You are here

  • Home
  • Archive
  • Volume 10, Issue 3
  • Exploratory analysis of the potential disconnect between objective inflammatory response and clinical response following certolizumab pegol treatment in patients with active axial spondyloarthritis

Article Text

Article menu

Download PDFPDF

Spondyloarthritis
Original research
Exploratory analysis of the potential disconnect between objective inflammatory response and clinical response following certolizumab pegol treatment in patients with active axial spondyloarthritis
  1. Martin Rudwaleit1,
  2. Helena Marzo-Ortega2,3,
  3. Victoria Navarro-Compán4,
  4. Rachel Tham5,
  5. Thomas Kumke6,
  6. Lars Bauer6,
  7. Natasha de Peyrecave7,
  8. Mindy Kim8 and
  9. Filip Van den Bosch9
  1. 1University of Bielefeld, Klinikum Bielefeld, Bielefeld, Germany
  2. 2Leeds Institute of Rheumatic and Musculoskeletal Medicine, University of Leeds, Leeds, UK
  3. 3NIHR Leeds Biomedical Research Centre, Leeds Teaching Hospitals NHS Trust, Leeds, UK
  4. 4Department of Rheumatology, La Paz University Hospital, IdiPaz, Madrid, Spain
  5. 5UCB Pharma, Slough, UK
  6. 6UCB Pharma, Monheim, Germany
  7. 7UCB Pharma, Brussels, Belgium
  8. 8UCB Pharma, Atlanta, Georgia, USA
  9. 9Department of Internal Medicine and Pediatrics, Ghent University and VIB Center for Inflammation Research, Ghent, Belgium
  1. Correspondence to Dr Martin Rudwaleit; MARTIN.RUDWALEIT{at}KLINIKUMBIELEFELD.DE

Abstract

Introduction This post hoc analysis evaluated the relationship between objective measures of inflammation and clinical outcomes following 12 weeks of certolizumab pegol (CZP) treatment in patients with active axial spondyloarthritis (axSpA).

Methods We report the proportion of patients achieving ≥50% and ≥75% improvements in clinical composite outcome measures of disease activity (Axial Spondyloarthritis Disease Activity Score [ASDAS], Bath Ankylosing Spondylitis Disease Activity Index [BASDAI]) and objective measures of inflammation (C reactive protein [CRP], Ankylosing Spondylitis spine MRI score [ASspiMRI-a] Berlin score and Spondyloarthritis Research Consortium of Canada [SPARCC] MRI Sacroiliac Joints [SIJ] score) following 12 weeks of CZP treatment. Data from two independent readers over four MRI reading campaigns were pooled using a mixed model with repeated measures for each variable.

Results 136 patients (radiographic axSpA [r-axSpA]: 76; non-radiographic axSpA [nr-axSpA]: 60) were included. Following CZP treatment, CRP, ASspiMRI-a Berlin score and SPARCC SIJ score were reduced by ≥50% in most patients (CRP: 136/136 [100.0%]; Berlin: 73/136 [53.7%]; SPARCC SIJ: 71/136 [52.2%]), and often by ≥75%. Less than half of patients with r-axSpA and nr-axSpA showed ≥50% reduction in clinical responses (BASDAI: 64/136 [47.1%]; ASDAS: 66/136 [48.5%]). These results were also observed at the individual patient level; ≥50% improvements in MRI/CRP inflammatory measures did not translate into similar improvements in clinical responses for most patients.

Conclusion There is a potential disconnect between objective measures of inflammation and clinical outcome responses in patients with axSpA. The use of only clinical response measures as trial endpoints may underestimate anti-inflammatory treatment effects.

Trial registration number NCT01087762.

  • axial spondyloarthritis
  • certolizumab pegol
  • inflammation
  • magnetic resonance imaging

Data availability statement

Data are available upon reasonable request. Qualified researchers whose proposed use of the data has been approved by an independent review panel will be given access to anonymised individual participant data and redacted study documents. Additional information is available at www.clinicalstudydatarequest.com.

http://creativecommons.org/licenses/by-nc/4.0/

This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.

https://doi.org/10.1136/rmdopen-2024-004369

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    WHAT IS ALREADY KNOWN ON THIS TOPIC

    • Axial spondyloarthritis (axSpA) clinical trials use validated composite outcome measures to assess disease activity and measure improvement with treatment. However, these measures are largely based on patient-reported symptoms, which may be subject to confounding factors and therefore may not adequately reflect the extent of the reduction in inflammation following treatment.

    WHAT THIS STUDY ADDS

    • We reveal a potential disconnect between objective measures of inflammation and clinical composite outcome responses in patients with axSpA.

    HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY

    • The use of only clinical measures of disease activity as clinical trial endpoints may underestimate objective anti-inflammatory treatment effects. This could have important implications for the clinical evaluation of patients with axSpA.

    Introduction

    Axial spondyloarthritis (axSpA) is a chronic inflammatory disease affecting the sacroiliac joints (SIJ) and spine with symptoms including chronic inflammatory back pain, morning stiffness and fatigue.1 Many patients also present with other musculoskeletal and extramusculoskeletal manifestations, including peripheral arthritis, enthesitis, dactylitis, anterior uveitis, psoriasis and inflammatory bowel disease.2–5

    AxSpA comprises two subpopulations: radiographic axSpA (r-axSpA; historically known as ankylosing spondylitis)6 and non-radiographic axSpA (nr-axSpA).7 8 In patients with either form of axSpA, uncontrolled inflammation may lead to structural damage and cause limitations in spinal mobility and physical function, and subsequently impair quality of life.9–12 Structural damage in patients with axSpA normally starts in the SIJ,1 11 progresses later to the spine and is relevant for long-term disease outcomes.1 Patients with nr-axSpA do not have evidence of definitive radiographic SIJ damage that fulfils the modified New York classification criteria, but often have substantial MRI inflammation of SIJ and other symptoms of axSpA that impact their quality of life.7 13

    Clinical trials in axSpA use validated composite outcome measures, such as Assessment of SpondyloArthritis international Society (ASAS) response criteria, Axial Spondyloarthritis Disease Activity Score (ASDAS) and Bath Ankylosing Spondylitis Disease Activity Index (BASDAI) to assess disease activity and measure improvement with treatment14 and these are also used in clinical practice to evaluate disease activity and response to treatment. However, these measures are largely based on patient-reported symptoms, which may be subject to confounding factors. A partial disconnect between patient-reported outcomes and presence of inflammation on MRI has been described in earlier studies in which it was demonstrated that high BASDAI or high ASDAS is not necessarily correlated with substantial inflammation on MRI and/or high CRP.15 16 On the other hand, the extent to which the reduction in inflammation, as evaluated by MRI (for the detection of bone marrow oedema) of the SIJ and spine, circulating inflammatory markers such as CRP level, and erythrocyte sedimentation rate correlates with improvements in BASDAI or ASDAS scores following anti-inflammatory treatment is less well studied. A previous study reported that change in BASDAI score at the group level significantly correlated with changes in SIJ MRI evaluations and CRP levels in patients with shorter symptom duration (<4 years) but not in patients with longer symptom duration (4 years or more).17

    We previously reported that in the phase 3 RAPID-axSpA trial, treatment with certolizumab pegol (CZP), an Fc-free, PEGylated tumour necrosis factor inhibitor (TNFi), resulted in improvements in disease activity and inflammation, as measured by composite outcomes measures and MRI.18 19 In this post hoc analysis of the RAPID-axSpA trial, we evaluated the relationship between improvement in objective measures of inflammation, as measured by MRI and CRP, and improvement in clinical composite outcome measures of disease activity in patients with active axSpA after 12 weeks of CZP treatment.

    Methods

    Study design

    RAPID-axSpA (NCT01087762) was a 204-week randomised, multicentre, phase 3 trial which evaluated the efficacy and safety of CZP in adult patients with active axSpA (either nr-axSpA or r-axSpA).18 The trial was double-blind and placebo (PBO)-controlled to week 24, dose-blind to week 48 and open-label to week 204. Enrolled patients had a clinical diagnosis of axSpA and an inadequate response or intolerance to at least one non-steroidal anti-inflammatory drug. Additionally, patients had to have CRP levels >upper limit of normal (ULN: 7.9 mg/L) and/or sacroiliitis on MRI according to the ASAS/Outcome Measures in Rheumatology (OMERACT) definition. Detailed inclusion and exclusion criteria have been reported previously.18

    At baseline, patients were randomised 1:1:1 to PBO, subcutaneous CZP 200 mg every 2 weeks (Q2W) or subcutaneous CZP 400 mg every 4 weeks (Q4W) until week 12 (figure 1). PBO-randomised patients who did not achieve an Assessment of SpondyloArthritis international Society 20% response (ASAS20) at weeks 14 and 16 underwent mandatory escape at week 16 and were randomised to active treatment in a dose-blind manner; at week 24, all remaining PBO-randomised patients were re-randomised 1:1 to CZP 200 mg Q2W or CZP 400 mg Q4W. At week 24, CZP-treated patients remained on their original randomised dose regimen. MRI scans of the spine and SIJ were performed at baseline, week 12 and week 48; however, since MRI was not mandatory in all patients in this trial, MRI data were not available for all patients. CRP samples were taken at screening, baseline and weeks 1, 2, 4, 8, 12, 16, 18, 20, 24, 26, 28, 36, 96 and 156 in all patients.

    Figure 1
    Figure 1

    Study design diagram. MRI set (n=136); includes patients evaluable for MRI and who have at least one MRI assessment. Outcomes are reported for patients in the MRI set initially randomised either to CZP, or to PBO before switching to CZP. aLoading dose of CZP 400 mg at baseline, week 2, week 4; bloading dose of PBO at baseline, week 2, week 4; cloading dose of CZP 400 mg at weeks 24, 26 and 28; dloading dose of CZP 400 mg at weeks 16, 18 and 20; epatients who failed to achieve ASAS20 responses at both weeks 14 and 16 were randomised in a blinded manner to escape treatment from week 16. axSpA, axial spondyloarthritis; CZP, certolizumab pegol; LD, loading dose; PBO, placebo; Q2W, every 2 weeks; Q4W, every 4 weeks; sc, subcutaneous.

    In these post hoc analyses, the pre-CZP period covers outcomes reported prior to CZP treatment while the post-CZP timepoint covers the following 12 weeks for patients in the MRI set initially randomised to either CZP, or to PBO before switching to CZP (figure 1).

    Outcomes

    We report the proportion of patients achieving ≥50% and ≥75% improvements in clinical composite outcome measures of disease activity after 12 weeks of CZP treatment: ASDAS (ranging from <1.3, indicating inactive disease, to >3.5, indicating very high disease activity)20 and BASDAI (ranging from 0 to 10). We also report the proportion of patients achieving ≥50% and ≥75% improvements in objective measures of inflammation: CRP concentration, Ankylosing Spondylitis spine MRI (ASspiMRI-a) Berlin score (ranging from 0 to 3 per vertebral unit with a maximum total score of 69)21 and Spondyloarthritis Research Consortium of Canada (SPARCC) MRI SIJ score.22 These improvement thresholds were chosen to allow comparison across all outcomes. Post-CZP improvements in these outcomes are also shown at the individual patient level in Manhattan plots, stratified by axSpA diagnosis (r-axSpA or nr-axSpA). The relative change from baseline in each of these outcomes was further categorised into change by <0% (worsening), change between 0 and <50%, change between 50 and <75% and change by ≥75%, reflecting different degrees of improvement.

    In patients with pre-CZP inflammation, defined as either active inflammation on MRI (Berlin ≥2 or SPARCC score ≥2) or substantially elevated CRP (≥15 mg/L; cut-off chosen as approximately half of patients in the RAPID-axSpA trial had CRP ≥15 mg/L),18 we report the proportion of patients achieving ≥50% and ≥75% improvements in CRP, Berlin score and SPARCC score. As a secondary analysis, we report post-CZP outcomes stratified by Assessment of SpondyloArthritis international Society 40% response (ASAS40) after 12 weeks of treatment.

    We also report change from baseline (CfB) to week 24 in BASDAI, ASDAS, Berlin score, SPARCC score and CRP stratified by sex.

    Statistical analysis

    To improve precision of patient-level estimates, scoring from two independent readers over four MRI reading campaigns was pooled in a mixed model with repeated measures for each response variable to correct for the variability in diagnostic error rates between readers. A linear mixed effect model was used to estimate empirical best linear unbiased predictor with visit timepoints as fixed effects, patients’ independent MRI reader scores from each campaign as random effects and patients’ response variable from pre-CZP timepoints (up to week 24, dependent on the first CZP treatment for those initially randomised to PBO) as the repeated measure.

    Results are descriptive in nature and are reported as predicted means and proportions; due to the post hoc nature of these analyses, no formal statistical comparisons were performed.

    Results

    Patient disposition and baseline characteristics

    In total, 315 patients were included in the RAPID-axSpA randomised set of whom 158 (50.2%) were included in the MRI substudy. Of these, 136 patients (76 r-axSpA; 60 nr-axSpA) had ≥1 pre-CZP and ≥1 post-CZP MRI assessment and were included in the MRI set. Pre-CZP demographics are presented in table 1. As expected, owing to the inclusion criteria of the trial, baseline clinical composite measures of disease activity and objective measures of inflammation indicated a high level of disease activity across patients. Patient characteristics and measures of clinical composite outcomes were similar between patients with r-axSpA and nr-axSpA, while objective measures of inflammation indicated greater inflammation in patients with r-axSpA compared with patients with nr-axSpA.

    View this table:
    Table 1

    Baseline patient characteristics

    In the MRI set, 55/136 (40.4%) of patients had a baseline CRP level >15 mg/L (the cut-off for substantially elevated CRP in this analysis) and 83.1% of patients had active SIJ or spine inflammation on MRI.

    Outcomes after 12 weeks of CZP treatment

    Outcomes in all patients

    Following 12 weeks of CZP treatment, objective measures of inflammation were reduced by ≥50% in the majority of patients with axSpA (CRP: 136/136 [100.0%]; Berlin: 73/136 [53.7%]; SPARCC: 71/136 [52.2%]), and often by ≥75% (CRP: 18/136 [13.2%]; Berlin: 53/136 [39.0%]; SPARCC: 50/136 [36.8%]), irrespective of pre-CZP value (figure 2). In comparison, fewer patients with axSpA showed a ≥50% reduction in clinical composite measures of disease activity (BASDAI: 64/136 [47.1%]; ASDAS: 66/136 [48.5%]) and very few showed ≥75% reduction (BASDAI: 14/136 [10.3%]; ASDAS: 8/136 [5.9%]). When stratified by axSpA type, broadly similar proportions of patients with r-axSpA and nr-axSpA achieved ≥50% reductions in clinical composite measures of disease activity; however, slightly fewer nr-axSpA patients experienced ≥50% reductions in objective measures of inflammation (figure 3). Similarly, while proportions of patients with r-axSpA and nr-axSpA achieving ≥75% reductions in clinical composite measures of disease activity were comparable, numerically fewer patients with nr-axSpA than r-axSpA achieved ≥75% reductions in SPARCC score (nr-axSpA: 15/60 [25.0%]; r-axSpA: 35/76 [46.1%]) and Berlin score (nr-axSpA: 20/60 [33.3%]; r-axSpA: 33/76 [43.4%]). These results were also observed at the individual patient level: ≥50% improvements in MRI/CRP inflammatory measures did not translate into similar improvements in clinical responses for most patients with either r-axSpA or nr-axSpA, as shown in the Manhattan plots (figure 4).

    Figure 2
    Figure 2

    Proportion of patients with axSpA achieving (A) ≥50% and (B) ≥75% reductions in clinical composite measures of disease activity and objective measures of inflammation. MRI set (n=136); includes patients evaluable for MRI and who have ≥1 pre-CZP and ≥1post-CZP MRI assessment. Includes patients with both nr-axSpA and r-axSpA irrespective of baseline CRP concentrations, ASspiMRI-a Berlin score and SPARCC SIJ score. ASDAS, Axial Spondyloarthritis Disease Activity Score; ASspiMRI-a, Ankylosing Spondylitis spine MRI score; axSpA, axial spondyloarthritis; BASDAI, Bath Ankylosing Spondylitis Disease Activity Index; CRP, C reactive protein; CZP, certolizumab pegol; nr-axSpA, non-radiographic axSpA; r-axSpA, radiographic axSpA; SIJ, sacroiliac joint; SPARCC, Spondyloarthritis Research Consortium of Canada.

    Figure 3
    Figure 3

    Proportion of patients with r-axSpA and nr-axSpA achieving ≥50% and ≥75% reductions in clinical composite measures of disease activity and objective measures of inflammation. MRI set (n=136); patients stratified by r-axSpA (n=76) and nr-axSpA (n=60). Includes patients evaluable for MRI and who have ≥1 pre-CZP and ≥1 post-CZP MRI assessment, irrespective of baseline CRP concentrations, ASspiMRI-a Berlin score and SPARCC SIJ score. ASAS40, Assessment of SpondyloArthritis international Society 40% response criteria; ASDAS, Axial Spondyloarthritis Disease Activity Score; ASspiMRI-a, Ankylosing Spondylitis spine MRI score; axSpA, axial spondyloarthritis; BASDAI, Bath Ankylosing Spondylitis Disease Activity Index; CRP, C reactive protein; CZP, certolizumab pegol; nr-axSpA, non-radiographic axSpA; r-axSpA, radiographic axSpA; SIJ, sacroiliac joint; SPARCC, Spondyloarthritis Research Consortium of Canada.

    Figure 4
    Figure 4

    Improvement in clinical composite measures of disease activity and objective measures of inflammation following 12 weeks of CZP treatment in (A) patients with r-axSpA and (B) patients with nr-axSpA. MRI set (n=136); includes patients evaluable for MRI and who have ≥1 pre-CZP and ≥1 post-CZP MRI assessment. Each bar represents a single patient in the cohort of 76 patients with r-axSpA and 60 patients with nr-axSpA. Across all five endpoints, an individual patient is indicated by an arrow (selected as an example in each cohort with mean [SE] pre-CZP and post-CZP values). The pre-CZP and post-CZP states are indicated by the step-line (the ‘skyline’, or bars [grey outline]) and the dark purple line within each bar (week 12 value), respectively; the space between these points reflects the change in response. Change in response has been grouped into four categories based on the percentage reduction, as shown by the figure legend. Data presented were obtained from a mixed model with repeated measures that was applied to each response variable. For SPARCC, Berlin and ASDAS response variables, a linear mixed effect model was used to estimate empirical best linear unbiased predictor with patients’ response variables from pre-CZP timepoints as random effects. For CRP and BASDAI response variables, a linear mixed effects model was used to estimate empirical best linear unbiased predictor with spatial autocorrelation (using nearby visits to estimate response variable values). ASDAS, Axial Spondyloarthritis Disease Activity Score; ASspiMRI-a, Ankylosing Spondylitis spine MRI score; axSpA, axial spondyloarthritis; BASDAI, Bath Ankylosing Spondylitis Disease Activity Index; CRP, C reactive protein; CZP, certolizumab pegol; nr-axSpA, non-radiographic axSpA; r-axSpA, radiographic axSpA; SIJ, sacroiliac joint; SPARCC, Spondyloarthritis Research Consortium of Canada.

    Outcomes in patients with pre-CZP inflammation

    A similar trend was observed in patients with pre-CZP inflammation (117/136; 86.0%), defined as either active inflammation on MRI (Berlin score ≥2 or SPARCC score ≥2) or substantially elevated CRP (≥15 mg/L). Objective measures of inflammation were reduced by ≥50% in the majority of patients (CRP: 117/117 [100.0%]; Berlin: 68/117 [58.1%]; SPARCC: 63/117 [53.8%]; figure 5). Around half of these patients had ≥50% reduction in clinical composite measures of disease activity (BASDAI: 56/117 [47.9%]; ASDAS: 60/117 [51.3%]).

    Figure 5
    Figure 5

    Proportion of patients with pre-CZP inflammation achieving (A) ≥50% and (B) ≥75% reductions in clinical composite measures of disease activity and objective measures of inflammation. MRI set (n=136); includes patients evaluable for MRI who have ≥1 pre-CZP and ≥1 post-CZP MRI assessment. Pre-CZP inflammation (n=117) is defined as either active inflammation on MRI (either ASspiMRI-a Berlin score ≥2 or SPARCC SIJ score ≥2) or substantially elevated CRP (≥15 mg/L). Includes patients with both nr-axSpA and r-axSpA. ASDAS, Axial Spondyloarthritis Disease Activity Score; axSpA, axial spondyloarthritis; ASspiMRI-a, Ankylosing Spondylitis spine MRI score; BASDAI, Bath Ankylosing Spondylitis Disease Activity Index; CRP, C reactive protein; CZP, certolizumab pegol; nr-axSpA, non-radiographic axSpA; r-axSpA, radiographic axSpA; SIJ, sacroiliac joint; SPARCC, Spondyloarthritis Research Consortium of Canada.

    A greater proportion of ASAS40 responders had ≥50% improvement in ASDAS after 12 weeks of CZP treatment compared with ASAS40 non-responders (75.4% in ASAS40 responders compared with 21.2% in ASAS40 non-responders; figure 6). Likewise, ≥50% improvements in BASDAI were seen in 73.8% of ASAS40 responders but in only 15.4% of ASAS40 non-responders. In contrast, there was little difference between ASAS40 responders and ASAS40 non-responders when comparing the reduction in CRP and MRI scores. Reduction in CRP by ≥75% was achieved in 9.2% of ASAS40 responders and 13.5% of ASAS40 non-responders, and all patients saw a reduction in CRP concentration of ≥50%, irrespective of ASAS40 response. Berlin score improved by ≥50% in 64.6% of ASAS40 responders and in 50.0% of ASAS40 non-responders. A greater proportion of ASAS40 non-responders achieved a ≥50% reduction in SPARCC (55.8%) compared with 52.3% of ASAS40 responders, and ≥75% reduction in SPARCC scores was seen in 35.4% and 38.5% of ASAS40 responders and ASAS40 non-responders, respectively (figure 6).

    Figure 6
    Figure 6

    Clinical composite measures of disease activity and objective measures of inflammation in patients with axSpA and pre-CZP inflammation, stratified by post-CZP ASAS40 achievement. MRI set (n=136); includes patients evaluable for MRI and who have ≥1 pre-CZP and ≥1 post-CZP MRI assessment. Includes patients with both nr-axSpA and r-axSpA. Pre-CZP inflammation is defined as either active inflammation on MRI (either ASspiMRI-a Berlin score ≥2 or SPARCC SIJ score ≥2) or substantially elevated CRP (≥15 mg/L). ASAS40, Assessment of SpondyloArthritis international Society 40% response criteria; ASDAS, Axial Spondyloarthritis Disease Activity Score; ASspiMRI-a, Ankylosing Spondylitis spine MRI activity score; axSpA, axial spondyloarthritis; BASDAI, Bath Ankylosing Spondylitis Disease Activity Index; CZP, certolizumab pegol; nr-axSpA, non-radiographic axSpA; r-axSpA, radiographic axSpA; SIJ, sacroiliac joint; SPARCC, Spondyloarthritis Research Consortium of Canada.

    Of the 117 patients with axSpA (r-axSpA and nr-axSpa) and pre-CZP inflammation, 36 (30.8%) achieved low inflammation (CRP <15 mg/L and SPARCC <2 and Berlin <2) after 12 weeks of CZP treatment (nr-axSpA: 13 (24.5%); r-axSpA: 23 (35.9%)).

    Outcomes stratified by sex

    At week 24 in patients who were randomised to CZP at baseline, clinical composite outcome measures of disease activity were reduced in both male (n=65) and female (n=40) patients. Mean (SD) BASDAI was 6.6 (1.0) at baseline and 3.5 (2.1) at week 24 in female patients (CfB [SD]: −3.1 [1.3]) and 6.5 (1.0) at baseline and 2.8 (1.8) at week 24 in male patients (CfB [SD]: −3.7 [1.6]). Mean (SD) ASDAS was 3.6 (0.7) at baseline and 2.0 (0.9) at week 24 in female patients (CfB [SD]: −1.6 [0.9]) and 4.0 (0.7) at baseline and 1.8 (0.8) at week 24 in male patients (CfB [SD]: −2.1 [1.0]).

    Objective measures of inflammation were also reduced in both male and female patients with axSpA at week 24. Mean (SD) Berlin score was 2.2 (3.7) at baseline and 0.8 (1.1) at week 24 in females (CfB [SD]: −1.4 [3.4]) and 6.7 (6.7) at baseline and 2.6 (3.5) at week 24 in males (CfB [SD]: −4.1 [5.3]). Mean (SD) SPARCC was 7.7 (10.6) at baseline and 3.7 (3.7) at week 24 in females (CfB [SD]: −4.0 [9.0]) and 8.1 (11.1) at baseline and 3.2 (5.5) at week 24 in males (CfB [SD]: −4.9 [8.6]). Mean (SD) CRP concentration was 11.2 (7.6) mg/L at baseline and 3.3 (2.4) mg/L at week 24 in females (CfB [SD]: −7.9 (5.3) mg/L) and 14.5 (12.0) mg/L at baseline and 4.6 (3.7) mg/L at week 24 in males (CfB [SD]: −9.9 [8.3] mg/L).

    Discussion

    In this post hoc analysis which included patients from across the axSpA spectrum, a partial disconnect between the impact of CZP treatment on objective measures of inflammation and clinical composite outcome measures of disease activity was observed, at the group and individual patient level. CZP treatment resulted in a greater proportion of patients showing reductions in objective measures of inflammation compared with improvements in clinical composite outcome measures of disease activity. For example, 39.0%, 36.8% and 13.2% of patients saw ≥75% reductions in Berlin score, SPARCC score and CRP, respectively, while a lower proportion, 10.3% and 5.9%, saw ≥75% reductions in BASDAI and ASDAS, respectively.

    In patients with pre-CZP inflammation, defined as either active inflammation on MRI (Berlin ≥2 or SPARCC ≥2) or substantially elevated CRP concentration (≥15 mg/L), a similar trend was observed in the overall group, most likely because the pre-CZP inflammation group represented the majority of patients (117/136; 86.0%). In the pre-CZP inflammation group, objective measures of inflammation were reduced by ≥50% in the majority of patients, while only around half had a ≥50% reduction in clinical composite measures of disease activity. When defining pre-CZP inflammation, the cut-off for sacroiliitis on MRI was chosen based on earlier studies.22–24 Meanwhile, the ≥15 mg/L substantially elevated CRP cut-off was chosen as approximately half of patients in the RAPID-axSpA trial had CRP concentrations ≥15 mg/L.18

    Patients’ baseline characteristics are important when considering improvements in objective measures of inflammation post treatment; in particular, patients who did not have inflammation on MRI at baseline do not have scope to improve MRI inflammatory measures during treatment. Likewise, in patients with low baseline CRP levels, there is little scope for improvement. Therefore, the reporting of proportions of patients demonstrating ≥50% and ≥75% improvements may underestimate the magnitude of the partial disconnect between improvement in clinical composite outcome measures (ASDAS, BASDAI) and objective measures of inflammation (CRP, SPARCC and Berlin). This disconnect is best visualised in the Manhattan plots which show improvements in both spinal and SIJ MRI scores by ≥75% in the majority of patients, particularly in the r-axSpA population, whereas ≥75% improvements in BASDAI and ASDAS were seen only in a minority of patients. With respect to CRP concentration, all patients with substantially elevated pre-CZP CRP improved by ≥50%, whereas only around half of patients improved by ≥50% in ASDAS and BASDAI. Of note, here we found similar changes in both clinical composite measures of disease activity and objective measures of inflammation in male and female CZP-randomised patients at week 24.

    The partial disconnect between clinical composite measures of disease activity and objective signs of inflammation was also illustrated when patients were stratified according to ASAS40 response, which is considered to represent a major, clinically relevant improvement in axSpA clinical trials. A reduction of ≥50% in BASDAI was seen in almost three-quarters of ASAS40 responders and in only 15.4% of ASAS40 non-responders (which is expected since BASDAI and ASAS40 response are measured using overlapping, mostly core axSpA patient-reported outcomes). At the same time, the proportion of patients achieving a reduction of ≥50% in objective measures of inflammation (CRP concentration, Berlin and SPARCC) were similar in ASAS40 responders and ASAS40 non-responders. This suggests that factors other than active inflammation impact a patient’s perception of their symptoms, physical function and overall disease severity. It should, however, be considered that it may take some time for improvements in inflammation to be translated into improvements in a patient’s perception of disease and symptom severity and ability to resume daily activities, which may have contributed to the disconnect between objective measures of inflammation and disease activity. Factors such as concomitant depression, among others, may prevent improvement in clinical outcome measures of disease activity (ASDAS, BASDAI) despite improvements in objective signs of inflammation.25 For example, a large study in the UK reported a lower ASAS40 response to 12 months of treatment with TNFis in patients with axSpA who suffered from concomitant fibromyalgia (32%) compared with those who did not (42%).26

    Previous cross-sectional and longitudinal observational studies have reported a low correlation between composite measures of disease activity and objective measures of inflammation (SIJ, MRI, CRP levels) in specific patient subgroups.15–17 Poor correlation between improvements of BASDAI scores and improvements of objective measures of inflammation has been reported in patients with axSpA with longer disease durations (≥4 years).17 In agreement with this, patients in this analysis had a median symptom duration of 8.1 years at baseline and demonstrated a disconnect between clinical composite disease severity scores and objective measures of inflammation. Here, symptom duration was longer in patients with r-axSpA (10.3 years) than in patients with nr-axSpA (5.9 years), as would be expected with the natural progression of axSpA. It is likely that there would also be a disconnect in some patients with shorter symptom duration and coexisting comorbidities such as fibromyalgia. This emphasises the importance of not only considering the improvements in objective measures of inflammation (MRI, CRP) but also considering patients’ symptom durations when evaluating clinical response to treatments for axSpA.

    In the 2022 update of the ASAS/European Alliance of Associations for Rheumatology recommendations for the management of axSpA, it is now recommended that in the absence of treatment response, diagnosis should be re-evaluated with the consideration of the presence of comorbidities.14 This highlights the need to consider objective measures of inflammation in these patients; in patients with active axSpA and objective signs of inflammation, biological therapy may be of benefit. Treating these objective signs of early inflammation is likely to reduce structural progression of axSpA and improve overall disease outcomes, despite any suboptimal responses captured by ASDAS and BASDAI, which capture more subjective symptoms of axSpA which are harder to treat, such as fatigue.27–30

    Limitations of this study include that the data were from a clinical trial with defined patient inclusion criteria and, therefore, may not reflect that of a real-world population. Indeed, for inclusion in the RAPID-axSpA trial, all patients had to have CRP levels >ULN (7.9 mg/L) or active inflammation on MRI according to the ASAS/OMERACT definition as an objective measure of disease activity.18 31 Here, 40.4% of patients had a CRP concentration ≥15 mg/L, a cut-off chosen to represent substantial elevation (based on the fact that approximately half of patients in the original RAPID-axSpA trial had CRP levels above this threshold).18 However, it should be acknowledged that this threshold is not supported by a specific definition within the field and other cut-off levels for the definition of substantially elevated CRP were not explored. Similarly, while reporting the proportions of patients who achieve thresholds of ≥50% and ≥75 reductions in objective measures of inflammation and clinical composite measures of disease activity is relatively common practice, their blanket application to different scales of reporting may not be sufficiently generalisable to derive definitive conclusions. Further, outcomes were only evaluated up to 12 weeks of treatment with CZP; evaluation of the disconnect into the longer term may be valuable. This being said, it has been previously shown that the majority of the reduction in inflammation on MRI occurs up to week 12 of treatment in patients with axSpA.32 Additional serum biomarkers of inflammation may have further supported the conclusions of this study, but these data were not collected during the RAPID-axSpA trial.

    The relatively limited sample size of patients who were evaluable for MRI and who had ≥1 pre-CZP and ≥1 post-CZP MRI assessment (and were thus included in the present analysis), relative to the total study sample in the RAPID-axSpA trial, must also be acknowledged as a limitation. It would have been interesting to have compared differences between those who had MRIs and those who did not. Finally, the scales of the five outcomes assessed did not have linear psychometric properties and ranges, and therefore a 50% or 75% reduction in one measure may not be proportional to that of another measure.

    To conclude, this post hoc analysis revealed a potential disconnect between objective measures of inflammation, as measured by clinical composite outcome responses of disease activity, and MRI and CRP concentration in patients with axSpA. In the majority of patients, CZP treatment resulted in a reduction of objective measures of inflammation, irrespective of improvements in clinical symptoms and measures of disease activity. The use of only clinical measures of disease activity as clinical trial endpoints may, therefore, underestimate objective anti-inflammatory treatment effects. This disconnect could have important implications for the clinical evaluation of patients with axSpA.

    Data availability statement

    Data are available upon reasonable request. Qualified researchers whose proposed use of the data has been approved by an independent review panel will be given access to anonymised individual participant data and redacted study documents. Additional information is available at www.clinicalstudydatarequest.com.

    Ethics statements

    Patient consent for publication

    Ethics approval

    Written informed consent was obtained from all patients. The study protocol, amendments and subject-informed consent were reviewed by national, regional, or Independent Ethics Committee or Institutional Review Board prior to implementation. RAPID-axSpA was conducted in accordance with local regulations, the International Council for Harmonisation Good Clinical Practice requirements and the Declaration of Helsinki. Participants gave informed consent to participate in the study before taking part.

    Acknowledgments

    The authors thank the patients, the investigators and their teams who took part in this study. The authors also acknowledge Simone E. Auteri, PhD, UCB Pharma, for publication coordination, and Erin Clarkson, BSc and Jane Spingardi, DPhil, Costello Medical, UK, for medical writing and editorial assistance based on the authors’ input and direction. This study was funded by UCB Pharma. This work was previously presented at the ACR Convergence 2022.33

    References

    1. ↵
      1. Navarro-Compán V,
      2. Sepriano A,
      3. El-Zorkany B, et al
      . Axial spondyloarthritis. Ann Rheum Dis 2021;80:1511–21. doi:10.1136/annrheumdis-2021-221035
      OpenUrlAbstract/FREE Full Text
    2. ↵
      1. Erol K,
      2. Gok K,
      3. Cengiz G, et al
      . Extra-articular manifestations and burden of disease in patients with radiographic and non-radiographic axial spondyloarthritis. Acta Reumatol Port 2018;43:32–9.
      OpenUrl
    3. ↵
      1. Schwartzman S,
      2. Ruderman EM
      . A road map of the axial spondyloarthritis continuum. Mayo Clin Proc 2022;97:134–45. doi:10.1016/j.mayocp.2021.08.007
      OpenUrl
    4. ↵
      1. Rademacher J,
      2. Poddubnyy D,
      3. Pleyer U
      . Uveitis in spondyloarthritis. Ther Adv Musculoskelet Dis 2020;12:1759720X20951733. doi:10.1177/1759720X20951733
    5. ↵
      1. Biggioggero M,
      2. Crotti C,
      3. Becciolini A, et al
      . The management of acute anterior uveitis complicating spondyloarthritis: present and future. Biomed Res Int 2018;2018:9460187. doi:10.1155/2018/9460187
    6. ↵
      1. Boel A,
      2. Molto A,
      3. van der Heijde D, et al
      . Do patients with axial spondyloarthritis with radiographic sacroiliitis fulfil both the modified New York criteria and the ASAS axial spondyloarthritis criteria? Results from eight cohorts. Ann Rheum Dis 2019;78:1545–9. doi:10.1136/annrheumdis-2019-215707
      OpenUrlAbstract/FREE Full Text
    7. ↵
      1. Rudwaleit M,
      2. Haibel H,
      3. Baraliakos X, et al
      . The early disease stage in axial spondylarthritis: results from the German Spondyloarthritis Inception Cohort. Arthritis Rheum 2009;60:717–27. doi:10.1002/art.24483
      OpenUrlCrossRefPubMedWeb of Science
    8. ↵
      1. Baraliakos X,
      2. Braun J
      . Non-radiographic axial spondyloarthritis and ankylosing spondylitis: what are the similarities and differences? RMD Open 2015;1:e000053. doi:10.1136/rmdopen-2015-000053
    9. ↵
      1. Braun J,
      2. Baraliakos X,
      3. Kiltz U
      . Treat-to-target in axial spondyloarthritis - what about physical function and activity? Nat Rev Rheumatol 2021;17:565–76. doi:10.1038/s41584-021-00656-5
      OpenUrlPubMed
    10. ↵
      1. Braun J,
      2. Sieper J
      . Ankylosing spondylitis. Lancet 2007;369:1379–90. doi:10.1016/S0140-6736(07)60635-7
      OpenUrlCrossRefPubMedWeb of Science
    11. ↵
      1. Hamilton-West KE,
      2. Quine L
      . Living with ankylosing spondylitis: the patient’s perspective. J Health Psychol 2009;14:820–30. doi:10.1177/1359105309341394
      OpenUrlCrossRefPubMedWeb of Science
    12. ↵
      1. Singh JA,
      2. Strand V
      . Spondyloarthritis is associated with poor function and physical health-related quality of life. J Rheumatol 2009;36:1012–20. doi:10.3899/jrheum.081015
      OpenUrlAbstract/FREE Full Text
    13. ↵
      1. van der Linden S,
      2. Valkenburg HA,
      3. Cats A
      . Evaluation of diagnostic criteria for ankylosing spondylitis. A proposal for modification of the New York criteria. Arthritis Rheum 1984;27:361–8. doi:10.1002/art.1780270401
      OpenUrlCrossRefPubMedWeb of Science
    14. ↵
      1. Ramiro S,
      2. van der Heijde D,
      3. van Tubergen A, et al
      . Higher disease activity leads to more structural damage in the spine in ankylosing spondylitis: 12-year longitudinal data from the OASIS cohort. Ann Rheum Dis 2014;73:1455–61. doi:10.1136/annrheumdis-2014-205178
      OpenUrlAbstract/FREE Full Text
    15. ↵
      1. Navarro-Compán V,
      2. Ramiro S,
      3. Landewé R, et al
      . Disease activity is longitudinally related to sacroiliac inflammation on MRI in male patients with axial spondyloarthritis: 2-years of the DESIR cohort. Ann Rheum Dis 2016;75:874–8. doi:10.1136/annrheumdis-2015-207786
      OpenUrlAbstract/FREE Full Text
    16. ↵
      1. Byravan S,
      2. Jain N,
      3. Stairs J, et al
      . Is there a correlation between patient-reported bath ankylosing spondylitis disease activity index (BASDAI) score and MRI findings in axial spondyloarthropathy in routine clinical practice? Cureus 2021;13:e19626. doi:10.7759/cureus.19626
    17. ↵
      1. Weiß A,
      2. Song I-H,
      3. Haibel H, et al
      . Good correlation between changes in objective and subjective signs of inflammation in patients with short- but not long duration of axial spondyloarthritis treated with tumor necrosis factor-blockers. Arthritis Res Ther 2014;16:R35. doi:10.1186/ar4464
    18. ↵
      1. Landewé R,
      2. Braun J,
      3. Deodhar A, et al
      . Efficacy of certolizumab pegol on signs and symptoms of axial spondyloarthritis including ankylosing spondylitis: 24-week results of a double-blind randomised placebo-controlled Phase 3 study. Ann Rheum Dis 2014;73:39–47. doi:10.1136/annrheumdis-2013-204231
      OpenUrlAbstract/FREE Full Text
    19. ↵
      1. van der Heijde D,
      2. Baraliakos X,
      3. Hermann K-GA, et al
      . Limited radiographic progression and sustained reductions in MRI inflammation in patients with axial spondyloarthritis: 4-year imaging outcomes from the RAPID-axSpA phase III randomised trial. Ann Rheum Dis 2018;77:699–705. doi:10.1136/annrheumdis-2017-212377
      OpenUrlAbstract/FREE Full Text
    20. ↵
      1. Machado PM,
      2. Landewé R,
      3. van der Heijde D
      . Ankylosing Spondylitis Disease Activity Score (ASDAS): 2018 update of the nomenclature for disease activity states. Ann Rheum Dis 2018;77:1539–40. doi:10.1136/annrheumdis-2018-213184
      OpenUrlCrossRefPubMed
    21. ↵
      1. Lukas C,
      2. Braun J,
      3. van der Heijde D, et al
      . Scoring inflammatory activity of the spine by magnetic resonance imaging in ankylosing spondylitis: a multireader experiment. J Rheumatol 2007;34:862–70.
      OpenUrlAbstract/FREE Full Text
    22. ↵
      1. van der Heijde D,
      2. Landewé R,
      3. Hermann K-G, et al
      . Is there a preferred method for scoring activity of the spine by magnetic resonance imaging in ankylosing spondylitis? J Rheumatol 2007;34:871–3.
      OpenUrlAbstract/FREE Full Text
    23. ↵
      1. Maksymowych WP,
      2. Østergaard M,
      3. Landewé R, et al
      . Impact of filgotinib on sacroiliac joint magnetic resonance imaging structural lesions at 12 weeks in patients with active ankylosing spondylitis (TORTUGA trial). Rheumatology (Oxford) 2022;61:2063–71. doi:10.1093/rheumatology/keab543
      OpenUrl
    24. ↵
      1. van den Berg R,
      2. de Hooge M,
      3. Bakker PAC, et al
      . Metric properties of the SPARCC score of the sacroiliac joints - data from baseline, 3-month, and 12-month followup in the SPACE cohort. J Rheumatol 2015;42:1186–93. doi:10.3899/jrheum.140806
      OpenUrlAbstract/FREE Full Text
    25. ↵
      1. Macfarlane GJ,
      2. Pathan E,
      3. Jones GT, et al
      . Predicting response to anti-TNFα therapy among patients with axial spondyloarthritis (axSpA): results from BSRBR-AS. Rheumatology (Oxford) 2020;59:2481–90. doi:10.1093/rheumatology/kez657
      OpenUrl
    26. ↵
      1. Macfarlane GJ,
      2. MacDonald RIR,
      3. Pathan E, et al
      . Influence of co-morbid fibromyalgia on disease activity measures and response to tumour necrosis factor inhibitors in axial spondyloarthritis: results from a UK national register. Rheumatology (Oxford) 2018;57:1982–90. doi:10.1093/rheumatology/key206
      OpenUrl
    27. ↵
      1. Furst DE,
      2. Louie JS
      . Targeting inflammatory pathways in axial spondyloarthritis. Arthritis Res Ther 2019;21:135. doi:10.1186/s13075-019-1885-z
      OpenUrl
    28. ↵
      1. Dougados M,
      2. Demattei C,
      3. van den Berg R, et al
      . Rate and predisposing factors for sacroiliac joint radiographic progression after a two-year follow-up period in recent-onset spondyloarthritis. Arthritis Rheumatol 2016;68:1904–13. doi:10.1002/art.39666
      OpenUrl
    29. ↵
      1. Bennett AN,
      2. McGonagle D,
      3. O’Connor P, et al
      . Severity of baseline magnetic resonance imaging-evident sacroiliitis and HLA-B27 status in early inflammatory back pain predict radiographically evident ankylosing spondylitis at eight years. Arthritis Rheum 2008;58:3413–8. doi:10.1002/art.24024
      OpenUrlCrossRefPubMedWeb of Science
    30. ↵
      1. Sepriano A,
      2. Ramiro S,
      3. Landewé R, et al
      . Is active sacroiliitis on MRI associated with radiographic damage in axial spondyloarthritis? Real-life data from the ASAS and DESIR cohorts. Rheumatology (Oxford) 2019;58:798–802. doi:10.1093/rheumatology/key387
      OpenUrlPubMed
    31. ↵
      1. Rudwaleit M,
      2. Landewé R,
      3. van der Heijde D, et al
      . The development of assessment of spondylo arthritis international society classification criteria for axial spondyloarthritis (part I): classification of paper patients by expert opinion including uncertainty appraisal. Ann Rheum Dis 2009;68:770–6. doi:10.1136/ard.2009.108217
      OpenUrlAbstract/FREE Full Text
    32. ↵
      1. Braun J,
      2. Baraliakos X,
      3. Hermann K-G, et al
      . Effect of certolizumab pegol over 96 weeks of treatment on inflammation of the spine and sacroiliac joints, as measured by MRI, and the association between clinical and MRI outcomes in patients with axial spondyloarthritis. RMD Open 2017;3:e000430. doi:10.1136/rmdopen-2017-000430
    33. ↵
      1. Rudwaleit M,
      2. Van den Bosch F,
      3. Marzo-Ortega H, et al
      . POS0683 AN exploratory analysis of the potential disconnect between objective inflammatory response and clinical response following certolizumab pegol treatment in patients with active axial spondyloarthritis. EULAR 2023 European Congress of Rheumatology, 31 May - 3 June. Milan, Italy; June 2023 doi:10.1136/annrheumdis-2023-eular.3141

    Footnotes

    • Contributors Substantial contributions to study conception and design; substantial contributions to analysis and interpretation of the data; drafting the article or reviewing it critically for important intellectual content; final approval of the version of the article to be published: all authors. MR is guarantor.

    • Funding This study was sponsored by UCB Pharma. This article was based on the original study RAPID-axSpA (NCT01087762) which was sponsored by UCB Pharma. Support for third-party writing assistance for this article, provided by Erin Clarkson, BSc, and Jane Spingardi, DPhil, Costello Medical, UK, was funded by UCB Pharma in accordance with Good Publication Practice (GPP) 2022 guidelines (https://www.acpjournals.org/doi/10.7326/M22-1460).

    • Competing interests MR: Speakers bureau from AbbVie, Boehringer Ingelheim, Chugai, Eli Lilly, Janssen, Novartis, Pfizer, UCB Pharma; consultant of AbbVie, Eli Lilly, Novartis, UCB Pharma. HM-O: Research grants from Janssen, Novartis, Pfizer and UCB Pharma. Speaking honoraria and/or consultancy fees from AbbVie, Amgen, Biogen, Eli Lilly, Janssen, MoonLake, Novartis, Pfizer, Takeda and UCB Pharma. HM-O is supported by the National Institute for Health Research (NIHR) Leeds Biomedical Research Centre (LBRC). The views expressed are those of the authors and not necessarily those of the (UK) National Health Service (NHS), the NIHR, or the (UK) Department of Health. VN-C: Speakers bureau for AbbVie, Eli Lilly, Fresenius Kabi, Janssen, MSD, Novartis, Pfizer and UCB Pharma; consultant for AbbVie, Eli Lilly, Galapagos, MoonLake, MSD, Novartis, Pfizer and UCB Pharma; grant/research support from AbbVie and Novartis. RT: Veramed statistical consultant for UCB Pharma. TK, LB, NdP and MK: Employee and stockholder of UCB Pharma. FvdB: Consultancy fees from AbbVie, Amgen, Eli Lilly, Galapagos, Janssen, Merck, Novartis, Pfizer and UCB Pharma; speakers bureau for AbbVie, Amgen, Bristol Myers-Squibb, Celgene, Janssen, Merck, Novartis, Pfizer and UCB Pharma.

    • Provenance and peer review Not commissioned; externally peer reviewed.