Machine learning algorithms as tools for identifying predictive autoantibody biomarkers in pemphigus vulgaris
Need to claim your poster? Find the KiKo table at the conference and they'll help
you get set up.
Presented at: Society for Investigative Dermatology 2025
Date: 2025-05-07 00:00:00
Views: 2
Summary: Abstract Body: While the role of autoantibodies (autoAbs) against desmoglein (Dsg)3 and -1 in Pemphigus vulgaris (PV) is well established, non-Dsg autoAbs are also being increasingly implicated but not yet extensively studied. Using multiplexed protein microarray technology, we examined reactivity to 52 putative PV relevant autoantigens in 633 serum samples (421 patients, 212 controls) stratified by HLA haplotype. Limma analysis revealed increased reactivity to 37 autoantigens, including IgE Fc, HLA-DRA, Human M1, -2, -3, -4, Thyroglobulin, and Dsg3/-1 in active PV (PVA, n=209) vs. controls (CR, n=212). HLA-negative patients (lacking the PV associated DRB1*0402/DQB1*0503 alleles, n=34) showed increased reactivity to 28 antigens compared to HLA-positive (n=169) patients, including annexin A9, Human M2, Thyroglobulin, and Dsg-1. Among HLA-positive patients, DQB1*0503 carriers exhibited heightened reactivity to 28 autoantigens compared to DRB1*0402 carriers, including TPO, Dsg3/-1/-4, Integrin alpha X, and C5a receptor 1. K-Nearest Neighbor (KNN) accurately distinguished PVA from controls (8/10), *0402-positive from *0503-positive (9/10), and PVA from remission patients (10/10). In a longitudinal analysis of 22 patients, autoAb reactivity varied over time across phases of disease activity, yet each patient retained a unique autoantigen profile signature. A core set of antigens with significant reactivity changes included Dsg2, HSP60, Human M2, HLA-DRA, FH, and IgE Fc. These data provide strong support for the presence and potential active role of both Dsg and non Dsg specific autoAbs in PV pathogenesis, and highlight the importance of HLA genetics in shaping autoimmune specificity. This information provides new insights into disease mechanisms and suggests predictive biomarkers for disease classification, clinical course and future individualized therapies. Victoria M. Hoffman<sup>1</sup>, Rebekah R. Schwartz<sup>1, 2</sup>, Kristina Seiffert-Sinha<sup>1</sup>, Animesh A. Sinha<sup>1</sup> 1. Dermatology, University at Buffalo Jacobs School of Medicine and Biomedical Sciences, Buffalo, NY, United States. 2. SUNY Upstate Medical University Hospital, Syracuse, NY, United States. Adaptive and Auto-Immunity