OP0113 WHOLE BLOOD RNA EXPRESSION IN CLINICALLY SUSPECTED ARTHRALGIA PATIENTS SHOWS A POTENTIAL ADDED VALUE IN PREDICTION OF INFLAMMATORY ARTHRITIS. (2nd June 2020)

• Record Type: Journal Article
• Authors: Niemantsverdriet, E.; Van den Akker, E.; Van den Eeden, S.; Geluk, A.; Van der Helm-van Mil, A.
• Is Part Of: Annals of the rheumatic diseases. Volume 79(2020)Supplement 1
• Page Start: 74
• Page End: 75
• Publication Date: 2020-06-02
• Subjects: Rheumatism -- Periodicals; 616.723005
• DOI: 10.1136/annrheumdis-2020-eular.3042
• Languages: English
• ISSNs: 0003-4967

Abstract

Abstract : Background: Clinically suspect arthralgia (CSA) can precede development of clinically evident inflammatory arthritis (IA). Autoantibody, C-reactive protein (CRP), and subclinical inflammation are known predictors, but risk estimation remains insufficiently accurate. Especially CRP has a small effect size and is inadequately reflective of inflammation that can be measured systemically. RNA expression in whole blood of patients with rheumatoid arthritis (RA) have shown differences compared to healthy individuals. Therefore, we hypothesized that differences in RNA expression can be found between CSA patients that do and do not progress to IA. Objectives: This study assessed whole blood RNA expression levels of inflammatory and immune genes as potential biomarkers for prediction of IA-development in patients with arthralgia. Methods: Between April 2012-March 2015, 234 patients were consecutively included in the Leiden CSA-cohort. Follow-up ended when patients developed clinically apparent IA (determined at physical examination), or else after 2-years. RNA expression in whole blood, at the moment of inclusion, was determined for 135 genes of the innate and adaptive immune system by dual color Reverse-Transcription Multiplex Ligation-dependent Probe Amplification (dcRT-MLPA) profiling. Cox proportional hazard models were used to associate time-to-event with gene expression level at inclusion, while adjusting for age, gender, and assay plate (model 1). The false discovery rate was used to correct for multiple testing. Genes with significantly different expression were subsequently studied for reproducibility by qPCR, and mutual independence in their association with IA-development. For the latter, we employed a forward selection strategy, starting with the most significantly associated gene and iteratively adding more genes. Resulting mutually independent genes were further investigated for their added predictive value over known risk factors, CRP, ACPA, and subclinical joint inflammation. Results: 21% of CSA-patients developed IA after mean 3.6 months (IQR:1.6-10.7) follow-up. After correction for multiple testing, six genes were significantly associated with IA-development (model 1), namely IFN-γ, PHEX, IGF-1, IL7R, CD19, or CCR7 (ordered by significance). For all six genes, a lower expression at inclusion was associated with an increased risk of IA-development. IFN-γ was only weakly expressed in peripheral blood, hampering the technical reproducibility between MLPA and qPCR results, and was excluded for further analyses. PHEX and IGF-1 were highly correlated (R 2 0.97) and only IGF-1, but not PHEX, was included in further analyses. Of the remaining significant genes (IGF-1, IL7R, CD19, CCR7), an independent association with IA-development was observed for IGF-1 and IL7R, but not for CD19 or CCR7. qPCR data of IL7R correlated with MLPA results (p<0.001), confirming the robustness of the transcriptomic outcome. Lastly, when analysing IGF-1 and IL7R with known clinical predictors, both IGF-1 and IL7R remained independently associated with progression to IA. Conclusion: Six genes were differentially expressed between CSA-patients that did or did not progress to IA. Expression of IGF-1 and IL7R showed added value to regularly used predictors. Validation is warranted. References: None. Disclosure of Interests: None declared