Anti-dsDNA antibody assay: high specificity and sensitivity with a filtration radioassay in comparison to low specificity with the standard ELISA.

OBJECTIVE To evaluate whether a new fluid-phase filtration radioassay possesses both high sensitivity and specificity compared with the currently used ELISA and Farr assays. METHODS Sequential sera (25 samples) from 9 patients with systemic lupus erythematosus (SLE), sera from 20 patients with SLE possessing anti-dsDNA antibodies by the Crithidia assay, 75 patients with rheumatoid arthritis possessing rheumatoid factors, 50 healthy control subjects, 767 from patients with type 1 diabetes, and a commercial standard serum sample were tested for anti-dsDNA antibodies with the 3 different assays. RESULTS Of serial dilutions of a standard anti-dsDNA antibody sample, only the highest positive sample (50 IU/ml) in the ELISA and the highest 2 positive samples (50 and 25 IU/ml) in the Farr assay were above the normal range. In contrast, all dilutions (to 2.5 IU/ml) of the standard anti-dsDNA antibody sample were above the normal range in the filtration radioassay. Using the values of 50 healthy control subjects in each assay to define the normal range, all 25 sequential sera from 9 patients with SLE were positive. In addition, 20/20 of the SLE individual sera, 2/75 (2.7%) of the RA sera, and 12/767 (1.6%) of the diabetes sera were positive (signal above normal range) in the filtration radioassay. The SLE sera were further examined in 2 additional assays, ELISA and Farr assay, and both assays were less sensitive and specific compared with the filtration radioassay. CONCLUSION The fluid-phase filtration radioassay demonstrated high sensitivity and specificity for the detection of anti-dsDNA antibodies in SLE, with the standard ELISA exhibiting lower specificity. We suggest that testing for anti-dsDNA antibodies can be improved using a fluid-phase filtration radioassay in comparison to commercial assays.