Aminoadamantanes for chronic hepatitis C

  • Review
  • Intervention

Authors

  • Mieke H Lamers,

    Corresponding author
    1. Radboud University Medical Center Nijmegen, Department of Gastroenterology and Hepatology, Nijmegen, Netherlands
    2. Copenhagen Trial Unit, Centre for Clinical Intervention Research, Department 7812, Rigshospitalet, Copenhagen University Hospital, The Cochrane Hepato-Biliary Group, Copenhagen, Denmark
    • Mieke H Lamers, Department of Gastroenterology and Hepatology, Radboud University Medical Center Nijmegen, Geert Grooteplein Zuid 10, Nijmegen, 6525 GA, Netherlands. m.lamers@mdl.umcn.nl.

    Search for more papers by this author
  • Mark Broekman,

    1. Radboud University Medical Center Nijmegen, Department of Gastroenterology and Hepatology, Nijmegen, Netherlands
    Search for more papers by this author
  • Joost PH Drenth,

    1. Radboud University Medical Center Nijmegen, Department of Gastroenterology and Hepatology, Nijmegen, Netherlands
    Search for more papers by this author
  • Christian Gluud

    1. Copenhagen Trial Unit, Centre for Clinical Intervention Research, Department 7812, Rigshospitalet, Copenhagen University Hospital, The Cochrane Hepato-Biliary Group, Copenhagen, Denmark
    Search for more papers by this author

Abstract

Background

Around 3% of the world's population (approximately 160 million people) are chronically infected with hepatitis C virus. The proportion of infected people who develop clinical symptoms varies between 5% and 40%. Combination therapy with pegylated interferon-alpha plus ribavirin eradicates the virus from the blood six months after treatment (sustained virological response) in approximately 40% to 80% of infected patients, depending on the viral genotype. New antiviral agents, such as boceprevir and telaprevir, in combination with standard therapy, can increase sustained virological response in genotype 1 infected patients to at least 70%. There is therefore an unmet need for drugs that can achieve a higher proportion of sustained virological response. Aminoadamantanes are antiviral drugs used for treatment of patients with chronic hepatitis C.

Objectives

To assess the beneficial and harmful effects of aminoadamantanes for patients with chronic hepatitis C infection by conducting a systematic review with meta-analyses of randomised clinical trials, as well as trial sequential analyses.

Search methods

We conducted electronic searches of the Cochrane Hepato-Biliary Group Controlled Trials Register (1996 to December 2013), the Cochrane Central Register of Controlled Trials (CENTRAL) 2013, Issue 11 of 12 (1995 to December 2013), MEDLINE (1946 to December 2013), EMBASE (1974 to December 2013), Science Citation Index EXPANDED (1900 to December 2013), the WHO International Clinical Trials Registry Platform (www.who.int/ictrp), Google Scholar, and Eudrapharm up to December 2013 and checked the reference lists of identified publications.

Selection criteria

Randomised clinical trials assessing aminoadamantanes in patients with chronic hepatitis C infection.

Data collection and analysis

Two authors independently extracted data. We assessed for risks of systematic errors ('bias') using the 'Risk of bias' tool. We analysed dichotomous data with risk ratio (RR) and continuous data with mean difference (MD) or standardised mean difference (SMD), both with 95% confidence intervals (CI). We used trial sequential analysis to assess the risk of random errors ('play of chance'). We assessed quality using the GRADE system.

Main results

We included 41 randomised clinical trials with 6193 patients with chronic hepatitis C. All trials had high risk of bias. All included trials compared amantadine versus placebo or no intervention. Standard antiviral therapy was administered equally to the intervention and the control groups in 40 trials. The standard antiviral therapy, which was administered to both intervention groups, was interferon-alpha, interferon-alpha plus ribavirin, and peg interferon-alpha plus ribavirin, depending on the time when the trial was conducted.

When we meta-analysed all trials together, the overall results demonstrated no significant effects of amantadine, when compared with placebo or no intervention, on our all-cause mortality or liver-related morbidity composite outcome (5/2353 (0.2%) versus 6/2264 (0.3%); RR 0.90, 95% CI 0.38 to 2.17; I² = 0%; 32 trials; very low quality). There was also no significant effect on adverse events (288/2869 (10%) versus 293/2777 (11%); RR 0.98, 95% CI 0.84 to 1.14; I² = 0%; 35 trials; moderate quality). We used both fixed-effect and random-effects meta-analyses. Amantadine, when compared with placebo or no intervention, did not significantly influence the number of patients who failed to achieve a sustained virological response (1821/2861 (64%) versus 1737/2721 (64%); RR 0.98, 95% CI 0.95 to 1.02; I² = 35%; 35 trials; moderate quality). However, in the subgroup using interferon plus ribavirin, amantadine decreased the number of patients who failed to achieve a sustained virological response (422/666 (63%) versus 447/628 (71%); RR 0.89, 95% CI 0.83 to 0.96; I² = 41%; 11 trials; low quality). Similar results were found for failure to achieve an end of treatment virological response. Amantadine, when compared with placebo or no intervention, significantly decreased the number of patients without normalisation of alanine aminotransferase (ALT) serum levels at the end of treatment (671/1141 (59%) versus 732/1100 (67%); RR 0.88, 95% CI 0.83 to 0.94; I² = 47%; 19 trials; low quality). Amantadine, when compared with placebo or no intervention, did not significantly influence the end of follow-up biochemical response (1133/1896 (60%) versus 1151/1848 (62%); RR 0.95, 95% CI 0.91 to 1.00; I² = 49%; 21 trials; low quality).

The observed beneficial effects could be true effects but could also be due to both systematic errors (bias) and random errors (play of chance). The latter is due to the fact that trial sequential analyses could not confirm or refute our findings. We were not able to perform meta-analyses for failure of histological improvement or quality of life due to a lack of valid data.

Authors' conclusions

This systematic review does not demonstrate any significant effects of amantadine on all-cause mortality or liver-related morbidity composite outcome and on adverse events in patients with hepatitis C; however, the median trial duration was 12 months, with a median follow-up of six months, which is not long enough to assess the composite outcome sufficiently. Overall, we did not see an effect of amantadine on failure to achieve a sustained virological response. Subgroup analyses demonstrated that the combination of amantadine plus interferon-alpha and ribavirin seems to increase the number of patients achieving a sustained virological response. This finding may be caused by both systematic errors (bias) and risks of random errors (play of chance), but it could also be real. Based on the results of the overall evidence, it appears less likely that future trials assessing amantadine for patients with chronic hepatitis C will show strong benefits. Therefore, it is probably advisable to wait for the results of trials assessing other direct-acting antiviral drugs. In the absence of convincing evidence of benefit, the use of amantadine is justified in the context of randomised clinical trials assessing the effects of combination therapy. We found a lack of evidence on other aminoadamantanes than amantadine.

Résumé scientifique

Aminoadamantanes dans l'hépatite C chronique

Contexte

Environ 3 % de la population mondiale (quelques 160 millions de personnes) sont porteurs d'une infection chronique par le virus de l'hépatite C. La proportion de personnes infectées qui développent des symptômes cliniques varie entre 5 % et 40 %. Un traitement combiné avec l'interféron alpha pégylé et la ribavirine élimine le virus du sang six mois après le traitement (réponse virologique soutenue) chez environ 40 % à 80 % des patients infectés, selon le génotype viral. De nouveaux agents antiviraux, tels que le bocéprévir et le télaprévir, en association avec le traitement standard, peuvent augmenter la réponse virologique soutenue chez les patients infectés par le génotype 1 jusqu'à 70 % au moins. Par conséquent, il existe un besoin non satisfait pour des médicaments qui peuvent atteindre une proportion plus élevée de réponse virologique soutenue. Les aminoadamantanes sont des médicaments antiviraux utilisés pour le traitement des patients atteints d'hépatite C chronique.

Objectifs

Évaluer les effets bénéfiques et nocifs des aminoadamantanes pour les patients atteints d'une infection par l'hépatite C chronique en réalisant une revue systématique avec méta-analyses des essais cliniques randomisés, ainsi que des analyses séquentielles d'essais.

Stratégie de recherche documentaire

Nous avons effectué des recherches électroniques dans le registre des essais contrôlés du groupe Cochrane sur les affections hépato-biliaires (de 1996 à décembre 2013), le registre Cochrane des essais contrôlés (CENTRAL) 2013, numéro 11 sur 12 (de 1995 à décembre 2013), MEDLINE (de 1946 à décembre 2013), EMBASE (de 1974 à décembre 2013), Science Citation Index EXPANDED (de 1900 à décembre 2013), le système d'enregistrement international des essais cliniques de l'OMS (www.who.int/ictrp), Google Scholar, et Eudrapharm jusqu'à décembre 2013 et avons examiné les références bibliographiques des publications identifiées.

Critères de sélection

Essais cliniques randomisés évaluant les aminoadamantanes chez les patients atteints d'hépatite C chronique.

Recueil et analyse des données

Deux auteurs ont indépendamment extrait les données. Nous avons évalué les risques d'erreurs systématiques (« biais ») en utilisant l'outil « Risque de biais ». Nous avons analysé les données dichotomiques avec le risque relatif (RR) et les données continues avec la différence moyenne (DM) ou la différence moyenne standardisée (DMS), tous avec des intervalles de confiance (IC) à 95 %. Nous avons utilisé l'analyse séquentielle des essais pour évaluer le risque d'erreurs aléatoires (« effet de hasard »). La qualité a été évaluée à l'aide du système GRADE.

Résultats principaux

Nous avons inclus 41 essais cliniques randomisés portant sur 6 193 patients atteints d'hépatite C chronique. Tous les essais présentaient un risque élevé de biais. Tous les essais inclus comparaient l'amantadine à un placebo ou à l'absence d'intervention. Un traitement antiviral standard était administré de manière égale aux groupes d'intervention et témoins dans 40 essais. Le traitement antiviral standard, administré aux deux groupes d'intervention, était l'interféron alpha, l'interféron alpha associé à la ribavirine ou le peg-interféron alpha associé à la ribavirine, selon le moment de réalisation de l'essai.

Lorsque nous avons effectué une méta-analyse de tous les essais, les résultats globaux n'ont démontré aucun effet significatif de l'amantadine, par rapport à un placebo ou à l'absence d'intervention, sur notre critère de jugement composite de la mortalité toutes causes confondues ou de la morbidité hépatique (5/2 353 (0,2 %) contre 6/2 264 (0,3 %) ; RR 0,90, IC à 95 % de 0,38 à 2,17 ; I² = 0 % ; 32 essais ; preuves de très faible qualité). Il n'y avait également aucun effet significatif sur les événements indésirables (288/2 869 (10 %) contre 293/2 777 (11 %) ; RR 0,98, IC à 95 % de 0,84 à 1,14 ; I² = 0 % ; 35 essais ; qualité modérée). Nous avons utilisé des méta-analyses à effets fixes et à effets aléatoires. L'amantadine, par rapport à un placebo ou à l'absence d'intervention, n'a pas eu d'influence significative sur le nombre de patients n'ayant pas obtenu une réponse virologique soutenue, (1 821/2 861 (64 %) contre 1 737/2 721 (64 %) ; RR 0,98, IC à 95 % de 0,95 à 1,02 ; I² = 35 % ; 35 essais ; qualité modérée). Cependant, dans le sous-groupe utilisant l'interféron plus ribavirine, l'amantadine a réduit le nombre de patients n'ayant pas obtenu une réponse virologique soutenue, (422/666 (63 %) contre 447/628 (71 %) ; RR 0,89, IC à 95 % de 0,83 à 0,96 ; I² = 41 % ; 11 essais ; faible qualité). Des résultats similaires ont été trouvés pour l'échec de l'obtention d'une réponse virologique à la fin du traitement. L'amantadine, par rapport à un placebo ou à l'absence d'intervention, a significativement réduit le nombre de patients sans normalisation des niveaux sériques de l'alanine aminotransférase (ALT) à la fin du traitement (671/1 141 (59 %) contre 732/1 100 (67 %) ; RR 0,88, IC à 95 % de 0,83 à 0,94 ; I² = 47 % ; 19 essais ; faible qualité). L'amantadine, par rapport à un placebo ou à l'absence d'intervention, n'a pas eu d'influence significative sur la réponse biochimique à la fin du suivi (1 133/1 896 (60 %) contre 1 151/1 848 (62 %) ; RR 0,95, IC à 95 % de 0,91 à 1,00 ; I² = 49 % ; 21 essais ; faible qualité).

Les effets bénéfiques observés pourraient être de véritables effets, mais ils pourraient également être dûs à la fois aux erreurs systématiques (biais) et aux erreurs aléatoires (effet de hasard). Cette dernière possibilité est due au fait que les analyses séquentielles d'essais n'ont pas pu confirmer ou réfuter nos résultats. Nous n'avons pas pu réaliser de méta-analyses pour l'échec d'amélioration histologique ou la qualité de vie en raison d'un manque de données valides.

Conclusions des auteurs

Cette revue systématique ne permet pas de démontrer un effet significatif de l'amantadine sur le critère de jugement composite de la mortalité toutes causes confondues ou de la morbidité hépatique, ni sur les événements indésirables, chez les patients atteints d'hépatite C ; cependant, la durée moyenne des essais était de 12 mois, avec un suivi médian de six mois, ce qui n'est pas assez long pour évaluer suffisamment ce critère de jugement composite. Dans l'ensemble, nous n'avons pas vu d'effet de l'amantadine sur l'échec de l'obtention d'une réponse virologique soutenue. Les analyses en sous-groupes ont démontré que l'association de l'amantadine et de l'interféron alpha plus la ribavirine semblait augmenter le nombre de patients obtenant une réponse virologique soutenue. Ce résultat peut être causé par les erreurs systématiques (biais) et les risques d'erreurs aléatoires (effet de hasard), mais il pourrait également être réel. D'après les résultats de l'ensemble des preuves, il semble moins probable que de futurs essais évaluant l'amantadine pour les patients atteints d'hépatite C chronique montreront de solides effets bénéfiques. Par conséquent, il est probablement préférable d'attendre les résultats d'essais évaluant d'autres médicaments antiviraux à action directe. En l'absence de preuves convaincantes d'un bénéfice, l'utilisation de l'amantadine est justifiée dans le contexte d'essais cliniques randomisés évaluant les effets d'un traitement combiné. Nous avons trouvé un manque de preuves concernant d'autres aminoadamantanes que l'amantadine.

Plain language summary

Aminoadamantanes for chronic hepatitis C

Background

Hepatitis C virus is mainly transmitted by contact with infected blood. Chronic hepatitis C infection affects around 3% of the world's population and progresses slowly. Most patients present without symptoms, or with symptoms like fatigue or liver-related morbidity (illness). Frequently, the disease is discovered by coincidence because of abnormal laboratory results. Between 5% and 40% of all infected patients will develop severe liver damage, which can cause severe liver-related morbidities and eventually death. Current treatment consists of pegylated interferon-alpha plus ribavirin, and in some groups of patients these two agents are administered in combination with antiviral drugs such as telaprevir or boceprevir. It is then possible to eradicate the virus from the blood in at least 70% of patients with chronic hepatitis C, but the clinical effects are not known.

Review questions and study characteristics

Only amantadine has been tested in randomised clinical trials including participants with chronic hepatitis C. The main goal of these trials was to investigate whether amantadine as a single therapy or amantadine in combination with other antiviral therapy, compared with placebo or no intervention (with or without antiviral therapy), could increase the proportion of patients with virus eradication from the blood. This review evaluates whether amantadine has any beneficial or harmful effect in patients with chronic hepatitis C. The primary outcomes were all-cause mortality or liver-related morbidity (combined outcome) and adverse events. The review includes 41 randomised clinical trials with a total of 6193 patients.

Key results and quality of evidence

This review shows that there seems to be no significant benefit of amantadine on hepatitis C-infected patients regarding all-cause mortality or liver-related morbidity. We were unable to assess the effect of amantadine on quality of life due to lack of data from the trials. Furthermore, amantadine did not increase the proportion of patients with a sustained virological response which is clearance of the virus from the blood six months after treatment. We considered all the included trials to have a high risk of bias. Accordingly, the evidence from this review does not support the routine clinical use of amantadine. There is some justification for amantadine to be used in future randomised clinical trials. We found no randomised clinical trials assessing other aminoadamantanes, for example rimantadine.

Résumé simplifié

Les aminoadamantanes pour l'hépatite C chronique

Contexte

Le virus de l'hépatite C est transmis principalement par contact avec du sang infecté. L'infection chronique par l'hépatite C touche environ 3 % de la population mondiale et progresse lentement. La plupart des patients présentent sans symptômes, ou avec des symptômes tels que la fatigue ou une morbidité (maladie) liée au foie. Souvent, la maladie est découverte par coïncidence en raison de résultats de laboratoire anormaux. Entre 5 % et 40 % de tous les patients infectés développent des lésions hépatiques graves, qui peuvent entraîner de graves morbidités liées au foie et finalement, le décès. Le traitement actuel consiste en l'interféron alpha pégylé associé à la ribavirine et, chez certains groupes de patients, ces deux agents sont administrés en association avec des médicaments antiviraux tels que le télaprévir ou le bocéprévir. Il est ainsi possible d'éradiquer le virus du sang chez au moins 70 % des patients atteints d'hépatite C chronique, mais les effets cliniques ne sont pas connus.

Questions de la revue et caractéristiques des études

Seule l'amantadine a été testée dans des essais cliniques randomisés portant sur des participants atteints d'hépatite C chronique. L'objectif principal de ces essais était de déterminer si l'amantadine en traitement unique ou en association avec d'autres traitements antiviraux, par rapport à un placebo ou à l'absence d'intervention (avec ou sans traitement antiviral), pouvait augmenter la proportion de patients ayant éradiqué le virus du sang. Cette revue évalue si l'amantadine a un quelconque effet bénéfique ou nocif chez les patients atteints d'hépatite C chronique. Les critères de jugement principaux étaient la mortalité toutes causes confondues ou la morbidité liée au foie (critère de jugement combiné) et les événements indésirables. La revue inclut 41 essais cliniques randomisés avec un total de 6 193 patients.

Résultats principaux et qualité des preuves

Cette revue montre que l'amantadine ne semble avoir aucun bénéfice significatif chez les patients infectés par l'hépatite C pour ce qui est de la mortalité toutes causes confondues ou de la morbidité liée au foie. Nous n'avons pas pu évaluer l'effet de l'amantadine sur la qualité de vie en raison du manque de données issues des essais. En outre, l'amantadine n'a pas augmenté la proportion de patients présentant une réponse virologique soutenue, qui correspond à l'absence du virus dans le sang six mois après le traitement. Nous avons considéré que tous les essais inclus présentaient un risque élevé de biais. En conséquence, les preuves issues de cette revue ne permettent pas de recommander l'utilisation clinique systématique de l'amantadine. L'utilisation de l'amantadine dans de futurs essais cliniques randomisés est justifiée dans une certaine mesure. Nous n'avons pas trouvé d'essais cliniques randomisés évaluant d'autres aminoadamantanes, par exemple la rimantadine.

Notes de traduction

Traduit par: French Cochrane Centre 31st July, 2014
Traduction financée par: Financeurs pour le Canada : Instituts de Recherche en Santé du Canada, Ministère de la Santé et des Services Sociaux du Québec, Fonds de recherche du Québec-Santé et Institut National d'Excellence en Santé et en Services Sociaux; pour la France : Ministère en charge de la Santé

Summary of findings(Explanation)

Summary of findings for the main comparison. 
  1. 1Quality of the evidence was downgraded by three points based on the risk of bias, presence of imprecision, and indirectness of evidence.
    2Quality of the evidence was downgraded by one or two points based on inconsistency of results and indirectness of evidence.
    3Investigators failed to report quality of life assessment.
    4Six trials included different quality of life scales/questionnaires.

Aminoadamantanes compared with placebo or no intervention for hepatitis C

Patient or population: patients with chronic hepatitis C

Settings: mainly outpatients in tertiary and teaching hospitals

Intervention: aminoadamantanes

Comparison: placebo or no intervention

OutcomesIllustrative comparative risks* (95% CI)Relative effect
(95% CI)
No of participants
(studies)
Quality of the evidence
(GRADE)
Assumed riskCorresponding risk
Placebo or no interventionAminoadamantanes

All-cause mortality or liver-related morbidity

Follow-up: 12 to 30 months

Study population

RR 0.90

(0.38 to 2.17)

4617
(32 trials)
⊕⊝⊝⊝
very low 1
3 per 1000 3 per 1000
(2 to 8)

Adverse events

Follow-up: 12 to 30 months

Study population

RR 0.98

(0.84 to 1.14)

5646
(35 trials)
⊕⊕⊕⊝
moderate2
106 per 1000 101 per 1000
(87 to 118)

Failure of end of treatment virological response

Absence of clearance of HCV RNA from the blood at end of treatment

Follow-up: 6 to 12 months

Study population

RR 0.95

(0.90 to 1.00)

4861
(30 trials)
⊕⊕⊕⊝
moderate2
534 per 1000 519 per 1000
(492 to 547)

Failure of sustained virological response

Absence of clearance of HCV RNA from the blood 6 months after treatment

Follow-up: 12 to 30 months

Study population

RR 0.98

(0.95 to 1.02)

5582
(35 trials)
⊕⊕⊕⊝
moderate2
639 per 1000 637 per 1000
(618 to 663)

Quality of life

Different QoL scales

Follow-up: 12 to 30 months

See commentSee comment 

1181

(6 trials)

⊕⊝⊝⊝
very low3,4

Failure of normalisation of ALT at end of treatment

Follow-up: 6 to 12 months

Study population

RR 0.88

(0.83 to 0.94)

2241
(19 trials)
⊕⊕⊝⊝
low2
666 per 1000 589 per 1000
(556 to 629)

Failure of normalisation of ALT at end of follow-up

Follow-up: 12 to 30 months

Study population

RR 0.95

(0.91 to 1.00)

3744
(21 trials)
⊕⊕⊝⊝
low2
623 per 1000 598 per 1000
(573 to 630)
*The basis for the assumed risk (e.g., the median control group risk across studies) is provided in footnotes. The corresponding risk (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI).
ALT: alanine aminotransferase; CI: confidence interval; HCV: hepatitis C virus; QoL: quality of life; RNA: ribonucleic acid; RR: risk ratio.
GRADE Working Group grades of evidence
High quality: Further research is very unlikely to change our confidence in the estimate of effect.
Moderate quality: Further research is likely to have an important impact on our confidence in the estimate of effect and may change the estimate.
Low quality: Further research is very likely to have an important impact on our confidence in the estimate of effect and is likely to change the estimate.
Very low quality: We are very uncertain about the estimate.

Background

Description of the condition

Hepatitis C virus was first described in 1989 (Choo 1989). Around 3% of the world's population is affected by chronic hepatitis C infection: on average approximately 160 million people (Sy 2006; Lavanchy 2011). Hepatitis C is a leading cause of liver-related morbidity and mortality, with hepatic fibrosis, end-stage liver cirrhosis, and hepatocellular carcinoma being the dominant clinical sequelae (Sy 2006).

Chronic hepatitis C infection progresses slowly, over a time frame of 15 to 50 years. Both prospective or retrospective studies, following cohorts of patients for decades, suggest that less than 10% of all infected individuals will develop end-stage liver disease. However, there are also publications that have reported on patients who developed cirrhosis two to three decades after infection, with a range of 0.5% to 39% (Koretz 1993; Kenny-Walsh 1999; Rodger 2000; Wiese 2000; Thein 2008; Seeff 2009). The incidence rate of hepatocellular carcinoma is 3 patients per 100,000 person-years in the USA (El-Serag 2003). Hepatitis C is responsible for one-third of hepatocellular carcinomas (El-Serag 2003). In cirrhotic hepatitis C patients, the annual occurrence of hepatocellular carcinoma is 1% to 4% (Lauer 2001). Furthermore, chronic hepatitis C infection is the most common indication for orthotopic liver transplantation (Kim 2009).

Hepatitis C is divided into six genotypes (Simmonds 2005). Genotypes 1 to 4 are the most common (Simmonds 2005). Several factors have an influence on achieving a sustained virological response to antiviral drugs; genotype is one of these factors (Asselah 2010). Genotypes 2 and 3 respond better to antiviral treatment than genotypes 1 and 4 (Asselah 2010).

In 1990, the antiviral drug interferon-alpha was approved for the treatment of chronic hepatitis C as monotherapy (Tine 1991). Interferon-alpha was administered subcutaneously in doses of more than or equal to 3 million units (MU) in the induction phase (over one to three months) and less than 3 MU in the maintenance phase (Tine 1991). Only 10% to 17% of patients achieved sustained virological response, compared to 1% to 3% of the patients receiving no intervention (Davis 1989; Myers 2002).

Antiviral drugs for patients with hepatitis C-related liver disease have improved considerably during the past two decades (Ghany 2009). In 1998, trials assessed the combination of interferon-alpha and ribavirin (Davis 1998; McHutchison 1998; Poynard 1998). This combination treatment resulted in an improved antiviral response in treatment-naive chronic hepatitis C-infected patients (Brok 2010) and in previously treated patients who had failed to respond to interferon-alpha monotherapy, compared with interferon-alpha alone (Brok 2010).

The success of antiviral therapy has been assessed by 'sustained virological response', that is clearance of hepatitis C ribonucleic acid (RNA) from the blood six months after treatment. Observational studies suggest that people with achieved sustained virological response have less disease progression and lower risk of hepatocellular carcinoma (Ueno 2009). Based on systematic reviews of randomised clinical trials comparing ribavirin plus interferon-alpha versus interferon-alpha alone, the combination of drugs seems to result in more patients with achieved sustained virological response, but we do not know if this results in less mortality or morbidity (Brok 2010). Accordingly, sustained virological response is a non-validated, putative, surrogate outcome measure (Gluud 2007). Furthermore, a recent trial has shown that there is increased mortality in patients who were retreated with interferon-alpha compared with non-treated patients (Di Bisceglie 2011). Other trials cannot confirm or invalidate this finding (Di Martino 2011).

The standard of treatment of chronic hepatitis C infection, according to guidelines, is a combination of pegylated interferon-alpha (peg interferon-alpha) and ribavirin (Ghany 2009; EASL 2011). The regimen can include either peg interferon-alpha-2b (Peg-Intron®, Schering Plough Corp., Kenilworth, NJ) or peg interferon-alpha-2a (Pegasys®, Hoffmann-La Roche, Nutley, NJ), both of which are administered subcutaneously (Awad 2010). The optimal dose of peg interferon-alpha-2b is 1.5 µg/kg/week (Awad 2010). Peg interferon-alpha-2a is administered at a fixed dose of 180 µg weekly (Awad 2010). Ribavirin is administered orally with weight-based total daily doses between 800 mg and 1200 mg (Brok 2009). Some 40% to 80% of chronic hepatitis C patients without co-infection with hepatitis B virus or human immunodeficiency virus (HIV) will achieve a sustained virological response after treatment with peg interferon-alpha and ribavirin (Simin 2007; Awad 2010).

Recently, a new class of drugs for hepatitis C genotype 1 has emerged. These drugs have to be given together with the current standard treatment. They are antiviral agents that inhibit the NS3/N4A serine protease of hepatitis C. This triple therapy can increase sustained virological response proportions to reach 70% to 80% (Bacon 2011; Jacobson 2011; Poordad 2011; Sherman 2011; Zeuzem 2011).

During the 1990s and 2000s, ribavirin was tested as monotherapy for chronic hepatitis C infection (Brok 2009). Ribavirin does not seem to have any major effect on the course of hepatitis C infection (Brok 2009).

Description of the intervention

Aminoadamantanes, such as amantadine and rimantadine, are another antiviral drug group and have also been investigated in several studies for treatment of patients with chronic hepatitis C (Brillanti 1999; Smith 2004). Aminoadamantanes have been investigated as oral monotherapy, administered mostly at a dose of 100 mg twice a day, and also in combination with interferon-alpha or ribavirin, or both. The benefits and harms of amantadine in patients with chronic hepatitis C infection have been explored previously in a meta-analysis (Deltenre 2004). The authors concluded that amantadine therapy had no effect in naive patients or relapsers. However, combination therapy of amantadine with interferon-alpha and ribavirin did improve sustained virological response proportions in non-responder patients.

How the intervention might work

Aminoadamantanes have been used for many years to prevent infection with influenza and have been shown to have activity against Flaviviridae, encompassing hepatitis C infection (Koff 1980). Known mechanisms of action of aminoadamantanes include inhibition of an early step in viral replication, most likely viral uncoating and interaction with the influenza A viral matrix protein (M2), which is important in virion budding (De Clercq 2001). The aminoadamantane amantadine acts in a similar way to ribavirin, which in monotherapy often improves liver biochemistry (Reichard 1991; Brok 2009). However, it is unclear whether aminoadamantanes reduce the hepatitis C viral load or improve liver biochemistry (Reichard 1993). Furthermore, it is unclear whether aminoadamantanes affect patient-relevant outcomes.

Why it is important to do this review

The combination therapy of peg interferon-alpha and ribavirin achieves virus eradication of approximately 40% to 80% (Simin 2007; Awad 2010). This indicates that there is an unmet need for drugs which can achieve a higher proportion of sustained virological response. With the new direct antiviral agents, higher proportions can be reached, but still not 100% (Bacon 2011; Jacobson 2011; Poordad 2011; Sherman 2011; Zeuzem 2011). Several studies have so far been published regarding the effects of aminoadamantanes. Our systematic review aims to assess the benefits and harms of aminoadamantanes. This systematic review may have practical implications for the way patients with chronic hepatitis C are treated.

We are aware of a meta-analysis by Deltenre 2004, who studied the benefits and harms of aminoadamantanes for patients with chronic hepatitis C. A total of 31 randomised clinical trials including 4831 patients with chronic hepatitis C infection were included in this meta-analysis. Since 2004, new randomised clinical trials of aminoadamantanes have been conducted and our review therefore includes all the trials identified both before and after this meta-analysis.

Objectives

To assess the beneficial and harmful effects of aminoadamantanes for patients with chronic hepatitis C infection by conducting a systematic review with meta-analyses of randomised clinical trials, as well as trial sequential analyses.

Methods

Criteria for considering studies for this review

Types of studies

Randomised clinical trials assessing aminoadamantanes in patients with chronic hepatitis C infection irrespective of duration of treatment, language, publication type or status, and blinding. We excluded quasi-randomised studies or other observational studies captured during the search process from the reporting of benefit but they were included for the reporting of harm. However, we did not conduct specific searches for the latter studies.

Types of participants

We included patients with chronic hepatitis C. The diagnosis was based on the presence of serum hepatitis C RNA plus elevated transaminases for more than six months, or chronic hepatitis documented by liver biopsy. We also included patients diagnosed with 'non-A, non-B' chronic hepatitis as some trials may have been conducted before hepatitis C RNA analyses were widely available.

Based on the existence of and response to previous antiviral treatment, we classified the included patients as naive (not previously treated with antivirals), relapsers (patients with a transient serological viral response to previous treatment with antivirals), or non-responders (patients without a serological viral response to previous treatment with antivirals).

We excluded patients with chronic hepatitis C who had undergone liver transplantation.

Types of interventions

We aimed to perform the following comparisons.

  • Aminoadamantanes versus placebo or no intervention.

  • Aminoadamantanes plus standard antiviral therapy versus standard antiviral therapy alone.

  • High-dose aminoadamantanes versus low-dose aminoadamantanes.

  • Long-duration aminoadamantanes versus short-duration aminoadamantanes.

Co-interventions were allowed if administered equally to the intervention groups.

Types of outcome measures

Primary outcomes
  1. All-cause mortality or liver-related morbidity as a composite outcome: number of patients who died or who developed, for example, cirrhosis (compensated or decompensated), ascites, hepatic encephalopathy, or hepatocellular carcinoma during treatment.

  2. Adverse events (according to the Code of Federal Regulations and ICH guidelines (ICH-GCP)): number of patients with either serious adverse events or treatment discontinuation due to any adverse event. An adverse event is defined as "Any untoward medical occurrence in a patient or clinical investigation subject administered a pharmaceutical product and which does not necessarily have to have a causal relationship with this treatment" (ICH-GCP 1997).

  3. Quality of life (as reported in the trials).

Secondary outcomes
  1. Failure of serum (or plasma) sustained virological response: number of patients with detectable hepatitis C RNA at least six months after treatment.

  2. Failure of end of treatment virological response: number of patients with detectable hepatitis C RNA at the end of treatment.

  3. Failure of histological response: number of patients without improvement of histology (inflammation score (grading) or fibrosis score (staging) as defined by the individual trials).

  4. Number of patients without normalisation of alanine aminotransferase (ALT) or aspartate transaminase (AST) serum levels or both (defined by the individual trials) at end of treatment and end of follow-up.

Search methods for identification of studies

Electronic searches

We searched the Cochrane Hepato-Biliary Group Controlled Trials Register (1996 to December 2013) (Gluud 2014), the Cochrane Central Register of Controlled Trials (CENTRAL) 2013, Issue 11 of 12 (1995 to December 2013), MEDLINE (1946 to December 2013), EMBASE (1974 to December 2013), and Science Citation Index EXPANDED (1900 to December 2013) (Royle 2003). We also searched the WHO International Clinical Trials Registry Platform (www.who.int/ictrp), Google Scholar, and Eudrapharm. We have provided the search strategies in Appendix 1. We performed the latest search in December 2013. We will improve the searches for any later updates, if necessary.

Searching other resources

We searched for further trials by reading the reference lists of the identified publications. We checked the retrieved review articles and meta-analyses in order to find randomised clinical trials not identified by the electronic searches. We searched the journals Hepatology and Journal of Hepatology for abstracts from various gastrointestinal meetings. We wrote to the principal authors of the identified randomised clinical trials to request additional information.

Data collection and analysis

Selection of studies

Two authors (ML, MB) independently inspected each reference identified by the searches and applied the inclusion criteria. For possibly relevant publications, or in cases of disagreement, we obtained the full article and inspected this independently. In cases where ML and MB still disagreed, CG was consulted.

Data extraction and management

Two authors (ML, MB) extracted data independently. In case of disagreement between the two authors, a third author (CG) arbitrated. We discussed the data extraction, documented decisions and, where necessary, contacted the authors of trials for clarification. Trials were identified by the name of the first author and year in which the trial was published in full and ordered chronologically.

We extracted, checked, and recorded the following data:

  • Characteristics of trials: date, location and setting; publication status; sponsor (specified, known, or unknown); duration of follow-up; bias domains; sample size calculation.

  • Characteristics of participants: number of participants in each group; age; sex; ethnicity; weight or body mass index; viral load at the beginning of treatment; degree of fibrosis at the beginning of treatment.

  • Characteristics of interventions: dose and duration of aminoadamantanes, and any co-interventions.

  • Characteristics of outcome measures: whenever possible, we recorded the number of events listed under 'outcomes' in each group of the trial; we extracted information about harms from observational studies.

We incorporated cross-over trials in meta-analysis by using the end of first period strategy, which indicates that the analysis is based on only the first period of the included trial.

Assessment of risk of bias in included studies

Methodological quality is defined as confidence that the design and reporting of a randomised clinical trial will restrict bias in the comparison of the intervention (Moher 1998). According to empirical evidence, risk of bias in a trial can be assessed using 'Risk of bias' domains (Schultz 1995; Moher 1998; Kjaergard 2001; Wood 2008; Lundh 2012; Savović 2012; Savović 2012a). These are the following.

Allocation sequence generation
  • Low risk of bias: sequence generation was achieved using computer random number generation or a random number table. Drawing lots, tossing a coin, shuffling cards, and throwing dice are adequate if performed by an independent research assistant not otherwise involved in the trial.

  • Uncertain risk of bias: the method of sequence generation was not specified.

  • High risk of bias: the sequence generation method was not random.

Allocation concealment
  • Low risk of bias: the participant allocations could not have been foreseen in advance of, or during, enrolment. Allocation was controlled by a central and independent randomisation unit. The allocation sequence was unknown to the investigators (for example, if the allocation sequence was hidden in sequentially numbered, opaque, and sealed envelopes).

  • Uncertain risk of bias: the method used to conceal the allocation was not described so that intervention allocations may have been foreseen in advance of, or during, enrolment.

  • High risk of bias: the allocation sequence was likely to be known to the investigators who assigned the participants.

Blinding of participants and personnel
  • Low risk of bias: it was described that both the participants and the personnel were blinded, and the method of blinding was described, so that knowledge of group assignment was adequately prevented during the trial.

  • Uncertain risk of bias: it was not described if the trial was blinded, or the trial was described as blind but the method of blinding was not described, so that knowledge of group assignment was possible during the trial.

  • High risk of bias: the trial was not blinded, so that the group assignment was known during the trial.

Blinded outcome assessment
  • Low risk of bias: outcome assessment was done blinded for all relevant outcomes, and the method of blinding was described, so that knowledge of group assignment was adequately prevented.

  • Unclear: it was not described if outcome assessment was blinded, or the outcome assessment was described as blind, but the method of blinding was not described, so that knowledge of group assignment was possible.

  • High risk of bias: outcome assessment was not blinded, so that the group assignment was known for outcome assessors.

Incomplete outcome data
  • Low risk of bias: missing data were unlikely to make treatment effects depart from plausible values. Sufficient methods, such as multiple imputation, have been employed to handle missing data.

  • Uncertain risk of bias: there was insufficient information to assess whether missing data in combination with the method used to handle missing data were likely to induce bias on the results.

  • High risk of bias: the results were likely to be biased due to missing data.

Selective outcome reporting
  • Low risk of bias: all outcomes were predefined (for example, in a published protocol) and reported, or all clinically relevant and reasonably expected outcomes were reported, which included all primary and secondary outcome measures as stated under 'Types of outcome measures'.

  • Uncertain risk of bias: it is unclear whether all predefined and clinically relevant and reasonably expected outcomes were reported, which included all primary and secondary outcome measures as stated under 'Types of outcome measures'.

  • High risk of bias: one or more clinically relevant and reasonably expected outcomes, which included all primary and secondary outcome measures as stated under 'Types of outcome measures', were not reported, and data on these outcomes were likely to have been recorded.

Vested interest bias
  • Low risk of bias: if the trial's source(s) of funding did not come from any parties that might have conflicting interests (e.g., an amantadine manufacturer), or if any academic, professional, financial, or other benefits to the person responsible for the trial were independent of the direction or statistical significance of the trial results.

  • Uncertain: if the source of funding was not clear, or if it was unclear if the person responsible for the trial stands to benefit according to the direction or statistical significance of the trial results.

  • High risk of bias: if the trial's source of funding had a conflict of interest, or if any academic, professional, financial, or other benefits to the person responsible for the trial were dependent of the direction or statistical significance of the trial results.

We assessed all trials for risk of bias. If we judged the risk of bias in a trial as 'uncertain' or 'high' for a domain, then we considered the trial to have 'high risk of bias'. If we judged a trial as low risk of bias in all seven domains, then we considered the trial as 'low risk of bias'. If we judged a trial as low risk of bias in at least the four domains random sequence generation, allocation concealment, blinding of participants and personnel, and blinding of outcome assessment, then we judged it to have 'lower risk of bias'. By using the term 'lower risk of bias', we wished to signal that we were well aware that such trials might indeed have risk of bias.

We handled reporting biases following the recommendations of The Cochrane Collaboration (Higgins 2011). We assessed funnel plot asymmetry (Higgins 2011), even though asymmetric funnel plots are not necessarily caused by publication bias and publication bias does not necessarily cause asymmetry in a funnel plot (Egger 1997).

Measures of treatment effect

The treatment effects in this meta-analysis are dichotomous or continuous. We expressed dichotomous data as risk ratio (RR) with 95% confidence intervals (CI). We derived the number needed to treat to benefit (NNTB) from the risk difference (RD) in case it was significant. For continuous data, we used the mean difference if the outcomes of the trials were measured in the same way. Where appropriate, we used the standardised mean difference to combine trials that measured the same outcome but using different methods.

Unit of analysis issues

As unit of analysis we used the reported outcomes per intervention group within the randomised clinical trials. In case no randomised clinical trials were identified, the results of the prospective cohort studies obtained with the search were to be presented in a narrative way in the 'Discussion' section of the review.

Dealing with missing data

We did the following to deal with missing data.

  • We contacted the original investigators to request missing data.

  • We performed sensitivity analyses to assess how sensitive our results were to reasonable changes in the assumptions that were made. We performed our analyses based on the intention-to-treat principle using imputation for the outcomes. We used the following scenarios (Hollis 1999).

    • Carry forward analysis: if participants had missing outcome data, we used the last reported observed response ('carry forward') in the nominator, and included all randomised participants in the denominator.

    • Extreme case analysis favouring the experimental intervention ('best-worst' case scenario): none of the drop-outs and participants lost from the experimental group but all of the drop-outs and participants lost from the control group experienced the outcome, including all randomised participants in the denominator.

    • Extreme case analysis favouring the control ('worst-best' case scenario): all drop-outs and participants lost from the experimental group but none from the control group experienced the outcome, including all randomised participants in the denominator.

Assessment of heterogeneity

We assessed heterogeneity using the Chi2 test of heterogeneity and quantified inconsistency with the I2 statistic (Higgins 2002). In cases of substantial heterogeneity, as measured by a Chi2 test with a P value less than 0.1 or an I2 statistic value greater than 70%, we did not conduct meta-analysis. We assessed sources of clinical, methodological, and statistical heterogeneity in subgroup analyses.

Assessment of reporting biases

This is described under 'Assessment of risk of bias in included studies'.

Data synthesis

For the statistical analyses, we used Review Manager 5.2 (RevMan 2012). We meta-analysed the data with both a random-effects model (DerSimonian 1986) and a fixed-effect model (DeMets 1987) to ensure the robustness of the results. In case of differences in the results that the two models may have produced, we presented the results using both methods. If there were no differences in the results, we presented the results of the fixed-effect model only (Higgins 2011). If there was considerable variation in the results, and particularly if the direction of effect was inconsistent, it may be misleading to quote the average value for the intervention effect; we therefore interpreted the meta-analyses with utmost care.

Trial sequential analysis

Trial sequential analysis is a tool for quantifying the statistical reliability of data in cumulative meta-analysis, adjusting for sparse data, and repetitive testing of accumulating data (Brok 2008; Wetterslev 2008; Brok 2009a; Thorlund 2009, Wetterslev 2009; Thorlund 2010). Trial sequential analysis is a methodology that combines the calculation of a required information size (the sample sizes of the trials in the meta-analysis ought to answer a research question reliably) with the threshold of statistical significance (CTU 2011; Thorlund 2011).

Our intention was to perform trial sequential analysis primarily on the data from the trials with low risk of bias (Brok 2008; Wetterslev 2008). However, we chose to carry out trial sequential analysis on all trials because there were only a few trials with lower risk of bias. We analysed the outcome measures using trial sequential analysis no matter whether they yielded a statistically significant result in the meta-analysis or not. We used the meta-analytic estimate of the control event proportions of all trials, independent of risk of bias, as the control event proportion in the trial sequential analysis. We used the intervention effect estimated in the meta-analysis using all trials or used an a priori intervention effect of 20% risk ratio reduction. The unit of the intervention effect was risk ratio reduction for all dichotomous data.

For each trial sequential analysis performed, we calculated a diversity-adjusted required information size based on the intervention effect suggested by the trials with low risk of bias (LBHIS) or an a priori intervention effect of 20% risk ratio reduction, a risk of type I error of 5% and a risk of type II error of 20% or 10% (Wetterslev 2009). We performed the diversity adjustment using the observed diversity adjustment factor (1/(1-D2)) where D2 is the estimated heterogeneity among all trials and with an a priori assumed final diversity of 50% (Wetterslev 2009).

Subgroup analysis and investigation of heterogeneity

We planned the following subgroup analyses.

  • Trials with low risk of bias compared to trials with high risk of bias.

  • Type of patients, regarding previous antivirals, naives, relapsers, and non-responders as three separate groups, e.g., naives compared to relapsers.

  • Type of patients, regarding genotype: genotype 1 compared to genotype non-1.

  • Type of patients, regarding degree of liver disease (inflammation score (grading) or fibrosis score (staging)).

  • Type of patients, regarding HIV or hepatitis B co-infection compared to patients without co-infection.

  • Type of patients, regarding age: children compared to adults.

  • Intervention: according to the type, dose, and duration of aminoadamantanes and other antiviral drugs.

We compared subgroups with a test of interaction (Altman 2003).

Sensitivity analysis

We identified suitable sensitivity analyses during the review process. For example, we used a sensitivity analysis when imputing missing data with replacement values.

Data analysis in the included trials was according to the intention-to-treat principle as well as 'as treated' (per protocol) analysis.

Summary of findings

We created a 'Summary of findings' table, presenting the results of our review outcomes (GRADEpro).

Results

Description of studies

See: Characteristics of included studies; Characteristics of excluded studies.

Results of the search

Our search strategy identified 639 publications of potential interest. After filtering for duplicates 290 publications remained. Of the remaining 290 publications, we excluded 214 after screening the title and abstract, among other reasons because they were reviews or because they did not describe a randomised clinical trial investigating the effect of aminoadamantanes in patients with chronic hepatitis C. The remaining 76 references described 41 unique randomised clinical trials (Figure 1).

Figure 1.

Flow diagram.

Twenty-five of the included trials were published in more than one publication. Six out of 41 randomised clinical trials were published as abstracts only (Cornberg 2000; Shakil 2000; Jorge 2001; Vardar 2001; Teuber 2002; Calay 2005).

When necessary, we contacted the primary or last authors for further information and data relating to the trials.

We searched for ongoing trials in the WHO International Clinical Trials Registry Platform (www.who.int/ictrp), Google Scholar, and Eudrapharm, but we did not identify any registered ongoing or planned trials.

Included studies

The included trials were 41 in total. Thirteen trials were conducted in Italy (Brillanti 1999; Brillanti 2000; Gaeta 2001; Mangia 2001; Tabone 2001; Bacosi 2002; Adinolfi 2003; Baisini 2003; Piai 2003; Angelico 2004; Ciancio 2006; Gramenzi 2007; Angelico 2008), seven trials were conducted in Germany (Cornberg 2000; Zeuzem 2000; Teuber 2001; Teuber 2002; Berg 2003; Teuber 2003; von Wagner 2008), four trials were conducted in the USA (Shakil 2000; Smith 2004; Thuluvath 2004; Herrine 2005), three trials were conducted in Switzerland (Sax 2001; Helbling 2002; Wenger 2007), and two trials were conducted both in France and the UK (Caronia 2001; Caronia 2001a; Calay 2005; Maynard 2006). Other trials were conducted in each of the following different countries: Argentina, Austria, Belgium, Brazil, Kuwait, Mexico, The Netherlands, Spain, Taiwan, and Turkey (see Characteristics of included studies).

The first of the included trials was published in 1999 (Brillanti 1999) and the last in 2012 (Pessoa 2012). Thirty-six trials had a parallel-group design with two intervention groups. Two trials included three intervention groups (Bacosi 2002; Gramenzi 2007) and two trials included four intervention groups (Herrine 2005; Salmeron 2007). One trial had a cross-over group design (Smith 2004).

A total of 6193 patients with chronic hepatitis C were randomised to an amantadine arm or a control arm in the 41 clinical trials.

Only one trial compared amantadine monotherapy with placebo without additional antiviral drugs (Smith 2004). Seventeen trials compared amantadine plus interferon-alpha versus placebo or no intervention plus interferon-alpha (Characteristics of included studies). One out of these 17 trials also compared amantadine plus interferon-alpha and ribavirin versus placebo or no intervention plus interferon-alpha and ribavirin (Salmeron 2007). Eleven trials reported on the comparison of amantadine plus interferon-alpha plus ribavirin versus placebo or no intervention plus interferon-alpha plus ribavirin (Characteristics of included studies). Twelve trials compared amantadine plus peg interferon-alpha plus ribavirin versus placebo or no intervention plus peg interferon-alpha plus ribavirin (Characteristics of included studies).

The amantadine dose was the same in each trial: 200 mg daily, except for one trial, which prescribed 400 mg per day (von Wagner 2008). The treatment duration of the trials varied from 6 to 12 months. A six-month post-treatment duration of follow-up was used in all trials, except for four trials which applied 12 months of post-treatment follow-up (Bacosi 2002; Adinolfi 2003; Yang 2003; van Soest 2010) and one trial which applied 18 months of post-treatment follow-up (Ciancio 2006). The details are displayed in Table 1.

Table 1. Summary of characteristics of the included trials
Trial Risk of bias Trial duration (months) Follow-up duration (months)
Amantadine versus placebo
Smith 2004Lower66
 
Amantadine plus interferon versus placebo or no intervention plus interferon
Angelico 2004High126
Bacosi 2002High1212
Baisini 2003High126
Caronia 2001High126
Caronia 2001aHigh126
Gaeta 2001High66
Helbling 2002High126
Jorge 2001High126
Mangia 2001High126
Salmeron 2007High126
Sax 2001High126
Shakil 2000High66
Tabone 2001High126
Teuber 2001High126
Vardar 2001High66
Yang 2003High612
Zeuzem 2000High126
 
Amantadine plus interferon plus ribavirin versus placebo or no intervention plus interferon plus ribavirin
Adinolfi 2003High1212
Berg 2003Lower126
Brillanti 1999High66
Brillanti 2000High126
Cornberg 2000High126
Gramenzi 2007High126
Piai 2003High126
Salmeron 2007High126
Teuber 2002High126
Teuber 2003High126
Thuluvath 2004High126
Wenger 2007High126
 
Amantadine plus peg interferon plus ribavirin versus placebo or no intervention plus peg interferon plus ribavirin
Angelico 2008High126
Calay 2005High126
Ciancio 2006High1218
Ferenci 2006High126
Hasan 2004High126
Herrine 2005High126
Langlet 2009High6/126
Maynard 2006High126
Mendez-Navarro 2010High126
Pessoa 2012High126
van Soest 2010High1212
von Wagner 2008High126

None of the trials compared one amantadine dose versus another. None of the trials compared head-to-head long-duration amantadine versus short-duration amantadine.

From the publications which reported the sex of the participants, more than 63% were males. All trials included adult patients, except for one trial which included children of one year or older (Smith 2004). Only one trial included HIV co-infected patients (Sax 2001). None of the trials included patients co-infected with hepatitis B.

Excluded studies

The excluded studies are listed under Characteristics of excluded studies and the reasons for exclusion are given there.

Risk of bias in included studies

We assessed risk of bias according to seven domains: random sequence generation; allocation concealment; blinding of participants and personnel; blinding of outcome assessment; handling of incomplete outcome data; selective outcome reporting; and vested interest bias. Other potential sources of bias for the individual trial, but not for the meta-analyses of such trials, were: baseline imbalance and early stopping.

We considered all included trials to have high risk of bias. We considered only two out of 41 trials as having lower risk of bias (Berg 2003; Smith 2004). Our statistical analyses are, therefore, based mainly on trials with high risk of bias. None of them had low risk of bias. For details of the judgements made for the individual trials, please see Figure 2 and Figure 3.

Figure 2.

'Risk of bias' graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.

Figure 3.

'Risk of bias' summary: review authors' judgements about each risk of bias item for each included study.

Allocation

The generation of the allocation sequence was adequately described in 20 trials (Characteristics of included studies). The remaining 21 trials were described as randomised but the method of random sequence generation was not described (Characteristics of included studies).

The method used to conceal allocation was adequately described in 14 trials (Characteristics of included studies). We judged the method for allocation concealment as unclear in 25 trials (Characteristics of included studies) and as high risk of bias in two trials (Caronia 2001; Caronia 2001a).

Blinding

The method of blinding of participants and personnel was adequately described in only eight trials (Zeuzem 2000; Teuber 2001; Helbling 2002; Berg 2003; Smith 2004; Thuluvath 2004; Ferenci 2006; van Soest 2010). We considered 33 trials as high risk of bias regarding blinding of participants and personnel (Characteristics of included studies). Three trials adequately described the method of blinding of outcome assessment (Caronia 2001a; Berg 2003; Smith 2004). Thus, the other 38 trials had high risk of bias (Characteristics of included studies). Only two trials had low risk of bias, with both blinding of participants and personnel and blinding of outcome assessments (Berg 2003; Smith 2004).

Incomplete outcome data

Incomplete data were addressed adequately in 15 trials (Brillanti 1999; Brillanti 2000; Cornberg 2000; Zeuzem 2000; Caronia 2001; Gaeta 2001; Mangia 2001; Sax 2001; Tabone 2001; Teuber 2001; Piai 2003; Yang 2003; Wenger 2007; Mendez-Navarro 2010; Pessoa 2012). In 26 trials there were risks of incomplete outcome data (Characteristics of included studies).

Selective reporting

Predefined, clinically relevant and reasonably expected primary and secondary outcomes were adequately assessed in only six clinical trials (Brillanti 2000; Zeuzem 2000; Teuber 2001; Berg 2003; Hasan 2004; Maynard 2006). Accordingly, there were risks of selective reporting of outcomes in 35 trials (Characteristics of included studies).

Other potential sources of bias

Five trials did not receive funding and were at low risk of bias regarding vested interests (Mangia 2001; Sax 2001; Tabone 2001; Teuber 2003; Ciancio 2006). Seventeen trials received funding from the medical industry. It was unclear whether trials received funding from the medical industry in 19 trials. We considered these last 36 trials as having high risk of bias because industrial sponsorship could introduce bias.

There were no baseline differences in any of the trials, except for two in which baseline imbalance was unknown (Sax 2001; Piai 2003). One trial stopped early due to poor results (Salmeron 2007).

Effects of interventions

See: Summary of findings for the main comparison

Amantadine versus placebo or no intervention

Primary outcomes
All-cause mortality or liver-related morbidity (composite outcome)

Thirty-two trials provided information on all-cause mortality or liver-related morbidity and could be included in the analyses. These 32 trials included one trial comparing amantadine versus placebo (Smith 2004), 14 trials comparing amantadine plus interferon-alpha versus placebo or no intervention plus interferon-alpha, 10 trials comparing amantadine plus interferon-alpha and ribavirin versus placebo or no intervention plus interferon-alpha and ribavirin, and eight trials comparing amantadine plus peg interferon-alpha and ribavirin versus placebo or no intervention plus peg interferon-alpha and ribavirin. It should be noted that one trial included four treatment groups and was part of two treatment subgroups (Salmeron 2007) (Analysis 1.1). The included trials reported five deaths or liver-related morbidities in 2353 (0.2%) participants in the amantadine group versus six out of 2264 (0.3%) patients in the control group (Analysis 1.1). Meta-analyses with both the fixed-effect model and random-effects model showed no significant effect of amantadine, when compared with placebo or no intervention, on all-cause mortality or liver-related morbidity (fixed-effect model: risk ratio (RR) 0.90, 95% confidence interval (CI) 0.38 to 2.17; I² = 0%) (Analysis 1.1).

The subgroup analyses stratifying the trials according to risk of bias and according to previous treatment and treatment response with antivirals (for example, naive or non-responder patients) did not reveal any significant subgroup differences in effect estimates for the risk of all-cause mortality or liver-related morbidity (Analysis 2.1; Analysis 4.1).

Inspection of the funnel plot did not suggest bias (Figure 4).

Figure 4.

Funnel plot of comparison: 4 Subgroup: trials at lower risk versus high risk of bias. Outcome: 4.1 All-cause mortality or liver-related morbidity.

Fourteen trials provided information on all-cause mortality or liver-related morbidity in patients treated with amantadine plus interferon-alpha versus placebo or no intervention plus interferon-alpha. In the amantadine group, two out of 813 (0.2%) patients died or experienced a liver-related morbidity and in the control group two out of 758 (0.3%) patients died or had a liver-related morbidity. Meta-analysis showed no significant effect of amantadine plus interferon-alpha when compared to placebo or no intervention plus interferon-alpha (fixed-effect model: RR 1.01, 95% CI 0.26 to 3.98; I² = 0%) (Analysis 1.1).

Zero deaths or liver-related morbidities were reported in 10 trials which conducted treatment with amantadine plus interferon-alpha and ribavirin compared with placebo or no intervention plus interferon-alpha and ribavirin (Analysis 1.1).

Eight trials reported on all-cause mortality or liver-related morbidity in patients treated with amantadine plus peg interferon-alpha and ribavirin versus placebo or no intervention plus peg interferon-alpha and ribavirin (Analysis 1.1). Three out of 933 (0.3%) patients treated with amantadine plus peg interferon-alpha and ribavirin, and three out of 897 (0.3%) patients treated with placebo or no intervention plus peg interferon-alpha and ribavirin, died or experienced liver-related morbidities. The risk ratio for this event was statistically non-significant when comparing amantadine plus peg interferon-alpha and ribavirin with placebo or no intervention plus peg interferon-alpha and ribavirin (fixed-effect model: RR 0.97, 95% CI 0.28 to 3.39; I² = 0%) (Analysis 1.1).

We considered only two trials as lower risk of bias, therefore we deemed it unnecessary to perform trial sequential analysis with these two trials only. Consequently, we performed trial sequential analysis on all included trials that reported on the composite outcome 'all-cause mortality or liver-related morbidity'. Due to lack of accurate reporting on all-cause mortality and liver-related morbidity in a number of trials, we were not able to gather enough information to support or refute the effect of amantadine on all-cause mortality or liver-related morbidity (see Figure 5).

Figure 5.

Trial sequential analysis of the random-effects meta-analysis of the effect of amantadine versus placebo or no intervention on all-cause mortality or liver-related morbidity in patients with chronic hepatitis C infection. The trial sequential analysis is performed with a type 1 error of 5% (two-sided), a power of 80%, an assumed control proportion of death or liver-related morbidity of 2%, and an anticipated relative risk reduction (RRR) of 20%. The diversity-adjusted required information size (DARIS) to detect or reject a RRR of 20%, with a between-trial heterogeneity of 0%, is estimated at 34,685 participants. The number of participants actually accrued is 2196, which is only 6% of the required information size. The blue cumulative Z-curve does not cross the red trial sequential monitoring boundaries for benefit or harm. Therefore, there is no evidence to support or refute the assumption that amantadine influences all-cause mortality or liver-related morbidity. The cumulative Z-curve does not reach the futility area (which is not even drawn by the program), demonstrating that further randomised trials may be needed.

Adverse events

We classified adverse events into two groups: number of patients with serious adverse events or number of patients with treatment discontinuation due to any adverse event.

Two-hundred and eighty-eight patients out of 2869 (10.0%) in the amantadine group with or without additional therapy versus 293 patients out of 2777 (10.6%) in the control placebo or no intervention group with or without additional therapy were reported to have either serious adverse events or treatment discontinuation due to any adverse event (Analysis 1.2).

The risk ratio for both events as a composite outcome was statistically non-significant when comparing amantadine with or without additional intervention versus placebo or no intervention with or without the same additional intervention (fixed-effect model: RR 0.98, 95% CI 0.84 to 1.14; I² = 0%; 5646 participants, 35 trials) (Analysis 1.2).

As there were no trials with low risk of bias, we performed trial sequential analysis on all included trials reporting on adverse events. Trial sequential analysis of these data supports the statistically non-significant finding (Figure 6).

Figure 6.

Trial sequential analysis of the random-effects meta-analysis of the effect of amantadine versus placebo or no intervention, in chronic hepatitis C-infected patients, on the number of patients experiencing a serious adverse event or the number of patients who had to discontinue treatment due to an adverse event. The trial sequential analysis is performed with a type 1 error of 5% (two-sided), a power of 80%, an assumed control proportion of number of patients experiencing a serious adverse event or who had to discontinue treatment due to an adverse event of 10%, and an anticipated relative risk reduction (RRR) of 20%. The diversity-adjusted required information size (DARIS) to detect or reject a RRR of 20%, with a between-trial heterogeneity of 0%, is estimated at 5787 participants. The number of participants actually accrued is 5272, which is 91% of the required information size. The blue cumulative Z-curve does not cross the red trial sequential monitoring boundaries for benefit or harm. Therefore, there is no evidence to support the assumption amantadine influences the number of patients experiencing a serious adverse event or who have to discontinue treatment due to an adverse event. The cumulative Z-curve does cross the trial sequential beta-spending monitoring boundaries and reach the futility area, demonstrating that no further randomised trials may be needed.

Inspection of the funnel plot did not suggest bias (Figure 7).

Figure 7.

Funnel plot of comparison: 4 Subgroup: trials at lower risk versus high risk of bias. Outcome: 4.2 Adverse events.

Quality of life

Only six trials reported on quality of life (Zeuzem 2000; Teuber 2001; Helbling 2002; Berg 2003; Smith 2004; Ferenci 2006). Three trials applied the 'Profile of Mood Status' scale (POMS) and the 'Everyday Life' questionnaire (EDLQ) (Zeuzem 2000; Teuber 2001; Berg 2003). The other three trials used a health-related quality of life (HRQoL) score (Ferenci 2006), VAS score (Helbling 2002), or the McMaster Quality of Life Survey (Smith 2004). We were not able to perform meta-analyses on quality of life due to a lack of valid data. Overall, we found no significant differences between treatment with amantadine when compared with placebo or no intervention in each separate trial.

Secondary outcomes
Failure of serum (or plasma) sustained virological response (sustained virological response)

Thirty-five trials provided information on patients who failed to achieve a sustained virological response. In the amantadine group, 1821 out of 2861 (63.6%) patients did not achieve a sustained virological response versus 1737 out of 2721 (63.8%) patients in the control group. Meta-analyses with both the fixed-effect model and random-effects model showed no significant effect of amantadine on failure to achieve a sustained virological response (fixed-effect model: RR 0.98, 95% CI 0.95 to 1.02; I² = 35%) (Analysis 1.4).

Thirteen trials reported on failure to achieve a sustained virological response in patients treated with amantadine plus interferon-alpha versus placebo or no intervention plus interferon-alpha (Analysis 1.4). Five-hundred and sixty-four patients failed to achieve a sustained virological response out of 687 patients (82.1%) in the amantadine group versus 514 patients out of 626 patients (82.1%) in the control group. Meta-analysis showed no significant effect of amantadine plus interferon-alpha compared with placebo or no intervention plus interferon-alpha (fixed-effect model: RR 0.99, 95% CI 0.94 to 1.04; I² = 37%) (Analysis 1.4).

Eleven trials provided information on failure to achieve a sustained virological response in patients treated with amantadine plus interferon-alpha and ribavirin versus placebo or no intervention plus interferon-alpha and ribavirin (Analysis 1.4). Four-hundred and twenty-two patients failed to achieve a sustained virological response out of 666 (63.4%) patients in the amantadine group versus 447 out of 628 (71.2%) patients in the control group. Meta-analysis with both the fixed-effect model and random-effects model showed a significant effect of amantadine plus interferon-alpha and ribavirin when compared with placebo or no intervention plus interferon-alpha and ribavirin (fixed-effect model: RR 0.89, 95% CI 0.83 to 0.96; I² = 41%) (Analysis 1.4).

We analysed the missing data using a best-worst case scenario (assuming that participants receiving amantadine with an unknown status for achieving a sustained virological response did achieve this and that all participants from the control group with an unknown status for achieving a sustained virological response did not). This reveals a statistically significant effect favouring amantadine in patients treated with amantadine plus interferon-alpha and ribavirin (best-worst case scenario: RR 0.69, 95% CI 0.56 to 0.85; 1294 participants, 11 trials) (Analysis 5.1). We also analysed the missing data using a worst-best case scenario (assuming that participants receiving amantadine with an unknown status for achieving a sustained virological response did not achieve this and that all participants from the control group with an unknown status for achieving this did). This analysis shows no significant differences (worst-best case scenario: RR 1.02, 95% CI 0.81 to 1.29; 1294 participants, 11 trials) (Analysis 5.1).

In 12 trials, 2975 patients were treated with amantadine plus peg interferon-alpha and ribavirin versus placebo or no intervention plus peg interferon-alpha and ribavirin (Analysis 1.4). Eight-hundred and thirty-five out of 1508 patients (55.4%) treated in the amantadine group compared with 776 out of 1467 patients (52.9%) in the control group failed to achieve a sustained virological response. The risk ratio for this event was statistically non-significant when comparing amantadine plus peg interferon-alpha and ribavirin therapy with placebo or no intervention plus peg interferon-alpha and ribavirin (fixed-effect model: RR 1.04, 95% CI 0.97 to 1.10; I² = 3%) (Analysis 1.4).

The subgroup analyses, stratifying the trials according to risk of bias, revealed no statistically significant differences in the risk ratio for failure to achieve a sustained virological response, with both the fixed-effect model and the random-effects model, when comparing trials with lower risk of bias (RR 0.85, 95% CI 0.70 to 1.03; 400 participants; one trial) to trials with high risk of bias (fixed-effect model: RR 1.00, 95% CI 0.96 to 1.03; 5182 participants, 35 trials) (Analysis 4.3).

Meta-analysis with both a fixed-effect model and random-effects model resulted in no significant difference in the effect estimates for the risk of failure to achieve a sustained virological response in the subgroup analysis of trials including genotype 1 patients (RR 1.00, 95% CI 0.94 to 1.06, I² = 6%) compared to trials including non-genotype 1 patients (RR 0.98, 95% CI 0.82 to 1.18; I² = 0%) (Analysis 3.1).

Lastly, subgroup analyses, stratifying the trials according to previous antiviral therapy, showed no statistically significant differences with both the fixed-effect model and the random-effects model (Analysis 2.3). Subgroup analyses regarding degree of liver disease, HIV or hepatitis B co-infection, and age could not be performed due to lack of data.

Inspection of the funnel plot did indicate bias (Figure 8).

Figure 8.

Funnel plot of comparison: 4 Subgroup: trials at lower risk versus high risk of bias. Outcome: 4.3 Failure of sustained virological response.

We performed trial sequential analysis on all trials, because we considered only two trials as lower risk of bias trials. Trial sequential analysis of the combined data supports the finding of no effect of amantadine, when compared with placebo or no intervention, on failure to achieve a sustained virological response (Figure 9). The result of the trial sequential analysis is shown by the cumulated Z-curve (blue curve), which does not cross the trial sequential alpha spending monitoring boundary (red inward sloping curve) and ends up in the futility area. This implies that there is no evidence for a beneficial effect of amantadine in preventing failure to achieve a sustained virological response.

Figure 9.

Trial sequential analysis of the random-effects meta-analysis of the effect of amantadine versus placebo or no intervention on the number of patients with chronic hepatitis C virus infection who failed to achieve a sustained virological response (SVR). The trial sequential analysis is performed with a type 1 error of 5% (two-sided), a power of 90%, an assumed control proportion of number of patients who failed to achieve a SVR of 64%, and an anticipated relative risk reduction (RRR) of 7%. The diversity-adjusted required information size (DARIS) to detect or reject a RRR of 7%, with a between-trial heterogeneity of 35%, is estimated at 7609 participants. The number of participants actually accrued is 5328, which is 70% of the required information size. The blue cumulative Z-curve does not cross the red trial sequential monitoring boundaries for benefit or harm. Therefore, there is no evidence to support the assumption that amantadine influences the number of patients who fail to achieve a SVR and it is likely that a 7% RRR in the number of patients who fail to achieve a SVR can be rejected with the chosen error risks. The cumulative Z-curve does reach the futility area, demonstrating that no further randomised trials may be needed.

We also performed trial sequential analysis on a subgroup, comparing failure to achieve a sustained virological response in patients treated with amantadine plus interferon-alpha and ribavirin with patients treated with interferon-alpha and ribavirin (Figure 10). There is no evidence to support or refute the assumption that amantadine influences the number of patients who fail to achieve a sustained virological response.

Figure 10.

Trial sequential analysis of the random-effects subgroup meta-analysis of the effect of amantadine plus interferon-alpha and ribavirin versus placebo or no intervention plus interferon-alpha and ribavirin on the number of patients with chronic hepatitis C virus infection who failed to achieve a sustained virological response (SVR). The trial sequential analysis is performed with a type 1 error of 5% (two-sided), a power of 90%, an assumed control proportion of number of patients who failed to achieve a SVR response of 71%, and an anticipated relative risk reduction (RRR) of 7%. The diversity-adjusted required information size (DARIS) to detect or reject a RRR of 7%, with a between-trial heterogeneity of 12%, is estimated at 4171 participants. The number of participants actually accrued is 1294, which is only 31% of the required information size. The blue cumulative Z-curve does not cross the red inward sloping trial sequential alpha-spending monitoring boundaries for benefit or harm. Therefore, there is no evidence to support the assumption that amantadine influences number of patients who fail to achieve a SVR and it is likely that a 7% RRR in the number of patients who fail to achieve a SVR on treatment with amantadine plus interferon-alpha and ribavirin can be rejected with the chosen error risks. The cumulative Z-curve does not reach the futility area (which is not even drawn by the program), demonstrating that further randomised trials may be needed.

Failure of end of treatment virological response

Thirty trials provided information on patients who failed to achieve an end of treatment virological response and could be included in the analyses (Analysis 1.3). In the amantadine group, 1288 out of 2483 patients (51.9%) did not achieve an end of treatment virological response versus 1268 out of 2378 patients (53.3%) in the control group. Meta-analyses with both the fixed-effect model and the random-effects model showed no significant effect of amantadine on achieving an end of treatment virological response (fixed-effect model: RR 0.95, 95% CI 0.90 to 1.00; I² = 43%) (Analysis 1.3).

Ten trials provided information on failure to achieve an end of treatment virological response in patients treated with amantadine plus interferon-alpha and ribavirin versus placebo or no intervention plus interferon-alpha and ribavirin (Analysis 1.3). Three-hundred and forty-nine patients failed to achieve an end of treatment virological response out of 625 patients (39.8%) in the amantadine group versus 386 out of 594 patients (65.0%) in the control group. Meta-analysis with both the fixed-effect model and the random-effects model showed a significant effect of amantadine plus interferon-alpha and ribavirin compared with placebo or no intervention plus interferon-alpha and ribavirin (fixed-effect model: RR 0.86, 95% CI 0.79 to 0.94; I² = 66%) (Analysis 1.3).

We analysed the data in a best-worst case scenario regarding missing data (assuming that participants with an unknown status for achieving an end of treatment virological response receiving amantadine did achieve this and that all participants from the control group with an unknown status for achieving an end of treatment virological response did not). This reveals a stronger positive statistical effect estimate favouring amantadine in patients treated with amantadine plus interferon-alpha and ribavirin (RR 0.58, 95% CI 0.52 to 0.65; 1219 participants, 10 trials) (Analysis 5.2). We also analysed the data in a worst-best case scenario regarding missing data (assuming that participants with an unknown status for an achieving end of treatment virological response receiving amantadine did not achieve this and that all participants from the control group with an unknown status for achieving an end of treatment virological response did). This reveals an effect favouring the control (worst-best case scenario: RR 1.20, 95% CI 1.08 to 1.34; 1219 participants, 10 trials) (Analysis 5.2).

Failure of histological response

Only three trials provided information on the number of patients without improvement of histology (Shakil 2000; Zeuzem 2000; Baisini 2003) (Analysis 1.5). They included 24, 93, and 119 patients. Only 74 out of these 219 patients underwent a liver biopsy before treatment and after treatment. We cannot meta-analyse or draw any conclusions from these data.

Number of patients without normalisation of serum ALT and/or AST levels at end of treatment and at end of follow-up

All trials that reported on biochemical response reported ALT levels only. Therefore, we have chosen only to provide ALT levels in this analysis.

Nineteen trials provided information on failure to achieve end of treatment biochemical response. In the amantadine group, 671 out of 1141 (58.8%) patients did not achieve end of treatment biochemical response versus 732 out of 1100 (66.5%) patients in the control group. Meta-analyses with both the fixed-effect model and random-effects model showed that amantadine significantly decreases the number of patients without normalisation of ALT serum levels at the end of treatment compared with placebo or no intervention (fixed-effect model: RR 0.88, 95% CI 0.83 to 0.94; I² = 47%) (Analysis 1.6).

In seven trials, 207 out of 418 (49.5%) patients treated with amantadine plus interferon-alpha and ribavirin compared with 247 out of 390 (63.3%) patients in the control group treated with placebo or no intervention plus interferon-alpha and ribavirin failed to achieve an end of treatment biochemical response (Analysis 1.6). Meta-analysis with both the fixed-effect model and the random-effects model showed a significant effect of amantadine plus interferon-alpha and ribavirin compared with placebo or no intervention plus interferon-alpha and ribavirin (fixed-effect model: RR 0.79, 95% CI 0.70 to 0.89; I² = 70%) (Analysis 1.6).

Furthermore, 21 trials provided information on patients who failed to achieve an end of follow-up biochemical response and could be included in the analyses (Analysis 1.7). In the amantadine group, 1133 out of 1896 (59.8%) patients did not achieve an end of follow-up biochemical response versus 1151 out of 1848 (62.3%) patients in the control group. Meta-analyses with both models showed no significant effect of amantadine on achieving an end of follow-up biochemical response (fixed-effect model: RR 0.95, 95% CI 0.91 to 1.00; I² = 49%) (Analysis 1.7).

Summary of findings

The Grading of Recommendations Assessment, Development, and Evaluation (GRADE) 'Summary of findings' table (Guyatt 2008) is shown in Summary of findings for the main comparison. We considered all outcomes for the 'Summary of findings' table except failure of histological response, due to lack of data.

Discussion

Summary of main results

We included 41 trials with a total of 6193 patients, which assessed the benefit and harm of amantadine when compared with placebo or no intervention in the treatment of patients with chronic hepatitis C. The effect of amantadine was evaluated in four different treatment strategies: monotherapy with amantadine, combination therapy of amantadine with interferon-alpha, combination therapy of amantadine plus interferon-alpha and ribavirin, and combination therapy of amantadine plus peg interferon-alpha and ribavirin. We carried out subgroup analyses according to a classification based on whether a patient had already received treatment for hepatitis C before and if so which treatment he/she had received, e.g., naive patients, relapsers, or non-responders. The present systematic review did not demonstrate any benefit of amantadine on all-cause mortality or liver-related morbidity for any of these treatment regimens or types of patients.

Our systematic review also showed that concomitant use of amantadine in the treatment of chronic hepatitis C is not associated with either an increase or a reduction in adverse events, defined as the number of patients who experienced a serious adverse event or had to discontinue treatment due to an adverse event. We confirmed these results by applying trial sequential analysis.

Moreover, amantadine did not decrease the overall proportion of patients who failed to achieve a sustained virological response. This finding was confirmed by a trial sequential analysis. However, in subgroup analysis we demonstrated that patients treated with a combination therapy of amantadine plus interferon-alpha and ribavirin had statistically significant less failure in achieving a sustained virological response. However, trial sequential analysis could not exclude risks of random errors (play of chance) and all trials had risks of systematic errors (bias). When applying further subgroup analysis with both the fixed-effect model and the random-effects model, stratifying trials according to previous antiviral therapy or genotype, there were no significant differences in the effect estimates for the risk of failure to achieve a sustained virological response.

Unfortunately, we were not able to identify any convincing benefits of amantadine when assessing histology, because only three trials reported failure of histological improvement. For quality of life, we also could not identify any convincing benefits because only six trials reported this outcome. We found a significant benefit of adding amantadine to interferon-alpha-based therapy for biochemical response at the end of treatment, but not for end of follow-up response.

Overall completeness and applicability of evidence

This systematic review examined the evidence from 41 randomised clinical trials on the treatment of hepatitis C. Despite efforts to obtain additional information from the authors we could not obtain all relevant data, hence not all trials reported on all of our predefined outcomes.

Thirty-two trials reported adequately on our primary outcomes of all-cause mortality or liver-related morbidity, and 35 trials reported on serious adverse events and treatment discontinuation due to an adverse event. Only six trials provided information on quality of life. Thirty-six trials reported on our first secondary outcome measure: failure to achieve a sustained virological response. Twenty-nine trials reported on failure of end of treatment virological response. Only two trials provided information on failure of histological improvement, another 17 trials reported on failure of biochemical response at the end of treatment, and 19 trials reported on failure of biochemical response at the end of follow-up.

It is questionable whether the included patients are representative of current practice. All trials included patients with positive serum hepatitis C RNA. However, there was heterogeneity among the trials due to the different disease severity of the patients at trial entry, differences in genotype (35 trials included a mixture of genotypes), and differences regarding previous antiviral treatment. Concerning sex and age, the trials seem representative of current clinical care: more than 63% of the included patients were male and all included adult patients, except for one trial which included children of one year or older (Smith 2004). However, only one trial included HIV co-infected patients (Sax 2001). None of the trials included patients co-infected with hepatitis B. There are therefore insufficient data on co-infected patients.

Quality of the evidence

We conducted this review according to theCochrane Handbook for Systematic Reviews of Interventions (Higgins 2011) and the Cochrane Hepato-Biliary Group Module (Gluud 2014). The results of our meta-analysis, however, are only as strong as the primary trials included.

The main limitation in the design and implementation of trials was the lack of clarity about the generation of the allocation sequence, concealment of allocation, and blinding. Of the 41 included trials, only 20 (49%) reported adequate allocation sequence generation, only 14 (35%) adequately reported allocation concealment, and only two (5%) reported blinding. Fifteen trials (37%) adequately addressed incomplete data, but only six trials (15%) reported all clinically relevant and reasonably expected outcomes. Also, only five trials (12%) appeared to be free of other components that could put them at risk of bias. Accordingly, 95% of trials were at high risk of bias. It is surprising to see that so many trials had high risk of bias, despite the repeated calls for improved trial quality both within and outside hepatology (Schultz 1995; Gluud 1998; Kjaergard 1999; Needleman 1999; Kjaergard 2001; Wood 2008; Savović 2012; Savović 2012a).

Regarding the precision of our results, some outcomes in our meta-analysis include few patients and few events, and thus have wide confidence intervals around the estimate of effect.

Potential biases in the review process

In this systematic review we performed a comprehensive literature search. As far as we know, we have found all the evidence available. A potential limitation of our literature search may be that we have not specifically searched for trials in the grey literature, which may have introduced a slight risk of bias into our meta-analysis (Egger 2003). However, this bias is unlikely to influence our results in a beneficial way as trials found in the grey literature rarely report beneficial effects.

Most of the included trials are of a relatively small size, especially those performed in the early 2000s. This increases the risk of an unrealistic estimate of the intervention effects due to bias (systematic errors) or chance (random errors). Risk of bias is known to have an impact on the estimated intervention effect, with trials at high risk of bias overestimating beneficial intervention effects and underestimating harmful effects (Schultz 1995; Moher 1998; Kjaergard 2001; Wood 2008; Lundh 2012; Savović 2012; Savović 2012a). We divided the analysis for all outcomes into trials with high risk of bias and trials with lower risk of bias trials to reveal any influence of bias on the effect estimates of our outcomes. Of the 41 included trials, only two had lower risk of bias. We did not observe an influence of bias on any of our analyses, but due to there being few trials with lower risk of bias these analyses do not have sufficient power. The estimated intervention effects for all significant beneficial findings may therefore possibly be due to systematic errors.

No statistical signs of publication bias or other bias were observed.

This review pooled data for all-cause mortality or liver-related morbidity from 32 trials involving 4617 patients. We also pooled data for serious adverse events or treatment discontinuation due to an adverse event from 35 trials involving 5646 patients. The median trial duration was 12 months, with a median follow-up of six months (four trials had a follow-up of 12 months, one trial had a follow-up of 18 months). For our primary outcome measure, all-cause mortality or liver-related morbidity, this is not sufficiently long, considering that the estimated median time in which hepatitis C progresses to cirrhosis is 15 years to 50 years (Koretz 1993; Kenny-Walsh 1999; Seeff 2009). Therefore, it is difficult to detect a significant difference in all-cause mortality and liver-related morbidity based on these trials. If aminoadamantanes have an effect on morbidity and mortality one prerequisite would be that they significantly affect virological load. However, we were unable to provide viral data to demonstrate that this was the case.

We used trial sequential analysis to cope with the risk of random error, which is higher when information sizes are small (Wetterslev 2008). Trial sequential analysis of the primary outcomes, all-cause mortality or liver-related morbidity and serious adverse events or treatment discontinuation due to an adverse event, and of the secondary outcome measure, sustained virological response, showed no significant effect estimates when we applied both the random-effects and fixed-effect models in patients treated with amantadine.

Heterogeneity among the trials might be due to differences in dose, duration, and type of interferon-alpha or peg interferon-alpha. Both evaluation of this and long-term follow-up studies could be useful. Also different definitions of non-responders were used in the trials, such as non-responder to previous interferon-alpha therapy alone or non-responder to combination therapy of interferon-alpha with ribavirin. Furthermore, there could be heterogeneity among trials due to the disease severity of patients at entry and differences in genotype, which can both affect the sustained virological response rates. To reflect our concern about heterogeneity, we conducted all analyses using both the fixed-effect model and the random-effects model. We only presented the results of the fixed-effect model if the results of the two models did not differ. We also considered other important and predefined trial-level covariates, including trial risk of bias, genotype distribution, and previous antiviral treatment. Subgroup analyses of other predefined covariates, such as degree of liver disease, could not be performed because of the lack of trials reporting on this outcome.

Lastly, we did not analyse the two amantadine modalities, amantadine hydrochloride and amantadine sulphate.

Agreements and disagreements with other studies or reviews

It is likely that less than 10% of all infected patients will develop end-stage liver disease. Overall, we found that amantadine did not show any benefit for all-cause mortality or liver-related morbidity. Most trials reported on the surrogate outcome measure sustained virological response, but as already mentioned, we do not know whether a sustained virological response results in less mortality or morbidity (Gluud 2007). An observation was that, while those treated with interferon-alpha and ribavirin were allegedly more likely to develop a sustained virological response if amantadine was added, there was no difference in all-cause mortality or liver-related morbidity (although this observation is certainly limited by the short follow-up periods). This is in accordance with a number of findings in patients with chronic hepatitis C showing that a sustained virological response may not be a valid surrogate marker of clinical outcomes for a number of antiviral drugs (Brok 2010; Koretz 2013; Gurusamy 2013; Hauser 2014; Hauser 2014a).

Considering failure to achieve a sustained virological response, we also found that amantadine did not show any benefit, except for in the subgroup patients treated with the combination therapy of amantadine plus interferon-alpha and ribavirin, in which amantadine seems to complement the lack of efficacy of interferon-alpha compared with peg interferon-alpha. However, this finding was not supported by the trial sequential analyses. This result is in accordance with the main findings of a recently published meta-analysis (Chen 2012), which suggests that there is no beneficial effect of adding amantadine to peg interferon-alpha plus ribavirin in naive hepatitis C genotype 1 patients. Our findings are contrary to the main findings of another meta-analyses (Deltenre 2004), which suggested a role for amantadine in non-responder patients. Furthermore, our results are also in contrast with another review, which suggests that there may be a limited role of combination therapy in naive patients (Lim 2005).

We have no evidence from randomised clinical trials on the long-term effects (more than one year) of amantadine on our primary outcomes. Long-term effects would be relevant in particular for outcomes such as all-cause mortality or liver-related morbidity.

Amantadine was generally well tolerated. We observed that amantadine was associated with non-serious adverse events and almost all trials in general reported similar frequencies and severities of adverse events in both amantadine groups versus control groups. This result is in accordance with a recently published Cochrane review of amantadine and rimantadine for influenza A in children and the elderly (Alves Galvão 2012). The result is also somewhat comparable to two other Cochrane reviews. The review of amantadine and rimantadine in influenza A in adults showed significantly more adverse effects in patients receiving amantadine compared with placebo, but no increased risk of serious adverse events (Jefferson 2012). The second review reported on amantadine in Parkinson's disease and found that there is not enough evidence from trials about the effects of amantadine for people with Parkinson's disease, but that adverse events in trials so far have not been severe (Crosby 2009). In our analysis, amantadine was administered with interferon-alpha or peg interferon-alpha with or without ribavirin, except for in one trial. Interferon-alpha-based therapy is typically associated with haematologic complications (i.e., neutropenia, thrombocytopenia), neuropsychiatric complications (i.e., memory and concentration loss, visual disturbances, headaches, depression, irritability), flu-like symptoms, hormonal complications (i.e., hypothyroidism, hyperthyroidism), gastrointestinal complications (i.e., nausea, vomiting, weight loss), and dermatologic complications (i.e., eczema, alopecia). The most well-known adverse effect of ribavirin is dose-dependent haemolytic anaemia but gastrointestinal adverse effects such as nausea are also reported (Chutaputti 2000; Soza 2002; Sulkowski 2004). In conclusion, both interferon-alpha and ribavirin are associated with a variety of adverse events of different severities, which may make it hard to detect less severe adverse events associated with amantadine. We cannot exclude the possibility of less severe adverse events with amantadine, for example gastrointestinal symptoms and insomnia.

Regarding tolerance of amantadine we have to take dosage into consideration. Only one trial used an amantadine dose of more than 200 mg per day (von Wagner 2008). One randomised clinical trial evaluated the safety and toxicity of amantadine in patients with chronic hepatitis C; it also investigated the maximum tolerable dose of amantadine (Smith 2004a). They reported an increase in biochemical response with higher daily doses of amantadine from 200 mg per day up to 500 mg per day in monotherapy. However, no statistically significant difference was found in alanine aminotransferase (ALT) values between those receiving 300 mg and those receiving higher doses of amantadine. Also, increasing the amantadine dose did not result in more patients achieving a sustained virological response, when comparing 200 mg per day with 300 mg to 500 mg per day (Smith 2004a).

Authors' conclusions

Implications for practice

This review shows that there seems to be no significant beneficial effect of amantadine on all-cause mortality or liver-related morbidity composite outcome, or on adverse events in hepatitis C-infected patients; although the timeframe for measuring the composite outcome was insufficient in the included randomised clinical trials. Furthermore, amantadine did not increase the proportion of patients with a sustained virological response. In the absence of convincing evidence of benefit, the use of amantadine is justified in the context of randomised clinical trials assessing the effects of combination therapy with peg interferon-alpha and ribavirin. We found no randomised clinical trials assessing other aminoadamantanes.

Implications for research

Given the results of our analysis, we cannot conclude whether new randomised clinical trials will or will not find any beneficial effect of amantadine on patients' survival in chronic hepatitis C patients. In subgroup analyses we observed that therapy with amantadine plus interferon-alpha and ribavirin compared with interferon-alpha and ribavirin seems to increase the number of patients with a sustained virological response, but this effect was not supported by our trial sequential analysis. We did not observe a similar finding when examining amantadine combined with peg interferon-alpha and ribavirin. Therefore, to prove the former effect, further randomised clinical trials would be required. We found no evidence for other aminoadamantanes. Based on the overall evidence, future trials assessing amantadine, or potentially other aminoadamantanes for patients with chronic hepatitis C, may not show strong benefits. Therefore, it is probably advisable to wait for the results of trials assessing other direct-acting antiviral drugs. Amantadine and other aminoadamantanes should only be used within randomised clinical trials; they do not appear to have a place in usual clinical practice. To our knowledge, no ongoing trials are investigating the effects of amantadine in hepatitis C patients. Any further trials should be designed according to the SPIRIT guidelines (SPIRIT 2013; SPIRIT 2013a), and conducted and reported according to the CONSORT Statement (Schulz 2012).

Acknowledgements

We thank Dimitrinka Nikolova, Bianca Hemmingsen, Maria Skoog, Jane Lindschou, and Luit Penninga from the Cochrane Hepato-Biliary Group and the Copenhagen Trial Unit for helpful discussions and comments. Furthermore, we thank Sarah Louise Klingenberg from the Cochrane Hepato-Biliary Group for helping with the electronic searches. We also thank the peer reviewers for their contribution to the protocol.

Peer reviewers: Sarel F Malan, South Africa; Manuel Romero-Gómez, Spain.
Contact editor: Ronald L Koretz, USA.

Data and analyses

Download statistical data

Comparison 1. Amantadine versus placebo or no intervention
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 All-cause mortality or liver-related morbidity324617Risk Ratio (M-H, Fixed, 95% CI)0.90 [0.38, 2.17]
1.1 Amantadine versus placebo or no intervention1152Risk Ratio (M-H, Fixed, 95% CI)0.36 [0.01, 8.71]
1.2 Amantadine plus interferon-alpha versus placebo or no intervention plus interferon-alpha141571Risk Ratio (M-H, Fixed, 95% CI)1.01 [0.26, 3.98]
1.3 Amantadine plus interferon-alpha and ribavirin versus placebo or no intervention plus interferon-alpha and ribavirin101064Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
1.4 Amantadine plus peg interferon-alpha and ribavirin versus placebo or no intervention plus peg interferon-alpha and ribavirin81830Risk Ratio (M-H, Fixed, 95% CI)0.97 [0.28, 3.39]
2 Adverse events355646Risk Ratio (M-H, Fixed, 95% CI)0.98 [0.84, 1.14]
2.1 Amantadine versus placebo or no intervention1152Risk Ratio (M-H, Fixed, 95% CI)0.22 [0.01, 4.43]
2.2 Amantadine plus interferon-alpha versus placebo or no intervention plus interferon-alpha141571Risk Ratio (M-H, Fixed, 95% CI)1.07 [0.73, 1.56]
2.3 Amantadine plus interferon-alpha and ribavirin versus placebo or no intervention plus interferon-alpha and ribavirin111289Risk Ratio (M-H, Fixed, 95% CI)0.78 [0.61, 1.00]
2.4 Amantadine plus peg interferon-alpha and ribavirin versus placebo or no intervention plus peg interferon-alpha and ribavirin102634Risk Ratio (M-H, Fixed, 95% CI)1.12 [0.90, 1.40]
3 Failure of end of treatment virological response304861Risk Ratio (M-H, Fixed, 95% CI)0.95 [0.90, 1.00]
3.1 Amantadine plus interferon-alpha versus placebo or no intervention plus interferon-alpha111129Risk Ratio (M-H, Fixed, 95% CI)0.94 [0.88, 1.01]
3.2 Amantadine plus interferon-alpha and ribavirin versus placebo or no intervention plus interferon-alpha and ribavirin101219Risk Ratio (M-H, Fixed, 95% CI)0.86 [0.79, 0.94]
3.3 Amantadine plus peg interferon-alpha and ribavirin verus placebo or no intervention plus peg interferon-alpha and ribavirin102513Risk Ratio (M-H, Fixed, 95% CI)1.03 [0.94, 1.13]
4 Failure of sustained virological response355582Risk Ratio (M-H, Fixed, 95% CI)0.98 [0.95, 1.02]
4.1 Amantadine plus interferon-alpha versus placebo or no intervention plus interferon-alpha131313Risk Ratio (M-H, Fixed, 95% CI)0.99 [0.94, 1.04]
4.2 Amantadine plus interferon-alpha and ribavirin versus placebo or no intervention plus interferon-alpha and ribavirin111294Risk Ratio (M-H, Fixed, 95% CI)0.89 [0.83, 0.96]
4.3 Amantadine plus peg interferon-alpha and ribavirin versus placebo or no intervention plus peg interferon-alpha and ribavirin122975Risk Ratio (M-H, Fixed, 95% CI)1.04 [0.97, 1.10]
5 Failure of histological response3236Risk Ratio (M-H, Fixed, 95% CI)1.01 [0.93, 1.09]
5.1 Amantadine plus interferon-alpha versus placebo or no intervention plus interferon-alpha3236Risk Ratio (M-H, Fixed, 95% CI)1.01 [0.93, 1.09]
6 Failure of normalisation of ALT at end of treatment192241Risk Ratio (M-H, Fixed, 95% CI)0.88 [0.83, 0.94]
6.1 Amantadine versus placebo or no intervention1152Risk Ratio (M-H, Fixed, 95% CI)0.82 [0.70, 0.94]
6.2 Amantadine plus interferon-alpha versus placebo or no intervention plus interferon-alpha91018Risk Ratio (M-H, Fixed, 95% CI)0.95 [0.87, 1.04]
6.3 Amantadine plus interferon-alpha and ribavirin versus placebo or no intervention plus interferon-alpha and ribavirin7808Risk Ratio (M-H, Fixed, 95% CI)0.79 [0.70, 0.89]
6.4 Amantadine plus peg interferon-alpha and ribavirin versus placebo or no intervention plus peg interferon-alpha and ribavirin2263Risk Ratio (M-H, Fixed, 95% CI)0.96 [0.78, 1.18]
7 Failure of normalisation of ALT at end of follow-up213744Risk Ratio (M-H, Fixed, 95% CI)0.95 [0.91, 1.00]
7.1 Amantadine plus interferon-alpha versus placebo or no intervention plus interferon-alpha8994Risk Ratio (M-H, Fixed, 95% CI)0.95 [0.88, 1.01]
7.2 Amantadine plus interferon-alpha and ribavirin versus placebo or no intervention plus interferon-alpha and ribavirin6784Risk Ratio (M-H, Fixed, 95% CI)0.82 [0.74, 0.92]
7.3 Amantadine plus peg interferon-alpha and ribavirin versus placebo or no intervention plus peg interferon-alpha and ribavirin71966Risk Ratio (M-H, Fixed, 95% CI)1.03 [0.95, 1.11]
Analysis 1.1.

Comparison 1 Amantadine versus placebo or no intervention, Outcome 1 All-cause mortality or liver-related morbidity.

Analysis 1.2.

Comparison 1 Amantadine versus placebo or no intervention, Outcome 2 Adverse events.

Analysis 1.3.

Comparison 1 Amantadine versus placebo or no intervention, Outcome 3 Failure of end of treatment virological response.

Analysis 1.4.

Comparison 1 Amantadine versus placebo or no intervention, Outcome 4 Failure of sustained virological response.

Analysis 1.5.

Comparison 1 Amantadine versus placebo or no intervention, Outcome 5 Failure of histological response.

Analysis 1.6.

Comparison 1 Amantadine versus placebo or no intervention, Outcome 6 Failure of normalisation of ALT at end of treatment.

Analysis 1.7.

Comparison 1 Amantadine versus placebo or no intervention, Outcome 7 Failure of normalisation of ALT at end of follow-up.

Comparison 2. Subgroup: naives, relapsers, non-responders
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Mortality or liver-related morbidity324617Risk Ratio (M-H, Fixed, 95% CI)0.90 [0.38, 2.17]
1.1 Naives173230Risk Ratio (M-H, Fixed, 95% CI)0.83 [0.27, 2.59]
1.2 Relapsers2101Risk Ratio (M-H, Fixed, 95% CI)0.0 [0.0, 0.0]
1.3 Non-responders121057Risk Ratio (M-H, Fixed, 95% CI)1.00 [0.14, 7.05]
1.4 Trials without differentiation according to previous antiviral treatment2229Risk Ratio (M-H, Fixed, 95% CI)1.05 [0.15, 7.47]
2 Adverse events355646Risk Ratio (M-H, Fixed, 95% CI)0.98 [0.84, 1.14]
2.1 Naives163141Risk Ratio (M-H, Fixed, 95% CI)1.02 [0.85, 1.22]
2.2 Relapsers2101Risk Ratio (M-H, Fixed, 95% CI)3.90 [0.46, 33.30]
2.3 Non-responders141482Risk Ratio (M-H, Fixed, 95% CI)0.77 [0.54, 1.10]
2.4 Trials without differentiation according to previous antiviral treatment4922Risk Ratio (M-H, Fixed, 95% CI)1.05 [0.71, 1.54]
3 Failure of sustained virological response355582Risk Ratio (M-H, Fixed, 95% CI)0.99 [0.95, 1.02]
3.1 Naives173804Risk Ratio (M-H, Fixed, 95% CI)1.01 [0.96, 1.06]
3.2 Relapsers4219Risk Ratio (M-H, Fixed, 95% CI)0.93 [0.71, 1.21]
3.3 Non-responders131412Risk Ratio (M-H, Fixed, 95% CI)0.97 [0.92, 1.02]
3.4 Trials without differentiation according to previous antiviral treatment3147Risk Ratio (M-H, Fixed, 95% CI)0.81 [0.64, 1.03]
4 Failure of end of treatment virological response304861Risk Ratio (M-H, Fixed, 95% CI)0.95 [0.91, 1.00]
4.1 Naives122571Risk Ratio (M-H, Fixed, 95% CI)0.97 [0.90, 1.04]
4.2 Relapsers2101Risk Ratio (M-H, Fixed, 95% CI)0.78 [0.49, 1.23]
4.3 Non-responders131412Risk Ratio (M-H, Fixed, 95% CI)0.95 [0.89, 1.02]
4.4 Trials without differentiation according to previous antiviral treatment4777Risk Ratio (M-H, Fixed, 95% CI)0.91 [0.75, 1.11]
5 Failure of histological response3236Risk Ratio (M-H, Fixed, 95% CI)1.01 [0.93, 1.09]
5.1 Naives3236Risk Ratio (M-H, Fixed, 95% CI)1.01 [0.93, 1.09]
6 Failure of normalisation of ALT at end of treatment192241Risk Ratio (M-H, Fixed, 95% CI)0.88 [0.83, 0.94]
6.1 Naives71246Risk Ratio (M-H, Fixed, 95% CI)0.90 [0.83, 0.99]
6.2 Relapsers124Risk Ratio (M-H, Fixed, 95% CI)0.8 [0.50, 1.28]
6.3 Non-responders9742Risk Ratio (M-H, Fixed, 95% CI)0.87 [0.79, 0.97]
6.4 Trials without differentiation according to previous antiviral treatment2229Risk Ratio (M-H, Fixed, 95% CI)0.82 [0.71, 0.96]
7 Failure of normalisation of ALT at end of follow-up213744Risk Ratio (M-H, Fixed, 95% CI)0.95 [0.91, 1.00]
7.1 Naives82050Risk Ratio (M-H, Fixed, 95% CI)0.95 [0.89, 1.02]
7.2 Relapsers177Risk Ratio (M-H, Fixed, 95% CI)0.97 [0.66, 1.43]
7.3 Non-responders10847Risk Ratio (M-H, Fixed, 95% CI)0.95 [0.88, 1.03]
7.4 Trials without differentiation according to previous antiviral treatment3770Risk Ratio (M-H, Fixed, 95% CI)0.95 [0.81, 1.12]
Analysis 2.1.

Comparison 2 Subgroup: naives, relapsers, non-responders, Outcome 1 Mortality or liver-related morbidity.

Analysis 2.2.

Comparison 2 Subgroup: naives, relapsers, non-responders, Outcome 2 Adverse events.

Analysis 2.3.

Comparison 2 Subgroup: naives, relapsers, non-responders, Outcome 3 Failure of sustained virological response.

Analysis 2.4.

Comparison 2 Subgroup: naives, relapsers, non-responders, Outcome 4 Failure of end of treatment virological response.

Analysis 2.5.

Comparison 2 Subgroup: naives, relapsers, non-responders, Outcome 5 Failure of histological response.

Analysis 2.6.

Comparison 2 Subgroup: naives, relapsers, non-responders, Outcome 6 Failure of normalisation of ALT at end of treatment.

Analysis 2.7.

Comparison 2 Subgroup: naives, relapsers, non-responders, Outcome 7 Failure of normalisation of ALT at end of follow-up.

Comparison 3. Subgroup: genotype 1 compared to genotype non-1
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Failure of sustained virological response355582Risk Ratio (M-H, Fixed, 95% CI)0.99 [0.95, 1.02]
1.1 HCV genotype 1132350Risk Ratio (M-H, Fixed, 95% CI)1.00 [0.94, 1.06]
1.2 HCV genotype non-18731Risk Ratio (M-H, Fixed, 95% CI)0.98 [0.82, 1.18]
1.3 Trials without differentiation according to genotype232501Risk Ratio (M-H, Fixed, 95% CI)0.98 [0.94, 1.02]
Analysis 3.1.

Comparison 3 Subgroup: genotype 1 compared to genotype non-1, Outcome 1 Failure of sustained virological response.

Comparison 4. Subgroup: trials at lower risk of bias compared to trials at high risk of bias
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 All-cause mortality or liver-related morbidity324617Risk Ratio (M-H, Fixed, 95% CI)0.90 [0.38, 2.17]
1.1 Trials with lower risk of bias2552Risk Ratio (M-H, Fixed, 95% CI)0.36 [0.01, 8.71]
1.2 Trials with high risk of bias304065Risk Ratio (M-H, Fixed, 95% CI)0.99 [0.39, 2.49]
2 Adverse events355646Risk Ratio (M-H, Fixed, 95% CI)0.98 [0.84, 1.14]
2.1 Trials with lower risk of bias2552Risk Ratio (M-H, Fixed, 95% CI)0.72 [0.53, 0.99]
2.2 Trials with high risk of bias335094Risk Ratio (M-H, Fixed, 95% CI)1.06 [0.89, 1.26]
3 Failure of sustained virological response355582Risk Ratio (M-H, Fixed, 95% CI)0.99 [0.95, 1.02]
3.1 Trials with lower risk of bias1400Risk Ratio (M-H, Fixed, 95% CI)0.85 [0.70, 1.03]
3.2 Trials with high risk of bias345182Risk Ratio (M-H, Fixed, 95% CI)1.00 [0.96, 1.03]
4 Failure of end of treatment virological response304861Risk Ratio (M-H, Fixed, 95% CI)0.95 [0.91, 1.00]
4.1 Trials with lower risk of bias1400Risk Ratio (M-H, Fixed, 95% CI)0.82 [0.65, 1.02]
4.2 Trials with high risk of bias294461Risk Ratio (M-H, Fixed, 95% CI)0.96 [0.92, 1.01]
5 Failure of histological response3236Risk Ratio (M-H, Fixed, 95% CI)1.01 [0.93, 1.09]
5.1 Trials with high risk of bias3236Risk Ratio (M-H, Fixed, 95% CI)1.01 [0.93, 1.09]
6 Failure of normalisation of ALT at end of treatment192241Risk Ratio (M-H, Fixed, 95% CI)0.88 [0.83, 0.94]
6.1 Trials with lower risk of bias2552Risk Ratio (M-H, Fixed, 95% CI)0.79 [0.68, 0.91]
6.2 Trials with high risk of bias171689Risk Ratio (M-H, Fixed, 95% CI)0.91 [0.85, 0.98]
7 Failure of normalisation of ALT at end of follow-up213744Risk Ratio (M-H, Fixed, 95% CI)0.95 [0.91, 1.00]
7.1 Trials with lower risk of bias1400Risk Ratio (M-H, Fixed, 95% CI)0.81 [0.67, 0.98]
7.2 Trials with high risk of bias203344Risk Ratio (M-H, Fixed, 95% CI)0.97 [0.92, 1.02]
Analysis 4.1.

Comparison 4 Subgroup: trials at lower risk of bias compared to trials at high risk of bias, Outcome 1 All-cause mortality or liver-related morbidity.

Analysis 4.2.

Comparison 4 Subgroup: trials at lower risk of bias compared to trials at high risk of bias, Outcome 2 Adverse events.

Analysis 4.3.

Comparison 4 Subgroup: trials at lower risk of bias compared to trials at high risk of bias, Outcome 3 Failure of sustained virological response.

Analysis 4.4.

Comparison 4 Subgroup: trials at lower risk of bias compared to trials at high risk of bias, Outcome 4 Failure of end of treatment virological response.

Analysis 4.5.

Comparison 4 Subgroup: trials at lower risk of bias compared to trials at high risk of bias, Outcome 5 Failure of histological response.

Analysis 4.6.

Comparison 4 Subgroup: trials at lower risk of bias compared to trials at high risk of bias, Outcome 6 Failure of normalisation of ALT at end of treatment.

Analysis 4.7.

Comparison 4 Subgroup: trials at lower risk of bias compared to trials at high risk of bias, Outcome 7 Failure of normalisation of ALT at end of follow-up.

Comparison 5. Subgroup: sensitivity analysis
Outcome or subgroup titleNo. of studiesNo. of participantsStatistical methodEffect size
1 Failure of sustained virological response112588Risk Ratio (M-H, Random, 95% CI)0.84 [0.71, 0.98]
1.1 Best-worst: amantadine plus interferon-alpha and ribavirin versus placebo or no intervention plus interferon-alpha and ribavirin111294Risk Ratio (M-H, Random, 95% CI)0.69 [0.56, 0.85]
1.2 Worst-best: amantadine plus interferon-alpha and ribavirin versus placebo or no intervention plus interferon-alpha and ribavirin111294Risk Ratio (M-H, Random, 95% CI)1.02 [0.81, 1.29]
2 Failure of end of treatment virological response102438Risk Ratio (M-H, Fixed, 95% CI)0.84 [0.78, 0.91]
2.1 Best-worst: amantadine plus interferon-alpha and ribavirin versus placebo or no intervention plus interferon-alpha and ribavirin101219Risk Ratio (M-H, Fixed, 95% CI)0.58 [0.52, 0.65]
2.2 Worst-best: amantadine plus interferon-alpha and ribavirin versus placebo or no intervention plus interferon-alpha and ribavirin101219Risk Ratio (M-H, Fixed, 95% CI)1.20 [1.08, 1.34]
Analysis 5.1.

Comparison 5 Subgroup: sensitivity analysis, Outcome 1 Failure of sustained virological response.

Analysis 5.2.

Comparison 5 Subgroup: sensitivity analysis, Outcome 2 Failure of end of treatment virological response.

Appendices

Appendix 1. Search strategies

Database Time span Search strategy
Cochrane Hepato-Biliary Group Controlled Trials Register1996 to December 2013(adaman* OR amantadin* OR symmetrel OR symandin* OR rimantadin* OR flumadin* OR methenamin*) AND ('hepatitis C' OR 'hep C' OR HCV)

Cochrane Central Register of Controlled Trials (CENTRAL) CENTRAL 2013, Issue 11 of 12, in The Cochrane Library (Wiley)

 

1995 to Issue 11 of 12, 2013

#1 MeSH descriptor: [Adamantane] explode all trees

#2 adaman* OR amantadin* OR symmetrel OR symandin* OR rimantadin* OR flumadin* OR methenamin*

#3 (#1 OR #2)

#4 MeSH descriptor: [Hepatitis C] explode all trees

#5 hepatitis C OR hep C OR HCV

#6 (#4 OR #5)

#7 (#3 AND #6)

MEDLINE (Ovid SP)1946 to December 2013

1. exp Adamantane/

2. (adaman* or amantadin* or symmetrel or symandin* or rimantadin* or flumadin* or methenamin*).mp. [mp=protocol supplementary concept, rare disease supplementary concept, title, original title, abstract, name of substance word, subject heading word, unique identifier]

3. 1 or 2

4. exp Hepatitis C/

5. (hepatitis C or hep C or HCV).mp. [mp=protocol supplementary concept, rare disease supplementary concept, title, original title, abstract, name of substance word, subject heading word, unique identifier]

6. 4 or 5

7. 3 and 6

8. (random* or blind* or placebo* or meta-analys*).mp. [mp=protocol supplementary concept, rare disease supplementary concept, title, original title, abstract, name of substance word, subject heading word, unique identifier]

9. 7 and 8

EMBASE (Ovid SP)

 

1974 to December 2013

1. exp amantadine/

2. exp rimantadine/

3. (adaman* or amantadin* or symmetrel or symandin* or rimantadin* or flumadin* or methenamin*).mp. [mp=title, abstract, subject headings, heading word, drug trade name, original title, device manufacturer, drug manufacturer, device trade name, keyword]

4. 1 or 2 or 3

5. exp hepatitis C/

6. (hepatitis C or hep C or HCV).mp. [mp=title, abstract, subject headings, heading word, drug trade name, original title, device manufacturer, drug manufacturer, device trade name, keyword]

7. 5 or 6

8. 4 and 7

9. (random* or blind* or placebo* or meta-analys*).mp. [mp=title, abstract, subject headings, heading word, drug trade name, original title, device manufacturer, drug manufacturer, device trade name, keyword]

10. 8 and 9

Science Citation Index Expanded

 

1900 to December 2013

#5 #4 AND #3

#4 TS=(random* or blind* or placebo* or meta-analys*)

#3 #2 AND #1

#2 TS=(hepatitis C or hep C or HCV)

#1 TS=(adaman* or amantadin* or symmetrel or symandin* or rimantadin* or flumadin* or methenamin*)

Contributions of authors

ML, MB, JD, and CG were involved in the study concept and design.
ML and MB screened the literature, selected publications for inclusion and exclusion according to the eligibility criteria, extracted data, and made the 'Risk of bias' judgements.
ML, MB, and CG analysed and interpreted the data and results.
ML drafted the manuscript and performed the meta-analyses.
JD and CG were involved in critical revision of the manuscript for important intellectual content.

Declarations of interest

M.H. Lamers: no declarations of interest.
Mark Broekman: no declarations of interest.
Joost PH Drenth: no declarations of interest.
Christian Gluud: no declarations of interest.

Sources of support

Internal sources

  • Radboud University Medical Center Nijmegen, Netherlands.

External sources

  • The Cochrane Hepato-Biliary Group, Denmark.

    The first author, Mieke H Lamers, worked on the review for three months at the CHBG Editorial Team offices.

Differences between protocol and review

We conducted sensitivity analysis only on the statistically significant findings and only using 'best-worse' case scenario and 'worst-best' case scenario analysis, in order to check the robustness of our analysis. We did not use 'poor outcome analysis' and 'good outcome analysis' and deleted this from our review.

We also assessed the risk of other sources of bias (baseline imbalance bias and early stopping bias) and described this in our Characteristics of included studies table. Both may bias the individual trial, but are unlikely to bias meta-analysis. Therefore, we reported this for the individual trials, but not for our meta-analyses.

We did not contact pharmaceutical companies who are involved in the production and assessment of aminoadamantanes.

We included a 'Summary of findings' table.

Characteristics of studies

Characteristics of included studies [ordered by study ID]

Adinolfi 2003

Methods

Randomised clinical trial in interferon non-responder patients

12 months therapy, 12 months follow-up

Participants

Country: Italy

114 patients were randomised

Inclusion criteria: chronic HCV with presence of serum HCV RNA, serum ALT levels persistently greater than 1.5 times the normal value during the follow-up period, previously received a course of recombinant or lymphoblastoid interferon-alpha 3 to 6 MU 3 times a week for at least 4 months, were considered as non-responders - that is, on no occasion had they had both serum HCV RNA clearance and normalisation of serum transaminase levels, liver biopsy in the 24 months before entering the study

Exclusion criteria: decompensated cirrhosis, cirrhosis with signs of portal hypertension, serum HIV or HBsAg positivity, serum markers of autoimmunity with or without associated disease, alcohol intake, serum haemoglobin concentration < 12 g/dl for women and < 13 g/dl for men, white cell count < 3000 mm³, platelet count < 100,000 mm³, other clinically significant diseases

Amantadine group: 24 patients, median age 50 (30 to 59) years, male/female = 16/8. Median serum ALT 105 (64 to 284) MU/mL, and a median basal viral load of 3.2 (0.8 to 28.6) eq/mL x 10⁶ copies per mL. Genotype 1 (n = 17) and genotype non-1 (n = 7). Histological staging (HAI): median 5.4 (4 to 10), 8 patients cirrhosis

Control group: 46 patients, median age 51 (30 to 60) years, male/female = 31/15. Median serum ALT 98 (62 to 308) MU/mL, and the median basal viral load of 3.0 (0.7 to 18.4) eq/mL x 10⁶ copies per mL. Genotype 1 (n = 33) and genotype non-1 (n = 13). Histological staging (HAI): median 5.3 (4 to 9), 12 patients cirrhosis

Interventions

Amantadine group: interferon-alpha-2b sc 3 MU daily for the first 4 weeks and subsequently 3 times a week, oral ribavirin at a daily dose of 1000 mg plus oral amantadine hydrochloride 200 mg/day administered in 2 doses of 100 mg

Control group: interferon-alpha-2b sc 3 MU daily for the first 4 weeks and subsequently 3 times a week plus oral ribavirin at a daily dose of 1000 mg

OutcomesMortality; liver-related morbidity; SAE; treatment discontinuation due to AE; composite outcome of number of patients with or without hepatitis C virus and ALT normalisation at end of treatment and end-of follow-up
Notes 
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskInsufficient information
Allocation concealment (selection bias)Unclear riskInsufficient information
Blinding of participants and personnel (performance bias)
All outcomes
High riskNot placebo-controlled
Blinding of outcome assessment (detection bias)
All outcomes
High riskNot placebo-controlled
Incomplete outcome data (attrition bias)
All outcomes
High riskDrop-outs not reported separately for 3 groups
Selective reporting (reporting bias)High riskOnly 1 outcome measure
Other biasUnclear risk

Vested interest bias: insufficient information

No baseline imbalance; sample size calculation was not reported; the trial was not stopped early

Angelico 2004

Methods

Randomised, open-label, controlled trial in naive patients in 14 centres

12 months therapy, 6 months follow-up

Participants

Country: Italy

181 patients were enrolled and started the initial 2-month treatment course with interferon-alpha-2a monotherapy. 17 patients dropped out within this period. The remaining 164 patients were randomised

Inclusion criteria: age between 18 and 65 years, presence of anti-HCV antibodies, positive serum HCV RNA by PCR, persistent elevation of serum ALT (≥ 1.5 times the upper limit of normal) during the 12 months prior to the study, and histological diagnosis of chronic hepatitis on liver biopsy sample taken in the preceding 6 months

Exclusion criteria: HBsAg or HIV positivity, recent or active alcohol and/or drug abuse, platelet count < 70,000/mL or leucocyte count < 3000/mL, histological evidence of cirrhosis, autoimmune or genetic liver diseases, other clinically significant diseases.

Amantadine group: 83 patients, mean age 39 ± 13 years, male/female = 61/22. Serum ALT 123 ± 73 MU/mL and the basal viral load 766 ± 747 x 10³ copies per mL. Genotype 1 (n = 45) and genotype non-1 (n = 38). Histological staging: median 1.7 ± 1.3

Control group: 81 patients, mean age 41 ± 12 years, male/female = 53/28. Serum ALT 110 ± 66 MU/mL and the basal viral load 738 ± 585 x 10³ copies per mL. Genotype 1 (n = 46) and genotype non-1 (n = 35). Histological staging: median 1.5 ± 1.1

Interventions

First there was an initial treatment course of 3 MU of recombinant interferon-alpha-2a, sc 3 times weekly for 2 months. Patients were then divided into 2 groups according to the serum HCV RNA status (HCV RNA-negative or HCV RNA-positive). Patients in each group were randomly assigned to receive:

Amantadine group: interferon-alpha-2a sc 3 times weekly plus amantadine, 200 mg po daily

Control group: interferon-alpha-2a sc 3 times weekly

In the HCV RNA-positive group, the dose of interferon-alpha-2a was increased to 6 MU 3 times weekly. At the end of month 6 of treatment, HCV RNA status was re-assessed. All HCV RNA-positive patients were withdrawn from therapy, whereas HCV RNA-negative patients continued treatment until month 12 according to their initial randomisation

OutcomesMortality; liver-related morbidity; SAE; treatment discontinuation due to AE; number of patients without SVR; number of patients with detectable HCV RNA at EOT; number of patients without normalisation of ALT at EOT and at EOFU
Notes
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskComputer-generated randomisation
Allocation concealment (selection bias)Low riskSealed envelopes
Blinding of participants and personnel (performance bias)
All outcomes
High riskNot placebo-controlled
Blinding of outcome assessment (detection bias)
All outcomes
High riskNot placebo-controlled
Incomplete outcome data (attrition bias)
All outcomes
High riskToo many drop-outs, not specified for what reasons and in which group: this can influence intervention effect
Selective reporting (reporting bias)High riskOutcome measures mentioned, but lack of other outcome measures
Other biasUnclear risk

Vested interest bias: insufficient information

No baseline imbalance; sample size calculation was reported and the trial was not stopped early

Angelico 2008

Methods

Randomised clinical trial in naive patients in 12 centres

The study was designed in 2001 for 48 weeks of treatment, 24 weeks of follow-up

Participants

Country: Italy and Sardinia

230 patients were randomised

Inclusion criteria: interferon-naive patients with chronic HCV, age 18 to 65 years, positive serum HCV RNA, elevated serum ALT (≥ 1.5 times the upper limit of normal in at least 2 determinations over the previous 6-month period)

Exclusion criteria: decompensated cirrhosis (presence or a history of ascites, gastrointestinal bleeding or hepatic encephalopathy); positive serum HBsAg, HIV co-infection; neutrophil count < 1500 cells/mm3; platelet count < 90 000 cells/mm3; haemoglobin levels < 12 g/dl (women) or < 13 g/dl (men); serum creatinine levels > 1.5 mg/dl; active alcohol or drug dependence; pregnancy or lactation; serological markers of autoimmunity; severe psychiatric disorders and cancer or severe pulmonary, renal or cardiac comorbidity. Cirrhotic patients were eligible only classified as Child–Pugh A

Amantadine group: 47 patients

Control group: 42 patients

Mean baseline characteristics for the whole group of 89 patients (actually 109, because the 20 patients who dropped out during the induction period belonged to this group according to ITT). Mean age 47.3 ± 12.1 years, male/female = 63/46, BMI 25.3 ± 3.5 kg/m2. Serum ALT was 117 ± 87 IU/l and the basal viral load was 939 ± 109 x 10³ IU per mL. Genotype 1 and 4 (n = 87) and genotype 2 and 3 (n = 22). Histological staging (Ishak): 2.3 ± 1.4

Interventions

Randomisation was performed after the assessment of EVR, defined as undetectable qualitative serum HCV RNA (< 50 IU/ml) after 12 weeks of induction monotherapy with peg interferon-alpha-2a (40 kDa) 180 µg/week sc

Patients who did not achieve EVR were randomised in a 1:1 ratio to add either:

Amantadine group: ribavirin, 800 mg/day, in divided doses and oral amantadine hydrochloride, 200 mg/day, for 36 additional weeks

Control group: oral ribavirin 800 mg/day, in divided doses, for 36 additional weeks

Patients who achieved EVR were randomised in a 1:1 ratio either to continue peg interferon-alpha-2a monotherapy or to add oral ribavirin, 800 mg/day, for 36 additional weeks

OutcomesMortality; liver-related morbidity; number of patients without SVR; number of patients with detectable HCV RNA at EOT
NotesAt 21 January 2012 ML sent email to angelico@med.uniroma2.it about treatment discontinuation due to SAE in each group
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskRandomisation was performed centrally using computer-generated lists and was stratified by individual centres and HCV genotypes (genotypes 1/4 versus genotypes non-1/4)
Allocation concealment (selection bias)Unclear riskRandom allocation to treatment groups
Blinding of participants and personnel (performance bias)
All outcomes
High riskNot placebo-controlled
Blinding of outcome assessment (detection bias)
All outcomes
High riskNot placebo-controlled
Incomplete outcome data (attrition bias)
All outcomes
High riskDrop-outs not described completely
Selective reporting (reporting bias)High riskPrimary outcome was mentioned, but other reasonably expected outcomes are missing
Other biasHigh risk

Vested interest bias: the study medication was provided by Roche Pharmaceuticals, Monza, Italy

No baseline imbalance; sample size calculation was reported; the trial was not stopped early

Bacosi 2002

Methods

Randomised clinical trial in non-responders or relapsers

12 months therapy, 12 months follow-up

Participants

Country: Italy

165 patients were screened (females n = 86; males n = 79) for 3 groups (55 patients each group)

Inclusion criteria: detectable, circulating HCV RNA; presence of chronic active liver disease already diagnosed on the grounds of laboratory and pathologic findings

Exclusion criteria: Child-Pugh score B or C, previous episode of gastrointestinal bleeding, disturbances of cardiac rhythm as determined by electrocardiogram and renal failure

Amantadine group: 38 patients, mean age 67 ± 4 years, male/female = 17/21. Serum ALT 2.6 ± 1.5-fold the upper limit of normal and the basal viral load 585 ± 257 x 10³ copies per mL. Genotype 1b was predominant (n = 32) with 4 patients with mixed genotypes. The other 6 patients had genotypes 2a (n = 3) and 2a-2c (n = 3). 1 patient cirrhosis

Control group: 39 patients, mean age 65 ± 2 years, male/female = 21/18. Basal viral load was 637 ± 452 x 10³ copies per mL, ALT was not provided. Genotype 1b was predominant (n = 31) associated with 1a in 3 cases; the remaining 8 patients had genotypes (2a (n=4), 2a-2c (n = 3) and 4 (n = 1). No patient with cirrhosis

Interventions

Amantadine group: interferon-alpha-n₃ 6 MU sc every other day until return to normal of ALT or a decrease in viral copies of at least 1 log unit (however, no longer than 3 months) then followed by 3 MU plus 200 mg/day amantadine orally

Control group: interferon-alpha-n₃ 6 MU sc every other day until return to normal of ALT or a decrease in viral copies of at least 1 log unit (however, no longer than 3 months) then followed by 3 MU

Another included group received only 100 mg amantadine oral twice daily

The duration of the trial treatment was 12 months; treatment, however, was planned to last for no more than 6 months if there was no significant decrease in viral load

OutcomesMortality; SAE; treatment discontinue due to AE; number of patients without SVR; number of patients with detectable HCV RNA at EOT; number of patients without normalisation of ALT at EOT and at EOFU
NotesML sent Dr Bacosi an email for additional information on 13 December 2011
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskInsufficient information
Allocation concealment (selection bias)Low riskClosed envelopes
Blinding of participants and personnel (performance bias)
All outcomes
High riskNot placebo-controlled
Blinding of outcome assessment (detection bias)
All outcomes
High riskNot placebo-controlled
Incomplete outcome data (attrition bias)
All outcomes
High riskDrop-outs not equally divided. Many patients dropped out after randomisation
Selective reporting (reporting bias)High riskNo clear primary and secondary outcome measures mentioned
Other biasUnclear risk

Vested interest bias: insufficient information

No baseline imbalance; sample size calculation was not reported; the trial was not stopped early

Baisini 2003

Methods

Randomised, open-label, controlled trial in naive patients involving 5 centres

12 months therapy, 6 months follow-up

Participants

Country: Italy

93 patients were randomised

Inclusion criteria: elevation of serum ALT > 2 times the upper limit of normal in at least 3 occasions over 12 months and with positive HCV RNA testing, histological evidence of chronic hepatitis as judged on liver biopsy performed no longer than 6 months prior to enrolment, confirmation of HCV infection by PCR

Exclusion criteria: age < 18 years or > 65 years, pregnancy or lack of appropriate contraceptive measures in women of child bearing age, previous treatment with antiviral or immunosuppressive drugs, current or previous drug addiction, alcoholism, positive HBsAg or HIV testing, histological evidence of cirrhosis, concomitant metabolic, autoimmune or neoplastic liver diseases, severe concomitant diseases other than liver disease, history of depression or psychiatric diseases, leukocyte count < 3000/dL, platelet count < 75,000/dL and serum albumin < 3 g/dL

Amantadine group: 48 patients, mean age 48 ± 1.8 years, male/female = 24/24. Serum ALT 130 ± 15 U/l and 32 patients had a basal viral load > 1 x 106 copies per mL. Genotype 1 (n = 22). Histological staging 1.7 ± 0.3

Control group: 45 patients, mean age 45 ± 1.8 years, male/female = 27/18. Serum ALT 115 ± 10 U/l and 32 patients had a basal viral load > 1 x 106 copies per mL. Genotype 1 (n = 19). Histological staging 1.3 ± 0.2

Interventions

i) Phase 1 (week 0/4): patients received either interferon-alpha lymphoblastoid 6 MU daily plus 100 mg amantadine twice daily; or the same dose of interferon-alpha alone (regimen B)

ii) Phase 2 (week 5/24): all patients in regimen A and in regimen B were shifted to receive interferon-alpha 6 MU 3 times a week while maintaining the amantadine dose as in phase 1 for patients allocated to regimen A
iii) Phase 3 (week 25/48): patients with serum ALT level lower than the upper limit of the normal range at the end of phase 2 were treated with a reduced dose of interferon-alpha dose, 3 MU 3 times a week while maintaining 100 mg twice daily amantadine for patients allocated to regimen A. Patients with abnormal ALT levels continued treatment as in phase 2

OutcomesMortality; liver-related morbidity; SAE; treatment discontinuation due to AE; number of patients without SVR; number of patients without improvement of histology; number of patients without normalisation of ALT at EOTand at EOFU
NotesAdditional information requested on 9 February 2012 from the last author Prof. Dr. A. Lanzini
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskThe method of sequence generation was not specified. Carried out using a blocked randomisation technique
Allocation concealment (selection bias)Unclear riskThe method of allocation concealment was not specified. Carried out using a blocked randomisation technique
Blinding of participants and personnel (performance bias)
All outcomes
High riskNot placebo-controlled
Blinding of outcome assessment (detection bias)
All outcomes
High riskNot placebo-controlled
Incomplete outcome data (attrition bias)
All outcomes
High riskPatients who terminated prematurely were not equally divided over the 2 treatment groups
Selective reporting (reporting bias)High riskThere was no protocol, but all the authors' study endpoints were discussed in the article. Not all reasonably expected outcomes were discussed
Other biasUnclear risk

Vested interest bias: unclear

No baseline imbalance; sample size calculation was reported; the trial was not stopped early

Berg 2003

Methods

Randomised, double-blind, placebo-controlled trial in naive patients in 8 centres

Patients were studied between December 1998 and June 2001

48 weeks therapy, 24 weeks follow-up

Participants

Country: Germany

400 patients were enrolled

Inclusion criteria: aged between 18 and 70 years with compensated chronic HCV infection who had not been previously treated with interferon-alpha, ribavirin, and/or amantadine, positive test for anti-HCV and HCV RNA by RT-PCR, elevated serum ALT levels for at least 6 months before initiation of treatment, and liver biopsy specimen taken in the preceding year of study entry showing chronic hepatitis

Exclusion criteria: decompensated liver disease, other causes of liver disease, hepatitis B infection, HIV infection, autoimmune disorders, haemoglobin values < 11 g/dL, white blood cell count < 3/nL, thrombocytopenia < 70/nL, other severe concurrent diseases, concurrent use of thiazide diuretics, pregnancy, or lactation period, alcohol or drug abuse or those unwilling to practice contraception

Amantadine group: 200 patients, mean age 41.7 ± 0.82 (18 to 70) years, male/female = 126/74. Mean weight 75.6 ± 1.0 (49 to 115) kg. Mean serum ALT 62.4 ± 4.03 (16 to 393) MU/mL and the mean basal viral load 5.99 ± 0.75 x 10⁶ copies per mL. Genotype 1 (n = 126), genotype 2 (n = 20), genotype 3 (n = 48), and genotype 4 (n = 6). Histological staging: F0 (n = 23), F1 (n = 79), F2 (n = 49), F3 (n = 35), and F4 (n = 14)

Control group: 200 patients, mean age 40.5 ± 0.79 (18 to 68) years, male/female = 127/73. Mean weight 73.7 ± 1.07 (47 to 125) kg. Mean serum ALT 62.8 ± 3.72 (16 to 369) MU/mL and the mean basal viral load 4.92 ± 0.59 x 10⁶ copies per mL. Genotype 1 (n = 129), genotype 2 (n = 12), genotype 3 (n = 47), genotype 4 (n = 9), and genotype 5 (n = 1). Histological staging: F0 (n = 28), F1 (n = 69), F2 (n = 52), F3 (n = 41), and F4 (n = 10)

Interventions

Amantadine group: total dose of 200 mg amantadine sulphate with interferon-alpha-2a plus 1000 to 1200 mg ribavirin per day orally adjusted according to body weight (1000 mg for weight < 75 kg and 1200 mg for weight ≥ 75 kg) for 48 weeks

Control group: matched placebo with interferon-alpha-2a plus 1000 to 1200 mg ribavirin per day orally adjusted according to body weight (1000 mg for weight < 75 kg and 1200 mg for weight ≥ 75 kg) for 48 weeks

For the first 2 weeks, 9 MU interferon-alpha daily, followed by 6 MU interferon-alpha daily for an additional 6 weeks, then 6 MU 3 times per week until week 24 and then 3 MU thrice weekly for a further 24 weeks

Psychological states were measured by the German adapted and validated version of the 'Profile of Mood States' (POMS) scale, which measures 4 factor scores for depression, fatigue, vigour, and anger. Furthermore, QoL was assessed by the 'Everyday Life' questionnaire (EDLQ), a German validated questionnaire related to the SF-36 Health Survey. The EDLQ assesses the following 6 subscales: body (e.g., make demands on body, concentrate on a task); mind (e.g., cope with illness, accept oneself); everyday life (e.g., solve daily problems, perform personal hygiene); social activity (e.g., get along with family, count on partner's help); zest for life (e.g., enjoy life); and medical treatment (e.g., believe in success of treatment)

OutcomesMortality; liver-related morbidity; SAE; treatment discontinuation due to AE; QoL; number of patients without SVR; number of patients with detectable HCV RNA at EOT; number of patients without normalisation of ALT at EOT and at EOFU
NotesStudy was supported by Merz + Co, Frankfurt a. M. and Hoffman-La Roche, Grenzach, Germany
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskIndependent central randomisation centre using a random number generator in fixed blocks of 4
Allocation concealment (selection bias)Low riskIndependent central randomisation centre using a random number generator in fixed blocks of 4
Blinding of participants and personnel (performance bias)
All outcomes
Low riskMatched placebo
Blinding of outcome assessment (detection bias)
All outcomes
Low riskMatched placebo
Incomplete outcome data (attrition bias)
All outcomes
High riskEach drop-out not explained/mentioned separately
Selective reporting (reporting bias)Low riskAll important outcome measures reported
Other biasHigh risk

Vested interest bias: supported by Merz + Co

No baseline imbalance; sample size calculation was reported; the trial was not stopped early

Brillanti 1999

Methods

Randomised clinical trial in interferon-alpha non-responders in 1 centre

Patients were enrolled between May and July 1996

6 months therapy, 6 months follow-up

Participants

Country: Italy

20 adult patients were randomised

Inclusion criteria: persistent elevations of serum ALT levels for at least the previous 6 months; liver biopsy obtained within the last 3 months before starting the previous interferon-alpha course, showing histological findings compatible with chronic viral hepatitis; the presence of antibodies to HCV by ELISA and of serum HCV RNA by PCR; the absence of circulating anti-interferon-alpha antibodies; no signs or symptoms of decompensated liver disease, other serious illnesses, or co-infection with HIV; previously been treated using 3 to 5 MU of recombinant or lymphoblastoid interferon-alpha on alternate days for 6 months, but neither biochemical nor virological response had been achieved; interferon-alpha had been discontinued at least 6 months before entry into study

Exclusion criteria: active hepatitis B virus infection, autoimmune hepatitis, alcoholic liver disease, and other possible causes chronic liver disease

Amantadine group: 10 patients, mean age 42.8 ± 2.5 years, male/female = 7/3. Mean weight not provided. Mean serum ALT 159.8 ± 22.9 MU/mL and the mean basal viral load 5.53 ± 0.22 x 10⁶ copies per mL. Genotype 1 (n = 4), genotype 2 (n = 3), and genotype 3 (n = 3). Histological staging: 5 patients cirrhosis

Control group: 10 patients, mean age 45.5 ± 5.2 years, male/female = 8/2. Mean weight not provided. Mean serum ALT 169.5 ± 49.6 MU/mL and the mean basal viral load 5.42 ± 0.19 x 10⁶ copies per mL. Genotype 1 (n = 4), genotype 2 (n = 4), and genotype 3 (n = 2). Histological staging: 4 patients cirrhosis

Interventions

Amantadine group: 100 mg oral amantadine per day plus 3 MU natural human leukocyte interferon-alpha-n3 on alternate days, plus 800 mg/day (if body weight < 75 kg) or 1000 mg/day (if body weight > 75 kg) ribavirin, given orally in 2 daily doses

Control group: 3 MU natural human leukocyte interferon-alpha-n3 on alternate days plus 800 mg/day (if body weight < 75 kg) or 1000 mg/day (if body weight > 75 kg) ribavirin, given orally in 2 daily doses

OutcomesMortality; liver-related morbidity; SAE; treatment discontinuation due to AE; number of patients without SVR; number of patients with detectable HCV RNA at EOT; number of patients without normalisation of ALT at EOT and at EOFU
Notes 
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskUsing restricted randomisation (permuted blocks) with serial entry, 10 individuals were randomly selected out of the set of 20 and allocated to the triple therapy group, and the other 10 were allocated to the double therapy group
Allocation concealment (selection bias)Unclear riskInsufficient information
Blinding of participants and personnel (performance bias)
All outcomes
High riskNot placebo-controlled
Blinding of outcome assessment (detection bias)
All outcomes
High riskNot placebo-controlled
Incomplete outcome data (attrition bias)
All outcomes
Low riskNo drop-outs
Selective reporting (reporting bias)High riskNo pre-planned outcome measures mentioned
Other biasUnclear risk

Vested interest bias: insufficient information

No baseline imbalance; sample size calculation was not reported; the trial was not stopped early

Brillanti 2000

Methods

Randomised clinical trial in interferon-alpha non-responders in 1 centre

Patients were enrolled starting in October 1996

12 months therapy, 6 months follow-up

Participants

Country: Italy

60 adult patients were randomised

Inclusion criteria: persistent elevations of serum ALT levels for at least the previous 6 months; liver biopsy obtained within the last 3 months before starting the previous interferon-alpha course, showing histological findings compatible with chronic viral hepatitis; the presence of antibodies to HCV by ELISA and of serum HCV RNA by PCR; the absence of circulating anti-interferon-alpha antibodies; previously been treated using 3 to 6 MU of recombinant or lymphoblastoid interferon-alpha on alternate days for 4 months, but neither biochemical nor virological response had been achieved; interferon-alpha had been discontinued at least 6 months, but not more than 12 months, before entering this study

Exclusion criteria: signs or symptoms of decompensated liver disease, co-infection with HIV, active hepatitis B virus infection, autoimmune hepatitis, alcoholic liver disease, other possible causes of chronic liver disease, other clinically significant diseases, haemoglobin concentration of < 12 g/dL in women and <13 g/dL in men, white cell count of < 3000 mm³, and platelet count of < 100,000 mm³

Amantadine group: 40 patients, median age 49 (28 to 70) years, male/female = 27/13. Weight not provided. Median serum ALT 124.5 (42 to 502) IU/L and the basal viral load 5.46 GMT (GMT = geometric mean titre of circulating HCV RNA as the antilog of the mean of the logarithmic transformed values of copies/mL). Genotype 1 (n = 23), genotype 2 (n = 11), genotype 3 (n = 3), and genotype 4 (n = 3). Histological staging: 10 patients cirrhosis

Control group: 20 patients, median age 47 (32 to 70) years, male/female = 13/7. Weight not provided. Median serum ALT 133 (58 to 404) IU/L and the basal viral load 5.5 GMT. Genotype 1 (n = 11), genotype 2 (n = 6), genotype 3 (n = 2), and genotype 4 (n = 1). Histological staging: 5 patients cirrhosis

Interventions

Amantadine group: oral amantadine hydrochloride administered twice daily at total dose of 200 mg plus 5 MU sc interferon-alpha-2b every other day and oral ribavirin 800 mg/day (if body weight < 75 kg) or 1000 mg/day (if body weight > 75 kg) ribavirin, given orally in 2 daily doses

Control group: 5 MU sc interferon-alpha-2b every other day and oral ribavirin 800 mg/day (if body weight < 75 kg) or 1000 mg/day (if body weight > 75 kg) ribavirin, given orally in 2 daily doses

OutcomesMortality; liver-related morbidity. SAE. Treatment discontinuation due to AE. Number of patients without SVR; number of patients with detectable HCV RNA at EOT; number of patients without normalisation of ALT at EOT; at EOFU
NotesSupported partially by a Research Grant from the Italian Ministry for the University and Scientific Research
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskWe used a restricted randomisation with a ratio of 2:1
Allocation concealment (selection bias)Unclear riskInsufficient information
Blinding of participants and personnel (performance bias)
All outcomes
High riskNot placebo-controlled
Blinding of outcome assessment (detection bias)
All outcomes
High riskNot placebo-controlled
Incomplete outcome data (attrition bias)
All outcomes
Low riskTreatment was not discontinued in any patient because of adverse events
Selective reporting (reporting bias)Low riskAll reasonably expected study endpoints were discussed in the article
Other biasUnclear risk

Vested interest bias: insufficient information

No baseline imbalance; sample size calculation was reported; the trial was not stopped early

Calay 2005

Methods

Multicentre, randomised, double-blind, placebo-controlled trial in naive patients

48 weeks treatment, 24 weeks follow-up

Participants

Country: France

269 patients were randomised, 253 really started with treatment

Patients with chronic HCV (proven by liver biopsy and positive for serum HCV RNA), previous treatment naive

Baseline characteristics were comparable in both groups:

Amantadine group: 128 patients, mean age 44.4 years, male/female = 78/50, BMI 23 kg/m2. Mean serum ALT was 2.3 times the upper limit of normal and the basal viral load was 1.3 MUI/mL. Genotype 1 (n = 88) and genotype non-1 (n = 40). Histological staging: 24 patients had extensive fibrosis and cirrhosis

Control group: 125 patients, mean age 45.6 years, male/female = 72/53, BMI 24 kg/m2. Mean serum ALT was 2.5 times the upper limit of normal and the basal viral load was 1.5 MUI/mL. Genotype 1 (n = 89) and genotype non-1 (n = 36). Histological staging: 22 patients had extensive fibrosis and cirrhosis

Interventions

Amantadine group: peg interferon-alpha-2b 1.5 μg/kg/week sc, ribavirin 800 to 1200 mg/day orally, with amantadine 200 mg/day for 48 weeks

Control group: peg interferon-alpha-2b 1.5 μg/kg/week sc, ribavirin 800 to 1200 mg/day orally, with placebo for 48 weeks

OutcomesNumber of patients without SVR
NotesML sent an email to dom-larrey@chu-montpellier.fr on 23 January 2012 about baseline characteristics
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskInsufficient information
Allocation concealment (selection bias)Unclear riskInsufficient information
Blinding of participants and personnel (performance bias)
All outcomes
Unclear riskInsufficient information, although placebo-controlled
Blinding of outcome assessment (detection bias)
All outcomes
Unclear riskInsufficient information, although placebo-controlled
Incomplete outcome data (attrition bias)
All outcomes
Unclear riskInsufficient information
Selective reporting (reporting bias)High riskDid not report on all reasonable important outcome measures
Other biasUnclear risk

Vested interest bias: research support by Schering-Plough, Roche

No baseline imbalance; sample size calculation was not reported; unknown if trial was stopped early

Caronia 2001

Methods

Randomised clinical trial in naive patients in 2 centres

Between 1997 and 1998

Treatment duration 48 weeks, follow-up 24 weeks

Participants

Country: United Kingdom

36 patients were randomised

Inclusion criteria: age between 18 and 70 years; liver biopsy taken within 18 months of randomisation showing chronic HCV with significant necro-inflammation (HAI grade > 3/18) and/or fibrosis (stage > 2/6) and ALT > 1.3 times the upper limit of normal within 6 months of randomisation

Exclusion criteria: patients with concomitant causes of liver disease, recent history of alcohol abuse (> 28 units per week within the last 6 months) or active intravenous drug use, and biopsy proven cirrhosis

Amantadine group: 18 patients, mean age 40 ± 12.2 years, male/female = 9/9. Serum mean ALT was 69 ± 43.9 U/L, the basal viral load was not provided. Genotype 1 (n = 9), genotype non-1 (n = 9). Histological staging was not provided, presence of cirrhosis = 0 patients

Control group: 18 patients, mean age 42 ± 14.3 years, male/female = 9/9. Serum ALT was 74 ± 52.6 U/L, the basal viral load was not provided. Genotype 1 (n = 10), genotype non-1 (n = 8). Histological staging was not provided, presence of cirrhosis = 0 patients

Interventions

Amantadine group: interferon-alpha-2a 4,5 MU sc 3 times weekly and amantadine hydrochloride, 200 mg oral daily, both for 48 weeks

Control group: interferon-alpha-2a 4,5 MU sc 3 times weekly for 48 weeks

OutcomesMortality; liver-related morbidity; treatment discontinuation due to AE
NotesAdditional information requested on 26 January 2012 from the last author Prof. Dr. G. Foster. Dr. Foster responded on 26January 2012. ML responded on 31 January 2012
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskComputer-generated random numbers
Allocation concealment (selection bias)High riskSealed envelopes. Not opaque
Blinding of participants and personnel (performance bias)
All outcomes
High riskNot placebo-controlled
Blinding of outcome assessment (detection bias)
All outcomes
High riskNot placebo-controlled
Incomplete outcome data (attrition bias)
All outcomes
Low risk1 patient in each group withdrew because of side effects
Selective reporting (reporting bias)High riskNot clearly mentioned what the primary and secondary outcome measures are. All the authors' study endpoints were discussed in the article. Not all reasonably expected outcomes were discussed
Other biasHigh risk

Vested interest bias: Roche

No baseline imbalance; sample size calculation was not reported; the trial was not stopped early

Caronia 2001a

Methods

Randomised clinical trial in naive patients in 14 centres

Between 1998 and 2000

Treatment duration 48 weeks, follow-up 24 weeks

Participants

Country: United Kingdom

143 patients were randomised

Inclusion criteria: age between 18 and 70 years; liver biopsy taken within 18 months of randomisation showing chronic hepatitis C with significant necro-inflammation (HAI grade > 3/18) and/or fibrosis (stage >2/6) and ALT > 1.3 times the upper limit of normal within 6 months of randomisation

Exclusion criteria: patients with concomitant causes of liver disease, recent history of alcohol abuse (> 28 units per week within the last 6 months) or active intravenous drug use

Amantadine group: 72 patients, mean age 43 ± 17.6 years, male/female = 45/27. Serum median ALT was 76 ± 10.6 U/L, the basal viral load was not provided. Genotype 1 (n = 19), genotype non-1 (n = 53). Histological staging was not provided, presence of cirrhosis = 6 or 7 patients

Control group: 71 patients, mean age 42 ± 21.6 years, male/female = 43/37. Serum ALT was 80 ± 39 U/L, the basal viral load was not provided. Genotype 1 (n = 20), genotype non-1 (n = 51). Histological staging was not provided, presence of cirrhosis = 7 patients

Interventions

Amantadine group: interferon-alpha-2a 4.5 MU sc 3 times weekly and amantadine hydrochloride, 200 mg oral daily, both for 48 weeks

Control group: interferon-alpha-2a 4.5 MU sc 3 times weekly for 48 weeks

OutcomesMortality; liver-related morbidity; treatment discontinuation due to AE
NotesAdditional information requested on 26 January 2012 from the last author Prof. Dr. G. Foster. Dr. Foster responded 26 January 2012. ML responded on 31 January 2012
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskComputer-generated random numbers
Allocation concealment (selection bias)High riskSealed envelopes. Not opaque
Blinding of participants and personnel (performance bias)
All outcomes
High riskNot placebo-controlled
Blinding of outcome assessment (detection bias)
All outcomes
Low riskIndependent laboratory performed all the PCR tests and all samples were provided to the laboratory in a coded, anonymous fashion
Incomplete outcome data (attrition bias)
All outcomes
High riskThe exact reasons for patients who terminated prematurely were not clearly explained
Selective reporting (reporting bias)High riskNot clearly described what the primary and secondary outcome measures are. All the authors' study endpoints were discussed in the article. Not all reasonably expected outcomes were discussed
Other biasHigh risk

Vested interest bias: Roche

No baseline imbalance; sample size calculation was not reported; the trial was not stopped early

Ciancio 2006

Methods

Randomised clinical trial in non-responders

Between May 2001 and December 2002 patients were included

12 months therapy, 6 months follow-up

Participants

Country: Italy

161 patients were randomised

Inclusion criteria: age > 18 and < 65 years; positive results for HCV RNA by PCR; chronic HCV at liver biopsy performed within 1 year before entry; previous non-response to combined therapy; abnormal ALT levels (at least 1.5 times upper limit of normal; range: 0 to 40 IU)

Exclusion criteria: previous course with peg interferon-alpha-based therapy; relapse after 1 or more interferon-alpha plus ribavirin courses; positive HBsAg test in serum; positive test for antibody to HIV; alcoholic liver disease; haemochromatosis; Wilson's disease; drug related liver disease; autoimmune hepatitis; haemoglobin level < 10 g/dL, platelet count < 70,000/mm3, white blood cell count < 3000/mm3, or granulocyte count < 1500/mm3; decompensated cirrhosis; intravenous drug abuse; abnormal serum uric acid level; presence of concomitant significant medical illness; history of haemolytic anaemia; a1-antitrypsin deficiency; obesity-induced liver disease; haemophilia; seizure disorders; ischaemic cardiovascular disease and severe mental depression. Pregnant women and patients unable to practice contraception during therapy and follow-up

Amantadine group: 80 patients, mean age 50 ± 11 (22 to 65) years, male/female = 59/21, BMI 24.8 ± 3.4 (17.3 to 33) kg/m2. Mean serum ALT was 116 ± 85 (43 to 335) IU/L, the basal viral load was 2.1 ± 3.1 (0.01 to 19) x 106 Eq/mL. Genotype 1 (n = 67), genotype 2 (n = 4), genotype 3 (n = 1), and genotype 4 (n = 8). Mean histological staging was 3 ± 1.5 (0 to 6), presence of cirrhosis = 11 patients

Control group: 81 patients, mean age 50 ± 11 (27 to 65) years, male/female = 60/21, BMI 24.9 ± 3.5 (17.6 to 34.2) kg/m2. Mean serum ALT was 127 ± 84 (39 to 770) IU/L, the basal viral load was 1.8 ± 3.1 (0.07 to 18) x 106 Eq/mL. Genotype 1 (n = 66), genotype 2 (n = 9), genotype 3 (n = 3), and genotype 4 (n = 3). Mean histological staging was 3 ± 1.5 (0 to 6), presence of cirrhosis = 7 patients

Interventions Amantadine group: 180 μg once weekly of peg interferon-alpha-2a plus ribavirin, either 1000 mg/day (body
weight < 75 kg) or 1200 mg/day (body weight > 75 kg) plus amantadine 200 mg daily for 12 months
Control group: 180 μg once weekly of peg interferon-alpha-2a plus ribavirin, either 1000 mg/day (body weight < 75 kg) or 1200 mg/day (body weight > 75 kg) for 12 months
OutcomesMortality; liver-related morbidity; SAE; treatment discontinuation due to AE; number of patients without SVR; number of patients with detectable HCV RNA at EOT
NotesML sent an email to g.saracco@tin.it on 23 January 2012 about treatment discontinuation due to AE in both groups. Dr. Saracco answered with additional information on 23 January 2012
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskEach patient was allocated through a concealed process, using a computerised program with block randomisation at a central location
Allocation concealment (selection bias)Low riskEach patient was allocated through a concealed process, using a computerised program with block randomisation at a central location
Blinding of participants and personnel (performance bias)
All outcomes
High riskNot placebo-controlled
Blinding of outcome assessment (detection bias)
All outcomes
High riskNot placebo-controlled
Incomplete outcome data (attrition bias)
All outcomes
Unclear riskNumbers of patients withdrawn due to adverse events in text does not match with number of patients in table. Unknown if all patients who withdrew from the study were reported
Selective reporting (reporting bias)High riskPrimary outcome mentioned, but lack of other outcome measures
Other biasLow risk

Vested interest bias: no external funding was received for this study

No baseline imbalance; sample size calculation was not reported; the trial was not stopped early

Cornberg 2000

Methods

Randomised, placebo-controlled trial in interferon non-responders

12 months therapy, 6 months follow-up

Participants

Country: Germany

26 patients were randomised

Amantadine group: 14 patients, mean age 38 years, male/female = 14/0. Serum ALT was not provided, the mean basal viral load was 1,099,643 copies per mL. Genotype 1a/b (n = 13), genotype non-1 (n = 1). Histological staging was not provided

Control group: 89 patients, mean age 41 years, male/female = 12/0. Serum ALT was not provided, the mean basal viral load was 1,420,417 copies per mL. Genotype 1a/b (n = 9), genotype non-1 (n = 3). Histological staging was not provided

Interventions

Amantadine group: interferon 2 weeks 10 MU daily, 2 weeks 5 MU daily, 8 weeks 3 MU daily followed by 3 MU every other day for further 9 months plus daily 1000 to 1200 mg ribavirin and 200 mg amantadine orally once daily

Control group: interferon 2 weeks 10 MU daily, 2 weeks 5 MU daily, 8 weeks 3 MU daily followed by 3 MU every other day for further 9 months plus daily 1000 to 1200 mg ribavirin and placebo orally once daily

OutcomesMortality; liver-related morbidity; SAE; treatment discontinuation due to AE; number of patients without SVR; number of patients with detectable HCV RNA at EOT
NotesML sent Dr. Cornberg an email on 12 January 202012 about the biochemical responses. Dr Cornberg responded the same day
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskInsufficient information
Allocation concealment (selection bias)Unclear riskInsufficient information
Blinding of participants and personnel (performance bias)
All outcomes
Unclear riskInsufficient information, although the study is placebo-controlled
Blinding of outcome assessment (detection bias)
All outcomes
Unclear riskInsufficient information, although the study is placebo-controlled
Incomplete outcome data (attrition bias)
All outcomes
Low risk26 patients were enrolled and completed treatment
Selective reporting (reporting bias)High riskOnly reported virological response
Other biasUnclear riskInsufficient information

Ferenci 2006

Methods

Randomised, double-blind, placebo-controlled study in naive patients

48 weeks therapy, 72 weeks follow-up (24 weeks follow-up also measured)

Participants

Country: Austria

233 patients screened and received interferon sensitivity test; 211 randomised

Inclusion criteria: treatment-naive patients with chronic HCV, genotype 1 infection, liver biopsy findings consistent with diagnosis of chronic HCV (obtained within 6 months), and elevated serum ALT activity (> 1.5 times the upper limit of normal) in the previous 6 months and during screening; haemoglobin values ≥ 12 g/dL (women) or ≥ 13 g/dL (men), leukocytes ≥ 3000/mL and platelets ≥ 100,000/mL

Exclusion criteria: refusal by women of child-bearing age or by sexually active patients to use effective contraception; pregnancy or breastfeeding; decompensated liver disease; coronary heart disease; co-infection with HIV or hepatitis B; overt psychiatric disorders; active alcohol or drug abuse; diabetes mellitus requiring medical therapy; autoimmune disorders and/or any other unstable medical condition not due to liver disease. Due to the potential adverse effects of amantadine, patients with Parkinson’s disease, narrow angle glaucoma or adenoma of the prostate gland

Amantadine group: 114 patients, mean age 45 ± 11 years, male/female = 68/46, mean BMI 25.5 ± 4.2 kg/m2. Median serum ALT was 47 (18 to 313) IU/L, the median basal viral load was 0.465 (0.023 to 3.82) x 106 IU/mL. Genotype 1a (n = 23), genotype 1b (n = 73), and genotype 1a and 1b (n = 18). Histological staging: F0-F2 = 83; F3 = 15; F4 = 16

Control group: 95 patients, mean age 44 ± 10 years, male/female = 65/30, mean BMI 25.7 ± 3.9 kg/m2. Median serum ALT was 54 (21 to 208) IU/L, the median basal viral load was 0.417 (0.0009 to 4.0) x 106 IU/mL. Genotype 1a (n = 27), genotype 1b (n = 43), and genotype 1a and 1b (n = 25). Histological staging: F0-F2 = 71; F3 = 10; F4 = 14

Interventions

Amantadine group: peg interferon-alpha-2a (40KD) 180 μg/week plus ribavirin 1000 to 1200 mg/day and oral amantadine 100 mg twice daily

Control group: peg interferon-alpha-2a (40KD) 180 μg/week plus ribavirin 1000 to 1200 mg/day and a matched placebo. Compliance was assessed by counting unused syringes and tablets at each visit

OutcomesMortality; liver-related morbidity; SAE; treatment discontinuation due to AE; QoL; number of patients without SVR
Notes
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskRandomisation was performed centrally using an adaptive biased coin design, stratified for study centre, the interferon sensitivity stratum, and fibrosis grade (Metavir 0/1/2 versus 3/4). Since this was a dynamic unrestricted procedure, the allocation sequence was produced during the study and unequal numbers of patients per treatment group were considered to be acceptable
Allocation concealment (selection bias)Unclear riskNot mentioned
Blinding of participants and personnel (performance bias)
All outcomes
Low riskMatched placebo
Blinding of outcome assessment (detection bias)
All outcomes
Unclear riskMatched placebo, but insufficient information
Incomplete outcome data (attrition bias)
All outcomes
High riskNo missing outcome data, but they used the last observation carried forward method
Selective reporting (reporting bias)High riskOutcome measures were reported, but no example histological outcomes and outcomes per treatment group (EOT in amantadine versus control group)
Other biasHigh risk

Vested interest bias: high risk: This study reported an unrestricted research grant from Roche Austria, Vienna

No baseline imbalance; sample size calculation was reported; the trial was not stopped early

Gaeta 2001

Methods

Randomised clinical trial in non-responders to interferon

6 months therapy, 6 months follow-up

Participants

Country: Italy

40 patients were randomised

Inclusion criteria: persistent serum ALT levels of > 1.5 times the upper normal limit; presence of anti-HCV antibodies and HCV RNA in serum; histological features of chronic hepatitis in a liver biopsy obtained in the previous 12 months; HCV genotype 1b

Exclusion criteria: age older than 60 years, decompensated cirrhosis, kidney disease, current use of antihistamine drugs, HBsAg or anti-HIV positivity and any of the major contraindications to interferon treatment

Amantadine group: 21 patients, mean age 44.7 ± 9.2 years, male/female = 14/7. Serum median ALT was 130 U/L and the basal viral load was 1.0 x 10⁶ copies per mL. Genotype 1b (n = 21). Histological staging: median 1.5; presence of cirrhosis = 1

Control group: 19 patients, mean age 48.4 ± 9.3 years, male/female = 12/7. Serum ALT was 134 U/L and the basal viral load was 1.2 x 10⁶ copies per mL. Genotype 1b (n = 19). Histological staging: median 1.5; presence of cirrhosis = 2

Interventions

Amantadine group: interferon-alpha-2a 4.5 MU sc daily for 4 weeks, followed by 6 MU sc thrice weekly for an additional 5 months and amantadine sulphate 100 mg orally twice daily for the complete 6 months

Control group: interferon-alpha-2a 4.5 MU sc daily for 4 weeks, followed by 6 MU sc thrice weekly for an additional 5 months

OutcomesMortality; treatment discontinuation due to AE; number of patients without SVR; number of patients with detectable HCV RNA at EOT; number of patients without normalisation of ALT at EOT and at EOFU
Notes
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskComputer-generated list
Allocation concealment (selection bias)Unclear riskInsufficient information
Blinding of participants and personnel (performance bias)
All outcomes
High riskNot placebo-controlled
Blinding of outcome assessment (detection bias)
All outcomes
High riskNot placebo-controlled
Incomplete outcome data (attrition bias)
All outcomes
Low riskNo missing data
Selective reporting (reporting bias)High riskNot clearly stated what the primary and secondary outcome measures are
Other biasUnclear risk

Vested interest bias: unclear

No baseline imbalance; sample size calculation was not reported; the trial was not stopped early

Gramenzi 2007

Methods

Randomised, multicentre study in interferon-alpha non-responders

Patients were enrolled between September 1998 and April 1999

48 weeks therapy, 24 weeks follow-up

Participants

Country: Italy

75 patients were randomised

Inclusion criteria: histologically proven chronic HCV non-responders to a previous course of 3 to 6 MU recombinant interferon-alpha 3 times a week for at least 16 weeks with or without ribavirin, failure to clear HCV RNA from serum and to normalise serum ALT during treatment period, with persistent positivity of HCV RNA for at least 12 months and persistent ALT levels greater than 1.5 times the normal value

Exclusion criteria: aged ≤ 18 and ≥ 64 years; decompensated liver disease; co-infection with HIV, HBsAg positivity; evidence of any cause of liver disease other than chronic hepatitis C; serum haemoglobin concentration of < 12 g/dL for women or < 13 g/dL for men; white cell count of < 3000/mm³; neutrophil count of < 1500/mm³; platelet count of < 70,000 mm³; presence of haemoglobinopathy or haemolytic anaemia; alcohol abuse; drug abuse; pregnancy; other clinically significant diseases

Amantadine group: 25 patients, mean age 50.1 ± 10.0 years, male/female = 15/10. Weight was not provided. Mean serum ALT 114.4 ± 69.8 U/L and mean basal viral load 2.3 ± 2.0 MEq/mL. Genotype 1 (n = 19), genotype 2 (n = 3), and genotype 3 (n = 3). Histological staging: 7 patients cirrhosis

Control group: 24 patients, mean age 49.7 ± 11.3 years, male/female = 16/8. Weight was not provided. Mean serum ALT 120.8 ± 77.9 U/L and mean basal viral load 2.3 ± 2.9 MEq/mL. Genotype 1 (n = 18), genotype 2 (n = 3), genotype 3 (n = 2), and genotype 4 (n = 1). Histological staging: 5 patients cirrhosis

Interventions

Amantadine group: oral amantadine hydrochloride administered twice daily at a total dose of 200 mg, plus 6 MU sc interferon-alpha-2a every other day for the first 4 weeks, followed by a dose of 3 MU per day for the remaining 44 weeks, and 15 mg/kg per day of oral ribavirin

Control group: 6 MU sc interferon-alpha-2a every other day for the first 4 weeks, followed by a dose of 3 MU per day for the remaining 44 weeks and 15 mg/kg per day of oral ribavirin

There was a third study group in this trial: 6 MU sc interferon-alpha-2a every other day for the first 4 weeks, followed by a dose of 3 MU per day for the remaining 44 weeks

OutcomesMortality; liver-related morbidity; SAE; treatment discontinuation due to AE; number of patients without SVR; number of patients with detectable HCV RNA at EOT; number of patients without normalisation of ALT at EOT and at EOFU
Notes
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskPatients were randomly assigned to one of 3 different treatment groups
Allocation concealment (selection bias)Unclear riskInsufficient information
Blinding of participants and personnel (performance bias)
All outcomes
High riskNot placebo-controlled
Blinding of outcome assessment (detection bias)
All outcomes
High riskNot placebo-controlled
Incomplete outcome data (attrition bias)
All outcomes
Unclear riskDrop-outs mentioned, but not divided by groups
Selective reporting (reporting bias)High riskOnly 1 study endpoint
Other biasUnclear risk

Vested interest bias: insufficient information

No baseline imbalance; sample size calculation was reported; the trial was not stopped early

Hasan 2004

Methods

Randomised clinical trial in non-responders

The study was conducted between March 2000 and February 2002

48 weeks therapy, 24 weeks follow-up

Participants

Country: Kuwait

63 patients were randomised

Inclusion criteria: age between 18 and 65 years; HCV RNA detectable in serum at concentrations > 3200 copies/mL (615 IU/mL) by a branched DNA assay within 3 months of enrolment; prior treatment for at least 6 months with a combination of unmodified interferon-alpha-2a or alpha-2b plus ribavirin; persistence of HCV RNA in serum at the end of combination therapy (non-responder); and evidence of chronic hepatitis with or without cirrhosis

Exclusion criteria: transient virological response during or at the end of combination therapy, followed by a relapse; clinical or biochemical evidence of hepatic decompensation; suspicion of hepatocellular carcinoma; white blood cell count < 2.5 × 109/l, haemoglobin < 110 g/l, platelet count < 60 × 109/l; serum creatinine > 140 μmol/l; alcohol or drug abuse; and severe comorbid medical or psychiatric conditions

Amantadine group: 42 patients, median age 42 (17 to 56) years, male/female = 34/8, BMI not provided. Median serum ALT 90 (62 to 184) IU/L, median basal viral load 2.1 (0.3 to 15) x 106 eq/mL. Genotype 1a/1b (n = 8), genotype 4 (n = 33), and genotype 1 (n = 1). Histological staging: F1/F2 = 22; F3/F4 = 20

Control group: 21 patients, median age 43 (20 to 61) years, male/female = 16/5, BMI not provided. Median serum ALT 96 (60 to 201) IU/L, median basal viral load 2.3 (0.6 to 17) x 106 eq/mL. Genotype 1a/1b (n = 4), genotype 4 (n = 17). Histological staging: F1/F2 = 12; F3/F4 = 9

Interventions

Patients were randomised in a 2:1 ratio:

Amantadine group: peg interferon-alpha-2b sc once weekly, at a dose of 1.5 μg/kg, ribavirin orally at a dose of 1000 mg or 1200 mg per day for patients weighing < 75 kg and ≥ 75 kg, and amantadine 200 mg/day
Control group: peg interferon-alpha-2b sc once weekly, at a dose of 1.5 μg/kg, ribavirin orally at a dose of 1000 mg or 1200 mg per day for patients weighing < 75 kg and ≥ 75 kg

OutcomesMortality; liver-related morbidity; SAE; treatment discontinuation due to AE; number of patients without SVR; number of patients with detectable HCV RNA at EOT; number of patients without normalisation of ALT at EOT and at EOFU
Notes
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskRandomly assigned, insufficient information
Allocation concealment (selection bias)Unclear riskRandomly assigned, insufficient information
Blinding of participants and personnel (performance bias)
All outcomes
High riskNot placebo-controlled
Blinding of outcome assessment (detection bias)
All outcomes
High riskNot placebo-controlled
Incomplete outcome data (attrition bias)
All outcomes
Unclear riskInsufficient information. The number of patients who completed the entire scheduled dose was reported, but information about withdrawals was missing, only 3 patients withdrew because of side effects, so it is not clear what happened to the other 2.
Selective reporting (reporting bias)Low riskReported on all important outcomes
Other biasUnclear risk

Vested interest bias unclear

No baseline imbalance; sample size calculation was reported; the trial was not stopped early

Helbling 2002

Methods

Double-blind, randomised, placebo-controlled trial in naive patients, in 28 centres

12 months therapy, 24 weeks follow-up

Participants

Country: Switzerland

254 patients were enrolled, 8 patients withdrew informed consent after baseline evaluation, but before starting treatment. 246 started treatment

Inclusion criteria: patients aged 18 to 65 years with biopsy-proven (within ≤ 2 years) chronic HCV who had never been treated before, exhibited elevated ALT within 6 months of entry on at least 2 occasions at least 1 month apart, and tested positive for HCV RNA in serum by RT-PCR

Exclusion criteria: any other cause of liver disease including HBV co-infection, and alcohol intake > 20 g/day in females and > 40 g/day in males; history of or actual decompensation of liver disease; cirrhosis ≥ 8 Child-Pugh points; leucocytes < 2000/μL, neutrophils < 50,000/μL, serum creatinine > 1.5 times upper limit of normal

Amantadine group: 121 patients, age 39 (20 to 66) years, male/female = 68/53. Serum ALT 101 U/L (34 to 421) and basal viral load 2.16 x 10⁶ copies per mL. Genotype 1 (n = 62), genotype 2 (n = 12), genotype 3 (n = 34), genotype 4 (n = 5), and genotype 6 (n = 1). The other 6 patients had genotypes 2a (n = 3) and 2a-2c (n = 3). Histological staging: mild = 59, moderate = 43, severe = 18

Control group: 125 patients, age 38 (20 to 65) years, male/female = 70/55. Serum ALT 111 U/L (30 to 768) and basal viral load 4.72 x 10⁶ copies per mL. Genotype 1 (n = 52), genotype 2 (n = 11), genotype 3 (n = 50), and genotype 4 (n = 6). Histological staging: mild = 29, moderate = 91, severe = 4

Interventions

Amantadine group: interferon-alpha-2a 6 MIU sc thrice weekly for 20 weeks, followed by 3 MIU sc thrice weekly for an additional 32 weeks and amantadine sulphate 100 mg oral twice daily

Control group: interferon-alpha-2a 6 MIU sc thrice weekly for 20 weeks, followed by 3 MIU sc thrice weekly for an additional 32 weeks and placebo oral twice daily

Treatment was stopped if after 10 weeks HCV RNA in serum remained detectable by RT-PCR

OutcomesMortality; SAE; treatment discontinuation due to AE; QoL. Number of patients without SVR; number of patients with detectable HCV RNA at EOT; number of patients without normalisation of ALT at EOT and at EOFU
NotesML sent an email to Prof Dr. Renner on 23 December 2011 about the drop-out rate. ML forwarded this email on 9 January 2012 to Dr. Helbling
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskRandomisation was carried out in blocks of 10 using random numbers stratified according to the presence/absence of cirrhosis
Allocation concealment (selection bias)Unclear riskInsufficient information
Blinding of participants and personnel (performance bias)
All outcomes
Low riskMatched placebo
Blinding of outcome assessment (detection bias)
All outcomes
Unclear riskHCV RNA was determined centrally. A single pathologist (CG), unaware of clinical data including treatment response, scored all pretreatment liver biopsies using the extended Knodell score
Incomplete outcome data (attrition bias)
All outcomes
High riskNumbers of patients who withdrew due to adverse events in text do not match the numbers of patients in the table. Unknown if all patients who withdrew from the study were reported
Selective reporting (reporting bias)High riskAll the authors' study endpoints were discussed in the article. Not all reasonably expected outcomes were discussed
Other biasHigh risk

Vested interest bias: Roche

No baseline imbalance; sample size calculation was reported; the trial was not stopped early

Herrine 2005

Methods

Randomised, controlled, multicentre trial in relapsers or patients who had a viral breakthrough

48 weeks therapy, 24 weeks follow-up

Participants

Country: United States of America

124 patients were randomised, 123 received at least 1 dose of study medication

Inclusion criteria: adult patients with serologic evidence of HCV infection, by a positive anti-HCV antibody test and detectable HCV RNA in serum, who had a virologic response during treatment with standard interferon-alpha-2b plus ribavirin and had relapsed after at least 24 weeks of treatment or had a virologic breakthrough while still on treatment; serum ALT activity above the upper limit of normal during the 6 months before entering the study; liver biopsy consistent with chronic HCV infection in the previous 36 months; and a minimum of 24 weeks since cessation of standard interferon-alpha-2b plus ribavirin treatment, with no interferon therapy during this time

Exclusion criteria: had received any systemic antiviral therapy within 24 weeks of the start of the study or were expected to need any systemic antiviral therapy during the study or had acute hepatitis A or B infection, HIV infection, decompensated liver disease, neutropenia (< 1500 neutrophils/mm³), anaemia (haemoglobin < 12 g/dL in women and < 13 g/dL in men), thrombocytopenia (platelets, < 90,000/mm3), serum creatinine level higher than 1.5 times the upper limit of normal, history of alcohol or drug abuse within 1 year of entry, history of severe psychiatric disease, serum α-fetoprotein level > 100 ng/mL, or substantial coexisting medical conditions

Amantadine group: 31 patients, mean age 46 years, male/female = 20/11, BMI not provided. Mean serum ALT 67 SE 9 U/L, mean AST 45 SE 6 U/L, basal viral load ≤ 800,000 IU/mL: 12, and > 800,000IU/mL: 19. Genotype 1 (n = 25) and genotype non-1 (n = 6). Histological staging: non-cirrhosis = 27; cirrhosis = 4

Control group: 32 patients, mean age 48 years, male/female = 24/8, BMI not provided. Mean serum ALT 75 SE 10 U/L, mean AST 60 SE 7 U/L, basal viral load ≤ 800,000 IU/mL: 14, and > 800,000IU/mL: 18. Genotype 1 (n = 25) and genotype non-1 (n = 7). Histological staging: non-cirrhosis = 23; cirrhosis = 9

Interventions

Patients were randomly assigned at a 1:1:1:1 ratio to:

Amantadine group: sc weekly injections of 180 μg peg interferon-alpha-2a plus orally administered ribavirin, 800 mg/day in split doses for patients weighing < 75 kg and 1000 mg/day in split doses for those weighing ≥ 75 kg, and amantadine 200 mg/day for 48 weeks

Control group: sc weekly injections of 180 μg peg interferon-alpha-2a plus orally administered ribavirin for 48 weeks in the same dosage as mentioned at the amantadine group

2 other intervention groups were: peg interferon-alpha-2a plus mycophenolate mofetil and peg interferon-alpha-2a plus amantadine, both also for 48 weeks in the same dosages as mentioned above, with a daily dose of mycophenolate mofetil of 1 g twice daily

Randomisation was stratified according to HCV genotype (type 1 versus non-type 1, with any patient positive for both type 1 and non-type 1 categorised as type 1), viral load (≤ 800,000 or > 800,000 IU/mL), and relapse versus breakthrough

OutcomesMortality; liver-related morbidity; SAE; treatment discontinuation due to AE; number of patients without SVR; number of patients with detectable HCV RNA at EOT; number of patients without normalisation of ALT at EOFU
Notes
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskInsufficient information
Allocation concealment (selection bias)Unclear riskInsufficient information
Blinding of participants and personnel (performance bias)
All outcomes
High riskNot placebo-controlled
Blinding of outcome assessment (detection bias)
All outcomes
High riskNot placebo-controlled
Incomplete outcome data (attrition bias)
All outcomes
Unclear riskInsufficient information. Withdrawals mentioned, but not the reason for withdrawal in all patients
Selective reporting (reporting bias)High riskNot every outcome we would suggest was reported on
Other biasHigh risk

Vested interest bias: high: research grant from Roche

No baseline imbalance; sample size calculation was reported; the trial was not stopped early

Jorge 2001

Methods

Randomised clinical trial in naive patients

12 months therapy, 24 weeks follow-up

Participants

Country: Argentina

91 patients were randomised

Amantadine group: 47 patients

Control group: 44 patients

Genotype 1 (63%), viral load, ALT, and necro-inflammatory/fibrosis scores level were similar in both groups

Interventions

Amantadine group: amantadine 200 mg daily and interferon-alpha-2a 6 MU daily for 4 weeks, 3 MU daily for 8 weeks, and 3 MU 3 times a week for 12 months

Control group: interferon-alpha-2a 6 MU daily for 4 weeks, 3 MU daily for 8 weeks, and 3 MU 3 times a week for 12 months

Treatment was discontinued in patients with detectable serum HCV RNA after treatment week 24

OutcomesNumber of patients without SVR
NotesML sent an email to Dr. Daruich on 12 January 2012 about virological EOT and biochemical responses
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskRandomly allocated, but method not described
Allocation concealment (selection bias)Unclear riskInsufficient information
Blinding of participants and personnel (performance bias)
All outcomes
High riskNot placebo-controlled
Blinding of outcome assessment (detection bias)
All outcomes
High riskNot placebo-controlled
Incomplete outcome data (attrition bias)
All outcomes
Unclear riskInsufficient information
Selective reporting (reporting bias)Unclear riskInsufficient information
Other biasUnclear riskInsufficient information

Langlet 2009

Methods

Multicentre, randomised clinical trial with parallel-group design in naive and relapsing patients

Trial duration: 24 or 48 weeks, follow-up 24 weeks

Participants

Country: Belgium, 37 centres

Number of patients randomised: 630 (actually 643, but 13 never took any study medication)

Amantadine group: mean age 43.74 ± 12.41 years, male/female: 189/127

Control group: mean age 45.48 ± 12.19 years, male/female: 173/141

Inclusion criteria: male and female patients ≥ 18 years of age; serological evidence of chronic HCV (anti-HCV antibody test), quantifiable serum HCV RNA of ≥ 600 IU/mL; elevated serum ALT activity documented on at least 2 occasions within the 6 months before randomisation; histological liver alterations consistent with chronic HCV; in case of cirrhosis, a compensated liver disease (Child-Pugh Grade A)

Exclusion criteria: non-responders to a previous therapy or had a relapse during a previous therapy (breakthrough) or after completion of any previous treatment other than interferon plus ribavirin; previous treatment with any systemic antiviral, anti-neoplastic, or immunomodulatory treatment within 6 months prior to the first dose of the study drug; chronic liver disease other than HCV; other clinically significant medical history or current disease; positive serology for HAV IgM, haemoglobin < 11 g/dL, neutrophil count < 1500 cells/mm3, platelet count < 90,000 cells/⁄mm3, and serum creatinine level > 1.5 times the upper limit of normal; pregnancy

Interventions

Amantadine group: peg interferon-alpha, ribavirin, and amantadine, n = 316

Control group: peg interferon-alpha and ribavirin, n = 314

Peg-INF was given sc at a dose of 180 μg in 0.5 mL, ribavirin was given twice daily at a total oral dose of 800 to 1200 mg daily according to body weight, amantadine was given orally 100 mg twice daily

Treatment was given for 24 or 48 weeks according to genotype

OutcomesSustained virological response; sustained biochemical response rate; early virological response rate; end of treatment virological response rate; mean reduction in HCV RNA
NotesAdditional information requested on 23 January 2012 from the last author, Prof. Dr. F. Nevens
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskRandomisation included a minimisation programme by study centre
Allocation concealment (selection bias)Unclear riskInsufficient information
Blinding of participants and personnel (performance bias)
All outcomes
High riskNot placebo-controlled
Blinding of outcome assessment (detection bias)
All outcomes
High riskNot placebo-controlled
Incomplete outcome data (attrition bias)
All outcomes
High riskNot equally matched drop-outs and insufficient information about reasons for drop-outs
Selective reporting (reporting bias)High riskAll outcome measures reported, but lacking some important outcome measures such as EOT ALT
Other biasHigh risk

Vested interest bias: high: Roche funded

No baseline imbalance; sample size calculation was reported; the trial was not stopped early

Mangia 2001

Methods

Randomised clinical trial in naive patients

Patients were recruited between June and December 1998

12 months therapy, 6 months follow-up

Participants

Country: Italy

200 patients were randomised

Inclusion criteria: raised ALT for at least 6 months, HCV RNA positive by PCR, liver biopsy performed within the previous 6 months before entry, consistent with chronic HCV

Exclusion criteria: decompensated cirrhosis, psychiatric conditions, diabetes, autoimmune diseases, concurrent hepatitis B or HIV infections, high alcohol intake, current intravenous drug use, previous treatment with interferon, pregnancy, or concomitant significant medical illness

Amantadine group: 99 patients, age 46 (19-67) year, male/female = 61/28. Serum ALT was not provided and basal viral load was 60 (0.3 to 400) x 10⁶ copies per mL. Genotype 1 (n = 52), genotype 2a (n = 34), genotype 3 (n = 7), genotype 4 (n = 6). Histological staging: 0/1 = 63, 2/3 = 36

Control group: 101 patients, mean age 48 (21 to 69) year, male/female = 71/30. Serum ALT was not provided and basal viral load was 58 (1.9 to 500) x 10⁶ copies per mL. Genotype 1 (n = 60), genotype 2a (n = 26), genotype 3 (n = 11), genotype 4 (n = 4). Histological staging: 0/1 = 52, 2/3 = 49

Interventions

Amantadine group: interferon-alpha-2a 6 MU sc thrice weekly plus amantadine 100 mg twice daily orally for 12 months

Control group: interferon-alpha-2a 6 MU sc thrice weekly for 12 months

OutcomesMortality; liver-related morbidity; SAE; treatment discontinuation due to AE; number of patients without SVR; number of patients with detectable HCV RNA at EOT; number of patients without normalisation of ALT at EOT and at EOFU
Notes
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskRandomisation was performed according to centre, in blocks of 10 patients; insufficient information on sequence generation
Allocation concealment (selection bias)Unclear riskNot described
Blinding of participants and personnel (performance bias)
All outcomes
High riskNot placebo-controlled
Blinding of outcome assessment (detection bias)
All outcomes
High riskNot placebo-controlled. A single pathologist who was unaware of the patients' treatment and response to therapy scored the pre-therapy liver biopsies for hepatic inflammation and fibrosis, according to Scheuer system
Incomplete outcome data (attrition bias)
All outcomes
Low riskFollow-up information was available for all patients, including those who did not complete the 12-month course of therapy
Selective reporting (reporting bias)High riskAll the authors' study endpoints were discussed in the article. Not all reasonable outcomes were discussed
Other biasLow risk

Invested interest bias: commercial kits for quantitative HCV RNA measurements by Roche. No further funding by manufacturers of interferon or amantadine

No baseline imbalance; sample size calculation was not reported; the trial was not stopped early

Maynard 2006

Methods

Multicentre, randomised double-blind, placebo-controlled trial in non-responders

Trial duration: 48 weeks, follow-up 24 weeks

Enrolment began in October 2000 and the trial completed in May 2003

Participants

Country: France, 23 centres

Number of patients randomised: 200 (actually 202, but 2 patients from the control group did not receive any medication)

Amantadine group: mean age 47.1 (27 to 66) years, male/female: 74/27

Control group: mean age 46.8 (23 to 66) years, male/female: 74/26

Inclusion criteria: failed to respond to a single previous 24-week cycle of interferon/ribavirin combination therapy (at least 3 MIU interferon-alpha 3 times weekly and ribavirin at a minimum dose of 600 mg/day) (non-response was defined as persistent HCV RNA in the serum during the last month of treatment); elevated serum ALT; detectable HCV RNA; neutrophil count ≥ 1000/mm³, platelet count ≥ 100 giga/L, haemoglobin ≥ 10 g/dL; post-treatment liver biopsy within a year had to show a METAVIR histological score ≥ A1F1 and < F4
Exclusion criteria: co-infection with HBV or HIV; any other cause of liver disease; active drug abuse or alcohol consumption > 40 g/day; other clinically significant history or current diseases; previous amantadine use, systemic immunosuppressive or antiviral treatment during the last 24 weeks, and those with a history of interferon and/or ribavirin intolerance

Interventions

Amantadine group: peg interferon-alpha-2b at a dose of 1.5 mg/kg per week sc plus oral ribavirin 800 to 1200 mg/day and oral amantadine hydrochloride 2 x 100 mg/day for 48 weeks, n = 101

Control group: the same dose of peg interferon-alpha-2b and ribavirin plus a placebo, n = 99

For both groups, the dose of ribavirin was adjusted according to body weight (800 mg up to 65 kg weight, 1000 mg between 65 and 85 kg, and 1200 mg for weight of 85 kg or more). All drugs were started and stopped at the same time. Treatment was administered for 48 weeks regardless of the virological response during therapy. At the end of this treatment period, patients underwent a liver biopsy and were followed up for 24 weeks

OutcomesSustained virological response; biochemical response at week 72 (ALT normalisation); histological benefit; tolerance - virological and biochemical responses during therapy at weeks 12, 24, and 48
NotesAdditional information requested on 23 January 2012 from the last author Prof. Dr. C. Trepo
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskRandomisation was done using a random permuted blocks method
Allocation concealment (selection bias)Low riskThe randomisation process was generated by the Department of Biostatistics, Hospices Civils de Lyon, Lyon, France
Blinding of participants and personnel (performance bias)
All outcomes
Unclear riskInsufficient information, although placebo-controlled
Blinding of outcome assessment (detection bias)
All outcomes
Unclear riskInsufficient information, although placebo-controlled
Incomplete outcome data (attrition bias)
All outcomes
High riskWithdrawals were reported, but not all the reasons for withdrawal
Selective reporting (reporting bias)Low riskAll outcome measures were reported
Other biasHigh risk

Vested interest bias: high: supported by Schering-Plough

No baseline imbalance; sample size calculation was reported; the trial was not stopped early

Mendez-Navarro 2010

Methods

Randomised, double-blind, placebo-controlled trial in naive patients

Trial duration: 48 weeks, follow-up 24 weeks

Randomisation between March 2003 and June 2005

Participants

Country: Mexico, 1 centre

Number of patients randomised: 124.

Amantadine group: mean age 44 ± 12.29 years, male/female: 29/32

Control group: mean age 46.2 ± 9.82 years, male/female: 26/37

Inclusion criteria: men and women age 18 to 65 years with genotype 1 HCV infection defined by the presence of an HCV antibody, HCV RNA positive by RT-PCR, and genotype 1 infection; elevated serum ALT levels (40 IU/l) for at least 6 months; patients with cirrhosis were included only if they were Child-Pugh Class A (compensated disease); of Latino ethnicity (self identified as "Latino or Hispanic") with Spanish as their primary language and were born in the Mexican Republic; not previously been treated with interferon, peg interferon-alpha, ribavirin, and/or amantadine; pre-treatment liver biopsy was encouraged but not required
Exclusion criteria: other causes of liver disease; HIV infection, hepatitis B infection; complication of portal hypertension (variceal bleeding, ascites, encephalopathy, Child-Pugh B or C, hepatocellular carcinoma); haemoglobin < 12 g/dl, platelets < 70,000 plt/mm3; pregnancy; other clinically significant diseases; alcohol or drug abuse; refusal to use contraception during treatment

Interventions

Patients were randomly assigned to receive:

Amantadine group: peg interferon-alpha-2a 180 μg/week plus 1000 to 1200 mg/day ribavirin according to body weight
(1000 mg if < 75 kg or 1200 mg if ≥ 75 kg) plus amantadine 200 mg orally daily (amantadine hydrochloride 100 mg tablets) for 48 weeks, n = 61

Control group: the same regimen of peg interferon-alpha-2a plus ribavirin for 48 weeks, n = 63

OutcomesSustained virological response; early virological response; end of treatment response
NotesAdditional information requested on 23 January 2012 from the first author, Dr. J. Mendez-Navarro. Dr. Mendez-Navarro responded on 26 January 2012. More information was requested on 26 January, and Dr. Mendez responded the same day
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskThe randomisation was 1:1 in a balanced design and the method for random sequence generation was a computer-based random number system
Allocation concealment (selection bias)Low riskCentral telephone allocation for concealment
Blinding of participants and personnel (performance bias)
All outcomes
High riskNot placebo-controlled
Blinding of outcome assessment (detection bias)
All outcomes
High riskNot placebo-controlled
Incomplete outcome data (attrition bias)
All outcomes
Low riskWithdrawals reported; no large differences in withdrawals between the 2 groups
Selective reporting (reporting bias)High riskOutcome measures as described in the methods are reported, but some important outcome measures, for example biochemical response, are missing
Other biasHigh risk

Vested interest bias: Chung has received a research grant from Roche

No baseline imbalance; sample size calculation was not reported; the trial was not stopped early

Pessoa 2012

Methods

Randomised, controlled, multicentre trial in non-responders and relapsers

Trial duration: 48 weeks, follow-up 24 weeks

First patient enrolled in June 2003 and last patient completed follow-up in November 2005

Participants

Country: Brazil

Number of patients randomised: 186 (106 non-responders and 80 relapsers); 182 actually received treatment

The population was predominantly male, of white race, with a mean age of ± 50 years and a baseline HCV RNA level ≥ 800,00 IU/mL

Inclusion criteria: adults with a positive anti-HCV antibody test, detectable HCV RNA in serum; elevated ALT serum levels on at least 2 occasions during the previous 6 months; liver biopsy result within the previous 35 months consistent with the diagnosis of chronic HCV; at least 24 weeks of previous treatment with interferon-alpha plus ribavirin of which the outcome was either virological non-response or virological relapse; previous course completed at least 12 weeks prior to enrolment

Exclusion criteria: co-infection with hepatitis A or B or HIV; neutrophil count < 1500 cells/mm³, serum creatinine level > 1.5 times the upper limit of normal, or haemoglobin level < 12 g/dL (women) or < 13 g/dL (men); serious chronic diseases including severe psychiatric disease or alcohol or drug abuse within 1 year; pregnant or breastfeeding women and male partners of pregnant women

Interventions

Amantadine group: peg interferon sc 180 μg/week plus oral ribavirin 1000 mg/day (body weight ≤ 75 kg) or 1200 mg/day (body weight > 75 kg) plus oral amantadine 200 mg/day for 48 weeks, n = 94 (n = 92 actually received at least 1 dose of treatment)

Control group: peg interferon sc 180 μg/week plus oral ribavirin 1000 mg/day (body weight ≤ 75 kg) or 1200 mg/day (body weight > 75 kg), n = 92 (n = 90 actually received at least 1 dose of treatment)

OutcomesSustained virological response; sustained biochemical response; early virological response; complete early virological response; safety: adverse events and laboratory abnormalities
NotesAdditional information requested on 25 January 2012 from second author, Prof. Dr. H. Cheinquer. Dr. Cheinquer responded on 25 January with information about drop-outs due to AE and information on random sequence generation and allocation concealment
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskRandomisation was provided by a computerised system hosted by the study contract research organisation
Allocation concealment (selection bias)Low riskDelivered by phone to the site
Blinding of participants and personnel (performance bias)
All outcomes
High riskNot placebo-controlled
Blinding of outcome assessment (detection bias)
All outcomes
High riskNot placebo-controlled
Incomplete outcome data (attrition bias)
All outcomes
Low riskMissing outcome data balanced in numbers across intervention groups, with similar reasons for missing data across groups
Selective reporting (reporting bias)High riskAll the outcomes mentioned in methods are reported, but information on biochemical EOT response is lacking
Other biasHigh risk

Vested interest bias: high: Roche funding

No baseline imbalance; sample size calculation was reported; the trial was not stopped early

Piai 2003

Methods

Randomised clinical trial in relapsers in 1 referral hepatologic centre

Patients were enrolled between January 1999 and May 2000 and were followed up until November 2001

12 months therapy, 6 months follow-up

Participants

Country: Italy

49 patients entered the first period of the study

Inclusion criteria: previously received 1 or more course of recombinant or lymphoblastoid interferon at a dose ranging from 3 to 6 MU 3 times per week for 6 to 12 months and who had normalised serum ALT and cleared serum HCV RNA by PCR on therapy but subsequently relapsed within 6 months after stopping treatment; age between 18 and 65 years; time between last course of interferon and the start of combination therapy < 12 months, liver biopsy before enrolment < 24 months; HCV genotype 1b

Exclusion criteria: decompensated liver disease; HIV and HBV co-infection; other clinically significant diseases; haemoglobin < 13 g/dl for males and < 12 g/dl for females; platelet count < 100,000 and WBC < 3000

Amantadine group: 12 patients, mean age 51.2 ± 4.4 years, male/female = 11/1. Mean BMI 27.5 ± 1.8 kg/m2. Mean serum ALT 195 ± 108 IU/L and basal viral load > 1 million n = 4. Genotype 1b (n = 12). Histological staging: mean fibrosis score 2.4 ± 1.2

Control group: 12 patients, mean age 49.3 ± 10.0 years, male/female = 9/3. Mean BMI 26.7 ± 2.7 kg/m2. Mean serum ALT 184 ± 115 IU/L and basal viral load > 1 million n = 4. Genotype 1b (n = 12). Histological staging: mean fibrosis score 1.9 ± 1.1

Interventions

In the first part of the study, all 49 relapsers were treated for 6 months with recombinant interferon-alpha-2b, administered sc at a dose of 3 MU thrice a week, together with ribavirin, given orally twice a day, at a total dosage adjusted according to body weight (1000 mg for patients weighing ≤ 75 kg and 1200 mg for those > 75 kg). During the second part of the study, 24 patients who showed no biochemical and virological response, were randomised to continue treatment for further 6 months in 2 arms:

Amantadine group: interferon-alpha-2b plus ribavirin in the above mentioned dosages, plus oral amantadine hydrochloride 200 mg daily

Control group: interferon-alpha-2b plus ribavirin in the above mentioned dosages

OutcomesMortality; liver-related morbidity; SAE; treatment discontinuation due to AE; number of patients without SVR; number of patients with detectable HCV RNA at EOT; number of patients without normalisation of ALT at EOT
NotesOn 9 January 2012 ML sent G. Piai an email about the number of patients in both groups with normal ALT 6 months after cessation of therapy
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskInsufficient information
Allocation concealment (selection bias)Unclear riskSealed envelopes; unknown if they were opaque
Blinding of participants and personnel (performance bias)
All outcomes
High riskNot placebo-controlled
Blinding of outcome assessment (detection bias)
All outcomes
High riskNot placebo-controlled, not mentioned in article
Incomplete outcome data (attrition bias)
All outcomes
Low riskAll non-responders completed the second part of therapy
Selective reporting (reporting bias)High riskAll the authors' study endpoints were discussed in the article. Not all reasonable outcomes were discussed
Other biasUnclear risk

The study appears to be free of other sources of bias, but insufficient information

Baseline imbalance unknown; sample size calculation was reported; the trial was not stopped early

Salmeron 2007

Methods

Randomised, parallel-group trial in interferon non-responder patients in 36 centres

Patients were recruited between 1999 and 2001 and the follow-up finished in March 2003

48 weeks therapy, 24 weeks follow-up

Participants

Country: Spain

378 patients were randomised

Inclusion criteria: serum HCV RNA positivity by PCR before the beginning of the treatment and serum ALT activity above the upper limit of normal with at least 1 value during the 6-month period preceding the initiation of test drug dosing. All patients had chronic hepatitis without cirrhosis in the biopsy. The biopsies had been carried out up to a maximum of 3 years before entering the study. A treatment-free interval of at least 6 months was necessary between the first and the second course

Exclusion criteria: age 60 years or older, evidence of any cause of liver disease other than chronic HCV (co-infection with hepatitis B virus or HIV, concomitant autoimmune disease or metabolic disease). Clinically significant cardiovascular, renal, haematological, rheumatological, neurological or psychiatric disease, systemic infections, neoplastic disease, organ grafts and systemic immunosuppressive treatment. Active alcohol (alcohol intake > 40 g/day in females and > 60 g/day in males) or drug abuse within the previous year. Pregnancy or lactation period. Haemoglobin levels < 12 g/dL, white cell count < 3000/mm³, granulocyte count < 1500/mm³ or platelet count < 100,000/mm³

Amantadine group + interferon: 111 patients, mean age 44.7 ± 9 years, male/female = 87/24. Mean weight 78 ± 13 kg. Mean serum ALT 133 ± 90 UI/L, mean serum AST 94 ± 81 UI/L, and high serum HCV RNA titre > 8 x 105 UI/mL was detected in 34 out of 78 patients. Genotype 1: 72 out of 88, genotype non-1: 16 out of 88. Histological staging was not provided

Control group (interferon): 53 patients, mean age 45 ± 8 years, male/female = 40/13. Mean weight 74 ± 11 kg. Mean serum ALT 135 ± 89 UI/L, mean serum AST 103 ± 88 UI/L, and high serum HCV RNA titre > 8 x 105 UI/mL was detected in 12 out of 43 patients. Genotype 1: 40 out of 44, genotype non-1: 4 out of 44. Histological staging was not provided

Amantadine group + interferon + ribavirin: 108 patients, mean age 45.3 ± 8 years, male/female = 87/21. Mean weight 78 ± 13 kg. Mean serum ALT 125 ± 80 UI/L, mean serum AST 97 ± 76 UI/L, and high serum HCV RNA titre > 8 x 105 UI/mL detected in 27 out of 80 patients. Genotype 1: 74 out of 82, genotype non-1: 8 out of 82. Histological staging was not provided

Control group (interferon + ribavirin): 106 patients, mean age 46 ± 9 years, male/female = 85/21. Mean weight 77 ± 13 kg. Mean serum ALT 124 ± 92 UI/L, mean serum AST 79 ± 75 UI/L, and high serum HCV RNA titre > 8 x 105 UI/mL was detected in 29 out of 81 patients. Genotype 1: 74 out of 85, genotype non-1: 11 out of 85. Histological staging was not provided

Interventions

Amantadine group + interferon: interferon-alpha-2a, 9 MUI/day sc for 4 weeks and 3 MUI 3 times a week for a further 44 weeks plus amantadine chloride, 100 mg twice per day

Control group (interferon): interferon-alpha-2a, 9 MUI/day sc for 4 weeks and 3 MUI 3 times a week for a further 44 weeks

Amantadine group + interferon + ribavirin: the same doses of interferon-alpha-2a plus amantadine 100 mg twice per day, and ribavirin 1000 to 1200 mg per day according to weight

Control group (interferon + ribavirin): the same doses of interferon-alpha-2a and ribavirin

OutcomesMortality; liver-related morbidity; SAE; treatment discontinuation due to AE; number of patients without SVR; number of patients with detectable HCV RNA at EOT
NotesML sent an email to Dr. Salmeron about the ALT values at EOT and at 6 months follow-up on 10 January 2012
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskPatients were selected randomly by central telephone
Allocation concealment (selection bias)Low riskPatients were selected randomly by central telephone
Blinding of participants and personnel (performance bias)
All outcomes
High riskNot placebo-controlled
Blinding of outcome assessment (detection bias)
All outcomes
High riskNot placebo-controlled
Incomplete outcome data (attrition bias)
All outcomes
High riskAll drop-outs were discussed, but not equally divided over groups
Selective reporting (reporting bias)High riskNot all of the study's prespecified primary outcomes have been reported. The biochemical response (normalisation of serum ALT) was not reported
Other biasHigh risk

Vested interest bias: Roche

No baseline imbalance; sample size calculation was reported; the trial was stopped early at 378 patients randomised (instead of 1100 patients calculated) due to poor results

Sax 2001

Methods

Randomised clinical trial in patients co-infected with HIV

Patients were recruited at 2 university outpatient clinics and were enrolled within a 4-month period

Trial duration: 12 months, follow-up 6 months

Participants

Country: Switzerland

7 patients were randomised: 3 female; mean age 40 years, range 28 to 54 years

Inclusion criteria: patient's triple antiretroviral treatment was unchanged for at least 2 months; > 200 CD4+ lymphocytes/l; < 50,000 HIV-1 RNA copies/mL; elevated transaminases for at least 6 months; biopsy results were compatible with HCV infection

Exclusion criteria: decompensated liver cirrhosis; additional liver diseases; ongoing illicit drug use; contraindications for interferon

Interventions

Amantadine group: interferon-alpha, 6 MU/day for 1 month and 6 MU thrice weekly for the remaining 11 months combined with amantadine sulphate 100 mg bid orally, n = 3

Control group: interferon-alpha, 6 MU/day for 1 month and 6 MU thrice weekly for the remaining 11 months alone, n = 4

OutcomesMortality; number of patients without SVR
NotesThe trial stopped early due to important toxicities and low tolerability of interferon-alpha
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskInsufficient information
Allocation concealment (selection bias)Unclear riskInsufficient information
Blinding of participants and personnel (performance bias)
All outcomes
High riskNot placebo-controlled
Blinding of outcome assessment (detection bias)
All outcomes
High riskNot placebo-controlled
Incomplete outcome data (attrition bias)
All outcomes
Low riskNo missing data
Selective reporting (reporting bias)High riskEndpoint not clearly stated. Not all reasonably expected outcomes were discussed
Other biasLow risk

No vested interest bias (low risk of bias regarding vested interest)

Baseline imbalance unknown; sample size calculation was not reported; the trial was not stopped early

Shakil 2000

Methods

Randomised clinical trial in naive patients

Trial duration: 24 weeks of therapy, 24 weeks follow-up

Participants

Country: United States of America

24 patients were randomised

Inclusion criteria: ≥ 18 years, elevated serum ALT levels, positive anti-HCV and HCV RNA in serum, and chronic hepatitis on liver biopsy

Exclusion criteria: HBsAg positivity, HIV, Child's B or C cirrhosis

Amantadine group: 12 patients, mean age 46 years, mean serum ALT 87 IU/L, mean viral load 83 x 105 eq/mL, and histological staging was 1.2. Genotype and male/female ratio were not provided

Control group: 12 patients, mean age 46 years, mean serum ALT levels were 72 IU/L, mean viral load was 76 x 105 eq/mL, and histological staging was 1.4. Genotype and male/female ratio were not provided

Interventions

Amantadine group: interferon-alpha-2a 3 MU sc 3 times a week, and amantadine 100 mg orally twice a day for 24 weeks

Control group: interferon-alpha-2a 3 MU sc 3 times a week for 24 weeks

OutcomesMortality; treatment discontinuation due to AE; number of patients without SVR; number of patients with detectable HCV RNA at EOT; number of patients without normalisation of ALT at EOT
NotesML sent an email to Dr. Shakil on 12 January 2012 about biochemical responses, SVR, and EOT response. Dr. Shakil responded on 13 January 2012. ML sent another email on 26January 2012 about SAE and death. Dr. Shakil responded on 3 1January 2012 (no SAE, no death)
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskInsufficient information
Allocation concealment (selection bias)Unclear riskInsufficient information
Blinding of participants and personnel (performance bias)
All outcomes
Unclear riskInsufficient information, although trial was placebo-controlled
Blinding of outcome assessment (detection bias)
All outcomes
Unclear riskInsufficient information, although trial was placebo-controlled
Incomplete outcome data (attrition bias)
All outcomes
High riskInsufficient information: 2 patients in the amantadine group withdrew, 4 in the placebo group withdrew; reasons unknown
Selective reporting (reporting bias)Unclear riskInsufficient information
Other biasUnclear riskInsufficient information

Smith 2004

Methods

Randomised, double-blind, placebo-controlled, cross-over trial in interferon failures or those not candidates for interferon

48 weeks therapy, 24 weeks follow-up

Participants

Country: United States of America

152 patients were enrolled in the study

Inclusion criteria: previous failed interferon, intolerant of interferon side effects, or not candidates for interferon therapy due to either depression, neutropenia, or thrombocytopenia. In patients who had previously been treated with interferon, a period of 6 months off therapy and a liver biopsy were required for enrolment; age between 1 and 65 years; patiens over the age of 65 years were eligible if chest x-ray, electrocardiogram, and creatinine clearance were normal prior to enrolment; abnormal liver enzymes, detectable HCV RNA; inflammation by liver biopsy; females of childbearing potential were required to use medially accepted contraceptive regimens if sexually active; normal laboratory values for albumin, prothrombin time, creatinine, haemoglobin, leukocyte count, antinuclear antibody, platelet count, and alpha-fetoprotein

Exclusion criteria: evidence of decompensated liver disease; other forms of liver disease; active HIV infection; other serious medical conditions; active using illicit drugs or alcohol; antiviral medications, oral steroids, immunosuppressive medications, or anticoagulation therapy

Amantadine group: 73 patients, age > 50 years, n = 13, male/female = 52/21. Serum ALT not provided and the viral load > 200 MEq/mL, n = 21. Genotype 1 (n = 57), genotype 2 (n = 9), genotype 3 (n = 6), and genotype 4 (n = 1). Histological staging: severe liver histology stage 3/4 = 34

Control group: 79 patients, age > 50 years, n = 18, male/female = 50/29. Serum ALT not provided and the viral load > 200 MEq/mL, n = 22. Genotype 1 (n = 59), genotype 2 (n = 9), genotype 3 (n = 5), and genotype 4 (n = 4), 2 patients could not be genotyped by 2 separate laboratories. Histological staging: severe liver histology stage 3/4 = 39

Interventions

Amantadine group: amantadine 100 mg by mouth twice daily

Control group: placebo twice daily

Both groups received amantadine or placebo for 6 months. After 6 months, patients receiving the placebo were crossed over to amantadine therapy for 6 months, while those on amantadine continued on this treatment for 6 additional months

OutcomesMortality; liver-related morbidity; SAE; treatment discontinuation due to AE; QoL; number of patients without normalisation of ALT at EOT
Notes

Amantadine was supplied by Endo Pharmaceuticals Inc. (Chadds Ford, Pa)

This trial used a 5-way stratification; with this amount of patients this could lead to over-stratification

ML sent an email to Dr. Smith on 9 January 2012 about exact biochemical and virological responses

Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskA stratified randomised scheme was invoked. SAS statistical software was used to generate the treatment codes within each of the 32 strata for implementation by the pharmacy
Allocation concealment (selection bias)Low riskSAS statistical software was used to generate the treatment codes within each of the 32 strata for implementation by the pharmacy
Blinding of participants and personnel (performance bias)
All outcomes
Low riskIdentical ghost capsules were filled by the pharmacist with amantadine and sucrose, so that neither staff nor patients could distinguish between placebo and active drug
Blinding of outcome assessment (detection bias)
All outcomes
Low riskIdentical ghost capsules were filled by the pharmacist with amantadine and sucrose, so that neither staff nor patients could distinguish between placebo and active drug
Incomplete outcome data (attrition bias)
All outcomes
Unclear riskAll drop-outs were mentioned, but uncertain if they were equally divided
Selective reporting (reporting bias)High riskAll the authors' study endpoints were discussed in the article. Not all reasonably expected outcomes were discussed
Other biasUnclear risk

Vested interest bias: unclear

No baseline imbalance; sample size calculation was reported; the trial was not stopped early

Tabone 2001

Methods

Randomised clinical trial in 9 different medical centres in naive patients

Patients were enrolled between September 1998 and April 1999

12 months therapy, 6 months follow-up

Participants

Country: Italy

180 patients were randomised

Inclusion criteria: positive for anti-HCV and for HCV RNA, liver biopsy within a year before entry in the study showing chronic hepatitis without cirrhosis, and serum ALT levels elevated at least 1.5 times the upper limit of normal (40 IU/L) on 3 determinations before enrolment

Exclusion criteria: chronic alcohol abuse, active drug addiction, hepatitis B or HIV co-infection, evidence of autoimmune disease, platelet count < 100,000/µL, leukocyte count < 2500/µL, other clinically significant diseases, and pregnancy

Amantadine group: 90 patients, mean age 42 ± 12 years, male/female = 62/28. Serum ALT 103 (56 to 400) U/L and median basal viral load 2.4 (0.2 to 32). Genotype 1 + 4 (n = 47), genotype 2 + 3 (n = 43). Histological staging: mean 3.4 ± 0.3

Control group: 90 patients, mean age 44 ± 12 years, male/female = 67/23. Serum ALT 114 (65 to 274) U/L and median basal viral load 2.54 (0.2-26). Genotype 1 + 4 (n = 53), genotype 2 + 3 (n = 37). Histological staging: mean 3.2 ± 0.3

Interventions

Amantadine group: interferon-alpha-2a 6 MU sc every other day for 6 months and then 3 MU sc every other day for the other 6 months plus amantadine 100 mg twice daily oral for 12 months

Control group: interferon-alpha-2a 6 MU sc every other day for 6 months and then 3 MU sc every other day for the other 6 months

OutcomesMortality; SAE; treatment discontinuation due to AE
Notes
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskRandomisation was centralised with a 1:1 ratio
Allocation concealment (selection bias)Low riskRandomisation was centralised with a 1:1 ratio
Blinding of participants and personnel (performance bias)
All outcomes
High riskNot placebo-controlled
Blinding of outcome assessment (detection bias)
All outcomes
High riskNot placebo-controlled
Incomplete outcome data (attrition bias)
All outcomes
Low riskDrop-outs mentioned; 5 in control group 8 in amantadine group
Selective reporting (reporting bias)High riskAll the authors' study endpoints were discussed in the article. Not all reasonably expected outcomes were discussed
Other biasLow risk

Vested interest bias: no support

No baseline imbalance; sample size calculation was reported; the trial was not stopped early

Teuber 2001

Methods

Randomised, placebo-controlled, double-blind trial in primary interferon-alpha in primary interferon-alpha non-responders

48 weeks therapy, 24 weeks follow-up

Participants

Country: Germany

55 patients were randomised

Inclusion criteria: non-response to previous interferon-alpha monotherapy with persistence of serum HCV RNA and a treatment-free interval of at least 24 weeks; elevated ALT levels; positive anti-HCV test; detectable serum HCV RNA; compensated liver disease; leukocyte count ≥ 2500/μL, platelet count ≥ 70,000/μL; aged between 18 and 70 years

Exclusion criteria: co-infection with hepatitis B and HIV, concomitant autoimmune disease, other clinically significant disease. Average daily intake of alcohol exceeding 50 g of ethanol or drug abuse within the previous year. Pregnancy and lactation period

Amantadine group: 59 patients, mean age 47.7 ± 10.5 years, male/female = 19/7. Serum ALT 73 ± 54 U/L, serum AST 38 ± 28 U/L, and basal viral load 630 ± 567 x 10³ copies per mL. Genotype 1 (n = 22), genotype non-1 (n = 4). Histological staging: non = 2, mild = 7, moderate = 10, severe = 7

Control group: 29 patients, mean age 45.7 ± 10.3 years, male/female = 17/12. Serum ALT 64 ± 44 U/L, serum AST 37 ± 25 U/L, and basal viral load 890 ± 823 x 10³ copies per mL. Genotype 1 (n = 27), genotype non-1 (n = 2). Histological staging: non = 2, mild = 9, moderate = 12, severe = 6

Interventions

Amantadine group: interferon-alpha-2a 6 MU sc thrice weekly for 24 weeks, followed by 3 MU sc thrice weekly for an additional 24 weeks and oral amantadine sulphate 100 mg twice daily

Control group: interferon-alpha-2a 6 MU sc thrice weekly for 24 weeks, followed by 3 MU sc thrice weekly for an additional 24 weeks and oral placebo twice daily

OutcomesMortality; liver-related morbidity; SAE; treatment discontinuation due to AE; QoL; number of patients without SVR; number of patients with detectable HCV RNA at EOT; number of patients without normalisation of ALT at EOT and at EOFU
Notes
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskRandom number generator in fixed blocks of 4 with a ratio of 1:1
Allocation concealment (selection bias)Unclear riskInsufficient information
Blinding of participants and personnel (performance bias)
All outcomes
Low riskPlacebo-controlled with a matched placebo
Blinding of outcome assessment (detection bias)
All outcomes
Unclear riskPlacebo-controlled, but insufficient information
Incomplete outcome data (attrition bias)
All outcomes
Low riskDrop-outs reported and equally divided over 2 groups
Selective reporting (reporting bias)Low riskAll important outcome measures were mentioned
Other biasUnclear risk

Vested interest bias: insufficient information

No baseline imbalance; sample size calculation was not reported; the trial was not stopped early

Teuber 2002

Methods

Randomised clinical trial in relapsing patients

48 weeks therapy, 24 weeks follow-up

Participants

Country: Germany

75 patients were randomised. 46 males, 29 females, mean age 43 years

Inclusion criteria: relapsing after primary successful antiviral treatment

Amantadine group: 41 patients, mean age, male/female, serum ALT, and viral load were not provided. Genotype distribution and histological staging also were not provided

Control group: 34 patients, mean age, male/female, serum ALT, and viral load were not provided. Genotype distribution and histological staging also were not provided

Interventions

Amantadine group: interferon-alpha-2b 5 MU daily for 4 weeks, 5 MU 3 times a week for 20 weeks followed by 3 MU 3 times a week for another 24 weeks in combination with daily 1000 to 1200 mg ribavirin plus 100 mg amantadine twice daily for 48 weeks

Control group: interferon-alpha-2b 5 MU daily for 4 weeks, 5 MU 3 times a week for 20 weeks followed by 3 MU 3 times a week for another 24 weeks in combination with daily 1000 to 1200 mg ribavirin for 48 weeks

Treatment was discontinued in patients with detectable serum HCV RNA after treatment week 24

OutcomesNumber of patients without SVR
NotesML sent an email to Dr. Teuber on 12 January 2012 about virological EOT and biochemical responses
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskInsufficient information
Allocation concealment (selection bias)Unclear riskInsufficient information
Blinding of participants and personnel (performance bias)
All outcomes
High riskNot placebo-controlled
Blinding of outcome assessment (detection bias)
All outcomes
High riskNot placebo-controlled
Incomplete outcome data (attrition bias)
All outcomes
Unclear riskInsufficient information
Selective reporting (reporting bias)Unclear riskInsufficient information
Other biasUnclear riskInsufficient information

Teuber 2003

Methods

Randomised controlled trial in non-responders to previous antiviral treatment in 10 centres

Patients were enrolled between July 1998 and September 1999

48 weeks therapy, 24 weeks follow-up

Participants

Country: Germany

225 patients were randomised

Inclusion criteria: chronic HCV not responding to 1 or more courses of interferon-alpha with a minimal total dose of at least 108 MU for 12 weeks (e.g., at least 3 x 3 MIU tiw) alone or in combination with ribavirin and/or amantadine; documented non-response with persistently detectable serum HCV RNA during the entire, most recent antiviral treatment; treatment-free interval of at least 6 months; positive anti-HCV antibody test; detectable serum HCV-RNA; elevated ALT levels; compensated liver disease; leucocyte count ≥ 2500/μl, platelet count ≥ 70,000/μl, haemoglobin ≥12.0 g/dL in females and haemoglobin ≥13.0 g/dL in males; and patient's age ≥18 years

Exclusion criteria: co-infection with hepatitis B virus or HIV types 1 and 2, concomitant autoimmune disease,
clinically significant cardiovascular, metabolic, renal, haematological, rheumatological, neurological or psychiatric disease, systemic infections, neoplastic disease, organ grafts, systemic immunosuppressive treatment, active alcohol or drug-abuse within the previous year, pregnancy or lactation period

Amantadine group: 115 patients, age 48 (20 to 72) year, male/female = 74/41. Median serum ALT 49 (19 to 254) U/l, median serum AST 27 (10 to 212) U/l, and median basal viral load 1.0 (0.04 to 268) x 10⁶ copies per mL. Genotype 1 (n = 102) and genotype non-1 (n = 13). Histological staging (n=110): mild = 56, moderate = 40, severe = 5, and cirrhosis = 9

Control group: 110 patients, age 46 (24 to 71) year, male/female = 69/41. Median serum ALT 50 (18 to 762) U/l, median serum AST 28 (10- to 1206) U/l, and median basal viral load 1.0 (0.02 to 22.7) x 10⁶ copies per mL. Genotype 1 (n = 97) and genotype non-1 (n = 13). Histological staging (n=105): mild = 57, moderate = 29, severe = 12, and cirrhosis = 7

Interventions

Amantadine group: 5 MU interferon-alpha-2b daily for the initial 4 weeks, followed by 5 MU interferon-alpha-2b thrice weekly sc for further 20 weeks and subsequently 3 MU interferon-alpha-2b thrice weekly sc for additional 24 weeks plus ribavirin 1000 to 1200 mg/day combined with amantadine sulphate 200 mg/day

Control group: 5 MU interferon-alpha-2b daily for the initial 4 weeks, followed by 5 MU interferon-alpha-2b thrice weekly sc for further 20 weeks and subsequently 3 MU interferon-alpha-2b thrice weekly sc for additional 24 weeks plus ribavirin 1000 to 1200 mg/day

After treatment week 24, antiviral treatment was only continued in patients with undetectable serum HCV RNA at treatment week 20

OutcomesMortality; treatment discontinuation due to AE; number of patients without SVR; number of patients with detectable HCV RNA at EOT; number of patients without normalisation of ALT at EOT and at EOFU
NotesML sent an email to Dr. Teuber on 11 January 2012 about SAE distribution, liver-related morbidity, and baseline characteristics
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskNot mentioned
Allocation concealment (selection bias)Unclear riskNot mentioned
Blinding of participants and personnel (performance bias)
All outcomes
High riskNot placebo-controlled
Blinding of outcome assessment (detection bias)
All outcomes
High riskNot placebo-controlled
Incomplete outcome data (attrition bias)
All outcomes
High riskPatients lost to follow-up were not equally divided over the 2 treatment groups. Not described why they were lost to follow-up
Selective reporting (reporting bias)High riskThere was no protocol, but all the authors' study endpoints were discussed in the article. Not all reasonably expected outcomes were discussed nor clearly stated which groups the patients were in (for example, SAE distribution)
Other biasLow risk

Vested interest bias: no support from pharmacy

No baseline imbalance; sample size calculation was reported; the trial was not stopped early

Thuluvath 2004

Methods

Randomised, double-blind, placebo-controlled trial in naive patients in 9 centres

48 weeks therapy, 24 weeks follow-up

Participants

Country: United States of America

171 patients were randomised into the study

Inclusion criteria: adult patients with chronic HCV who had no previous treatment for HCV; HCV RNA detectable by PCR, evidence of liver disease ALT or AST above the upper limit of normal, liver biopsy (within 3 months), and no known contraindications to treatment with interferon, ribavirin, or amantadine. Stress testing was required for patients at high risk for coronary artery disease, and only patients demonstrating euthyroid state were enrolled

Exclusion criteria: haemolytic anaemia, hepatocellular carcinoma, renal failure, and seizure disorders; concomitant hepatitis B virus or HIV infection, immunosuppressed state, active substance abuse, decompensated liver disease, major psychiatric disorders, life expectancy less than 5 years, or daily alcohol intake over 10 g/day; haemoglobin < 12 g/dl, white blood cell count < 3000, platelet count < 70,000, serum bilirubin > 3 mg/dl, serum creatinine > 1.2 mg/dl, and a positive pregnancy test. Women and men of childbearing age were required to practice medically acceptable methods of contraception

Amantadine group: 85 patients, age < 50 years: 70, age > 50 years: 15; male/female = 45/40. Serum ALT was 103 ± 126 U/L and viral load < 1 x 106 copies per mL: 51, viral load > 1 x 106 copies per mL: 34. Genotype 1a/b (n = 74), genotype 2/3 (n = 9), unable to genotype (n = 2). Histological staging: minimal/no fibrosis = 69, cirrhosis/septate fibrosis = 16

Control group: 86 patients, age < 50 years: 65, age > 50 years: 21; male/female = 55/31. Serum ALT was 90 ± 66 U/L and viral load < 1 x 106 copies per mL: 49, viral load > 1 x 106 copies per mL: 37. Genotype 1a/b (n = 71), genotype 2/3 (n = 13), unable to genotype (n = 2). Histological staging: minimal/no fibrosis = 64, cirrhosis/septate fibrosis = 22

Interventions

Amantadine group: interferon-alpha-2b sc 3 million units 3 times a week, ribavirin 1000 to 1200 mg (based on body weight) daily in divided doses, and amantadine hydrochloride 100 mg twice daily

Control group: interferon-alpha-2b sc 3 million units 3 times a week, ribavirin 1000 to 1200 mg (based on body weight) daily in divided doses, and placebo twice daily

OutcomesMortality; liver-related morbidity; SAE; treatment discontinuation due to AE; number of patients without SVR; number of patients with detectable HCV RNA at EOT
NotesML sent an email to Dr Thuluvath about the biochemical response on 10 January 2012
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskNot mentioned
Allocation concealment (selection bias)Low riskPatients were randomly assigned (central randomisation at Johns Hopkins University)
Blinding of participants and personnel (performance bias)
All outcomes
Low riskThe pharmacy department at the Johns Hopkins Hospital was responsible for randomisation and supplying amantadine or identical placebo to all centres. Unblinding of amantadine was done only when all patients completed treatment or if any patient experienced unexpected side effects (this was not necessary as there were no serious adverse events)
Blinding of outcome assessment (detection bias)
All outcomes
Unclear riskInsufficient information, although placebo-controlled
Incomplete outcome data (attrition bias)
All outcomes
Unclear riskAll drop-outs were discussed, but not clearly mentioned which patients in each group stopped for which reason
Selective reporting (reporting bias)High riskNot all the secondary outcome measures are well described; very sober description of ALT normalisation, no description of histological findings
Other biasHigh risk

Vested interest bias: Schering-Plough

No baseline imbalance; sample size calculation was reported; the trial was not stopped early

van Soest 2010

Methods

Randomised, double-blind, placebo-controlled trial in naive patients

Trial duration: 52 weeks, follow-up 52 weeks

Conducted from January 2001 to July 2007

Participants

Country: the Netherlands, 26 centres

Number of patients randomised: 321, only 297 really received allocated intervention

Amantadine group: mean age 42.6 ± 9.1 years, male/female: 108/32

Control group: mean age 43.8 ± 9.2 years, male/female: 105/48

Inclusion criteria: previously untreated adult patients who tested positive for serum HCV antibodies and HCV RNA; ALT and/or AST elevated at least once within 6 months before inclusion; liver biopsy (performed within 1 year before entry) consistent with chronic viral hepatitis; minimal baseline haematological values were: haemoglobin 6.5 mmol/L, white blood cells 2.5 × 109 L−1,
neutrophils 1.5 × 109 L−1, platelets 70 × 109 L−1 and serum creatinine < 150 mol/L
Exclusion criteria: Child-Pugh classification B or C; HIV co-infection; active uncontrolled psychiatric disorders; significant dysfunction of the central nervous system; chemotherapy and/or systemic antiviral treatment in the preceding 6 months; other serious disease; pregnancy or intention to get pregnant or unwillingness to use contraception; (former) drug users could be included if stable psychosocial situation, support and housing were available

Interventions

Amantadine group: peg interferon-alpha-2b, ribavirin, plus amantadine hydrochloride for 48 weeks, n = 144

Control group: peg interferon-alpha-2a, ribavirin, plus oral placebo of identical shape and taste was added for 48 weeks, n = 153

Both treatment groups received the same interferon-alpha induction therapy (from day 1 combined with ribavirin), consisting of interferon-alpha-2b 10 MIU/day sc during the first 6 days, followed by 5 MIU/day for the next 6 days, followed by peg interferon-alpha-2b 1.5 g/kg/week sc up to 26 weeks and 1.0 g/kg/week from week 26 to week 52. Oral ribavirin was given during the entire 52-week treatment period in 2 different doses: 1000 mg/day for body weight < 75 kg and 1200 mg/day for body weight ≥ 75 kg. In the triple therapy group, oral amantadine hydrochloride 100 mg twice daily was added

OutcomesSustained virological response, 1 year after cessation of the study medication; virologic response rates (negative HCV RNA at week 24); breakthrough rates (negative HCV RNA at week 24 and positive HCV RNA at week 52); relapse rates (negative HCV RNA at week 24 and 52; positive HCV RNA at week 104)
Notes
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskCentral randomisation was implemented by the pharmacist of the co-ordinating academic centre using a block size of 4
Allocation concealment (selection bias)Low riskCentral randomisation was implemented by the pharmacist of the co-ordinating academic centre using a block size of 4
Blinding of participants and personnel (performance bias)
All outcomes
Low riskOral placebo of identical shape and taste
Blinding of outcome assessment (detection bias)
All outcomes
Unclear riskInvestigators and patients were blinded to treatment assignment during the entire study and follow-up period
Incomplete outcome data (attrition bias)
All outcomes
Unclear riskPatient withdrawal is mentioned, but the reason for withdrawal is not stated clearly for every patient
Selective reporting (reporting bias)High riskAll outcome measures reported; lack of biochemical response
Other biasHigh risk

Vested interest bias: Schering-Plough

No baseline imbalance; sample size calculation was reported; the trial was not stopped early

Vardar 2001

Methods

Randomised controlled trial in naive patients in Turkey

6 months therapy, 6 months follow-up

Participants

Country: Turkey

33 patients were randomised. They had biopsy-proven chronic HCV with raised ALT values equal to or greater than 1.5 times the upper normal limit, positive serum HCV RNA by PCR testing, and compensated liver disease

Amantadine group: 19 patients, mean age, male/female, serum ALT, viral load, and histological staging were not provided. All patients had genotype 1b

Control group: 14 patients, mean age, male/female, serum ALT, viral load, and histological staging were not provided. All patients had genotype 1b

There was no difference in the means for age, sex, and initial serum transaminase values between the 2 groups

Interventions

Amantadine group: interferon 3 MU 3 times per week plus amantadine 200 mg per day combination therapy for 6 months

Control group: interferon 3 MU 3 times per week for 6 months

Follow-up period was 6 months

OutcomesNumber of patients without SVR; number of patients with detectable HCV RNA at EOT
NotesML sent an email to Dr. Vardar on 12 January 2012 about biochemical responses
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Unclear riskInsufficient information
Allocation concealment (selection bias)Unclear riskInsufficient information
Blinding of participants and personnel (performance bias)
All outcomes
High riskNot placebo-controlled
Blinding of outcome assessment (detection bias)
All outcomes
High riskNot placebo-controlled
Incomplete outcome data (attrition bias)
All outcomes
Unclear riskInsufficient information
Selective reporting (reporting bias)High riskOnly virological response
Other biasUnclear riskInsufficient information

von Wagner 2008

Methods

Randomised, double-blind, placebo-controlled trial in naive patients

Trial duration: 48 weeks, follow-up 24 weeks

Participants

Country: Germany, 5 centres

Number of patients randomised: 705

Amantadine group: mean age 46.3 ± 12.0 years, male/female: 185/167

Control group: mean age 45.4 ± 12.2 years, male/female: 183/169

Inclusion criteria: male and female patients older than 18 years with compensated chronic genotype HCV-1-infection not previously treated with interferon-alpha or ribavirin; positive anti-HCV antibody and HCV RNA (600 IU/mL by quantitative RT-PCR); liver biopsy taken within 24 months before the screening visit showing chronic hepatitis; at least 1 serum ALT level elevated during the screening period; baseline neutrophil and platelet counts ≥ 1500/L and 90,000/L; haemoglobin values ≥ 12 g/dL for females and ≥ 13 g/dL for males
Exclusion criteria: any other cause of liver disease or other relevant disorders including HIV or hepatitis B virus co-infection; other clinically significant disease; excessive daily intake of alcohol, or drug abuse within the past year; pregnancy and lactation, and male partners of pregnant women; higher degree of atrioventricular block, bradycardia (heart rate 55 beats/minute), an implanted pacemaker, prolonged Q-T-interval, or a U wave in electrocardiogram, or concomitant intake of medication with long Q-T-interval as a known side effect, concomitant medication with thiazides, known history of severe ventricular arrhythmia

Interventions

Patients were randomly assigned to receive:

Amantadine group: amantadine-sulphate 400 mg/day orally in combination with peg interferon-alpha-2a 180 μg once per week sc plus ribavirin 1000 to 1200 mg/day orally according to body weight (75 kg: 1000 mg; 75 kg: 1200 mg), n = 353

Control group: placebo plus the same regimen peg interferon-alpha-2a plus ribavirin for 48 weeks, n = 352

Before onset of antiviral treatment with peg interferon-alpha-2a and ribavirin, amantadine/placebo was dose escalated within 2 weeks in 100 mg steps weekly starting at 200 mg/day

OutcomesSustained virological response
NotesAdditional information requested on 23 January 2012 from the last author, Prof. Dr. S. Zeuzem. Prof Zeuzem answered on 24 January with information about random sequence generation and allocation concealment. ML sent another email requesting the number of patients with liver-related morbidity. Prof. Zeuzem answered on 26 January 2012
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskComputer random number generator
Allocation concealment (selection bias)Low riskCentral allocation using telephone
Blinding of participants and personnel (performance bias)
All outcomes
Unclear riskPlacebo-controlled, but insufficient information
Blinding of outcome assessment (detection bias)
All outcomes
Unclear riskPlacebo-controlled, but insufficient information
Incomplete outcome data (attrition bias)
All outcomes
High riskWithdrawals/loss to follow-up not equally divided
Selective reporting (reporting bias)Unclear riskOnly 1 outcome measure, but more outcome measures reported in results. No information about histological improvement
Other biasHigh risk

Vested interest bias: high: Roche

No baseline imbalance; sample size calculation was not reported; the trial was not stopped early

Wenger 2007

Methods

Randomised controlled study in interferon-alpha non-responders on behalf of the Swiss Association of the Study of the Liver in 11 centres

Patients were recruited between August 1999 and June 2001

48 weeks therapy, 6 months follow-up

Participants

Country: Switzerland

32 patients were recruited for the pilot phase, of whom 2 (1 in the triple, 1 in the combination therapy group) withdrew written consent after baseline evaluation but before starting treatment. 30 patients started the pilot phase

Inclusion criteria: patients of both genders, aged 18 to 65 years, with chronic HCV who had previously failed to respond to interferon-alpha-2a or -2b given in a dose of 3 to 6 MU 3 times weekly for at least 12 weeks, elevated ALT within 12 months of entry on at least 2 occasions, positive HCV RNA test in serum by RT-PCR within 2 months of entry, and a liver biopsy within 5 years before entry consistent with chronic HCV

Exclusion criteria: any other cause of liver disease including hepatitis B virus co-infection (HBsAg positive) and alcohol intake (> 20 g/day in females and > 40 g/day in males); a history of or actual decompensation of liver disease (ascites, variceal bleeding or encephalopathy); cirrhosis ≥ 8 Child-Pugh points; other clinically relevant disorders including cardiovascular, pulmonary, renal, metabolic, haematological, rheumatologic, neurological and psychiatric diseases, autoimmune disorders, HIV infection, immunosuppression within 12 months of entry, organ transplantation, malignant neoplastic disease within 2 years of study entry, illicit drug use within 1 year of study entry or psychosocial instability, pregnancy or lactation, refusal to practice effective contraception during treatment and follow-up, or treatment with any investigational drug within 6 months of study entry; leucocytes < 2000/µL, neutrophils < 1000/µL, platelets < 50,000/µL, serum creatinine > 1.5 times upper limit of normal, elevated thyroid-stimulating hormone, alfa-fetoprotein above normal limits and/or focal lesion on ultrasound performed within 1 month of study entry.

Amantadine group: 16 patients, median age 47 (28 to 65) years, male/female = 13/3, BMI 25 (20 to 33) kg/m2. Median serum ALT was 77 (48 to 567) U/l and the median basal viral load was 4.3 (0.16 to 25) x 10⁶ copies per mL. Genotype 1 + 4 (n = 9) and genotype 2 + 3 (n = 7). Histological staging: 3 patients had cirrhosis

Control group: 14 patients, median age 45 (23 to 59) years, male/female = 12/2, BMI 26 (21 to 40) kg/m2. Median serum ALT was 89 (53 to 397) U/l and the median basal viral load was 2.0 (0.12 to 26.2) x 10⁶ copies per mL. Genotype 1 + 4 (n = 11) and genotype 2 + 3 (n = 3). Histological staging: 1 patient had cirrhosis

Sample size calculation was not mentioned

Interventions

Amantadine group: interferon-alpha-2a 6 MIU sc daily for 4 weeks, followed by 6 MIU sc tiw for an additional 44 weeks, ribavirin (< 75 kg: 1000, ≥ 75 kg: 1200 mg), plus amantadine sulphate 100 mg po twice daily

Control group: interferon-alpha-2a plus ribavirin in the above mentioned doses

Treatment was stopped, if after 4 weeks HCV RNA in serum remained detectable by RT-PCR (detection limit: 1000 copies/mL)

Patients were followed for 24 weeks after stopping therapy

OutcomesMortality; liver-related morbidity; SAE; treatment discontinuation due to AE; number of patients without SVR; number of patients with detectable HCV RNA at EOT; number of patients without normalisation of ALT at EOT and at SVR
NotesML sent an email about the biochemical response at EOT and 24 weeks after stopping therapy to Dr. Beat Mullhaupt on10 January 2012. Dr. Mullhaupt responded on 11 January 2012 with the following information: biochemical response at EOT in the triple therapy 4 out of 16, sustained biochemical response 3 out of 16. In the double therapy arm EOT was 3 out of 14, as well as the sustained biochemical response
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskPatients were randomised with a ratio of 1:1. Randomisation was carried out in blocks of 10 using random numbers
Allocation concealment (selection bias)Unclear riskInsufficient information, method is not described
Blinding of participants and personnel (performance bias)
All outcomes
High riskNot placebo-controlled
Blinding of outcome assessment (detection bias)
All outcomes
High riskNot placebo-controlled
Incomplete outcome data (attrition bias)
All outcomes
Low riskDrop-outs were discussed
Selective reporting (reporting bias)High riskNo ALT description and no QoL description, although mentioned as endpoint in the methods section
Other biasHigh risk

Vested interest bias: interferon-alpha-2a (Roferon® A), ribavirin and amantadine sulphate (PK Merz®) were provided by Roche Pharma (Schweiz) AG, Reinach, Switzerland

Baseline balance is questionable: baseline variables were similar in both groups, except genotype 1 and 4 infection and cirrhosis tended to be slightly more prevalent in the double and triple therapy group.

No baseline imbalance; sample size calculation was not reported; the trial was not stopped early

Yang 2003

Methods

Randomised, double-blind, placebo-controlled trial in naive patients

Patients were studied between October 1996 and October 1998

24 weeks therapy, 1 year follow-up

Participants

Country: Taiwan, 1 centre

Number of patients randomised: 30

Amantadine group n = 15: age 42 ± 9 years, male/female: 12/3

Control group n = 15: age 38 ± 7 years, male/female: 12/3

Inclusion criteria: naive patients aged between 20 and 55 years; positive for anti-HCV antibodies; abnormal serum ALT levels for more than 6 months, at least 3 documented occasions higher than twice the upper limit of normal (< 3 IU/L) with 1 month apart, within 6 months prior to enrolment; liver biopsy, within 1 month before start of treatment, to confirm chronic hepatitis without cirrhosis

Exclusion criteria: alcoholic, no intravenous drug abusers or homosexuals; hepatotoxic drugs, herb medicine, and immunosuppressive therapy within the past 6 months; decompensated liver function, cirrhosis; other diseases, i.e., chronic renal failure, neurological disorders, chronic hepatitis B, autoimmune; pregnancy; white cells and platelet abnormalities

Interventions

Patients were randomly assigned to receive:

Amantadine group: 4.5 MU recombinant interferon-alpha-2a sc thrice weekly and oral amantadine twice daily 100 mg, n = 15

Control group: 4.5 MU recombinant interferon-alpha-2a sc thrice weekly and oral placebo twice daily for 24 weeks, n = 15

Outcomes

Complete response: normalisation of serum ALT levels together with the absence of serum HCV RNA by the end of treatment = composite outcome;

sustained complete response: the continuation of the remission 12 months after the end of treatment

NotesAdditional information requested on 23 January 2012 to the first author Dr. S. Yang. Dr. Yang answered on 27 January 2012
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskComputer random number generator
Allocation concealment (selection bias)Low riskBy pharmacy. Only the pharmacologist in charge knew the sequence. The patients received amantadine or placebo from the pharmacy
Blinding of participants and personnel (performance bias)
All outcomes
Unclear riskInsufficient information, although placebo-controlled
Blinding of outcome assessment (detection bias)
All outcomes
Unclear riskInsufficient information, although placebo-controlled
Incomplete outcome data (attrition bias)
All outcomes
Low risk1 patient lost to follow-up 2 months after treatment. All others finished follow-up period
Selective reporting (reporting bias)High riskNo information about separate biochemical outcome
Other biasUnclear risk

Vested interest bias: unknown

No baseline imbalance; sample size calculation was not reported; the trial was not stopped early

Zeuzem 2000

  1. a

    AE: adverse event
    ALT: alanine aminotransferase
    AST: aspartate transaminase
    bid: twice a day
    BMI: body mass index
    EDLQ: everyday life questionnaire
    ELISA: enzyme-linked immunosorbent assay
    EOFU: end of follow-up
    EOT: end of treatment
    EVR: early virological response
    HAI: histology activity index
    HAV: hepatitis A virus
    HBsAg: hepatitis B surface antigen
    HBV: hepatitis B virus
    HCV: hepatitis C virus
    HIV: human immunodeficiency virus
    HRQoL: health-related quality of life
    ITT: intention-to-treat
    MIU: million international units
    MU: million units
    NS: non-significant
    PCR: polymerase chain reaction
    po: orally
    POMS: profile of mood status scale
    QoL: quality of life
    RNA: ribonucleic acid
    RT-PCR: real-time polymerase chain reaction
    SAE: serious adverse event
    sc: subcutaneous
    SVR: sustained virological response
    tiw: three times weekly
    U/L: units per liter
    VAS: visual analogue scale
    WBC: white blood cells

Methods

Randomised, double-blind, placebo-controlled trial in naive patients in Germany

Patients were enrolled between March and October 1997

48 weeks therapy, 24 weeks follow-up

Participants

Country: Germany

120 patients were enrolled, 1 patient did not receive treatment. 119 started treatment

Inclusion criteria: patients aged 18 to 70 years with compensated chronic, HCV infection not previously treated with interferon, ribavirin, and/or amantadine. Tested positive for anti-HCV and HCV RNA by RT-PCR. A liver biopsy within a year of study entry showing chronic hepatitis, and had elevated serum ALT levels for at least 6 months before initiation of treatment. Entry leucocyte count had to be ≥ 2500/μL, platelets > 70,000/μL

Exclusion criteria: any other cause of liver disease or other relevant disorders, including HIV or hepatitis B co-infection. Evidence or history of autoimmune disease. Other clinically significant diseases. Average daily intake of alcohol exceeding 50 g of ethanol or drug abuse within the previous year. Pregnancy and lactation period

Amantadine group: 26 patients, mean age 42.1 ± 12.9 years, male/female = 37/22. Serum ALT was 57.5 ± 39.0 U/L and the basal viral load was 7.8 ± 8.5 x 10⁶ copies per mL. Genotype 1 (n = 42), genotype 2 (n = 3), genotype 3 (n = 13), genotype 4 (n = 1). Histological staging: non = 8, mild = 25, moderate = 18, severe = 8

Control group: 60 patients, mean age 41.6 ± 10.3 years, male/female = 36/24. Serum ALT was 59.6 ± 36.0 U/L and the basal viral load was 7.4 ± 9.8 x 10⁶ copies per mL. Genotype 1 (n = 40), genotype 2 (n = 3), genotype 3 (n = 15), genotype 4 (n = 1). Histological staging: non = 2, mild = 28, moderate = 22, severe = 8

Interventions

Amantadine group: interferon-alpha-2a 6 MU sc thrice weekly for 24 weeks, followed by 3 MU sc thrice weekly for an additional 24 weeks and amantadine sulphate 100 mg po twice daily

Control group: interferon-alpha-2a 6 MU sc thrice weekly for 24 weeks, followed by 3 MU sc thrice weekly for an additional 24 weeks and placebo po twice daily

OutcomesMortality; SAE; treatment discontinuation due to AE; QoL; number of patients without SVR; number of patients with detectable HCV RNA at EOT; number of patients without improvement of histology; number of patients without normalisation of ALT at EOTand at EOFU
Notes
Risk of bias
BiasAuthors' judgementSupport for judgement
Random sequence generation (selection bias)Low riskRandom number generator in fixed blocks of 4 with a ratio of 1:1
Allocation concealment (selection bias)Unclear riskInsufficient information
Blinding of participants and personnel (performance bias)
All outcomes
Low riskPlacebo-controlled with matched placebo
Blinding of outcome assessment (detection bias)
All outcomes
Unclear riskPlacebo-controlled, but insufficient information
Incomplete outcome data (attrition bias)
All outcomes
Low riskNo missing data, equally divided
Selective reporting (reporting bias)Low riskNo selective reporting
Other biasHigh risk

Vested interest bias: Merz + Roche

No baseline imbalance; sample size calculation was reported; the trial was not stopped early

Characteristics of excluded studies [ordered by study ID]

StudyReason for exclusion
Buggisch 2009Unknown how many patients were randomised to each group
Di Bisceglie 2001Compared amantadine plus peg interferon-alpha and ribavirin with amantadine and peg interferon-alpha
Mendez-Navarro 2010aNot randomised: erratum
Nakamura 2003Not randomised
Popovic 2000Data in text and table are not comparable and reproducible
Quarantini 2006Does not report one of our outcome measures
Schories 2003Does not report one of our outcome measures
Torre 1999Does not report one of our outcome measures
Zilly 2002Not randomised

Ancillary