SEARCH

SEARCH BY CITATION

Summary

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

Background

There is no satisfactory treatment for patients with hepatitis D (HDV).

Aim

To evaluate treatment for HDV using meta-analysis.

Methods

Medline, Scopus, Cochrane Library and ISI Web of Knowledge searches using the textwords ‘Hepatitis D’, ‘therapy’, “interferon”, “peginterferon”, “pegylated interferon”, “lamivudine”, “pegifn”, “ifn” and “Hepatitis D”, and abstracts from major Gastroenterology/Liver meetings. Endpoints: end of treatment biochemical (biochemical EOT) and virological response (virological EOT), end of follow-up virological response (EOFUP VR), histological improvement and intrahepatic HDAg clearance.

Results

We included randomised clinical trials (RCTs) comparing Group A: interferon-A (IFNa) vs. no treatment (three RCTs, n ;= ;137 patients), Group B: low dose vs. high dose IFNa (two RCTs, n ;= ;60), Group C: IFNa ;+ ;lamivudine vs. IFNa (two RCTs, n ;= ;48) and Group D: pegylated IFNa (PEG-IFNa) vs. other medications (two RCTs, n ;= ;157). Group A. IFNa was better for biochemical EOT [OR, 0.11 (95% CI, 0.04–0.2)] and virological EOT [OR, 0.08 (95% CI, 0.03–0.2)], but not for EOFUP VR. Group B. High dose IFNa was better for biochemical EOT [OR, 0.24 (95% CI,0.08–0.73)] and virological EOT [OR, 0.27 (95% CI, 0.1–0.74)]. Group C. There was a trend favouring histological improvement [OR, 2.9 (95% CI, 0.6–13.4)]. Group D. PEG-IFNa was better for virological EOT [OR, 0.419 (95% CI, 0.18–0.974)], EOFUP VR [OR, 0.404 (95% CI, 0.189–0.866)] and improvement in necroinflammatory activity [OR, 0.308 (95% CI, 0.129–0.732)].

Conclusions

Long-term suppression of HDV RNA by IFNa is not maintained despite an end of treatment response; adding lamivudine is not beneficial. PEG-IFNa is superior to other medications with respect to EOT and EOFUP. New RCTs should test combinations of PEG-IFNa and newest antivirals.


Introduction

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

Hepatitis D is caused by a defective virus, the hepatitis D virus (HDV), which requires the presence of the hepatitis B virus (HBV) to complete its life cycle. Chronic HDV infection runs a more progressive course compared with chronic HBV; however, disease course has a wide range.[1, 2] In a recent trial[3] (n ;= ;299, mean follow-up period: 233 ;months), cirrhosis was diagnosed in 186 (62%) patients with an incidence rate of 4% per year; liver decompensation developed in 54 (18%) patients and HCC in 46 (15%) patients (incidence rate 2.8% per year). The main cause of death was liver failure (59%). Persistent HDV replication and the presence of HBV viremia are associated with the occurrence of liver failure and lower survival rates,[3, 4] whereas HDV genotype seems to determine disease progression.[5, 6] There is currently no satisfactory medical treatment for HDV infection.[7]

The only medication of proven efficacy against chronic HDV infection is IFNa,[8-10] and its efficacy is related to dose.[8, 11] High doses of interferon-A (IFNa) achieve significantly better biochemical and virological responses compared with low doses. Although a biochemical, virological and histological response can be achieved at the end of treatment, virological relapse is extremely common after stopping treatment.[8, 12, 13] Standard IFN has been replaced by its pegylated form (PEG-IFN), which is administered weekly due to its longer half-life. The data regarding the efficacy of PEG-IFN for treating hepatitis D are scanty[10, 14-16]; some report better results compared with standard IFNa.[14-16] Treatment with nucleos(t)ide analogues has also been evaluated as monotherapy or combined with IFNa or PEG-IFNa. In individual trials, famciclovir, lamivudine, adefovir dipivoxil, entecavir, alone or combined with IFNa, have shown no efficacy[17-21] or superiority[22, 23] compared with IFNa. In addition, treatment with ribavirin monotherapy[24, 25] or combined with PEG-IFNa,[14] failed to improve biochemical or virological response.

We systematically evaluated the efficacy of current antiviral treatment for chronic hepatitis D, using a meta-analysis of randomised controlled trials to clarify and compare the effects of different treatment options on the biochemical and virological response, and also the histological improvement at the end of treatment, as well as on the long-term virological response.

Methods

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

Data identification

We searched MEDLINE database, Scopus, Cochrane Library and ISI Web of Knowledge search service from January 1971 to October 2011 using the “Mesh” terms “Hepatitis D” and “therapy”, as well as “interferon” or “peginterferon” or “pegylated interferon” or “lamivudine” and “hepatitis D”, as well as the truncated terms “pegifn” or “ifn” and “hepatitis D”. We manually searched (CT and MK) all review articles (English and non-English), and retrieved original studies and all abstracts from the major Gastroenterology and Liver meetings held between January 1971 and October 2011 of the American Association for the Study of Liver Diseases and the European Association for the Study of the Liver and those from the Digestive Diseases Weeks and the United European Gastroenterology Weeks.

Inclusion criteria

Studies included fulfiled the following criteria: (i) a published article or abstract, (ii) a randomised design irrespective of blinding, (iii) antiviral treatment for at least 12 ;months, follow-up for at least 6 ;months and sufficient data to evaluate treatment response, (iv) at least one of the following clinical end points: end of treatment complete biochemical response (biochemical EOT), end of treatment virological response (virological EOT), end of follow-up virological response (EOFUP VR), histological improvement and intrahepatic HDAg clearance.

Exclusion criteria

Studies with (i) a nonrandomised design, (ii) comparison between different treatment regimens and (iii) lack of data regarding treatment response were excluded from the meta-analysis. The flow diagram of the studies excluded from this analysis is shown in Figure ;1. The decision to include or exclude a study was made before performing the meta-analysis by CT, MK and AB.

image

Figure 1. Flow diagram of the studies considered for meta-analysis.

Download figure to PowerPoint

Data extraction

Data were extracted independently by CT, MK and AB using a predefined review form, and disagreement was resolved by consensus.

Study groups

We classified the included trials in four groups based on the following treatment comparisons: Group A: IFNa for 12 ;months vs. no treatment. Group B: high doses of IFNa (9–18 ;MU) vs. low doses (3 ;MU) for 12 ;months. Group C: IFNa combined with lamivudine vs. IFNa monotherapy for 12 ;months. Group D: PEG-IFNa monotherapy vs. other treatment regimens.

End points for analysis

(i) End of treatment complete biochemical response (biochemical EOT), (ii) end of treatment virological response (virological EOT), (iii) end of follow-up virological response (EOFUP VR), (iv) histological improvement and (v) intrahepatic HDAg clearance. CT, MK and AB independently assigned a methodological quality score based on a five-point quality scale by Kjaergard et ;al.,[26] evaluating the trial's design, conduct, analysis and presentation, minimised or avoided biases in the trial's intervention comparisons.[27] The quality score was ranked as low (≤2 points) or high (≥3 points).

Definitions

Biochemical EOT: normal serum alanine aminotransferase (ALT) at the end of treatment. Virological EOT: negative serum HDV RNA at the end of treatment. EOFUP VR: negative serum HDV RNA at the end of follow-up. Histological improvement: difference of at least two points between the total Knodell scores and of at least one point in fibrosis score between paired biopsies before and at the end of treatment respectively. Available data on improvement of fibrosis were available only for Group D (trials of PEG-IFNa). Intrahepatic HDAg clearance was undetectable HD antigen in hepatic cells at the end of treatment.

Statistical analysis

We used risk odds ratios (fixed models) as our main effect measure to make comparisons across the groups easier to interpret.[28] The Peto method[29] was used to estimate study-specific and common 95% confidence interval, as well as odds ratio. Where statistical significance was present, we calculated the number of patients needing treatment to prevent one outcome event (NNT) and the ‘publication bias assessment’ (minimum number of negative or null studies required to render the meta-analytical results statistically nonsignificant). The ‘publication bias assessment’ was calculated according to Klein formula.[30] We assessed heterogeneity using the Q-test (P ;< ;0.05) evaluating whether treatment effect variations within each trial group was greater than expected, and report it only if significant. All statistical analyses and plots were performed by specific computer software, previously used for meta-analyses, designed by G. Leandro.[31]

Results

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

Descriptive assessment

Through database searching, we identified 12 randomised clinical trials (RCTs) concerning antiviral treatment for hepatitis D. We excluded four[32-34] from the analysis because they did not fulfil the inclusion criteria (different treatment regimens or treatment duration). Gunsar F et ;al.[32] randomised patients in IFNa-2a monotherapy or its combination with ribavirin for 24 ;months. EOFUP VR was achieved in 20% and 23.5% of patients respectively. Porres JC et ;al.[33] compared treatment outcomes between IFNa-2c for 6 ;months and no treatment. IFNa-2c-treated patients had 28.6% EOFUP VR, whereas none of the untreated patients sustained virological response. Borghesio E et ;al.[34] treated patients with lymphoblastoid IFNa in high dose (10 ;MU) or low dose (5 ;MU). No data on EOFUP VR were reported. Rosina F et ;al.[35] randomised patients in IFNa-2b (n ;= ;12) for 3 ;months or no intervention (n ;= ;12). EOFUP VR was observed in 8.3% and none of the patients respectively (Table S1).

We finally included in the analysis eight RCTs comprising 402 patients.[8, 9, 11, 13, 14, 16, 21, 36] In three studies,[8, 9, 13] IFNa monotherapy was compared with no intervention (Group A, n ;= ;137 patients, Table S2). In two studies,[8, 11] high dose was compared with low dose of IFNa (Group B, n ;= ;60, Table S3). In two trials,[21, 36] combined treatment of IFNa and lamivudine was compared with IFNa monotherapy (Group C, n ;= ;48, Table S4). Lastly, in two trials[14, 16] PEG-IFNa was compared with other treatment options (PEG-IFNa combined with ribavirin, PEG-IFNa combined with adefovir dipivoxil and adefovir dipivoxil monotherapy) (Group D, n ;= ;157, Table S5). All patients received treatment for at least 12 ;months. Median follow-up was 22 (6–32), 25 (18–32), 21.6 (6–37.2) and 6 ;months for Group A, B, C and D respectively. Descriptive results are shown in Tables S6–S9. All but three studies[16, 21, 36] had adequate methodological quality (Table ;1).

Table 1. Methological quality of the studies included in the meta-analysis
StudyGeneration of the allocation sequenceDouble bindingFollow-upTotal score
Farci P, et ;al.[8]Adequate (computer generated)No blindingAdequate3
Madejon A, et ;al.[11]Adequate (computer generated)Inadequate (not described)Adequate4
Gaudin JL, et ;al.[9]Adequate (computer generated)No blindingAdequate3
Rosina F, et ;al.[13]Adequate (computer generated)No blindingAdequate3
Niro GA, et ;al.[14]Adequate (computer generated)No blindingAdequate3
Wedemeyer H, et ;al.[16]Inadequate (not described)No blindingAdequate2
Yurdaydin C, et ;al.[21]Inadequate (not described)No blindingAdequate2
Canbakan B, et ;al.[35]Inadequate (not described)No blindingAdequate2
Prevalence of cirrhosis across trials

Group A: The proportion of patients with liver cirrhosis was 60.7% (17 of 28),[8] 59% (13 of 22)[9] and 31.5% (17/54).[13] Group B: 64.3% (18 of 28)[8] and 12.5% (4 of 32)[11] of patients had liver cirrhosis. Group C: Only one study[34] reports data on the prevalence of cirrhosis [23% (6 of 26)]. Group D: Liver cirrhosis was present in 73.7% (28 of 38)[14] and 16.7% (9 of 54) of patients.[16]

Risk of bias in included studies

Five trials[8, 9, 11, 13, 14] reported adequate generation of the allocation sequence; only one[11] reported adequate allocation concealment. Another study[21] reported no blinding of participants, personnel or outcome assessors, whereas the other studies provided insufficient information to permit judge blinding or not.

Biochemical EOT

Group A (Figure S1): All three trials report data regarding this endpoint. There was a significant difference for biochemical EOT favouring IFNa monotherapy [OR, 0.11 (95% CI, 0.04–0.2)]. From the IFNa group (n ;= ;70), 26 (37%) patients achieved biochemical response at the end of treatment compared with only 2 (3%) patients in the no-treatment group (n ;= ;67). Group B (Figure S2): High dose of IFNa was significantly better than low dose [OR, 0.24 (95% CI, 0.08–0.73)]. Biochemical EOT was found in 50% (15 of 30) patients from the high dose-IFNa group compared with 20% (6 of 30) from the low dose group. Group C: There was no significant difference regarding biochemical EOT between combination treatment and IFNa monotherapy [OR, 1.4 (95% CI, 0.4–4.5)]. Biochemical EOT was observed in 61% (17 of 28) patients in combination treatment arm and in 50% (10 of 20) patients in IFNa monotherapy arm. Group D (Figure S3): There was no significant difference between the two treatment arms [OR, 0.761 (95% CI, 0.374–1.551)]. Biochemical EOT was achieved in 30% (22 of 74) patients in PEG-IFNa monotherapy group compared with 25% (21 of 83) of patients who received other treatment.

Virological EOT

Group A (Figure ;2): The efficacy of IFNa was significantly better compared with no intervention [OR, 0.08 (95% CI, 0.03–0.2)]. Fifty-one percent (26 of 51) of patients in IFNa group had negative HDV RNA at the end of treatment compared with 6% (3 of 52) in the group with no intervention. Group B (Figure ;3): High dose IFNa had better efficacy in clearing HDV RNA at the end of treatment [OR, 0.27 (95% CI, 0.1–0.74)]: 70% (21 of 30) of patients received high dose IFNa compared with 37% (11 of 30) with low dose therapy. Group C: 45% (9 of 20) and 57% (16 of 28) of patients from IFNa monotherapy and IFNa plus lamivudine group, respectively, achieved virological EOT. There was no significant difference between the two treatment arms [OR, 1.6 (95% CI, 0.5–5)]. Group D (Figure ;4): 23% (17 of 74) patients who received PEG-IFNa monotherapy achieved statistically significant better virological response at the end of treatment compared with 11% (9 of 83) of patients from the group who received other treatment: OR, 0.419 (95% CI, 0.18–0.974).

image

Figure 2. Virological response at the end of treatment: forest plot of trials of IFNa monotherapy vs. no intervention. Data were expressed as OR (95% CI) in a log scale. HD, high dose; LD, low dose.

Download figure to PowerPoint

image

Figure 3. Virological response at the end of treatment: forest plot of trials of high doses of IFNa vs. low doses. Data were expressed as OR (95% CI) in a log scale. IFN, interferon; IFN HD, IFN high dose; IFN LD, IFN low dose.

Download figure to PowerPoint

image

Figure 4. Virological response at the end of treatment: forest plot of trials of PEG-IFNa vs. other medications. Data were expressed as OR (95% CI) in a log scale. PEG-IFNa, pegylated interferon-A; Co, combination treatment; Ad, adefovir monotherapy.

Download figure to PowerPoint

EOFUP VR

Group A (Figure S4): There was no significant difference between IFNa and no treatment, regarding EOFUP VR [OR, 0.49 (95% CI, 0.07–3.3)]. 6% (3 of 51) and 4% (2 of 52) patients, respectively, had undetectable HDV RNA at the end of follow-up. Group B: There were no significant differences [OR, 0.52 (95% CI, 0.1–2.5)]: 19% (5 of 26) and 11% (3 of 27) from high dose and low dose arm achieved virological response at the end of follow-up respectively. Group C: Similarly, the addition of lamivudine to IFNa treatment did not result in significant differences regarding EOFUP VR. Sustained virological response was observed in 30% (6 of 20) patients with monotherapy and in 32% (9 of 28) with combination therapy. Group D (Figure ;5): There was a significant difference [OR, 0.404 (95% CI (0.189–0.866)] favouring PEG-IFNa: 30% (22 of 74) patients from PEG-IFNa group vs. 15% (12 of 83) patients from the group who received other therapy (PEG-IFNa combined with ribavirin, PEG-IFNa combined with adefovir dipivoxil and adefovir dipivoxil monotherapy) achieved virological response at the end of follow-up.

image

Figure 5. Virological response at the end of follow-up: forest plot of trials of PEG-IFNa vs. other medications. Data were expressed as OR (95% CI) in a log scale. PEG-IFNa, pegylated interferon-A; Co, combination treatment; Ad, adefovir monotherapy.

Download figure to PowerPoint

Histological improvement

Group A (Figure S5): The difference just failed to achieve statistical significance for histological improvement with IFNa treatment [OR, 0.45 (95% CI, 0.19–1.05)]. Histology was improved in 48% (24 of 50) of patients with IFNa and in 30% (14 of 47) of patients with no treatment. Group B: Meta-analysis could not be performed as only one trial reported data regarding this endpoint[8]: 27% (8 of 30) of patients from high dose group and 18% (5 of 28) from low dose group had histological improvement. Group C (Figure S6): A nonsignificant trend favouring combination treatment of IFNa and lamivudine was found [OR, 2.9 (95% CI, 0.6–13.4)]. Histology was improved in 52% (11 of 21) from the combination group and 29% (4 of 14) from the monotherapy group. Group D: Regarding improvement in histological activity (Figure S7), PEG-IFNa was more effective compared with other medications [OR, 0.308 (95% CI, 0.129–0.732)]: 54% (22 of 41) of patients receiving PEG-IFNa had improvement in necroinflammatory activity, whereas only 26% (11 of 43) patients receiving other therapy (PEG-IFNa combined with ribavirin, PEG-IFNa combined with adefovir dipivoxil and adefovir dipivoxil monotherapy). Regarding improvement in fibrosis (Figure S8), the analysis showed no significant difference [OR, 1.21 (95% CI, 0.432–3.393)]: 20% (8 of 41) and 23% (10 of 43) showed such improvement.

Intrahepatic HDAg clearance

Group A: Only one[13] of three trials report data regarding this endpoint. No analysis could be made in this study group. Group B (Figure S9): High dose IFNa was not significantly different compared with low dose [OR, 0.3 (95% CI, 0.04–1.9)]. Thirteen percent (4 of 30) patients and 4% (1 of 28) from high dose and low dose group, respectively, cleared HD Ag in the post-treatment liver biopsies. Group C, D: Meta-analysis could not be performed as none of the trials reported data regarding this endpoint.

No significant heterogeneity was found for all endpoints in all four groups. Using cutoff of 60% in the prevalence of cirrhosis, subgroup meta-analysis for each endpoint in all four groups showed no difference, except for EOFUP VR in group D (Figure ;6).

image

Figure 6. Virological response at the end of follow-up: subgroup meta-analysis of trials of PEG-IFNa vs. other medications, using a cut-off of 60% in the prevalence of cirrhosis. Data were expressed as OR (95% CI) in a log scale. PEG-IFNa, pegylated interferon-A; Co, combination treatment; Ad, adefovir monotherapy.

Download figure to PowerPoint

Discussion

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

To our knowledge, there is no other meta-analysis that investigates different treatment options and outcomes in CHD, except one older meta-analysis[37] reporting response to IFNa therapy compared with no intervention. In this meta-analysis, there was no evidence of a sustained virological response for standard IFNa therapy for chronic HDV infection, as well as no improvement in treatment response with the addition of lamivudine. In contrast, there was superiority of PEG-IFNa regarding virological response at the end of therapy and 6 ;months thereafter compared with other treatment.

Although IFNa is effective in terms of EOT response, nearly all patients had virological relapse between 2 and 6 ;months after stopping treatment.[8] There is no randomised trial comparing the efficacy between a duration of 1 ;year of IFN therapy and a longer treatment period. However, some small studies[32, 38, 39] showed that 2 ;years are not superior to 1 ;year of IFN monotherapy. PEG-IFN has been used in four studies[10, 14-16] in chronic hepatitis D, and the reported SVR rates of 17–43% demonstrate better response compared with standard IFN (Table S10). These data suggest that the best therapeutic option for CHD at present is PEG-IFN, although there is no randomised controlled trial comparing its efficacy against placebo/no treatment or standard IFNa.

Lamivudine combined with standard IFNa and nucleos(t)ide analogues, and ribavirin, alone or combined with standard or pegylated IFN, failed to improve virological response rates. Similarly, our analysis favours PEG-IFN alone compared with other therapies (combined with adefovir or ribavirin or adefovir alone), with respect to virological response 6 ;months after stopping therapy. Taking into consideration their ineffectiveness, it seems that combined treatment of interferon with nucleoside analogues has no role in the management of chronic hepatitis delta. However, Wedemeyer et ;al.[16] has recently shown a decline in HbsAg levels of more than 1 log10 ;IU/mm from baseline to week 48 in a larger proportion of patients treated with PEG-IFN and adefovir, compared with the group treated with PEG-IFN or adefovir alone (P ;< ;0.001 and P ;= ;0.01 respectively). Decline of HbsAg levels is a positive predictive factor for successful treatment in HBeAg positive chronic hepatitis B.[40] Therefore, combination treatment might favour HBV/HDV co-infected patients with high viral load and/or positive e antigen,[41] but the HbsAg decline at present in HDV hepatitis has not been shown to be a surrogate marker.

Baseline and treatment factors predicting the response to IFN therapy are poorly defined. According to one trial,[21] patients who were treatment naïve or had low baseline levels of γGT, tended to respond better to IFN therapy. Furthermore, disease of short duration is reported to favour response to treatment.[21, 42] Other pretreatment predictors of response to PEG-IFN therapy are absence of cirrhosis and HDV RNA titles higher than 2.2 ;× ;107 copies/mL.[43, 44] Hughes et ;al.[43] also reported that the persistence of anti-HD IgM at the end of treatment was a poor prognostic factor of response. No data are available regarding the role of HDV genotype on treatment outcome. We examined all aforementioned predictors for subgroup analyses. However, data were available only to evaluate cirrhosis. We found significantly better sustained virological response after treatment with pegylated interferon in the trials where the prevalence of cirrhotics was lower than 60%. This finding may be consistent with the fact that the presence of advanced fibrosis is a negative predictive factor of response to IFN and PEG-IFN antiviral treatment in both chronic hepatitis B and C.[45]

Regarding histological improvement, our analysis showed a trend to histological improvement of necroinflammatory activity with IFNa therapy, as well as with the combination of IFNa and lamivudine, but the differences were not significant. We were unable to compare fibrosis improvement in the first three study groups due to lack of data. PEG-IFN was found to improve histological activity more than other medications, but no differences in fibrosis could be demonstrated. Wedemeyer et ;al.[16] found that more patients who received combination of PEG-IFN and adefovir had a worsening of fibrosis score (despite an improvement in HbsAg titre), and a higher number of patients receiving PEG-IFNa monotherapy had a worsening of scores for histological activity. However, the authors noted that the second biopsies were performed at the end of treatment and not after longer follow-up, and so, these findings need further evaluation.

The results of our analysis are limited by various factors. First, the small number of patients included, especially in trials, comparing high vs. low dose of IFNa and IFNa monotherapy vs. combination with lamivudine. Secondly, three[16, 21, 36] of eight trials did not have adequate methodological quality. Thirdly, the effect of IFNa on fibrosis could not be explored due to the lack of sufficient data. Finally, duration of follow-up varied among the included trials (6–90 ;months) in this meta-analysis.

In conclusion, our meta-analysis clearly shows that IFNa, although achieves high rates of biochemical and virological response, especially at high doses, does not have a sustained efficacy after the end of treatment. In addition, we showed that the addition of lamivudine as well as of ribavirin and adefovir do not contribute to the antiviral effects of IFNa and PEG-IFNa respectively. We should note that because of the methodological problems of our analysis (small numbers of patients included, lack of designed studies for the specific outcomes, not universally accepted definitions for virological response to treatment) and the limited number of studies available, the results of this meta-analysis should be interpreted with caution. Taking into consideration the high risk of adverse events with interferon therapy, predictive factors of response should be defined. Furthermore, randomised trials should be performed to determine the appropriate duration of treatment and the effectiveness of combination treatment of PEG-IFN with the newest nucleoside analogues (entecavir, tenofovir) in HBV/HDV co-infection. Alternative antiviral targets may be important to be investigated. Prenylation inhibitors reduce virus assembly in vitro[46, 47] and in vivo,[48] but their development is still in the preclinical stage.

Acknowledgement

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

Declaration of personal interests: Dr CK Triantos has served as a speaker for Gilead. Declaration of funding interests: None.

References

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

Supporting Information

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgement
  8. References
  9. Supporting Information
FilenameFormatSizeDescription
apt4993-sup-0001-TableS1.docWord document31KTable S1. Characteristics of excluded randomised trials regarding treatment of chronic hepatitis D [NT: no treatment, n: number of patients, IFNa: interferon-alfa, RBV: ribavirin, EOT: end of treatment, EOFUP VR: virological response at the end of follow-up, m: months, FU: follow-up, X: time until normalisation of alanine aminotransferase (ALT), NR: not reported].
apt4993-sup-0002-TableS2.docWord document28KTable S2. Characteristics of randomised trials comparing IFNa monotherapy vs. no treatment (NT) (IFNa: interferon-alfa, m: months, FU: follow-up, t.i.w: three times a week).
apt4993-sup-0003-TableS3.docWord document26KTable S3. Characteristics of randomised trials comparing high dose (HD) of IFNa vs. low dose (LD) (IFNa: interferon-alfa, m: months, FU: follow-up, t.i.w: three times a week).
apt4993-sup-0004-TableS4.docWord document28KTable S4. Characteristics of randomised trials comparing IFNa monotherapy vs. IFNa plus lamivudine (LAM) (IFNa: interferon-alfa, m: months, FU: follow-up, d: day, t.i.w: three times a week).
apt4993-sup-0005-TableS5.docWord document30KTable S5. Characteristics of randomised trials comparing PEG-IFNa monotherapy vs. other treatment options (PEG-IFNa: pegylated interferon-alfa, RIBA: ribavirin, ADV: adefovir dipivoxil, m: months, FU: follow-up, d: day, w: weekly, Co: combination treatment, Ad: adefovir monotherapy).
apt4993-sup-0006-TableS6.docWord document32KTable S6. Results in randomised trials comparing IFNa monotherapy vs. no treatment (NT) (LD: low dose, HD: high dose, EOT BR: end of treatment biochemical response, EOT VR: end of treatment virological response, EOFUP VR: end of follow-up virological response, HAI: Histological Activity Index, NNT: number needed to treat, PBA: publication bias assessment).
apt4993-sup-0007-TableS7.docWord document31KTable S7. Results in randomised trials comparing low doses (LD) of IFNa vs. high doses (HD) (EOT BR: end of treatment biochemical response, EOT VR: end of treatment virological response, EOFUP VR: end of follow-up virological response, HAI: Histological Activity Index, NNT: number needed to treat, PBA: publication bias assessment).
apt4993-sup-0008-TableS8.docWord document30KTable S8. Results in randomised trials comparing IFNa monotherapy vs. combination treatment of IFNa and lamivudine (LAM) (EOT BR: end of treatment biochemical response, EOT VR: end of treatment virological response, EOFUP VR: end of follow-up virological response, HAI: Histological Activity Index, NNT: number needed to treat, PBA: publication bias assessment).
apt4993-sup-0009-TableS9.docWord document33KTable S9. Results in randomised trials comparing PEG-IFNa monotherapy vs. other treatments (EOT BR: end of treatment biochemical response, EOT VR: end of treatment virological response, EOFUP VR: end of follow-up virological response, HAI: Histological Activity Index, Co: combination treatment, Ad: adefovir monotherapy, NNT: number needed to treat, PBA: publication bias assessment).
apt4993-sup-0010-TableS10.docWord document32KTable S10. Studies of treatment of chronic HDV infection with pegylated interferon (PEG-IFN) (RCT: randomised controlled trial, RIBA: ribavirin, SVR: sustained virological response, ADV: adefovir dipivoxil).
apt4993-sup-0011-FigS1.pngPNG image21KFigure S1. Biochemical response at the end of treatment: Forest plot of trials of IFNa monotherapy vs. no intervention. Data were expressed as OR (95% CI) in a log scale.
apt4993-sup-0012-FigS2.pngPNG image41KFigure S2. Biochemical response at the end of treatment: Forest plot of trials of high doses of IFNa vs. low doses. Data were expressed as OR (95% CI) in a log scale. IFN HD: IFN high dose, IFN LD: IFN low dose.
apt4993-sup-0013-FigS3.pngPNG image21KFigure S3. Biochemical response at the end of treatment: Forest plot of trials of PEG-IFNa vs. other medications. Data were expressed as OR (95% CI) in a log scale.
apt4993-sup-0014-FigS4.pngPNG image2KFigure S4. Virological response at the end of follow-up: Forest plot of trials of IFNa monotherapy vs. no intervention. Data were expressed as OR (95% CI) in a log scale.
apt4993-sup-0015-FigS5.pngPNG image2KFigure S5. Histological improvement at the end of treatment: Forest plot of trials of IFNa monotherapy vs. no intervention. Data were expressed as OR (95% CI) in a log scale.
apt4993-sup-0016-FigS6.pngPNG image2KFigure S6. Histological improvement at the end of treatment: Forest plot of trials of IFNa plus lamivudine vs. IFNa monotherapy. Data were expressed as OR (95% CI) in a log scale.
apt4993-sup-0017-FigS7.pngPNG image21KFigure S7. Improvement in histological activity at the end of treatment: Forest plot of trials of PEG-IFNa vs. other medications. Data were expressed as OR (95% CI) in a log scale.
apt4993-sup-0018-FigS8.pngPNG image21KFigure S8. Improvement in fibrosis at the end of treatment: Forest plot of trials of PEG-IFNa vs. other medications. Data were expressed as OR (95% CI) in a log scale.
apt4993-sup-0019-FigS9.pngPNG image1KFigure S9. Intrahepatic HDAg clearance: Forest plot of trials of high doses of IFNa vs. low doses. Data were expressed as OR (95% CI) in a log scale. IFN HD: IFN high dose, IFN LD: IFN low dose.

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.