Background/Purpose:

TNF inhibition has revolutionised the treatment of axial spondyloarthritis (AxSpA). Much of what we know about the use of these drugs comes from randomised controlled trials (RCTs). However, RCT populations are highly selected, often from specialist centres, and there are potential threats to external validity – i.e. that the treatment effect in this sub-group of patients is not representative of the wider eligible patient population.

The aim of the current study, using a large, nationally representative patient registry, was to (a) determine the effectiveness of TNF inhibition in a large ‘real-world’ patient population; and (b) examine predictors of treatment failure.

Methods:

The British Society for Rheumatology (BSR) Biologics Register for Ankylosing Spondylitis recruits patients with AxSpA newly starting TNF inhibition. At baseline, clinical and patient-reported data is collected from the medical notes and patient questionnaires. Patients are followed up at 3, 6 and 12 months, and treatment failure was defined as failure to achieve an ASAS20 response by 1yr – a composite outcome comprising patient global assessment, pain, function, and inflammation.

Logistic regression was used to determine the relationship between baseline clinical and patient-reported characteristics and treatment failure.

Results:

Data were available on 354 patients recruited between Apr-2013 and Sep-2016, of whom 127 (36%) had failed to achieve an ASAS20 response by 1yr. There was no difference in age between those who did / did not achieve an ASAS20 response, nor by gender.

Patients with peripheral joint disease were significantly less likely to achieve ASAS20 response than other patients (OR: 1.90; 95%CI: 1.13-3.19). There was some evidence that patients with non-radiographic disease and those who were HLA-B27 negative were less likely to respond than other patients, although this did not reach statistical significance (see Table). Similarly, no clear or significant association was observed with disease activity or function.

Conclusion:

We have shown, in this real-world AxSpA patient population receiving their first TNF inhibitor, around two-thirds achieve satisfactory treatment response within 1yr. This is reassuringly comparable to the proportion of responders in randomised clinical trial populations. There are few strong predictors of treatment response and it may be that some patients have other conditions (e.g. fibromyalgia) that may be less likely to respond to biologic therapy. Early identification of patients who fail to respond within 1yr remains a clinical challenge, emphasising the importance of early monitoring after commencing these therapies.

« Back to 2017 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/real-world-effectiveness-of-tnf-inhibition-in-spondyloarthritis-data-from-a-large-nationwide-prospective-cohort-the-british-society-for-rheumatology-biologics-register-for-ankylosing-spondylit/