POS0245 BLOOD-BASED EXTRACELLULAR MATRIX BIOMARKERS CAN IDENTIFY ENDOTYPES OF PATIENTS WITH AXSPA AND RESPONDERS TO ADALIMUMAB TREATMENT

H. Port, F. Christiansen, S. Holm Nielsen, P. Frederiksen, A. C. Bay-jensen, M. Karsdal, I. J. Sørensen, M. Østergaard, S. Juhl Pedersen

Research output: Contribution to journalConference abstract in journalResearch

Abstract

Background Axial spondyloarthritis (axSpA) is a chronic inflammatory disease associated with extracellular matrix (ECM) remodelling of the cartilage, bone, and connective tissues. Adalimumab (ADA) is an effective treatment, but not all patients respond, and this may relate to subtypes of the disease (endotypes). Serological quantification of ECM-mediated biomarkers may be useful to identify axSpA endotypes and monitor treatment response to ADA.

Objectives 1) To identify endotypes of patients with axSpA using blood-based ECM biomarkers at baseline and 2) To investigate differences in response to ADA by BASDAI and ASDAS criteria within the endotypes.

Methods ECM biomarkers were measured in serum from patients with axSpA in the three studies (MASH (n=41), DANISH (n=49) and ASIM (n= 45)) at baseline [1–3]. MASH was a cross-sectional study while in DANISH and ASIM patients were randomised to receive treatment with ADA 40 mg or placebo every other week (e.o.w.) for 6 or 12 weeks (ASIM and DANISH, respectively) followed by ADA 40 mg e.o.w. for an additional 18 or 12 weeks (ASIM and DANISH, respectively). Biomarkers of type II collagen formation (PRO-C2), type I collagen degradation (C1M), inflammation (CRP, CRPM) citrullinated and degraded vimentin (VICM) and neutrophil activity (CPa9-HNE) were measured by immunoassays. Biomarker data was log-transformed, standardized by mean centering and scaled by the standard deviation prior to principal component analysis (PCA) and K-means clustering. Response to ADA based on BASDAI50 response, ASDAS clinical important improvement (CII) and major improvement (MI) at study week 24 was compared in the PCA components and between clusters using Mann-Whitney tests. Key demographic parameters were also compared between clusters using Mann-Whitney and chi-squared tests.

Results The variability of baseline ECM biomarker data among patients with axSpA was mainly explained by two dimensions (PC1 and PC2). Type I collagen degradation and inflammation biomarkers (C1M and CRP), reflecting tissue inflammation, were the primary contributors to PC1, whereas type II collagen formation (PRO-C2), reflecting cartilage turnover, contributed the most to PC2 (Figure 1A). In ADA-treated patients, BASDAI50 responders, patients with ASDAS CII, and ASDAS MI had a higher score in PC1 compared to BASDAI50 non-responders, patients with no ASDAS CII and no ASDAS MI, respectively (p=.05, p<.001, p<.001). No differences were observed in PC2. Two distinct clusters were identified from the PC1 and PC2 (Figure 1). Patients in Cluster 1 had higher levels of ECM biomarkers, a greater BASDAI50 response compared to those in Cluster 2 (89.5% vs. 73.2%) and a higher % of patients with ASDAS CII and ASDAS MI (73.7% vs. 32.1% and 92 % vs. 73%, all p<.05). No significant differences were observed in the demographic parameters between Clusters.

Conclusion The PCA analysis identified two orthogonal dimensions related to: inflammation (driven by C1M and CRP) and cartilage turnover (driven by PRO-C2). Two clusters were determined. Patients from Cluster 1, associated with high inflammation, demonstrated a greater BASDAI50 response and ASDAS improvement when treated with ADA than those from Cluster 2. These findings may help profiling treatment response within patients with axSpA.
Original languageEnglish
JournalAnnals of the Rheumatic Diseases
Volume82
Issue numberSuppl 1
Pages (from-to)356-357
Number of pages2
ISSN0003-4967
DOIs
Publication statusPublished - 2023

Cite this