Anti-phospholipid IgG antibodies detected by line immunoassay differentiate patients with anti-phospholipid syndrome and other autoimmune diseases

PURPOSE: Anti-phospholipid antibodies (aPL) analyzed by line immunoassay (LIA) can recognize beta(2)-glycoprotein I (β(2)GPI) domain 1 (D1) epitopes depending on β(2)GPI binding to distinct phospholipids. The aPL LIA was compared with consensus ELISA to investigate whether both techniques can discriminate anti-phospholipid syndrome (APS) patients from aPL-positive, systemic autoimmune rheumatic diseases (SARD) patients without clinical symptoms of APS and controls. METHODS: Thirty-four APS patients (14 arterial/venous thrombosis, 16 pregnancy morbidity, and 4 both), 41 patients with SARD lacking clinical APS criteria but demonstrating positivity for anti-β(2)GPI (aβ(2)GPI) IgG, and 20 healthy subjects (HS) were tested for aPL to cardiolipin (aCL), phosphatidic acid, phosphatidylcholine, phosphatidylethanolamine, phosphatidylglycerol (aPG), phosphatidylinositol, phosphatidylserine, β(2)GPI, prothrombin, and annexin V by LIA. Samples were also tested for aCL, aβ(2)GPI, aβ(2)GPI-domain 1 (aD1), and aβ(2)GPI-domains 4–5 (aD4–5) by ELISA and for lupus anti-coagulant. RESULTS: Comparison of LIA with ELISA revealed a good agreement for the consensus criteria aPL aβ(2)GPI and aCL IgG (kappa = 0.69, 0.68, respectively) and a moderate agreement for IgM (kappa = 0.52, 0.49, respectively). Regarding ELISA, aD1/aD4–5 demonstrated the best performance of differentiating APS from asymptomatic SARD [area under the curve (AUC): 0.76]. aPG IgG had the best performance by LIA (AUC: 0.72) not significantly different from aD1/aD4–5. There was a good agreement for aPG IgG with aD1/aD4–5 (kappa = 0.71). CONCLUSIONS: aD1/aD4–5 (ELISA) and aPG IgG (LIA) differentiate APS from SARD patients. PG appears to interact with β(2)GPI of APS patients and exposes D1 thereof for disease-specific aPL binding in LIA.