General Information
Conflict of Interest
Request Clinical Trials
Associated Trial(s):- NCT00207701 - A Randomized, Double-blind Trial of the Efficacy of REMICADE (Infliximab) Compared With Placebo in Subjects With Ankylosing Spondylitis Receiving Standard Anti-inflammatory Drug Therapy
- NCT00265083 - A Multicenter, Randomized, Double-blind, Placebo-controlled Trial of Golimumab, a Fully Human Anti-TNFa Monoclonal Antibody, Administered Subcutaneously, in Subjects with Active Ankylosing Spondylitis
- NCT01453725 - A Multicenter, Randomized, Double-blind, Placebo-controlled Study of the Effect of Golimumab Administered Subcutaneously in Subjects With Active Axial Spondyloarthritis (Also Known as MK-8259-006-02)
- NCT01248793 - A Phase 3, Multicenter, Randomized, Double-blind, Placebo-controlled Study Evaluating the Efficacy and Safety of Golimumab in the Treatment of Chinese Subjects with Ankylosing Spondylitis
- NCT02186873 - A Study of Golimumab in Participants With Active Ankylosing Spondylitis
Request Clinical Trials
Data Request Status
Status: OngoingResearch Proposal
Project Title: Efficacy of b/tsDMARDs in patients with axial spondyloarthritis with elevated and not elevated CRP
Scientific Abstract:
Background: Elevated CRP predicts good response to bDMARDs in axSpA, mainly based on TNFi trials. Evidence for IL-17i is limited; in COAST-V, CRP+ patients responded better, but CRP− patients still showed numerical benefit over placebo. Small subgroup sizes and lack of combined CRP/MRI analyses limit conclusions.
Objective: To assess whether treatment efficacy of b/tsDMARDs differs between CRP+ and CRP− patients, considering MRI status as a potential effect modifier or confounder.
Study Design: Systematic review with meta-analysis and individual patient data (IPD) meta-analysis of RCTs.
Participants: Adults (>=18 years) with axSpA enrolled in placebo-controlled RCTs of approved bDMARDs (TNFi, IL-17i) or tsDMARDs (JAK inhibitors). Patients will be classified as CRP+/CRP− and MRI+/MRI−.
Primary Outcome Measure: ASAS40 response at the study's primary timepoint (12--16 weeks).
Secondary Outcome Measures: ASAS20, ASAS5/6, ASDAS changes and response states, BASDAI, BASFI, BASMI, CRP levels, SPARCC MRI scores, and quality-of-life indices.
Statistical Analysis: For each trial, relative risks for CRP+ vs CRP− (and MRI+ vs MRI−) and relative risk ratios (RRRs) comparing treatment vs placebo will be computed. Continuous outcomes will use ANCOVA-based differences in differences. Study-level estimates will be pooled using random-effects models. IPD meta-analysis will test for CRPxMRI interaction; if significant, analyses will be stratified by MRI status; otherwise, models will adjust for MRI and trial-level covariates.
Brief Project Background and Statement of Project Significance:
In patients with axial spondyloarthritis (axSpA), C-reactive protein (CRP) has been shown to predict a good response to biological disease-modifying antirheumatic drugs (bDMARDs).[1] This evidence has stemmed from studies with tumor necrosis factor inhibitors (TNFi).[2-5]
While there are less data on IL-17i, the subgroup analysis of the COAST-V trial showed a higher response in patients with elevated CPR (CRP+) compared to those without elevated CRP (CRP-).[6] However, in the CRP- group, the response achieved with ixekizumab was numerically higher than that of placebo, which in turn was comparable to that of adalimumab in CRP- (ASAS40 34.5% for ixekizumab, 19.2% placebo and 21.1% adalimumab, with n=29 for ixekizumab, n=26 for placebo and n=38 for adalimumab). Based on these findings and their own clinical experience, some authors have proposed a differentiating role for IL-17i in patients who are CRP-. The claim is that, unlike TNFi, the efficacy of IL-17i may not depend on the initial inflammation status, as assessed by CRP levels or MRI findings.[7] However, the existing data are insufficient to confirm this claim.
Ideally, such a question would be investigated in a randomised clinical trial (RCT), with randomisation stratified for baseline CRP+/- and MRI of the sacroiliac joints (SIJ) (MRI +/-), as the latter is also a known predictor of response to treatment.[5,8,9] However, such a randomised clinical trial is very costly and challenging. Subgroup analyses of each trial represent challenges. First, the groups, particularly CRP- or MRI- tend to be small. In trials with nr-axSpA, the combination of both, i.e. CRP- and MRI-, does not even exist. Additionally, because of the small group size, only one parameter is considered e.g. CRP, without considering the presence of inflammation on MRI, which is a known predictor for response.
Without the possibility of running an RCT to address this question, the second-best option is to systematically conduct post-hoc analyses of existing trials, across drug classes, and to try to pool data across trials of the same drug class in order to have more power for this comparison. Moreover, such analysis should also consider the role of MRI, either as a potential effect modifier or confounder of the efficacy of b/tsDMARDs in the subgroups of CRP+/CRP-.
Specific Aims of the Project:
To evaluate the impact of CRP+/- on clinical response in patients with axSpA treated with bDMARDs and targeted synthetic DMARDS (tsDMARDs):
a) To investigate whether there is a different efficacy of b/tsDMARDs (and drug classes: TNFi, IL-17i, JAKi) in CRP+ vs CRP- patients;
b) To investigate whether there is a different efficacy of b/tsDMARDs (and drug classes: TNFi, IL-17i, JAKi) in MRI+ vs MRI- patients;
c) To investigate whether the efficacy of b/tsDMARDs (and drug classes: TNFi, IL-17i, JAKi) in CRP+ vs CRP- patients is different according to the MRI+/-;
d) To assess the efficacy of b/tsDMARDs in CRP subgroups taking MRI subgroups into account (subgroups according to MRI and CRP) or analyses corrected for MRI, depending on whether there is effect modification or confounding;
Study Design: Meta-analysis (analysis of multiple trials together)
What is the purpose of the analysis being proposed? Please select all that apply.: New research question to examine treatment effectiveness on secondary endpoints and/or within subgroup populations Research on clinical prediction or risk prediction
Software Used: STATA
Data Source and Inclusion/Exclusion Criteria to be used to define the patient sample for your study:
Eligible studies will include RCTs involving adult patients (aged >=18 years) diagnosed with axSpA, treated with bDMARDs or tsDMARDs compared to placebo. Interventions will include treatment with approved bDMARDs --adalimumab, etanercept, infliximab, golimumab, certolizumab pegol, secukinumab, ixekizumab and bimekizumab--and tsDMARDs--tofacitinib, or upadacitinib.
Patients will be categorized as being CRP+ (baseline CRP >5mg/L) or CRP- (baseline CRP 5mg/L). To ensure meaningful subgroup analyses, studies will be required to have at least 20% of patients within each of the two CRP subgroups.
Additionally, patients will be categorized as being MRI+ (baseline MRI-SIJ+ according to the ASAS positive definition or SPARCC 2) or MRI- (baseline MRI-SIJ not positive or SPARCC<2). As the main focus of the analysis will be the CRP, no patient selection will be made based on the number of patients within each MRI subgroup. However, if included studies have less than 20% of patients within each MRI subgroup, a sensitivity analysis will be conducted, excluding these studies.
De-identified individual patient data (IPD) will be obtained from sponsors of industry-funded RCTs, where available. Approval will be requested from data-sharing committees of the respective companies through platforms such as Vivli, the Yale University Open Data Access (YODA) Project, ClinicalStudyDataRequest.com (CSDR). For investigator-initiated trials not sponsored by industry, we will contact the principal investigators to obtain the necessary data. Data will be obtained by analyzing IPD and, at the trial level, compiling aggregated information based on a standardised data extraction sheet. Subsequently, studies will be merged into a single dataset to allow for IPD meta-analysis.
Primary and Secondary Outcome Measure(s) and how they will be categorized/defined for your study:
The primary outcome will be the proportion of patients achieving ASAS40 response at the time of the primary outcome assessment of each study (expected 12-16 weeks). This outcome was selected as it is nowadays the most common primary endpoint of RCTs. Secondary outcomes will include ASAS20, ASAS 5/6 response, ASAS partial remission, change in the Axial Spondyloarthritis Disease Activity Score (ASDAS), ASDAS clinically important improvement (improvement >=1.1), ASDAS major improvement (improvement >=2.0), achievement of ASDAS low disease activity or inactive disease (<2.1), achievement of ASDAS inactive disease (<1.3), change in the Bath Ankylosing Spondylitis Disease Activity Index (BASDAI), BASDAI50 response, change in CRP serum levels, change in MRI SpondyloArthritis Research Consortium of Canada (SPARCC) score (SIJs and spine), functional assessment (change in the Bath Ankylosing Spondylitis Functional Index [BASFI]), spinal mobility (change in the Bath Ankylosing Spondylitis Metrology Index [BASMI]), and quality of life measures (change in the ASAS Health Index [ASAS HI], Ankylosing Spondylitis Quality of Life [ASQoL], European Quality of Life-5 Dimensions [EQ-5D], and the 36-Item Short Form Health Survey physical and mental component scores [SF-36]).
Main Predictor/Independent Variable and how it will be categorized/defined for your study: Data on study characteristics, patient demographics (age, sex), disease characteristics (symptom duration, disease subtype--radiographic axSpA [r-axSpA] or non-radiographic axSpA [nr-axSpA]--, HLA-B27 status, baseline CRP+, baseline MRI+, bDMARD-insufficient responders (IR), treatments (intervention and comparator) and outcomes will be collected. Outcomes will be collected for the subgroups defined based on CRP+/-, MRI+/- and their combination CRP+/MRI+, CRP+/MRI-, CRP-/MRI+, CRP-/MRI- (the latter only available in r-axSpA studies).
Other Variables of Interest that will be used in your analysis and how they will be categorized/defined for your study: Data on study characteristics, patient demographics (age, sex), disease characteristics (symptom duration, disease subtype--radiographic axSpA [r-axSpA] or non-radiographic axSpA [nr-axSpA]--, HLA-B27 status, baseline CRP+, baseline MRI+, bDMARD-insufficient responders (IR), treatments (intervention and comparator) and outcomes will be collected.
Statistical Analysis Plan:
The analysis will involve three steps: 1) individual study-level analysis; 2) meta-analysis of all included studies with aggregate data per study; and 3) individual patient data meta-analysis of included studies.
1) Individual study-level analysis
For the individual study-level analysis, for each trial and for each treatment arm, relative risks (RRs) will be calculated as the ratio of outcomes between patients with CRP+ vs CRP- (the same for MRI+ vs MRI-). Following this, relative risk ratios (RRRs), together with their confidence interval (CIs), will be computed to represent the treatment effect (b/tsDMARD vs placebo) between the two groups (CRP+ vs CRP-). This will allow to assess whether CRP+ (vs CRP-) influences the response to b/tsDMARD therapy relative to placebo, indicating whether baseline CRP affects treatment response to b/tsDMARD therapy.[12,13] For continuous outcomes, mean differences and their standard deviation (SD) will be calculated for each treatment group. Subsequently, differences in differences (DiD) will be calculated to represent the difference in treatment effects between CRP+ and CRP- subgroups. The same will be done for MRI+ vs MRI- and then for the subgroups combining both CRP+ and MRI+ (so CRP+/MRI+, CRP+/MRI-, CRP-/MRI+, CRP-/MRI-). The last group will only be included in trials with patients with r-axSpA and will serve as the reference. In nr-axSpA, the CRP-/MRI+ group will serve as the reference. For binary endpoints, an RRR above 1 will indicate better
outcomes in CRP+ (or MRI+), while an RRR below 1 indicates worse outcomes in CRP+ (or MRI+) patients.
Additionally, models will be constructed to compute relative Odds Ratio (OR) for binary outcomes (logistic regression) or relative differences for continuous outcomes (analysis of covariance -- ANCOVA) for CRP+, which allow for adjusting for the main stratification factors of screening of the different trials, treatment, interaction between treatment and CRP+ and also for MRI+/- and baseline value of the respective continuous outcome variable (ANCOVA only).
An Intention-to-treat approach will be applied, and missing data will be managed through imputation methods following the imputation for the trial's main analysis. Most often, this means non-responder imputation for dichotomous outcomes and multiple imputation for continuous outcomes.
2) Meta-analysis of all included studies with aggregate data per study
As a second step, the individual above-mentioned aggregate estimates from each trial and for each outcome (RRR/relative OR, DiD/relative difference) will be pooled. A meta-analysis will be performed using random-effects models to account for variability across studies.
Although an individual patient data meta-analysis is planned (step 3), this second step is conducted, as it is possible that not all trials are available for step 3 (e.g. if trials made available in different data sharing platforms, e.g. Vivli and CSDR), and with this step 2 we can include all trials in the meta-analysis.
3) Individual patient data meta-analysis of included studies
As a third step, individual patient data meta-analysis will be conducted. This will allow including all studies into one single dataset and pooling their results. This will allow having more power for the subgroup analyses. This will be done while maintaining trial membership for each patient with a one-stage hierarchical approach. This means that a multilevel analysis will be conducted, considering the individual studies as a level in the analysis. Analyses will be conducted for all b/tsDMARDs together and for each drug class separately -- TNFi, IL-17i, JAKi. Analyses will not be conducted at the level of the individual drugs. Relative OR/relative differences for b/tsDMARD vs PBO in CRP+ vs CRP- will be computed. An interaction between CRP+ and MRI+ will be tested. If statistically significant and clinically relevant, models will be stratified for MRI+ to allow for assessing the effect of CRP+ vs CRP- in those who are MRI+ and MRI- separately. In case of no relevant interaction, then models will be adjusted for MRI+, as well as bDMARD-IR, sex, trial and the main stratification factors of screening of the different trials.
Other interactions with CRP+ will be analysed, namely r-axSpA/nr-axSpA, drug class (TNFi, IL-17i and JAKi), bDMARD-IR.
Narrative Summary: Inflammatory status, particularly CRP levels, predicts treatment response to bDMARDs in axSpA, mainly supported by TNFi trials. Evidence for IL-17i is more limited: in COAST-V, CRP+ patients responded better, yet ixekizumab also showed numerical benefit over placebo in CRP- patients, suggesting a potential effect independent of baseline inflammation. However, current data are insufficient to confirm this hypothesis. Ideally, a stratified RCT would address this, but such a study is unlikely. Existing subgroup analyses are limited by small CRP- and MRI- groups, and by not accounting for MRI status, another key predictor of response. Therefore, the most feasible approach is a systematic, cross-drug, post-hoc analysis of existing RCTs--pooling data where possible and evaluating the role of CRP and MRI as effect modifiers or confounders.
Project Timeline:
The present study will be carried out over a period of 15 months after we obtain the data from all eligible RCTs:
- Submission of the proposal to Vivli
- 6 months: data management, preparation of the datasets to perform the statistical analysis
- 6 months: statistical analysis, discussion and interpretation of findings
- 3 months: abstract and manuscript preparation
Dissemination Plan:
Target audience: Rheumatologist all over the world.
Presentation in the European and American Congress of Rheumatology.
Manuscript publication in a high impact journal of rheumatology (i.e. Annals of Rheumatic Diseases)
Bibliography:
1. Ramiro S, Nikiphorou E, Sepriano A, et al. ASAS-EULAR recommendations for the management of axial spondyloarthritis: 2022 update. Ann Rheum Dis 2023;82:19-34
2. Arends S, Brouwer E, van der Veer E, et al. Baseline predictors of response and discontinuation of tumor necrosis factor-alpha blocking therapy in ankylosing spondylitis: a prospective longitudinal observational cohort study. Arthritis research & therapy 2011;13:R94
3. Glintborg B, Ostergaard M, Krogh NS, et al. Predictors of treatment response and drug continuation in 842 patients with ankylosing spondylitis treated with anti-tumour necrosis factor: results from 8 years’ surveillance in the Danish nationwide DANBIO registry. Ann Rheum Dis 2010;69:2002-8
4. Rudwaleit M, Schwarzlose S, Hilgert ES, et al. MRI in predicting a major clinical response to anti-tumour necrosis factor treatment in ankylosing spondylitis. Ann Rheum Dis 2008;67:1276-81
5. Sieper J, van der Heijde D, Dougados M, et al. Efficacy and safety of adalimumab in patients with non-radiographic axial spondyloarthritis: results of a randomised placebo-controlled trial (ABILITY-1). Ann Rheum Dis 2013;72:815-22
6. Maksymowych WP, Bolce R, Gallo G, et al. Ixekizumab in radiographic axial spondyloarthritis with and without elevated C-reactive protein or positive magnetic resonance imaging. Rheumatology 2022;61:4324-34
7. Goupille P, Wendling D. Initial C-reactive protein level: Discriminating between anti-TNF and anti-IL-17 agents as the first biologic treatment for axial spondyloarthritis? Joint, bone, spine : revue du rhumatisme 2024;91:105662
8. Dougados M, van der Heijde D, Sieper J, et al. Symptomatic efficacy of etanercept and its effects on objective signs of inflammation in early nonradiographic axial spondyloarthritis: a multicenter, randomized, double-blind, placebo-controlled trial. Arthritis & rheumatology 2014;66:2091-102
9. Sieper J, van der Heijde D, Dougados M, et al. A randomized, double-blind, placebo-controlled, sixteen-week study of subcutaneous golimumab in patients with active nonradiographic axial spondyloarthritis. Arthritis & rheumatology 2015;67:2702-12
10. Ortolan A, Webers C, Nikiphorou E. Peripheral manifestations in axial and peripheral spondyloarthritis: a systematic literature review on their assessment in clinical trials. Ann Rheum Dis 2025;84:2135-6
11. Webers C, Ortolan A, Nikiphorou E. The effect of biological and targeted synthetic DMARDs on peripheral manifestations in axial spondyloarthritis: a systematic literature review. Ann Rheum Dis 2025;84:1035-6
12. Capelusnik D, Benavent D, van der Heijde D, et al. Treating spondyloarthritis early: does it matter? Results from a systematic literature review. Rheumatology 2022
13. Benavent D, Navarro-Compan V, Capelusnik D, et al. Does symptom duration impact on treatment response in axial spondyloarthritis? A meta-analysis of randomized controlled trials. Rheumatology 2025