Discrepancies between two immunoassays for the determination of MPO and PR3 autoantibodies

Abstract

BackgroundTesting for autoantibodies to myeloperoxidase (MPO) and proteinase 3 (PR3) is part of anti-neutrophil cytoplasmic antibodies (ANCA) test that aids the diagnosis of a number of autoimmune diseases including small-vessel vasculitis. We characterized the differences between two automated immunoassays at three facilities for measuring MPO- and PR3-ANCA autoantibodies. Methods117 serum samples were analyzed for MPO and PR3 autoantibodies. The INOVA QUANTA Lite® IgG assay (INOVA Diagnostics) were performed at two facilities and the Bio-Plex® 2200 Vasculitis Panel (Bio-Rad) were performed at a third reference lab. The results were compared both qualitatively (between INOVA QUANTA Lite® and Bio-Plex® methods) and quantitatively (between two sites performing INOVA QUANTA Lite® assays). ResultsComparison of the INOVA QUNATA Lite® assays at two different facilities (n=36) demonstrated high concordance (97.2% for MPO and 94.4% for PR3) and quantitative correlation (R2=0.973 for MPO and R2=0.935 for PR3). Conversely, INOVA QUNATA Lite® and Bio-Plex® methods showed poor concordance at 70.4% for MPO (n=81; 95% CI: 59.7% to 79.2%) and at 76.5% for PR3 (n=81; 95% CI: 66.2% to 84.4%). ConclusionThis study demonstrated low concordance between two methods for MPO-ANCA and PR3-ANCA measurements. Given the discrepancies, the performance of different autoantibody immunoassay methods should be taken into consideration when evaluating MPO-ANCA and PR3-ANCA results.