Evaluation of the placebo and treatment effect overtime in randomised clinical trials evaluating the efficacy of biologics in axial spondyloarthritis: systematic review and meta-analysis

Léa Perrot; Anna Molto; Caroline Paris; Robinson Gravier Dumonceau; Maxime Dougados; Thao Pham

doi:10.1016/j.ard.2025.11.010

Evaluation of the placebo and treatment effect overtime in randomised clinical trials evaluating the efficacy of biologics in axial spondyloarthritis: systematic review and meta-analysis

Ann Rheum Dis. 2025 Dec 24:S0003-4967(25)04536-4. doi: 10.1016/j.ard.2025.11.010. Online ahead of print.

Authors

Léa Perrot¹, Anna Molto², Caroline Paris³, Robinson Gravier Dumonceau⁴, Maxime Dougados⁵, Thao Pham³

Affiliations

¹ Aix Marseille University, Rheumatology Department, Sainte Marguerite Hospital, APHM, Marseille, France. Electronic address: lea.perrot2@gmail.com.
² Université Paris Cité, Service de Rhumatologie, Hôpital Bichat, APHP.Nord, Paris, France; Université Paris Cité and Université Sorbonne Paris Nord, Inserm, INRAE, Centre for Research in Epidemiology and Statistics (CRESS), Paris, France.
³ Aix Marseille University, Rheumatology Department, Sainte Marguerite Hospital, APHM, Marseille, France.
⁴ Équipe QuanTIM, UMR SESSTIM 1252 INSERM/IRD/Université AMU, Marseille, France.
⁵ Rheumatology Department, Cochin Hospital, APHP.Centre, Paris, France.

PMID: 41448991
DOI: 10.1016/j.ard.2025.11.010

Abstract

Objectives: To evaluate temporal trends in treatment efficacy and placebo responses in randomized controlled trials in axial spondyloarthritis (axSpA).

Methods: We conducted a systematic review of RCTs in MEDLINE up to July 2024 (PROSPERO: CRD42024563776). Eligible trials compared tumour necrosis factor inhibitors (adalimumab, certolizumab pegol, etanercept, infliximab, golimumab) or interleukin-17 inhibitors (bimekizumab, ixekizumab, secukinumab) versus placebo in axSpA. Two reviewers independently selected studies and extracted data. Primary outcomes were temporal trends in placebo response for Assessment of SpondyloArthritis International Society 20% and 40% response criteria (ASAS20/40), axSpA Disease Activity Score (ASDAS) Clinically Important Improvement, and ASDAS Inactive Disease (ASDAS-ID) at each trial's primary endpoint. Random-effects meta-regressions estimated annual changes in treatment effects (odds ratios) and in arm-specific response rates, adjusted for age, sex, radiographic status, region, route of administration, baseline Bath Ankylosing Spondylitis Disease Activity Index score, baseline C-reactive protein, and disease duration.

Results: Thirty-one RCTs (n = 6437) were included. Treatment effect declined over time for ASAS20 (-5% per year, P < .001) and ASAS40 (-3.7%, P = .04), driven by increased placebo response (ASAS20: +1% per year, P < .0001) and a decline in treatment response for ASAS40. In contrast, ASDAS-based treatment effects were stable, despite a small increase in placebo ASDAS-ID (+0.19% per year, P < .05).

Conclusions: ASAS20/40 treatment effects have declined over time in axSpA trials, consistent with a fading of reported effectiveness, whereas ASDAS-based endpoints remain stable, indicating greater robustness to placebo effects and temporal trends.