Pos0359 molecular Profiling of Radiographic Axial Spondyloarthritis Patients Reveals an Association Between Innate and Adaptive Cell Populations and Therapeutic Response to Adalimumab

Abstract

Background:The response to treatment in spondylarthropaties is heterogeneous, due to factors yet to be better described. For that reason, it is important to find tools that might help clinicians to decide what is the best available therapeutic option for each patient.Objectives:The goal of this study is to use comprehensive molecular profiling to characterize clinical response to therapy in a real-world setting. Specifically, to identify molecular biomarkers differentiating good responders and non-responders to TNF inhibitors (TNFi) treatment, using adalimumab, in radiographic axial spondyloarthritis | ankylosing spondylitis (r-axSpA|AS) patients context.Methods:Whole-blood mRNA and plasma proteins were measured in a cohort of biologic naïve r-axSpA|AS patients (n = 35) from the Bioefficacy study (Biomarkers identification of anti-TNF alpha agent efficacy in AS patients using RNA sequencing and mass spectrometry), pre and post (14 weeks) TNFi treatment using adalimumab. Response to treatment was categorized according to ASAS20. Results of differential expression analysis were used to identify the most enriched pathways and in predictive models to distinguish responses to TNFi.Results:A treatment-related signature, independent of the type of response, suggests a reduction in inflammatory disease activity. We found genes and proteins robustly differentially expressed between baseline and week 14 in responders, including the GWAS AS-associated genes TNFRSF1A, FCGR2A, TYK2, TBKBP1, IL1R1, IL6R, ICOSLG, IL7R, HHAT and LTBR. Moreover, CRP and HP proteins showed strong and early decrease in the plasma of AS patients, while a cluster of apolipoproteins (APO1, APO2, APO3) showed an increased expression at week 14. Good responders to TNFi treatment tend to have higher expression of innate immunity genes at baseline, and lower expression of markers associated with adaptive immunity, particularly B-cells. A logistic regression model incorporating ASDAS-CRP, gender and Gene x, the top differentially expressed gene at baseline between responders and non-responders, enabled an accurate prediction of response to adalimumab in our cohort (AUC=0.97).Conclusion:Differences in disease activity and/or innate/adaptive immune cell type composition at baseline may be a major contributor to response to adalimumab in r-axSpA|AS. Alternatively, a model including clinical and gene expression variables could be considered, particularly in patients with mild disease activity.Disclosure of Interests:None declared