OnCo
ideasIdea

Humanised mice with an immune system matched to the tumour donor

Most cancer drugs are tested in mice with no immune system, then given to people who have one. Mice carrying the same patient's immune cells and tumour would be a fairer test.

Standard PDX models use immunodeficient mice, which cannot evaluate immunotherapy or immune-mediated toxicity. Humanised immune system mice reconstituted with HLA-matched or autologous haematopoietic cells, ideally with human cytokine knock-in backgrounds to support myeloid development, allow tumour and immune system from the same donor. Cost and reproducibility have kept them niche.

Hypothesis
Autologous or HLA-matched humanised models reproduce the individual patient's response and immune-related toxicity to checkpoint blockade better than chance, measured against the patient's actual outcome.
Rationale
Immune context determines the outcome of most modern oncology drugs; syngeneic mouse tumours are immunologically unlike human cancers and have repeatedly over-predicted benefit.
What would test it
Build 40 paired autologous models from patients starting checkpoint therapy and score concordance between model response and patient outcome, pre-registering the concordance threshold.
Maturity
preclinical evidence
Who has to act
research
Cost to try
Medium ($1M to $50M)
Years to first evidence
5
Bottlenecks it attacks

Connected

5top