B-073 The Clinical Utility of Reporting AC-2 Pattern With Anti-DFS70 Antibody

Abstract

Abstract Background The dense fine speckled (DFS) pattern (AC-2) is often observed in antinuclear antibody (ANA) indirect immunofluorescence (IIF) assay, and the related anti-DFS70 antibody is known to be negatively associated with a systemic autoimmune rheumatic disease (SARD). Identifying the AC-2 pattern is challenging, as it could be generally confused with the AC-1, AC-4, AC-5, or AC-29 patterns displaying stained metaphase plates or speckled nuclei. Since the anti-DFS70 antibody is not included in routine autoantibody tests, the clinical utility of the AC-2 pattern and anti-DFS70 antibody is not adequately applied in the laboratory and clinical practice. We evaluated the prevalence of anti-DFS70 and other autoantibodies in samples presenting AC-2 pattern in ANA IIF assay. Additionally, we analyzed the association of anti-DFS70 antibodies and clinical features in patients with AC-2 patterns. Method We retrospectively reviewed the ANA IIF images of 1839 samples made by an automated ANA analyzer classified as AC-1, AC-2, AC-4, AC-5, and AC-29 patterns. AC-1, AC-2, AC-4, AC-5, and AC-29 patterns comprised 60.5% of all ANA positives (1839/3039). Among them, the AC-2 pattern was identified in 18.5% (341/1839) by two experts, and only 197 with residual samples were available for further evaluation. Those 197 samples were tested for anti-DFS70 antibodies with enzyme-linked immunosorbent assay (ELISA) and line immunoblot assay (LIA). They were also tested for 22 other autoantibodies (dsDNA, nucleosomes, histones, SS-A/Ro-52, SS-B, RNP/Sm, Sm, Mi-2α, Mi-2β, Ku, CENP-A, CENP-B, Sp100, PML, Scl-70, PM/Scl100, PM/Scl75, RP11, RP155, gp210, PCNA) with LIA. Clinical information from 197 patients was obtained from a retrospective review of electronic medical records. Results Anti-DFS70 antibody prevalence by ELISA and LIA was 32.0% (63/197) and 19.8% (39/197), respectively. The agreement percentage between the two assays was 81.7%, with a kappa value of 0.53 (95% confidence interval: 0.40–0.66). Of the 69 DFS70+ samples (as per ELISA or LIA), the coexistence of other autoantibodies was found in 81.2% (56/69). Of the 128 DFS70- samples, 90.6% (116/128) revealed other autoantibodies. The three most frequent autoantibodies (except anti-DFS70) in 197 patients with AC-2 patterns were anti-SS-A/Ro-52 (30.5%, 60/197), anti-nucleosomes (27.4%, 54/197), and anti-histones (20.3%, 40/197). There was no significant difference in the prevalence of the other 22 autoantibodies between DFS70+ and DFS70− patients. The anti-DFS70 prevalence in the SARD group was 33.3% (50/150), and in the non-SARD was 40.4% (19/47) without significant difference. However, of the 69 DFS70+ patients, the isolated anti-DFS70 was found more frequently in the non-SARD group (36.8%, 7/19) than in the SARD group (12.0%, 6/50) (P = 0.0346). Conclusion The anti-DFS70 positive rate in AC-2 pattern samples is lower than a third, and the prevalence is affected by different detection assays. Along with the AC-2 pattern, just reporting the presence of anti-DFS70 is not valuable for SARD exclusion. Therefore, it is important to co-examine other autoantibodies, such as anti-SS-A, anti-nucleosomes, and anti-histones, to confirm the isolated anti-DFS70.