| Figure 1. Flow diagram. |
| 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. |
| 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. |
| 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. |
| 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. |
| Figure 7. Funnel plot of comparison: 4 Subgroup: trials at lower risk versus high risk of bias. Outcome: 4.2 Adverse events. |
| 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. |
| 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. |
| 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. |
| 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. |
| 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. |
| Analysis 3.1. Comparison 3 Subgroup: genotype 1 compared to genotype non-1, Outcome 1 Failure of sustained virological response. |
| 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. |
| 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. |