Preclinical mouse models of multiple myeloma: a clinically oriented guide to model selection, validation, and translational application

Multiple myeloma (MM) remains incurable, and progress in drug development and understanding its pathogenesis relies on mouse models that replicate the patient’s disease course. Most existing reviews merely catalogue model phenotypes, rarely include post-2020 work, and provide no workable standards for model selection or cross-validation. Here, we summarize the MM mouse models developed or refined over the past five years and compare the advantages and limitations of the four main categories: syngeneic, xenograft, genetically engineered, and humanized. Given the frequency of false-positive preclinical efficacy signals, we propose a “2 + 1” cross-model validation framework in which any candidate therapy should be tested in at least two mechanistically distinct models, one of which must be immunocompetent. We also constructed a four-dimensional Clinical Translation Scoring System (CTSS) that rates each model based on genomic fidelity (25%), immune integrity (25%), bone disease reproducibility (20%), and disease course dynamics (30%). The CTSS is an exploratory tool developed by our team. Its weights can be calibrated to the central question of a given study, although it still requires validation in multicenter studies.