Background A recent proof of concept, open label trial (TOPAS)¹ demonstrated the efficacy of ustekinumab (a monoclonal antibody against IL12p40) in patients with active ankylosing spondylitis (AS).

Objectives To identify novel serum biomarkers associated with disease activity and response to ustekinumab treatment in patients with AS.

Methods We profiled 20 AS patients from the TOPAS study¹ (treated with 90 mg ustekinumab SC at baseline, wk 4 and wk 16) and compared to an independent cohort of 30 healthy normal volunteers (HNV). Samples collected at wk0 and wk24 were assayed using the SOMAmer™ (Slow Off-rate Modified Aptamer)-based proteomic assay², which measured 1129 analytes. A general linear model was applied to identify analytes that differed between AS and HNV and within subjects before and after treatment using a fold change (FC) cutoff of FC>1.3 and p value cutoff of p<0.05. Correlation between serum markers and disease activity scores was evaluated by the Spearman rank correlation test. Results were interpreted via Ingenuity Pathway Analysis (IPA). Disease activity was specified by Bath Ankylosing Spondylitis Disease Activity Index (BASDAI), Ankylosing Spondylitis Disease Activity Score (ASDAS) and Osteitis MRI Scores (Berlin method) for sacroiliac joints and spine. Clinical response to ustekinumab was defined as 50% improvement in BASDAI at wk24 as compared to baseline.

Results 113 analytes were differentially expressed in AS compared to HNV. The top dysregulated pathways were Acute Phase Response Signaling, Role of Macrophages, Fibroblasts and Endothelial Cells and Inflammation. Following ustekinumab treatment, proteins involved in Acute Phase Response Signaling were significantly modulated and C-reactive Protein (CRP) decreased (p=0.009) in BASDAI 50 responders but not non-responders. At baseline, Stanniocalcin-1 (STC1, p=0.0012), Thyroid Stimulating Hormone (p=0.0051), Tumor Necrosis Factor Receptor Superfamily 2 (TNFRSF2, p=0.0132) and Spondin-1 (SPON1, p=0.0135) correlated to BASDAI score, and Interleukin 13 Receptor alpha 1 (IL13RA1, p=0.0019), and Bone Morphogenetic Protein Receptor 1A (BMPR1A, p=0.0036) correlated to ASDAS. Chk protein kinase 1 (CHEK1, p=0.0017), and ADAM metallopeptidase 5 (ADAMST5, p=0.0024) correlated to baseline osteitis on MRI and markers known to be involved in bone remodeling i.e. Dickkopf Like Protein 1 (DKKL1, p=0.0092), Semaphorin 3E (SEMA3E, p=4.29E-05), SPON1 (p=0.0002), ADAM12 (p=0.0002), and CLIC1 (p=0.0003) correlated with wk24 change in osteitis and spinal osteitis scores. Baseline levels of acute phase markers showed a trend for association with response. BASDAI 50 responders had decreased Complement component 3 (C3, p=0.0120) and fibronectin (FN, p=0.0131), increased Haptoglobin (HP, p=0.0095) and predicted inhibition of Colony Stimulating Factor 1 (CSF-1, implicated in osteoclast differentiation) at baseline.

Conclusions Evidence for significant dysregulation of inflammatory pathways and bone remodeling can be demonstrated from the serum of AS patients. Novel serum markers associated with AS disease activity and response to ustekinumab, were identified. We plan for future assessment of these markers in an expanded placebo controlled cohort.

References

1. Ann Rheum Dis 2014;73:817–823.

2. PLoS ONE 5:e15004.

Disclosure of Interest B. Dasgupta Employee of: Janssen Research & Development, LLC., S. Telesco Employee of: Janssen Research & Development, LLC, J. Sieper Grant/research support from: Janssen Research & Development, LLC., D. Poddubnyy Grant/research support from: Janssen Research & Development, LLC