Agreement between local and central anti-synthetase antibodies detection: results from the Classification Criteria of Anti-Synthetase Syndrome project biobank

OBJECTIVES

The CLASS (Classification Criteria of Anti-Synthetase Syndrome) project is a large international multicentre study that aims to create the first data-driven anti-synthetase syndrome (ASSD) classification criteria. Identifying anti-aminoacyl tRNA synthetase antibodies (anti-ARS) is crucial for diagnosis, and several commercial immunoassays are now available for this purpose. However, using these assays risks yielding false-positive or false-negative results, potentially leading to misdiagnosis. The established reference standard for detecting anti-ARS is immunoprecipitation (IP), typically employed in research rather than routine autoantibody testing. We gathered samples from participating centers and results from local anti-ARS testing. As an "ad-interim" study within the CLASS project, we aimed to assess how local immunoassays perform in real-world settings compared to our central definition of anti-ARS positivity.

METHODS

We collected 787 serum samples from participating centres for the CLASS project and their local anti-ARS test results. These samples underwent initial central testing using RNA-IP. Following this, the specificity of ARS was reconfirmed centrally through ELISA, line-blot assay (LIA), and, in cases of conflicting results, protein-IP. The sensitivity, specificity, positive likelihood ratio and positive and negative predictive values were evaluated. We also calculated the inter-rater agreement between central and local results using a weighted κ co-efficient.

RESULTS

Our analysis demonstrates that local, real-world detection of anti-Jo1 is reliable with high sensitivity and specificity with a very good level of agreement with our central definition of anti-Jo1 antibody positivity. However, the agreement between local immunoassay and central determination of anti-non-Jo1 antibodies varied, especially among results obtained using local LIA, ELISA and "other" methods.

CONCLUSIONS

Our study evaluates the performance of real-world identification of anti-synthetase antibodies in a large cohort of multi-national patients with ASSD and controls. Our analysis reinforces the reliability of real-world anti-Jo1 detection methods. In contrast, challenges persist for anti-non-Jo1 identification, particularly anti-PL7 and rarer antibodies such as anti-OJ/KS. Clinicians should exercise caution when interpreting anti-synthetase antibodies, especially when commercial immunoassays test positive for non-anti-Jo1 antibodies.

AB0554 CONCORDANCE IN THE CLASSIFICATION CRITERIA IN A PATIENT COHORT WITH IDIOPATHIC INFLAMMATORY MYOPATHIES (IIM)

Background:

Differents classifications criteria for IIM exist. New classification criteria are superior to previous in capturing different subgroups of IIM.

Objectives:

To compare the concordance and disparity of Bohan and Peter and EULAR/ACR 2017 classification criteria at the level of diagnostic certainty, in a group of chilean patients with IIM.

Methods:

40 adults patients with IIM (27 Dermatomyositis (DM) and 13 Polimyositis (PM)), according Bohan and Peter criteria were analized. The patients were reclassified with EULAR/ACR criteria. The level of diagnostic certainty (definitive, probable and possible) was registered for both Classification Criteria. The concordance and disparity between criteria was evaluated. Concordance and disparity analysis were made considering the strict agreement between level of certainty of both criteria, using Cohen´s Kappa coefficient. The analysis was done for the complete cohort and for separated groups.

Patients with discordance belonging to the same subgroup were evaluated using contingency tables. The direction of the change (gain or lost of certainty) and the relation with diagnostic subgroup was also analized. Descriptive statistics is expressed as diagnostic categories, number of patients and rates.

Results:

For the complete cohort and for DM and PM groups the concordance was low. For 27 patients with DM, the observed concordance rate was 63% (16 definitives, 1 probable). The observed disparity rate was 37%. The direction of the change was gain of one level of certainty in 14.5% and two levels in 22.2% of patients applying EULAR/ACR criteria compared to Bohan y Peter criteria. For 13 pacients with PM, the observed concordance rate was 46% (3 definitives and 3 probables). The observed disparity rate was 54%. The direction of the change was loss of certainty. The loss of certainty was one level in 85,7% (one patient change from probable to posible). Only one patiente had gain of certainty of one level (14.3%).

Conclusion:

The strict concordance between both classification criteria was low. The observed concordances were better in patients with DM that PM. The disparities involved gain of level of diagnostic certainty in DM patients, while in PM patients there was mostly lost of level of certainty.

References:

[1]Bohan A, Peter JB. Polymyositis and dermatomyositis (first of two parts). NEJM 1975;292:344-7.

[2]Bohan A, Peter JB. Polymyositis and dermatomyositis (second of two parts). NEJM 1975;292:403-7.

[3]2017 European League Against Rheumatism/American Colllege of Rheumatology Classification Criteria for Adult and Juvenile Idiopathic Inflammatory Myopathies and Their major Subgroups. Arthritis Rheum 2017;0:1-12.

Disclosure of Interests: :

None declared