Clinical laboratory experience with enzyme immunoassays for the detection of auto-antibodies to nuclear antigens

Abstract

Objective: We introduced enzyme immunoassays (EIAs) for detecting auto-antibodies to nuclear antigens in 1999 and discontinued testing for these autoantibodies (including anti-nuclear antibodies, ANA) by indirect fluorescence microscopy. During the past year, we monitored the performance of the new EIA assays to determine their reliability, and to validate the decision to switch to EIA methodology.Methods: Tests for ANA, anti-dsDNA, and anti-DNA antibodies (screening and discrete assays) were performed by EIAs according to the manufacturer's recommended procedures (Helix Diagnostics Inc., West Sacramento, CA). Results were scored as follows: for ANA: <1 U, negative; 1–3 U, weakly positive; 3–6 U positive; and >6 U, strongly positive; and, for anti-dsDNA <25 U, negative; 25–30 U, borderline; and >30 U, positive; and for anti-ENA: <20.0 U, negative; 20–25.9 U, borderline; and >26.0 U, positive. For the calendar year 2000, we monitored the frequencies of negative, weakly positive, positive and strongly positive ANA results for all sera submitted for testing across 3 different lots of kit reagents. We also evaluated a study group of 518 sera from patients seen locally that were positive or borderline positive for the presence of anti-ENA antibodies on the screening assay. For this group of test results, we calculated the frequencies of each discrete positive ENA specificity, and the frequencies of positive anti-dsDNA and ANA results.Results: The ANA test was very stable from lot-to-lot. The month-to-month frequencies for ANA results in each category were as follows: negative, 70+ or −4.1%; weakly positive, 19+ or −1.8%; positive, 6+ or −0.9%; and strongly positive, 4+ or −0.7%. For the 7 months period from January through July, 2000, the cumulative frequency of positive and borderline results for the ENA screen assay for patients seen locally was 11.0%. ANA test results were available in 391 of 518 sera with borderline or positive ENA screen results. The ANA was >1 U in 93% of these cases, and >3.0 U in 76% of cases. In no case did we find a positive anti-dsDNA result in a specimen that had an ENA screen result >20 U and an ANA result <1 U. The vast majority of positive anti-dsDNA results (94%) were observed in sera that had ANA results >6 U. Similarly, the frequencies of positive anti-ENA antibodies varied directly with the level of ANA: 80% of positive ENA results occurred in sera with ANA levels >3.0 U and 63% had levels >6.0 U. The ENA screen assay performed well in identifying sera that one or more discrete ENA antibodies: 87% of sera had at least one positive discrete ENA result and 100% of sera had at least one discrete ENA result >20 U. The frequencies of individual ENA antibody specificities varied widely in sera that were presumptively positive for ENA antibodies on the screening assay. The individual frequencies were as follows: anti-SSA, 59%; anti-SSB, 22%, anti-Sm/RNP, 9%; anti-RNP, 27%; anti-Scl 70, 12%; and anti-Jo 1, 6.3%Conclusions: Each of the EIA assays mentioned above performed well in day-to-day clinical practice. The results indicate that the ANA and ENA screening assays are reliable for detecting the presence of antibodies to dsDNA and discrete ENA antigens, respectively. The finding of a direct correlation between the level of ANA and the presence of anti-dsDNA or ENA antibodies confirms the reliability of the ANA EIA as a screening test for marker auto-antibodies.