Introduction: Testing for autoantibodies to extractable nuclear antigens (ENA) is essential in the investigation of connective tissue disease. Counterimmunoelectrophoresis is an early described testing methodology for antibodies to ENAs, but is labour-intensive, only moderately sensitive, and reliant on high-quality reference sera. Enzyme-linked immunosorbent assay (ELISA) is automatable for relatively high sample throughput, but has issues with false positives. The addressable laser bead immunoassay (ALBIA) is a multiplex technology which can assess several antibody specificities simultaneously on a small serum sample. We report performance of an ALBIA system compared with CIE and ELISA.
Methods: Samples from 100 systemic sclerosis patients attending Royal Free Hospital in 2007 and 99 SLE patients attending St Thomas's Hospital in 2007-2008 were studied. All samples were tested for antibodies to RNP, Sm, Ro, La, Scl-70, Jo-1 by in-house CIE, FIDIS™ ALBIA (BMD, France), and ELISAs (Phadia, Germany). Cohen's kappa coefficient was used to examine agreement of the different assay methods for the same antibody. McNemar's test was used to detect differences between methodologies.
Results: One sample was positive for anti-Jo-1 by CIE, & confirmed by ALBIA & ELISA. All 198 remaining samples were anti-Jo-1 negative by all 3 methods. With respect to RNP, Ro, La, Scl-70 antibodies, there was good agreement in assay performance between CIE, ALBIA, and ELISA. For Sm, agreement was less good between CIE and ELISA (kappa 0.491), and ALBIA and ELISA (kappa 0.403). Using McNemar's test performance was no different between the 3 assays, with the following exceptions: between CIE and ELISA for Ro-60 (p<0.01) and RNP (p<0.05), and between ALBIA and ELISA for RNP (p<0.01).
Conclusions: The FIDIS™ ALBIA produced similar level of performance as CIE, but with advantages of automation, and less dependence on highly skilled operators. ALBIA represents a potential advancement applicable to routine Immunology diagnostics.
Copyright © 2011. Published by Elsevier B.V.