Background Idiopathic inflammatory myopathies (IIM) are rare autoimmune diseases characterized by proximal muscle weakness and muscle inflammation. Recent studies suggest a pathogenic role for type I interferon (IFN) in IIM, particularly within the subgroup dermatomyositis (DM). Expression of type I IFN and IFN-regulated genes, the IFN signature, in muscle tissue and skin biopsies from DM patients have been reported. Studies indicate that an IFN signature in peripheral blood cells correlates to disease activity although results are contradictory. Autoantibodies are a characteristic feature of IIM, and immune complexes have been demonstrated to induce type I IFN expression in target cells. Its effect and correlation with clinical phenotype are, however, poorly understood.

Objectives The aim of this study was to examine if the type I IFN activity of sera from IIM patients correlates with disease activity, clinical manifestations, autoantibody profile, or HLA haplotype.

Methods Clinical, serological, genetic and laboratory data were collected from 132 IIM patients recruited from Karolinska University Hospital, with diagnoses DM, polymyositis, inclusion body myositis and juvenile DM. Patient sera were assessed for their type I IFN activity using an in vitro system with a cell line. Expression of type I IFN-inducible genes, IFIT1, IFI44, and MX1, were quantified using real time-PCR. Two groups of patients, IFN+ (n=13) and IFN- (n=119), were categorized based on IFN score, the sum of individual gene expression scores. Differences between the two groups were assessed for clinical, serological and genetic variables, as well as correlation between IFN score and variables.

Results The proportion of the IIM subgroups within the IFN+ and IFN- patient groups was equal. No correlation between type I IFN activity in sera and disease activity was found for any of the IIM subgroups, and no significant differences between the two patient categories for the assessed clinical or laboratory variables were found. Autoantibody analysis demonstrated that significantly more IFN+ patients were positive for anti-nuclear antibodies (ANA) compared to IFN- patients (p=0.001). HLA-DRB1 typing of patients revealed no differences between the two patients groups. Interestingly, the IFN- patients had a significantly higher prednisolone dose compared to the IFN+ patients (p=0.002). Moreover, a trend for higher self-perceived pain among the IFN+ patients was found.

Conclusions Type I IFN activity in patient sera can be found for all subgroups of IIM and does not correlate with disease activity. The ability to induce IFN-regulated gene expression was found to be associated with presence of ANA. This is in line with reports demonstrating nucleic acid containing immune complexes as endogenous inducers of type I IFN. We will further delineate the specificities of the ANA involved. In agreement with studies showing suppression of type I IFNs by glucocorticoids, the type I IFN activity in sera was lower for patients receiving higher dose of cortisone treatment.

Disclosure of Interest L. Ekholm: None Declared, A. Tjärnlund: None Declared, C. Mavragani Grant/Research support from: S.Niarchos Foundation Grant, P. Charles: None Declared, L. Padyukov: None Declared, M. Crow Consultant for: Astra Zeneca,BMS, Idera, Vertex, I. Lundberg Shareholder of: Pfizer, Grant/Research support from: BMS