The Limitations of Immunodeficient Models in Preclinical Research

This direction is merged into another direction: immunodeficient model limitations. Its last version stays readable and citable here.

Version 1 of 145 papers48 conceptsfrozen 2026-10-04

For decades, the gold standard for testing new cancer therapies has been the immunodeficient xenograft model. By using mice that lack a functional immune system—such as "nude" or athymic mice—researchers can successfully grow human tumors that would otherwise be rejected. However, a growing body of evidence suggests that this convenience comes at a steep price: a "predictive gap" where therapies that look miraculous in the lab fail to translate into clinical success.

The core of the problem is that modern medicine is increasingly moving toward therapies that don't just kill cells, but modulate the environment around them. When we remove the immune system to facilitate tumor growth, we inadvertently delete the most important "conversation partner" the drug has in the body.

The Missing Crosstalk in the Microenvironment

In a living patient, a tumor is not an island; it is an ecosystem. Many drugs work by disrupting the "crosstalk" between malignant cells and the surrounding stroma. For example, Li Wang and colleagues demonstrated in Bevacizumab potentiates chemotherapeutic effect on T-leukemia/lymphoma cells by direct action on tumor endothelial cells that Bevacizumab does more than starve vessels; it alters adhesion molecules to trigger vessel collapse. Yet, in immunodeficient models, the absence of a complete immune microenvironment means we cannot see how these treatments interact with the host's natural defenses.

Similarly, Yang Du showed that Zoledronic acid prevents mesenchymal stem cells from recruiting pro-tumor macrophages via MCP-1 signaling (Zoledronic acid prevents the tumor-promoting effects of mesenchymal stem cells via MCP-1 dependent recruitment of macrophages). Because immunodeficient models lack T-cells and B-cells, the influence of the adaptive immune system on this "stromal reprogramming" remains an invisible variable.

Why Immunotherapies Struggle to Translate

The predictive gap is most visible in the field of immunotherapy. A systematic review of CAR-T cell therapy clinical trials highlights that the heavy reliance on immunodeficient xenografts contributes to the high failure rate of therapies transitioning to human trials for solid tumors. Without a functional immune environment, these models cannot simulate the immunosuppressive barriers—like "bystander" recruitment or T-cell exhaustion—that occur in humans.

Consider the development of bispecific antibodies like HX009. As Hang Ke notes in HX009, a novel BsAb dual targeting PD1 x CD47, demonstrates potent anti-lymphoma activity in preclinical models, standard xenografts make it difficult to isolate how different immune checkpoints interact in a living organism. When testing "living drugs" like CAR-T or CAR-NK cells, Tetsuro Tachi argues that immunodeficient mice simply cannot predict how these cells will perform in the hostile, immunosuppressive environment of a real human brain tumor (Antitumor effects of intracranial injection of B7-H3-targeted Car-T and Car-Nk cells in a patient-derived glioblastoma xenograft model).

Beyond Oncology: The Systemic Blind Spot

This limitation isn't restricted to cancer. In virology, researchers using isolated human airway cells or organoids to study SARS-CoV-2 find that these models lack the systemic complexity of the full immune system. Lucy Thorne and colleagues point out in Evolution of enhanced innate immune evasion by SARS-CoV-2 that localized innate immune suppression can only be fully understood when we see how it affects the recruitment and coordination of adaptive immune cells—something impossible in a dish.

Even in metabolic and inflammatory research, the model matters. Jane F. Ferguson found that high dietary salt induces hypertension through gut-driven T-cell responses (High dietary salt–induced DC activation underlies microbial dysbiosis-associated hypertension). However, using germ-free or immunodeficient models introduces bias because their immune systems are fundamentally naive, potentially leading to physiological responses that don't represent a real-world host.

Toward Immunocompetent Modeling

To bridge this gap, the field is shifting toward more sophisticated models that prioritize immune context:

  1. Humanized Mice: Models like the one developed by Tingting Huang attempt to reconstitute human immune populations within mice to test immunotherapies for leukemia (A murine model to evaluate immunotherapy effectiveness for human Fanconi anemia-mutated acute myeloid leukemia).
  2. Stochastically Emergent Tumors (SETs): Ruhollah Moussavi-Baygi and colleagues are moving away from "implanted" tumors toward models where tumors emerge naturally in immunocompetent environments, allowing researchers to see how immune surveillance shapes the cancer's evolution (Stochastically Emergent Tumors offer in vivo whole genome interrogation of cancer evolution from non-malignant precursors).
  3. Modular Platforms: Platforms like ProMPt, described by Nicole Pandell, combine specific clinical mutations with an immunocompetent environment to identify vulnerabilities that would be invisible in a standard xenograft (ProMPt: A modular preclinical platform for functional modelling of prostate cancer heterogeneity and therapeutic vulnerabilities).

The intuition is simple: if we want to know how a drug will work in a human, we must test it in an environment that includes the human's most powerful ally and most complex variable—the immune system. Without it, we are only seeing half the picture.

Papers in this version

History & citationVersion 1 · 2026-10-04

Merged into another direction · Permanent link

View frozen version 1

Cite this direction

Latest version (a living page):

@misc{lacuna_the_limitations_of_immunodeficient_model_d8adaf0f,
  title = {The Limitations of Immunodeficient Models in Preclinical Research},
  author = {{Lacuna}},
  howpublished = {Lacuna research direction},
  year = {2026},
  url = {https://lacuna.tiptreesystems.com/direction/the-limitations-of-immunodeficient-models-in-preclinical-research/rdr_d8adaf0fc63b40b8b6538dbba996449b},
  note = {Latest version of a living research direction; cite a version for a fixed text. Accessed 2026-11-19},
}

This exact version (version 1, frozen text):

@misc{lacuna_the_limitations_of_immunodeficient_model_d8adaf0f_v1,
  title = {The Limitations of Immunodeficient Models in Preclinical Research},
  author = {{Lacuna}},
  howpublished = {Lacuna research direction},
  year = {2026},
  url = {https://lacuna.tiptreesystems.com/direction/the-limitations-of-immunodeficient-models-in-preclinical-research/rdr_d8adaf0fc63b40b8b6538dbba996449b/v/1},
  note = {Version 1, frozen 2026-10-04. Accessed 2026-11-19},
}