{
  "abstract": "Objectives To investigate the antigen specificity and clinical significance of IgA autoantibody subclasses (IgA1 and IgA2) in antiphospholipid syndrome (APS) patients.Methods We measured IgA1 and IgA2 antibodies against β2-glycoprotein I (aβ2GPI) and its domain I (aDI) using in-house ELISAs on stored sera from APS patients and healthy controls (HC). Positivity was determined based on a cut-off calculated as the mean +3 standard deviations (+3SD) of HC values. Fisher’s exact test was applied to identify clinical associations without implying causality.We then constructed a prediction model to differentiate pregnancy-related complications (including fetal loss and pregnancy-associated thrombosis) from non-pregnancy-related thrombotic events. Candidate predictors included criteria for APS antibodies (IgG anti-β2GPI, IgG anti-cardiolipin and lupus anticoagulant) alongside clinical factors (age, ethnicity, smoking status, hypertension, hyperlipidemia, and prior thrombosis). Features were screened using a Random Forest algorithm, followed by LASSO-penalised multivariable logistic regression to estimate the adjusted odds ratios (ORs).Results We analysed 42 APS patients and 7 HC. IgA subclass binding displayed distinct autoantigen preferences: a significantly higher proportion of patients tested positive for IgA2 aβ2GPI antibodies (83.3%) compared with IgA1 aβ2GPI (33.3%, McNemar’s exact test p=0.005) and IgA1 aDI (28.6%, p=0.0002)( figure 1 A-B). IgA2 aDI antibodies showed no significant enrichment. Three variables emerged as important in distinguishing pregnancy-related complications from thrombosis: smoking, IgA1 aDI, and hypertension (figure 1C). In the final LASSO-penalised multivariable model (AUROC 0.84, 95% CI:0.73-0.91) (figure 1D), IgA1 aDI was the only positive predictor of pregnancy-related complications (OR 5.48, p=0.0228). Traditional thrombosis risk factors were negatively associated with pregnancy complications (i.e., associated with the thrombosis group): smoking (OR 0.20, p=0.0221) and hypertension (OR 0.31, p=0.0377).Abstract PO:01:013 Figure 1(A-D): IgA subclass-specific antigen recognition and its associations with pregnancy-related complications in APS. IgA1 and IgA2 levels targeting β2GPI (fig A.) and DI (fig B) were measured (OD at 450nm).Random Forest (RF) model (fig C.) showing variable importance scores for key predictors of pregnancy-related complications. Forest Plot of odds ratios (ORs) and 95% confidenceintervals (95%Cls) (fig D.) for the top predictors distinguishing pregnancy loss from thrombosis were identified in the RF model, as shown. The numbers of patients for each variable are displayed. Positivity thresholds were set at mean +3SD of HC; error bars represent standard deviation; percentage positivity is given in green.Conclusions This study suggests that the IgA2 subclass in APS preferentially binds to β2GPI. IgA1 aDI emerged as a significant predictor of pregnancy-related complications, highlighting its potential as a biomarker for risk stratification. These findings advance our understanding of APS pathogenesis and underscore the value of IgA subclasses in enhancing diagnostic accuracy, risk assessment, and developing targeted therapeutic strategies.",
  "authors": [
    {
      "affiliations": [
        "University College London and Department of Aging, Rheumatology and Regenerative Medicine, London, UK"
      ],
      "name": "Emome Thingujam"
    },
    {
      "affiliations": [
        "University College London and Department of Aging, Rheumatology and Regenerative Medicine, London, UK"
      ],
      "name": "Dan Mccluskey"
    },
    {
      "affiliations": [
        "University College London and Department of Aging, Rheumatology and Regenerative Medicine, London, UK"
      ],
      "name": "Thomas Mcdonnell"
    },
    {
      "affiliations": [
        "University College London and Department of Aging, Rheumatology and Regenerative Medicine, London, UK"
      ],
      "name": "Muhammad RA Shipa"
    },
    {
      "affiliations": [
        "University College London and Department of Aging, Rheumatology and Regenerative Medicine, London, UK"
      ],
      "name": "Michael R Ehrenstein"
    }
  ],
  "title": "PO:01:013 Subclass specific IgA-autoantibody profiles and their association with pregnancy morbidity in antiphospholipid syndrome",
  "uid": "bc1fae29-7a2c-5dc5-8739-a4f607a87eb1"
}