OBJECTIVES : Tumour necrosis factor inhibitor (TNFi) withdrawal in patients with non-radiographic axial spondyloarthritis (nr-axSpA) following sustained remission is predominantly associated with disease flares requiring re-establishment of treatment. This study compared continuous TNFi therapies in nr-axSpA patients who have achieved sustained remission.

METHODS : A systematic literature review (SLR) was performed to identify randomised controlled trials evaluating adult axSpA patients who failed ≥1 non-steroidal anti-inflammatory drug. Of the identified trials, a subset evaluated the efficacy of continuous versus withdrawal of TNFi therapy in nr-axSpA patients who responded to treatment. Bayesian indirect treatment comparisons (ITCs) were performed to indirectly compare treatments for sustained remission and flare, measured using odds ratios (OR) with 95% credible intervals (CrIs).

RESULTS : The SLR identified one trial designed to evaluate withdrawal (ABILITY-3); after assessment, it was determined to be comparable for ITC with C-OPTIMISE. In C-OPTIMISE, after 48 weeks of treatment with certolizumab pegol (CZP) 200mg every two weeks (Q2W), patients achieving sustained remission were randomised to CZP 200mg Q2W, CZP 200mg every four weeks (Q4W), or placebo (PBO); analyses included only the nr-axSpA subgroup (data on file). In ABILITY-3, patients responding to 28 weeks of adalimumab (ADA) 40mg Q2W were randomised to ADA 40mg Q2W or PBO. In ITC, patients receiving CZP 200mg Q2W and CZP 200mg Q4W had significantly higher odds of maintaining remission compared to ADA (OR [95% CrI]: 6.70 [2.23–21.96] and 4.08 [1.45–12.21], respectively). Patients receiving CZP 200mg Q2W and CZP 200mg Q4W had significantly lower odds of flares 48 weeks post-randomisation compared to ADA at 40 weeks (OR [95% CrI]: 0.15 [0.05–0.44] and 0.25 [0.08–0.69], respectively).

CONCLUSIONS : In nr-axSpA patients who responded to TNFi regimens, continued CZP treatment at both full and reduced maintenance doses showed improved odds of reduced flares and sustained remission compared to continued ADA.