Does symptom duration impact on treatment response in axial spondyloarthritis? A meta-analysis of randomized controlled trials

Abstract Objectives The ASAS consensus defines ‘early axial spondyloarthritis (axSpA)’ as symptom duration ≤2 years, a definition derived from expert opinion due to limited evidence. We performed a meta-analysis of randomized placebo-controlled trials (RCTs) of biologic and targeted synthetic DMARDs (bDMARDs/tsDMARDs) in axSpA to assess the impact of symptom duration on treatment response. Methods RCTs in adult axSpA patients comparing b/tsDMARDs to placebo were identified through systematic literature reviews supporting the ASAS-EULAR management recommendations. Primary outcome was ASAS40 response, with secondary outcomes including disease activity, function and quality of life at each trial’s primary end point. Relative risks (RRs) comparing two groups (shorter vs longer symptom duration) were calculated for efficacy outcomes for each arm and for several thresholds of symptom duration (1–5 years). RR ratios (RRRs) were calculated as the ratio of RR of the b/tsDMARD and placebo arms. Pooled effects were obtained with random-effects meta-analysis. Results Eleven bDMARD RCTs, comprising 3272 axSpA patients, were included. No studies involving tsDMARDs met the eligibility criteria. At the 2-year threshold, pooled RR for ASAS40 response was 2.04 (95% CI, 1.48–2.84) in early axSpA and 2.28 (95% CI, 1.65–3.15) in established disease, yielding an RRR of 0.89 (95% CI, 0.63–1.26), with no statistical significance. There were no significant differences between efficacy in shorter and longer symptom duration for other outcomes or across other symptom duration thresholds (3–5 years). Conclusions Based on a meta-analysis of 11 trials, no significant differences were found in the efficacy of bDMARDs compared with placebo in patients with axSpA with shorter vs longer symptom duration, across thresholds from 2 to 5 years.