Background: Planning for outbreaks of influenza is a high priority public health issue for national governments. Neuraminidase inhibitors (NIs) are thought to help reduce the symptoms of influenza with several possible mechanisms proposed. NIs have been stockpiled with a view to their widespread use in the event of a pandemic. However, the evidence base for this class of agents remains a source of debate. In a previous review we have documented substantial risks of publication bias of trials of NIs for influenza (60% of patient data from phase III treatment trials of oseltamivir have never been published) and reporting bias in the published trials. Our confidence in the conclusions of previous versions of this review has been subsequently undermined. Since we have become aware of a large number of unpublished trials of NIs in the management of influenza, this review updates and merges existing reviews in this area.
Objectives: To review clinical study reports of placebo-controlled randomised trials, regulatory comments and reviews ('regulatory information') of the effects of the NIs oseltamivir and zanamivir for influenza in all age groups and appraise trial programmes, rather than single studies.Clinical study reports are very detailed, unpublished clinical trial data containing in-depth descriptions of protocol rationale, methods analysis plans, trial results and organisational documents (such as contracts). A series of clinical studies designed and conducted by one sponsor represents a trial programme of a drug indication (for example treatment of influenza).
Search methods: We searched trial registries, cross-referencing published and unpublished sources and corresponded with manufacturers and regulators. We searched the archives of the US Food and Drug Administration (FDA) and European and Japanese regulators. The evidence in this review reflects searches to obtain relevant information up to 12 April 2011.
Selection criteria: We included regulatory information based on assessments of randomised controlled trials (RCTs) conducted in people of any age who had either confirmed or suspected influenza, or who had been exposed to influenza in the local community or place of residence. We included information which had been made available by our deadline.
Data collection and analysis: We indexed regulatory information in two purpose-built instruments and reconstructed trials using CONSORT statement-based templates. To progress to Stage 2 (full analysis) we sought manufacturer explanations of discrepancies in the data. GlaxoSmithKline (GSK) offered us individual patient data and responded to our queries, but Roche did not provide us with complete clinical study reports. In Stage 2 we intended to analyse trials with validated data (i.e. assuming our validation questions aimed at clarifying omissions and discrepancies were resolved). No studies progressed to Stage 2. We carried out analyses of the effects of oseltamivir on time to first alleviation of symptoms and hospitalisations using the intention-to-treat (ITT) population and tested five hypotheses generated post-protocol publication.
Main results: We included and analysed data from 25 studies (15 oseltamivir and 10 zanamivir studies). We could not use data from a further 42 studies due to insufficient information or unresolved discrepancies in their data. The included trials were predominantly conducted in adults during influenza seasons in both hemispheres. A small number of studies were conducted in older people residing in care homes and in people with underlying respiratory diseases. The studies had adequate randomisation and blinding procedures, but imbalances in the analysis populations available (ITT influenza-infected) left many of the studies at risk of attrition bias. All the studies were sponsored by manufacturers of NIs. Time to first alleviation of symptoms in people with influenza-like illness symptoms (i.e. ITT population) was a median of 160 hours (range 125 to 192 hours) in the placebo groups and oseltamivir shortened this by around 21 hours (95% confidence interval (CI) -29.5 to -12.9 hours, P < 0.001; five studies) but there was no evidence of effect on hospitalisations based on seven studies with a median placebo group event rate of 0.84% (range 0% to 11%): odds ratio (OR) 0.95; 95% CI 0.57 to 1.61, P = 0.86). These results are based on the comprehensive ITT population data and are unlikely to be biased. A post-protocol analysis showed that participants randomised to oseltamivir in treatment trials had a reduced odds being diagnosed with influenza (OR 0.83; 95% CI 0.73 to 0.94, P = 0.003; eight studies), probably due to an altered antibody response. Zanamivir trials showed no evidence of this. Due to limitations in the design, conduct and reporting of the trial programme, the data available to us lacked sufficient detail to credibly assess a possible effect of oseltamivir on complications and viral transmission. We postponed analysis of zanamivir evidence because of the offer of individual patient data (IPD) from its manufacturer. The authors have been unable to obtain the full set of clinical study reports or obtain verification of data from the manufacturer of oseltamivir (Roche) despite five requests between June 2010 and February 2011. No substantial comments were made by Roche on the protocol of our Cochrane Review which has been publicly available since December 2010.
Authors' conclusions: We found a high risk of publication and reporting biases in the trial programme of oseltamivir. Sub-population analyses of the influenza infected population in the oseltamivir trial programme are not possible because the two arms are non-comparable due to oseltamivir's apparent interference with antibody production. The evidence supports a direct oseltamivir mechanism of action on symptoms but we are unable to draw conclusions about its effect on complications or transmission. We expect full clinical study reports containing study protocol, reporting analysis plan, statistical analysis plan and individual patient data to clarify outstanding issues. These full clinical study reports are at present unavailable to us.