Background: The differential diagnosis between idiopathic non-specific interstitial pneumonia (INSIP) and idiopathic pulmonary fibrosis (IPF) is often difficult and requires invasive surgical lung biopsy, when high resolution computed tomographic appearance of the patient's lung does not show the typical usual interstitial pneumonia (UIP)-pattern. Because involvement of autoimmunity has been suggested in the pathogenesis of INSIP, it is expected that some unknown autoantibodies may develop as non-invasive diagnostic markers.

Objectives: This study was conducted to determine INSIP-specific autoantibodies and measure them among various lung diseases including bacterial pneumonia, pulmonary tuberculosis, and connective tissue diseases with interstitial lung diseases (ILD).

Methods: Autoantibodies in the sera obtained from patients of INSIP, IPF, autoimmune pulmonary alveolar proteinosis (aPAP), sarcoidosis, and healthy controls were comprehensively detected by using ProtoArray® (Invitrogen) and specified by the scoring method reported previously¹. Immunohistochemistry was performed on the specimens of surgical lung biopsy from INSIP and IPF patients. The normal parts of surgically resected lung tissue due to lung cancer was used as control. We developed direct Enzyme-linked immunosorbent assay (ELISA) and flow cytometric analysis using anti-myxovirus resistance protein-1 (Mx1) overexpressing cells² to measure the serum titers of anti-Mx1 autoantibodies among various lung diseases.

Results: We identified a group of autoantibodies specifically associated with INSIP. Among them we focused on anti-Mx1 antibody. Mx1 is one of the main interferon-inducible antiviral effectors. Immunoprecipitation using a set of INSIP and IPF sera proved the existence of anti-Mx1 antibodies only in INSIP patients. Immunohistochemistry showed Mx1 was expressed in Clara cells, type II alveolar epithelial cells, and alveolar macrophages in control lung tissue. In both INSIP and IPF, the expression of Mx1 increased and localized to hyperplastic type II alveolar epithelial cells and alveolar marcophages aggregated in the air space. Both ELISA and flow cytometric analysis showed anti-Mx1 antibody differentiated INSIP from IPF with high specificity and moderate sensitivity. Among connective tissue diseases, anti-Mx1 antibody was only elevated among patients who were complicated with ILD, suggesting the close relationship of this autoantibody to ILD.

Conclusions: These data support the value of anti-Mx1 antibody as a diagnostic biomarker for INSIP in the setting of IIPs. The involvement of this autoantibody in the pathogenesis of ILD must be studied in the future.

Disclosure of Interest: None declared

DOI: 10.1136/annrheumdis-2014-eular.2650