POS0975 VALIDATION OF INTERFERON STIMULATED GENE SCORES, PREVIOUSLY DEVELOPED IN PERIPHERAL BLOOD MONONUCLEAR CELLS, IN WHOLE BLOOD SAMPLES

Background A recent EULAR Task force identified critical variability in interferon (IFN) assay methods and defined a research agenda to improve harmonise assays. IFN scores B, developed using stimulated genes (ISG) measured peripheral blood mononuclear cells (PBMC) differentiate between clinical subsets of patients with autoimmune disease [1]. Whole (WB) samples are more commonly used practice trials. Regardless sample type, real time polymerase chain reaction (RT-PCR) experiments require suitable reference gene (RG) which is uniformly expressed across different experimental conditions order normalise target expression quantities [2]. Objectives (i) To determine whether type affects the previously scores; (ii) if so, develop revised WB; (iii) identify best RG for future RT-PCR experiments. Methods Samples were collected from participants enrolled 5 studies; DEFINITION, CONVAS, USEFUL, MASTERPLANs BRAGGSS. PBMCs separated leucosep tubes WB was Tempus™ Blood RNA Tubes. Undetected delta Ct values singly imputed. B calculated paired PBMC agreement types assessed via Bland-Altman (BA) method. New derived exploratory factor analysis (EFA) 31 ISG as per original work Separation subgroups visualized violin plots effect sizes Hedges’ g. identification 16 candidates online software. Results We 342 systemic lupus erythematosus (SLE), 6 ANA positive at risk progressors, 7 non-progressors, 30 rheumatoid arthritis (RA) 10 healthy controls (HC). Subsets comparison (n=45) (n=22). Comparing score versus matched resulted bias -1.44, limits (LOA 0.08, -2.96), spread data on BA plot showed systematic difference. The same gave -1.23 0.26, 2.73), suggested differences not systematic. EFA produced two-factor solution explained 97% variation. first (score C) contained 3 second D) 4 (Table 1), compared (12 ISG) (14 Both potential HC those SLE or RA, but neither RA had done (Figure 1). YWHAZ, PGK1 GUSB most stable RG. Others such ACTB performed poorly Conclusion may be interchangeably calculate A, B. This suggests an influence neutrophils, including demethylation neutrophil cytosine-guanine sites, Score [3]. In we further two differentiated HC, although distinguished RA. Further needed assess performance A-D samples. Minimum Information Publication Quantitative Real-Time PCR Experiments recommend ≥3 avoid bias[4]; found that range conditions. References [1]El-Sherbiny, Y.M. et al., Sci Rep, 2018. 8(1): p. 5793. [2]Schmittgen, T.D. K.J. Livak, Nat Protoc, 2008. 3(6): 1101-8. [3]Coit, P., J Autoimmun, 2015. 58: 59-66. [4]Bustin, S.A., BMC Molecular Biology, 2010. 11(1): 74. Table 1. Interferon selected whole blood. Variable C D CCL8 1.01 IFI27 0.84 CXCL10 0.80 LAMP3 0.73 ISG15 0.66 CASP1 0.93 CEACAM1 0.89 TRIM38 SOCS1 0.58 Figure Violin (bold dashed line = median, dotted lines upper lower quartiles) presented top value 95% CI underneath. Acknowledgements: NIL. Disclosure Interests None Declared.