This review concluded that therapy with adalimumab or infliximab may have a beneficial effect compared with standard care on outcome measures for induction and maintenance therapy for Crohn's disease. Issues with the reliability of the trial evidence were documented by the authors and their conclusion rests on a well-conducted review and appears likely to be reliable.

To assess the effectiveness and cost-effectiveness of infliximab in adults and children with moderate to severe Crohn's disease who were refractory to or intolerant of conventional treatment and in adults with fistulising Crohn's disease, and to assess the effectiveness and cost-effectiveness of adalimumab in adults with moderate to severe Crohn's disease who were refractory to or intolerant of conventional treatment.

MEDLINE, EMBASE and Cochrane Central Register of Controlled Trials (CENTRAL) were searched up to June 2007. Research registries were searched, with conference abstracts from five relevant societies for 2006 and 2007. The search strategy was reported. Relevant websites including those of the European Medicines Agency and the US Food and Drug Administration were also searched; references of relevant studies were checked, experts were contacted and submissions from industry considered. There were no language restrictions.

Randomised controlled trials (RCTs) that enrolled adults aged at least 18 years or children aged six to 17 years with moderate to severe Crohn's disease who were refractory to or intolerant of conventional treatment were eligible for inclusion. Adults with fistulising or non-fistulising Crohn's disease were eligible. Moderate to severe disease was defined as reported by trial authors or having an average Crohn's Disease Activity Index (CDAI) of at least 220.

Trials were required to assess tumour necrosis factor-alpha (TNF-alpha) inhibitors (adalimumab or infliximab) at any dosage and on any regimen compared with conventional treatment without TNF-alpha inhibitors. Conventional treatment could include: no treatment; placebo; dietary interventions; pharmacotherapies such as aminosalicylates, methotrexate, corticosteroids, azathioprine, metronidazole; or surgical treatment. Trials that compared adalimumab and infliximab with each other were also eligible, as were those that compared different dosages or regimens of the same drug. At least one of the following outcomes had to be reported: overall survival; progression-free survival; quality of life; disease activity; need for surgery; hospitalisation; adverse effects. Trials of both induction and maintenance of remission were eligible.

All except two of the included trials compared adalimumab or infliximab with placebo; two trials of infliximab enrolled patients with fistulising disease and two trials were in paediatric patients (both dose-finding studies of infliximab and neither placebo controlled). All were multi-centre trials conducted mainly in North America or Europe. Exclusion criteria that related to previous or concomitant medications varied but most patients were receiving additional treatments.

Two reviewers independently assessed the studies for inclusion; disagreements were resolved by discussion or referred to a third reviewer.

Studies were appraised for quality using criteria based on Cochrane Collaboration guidelines. These comprised: randomisation, allocation concealment, blinding, follow-up, comparability of groups, use of intention-to-treat analysis, power calculation and selective reporting.

Quality was assessed by one reviewer and checked by a second. Discrepancies were resolved by discussion or referral to a third reviewer.

Data extraction was undertaken for risk differences (RD) or risk ratios (RR) with 95% confidence intervals (CI) or to permit the calculation of these for the outcomes of percentage response, response 70, response 100 and remission. Mean differences were calculated for the Crohn's Disease Activity Index and other continuous measures. Data extraction included the extraction of numerical information from graphs.

One reviewer performed the data extraction using a standardised data extraction form, and another reviewer checked it. Discrepancies were resolved through discussion with a third reviewer.

Due to clinical heterogeneity, the studies were combined in a narrative synthesis structured by the aims of therapy (induction or maintenance) and characteristics of patients (non-fistulising versus fistulising adult patients or paediatric patients).The effect of treatment dose and differences in study duration (range four to 56 weeks) were examined.

Eleven RCTs were included in the review. There were concerns about the quality of the trials, in particular those that assessed maintenance therapy. These centred on division of patients into responders and non-responders at different time points, high rates of scheduled cross-overs between treatment groups and unclear handling of data (including missing data). The use of responder only data in maintenance trials was a particular issue.

Adult patients with non-fistulising disease

In two induction trials between 6% and 24% more patients treated with adalimumab achieved remission compared with placebo; in one infliximab trial between 21% and 44% more patients achieved remission. In maintenance trials benefits of between 24% and 29% more patients achieved remission for adalimumab (two trials) and between 14% and 24% for infliximab (two trials). These results were based on changes in the Crohn's Disease Activity Index and reflected results across a range of doses.

Adult patients with fistulising disease

One induction trial showed that between 29% and 42% more patients achieved a 50% reduction in fistulas compared with placebo; 23% more patients in a maintenance trial achieved this reduction compared with placebo. Subgroup results from two additional RCTs provided contradictory indications as to whether adalimumab conferred any benefit.

Children

It appeared likely that infliximab therapy for children offered a benefit of increased remission but the lack of a placebo group in either the induction or the maintenance trial made this difficult to determine.

Adverse events

There were few differences in selected adverse events between groups.

Other outcomes were also reported as were some results for patients who did not respond to treatment.

A Markov model was used to estimate the costs of adalimumab and infliximab; this indicated that both adalimumab and infliximab were cost-effective for induction therapy in severe Crohn's disease and adalimumab was was cost-effective in moderate Crohn's disease. Neither drug was cost-effective as maintenance therapy for either moderate or severe disease. Total costs to the NHS were estimated at between £17 million and £92 million for induction treatment in severe disease only; targeted treatment would reduce these estimates. Maintenance therapy would have a much higher total cost, at between £140 million and £200 million for one year.

Anti-TNF therapy with adalimumab or infliximab may have a beneficial effect compared with standard care on outcome measures for induction and maintenance therapy for Crohn's disease. However, multiple methodological problems with the available evidence meant that the relative efficacy of the medications could not be determined.

The review questions and inclusion criteria were clear. The search was thorough and did not include restrictions on language or publication status. The authors reported using methods designed to reduce reviewer bias and error at all stages of the review process. Validity was assessed appropriately and key concerns incorporated into the synthesis and conclusions. The decision to use a narrative synthesis reflected the nature of the included trials including the concerns about the quality of the evidence and the small numbers of patients. The authors' conclusions reflect the results of the review and appear likely to be reliable.

Practice: The authors did not state any implications for practice.

Research: The authors identified the need for independently funded RCT research on the effectiveness of anti-TNF therapy for Crohn's disease. They made detailed recommendations for the types of trials which were required and the methods which should be adopted in their design and conduct; these were documented in the report.

National Institute for Health Research through the Health Technology Assessment programme

This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.