprevir or boceprevir + peg-interferon + ribavirin) and dual therapy (peg-interferon + ribavirin) and (iii) measured the outcome using sustained virologic response (SVR).

Results

A total of 4421 patients from 10 evaluated articles were included in the meta-analysis. Overall, triple therapy was significantly associated with a higher achievement of SVR than dual therapy: (i) telaprevir-based triple therapy in treatment-naïve patients (relative risk [RR] = 1·62; 95% confidence interval [CI], 1·47–1·78), (ii) telaprevir-based triple therapy in treatment-experienced patients (RR = 3·85; 95% CI, 3·03–4·90), (iii) boceprevir-based triple therapy in treatment-naïve patients (PR = 1·70; 95% CI, 1·56–1·86) and (iv) boceprevir-based triple therapy in treatment-experienced patients (RR = 2·98; 95% CI, 2·29–3·87). Although black and non-black patients demonstrated the higher rates of achieving SVR with triple therapy compared to dual therapy, the rates of SVR were still lower among black patients than among non-black patients. Patients on triple therapies had the significantly increased incidences of treatment discontinuation attributable to adverse events and serious adverse events when compared to dual therapy, especially treatment-experienced patients.

What is new and conclusions

Regarding achieving SVR, triple therapies including either PI are superior to dual therapy for both treatment-naïve and treatment-experienced patients.


What is known and objective

  1. Top of page
  2. Summary
  3. What is known and objective
  4. Methods
  5. Results
  6. Discussion
  7. What is new and conclusions
  8. Acknowledgements
  9. References

Hepatitis C virus (HCV) infections are a significant global health problem. Worldwide, an estimated 160 million people (2·35% of the world population) are infected with HCV.[1] Hepatitis C is also the most common chronic bloodborne infection in the United States, affecting an estimated 3·2 million people.[2] HCV infection is one of the critical indicators for cirrhosis, decompensated liver disease, hepatocellular carcinoma and liver transplantation.[3-6] Based on genetic differences between HCV isolates, the HCV species is classified into at least six major genotypes, each of which has several subtypes. Genotype 1 is the most common HCV genotype in the United States, followed by genotypes 2 and 3.[7] HCV genotypes 4, 5 and 6 are less common but are becoming more frequently observed in the United States.[7, 8]

Dual therapy, which is a combination of pegylated interferon (peg-IFN)-alfa and ribavirin, has been the standard of care (SOC) in the treatment for hepatitis C for many years.[9, 10] Dual therapy is administered in patients with HCV genotypes 1, 4, 5, or 6 for 48 weeks and in those with genotypes 2 or 3 for 24 weeks.[9-11] The response to HCV treatment is defined by change in the HCV-RNA level during and after treatment and measured by HCV viral load tests at different times: (i) rapid virologic response (RVR) defined as an undetectable HCV-RNA after 4 weeks of therapy, (ii) early virologic response (EVR) defined as an undetectable HCV-RNA or at least a 2-log10 IU/mL reduction in HCV-RNA at 12 weeks of therapy, (iii) end of treatment (ETR) defined as an undetectable HCV-RNA at completion of therapy and (iv) sustained virologic response (SVR) defined as an undetectable HCV-RNA in peripheral blood 24 weeks after the end of treatment. The primary aim in treating HCV infection is to achieve an SVR, which is considered a virologic ‘cure’.[12-14] Generally, patients with genotypes 2 or 3 are adequately treated by dual therapy for 24 weeks; the rate of SVR following this 24-week period approaches 80%.[15] However, SVR is achieved in only 42–52% of patients who are infected with HCV genotype 1 and who are treated with dual therapy for 48 weeks.[16]

Accordingly, a shift towards therapies that more effectively inhibit HCV is essential. Novel treatments, NS3/4A protease inhibitors (PIs), have been identified as the life cycle of the HCV has become known. PIs prevent HCV replication by directly inhibiting the action of NS3/4A.[17, 18] Incivek® (Vertex Pharmaceuticals Inc., Cambridge, MA, USA) (telaprevir) and Victrelis® (Merck & Co., Inc., Whitehouse Station, NJ, USA) (boceprevir) are currently the only approved NS3/4A PIs. The U.S. Food and Drug Administration (FDA) and the European Medicines Agency (EMA) have approved these drugs for the treatment of patients infected with HCV genotype 1.[19] The addition of telaprevir and boceprevir to dual therapy strongly improves the odds of achieving an SVR in treatment-naïve HCV genotype 1 patients as well as in prior non-responders and relapsers when compared with standard therapy.[20] In 2011, boceprevir was approved by the FDA for the treatment of chronic HCV genotype 1 infection, in combination with peg-IFN-a and ribavirin, in adults (≥18 years and older) with compensated liver disease, including cirrhosis, who are previously untreated or those in whom previous IFN and ribavirin therapy has failed.[21, 22] Telaprevir, in combination with peg-IFN-alfa and ribavirin, was approved by the FDA in 2011 for the treatment of genotype 1 chronic HCV in adult patients with compensated liver disease, including cirrhosis, who are treatment-naïve or who have been previously treated with interferon-based treatment, including prior null responders, partial responders and relapsers.[23, 24]

Several randomized controlled trials (RCTs) have demonstrated the efficacy and safety of telaprevir and boceprevir. Additionally, five meta-analysis studies have been conducted on PIs; two of those studies were limited to only telaprevir and summarized the efficacy and safety of telaprevir,[25, 26] one of those studies conducted a systematic review/meta-analysis of both telaprevir and boceprevir in only treatment-naïve patients,[27] and the others used indirect methods to compare telaprevir and boceprevir.[28, 29] To our knowledge, no researchers to date have compared the efficacy and safety of dual therapy and triple therapy with telaprevir and boceprevir in both treatment-naïve and treatment-experienced patients. Therefore, there is a need to summarize the efficacy and safety of both triple therapies, including telaprevir or boceprevir, in comparison with dual therapy not only in treatment-naïve patients but also in treatment-experienced patients. In addition, although race/ethnicity is known to affect the treatment of patients with chronic hepatitis C genotype 1,[30] a meta-analysis using subgroup analysis based on patients' race/ethnicity has not been conducted. Therefore, the primary objective was to compare the efficacy and safety of triple therapies including either telaprevir or boceprevir to dual therapy in patients with chronic hepatitis C genotype 1. Furthermore, the secondary objective was to conduct subgroup analyses to make comparisons based on patients' race (black patients vs. non-black patients).

Methods

  1. Top of page
  2. Summary
  3. What is known and objective
  4. Methods
  5. Results
  6. Discussion
  7. What is new and conclusions
  8. Acknowledgements
  9. References

Search strategy

We conducted a Medical Literature Analysis and Retrieval System Online (MEDLINE) search for RCTs of NS3/4A PIs for genotype 1 HCV infection through November 2012. The search terms were ‘boceprevir’, ‘telaprevir’ and ‘NS3/4A protease inhibitors’. We also searched for completed but unpublished trials on websites (http://www.clinicaltrials.gov and http://www.clinicalstudyresults.org). In addition, we searched the reference lists of these articles.

Inclusion criteria

Both published and unpublished RCTs were selected if they: (i) had patients with chronic hepatitis C genotype 1, (ii) compared triple therapies (telaprevir or boceprevir + peg-interferon + ribavirin) and dual therapy (peg-interferon + ribavirin), (iii) measured the outcome using sustained virologic response (SVR) and (iv) were published in English.

Validity assessment

For assessment of the quality of RCTs chosen for inclusion in our meta-analysis, the validated Jadad scale and the Cochrane Collaboration's tool for assessing risk of bias were used.[31, 32] The validated Jadad scale assesses inherent controllers of bias by using the following quality assessment criteria: (i) the method to generate the sequence of randomization was described and found to be appropriate, (ii) the method of double blinding was described and found to be appropriate, and (iii) withdrawals and drop-outs were described. A possible aggregate score between 0 and 5 was calculated for each RCT included in the analysis, with a score of less than 3 points signifying a lower methodological quality.[31] Additionally, a sensitivity analysis was conducted by excluding RCTs having a lower quality based on the Jadad scale. The Cochrane Collaboration's tool for assessing risk of bias has six domains: sequence generation, allocation concealment, blinding, incomplete outcome data, selective outcome reporting and other sources of bias.[32]

Data extraction

Two reviewers (CP and SJ) screened abstracts according to the inclusion criteria. Full-text articles were retrieved and reviewed if a decision on inclusion could not be made solely based on the abstract. Items determined pertinent to study characteristics, patient characteristics and outcome results were extracted. Differences were resolved by consensus between the two reviewers.

The efficacy outcome variable extracted for this study was the rate of SVR for triple therapy compared with dual therapy. To evaluate overall safety data, the following events were extracted. First, data on discontinuation due to adverse events were retrieved. In triple therapy as compared with dual therapy, the proportions of patients who discontinued the study drug owing to adverse events were compared. Second, the proportions of patients with serious adverse events, which were defined as the number of participants who experienced one or more serious adverse events per total number of participants, were compared. Third, data on the occurrence of specific adverse events, which were defined as the most common adverse events related to the use of both PIs, were extracted. The specific adverse events included anaemia, rash, pruritus, nausea, diarrhoea and dysgeusia (distortion of the sense of taste).

When calculating the pooled estimates, the following treatment regimens in each trial were excluded. According to FDA-approved dosages, telaprevir must be administered with dual therapy (peg-interferon + ribavirin) for all treatment-naïve and treatment-experienced patients for 12 weeks, followed by a response-guided regimen of either 12 or 36 additional weeks of dual therapy depending on treatment response.[23] Based on this, triple therapy regimens where telaprevir was administered for 8 weeks or 24 weeks with dual therapy were excluded. Triple therapy regimens in which peg-interferon + ribavirin was administered for a shorter duration (12 weeks) were also excluded when combining data. Additionally, dual therapy regimens of telaprevir and peg-interferon were excluded because these regimens did not meet our study objectives.

Statistical analyses

All efficacy and safety outcomes were dichotomous variables. Therefore, the relative risk (RR) and 95% confidence interval (CI) were calculated to compare the efficacy and safety of these PIs in patients with chronic hepatitis C genotype 1. In addition to RR, the number needed to treat (NNT) for the efficacy outcome and the number needed to harm (NNH) for the main safety outcomes were calculated as the reciprocal of the pooled risk difference (RD). NNTs were rounded up, and NNHs were rounded down to the nearest whole numbers.

Overall, random-effects models were used by considering the differences in treatment regimens and populations' baseline characteristics among included trials. We used the Q-statistic test and I2 index to assess heterogeneity among effect sizes. Significance of the Q-statistic test (< 0·05) and an I2 value of 50% and greater indicate a substantial level of heterogeneity.[33] Funnel plots and the Egger's weighted regression statistics were used to identify the potential presence of publication bias. All analyses were performed using comprehensive meta analysis Version 2.0, biostat (Englewood, NJ, USA) and stata/ic 12.1 (Stata Corp., College Station, TX, USA).

Results

  1. Top of page
  2. Summary
  3. What is known and objective
  4. Methods
  5. Results
  6. Discussion
  7. What is new and conclusions
  8. Acknowledgements
  9. References

Study and patient characteristics

A total of 414 potentially relevant articles were identified for possible inclusion in the meta-analysis. Through a review of these selected abstracts independently by two reviewers, 53 were retrieved for further evaluation, and of these, 10 studies met all of the predefined inclusion criteria.[34-43] Figure 1 represents a flow diagram of the study search strategy and review process. Overall, a total of 5186 genotype 1 chronic hepatitis C patients were included in the meta-analyses. Participant baseline characteristics of the included studies were extracted and are summarized in Table 1.

Table 1. Characteristics of randomized controlled trials included in the meta-analysis
StudyClinical trial phasePopulation descriptionPrevious treatmentNo. of participantsMean ageHepatitis C subtypeHCV-RNA >800 000 (%)HCV-RNA log10 (IU/mL)
  1. FAS, full analysis set; ITT, intention to treat.

Telaprevir
Kumada et al.[34]3N/ANaïve18953·7a, bN/A6·8
Jacobson et al.[35]3FASNaïve108849·0a, b, unknown77·06·3
Hezode et al.[36]2FASNaïve32345·0a, b83·36·4
McHutchison et al.[37]2FASNaïve25049·3a, b, unknown87·36·6
Zeuzem et al.[38]3FASExperienced66250·8a, b, c, unknown88·46·6
McHutchison et al.[39]2FASExperienced22950·5a, b, unknown92·26·7
Boceprevir
Poordad et al.[40]3FASNaïve109749·3a, b, unknown85·3N/A
Kwo et al.[41]2ITTNaïve52047·5a, b, unknownN/A6·6
Bacon et al.[42]3FASExperienced40352·7a, b, unknown87·8N/A
Flamm et al.[43]3FASExperienced20152·5a, b, unknown81·2N/A
image

Figure 1. Flow diagram of study search strategy and review process.

Download figure to PowerPoint

Quality assessment of published articles included in the meta-analyses is presented in Table 2. The Jadad scores had a median of 4·2 and a range from 2 to 5. Overall, high methodological quality was shown among the 10 published RCTs. Specifically, nine studies were considered of high methodological quality, whereas only one study was classified as a low methodological quality RCT. Additionally, quality assessment using the Collaboration's tool for assessing risk of bias also showed that plausible bias was unlikely to seriously change the results.

Table 2. Quality assessment of randomized controlled trials included in the meta-analysis
 Jadad ScoreCochrane collaboration's tool for assessing risk of bias
Adequate sequence generationAllocation concealmentBlindingIncomplete outcome data AddressedFree of selective reportingFree of other bias
Telaprevir
Kumada et al.[34]2UnclearUnclearUnclearYesYesYes
Jacobson et al.[35]4UnclearUnclearYesYesYesYes
Hezode et al.[36]5YesUnclearYesYesYesYes
McHutchison et al.[37]4UnclearUnclearYesYesYesYes
Zeuzem et al.[38]5YesUnclearYesYesYesYes
McHutchison et al.[39]4UnclearUnclearYesYesYesYes
Boceprevir
Poordad et al.[40]5YesUnclearYesYesYesYes
Kwo et al.[41]3YesUnclearNoYesYesYes
Bacon et al.[42]5YesUnclearYesYesYesYes
Flamm et al.[43]5YesUnclearYesYesYesYes

Efficacy

Table 3 summarizes the rates of patients who achieved SVR for each RCT included in this study. In the 10 RCTs, the rates of achieving SVR were relatively lower in dual therapy compared to triple therapy. With dual therapy for 48 weeks, SVR was achieved in only 37·5–49·2% of treatment-naïve patients and 14·0–21·3% of treatment-experienced patients, such as relapsers, non-responders or partial responders. The rates of patients who achieved SVR with triple therapy were (i) 60·8–74·7% for telaprevir-based regimens in treatment-naïve patients, (ii) 51·3–66·3% for telaprevir-based regimens in treatment-experienced patients, (iii) 54·2–74·8% for boceprevir-based regimens in treatment-naïve patients and (iv) 58·6–66·5% for boceprevir-based regimens in treatment-experienced patients.

Table 3. Rates of sustained virologic response (SVR) of randomized controlled trials included in the meta-analysis
StudyArmNo. participants analysed (n)No. participants achieving SVR (n)Rates of SVR (%)
  1. a

    These arms were excluded in the meta-analysis because they did not meet the inclusion criteria.

  2. PR48, placebo + peg-interferon alfa (peg-IFN) + ribavirin (RBV) during 12 weeks, followed by peg-IFN + RBV during 36 weeks; T8PR24/48, TEL + peg-IFN + RBV during 8 weeks, followed by placebo + peg-IFN+RBV during 4 weeks, followed by peg-IFN + RBV from 12 to 36 weeks; T12P12, telaprevir (TEL) + peg-IFN during 12 weeks; T12PR12, TEL + peg-IFN+ RBV during 12 weeks; T12PR24, TEL + peg-IFN+RBV during 12 weeks, followed by peg-IFN + RBV during 12 weeks; T12PR48, TEL + peg-IFN+RBV during 12 weeks, followed by peg-IFN + RBV during 36 weeks; T24PR48, TEL + peg-IFN + RBV during 24 weeks, followed by peg-IFN + RBV during 24 weeks; T24P48, TEL + peg-IFN during 24 weeks; B24PR28/48, peg-IFN+RBV during 4 weeks, followed by boceprevir (BOC) + peg-IFN + RBV during 24 weeks, followed by peg-IFN + RBV from 0 to 20 weeks; B44PR48, peg-IFN + RBV during 4 weeks, followed by BOC + peg-IFN + RBV during 44 weeks; B44PR48, peg-IFN + RBV during 4 weeks, followed by BOC + peg-IFN + RBV from 44 to 48 weeks.

Telaprevir
Kumada et al.[34]Arm 1 (Experimental: T12PR24)1269273·0
Arm 2 (Comparator: PR48)633149·2
Jacobson et al.[35]Arm 1 (Experimental 1: T12PR24 or T12PR48)36327174·7
Arm 2 (Experimental 2: T8PR24 or T8PR48)a36425068·7
Arm 3 (Comparator: PR48)36115843·8
Hezode et al.[36]Arm 1 (Experimental 1: T12PR24)815669·1
Arm 2 (Experimental 2: T12PR12)a824959·8
Arm 3 (Experimental 3: T12P12)a782835·9
Arm 4 (Comparator: PR48)823846·3
McHutchison et al.[37]Arm 1 (Experimental 1: T12PR24)794860·8
Arm 2 (Experimental 2: T12PR48)795367·1
Arm 3 (Experimental 3: T12PR12)a17635·3
Arm 4 (Comparator: PR48)753141·3
Zeuzem et al.[38]Arm 1 (Experimental 1: T12PR48)26617164·3
Arm 2 (Experimental 2: Lead-in T12PR48)26417566·3
Arm 3 (Comparator: PR48)1322216·7
McHutchison et al.[39]Arm 1 (Experimental 1: T12PR24)1155951·3
Arm 2 (Experimental 2: T24PR48)a1136053·1
Arm 3 (Experimental 3: T24P24)a1112724·3
Arm 4 (Comparator: PR48)1141614·0
Boceprevir
Poordad et al.[40]Arm 1 (Experimental 1: B24PR28 or B24PR48)36823363·3
Arm 2 (Experimental 2: B44PR48)36624266·1
Arm 3 (Comparator: PR48)36313737·7
Kwo et al.[41]Arm 1 (Experimental 1: B24PR28)1035856·3
Arm 2 (Experimental 2: B44PR48)1037774·8
Arm 3 (Experimental 3: B28PR28)1075854·2
Arm 4 (Experimental 4: B48PR48)1036967·0
Arm 5 (Comparator: PR48)1043937·5
Bacon et al.[42]Arm 1 (Experimental 1: B32PR36 or B32PR48)1629558·6
Arm 2 (Experimental 2: B44PR48)16110766·5
Arm 3 (Comparator: PR48)801721·3
Flamm et al.[43]Arm 1 (Experimental: B44PR48)1348664·2
Arm 2 (Comparator: PR48)671420·9

Figure 2 summarizes the effects of triple therapy on SVR when compared with dual therapy. Overall, triple therapy, regardless of PI type and patients' treatment experience, was significantly associated with a higher achievement of SVR than dual therapy. The probability of achieving SVR with triple therapy compared to dual therapy was statistically significantly higher in 4 groups based on PI type and patients' treatment experience: (i) telaprevir-based triple therapy in treatment-naïve patients (RR = 1·62; 95% CI, 1·47–1·78; Q = 1·68, d.f. = 4, = 0·79; I2 = 0%), (ii) telaprevir-based triple therapy in treatment-experienced patients (RR = 3·85; 95% CI, 3·03–4·90; Q = 0·07, d.f. = 2, = 0·97; I= 0%), (iii) boceprevir-based triple therapy in treatment-naïve patients (RR = 1·70; 95% CI, 1·56–1·86; Q = 3·36, d.f. = 5, = 0·64; I2 = 0%) and (iv) boceprevir-based triple therapy in treatment-experienced patients (RR = 2·98; 95% CI, 2·29–3·87; Q = 0·18, d.f. = 2, = 0·92; I2 = 0%). Because one RCT that included an analysis for telaprevir-based triple therapy in the treatment-naïve patients' group was classified as a low-quality study, a sensitivity analysis was conducted for this group. When excluding this study, the RR was 1·64 (95% CI, 1·48–1·82; Q = 1·25, d.f. = 3, = 0·74; I2 = 0%). In these four groups, no statistically significant heterogeneity was found.

image

Figure 2. Relative risk [95% confidence intervals (CI)] for sustained virologic response (SVR) for triple therapy vs. dual therapy in patients with hepatitis C type 1.

Download figure to PowerPoint

Figs 3 and 4 show the rates of patients who achieved SVR and relative risks for SVR by subgroup analyses according to patients' race. Data on racial differences were retrieved from 5 of the 10 RCTs: 2 RCTs on telaprevir-based triple therapy in treatment-naïve patients, 1 RCT of telaprevir-based triple therapy in treatment-experience patients and 2 RCTs of boceprevir-based triple therapy in treatment-naïve patients. When compared with dual therapy, both black and non-black patients demonstrated the higher rates of achieving SVR with triple therapy: (i) telaprevir-based triple therapy in treatment-naïve patients (black: RR = 2·61; 95% CI, 1·34–5·80 and non-black: RR = 1·59; 95% CI, 1·42–1·79), (ii) telaprevir-based triple therapy in treatment-experienced patients (black: RR = 2·22; 95% CI, 0·24–20·57 and non-black: RR = 3·73; 95% CI, 2·26–6·15) and (iii) boceprevir-based triple therapy in treatment-naïve patients (black: RR = 2·29; 95% CI, 1·61–3·25 and non-black: RR = 1·65; 95% CI, 1·51–1·80). However, when dual therapy was administered, the rates of patients who achieved SVR were lower among black patients than among non-black patients. Specifically, the range of SVR rates was from 10·0 to 25·0% among black patients and from 40·2 to 46·2% among non-black patients. A similar trend in results also was seen with triple therapy. The ranges of patients who achieved SVR were (i) 22·2–61·5% among black patients and (ii) 53·8–78·4% among non-black patients.

image

Figure 3. Rates of sustained virologic response (SVR) and relative risk [95% confidence intervals (CI)] for SVR for telaprevir-based triple therapy vs. dual therapy in patients with hepatitis C type 1 by subgroup analyses of race.

Download figure to PowerPoint

image

Figure 4. Rates of sustained virologic response (SVR) and relative risk [95% confidence intervals (CI)] for SVR for boceprevir-based triple therapy vs. dual therapy in patients with hepatitis C type 1 by subgroup analyses of race.

Download figure to PowerPoint

Safety

Discontinuation owing to adverse events

Data on discontinuation due to adverse events were retrieved from 8 of the 10 trials. In treatment-experienced patients, telaprevir-based triple therapy was associated with a higher probability of discontinuation attributable to adverse events when compared with dual therapy in treatment-experienced patients (RR = 3·38; 95% CI, 1·88–6·09). However, there was no statistical difference between telaprevir-based triple therapy and dual therapy in treatment-naïve patients (RR = 1·33; 95% CI, 0·86–2·05). Regarding boceprevir-based triple therapy, the probabilities of discontinuation in both treatment-naïve patients (RR = 1·74; 95% CI, 1·15–2·64) and treatment-experienced patients (RR = 3·94; 95% CI, 1·82–8·49) were significantly higher than those in dual therapy (Table 4).

Table 4. Pooled relative risk estimates [95% confidence intervals (CI)] for discontinuation owing to adverse events and serious adverse events for triple therapy vs. dual therapy in patients with hepatitis C type 1
 No. of studies (Comparisons)aNo. of participantsI2 (%)Overall effectRelative risk (95% CI)
  1. a

    Number of comparisons can exceed the number of studies because several studies had multiple comparisons.

Discontinuation owing to adverse events
Telaprevir
Treatment naïve3 (4)107642Z = 1·28 (= 0·200)1·33 (0·86, 2·05)
Treatment experienced2 (3)10230Z = 4·06 (< 0·001)3·38 (1·88, 6·09)
Boceprevir
Treatment naïve1 (4)8320Z = 2·06 (= 0·009)1·74 (1·15, 2·64)
Treatment experienced2 (3)6840Z = 3·49 (< 0·001)3·94 (1·82, 8·49)
Serious adverse events
Telaprevir
Treatment naïve4 (5)13840Z = 1·56 (= 0·120)1·31 (0·93, 1·84)
Treatment experienced2 (3)10230Z = 5·86 (< 0·001)3·00 (2·08, 4·33)
Boceprevir
Treatment naïve2 (6)22920Z = 1·88 (= 0·061)1·28 (0·99, 1·66)
Treatment experienced2 (3)6840Z = 2·12 (= 0·034)1·81 (1·05, 3·14)
Serious adverse events

The data of nine trials were retrieved to determine the probability of serious adverse events. When triple therapies of either PI were employed in treatment-naïve patients, although higher probabilities of serious adverse events were detected, the differences between triple and dual therapies were not statistically significant: (i) telaprevir-based triple therapy (RR = 1·31; 95% CI, 0·93–1·84) and (ii) boceprevir-based triple therapy (RR = 1·28; 95% CI, 0·99–1·66). However, in treatment-experienced patients, triple therapy with either PI had increased probabilities of serious adverse events: (i) telaprevir-based triple therapy (RR = 3·00; 95% CI, 2·08–4·33) and (ii) boceprevir-based triple therapy (RR = 1·81; 95% CI, 1·05–3·14) (Table 4).

Specific adverse events

Table 5 shows pooled RR estimates for the principal adverse events associated with telaprevir or boceprevir treatments. A total of 10 trials were included for data analyses of each adverse event. Regardless of PI type and patients' treatment experience, regimens of triple therapy were associated with increased probabilities of anaemia in comparison with dual therapy. The probability of incidences of rash and pruritus was higher in telaprevir-based triple therapy than in dual therapy in both treatment-naïve patients (RR, 1·51; 95% CI, 1·31–1·74 and RR, 1·47; 95% CI, 1·20–1·81, respectively) and treatment-experienced patients (RR, 2·01; 95% CI, 1·59–2·55 and RR, 1·88; 95% CI, 1·55–2·29, respectively). Higher probabilities of rash were observed in both boceprevir-based triple therapies regardless of patients' treatment experience: treatment-naïve patients (RR, 1·22; 95% CI, 1·03–1·44) and treatment-experienced patients (RR, 3·05; 95% CI, 1·74–5·36). However, there were no statistically significant differences in the incidence of pruritus between boceprevir-based triple therapy and dual therapy. Regarding nausea, a similar trend in results was seen as with pruritus. The probability of nausea was higher in only telaprevir-based triple therapy compared to dual therapy in treatment-naïve patients. Overall, triple therapy was related to an increased probability of dysgeusia except with treatment-experienced patients on telaprevir-based regimens.

Table 5. Pooled relative risk estimates (95% confidence intervals) for adverse events for triple therapy vs. dual therapy in patients with hepatitis C type 1
 No. of Study (Comparison)aNo. of ParticipantsAnaemiaRashPruritusNauseaDiarrhoeaDysgeusia
  1. a

    Number of comparisons can exceed the number of studies because several studies had multiple comparisons.

Telaprevir
Treatment naïve4 (5)13841·44 (1·12, 1·85)1·51 (1·31, 1·74)1·47 (1·20, 1·81)1·48 (1·26, 1·75)1·18 (0·99, 1·40)1·60 (1·11, 2·32)
Treatment experienced2 (3)10232·40 (1·81, 3·17)2·01 (1·59, 2·55)1·88 (1·55, 2·29)1·30 (1·04, 1·63)1·81 (1·37, 2·36)1·51 (0·78, 2·93)
Boceprevir
Treatment naïve2 (6)22921·65 (1·49, 1·83)1·22 (1·03, 1·44)0·99 (0·85,1·15)1·07 (0·97, 1·17)1·15 (0·99, 1·33)2·41 (2·06, 2·83)
Treatment experienced2 (3)6841·90 (1·45, 2·48)3·05 (1·74, 5·36)1·10 (0·77, 1·58)1·21 (0·98, 1·49)1·77 (1·14, 2·76)3·37 (2·34, 4·84)

Numbers needed to treat/harm

Table 6 summarizes the number of patients needed to treat for SVR and the number of patients needed to harm for discontinuation owing to adverse events and serious adverse events with triple therapy compared to dual therapy. Regardless of PI type, treatment-naïve patients were more likely to have higher NNTs than treatment-experienced patients. In treatment-naïve patients, NNTs were 4 (95% CI, 4, 5) in both telaprevir-based triple therapy and boceprevir-based triple therapy. In treatment-experienced patients, NNTs were 3 in telaprevir-based triple therapy (95% CI, 2, 3) and boceprevir-based triple therapy (95% CI, 3, 3). Both NNHs for discontinuation owing to adverse events and serious adverse events in treatment-naïve patients were more likely to be higher than those in treatment-experienced patients.

Table 6. Number needed to treat (NNT) or number needed to harm (NNH) (95% confidence intervals (CI)) for sustained virologic response (SVR), discontinuation owing to adverse events and serious adverse events for triple therapy vs. dual therapy in patients with hepatitis C type 1
 NNT (95% CI)NNH (95% CI)
SVRDiscontinuation owing to adverse eventsSerious adverse events
Telaprevir
Treatment naïve4 (4, 5)25 (11, ∞)38 (18, ∞)
Treatment experienced3 (2, 3)11 (8, 19)6 (3, 111)
Boceprevir
Treatment naïve4 (4, 5)18 (10, 125)47 (22, ∞)
Treatment experienced3 (3, 3)11 (7, 22)16 (9, 58)

Publication bias

Taking into consideration that SVR is the main outcome variable, neither the Egger's weighted regression statistic (= 0·340) nor inspection of funnel plots for asymmetry revealed any indication of publication bias.

Discussion

  1. Top of page
  2. Summary
  3. What is known and objective
  4. Methods
  5. Results
  6. Discussion
  7. What is new and conclusions
  8. Acknowledgements
  9. References

The combination of peg-IFN-alfa and ribavirin was the SOC in the treatment for hepatitis C for many years. The PIs, telaprevir and boceprevir, are a new therapeutic class of oral medications for the management of this disease. With the FDA approval of telaprevir and boceprevir in 2010, we can achieve greater treatment success than before, and the new SOC has been changed from dual to triple therapy.[20]

The efficacy results of this meta-analysis demonstrated that triple therapies including telaprevir or boceprevir, peg-interferon and ribavirin were more likely to achieve an SVR than dual therapies including peg-interferon and ribavirin. These findings from the present analyses were consistent with those of Cooper et al. and Dang et al., which reported that patients provided a standard-dose duration or response-guided therapy duration of triple therapy are more likely to achieve an SVR, when compared with those treated with dual therapy.[25, 27] Additionally, Cooper et al.[28] have concluded that there was no difference in achieving SVR according to the PI type. When compared to dual therapy, NNTs for SVR with triple therapy were lower in treatment-experienced patients than in treatment-naïve patients.

One of the main strengths of this meta-analysis is that this study benefits from the lack of heterogeneity. All analyses in this meta-analysis were based on the four groups according to PI type and patients' treatment experience. Given the results of the Q statistics and the I2 values in the four groups, no statistically significant heterogeneity was found when considering SVR as the main outcome variable. This provides assurance that each individual study included in each group had similar characteristics, making it acceptable to combine data into the same group.[44]

Another main strength of this study was the subgroup analyses according to patients' race. To our knowledge, this is the first meta-analysis that has compared SVR rate between black and non-black patients. These results can provide additional information regarding the use of telaprevir and boceprevir for patients with hepatitis C. Black patients with chronic hepatitis C have been considered to be a difficult-to-treat population in that they have a lower rate of SVR when receiving treatment with dual therapy than does the white population.[30, 45] When triple therapy was administered, the rate of SVR was substantially increased among black patients (although it was still lower among black patients than among non-black patients). Among black patients, only 10·0–25·0% of patients achieved SVR with dual therapy, but 22·2–61·5% of patients achieved SVR with triple therapy. Based on our pooled estimate, the probabilities of SVR with triple therapies including either telaprevir or boceprevir were approximately two to four times higher than those with dual therapies among black patients. Thus, this improvement of treatment effect should be considered in the treatment for chronic hepatitis C.

Consistent with our results, Cooper et al. claimed that triple therapies of both PIs were associated with an increased discontinuation of treatment when compared with dual therapy,[46] and Dang et al. have linked telaprevir with significantly higher incidences of serious adverse events and with discontinuation due to adverse events.[25] Our results on safety demonstrated that triple therapies were more likely than dual therapies to have higher incidences of discontinuation of therapy due to adverse events. In addition, we found that patients using triple therapies were more likely to suffer from serious adverse events, especially in treatment-experienced patients, who were more likely to have lower NNHs than in treatment-naïve patients regardless of PI type.

This meta-analysis estimated various pooled effect sizes regarding specific adverse events. According to American Association for the Study of Liver Disease (AASLD) guidelines, rash, anaemia, pruritus, nausea and diarrhoea were more commonly associated with the administration of telaprevir, whereas anaemia and dysgeusia were the most common adverse events associated with the use of boceprevir.[20] Accordingly, this meta-analysis estimated pooled RRs for these six adverse events (anaemia, rash, pruritus, nausea, diarrhoea and dysgeusia). First, anaemia was more commonly found with triple therapy including either PI than with dual therapy in both treatment-naïve and treatment-experienced patients. The guidelines and previous meta-analysis found similar results.[18, 27] Second, we found the significant increased probabilities of incidences of rash and pruritus in telaprevir-based triple therapy compared to dual therapy regardless of the patients' treatment experience. However, when administering boceprevir-based triple therapies, the probability of incidences of rash was significantly increased compared to dual therapy, but the probabilities of incidences of pruritus were not different from dual therapy regardless of patients' treatment experience. Third, regarding gastrointestinal disorders, significant increases in the probabilities of nausea and diarrhoea were associated with only telaprevir-based triple therapies when compared to dual therapy. However, these probabilities were not significantly increased in boceprevir-based triple therapy. These results were consistent with the findings in the study by Ghany et al.,[20] which demonstrated that only telaprevir was associated with nausea and diarrhoea. Finally, the incidences of dysgeusia were more common in boceprevir-based triple therapy than in dual therapy for both treatment-naïve and treatment-experienced patients. Several previous studies reported that boceprevir was significantly associated with a higher incidence of dysgeusia.[18, 20, 27] In addition, this study found that the incidence of dysgeusia was significantly higher for only treatment-naïve patients when administering telaprevir-based triple therapy.

This study has several limitations. First, multiple treatment regimens (triple therapy regimens) were compared to the same control regimen (dual therapy) in 5 of 10 RCTs. Thus, the dual therapy control group was double-counted in the analyses. In this case, the effect measures might be dependent on the use of a same control group, which means that these RCTs were overweighted in the analysis because these trials were counted as multiple trials. Future meta-analyses might be considered to overcome this limitation, by restricting the analysis to a single triple therapy regimen from each trial. Second, this meta-analysis includes only six RCTs for telaprevir and four RCTs for boceprevir because few RCTs have been conducted with these two drugs. Third, several meaningful subgroup analyses according to subtype of hepatitis C virus genotype 1, the presence of cirrhosis and hepatitis C viral load were not conducted because of limited information from the included RCTs. Fourth, generalizability to typical patient care settings is limited because the meta-analysis included only RCTs; observational studies were excluded. Finally, another limitation pertains to the main efficacy outcome. Although the SVR rate is an important outcome used in assessing the treatment effect in chronic hepatitis C, it is not a long-term outcome, such as liver-related mortality and liver transplantation. No RCT to date has assessed long-term outcomes.

What is new and conclusions

  1. Top of page
  2. Summary
  3. What is known and objective
  4. Methods
  5. Results
  6. Discussion
  7. What is new and conclusions
  8. Acknowledgements
  9. References

This meta-analysis compared the efficacy and safety outcomes of telaprevir and boceprevir in triple therapies to those of dual therapy in patients with chronic hepatitis C genotype 1. To the best of our knowledge, this is the first meta-analysis to report race differences in achieving the efficacy outcome. In conclusion, this meta-analysis suggests that triple therapies including either telaprevir or boceprevir are superior to dual therapy for both treatment-naïve patients and treatment-experienced patients in achieving better efficacy outcomes. However, triple therapies also have higher incidence rates of discontinuation due to adverse events, serious adverse events and other specific adverse events, such as anaemia and rash. In future, this new class of HCV agents will need continued evaluation for long-term efficacy and safety.

References

  1. Top of page
  2. Summary
  3. What is known and objective
  4. Methods
  5. Results
  6. Discussion
  7. What is new and conclusions
  8. Acknowledgements
  9. References