A contemporary analysis of solid-phase antiphospholipid antibody testing: implications for classification or...

Antiphospholipid antibody (aPL) testing is required for diagnosis of antiphospholipid syndrome (APS). Although they have different uses, clinicians sometimes (inappropriately) use APS classification criteria as diagnostic criteria. We evaluated interlaboratory performance of testing for anticardiolipin (aCL) and anti-beta-2-glycoprotein (aβ2GPI) from participants of the Royal College of Pathologists of Australasia Quality Assurance Programs Immunology program over the last 3 years, from information in the final reports. We analysed data for aCL immunoglobulin (Ig)G, aCL IgM and aβ2GPI IgG testing, as representing the main aPL assays. These were assessed according to reagent type and subtyped into enzyme-linked immunosorbent assay (ELISA) versus non-ELISA solid-phase assays. We assessed results for interlaboratory consensus regarding aPL-positive versus aPL-negative samples and quantitative data, including mean, median, coefficient of variation (CV), and standard deviation. Universal consensus was only achieved in approximately one-third (39%) of samples over the reporting period, while 80% participant agreement occurred in 87.5% of tests overall, with results comparable between analytes (22/24=92% IgG aCL; 19/24=79% IgM aCL; 22/24=92% IgG aβ2GPI). CVs were often above 50% for aPL-positive samples for most ELISA assays, but often <20% for non-ELISA methods, with the poorest overall harmonisation for aCL IgM testing. There was an emerging trend for reduced usage of ELISA-based assays, with currently fewer than 25% of participants using these assays. In conclusion, variability persists in the reporting of aCL and aβ2GPI, with limited consensus and an emerging trend to uptake of non-ELISA solid-phase reagents, with reduced CV trends, highlighting the disconnect between APS classification and diagnostic criteria. Further harmonisation of solid-phase testing for real-world diagnostic applications is also required.