Abstract Kawasaki disease (KD) is a pediatric systemic vasculitis of unknown etiology. Although infections are thought to trigger the disease, the antigen-specific adaptive immune responses remain poorly characterized. This study aimed to comprehensively profile the adaptive immune response in KD patients using omics data to improve diagnosis and to assess how faithfully the Candida albicans water-soluble fraction (CAWS) mouse model recapitulates human KD immunopathology. We performed high-throughput sequencing of B-cell and T-cell receptor repertoires (BCR/TCR) in clinical cohorts comprising KD patients, febrile controls, and healthy children. A CAWS-induced vasculitis mouse model was established for comparative pathological and immunological analysis. We assessed immune repertoire diversity, clonal expansion, V(D)J gene usage, and clonal distribution. A previously developed physics and mathematics based model, was applied to classify immune states. KD patients showed oligoclonal expansion in the Ig-A and TCR-β repertoires, accompanied by significantly reduced diversity, consistent with antigen-driven clonal selection. The model effectively discriminated KD patients from control groups and identified misclassified febrile cases. Although the CAWS model recapitulated human-like coronary vasculitis and immune infiltration, it did not mirror the oligoclonal immune response seen in patients. Instead, the model exhibited polyclonal B-cell activation, increased Ig-G diversity, and predominant Ig-M expansion. Our findings reveal antigen-specific adaptive immune mechanisms in KD and provide an immune repertoire-based framework for diagnostic discrimination using minimal sample input. The marked divergence between human and mouse immune responses highlights the limitations of current animal models and underscores the need for more human-relevant systems to study KD pathogenesis and therapeutic strategies.
Ma et al. (Sun,) studied this question.