Detection of Anti-Cardiolipin and Anti-β2glycoprotein I Antibodies Differs between Platforms without Influence on Association with Clinical Symptoms

Walid Chayoua, Hilde Kelchtermans, Gary W. Moore, Jean-Christophe Gris, Jacek Musial, Denis Wahl, Stephane Zuily, Francesca Gianniello, Pierre Fontana, Jasper Remijn, Rolf T. Urbanus, Bas de Laat, Katrien M. J. Devreese*

*Corresponding author for this work

Research output: Contribution to journalArticleAcademicpeer-review


Background The anti-phospholipid syndrome (APS) is characterized by thrombosis and/or pregnancy morbidity with persistent presence of anti-phospholipid antibodies (aPL). Laboratory criteria include aPL detection by coagulation tests for lupus anticoagulant (LAC) or solid phase assays measuring anti-beta 2 glycoprotein I (a beta 2GPI) or anticardiolipin (aCL) immunoglobulin (Ig) G/IgM antibodies. External quality control programs illustrate that commercially available aPL assays produce variable results.

Objective We aimed to investigate the agreement and diagnostic accuracy of solid phase assays.

Materials and Methods In thismulti-centre study, 1,168 patient samples were tested on one site for aCL and a beta 2GPI IgG/IgM antibodies by four solid phase test systems. Samples included APS patients, controls and monoclonal antibodies (MoAB) against different epitopes of beta 2GPI. LAC was determined by the local centre.

Results aCL IgM assays resulted in the most discrepancies (60%), while aCL IgG and a beta 2GPI IgM assays resulted in lower discrepancies (36%), suggesting better agreement. Discrepant samples displayed lower median aPL titers. Dependent on the solid phase test system, odds ratios (ORs) for thrombosis and pregnancy morbidity ranged from 1.98 to 2.56 and 3.42 to 4.78, respectively. Three platforms showed lower sensitivity for MoAB directed against the glycine (Gly) 40-arginine (Arg) 43 epitope of domain I of beta 2GPI.

Conclusion Poor agreement was observed between different commercially available aCL and a beta 2GPI IgG/IgM assays, hampering uniformity in the identification of aPL-positive patients. Clinical association was globally concordant between solid phase test systems considering results of the four aPL together. An assay sensitive in detecting the MoAB against Gly40-Arg43 of domain I of beta 2GPI reached the highest OR for thrombosis.

Original languageEnglish
Pages (from-to)797-806
Number of pages10
JournalThrombosis and Haemostasis
Issue number5
Publication statusPublished - May 2019


  • solid phase assays
  • anti-cardiolipin
  • beta 2 glycoprotein I
  • epitope
  • thrombosis

Cite this