Autoantibodies to double-stranded DNA-Intermethod comparison between four commercial immunoassays and a research biosensor-based device

Patients with systemic lupus erythematosus (SLE) often develop a wide variety of serological manifestations including the presence of antibodies to double-stranded DNA (anti-dsDNA). Positivity for anti-dsDNA constitutes one of the laboratory criteria for the diagnosis of SLE and is therefore clinically relevant. We analyzed the diagnostic accuracies of four commercial anti-dsDNA immunoassays and compared the results with a recently established surface plasmon resonance (SPR) biosensor chip with covalently chip-immobilized dsDNA. The anti-dsDNA measurements were performed retrospectively in 50 patients with clinically proven SLE, 39 patients with other autoimmunopathies and 20 healthy controls. Data were evaluated by Receiver-Operator Characteristic (ROC) analysis, with special regard to SLE patients suffering from lupus nephritis. The ROC analyses for the four immunoassays and the SPR biosensor resulted in the following area-under-the-curve (AUC) and diagnostic efficiency (DE) values in descending order: Bindazyme AUC, 0.89; DE, 0.88; ELiA AUC, 0.89; DE, 0.86; SPR biosensor AUC, 0.82; DE, 0.80; Farrzyme AUC, 0.77; DE, 0.77; Farr AUC, 0.77; DE, 0.70. When considering the 22 nephritis SLE patients the following AUC were observed: Bindazyme 0.98; EliA 0.95; SPR biosensor 0.93; Farr 0.89; Farrzyme 0.88. Although various methodologies for the determination of anti-dsDNA were compared, the overall diagnostic accuracy was found satisfactory in all immunoassays. Best data were found for the Bindazyme assay. We referenced the measurements to our in-house SPR biosensor device which showed good AUC and DE values. When optimized, this technique, allowing to monitor antigen/antibody interactions in real-time, may add a new analytical quality to the existing methods, potentially beneficial in diagnosis and clinical monitoring of SLE. Lupus (2010) 19, 957-964.