medwireNews: The response to tumor necrosis factor (TNF) inhibitors among patients with axial spondyloarthritis (axSpA) is lower in the real world than in clinical trials regardless of whether individuals would have met trial eligibility criteria, study findings indicate.
Gareth Jones (University of Aberdeen, UK) and colleagues say their findings have “important implications for the generalisability of trial results, and the cost-effectiveness of TNF [inhibitor] agents.”
They add: “The rheumatologist needs to consider that the proportion of patients who achieve a satisfactory treatment response to TNF inhibition will be lower than might be expected from clinical trials.”
Jones and team studied data for 2419 participants (mean age 48 years, 68% men) of the British Society for Rheumatology Biologics Register for Ankylosing Spondylitis (BSRBR-AS), a real-world cohort of axSpA patients who were TNF inhibitor-naïve at recruitment.
Of these, 816 (34%) were starting treatment with a TNF inhibitor, most commonly adalimumab (64%). On average, the individuals starting TNF inhibition were younger (44 vs 50 years) with a shorter symptom duration (15 vs 22 years) than those not starting such treatment, but had more active disease, with a BASDAI score of 6.4 versus 4.0, a BASFI score of 6.2 versus 3.8, and a BASMI score of 4.2 versus 3.6.
Among the individuals commencing TNF inhibitors, 41% met eligibility criteria for at least one of 14 relevant trials identified from a recent Health Technology Assessment of TNF inhibitors for ankylosing spondylitis and non-radiographic axSpA. But the researchers note that the proportion eligible differed according to the agent being trialed, at 64% for etanercept, 50% for certolizumab pegol, and 30% for adalimumab.
As reported in the Annals of the Rheumatic Diseases, the trial-eligible individuals from the BSRBR-AS cohort had higher disease activity (BASDAI 6.9 vs 6.1) and poorer function (BASFI 6.6 vs 6.0) than the individuals who would not have met trial eligibility criteria but were otherwise broadly similar.
The response rate among the BSRBR-AS individuals was 51.3% according to ASAS20 criteria. By comparison, the response rate was significantly higher, at 61.7%, among the 1401 participants of the published trials.
The response rate was 50.0% among the BSRBR-AS participants who met trial eligibility criteria and 52.0% among those who did not meet any of the criteria, indicating that “[p]otential eligibility for trials did not influence treatment response,” Jones et al remark.
Discussing the reasons for the lower response rates in the real-world, the authors “hypothesise that there is a selection bias into randomised clinical trials favouring the more educated and more affluent, and those with better mental health, and that this results in trial participants having a superior chance of positive outcome, compared with the real-world patients that they ostensibly represent.”
medwireNews is an independent medical news service provided by Springer Healthcare. © 2020 Springer Healthcare part of the Springer Nature Group