Immunogenicity of Therapeutic Antibodies: Mechanisms, Prediction, and Mitigation Strategies in the Era of Personalized Biologics

Therapeutic monoclonal antibodies have transformed treatment across cancer, autoimmune disorders, and infectious diseases, yet their clinical utility remains challenged by immunogenicity and the resulting formation of anti-drug antibodies (ADAs), which can alter pharmacokinetics, neutralise therapeutic activity, and cause serious adverse events. This review synthesises current evidence on the mechanisms, prediction frameworks, and mitigation strategies relevant to therapeutic antibody immunogenicity, with particular focus on identifying underexplored risk dimensions not addressed by existing models. We conducted a comprehensive narrative review of peer-reviewed literature covering T-cell-dependent and T-cell-independent ADA formation pathways, multifactorial determinants of immunogenic risk, in silico prediction tools including NetMHCIIpan, SITA, DeepImmuno, and PRIME, in vitro assays including MAPPs and DC-T cell co-culture systems, and engineering and clinical mitigation strategies. Persistent challenges across the field include systematic overprediction, inadequate modelling of conformational B-cell epitopes, HLA diversity, and lack of data standardisation. Critically, molecular mimicry — structural similarity between therapeutic epitopes and pathogen-derived or self-peptides — emerges as a mechanistically distinct and currently invisible risk axis that may explain a subset of inter-patient variability in ADA incidence unaccounted for by existing sequence-based frameworks. Next-generation immunogenicity prediction requires multimodal approaches integrating structural epitope validation, patient HLA profiling, infection history, and machine learning to achieve biologically complete risk stratification.