Article Text

PDF
AB1071 Antiphospholipid Syndrome Alliance for Clinical Trials and International Networking (APS Action) Core Laboratory Validation Exercise: Comparison of Chemiluminescence Immunoassay (CIA) and Enzyme-Linked Immunosorbant Assay (ELISA)
  1. M.L. Bertolaccini1,
  2. R. Willis2,
  3. G. Lakos3,
  4. D. Andrade4,
  5. V. Pengo5,
  6. A. Banzato5,
  7. S. Krilis6,
  8. D. Erkan7
  9. on behalf of APS ACTION
  1. 1Lupus Research Unit, King's College London, London, United Kingdom
  2. 2Department of Internal Medicine, University of Texas Medical Branch, Galveston
  3. 3Inova Diagnostics Inc, San Diego, United States
  4. 4Department of Rheumatology, University of São Paulo School of Medicine, Sao Paolo, Brazil
  5. 5Department of Cardiac Thoracic and Vascular Sciences, University of Padua, Padua, Italy
  6. 6Department of Infectious Diseases, Immunology and Sexual Health, St. George Hospital, Sydney, Australia
  7. 7Barbara Volcker Center for Women and Rheumatic Diseases, Hospital For Special Surgery, New York, United States

Abstract

Background APS ACTION International Clinical Database and Repository was created to study the natural course of disease over 10 years in persistently aPL-positive patients with/without other systemic autoimmune diseases. The protocol involves testing for anticardiolipin antibodies (aCL) and anti-β2glycoprotein I antibodies (aβ2GPI) at five different international core laboratories in samples collected at multiple sites. Although all core laboratories use the same ELISA kits, APS ACTION core laboratory directors agreed that an exchange of samples to assess intra- and inter-laboratory variability is crucial to establish whether an acceptable agreement among the core labs can be achieved. As part of the this validation exercise, two different assays were used.

Objectives To assess between-laboratory agreement obtained with a novel antiphospholipid (aPL) chemiluminescence immunoassay (CIA), and compare the results to those obtained with ELISA assays.

Methods Blinded serum samples from aPL positive patients and controls (n=30) were distributed to the five APS ACTION core laboratories. All samples were tested for IgG, IgM and IgA aCL and aβ2GPI with the FDA cleared QUANTA Flash® CIA and QUANTA Lite® ELISA kits (Inova Diagnostics). For every sample, the average, SD and %CV of the values were calculated. Between laboratory precision was considered acceptable if %CV was less than 20%.

Results Categorical agreement between the two methods was 100%, 97% and 77% for IgG, IgM and IgA aCL, respectively, and 87%, 93% and 77% for IgG, IgM and IgA ab2GPI, respectively. The correlation between quantitative results was good (Spearman rho 0.917, 0.809 and 0.893 for IgG, IgM and IgA aCL, respectively; rho 0.938, 0.766 and 0.78 for IgG, IgM and IgA anti-b2GPI. All p values <0.0001). Only CIA met the <20% total CV acceptance criteria for all the samples with all the assays. %CV was higher for ELISA and, in some cases, well outside the acceptable range (Table).

Conclusions IgG and IgM aCL and aβ2GPI results showed good agreement between CIA and ELISA. Agreement was lower for IgA assays due to borderline samples being positive for CIA (a very sensitive test) and negative for the ELISA. Based on the evaluations performed in the APS ACTION Core Laboratories, CIA and ELISA tests displayed substantially equivalent performance for the detection of aCL and aβ2GPI. QUANTA Flash® CIA, a fully automated assay, however, showed higher level of reproducibility, making inter-laboratory comparisons more reliable.

Disclosure of Interest M. L. Bertolaccini: None declared, R. Willis: None declared, G. Lakos Employee of: Inova Diagnostics Inc, D. Andrade: None declared, V. Pengo: None declared, A. Banzato: None declared, S. Krilis: None declared, D. Erkan: None declared

Statistics from Altmetric.com

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.