Solid phase assays versus automated indirect immunofluorescence for detection of antinuclear antibodies

Solid phase assays (SPAs) and automated microscope systems are increasingly used to screen for antinuclear antibodies (ANAs). The goal of this study was to evaluate the performance of three automated ANA screening assays; NOVA Lite HEp-2 using NOVA View® (NV, Inova Diagnostics), an automated indirect immunofluorescence method, EliA™ CTD Screen (Fluorescence Enzyme Immunoassay, FEIA; Thermo Fisher) and QUANTA Flash® CTD Screen Plus (Chemiluminescence immunoassay, CIA; Inova Diagnostics). The assays were performed on 480 diagnostic samples from patients with an ANA-associated rheumatic disease (AARD; systemic lupus erythematosus, primary Sjögren's syndrome, systemic sclerosis, inflammatory myopathy, mixed connective tissue disease) and on 767 samples from diseased and healthy controls. Using cut-offs proposed by the manufacturers, the sensitivity was 95%, 80.5% and 86% for NV, FEIA and CIA, respectively. The corresponding specificity was 61% (NV), 97.5% (FEIA) and 88% (CIA). The sensitivity associated with a specificity of ~95% was 79%, 82% and 78% for NV, FEIA, and CIA, respectively. Receiver operating characteristics (ROC) curve analysis revealed no differences in area under the curve (AUC) between the 3 assays when all diseases were grouped. For Sjögren's syndrome, the AUC was higher for SPAs than for NV, whereas for systemic sclerosis, the AUC was higher for NV than for CIA. For all assays, the likelihood ratio for AARD increased with increasing antibody levels and for double positivity of NV with SPA. In conclusion, the performance of automated SPA and IIF was assay- and disease-dependent. Taking into account antibody levels and combining IIF with SPA adds value.