Background and objectives:

Systematic reviews and economic modelling of clinical effectiveness and cost-effectiveness of therapeutic monitoring of tumour necrosis factor alpha (TNF-α) inhibitors [using LISA-TRACKER^® enzyme-linked immunosorbent assay (ELISA) kits (Theradiag, Marne La Vallee, France, or Alpha Laboratories, Heriot, UK), TNF-α-Blocker ELISA kits (Immundiagnostik AG, Bensheim, Germany) and Promonitor^® ELISA kits (Proteomika, Progenika Biopharma, Bizkaia, Spain)] versus standard care for Crohn’s disease (CD).

Methods:

Multiple electronic databases were searched from inception to December 2014 in order to identify primary studies and meta-analyses.

Population:

Patients with moderate to severe active CD treated with infliximab (IFX) (Remicade^®, Merck Sharp & Dohme Ltd, Kenilworth, NJ, USA) or adalimumab (ADA) (Humira^®, AbbVie Inc., North Chicago, IL, USA).

Intervention:

Monitoring of serum anti-TNF-α (IFX or ADA) and/or of anti-drug antibody levels using test assays with a test–treatment algorithm.

Comparator:

Standard care.

Outcomes:

Any patient-related outcome, test agreement and cost-effectiveness estimates. The quality assessments used recognised checklists (Quality Assessment of Diagnostic Accuracy Studies-2, Cochrane, Philips and Consolidated Health Economic Evaluation Reporting Standards). Evidence was synthesised using narrative review and meta-analysis. A Markov model was built in TreeAge Pro 2013 (TreeAge Software, Inc., Williamstown, MA, USA). The model had a 4-week cycle and a 10-year time horizon, adopted a NHS and Personal Social Services perspective and used a linked evidence approach. Costs were adjusted to 2013/14 prices and discounted at 3.5%.

Results:

We included 68 out of 2434 and 4 out of 2466 studies for the clinical effectiveness and cost-effectiveness reviews, respectively. Twenty-three studies comparing test methods were identified. Evidence on test concordance was sparse and contradictory, offering scant data for a linked evidence approach. Three studies [two randomised controlled trials (RCTs) and one retrospective observational study] investigated outcomes following implementation of a test algorithm. None used the specified commercial ELISA immunoassay test kits. Neither of the two RCTs demonstrated clinical benefit of a test–treatment regimen. A meta-analysis of 31 studies to estimate test accuracy for predicting clinical status indicated that 20–30% of test results are likely to be inaccurate. The four cost-effectiveness studies suggested that testing results in small cost reductions. In the economic analysis the base-case analysis showed that standard practice (no testing/therapeutic monitoring with the intervention tests) was more costly and more effective than testing for IFX. Sensitivity and scenario analyses gave similar results. The probabilistic sensitivity analysis indicated a 92% likelihood that the ‘no-testing’ strategy was cost-effective at a willingness to pay of £20,000 per quality-adjusted life-year.

Strengths and limitations:

Rigorous systematic reviews were undertaken; however, the underlying evidence base was poor or lacking. There was uncertainty about a linked evidence approach and a lack of gold standard for assay comparison. The only comparative evidence available for economic evaluation was for assays other than the intervention assays.

Conclusions:

Our finding that testing is not cost-effective for IFX should be viewed cautiously in view of the limited evidence. Clinicians should be mindful of variation in performance of different assays and of the absence of standardised approaches to patient assessment and treatment algorithms.

Future work recommendations:

There is substantial variation in the underlying treatment pathways and uncertainty in the relative effectiveness of assay- and test-based treatment algorithms, which requires further investigation. There is very little research evidence on ADA or on drug monitoring in children with CD, and conclusions on cost-effectiveness could not be reached for these.

Study registration:

This study is registered as PROSPERO CRD42014015278.

Funding:

The National Institute for Health Research Health Technology Assessment programme.

Article history

The research reported in this issue of the journal was commissioned and funded by the HTA programme on behalf of NICE as project number 14/69/03. The protocol was agreed in January 2015. The assessment report began editorial review in May 2015 and was accepted for publication in December 2015. The authors have been wholly responsible for all data collection, analysis and interpretation, and for writing up their work. The HTA editors and publisher have tried to ensure the accuracy of the authors’ report and would like to thank the reviewers for their constructive comments on the draft document. However, they do not accept liability for damages or losses arising from material published in this report.

Declared competing interests of authors

Aileen Clarke is a member of the National Institute for Health Research, Health Technology Assessment and Efficacy and Mechanism Evaluation Editorial Boards. Aileen Clarke and Sian Taylor-Phillips are partly supported by the National Institute for Health Research Collaboration for Leadership in Applied Health Research and Care West Midlands at the University Hospitals Birmingham NHS Foundation Trust.