Introduction to Humanized Xenograft Models

Humanized mouse xenograft models represent a major advancement in preclinical immuno-oncology, enabling the study of human tumor–immune system interactions in vivo. Traditional xenograft models, though highly valuable for evaluating tumor growth and therapeutic response, lack a functional human immune system due to the immunodeficient status of the host mice. This absence of adaptive immunity precludes the study of immunotherapies, immune checkpoint blockade, and tumor immunoediting. To address these limitations, researchers have developed humanized mouse models—immunodeficient mice engrafted with functional human immune cells—allowing co-engraftment of human tumors and the investigation of immune-mediated mechanisms in a physiologically relevant context.

Humanized xenograft models are established by introducing components of the human immune system into mice such as NSG (NOD scid gamma) or NOG (NOD/Shi-scid IL2Rγnull) strains, followed by implantation of human tumor tissue or cell lines. These models enable real-time evaluation of human immune responses to cancer, including T-cell infiltration, antigen presentation, cytokine signaling, and immune checkpoint regulation. They are critical tools in the preclinical development of monoclonal antibodies, CAR-T and CAR-NK therapies, bispecific T-cell engagers (BiTEs), and immune modulating biologics.

Human Immune System Reconstitution Methods

Humanized mouse models are created by engrafting immunodeficient mice with human hematopoietic stem cells (HSCs), peripheral blood mononuclear cells (PBMCs), or bone marrow cells. Each engraftment strategy offers distinct immunological profiles, timelines, and experimental outcomes. PBMC-engrafted models produce mature T cells within 2–3 weeks and are ideal for short-term studies of T cell–mediated cytotoxicity and immune checkpoint inhibition. However, they also lead to rapid onset of xenogeneic graft-versus-host disease (GvHD), limiting their use for long-term experiments.

In contrast, HSC-engrafted models support multilineage hematopoiesis and give rise to more stable, long-term reconstitution of adaptive and innate immune compartments, including T cells, B cells, dendritic cells, monocytes, and NK cells. These models are better suited for evaluating chronic immune responses, immunological memory, and tolerability of repeated therapeutic dosing. The development of BLT (bone marrow–liver–thymus) models, which incorporate human thymic tissue, further enhances T cell education and HLA-restricted antigen presentation, increasing the fidelity of the immune response to human tumors.

Co-Engraftment of Human Tumors and Immune Cells

The central feature of humanized xenograft models is their ability to support simultaneous engraftment of human immune and tumor tissues within a single host. This dual-engraftment system creates a chimeric in vivo environment in which immune–tumor interactions can be studied longitudinally and manipulated therapeutically. Human tumor cell lines or patient-derived xenograft (PDX) fragments are implanted subcutaneously or orthotopically into the humanized mouse. Over time, these tumors are infiltrated by human immune cells, enabling functional assays of T cell activation, immune checkpoint expression, tumor-infiltrating lymphocytes (TILs), and immunosuppressive cell populations such as regulatory T cells (Tregs) and myeloid-derived suppressor cells (MDSCs).

This approach has proven especially useful for evaluating the efficacy and mechanism of action of immunotherapies. For example, PD-1/PD-L1 inhibitors can be tested in HSC-humanized mice bearing PD-L1-expressing tumors to observe T cell reactivation, cytokine secretion, and tumor regression. Similarly, bispecific antibodies targeting CD3 and tumor-associated antigens can be evaluated for their capacity to engage T cells and induce tumor cell lysis. In adoptive cell therapy studies, humanized mice allow assessment of CAR-T or CAR-NK cell trafficking, expansion, tumor infiltration, and cytokine release.

Applications in Immuno-Oncology Drug Development

Humanized xenograft models are now indispensable for the preclinical evaluation of immunotherapeutic strategies. They provide a platform to test novel immune checkpoint inhibitors beyond PD-1/PD-L1 and CTLA-4, including emerging targets such as LAG-3, TIM-3, TIGIT, VISTA, and CD47. Additionally, these models allow comparative analysis of combination therapies, such as checkpoint inhibitors with chemotherapy, oncolytic viruses, or targeted agents, under human-relevant immune conditions.

Another major advantage is the ability to model immune-related adverse events (irAEs) and cytokine release syndrome (CRS), which are not observable in traditional immunodeficient xenografts. Humanized mice receiving CAR-T cell therapy, for instance, can develop CRS-like profiles with elevated IL-6 and IFN-γ, allowing early safety assessments. Humanized models also support longitudinal tracking of minimal residual disease, tumor relapse, and immune escape, which are critical considerations in the clinical translation of immunotherapies.

As the field moves toward increasingly personalized and tumor-specific immune interventions, humanized xenograft platforms are being integrated with next-generation PDX models, CRISPR-engineered tumor variants, and immune-repertoire sequencing to create highly defined and mechanistically informative preclinical studies. They are also instrumental in biomarker discovery, including assessment of neoantigen burden, tumor mutational load, and immune gene signatures.

Advancing Humanized Xenograft Research in Oncology

With the growing emphasis on immuno-oncology across pharmaceutical and academic sectors, humanized xenograft models are at the forefront of translational cancer research. These models enable biologically accurate exploration of human immune mechanisms, accelerating the development and de-risking of immune-based therapies prior to clinical trials. Despite their complexity and higher cost relative to traditional xenograft models, the predictive value and mechanistic depth they provide justifies their increasing adoption.

Innovations in humanization protocols, such as knock-in of human cytokine genes and expression of HLA alleles in host mice, are further improving the robustness and reproducibility of these systems. As experimental refinement continues, humanized xenograft models are expected to play a decisive role in immunotherapeutic drug pipelines, from early target validation to late-stage regulatory submissions.