Humanized immune system mice (HIS mice)

Humanized immune system (HIS) model mice continue to evolve as a core platform for the preclinical evaluation of immunotherapy. This article provides an overview of their fundamentals, standard protocols, and the trends in next-generation models, along with key points for their application in drug discovery research.

Why is the reconstruction of the "human immune system" necessary?

The background requiring humanized immune system (HIS) mice is the challenge that conventional animal models cannot sufficiently reproduce human-specific immune responses. In particular, advanced immunotherapies such as immune checkpoint inhibitors, CAR-T cells, and bispecific antibodies rely heavily on the precise interactions between human immune cells.It is difficult to accurately evaluate the efficacy and safety of drugs in the mouse immune system.As a result, HIS mice, which possess human-derived immune cells in their bodies, have become the global standard in drug discovery research as a model capable of reproducing human immune responses *in vivo*. They are also widely utilized as indispensable tools for elucidating human-specific immune mechanisms and verifying novel therapeutic strategies in research on tumor immunity, autoimmune diseases, and infectious diseases.

Standard Review: HIS Model Basics and Standard Protocols

Understanding the humanized immune system mouse (HIS model) is essential, requiring a grasp of basic concepts and standard production protocols. Here, we select a representative and widely referenced review to summarize the key points of the model's fundamental structure and reproducible methods.

[Standard 1] Classification of HIS Models and Establishment of Platforms

• Paper title: Humanized mice in cancer immunotherapy.
• Author/Journal: Shultz LD, et al. / Nature Reviews Cancer

This review by Shultz et al. is,Comprehensive review that greatly contributed to the establishment of humanized immune system mice (HIS model)This document systematically organizes the background of developing immunodeficient mouse strains, particularly those centered on NSG mice, and the characteristics of the HSC (CD34+ hematopoietic stem cell) transplantation model and PBMC transplantation model built upon them. The HSC model, while capable of multilineage differentiation and long-term immune reconstitution, faces the challenge of requiring time for maturation. In contrast, the PBMC model can rapidly reproduce human T cell responses but is limited to short-term evaluation due to the high risk of GVHD. This paper clearly compares the advantages and limitations of each model, providing guidelines for selecting the appropriate platform based on research objectives.

[Standard 2] Validation of predictive ability in preclinical studies

• Humanized mouse models for the study of human immunology and personalized medicine.
• Author(s)/Journal Title: Walsh NC, et al. / Annual Review of Immunology

This review by Walsh et al.A systematic review examining the extent to which humanized mice (HIS models) can predict human responses in preclinical studies.This report organizes the reproducibility of immune responses using HSC and PBMC transplant models, and their applications in tumor immunity and infectious disease models, demonstrating their usefulness in reproducing human-specific immune responses. In particular, it emphasizes the correlation with clinical outcomes observed to a certain extent in the evaluation of immune checkpoint inhibitors and cell therapies. On the other hand, limitations affecting predictive power, such as discrepancies in cytokine environments and the difficulty of complete immune cell maturation, are also pointed out, and directions for model improvement are discussed.

Latest Review: Trends in Next-Generation HIS Models (2024-2026)

Recent research on HIS models has evolved from the traditional "transplantation of human immune cells" towards recreating a physiological environment closer to that of humans. Here, we present a review of the trends for 2024-2026.

[Trend 1] Next-generation (Cytokine-supported) HIS mice

• Key point: Differentiation and maturation of myeloid cells (macrophages, DCs) and NK cells, which were insufficient in conventional models.
• Recommended Literature: Willis E, et al. Toxicologic Pathology (2024): Comparison of NSG-SGM3 and NOG-EXL Mice and Features of Myeloid Expansion
  Stocks H, et al. review on myeloid-engrafting humanized mice (2025): Advances and challenges in cytokine-induced models

Next-generation (cytokine-supported) HIS mice areSignificantly improved the reconstitution of myeloid cells and NK cells, which was insufficient in conventional modelsThey are attracting attention for their ability to reproduce human immune responses. Representative models such as NSG-SGM3 and NOG-EXL have been developed by introducing human cytokines like IL-3, GM-CSF, M-CSF, and SCF. This promotes the differentiation and maturation of human-derived macrophages, dendritic cells, and monocytic cells, thereby improving the reproducibility of the innate immune system. These models enable the analysis of the behavior of immunosuppressive macrophages and myeloid-derived suppressor cells (MDSCs) in the tumor microenvironment (TME), contributing to the elucidation of the mechanisms of action of immunotherapies. On the other hand, NSG-SGM3 has been reported to have issues such as myeloid overactivation and short lifespan due to excessive cytokine expression, making comparative studies with improved strains like NOG-EXL, which exhibit more physiological expression, crucial.

[Trend 2] PDX x Spatial Omics x AI

• Key Points: Overcoming GVHD (Graft-versus-Host Disease) and Long-Term Follow-up
• Recommended Reading: Voabil P, et al. Cell (2021): Foundational research on PDX and immune microenvironment remodeling.
  Bagaev A, et al. Nature Medicine (2021): Spatial immune subtyping of the TME and AI analysis
  Nature Reviews Cancer, Cancer Discovery, etc.

Combining PDX (Patient-Derived Xenograft) and HIS models with integrated spatial omics and AI analysis is the next-generation trend in tumor immunology research.and is attracting attention. In conventional PBMC transplantation systems, GVHD (Graft-versus-Host Disease) developed early, making long-term observation difficult. However, with HSC transplantation and the use of improved immunodeficient strains, it has become possible to maintain a more stable human immune environment. By visualizing the spatial arrangement and interactions of immune cells within the tumor microenvironment (TME) at high resolution using spatial transcriptomics and other methods, and by comprehensively elucidating the mechanisms of immune response and therapeutic resistance through pattern analysis using AI, it has become possible. The ability to track long-term tumor progression and treatment response represents a significant advancement in the application to personalized medicine.

Disease Area Specific Major Review List

The optimal HIS model varies greatly depending on the disease area and the modality being evaluated. Each model has characteristics in terms of the immune functions it can reproduce and the phenomena it can observe, making it important to select the one that is appropriate for the intended purpose. Here, we organize recommended HIS models and evaluation points to focus on in reviews for each major application area.

Strategic Perspective: Improving Seed Exploration Accuracy

To improve the accuracy of seed exploration using the HIS model,Experimental design that assumes variability between donors is essential.Because human immune cells vary significantly in phenotype and function by donor, evaluations that rely on a single donor have limitations in reproducibility and external validity. For this reason, it is important to standardize multi-donor validation to understand the consistency and range of variability in responses. Furthermore, by combining with PDX, it is possible to select leads with higher predictive power closer to the clinic by evaluating the interaction between patient-derived tumors and diverse immune backgrounds.

Summary

We have looked at the necessity of HIS models, from basic classification and standard protocols to the advancement of next-generation models. In addition to understanding the characteristics of HSC and PBMC models, cytokine-driven and PDX-integrated models have greatly improved the reproducibility of human immune responses and the predictive accuracy in preclinical studies. On the other hand, challenges such as inter-donor variability and GVHD remain important, and multi-donor validation and appropriate model selection are essential for improving the accuracy of seed discovery.