320. TYPE 1 INTERFERON EXPRESSION IS ASSOCIATED WITH AUTOANTIBODIES ACROSS SYSTEMIC AUTOIMMUNE DISEASES: RESULTS FROM THE LUPUS EXTENDED AUTOIMMUNE PHENOTYPE STUDY

Abstract

Background: Type I interferons have been implicated in the pathogenesis of systemic lupus erythematosus (SLE), but much less is known about their role in other connective tissue diseases (CTDs). We aimed to determine the prevalence of a whole-blood type I interferon signature across CTDs, and to identify clinical and serological factors associated with elevated interferon pathway activity. Methods: Subjects were recruited from Central Manchester University Hospitals, UK between May 2014 and March 2016. Patients were categorized by their physician diagnosis into SLE, undifferentiated CTD (UCTD), mixed CTD (MCTD), Sjogren's syndrome (SS) and systemic sclerosis (SSc). Autoantibodies (ds-DNA, Ro, La, Smith, U1RNP, CCP, chromatin, centromere, Jo-1 and Scl-70) were measured using multiplex ELISA. RT-qPCR was performed on cDNA derived from whole blood and the median fold change of six interferonsimulated genes (IFI27, IFI44L, IFIT1, ISG15, RSAD2, SIGLEC1) was compared with the median of healthy controls, to create an interferon sensitive gene (ISG) score for each patient. Scores higher than the mean of the controls plus two standard deviations (>2.466) were designated as a positive score. Results: We recruited 95 subjects with a median (IQR) age of 48.3 (32.9, 57.4) years. 88 (92.6%) were female, 72/94 (77%) were Caucasian and the median disease duration was 7.68 (3.06, 14.81) years. The most commonly present autoantibodies were anti-U1RNP (26.3%), anti-Ro/SSA (25.2%) and anti-dsDNA (24.2%). In total, 60/95 (63%) subjects had at least 1 positive autoantibody from the above list. Across all subjects, 33/95 (34.7%) had a positive ISG score. The median ISG scores was higher in patients with Sjogren's syndrome or MCTD, compared to healthy controls (Kruskal-Wallis, p=0.005 across all groups). All 3 SSc patients had a negative ISG score. In univariate logistic regression models (excluding anti-centromere and anti-Jo1 due to low numbers), a positive ISG was significantly associated with the presence of ds-DNA, Smith, Ro, RNP and chromatin antibodies. After adjustment for age, gender, ethnicity (Caucasian or non-Caucasian) disease duration and clinical diagnosis, the presence of rheumatoid factor and anti-Smith or -Ro antibodies remained associated with a positive ISG score. A significant association was also observed between the number of autoantibodies (range 0-5) and a positive ISG score [OR 2.4 (1.64, 3.50), p<0.001]. In a multivariable logistic regression model this observation remained significant after adjustment for age, gender, ethnicity and diagnosis [OR 1.8 (1.24, 2.88), p=0.003]. Conclusion: Expression of a type I interferon signature differs across CTD subtypes and is not observed in UCTD and SSc. The strongest factor associated with a positive ISG score was the type and number of autoantibodies, especially those binding to RNA antigens.