Background/Purpose: Although immunoprecipitation (IP) remains the gold standard for detecting myositis autoantibodies, it is technically demanding and not widely available. Line immunoassay (LIA) has become a primary diagnostic tool in many centers but is prone to false positives and negatives, which may lead to inappropriate treatment—particularly in anti-MDA5 patients at risk of rapidly progressive interstitial lung disease (RPILD). Commercial anti-MDA5 ELISA tests have shown comparable performance to IP and may offer prognostic value, especially with additional sample dilution. We aim to validate the performance of LIA for detecting anti-MDA5 using commercial ELISA as reference and explore the clinical utility of ELISA in a Taiwanese medical center.

Methods: Our center uses the Euroimmun 16-antigen LIA panel for myositis antibodies. Samples with anti-MDA5 positivity on LIA and those with clinical features suggestive of anti-MDA5 disease (e.g., RPILD, cutaneous findings) but LIA-negative were further tested using a commercial anti-MDA5 ELISA (MBL, Japan). ELISA was performed at 1:101 dilution per the manufacturer’s protocol and at 1:5050 per Gono et al. Two rheumatologists confirmed the final diagnosis based on clinical course.

Results: We analyzed 40 samples from 33 patients. LIA had 4 false positives and 2 false negatives, yielding a sensitivity of 0.93 and specificity of 0.64. Three of four false positives showed moderate or strong LIA signals. Restricting interpretation to moderate/strong positives improved specificity (0.73) but reduced sensitivity (0.62). Both false negatives occurred in RPILD cases: one patient survived after early immunotherapy guided by timely ELISA, while the other died before ELISA results were available. Among the four false positives, two had mechanic’s hands but lacked typical anti-MDA5 features, and the other two had ILD due to other causes. Among ELISA-positive samples, 45% (13/29) exceeded the 150 IU upper limit at 1:101 dilution, masking antibody titer differences. Dilution to 1:5050 revealed clearer stratification (Figure 1B), with higher titers more strongly associated with poor outcomes (median 1515 vs. 882, p=0.059). In comparison, neither the 1:101 dilution (164 vs. 145 IU, p=0.27) nor LIA signal (64 vs. 57.5, p=0.59) significantly predicted outcome (Figure 2).

Conclusion: Commercial ELISA is a valuable confirmatory tool for anti-MDA5 diagnosis, especially where IP is unavailable and LIA is the standard. ELISA reduces misclassification by LIA and improves treatment guidance. High antibody titers often exceed the ELISA kit’s standard detection range, and additional dilution (1:5050) enhances outcome prediction and better identifies high-risk patients.

(A) The correlation between LIA signal intensity and ELISA (standard dilution 1:101).

(B) The correlation ELISA with standard dilution (1:101) and dilution of 1:5050.

Abbreviations: LIA: Line blotting assay, FN: false negative, FP: false positive

Figure 2: Association of antibody signal intensity (LIA)/ELISA titers and prognosis of patients (morality vs alive).

« Back to ACR Convergence 2025

ACR Meeting Abstracts - https://acrabstracts.org/abstract/validation-and-clincial-use-of-anti-mda5-test-lia-versus-elisa/