The adaptive immune receptor repertoire(AIRR), comprises T-cell receptors(TCR) and B-cell receptors(BCR) that form the molecular foundation of antigen-specific immunity and immunological memory. Autoimmune diseases arise from the breakdown of immune tolerance, leading to expansion and persistence of autoreactive lymphocyte clones targeting self-antigens. This review aims to systematically summarize dynamic changes in TCR and BCR repertoires across major autoimmune diseases, explore their diagnostic, prognostic, and therapeutic implications, and identify current limitations hindering clinical translation of AIRR profiling.

This review synthesizes findings from high-throughput AIRR sequencing studies in rheumatoid arthritis(RA), systemic lupus erythematosus(SLE), Sjögren’s disease(SjD), type 1 diabetes(T1D), and ankylosing spondylitis(AS). The methodological framework encompasses the full analytical pipeline. Sample types include genomic DNA for unbiased repertoire assessment and complementary DNA for functional expression profiling. Unique molecular identifiers correct PCR amplification bias. Library preparation strategies include 5’ RACE, preferred for BCRs due to somatic hypermutation; multiplex PCR for higher sensitivity with low-input samples; and targeted enrichment. Sequencing platforms range from short-read technologies like Illumina to long-read platforms including PacBio HiFi and Oxford Nanopore, which enable full-length VDJ assembly. Bioinformatic analysis employs standardized pipelines such as MiXCR for germline gene assignment, CDR3 identification, and mutation quantification.

Beyond traditional bioinformatics, machine learning has emerged as a transformative approach for extracting high-order functional signals from AIRR-seq data. Deep learning architectures, including convolutional neural networks and transformer-based models, automatically learn hierarchical features from large-scale immune repertoire data without manual feature engineering. These models have been successfully applied to identify abnormally expanded TCR and BCR clones in autoimmune diseases by learning discriminative patterns from sequence composition, VDJ usage, and CDR3 physicochemical properties. For antigen specificity prediction, attention-based models capture contextual relationships within CDR3 sequences to predict TCR–peptide–MHC binding. In antibody discovery, generative models are employed to design novel antibody sequences with optimized binding properties. For disease diagnosis, supervised learning approaches trained on repertoire-wide features achieve accurate classification of autoimmune disease states versus healthy controls, and some models can stratify patients by disease activity or treatment response. Integration of repertoire data with single-cell transcriptomics further enhances predictive performance by enabling joint modeling of clonal expansion and functional phenotypes.

Disease-associated repertoire remodeling is characterized by oligoclonal expansion of autoreactive T and B cell clones, skewed VDJ usage, reduced diversity, and enrichment of public clonotypes. Disease-specific features are evident for both compartments. In RA, TCR clonal conservation occurs across joints with CD8-positive tissue-resident T cells in synovium, while BCR analysis reveals expanded IGHV4-39 usage with somatic hypermutation in anti-citrullinated protein antibody-positive patients. In SLE, TCR repertoires exhibit oligoclonal expansion during flares with cytotoxic differentiation, whereas BCR repertoires show early expansion of autoreactive IGHV4-34 clones. In SjD, TCR analysis demonstrates clonal expansion of CD4-positive T cells targeting salivary gland autoantigens MAP3K4 and DDIAS, with increased similarity among expanded CDR3 sequences in patients carrying the HLA-DR3 risk haplotype, and BCR analysis demonstrates restricted IGHV4-34 usage in plasmablasts. In T1D, insulin-reactive TCR beta chain clones persist throughout disease progression, while BCR analysis reveals polyreactive insulin-binding B cells with biased light chain usage. In AS, the canonical TRBV9-J2S3 TCR rearrangement recognize HLA-B27-restricted epitopes, and BCR analysis demonstrates expanded double-negative B cell populations with IgA isotype skewing.

Spatiotemporal heterogeneity is evident for both compartments. Substantial divergence exists between peripheral blood and affected tissues, with inflamed target organs harboring more focused, clonally expanded T and B cell populations. Repertoire dynamics evolve across disease stages, from early clonal expansion to diversity contraction during flares and partial restoration in remission.

Translational applications are increasingly recognized. Disease-associated clonotypes serve as biomarkers for early diagnosis, patient stratification, and treatment monitoring. In RA, high TRBV6-6 usage predicts methotrexate non-responsiveness. In SLE, contraction of the IGHV3 gene family correlates with glucocorticoid sensitivity. In SjD, machine learning models using TCR clonotype panels achieve high diagnostic accuracy. Therapeutically, precise targeting of autoreactive clones is emerging through engineered TCR-Treg and CAR-Treg cells for antigen-specific immune regulation, peptide-based strategies that competitively modulate MHC class II antigen presentation by reducing peptide-binding，anti-idiotype antibodies selectively eliminating IGHV4-34 clones, and humanized anti-TRBV9 therapy demonstrating clinical efficacy in AS.

AIRR sequencing has transformed our understanding of autoimmune diseases by revealing autoreactive clone dynamics. Characteristic alterations in TCR and BCR repertoires offer insights into disease pathogenesis, serve as promising biomarkers, and open new avenues for precision immunotherapies. Future efforts must prioritize standardized protocols following MIARR guidelines, multi-center collaborative cohorts, integration of single-cell multi-omics, and prospective validation studies. Addressing these priorities will be essential to realize the full potential of AIRR as a diagnostic, prognostic, and therapeutic tool in autoimmune diseases.