SEARCH

SEARCH BY CITATION

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Background

Recent studies suggested that IL28B polymorphisms may affect rapid and sustained virological response rates in HCV patients infected with genotype 2 or 3.

Aim

To assess the role of IL28B polymorphisms on the virological response in HCV-2 and -3 patients.

Methods

We performed meta-analysis of studies evaluating the impact of rs12979860 and rs8099917 polymorphisms on rapid and sustained virological response in HCV-2 or -3 patients.

Results

Twenty-three studies involving 3042 patients were included. The first meta-analysis evaluated the impact of rs12979860 polymorphism and included 1963 patients. When compared with rs12979860 CT/TT patients, CC patients had a higher rapid virological response rate (mean difference: 12.9%, 95% CI: 6.5–19.4%, P < 0.001) and a higher sustained virological response rate (mean difference: 4.9%, 95% CI: 0.1–9.8%, P = 0.046). The second meta-analysis evaluated the impact of rs8099917 polymorphism and included 2246 patients. When compared with rs8099917 TG/GG patients, TT patients had a higher rapid virological response rate (mean difference: 14.8%, 95% CI: 7.2–22.4%, P < 0.001) and a higher sustained virological response rate (mean difference: 5.5%, 95% CI: 0.4–10.6%, P = 0.033). When considering only patients treated for 24 weeks, results were unchanged. No potential sources of between-study heterogeneity were identified.

Conclusions

Favourable IL28B polymorphisms are associated with higher rapid and sustained virological response rates in HCV-2 and -3 patients. However, as the impact on a sustained response is very limited, it is unlikely that IL28B polymorphisms provide additional predictive value when considering other predictors of a sustained response.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

In patients with chronic hepatitis C virus (HCV) infection, response rates to interferon-based therapy are heterogeneous and HCV genotypes have been concordantly shown to be important baseline variables influencing treatment outcome. HCV-2 and -3 are considered ‘easy to treat’ genotypes, as a sustained virological response (SVR) is observed in 70–80% of patients with 24-week pegylated interferon (Peg-IFN) and ribavirin combination therapy.[1] Recently, telaprevir and boceprevir, two protease inhibitors, were approved for HCV-1-infected patients in combination with Peg-IFN and ribavirin and led to substantial improvement in the SVR rate.[2, 3] However, for non-1 genotypes, Peg-IFN and ribavirin remain the standard therapy. Four independent genome-wide association studies (GWAS) identified several single nucleotide polymorphisms (SNPs) near the IL28B gene (encoding IFN-λ3) as being strongly associated with spontaneous and treatment-induced clearance of HCV infection.[4-7] In HCV-1 patients, favourable IL28B variations were strongly associated with on-treatment viral kinetics and approximately twofold increased SVR rates, and IL28B variations were shown to be a strong pre-treatment predictor of virological response.[8] The influence of IL28B variations on treatment outcome was less convincing in HCV-2 and -3 patients; indeed, studies of IL28B in HCV-2 and -3 included only small numbers of patients and produced conflicting data.[9-15]

Meta-analysis is a quantitative technique for therapeutic evaluation when controversy persists despite reports from several clinical trials.[16] It is particularly useful when trials have insufficient statistical power, as pooling of trial subjects decreases random errors. In this study, we performed a meta-analysis of trials evaluating the influence of IL28B variations on rapid virological response (RVR) and SVR rates in HCV-2 and -3 patients.

Methods

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Literature search

Medline, Cancerlit, Embase and manual searches were combined.[17] Data abstraction was done independently by two investigators (J.S. and B.G.G.) using standardised data collection forms. Discrepancies in data interpretation were resolved by discussion, re-review of the studies and consultation with two other authors (C.M. and P.D.) when necessary. Search terms were hepatitis C, treatment, IL28B polymorphisms, IFN-λ3, RVR and SVR. General reviews and references from published trials were used. The two observers also extensively screened all abstracts presented in English at liver and gastroenterology congresses over the last 3 years.

Criteria for inclusion and exclusion of studies

Inclusion and exclusion criteria were defined prior to commencement of the literature search. For inclusion, a study had to (i) include patients with chronic HCV-2 or -3 infection treated with IFN and ribavirin or Peg-IFN and ribavirin combination therapy; (ii) compare virological response rates in patients with favourable IL28B polymorphisms for SVR, defined as either the rs12979860 CC genotype or the rs8099917 TT genotype, and in patients without favourable IL28B polymorphisms for SVR defined as either the rs12979860 CT/TT genotype or the rs8099917 TG/GG genotype; (iii) have been published as a full-length article or presented as an abstract at an international congress using English as the official language (AASLD Annual Meeting, EASL Annual Meeting, DDW Annual Meeting) in 2009, 2010 and 2011.

We excluded (i) studies without information on RVR or SVR rates or IL28B polymorphisms, not enabling comparison of RVR or SVR rates between patients with and without favourable IL28B polymorphisms; (ii) studies on HIV-co-infected patients; (iii) studies on liver transplant recipients; and (iv) studies on paediatric patients. When several publications existed concerning the same study population, only the most recent was taken into account.

Endpoints and criteria for combinability

Endpoints were defined prior to the beginning of meta-analysis. Endpoints were RVR and SVR. As RVR and SVR rates were studied according to two different SNPs (rs12979860 and rs8099917), we performed two meta-analyses to assess the impact of each polymorphism on virological response. Each meta-analysis included studies evaluating the efficacy of anti-viral therapy in patients with and without favourable IL28B polymorphisms. In a first step, overall meta-analysis was performed. This analysis combined studies that included patients treated for less than 24 weeks. In a second step, sensitivity analyses restricted to studies including patients treated for 24 weeks were performed. Overall analysis also combined studies that included patients infected with genotypes 2 and 3. In a third step, sensitivity analyses that separately considered patients infected with genotype 2 and those infected with genotype 3 were performed. Sensitivity analyses were performed when more than three studies could be analysed.

Statistical methods

We used a random effects model to obtain a summary estimate of RVR and SVR rates between patients with and without favourable IL28B polymorphisms. Peto[18] and DerSimonian and Laird methods[19] were used. When virological responses were given as percentages, the number of patients was extrapolated.

In the DerSimonian and Laird method, studies are considered a random sample from a group of studies. The random effect model incorporates the heterogeneity of the studies. The overall treatment effect is estimated by a weighted average of individual effects, with weights inversely proportional to the variance of the observed effects. The estimate is expressed by the difference in percentages (risk of treated/risk of controls). A significance level of 5% was taken as the alpha risk. Comparison of percentages between groups was performed using their 95% confidence interval. Our assumption of heterogeneity was tested for each planned analysis using the DerSimonian and Laird method based on work first presented by Cochran.[20] To assess the extent of publication bias, the Egger test was used.[21] Detailed results are given only for DerSimonian and Laird analysis. All procedures and calculations used in meta-analyses were made following methodology reported elsewhere.[22-24]

All results in RVR and SVR rates present the mean difference as being the effect (proportion having a positive outcome) on patients with favourable IL28B polymorphisms, minus the effect (proportion having a positive outcome) on patients without favourable IL28B polymorphisms. All statistical analyses were performed using WeasyMA 2002 (ClinInfo SA and EA643, Faculté RTH Laennec, Lyon, France).

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Among 98 references identified, 75 studies were excluded for the following reasons: 45 did not provide information on RVR or SVR rates and IL28B polymorphisms, therefore not permitting comparison of RVR or SVR rates between patients with and without favourable IL28B polymorphisms, 18 were performed in HIV-co-infected patients, 10 were performed in liver transplant recipients and 2 were performed in paediatric patients (see supplementary Table S1, showing excluded studies and reasons for exclusion, published online).

Twenty-three studies involving 3042 patients were included in the analysis[7, 9, 11-15, 25-40] (Table 1). Six studies evaluated virological response according to rs12979860 polymorphism,[9, 26, 29, 32, 35, 36] seven according to rs8099917 polymorphism[7, 11, 15, 34, 37-39] and 10 according to both polymorphisms.[12-14, 25, 27, 28, 30, 31, 33, 40] The main characteristics of these studies are given in Table 1. In 22 studies, patients were treated with Peg-IFN-α2a or -α2b and ribavirin. The final study included patients treated with Peg-IFN and ribavirin combination therapy and patients treated with IFN and ribavirin combination therapy.[12]

Table 1. Characteristics of the 23 studies included
ReferenceRaceHCV genotypeNo. of patients included in meta-analysisNo. of patients with rs12979860 CC genotypeNo. of patients with rs12979860 CT/TT genotypeNo. of RVR among patients with rs12979860 CC genotypeNo. of RVR among patients with rs12979860 CT/TT genotypeNo. of SVR among patients with rs12979860 CC genotypeNo. of SVR among patients with rs12979860 CT/TT genotypeNo. of patients with rs8099917 TT genotypeNo. of patients with rs8099917 TG/GG genotypeNo. of RVR among patients with rs8099917 TT genotypeNo. of RVR among patients with rs8099917 TG/GG genotypeNo. of SVR among patients with rs8099917 TT genotypeNo. of SVR among patients with rs8099917 TG/GG genotype
  1. Afr, African; Am, American; Eur, European; G, genotype; NA, not assessed; Peg-IFN, pegylated interferon; RVR, rapid virological response; SVR, sustained virological response.

  2. a

     Of the 66 patients enrolled in the study, 47 patients who received Peg-IFN and ribavirin were included in the analysis.

  3. b

     This study also included non-G2/G3 patients; only G2 and G3 patients were considered in the analysis.

  4. c

     Of the 72 patients enrolled in the study, 71 patients with RVR and SVR data were included in the analysis; in this study, patients were randomised to different treatment regimens after 6 weeks of Peg-IFN and ribavirin combination therapy including one Peg-IFN monotherapy arm; other patients received 24–72 weeks of Peg-IFN and ribavirin combination therapy.

  5. d

     Of the 56 patients enrolled in the study, SVR data were available in 15 rs12979860 CC patients and 40 rs12979860 CT/TT patients.

  6. e

     Of the 109 patients enrolled in the study, 7 G2 patients and 21 G3 patients with SVR data were included in the analysis.

  7. f

     SVR data related to the rs12979860 polymorphism could not be included in the analysis because of discordances between results provided in Table 2 and those provided in Supplementary Table S4 of the cited article.

  8. g

     Of the 74 patients enrolled in the study, 15 patients with available data for the SNP rs8099917 were included in the analysis.

  9. h

     Of the patients enrolled in the study, 18 patients with SVR data were included in the analysis.

  10. i

     This study included 719 patients among whom 559 received IFN monotherapy; among the 160 participants treated with Peg-IFN and ribavirin and enrolled in the study, 130 patients with available SVR data according to rs8099917 polymorphism were included in the analysis.

  11. j

     175 patients received a 12-week course of treatment.

  12. k

     RVR data were available in 148 rs12979860 CC patients and 187 rs12979860 CT/TT patients.

  13. l

     122 RVR patients received a 12-week course of treatment.

  14. m

     This study included patients from two previous trials; in one of them, all RVR patients received a 14-week course of treatment; in the other, RVR patients were randomised to a 14-week or a 24-week course of treatment.

  15. n

     Of the 2112 patients enrolled in the study, only 29 G2 patients treated with Peg-IFN and ribavirin were included in the analysis.

  16. o

     This study enrolled 1362 patients, some of whom were co-infected with human immunodeficiency virus; only 230 mono-infected G2 and G3 patients treated with Peg-IFN and ribavirin were included in the analysis.

  17. p

     RVR data were available in 88 rs8099917 TT patients and in 24 rs8099917 TG/GG patients.

  18. q

     This study also included patients treated with IFN and ribavirin combination therapy.

  19. r

     Data extrapolated from the whole cohort of 267 enrolled patients.

  20. s

     This study enrolled 35 patients who received only 400 mg of ribavirin per day; of the 72 patients enrolled in the study, rs12979860 and rs8099917 polymorphisms were available for 71 patients.

  21. t

     SVR data were available in 25 rs12979860 CC patients and 44 rs12979860 CT/TT patients.

  22. u

     SVR data were available in 39 rs8099917 TT patients and 30 rs8099917 TG/GG patients.

  23. v

     This study enrolled patients who received only 400 mg of ribavirin per day.

  24. w

     RVR data were available in 81 rs12979860 CC patients and 114 rs12979860 CT/TT patients.

  25. x

     RVR data were available in 112 rs8099917 TT patients and 83 rs8099917 TG/GG patients.

Amanzada et al.[25]UnreportedG347a2423NANA20193611NANA2910
Bitetto et al.[26]CaucasianG2/G3b1014457NANA3651NANANANANANA
Bochud et al.[27]CaucasianG2/G3b71c452640194022521946144616
Cavalcante et al.[28]Afr/Eur/AmG2/G3b5615d40dNANA11163620NANA226
de Rueda et al.[29]CaucasianG2/G3b502525NANA2120NANANANANANA
Dill et al.[30]UnreportedG2b7e16NANA0434NANA13
UnreportedG3b21e813NANA88138NANA115
Fattovich et al.[31]CaucasianG2b104396537583759545049465046
CaucasianG3b5528272420NAfNAf391633113414
Giannini et al.[32]UnreportedG2b18513NANA411NANANANANANA
Halfon et al.[33]CaucasianG2/G3b421527NANA12193012NANA229
Itoh et al.[34]AsianG215gNANANANANANA132NANA100
Jung et al.[35]AsianG2b18h171NANA161NANANANANANA
Kawaoka et al.[11]AsianG2130iNANANANANANA10624NANA8316
Lindh et al.[36]CaucasianG2/G3341j15019199k108k105132NANANANANANA
Mangia et al.[9]CaucasianG2/G3268l1001685910682119NANANANANANA
Moghaddam et al.[13]UnreportedG3281m1291521089399127201801564516165
Ochi et al.[37]AsianG2b29nNANANANANANA263NANA241
Rauch et al.[7]CaucasianG2/G3b230oNANANANANANA14783NANA12766
Sakamoto et al.[38]AsianG2129NANANANANANA1002958p10p8117
Sarrazin et al.[12]CaucasianG2b58q27317687r2625130r75rNANA104r56r
CaucasianG3b147q6087  5059      
Scherzer et al.[14]CaucasianG371s2744212219t34t4031NANA29u24u
Sinn et al.[39]AsianG2b63NANANANANANA549497439
Stättermayer et al.[40]UnreportedG2/G3b208v8712161w60w70901189081x40x9565
Yu et al.[15]AsianG2482NANANANANANA432503683638643

Meta-analysis of studies evaluating RVR and SVR rates in HCV-2 and -3 patients with the rs12979860 CC or CT/TT genotype

Overall analysis

Sixteen studies involving 1963 patients were included in the analysis[9, 12-14, 25-33, 35, 36, 40] (Table 1). A total of 846 patients had the rs12979860 CC genotype (43% of the population studied, ranging from 27 to 94% among studies), whereas 1117 patients had the CT/TT genotype. When compared with CT/TT patients, CC patients had a higher RVR rate (mean difference: 12.9%, 95% CI: 6.5–19.4%, P < 0.001) (Figure 1a) and a higher SVR rate (mean difference: 4.9%, 95% CI: 0.1–9.8%, P = 0.046) (Figure 1b).

image

Figure 1. Overall analysis of studies evaluating virological response rates in HCV-2 and -3 patients with the rs12979860 CC or CT/TT genotype. (a) Overall analysis of studies evaluating RVR rates. (b) Overall analysis of studies evaluating SVR rates.

Download figure to PowerPoint

Using Cochran's Q test, no statistical heterogeneity was found between the included studies in terms of clinical outcome, as indicated in each figure. The funnel plot of trials included in overall analysis is given in Figure S1. The Egger test did not identify significant asymmetry (P = 0.2). However, that test in itself could not formally exclude the presence of a publication bias.

Significant results using the DerSimonian method were also significant using the Peto method (data not shown). Significant results using the random effects model were also significant using the fixed effects model (data not shown).

Sensitivity analyses restricted to studies including patients treated for 24 weeks

Thirteen studies involving 1073 patients were included in the analysis[12, 14, 25-33, 35, 40] (Table 1). A total of 467 patients had the rs12979860 CC genotype (44% of the population studied), whereas 606 patients had the CT/TT genotype. When compared with CT/TT patients, CC patients had a higher RVR rate (mean difference: 14.3%, 95% CI: 8.0–20.6%, P < 0.001) and a higher SVR rate (mean difference: 6.4%, 95% CI: 0.9–12.0%, P = 0.023).

Sensitivity analyses according to genotype (2 or 3)

Five studies involving 205 patients infected with genotype 2 were included in the analysis[12, 30-32, 35] (Table 1). A total of 89 patients had the rs12979860 CC genotype (43% of the population studied), whereas 116 patients had the CT/TT genotype. The SVR rate did not differ between CC and CT/TT patients (mean difference: 6.7%, 95% CI: −7.1 to 20.4%, P = 0.3). Sensitivity analyses for RVR restricted to studies including patients infected with genotype 2 were not feasible, as not enough studies provided specific data.

Five studies involving 565 patients infected with genotype 3 were included in the analysis[12-14, 25, 30] (Table 1). A total of 246 patients had the rs12979860 CC genotype (44% of the population studied), whereas 319 patients had the CT/TT genotype. The SVR rate did not differ between CC and CT/TT patients (mean difference: 5.3%, 95% CI: −7.4 to 18%, P = 0.4). Sensitivity analyses for RVR restricted to studies including patients infected with genotype 3 were not feasible, as not enough studies provided specific data.

Meta-analysis of studies evaluating RVR and SVR rates in HCV-2 and -3 patients with the rs8099917 TT or TG/GG genotype

Overall analysis

Seventeen studies involving 2246 patients were included in the analysis[7, 11-15, 25, 27, 28, 30, 31, 33, 34, 37-40] (Table 1). A total of 1630 patients had the rs8099917 TT genotype (73% of the population studied, ranging from 57 to 90% among studies), whereas 616 patients had the TG/GG genotype. When compared with TG/GG patients, TT patients had a higher RVR rate (mean difference: 14.8% (95% CI: 7.2–22.4%, P < 0.001) (Figure 2a) and a higher SVR rate (mean difference: 5.5%, 95% CI: 0.4–10.6%, p = 0.033) (Figure 2b).

image

Figure 2. Overall analysis of studies evaluating virological response rates in HCV-2 and -3 patients with rs8099917 TT or TG/GG genotype. (a) Overall analysis of studies evaluating RVR rates. (b) Overall analysis of studies evaluating SVR rates.

Download figure to PowerPoint

Using Cochran's Q test, no statistical heterogeneity was found between the included studies in terms of clinical outcome, as indicated in each figure. The funnel plot of trials included in overall analysis is given in Figure S2. The Egger test did not identify significant asymmetry (P = 0.3).

Significant results using the DerSimonian method were also significant using the Peto method (data not shown). Significant results using the random effects model were also significant using the fixed effects model (data not shown).

Sensitivity analyses restricted to studies including patients treated for 24 weeks

Sixteen studies involving 1965 patients were included in the analysis[7, 11, 12, 14, 15, 25, 27, 28, 30, 31, 33, 34, 37-40] (Table 1). A total of 1429 patients had the rs8099917 TT genotype (73% of the population studied), whereas 536 patients had the TG/GG genotype. When compared with TG/GG patients, TT patients had a higher RVR rate, with a mean difference of 13.5% (95% CI: 4.9–22.0%, P = 0.002) and a higher SVR rate (mean difference: 6.3%, 95% CI: 0.8–11.8%, P = 0.024).

Sensitivity analyses according to genotype (2 or 3)

Eight studies involving 959 patients infected with genotype 2 were included in the analysis[11, 15, 30, 31, 34, 37-39] (Table 1). A total of 788 patients had the rs8099917 TT genotype (82% of the population studied), whereas 171 patients had the TG/GG genotype. There was no significant difference in the RVR rate (mean difference: 9.8%, 95% CI: −1.6 to 21.1%, P = 0.09) or the SVR rate (mean difference: 10.1%, 95% CI: −2.1 to 22.3%, P = 0.1) between TT patients and TG/GG patients.

Five studies involving 473 patients infected with genotype 3 were included in the analysis[13, 14, 25, 30, 31] (Table 1). A total of 328 patients had the rs8099917 TT genotype (69% of the population studied), whereas 145 patients had the TG/GG genotype. There was no difference in the SVR rate in TT patients compared with TG/GG patients (mean difference: −1.9%, 95% CI: −9.4% to 5.6%, P = 0.6). Sensitivity analyses for RVR restricted to studies including patients infected with genotype 3 were not feasible, as not enough studies provided specific data.

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

This meta-analysis sought to evaluate the impact of IL28B polymorphisms on the efficacy of anti-viral treatment in HCV patients infected with genotype 2 or 3. It included 23 studies involving 3042 patients. As studies evaluated SVR according to two different polymorphisms, we performed one analysis of virological response rates according to rs12979860 polymorphism and another according to rs8099917 polymorphism. These meta-analyses demonstrated that in patients with favourable IL28B polymorphisms (either rs12979860 CC or rs8099917 TT patients), RVR rates were increased by 13–15% and SVR rates by 5%. Sensitivity analyses restricted to studies including patients treated for 24 weeks provided similar results.

IL28B variations seem to be a weaker pre-treatment predictor of virological response in HCV-2 and -3 patients than in HCV-1 patients. This is supported by comparison of the odds ratio of favourable polymorphisms for SVR in HCV-2 and -3 patients (1.3 and 1.5 for rs12979860 CC and rs8099917 polymorphisms, respectively) and in HCV-1 patients (5.2 for rs12979860 CC polymorphism[8]). Therefore, although a limited impact on SVR exists in patients infected with genotype 2 or 3, the clinical use of the IL28B polymorphism in the individual patient in daily practice seems very limited, and other predictors of SVR such as viral load and fibrosis stage are probably more relevant. Thus, we believe that the IL28B polymorphism will have no impact for the clinician to initiate an anti-viral therapy in those patients.

The mechanism by which IL28B variants influence viral kinetics remains undetermined. The IL28B gene forms a cytokine gene cluster with IL28A and IL29 on human chromosome 19. It encodes for IFN-λ. IFN-λ signalling is initiated through a membrane receptor system complex distinct from that of IFN-α,[41, 42] but has biological activities similar to those of IFN-α/β.[43] Its expression depends on the same stimuli and signal transduction pathways as those involved in IFN-α/β, which leads to induction of several hundred IFN-stimulated genes (ISG).[44] Although it is well known that genotypes 2 and 3 are more responsive to IFN than genotype 1, the mechanisms by which the genotype interferes with the response to anti-viral treatment remain speculative. One hypothesis is that certain viral genotypes (like genotype 1) are inherently more likely to induce ISG preactivation,[44, 45] which prevents further induction with IFN-based therapy and reduces the chances of a virological response.[46, 47] As increased levels of ISG preactivation have been observed in the presence of unfavourable IL28B variants,[48, 49] this hypothesis offers an explanation linking genotype, IL28B polymorphisms, ISG preactivation levels, and the likelihood of a virological response. Once present, ISG preactivation could be the strongest predictor of virological response regardless of the viral genotype or IL28B polymorphism.[44]

This study did not suffer from classical limitations of meta-analyses. None of the potential sources of between-study heterogeneity were identified and the funnel plot of trials disclosed a symmetric pattern, suggesting the absence of publication bias. Unfortunately, we were unable to perform sensitivity analysis according to the presence of other independent predictive factors of SVR such as fibrosis or level of viraemia. Finally, sensitivity analyses that separately considered HCV-2 and HCV-3 patients included a limited number of patients and did not enable drawing firm conclusions.

In conclusion, this meta-analysis indicates that favourable IL28B polymorphisms are associated with higher RVR and SVR rates in patients infected with genotype 2 or 3. However, as the impact on SVR is very limited, it is unlikely that IL28B polymorphisms provide additional predictive value when considering other predictors of SVR.

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Declaration of personal and funding interests: None.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Supporting Information

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information
FilenameFormatSizeDescription
apt5197-sup-0001-FiguresS1.docxWord document20KFigure S1. Funnel plot for SVR in overall analysis in HCV-2 and -3 patients with the rs12979860 CC or CT/TT genotype.
apt5197-sup-0002-FiguresS2.docxWord document14KFigure S2. Funnel plot for SVR in overall analysis in HCV-2 and -3 patients with the rs8099917 TT or TG/GG genotype.
apt5197-sup-0002-TableS1.docWord document53KTable S1. List of excluded references and reasons for exclusion.

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.