HLA variation analysis and autoantigen epitope prediction in idiopathic inflammatory myopathies

OBJECTIVE: To identify HLA associations with clinical and autoantibody-defined subgroups of idiopathic inflammatory myopathy (IIM) and predict potential immunogenic epitopes presented by subgroup-specific HLA alleles. METHODS: An observational case-control study was conducted including 147 biopsy-confirmed IIM patients and 114 ethnically matched healthy controls. Patients were classified into clinical and autoantibody-defined subgroups. High-resolution HLA genotyping was performed, and allele/amino acid frequencies were compared between the groups. Epitope prediction was carried out for key HLA-associated autoantibody subgroups using in silico tools assessing antigenicity, allergenicity, toxicity and molecular docking with HLA alleles. RESULTS: HLA-C*07:01 was protective in IIM (Pa = 1.23 × 10-2, OR = 0.17, 95% CI 0.06-0.41). Subgroup analysis showed associations of HLA-A*33:03 with DM (Pa = 3.97 × 10-3, OR = 14.8, 95% CI 4.38-68.07) and DQB1*06:03 with PM (Pa = 2.40 × 10-2, OR = 7.24, 95% CI 2.42-22.61). The anti-Mi2 subgroup showed strong associations with A*33:03 (Pa = 2.82 × 10-3, OR = 19.73, 95% CI 5.11-98.01), DQA1*02:01 (Pa = 1.03 × 10-2, OR = 10.28, 95% CI 2.33-53.75) and DQB1*03:03 (Pa = 1.98 × 10-2, OR = 8.99, 95% CI 2.71-30.51). Anti-Ro52 was associated with B*08:01 (Pa = 2.94 × 10-3, OR = 11.67, 95% CI 3.56-39.83). Fine mapping pinpointed phenylalanine at position 25 of DQA1 to the risk allele DQA1*02:01 in anti-Mi2a/b autoantibody positive patients. From 49 predicted Mi-2 peptides, 'YYKYILTR', 'QAFSRAHR' and 'DYWEKLLR' bound stably to A*33:03 and were non-allergenic and non-toxic. CONCLUSION: In the anti-Mi2 subgroup, we identified novel HLA alleles, amino acid variants and immunogenic epitopes, supporting a role for HLA-restricted antigen presentation in disease pathogenesis and indicating candidate targets for antigen-specific immunotherapy. However, the HLA/epitope findings related to anti-Mi2 are exploratory due to limited power, and require confirmation in larger cohorts together with experimental validation.