ideasIdea
Run the mouse or organoid trial at the same time as the human trial
Instead of testing a drug in lab models first and hoping the results carry over, build the same models from trial participants and run both experiments in parallel to see how well the models predict.
Co-clinical trials generate patient-derived organoids or xenografts from enrolled participants and treat them with the trial regimen, with model results locked and blinded until the clinical readout. This produces a prospective, unbiased estimate of model predictive value, which the field currently lacks, and gives a resource for studying the responders and non-responders.
Hypothesis
Blinded, prospectively locked avatar predictions achieve a positive predictive value above 70 percent for clinical response in at least one modality, establishing which model system deserves decision-making weight.
Rationale
Retrospective concordance figures for organoids are encouraging but subject to selection and reporting bias. Only prospective locking measures true predictive value, as was done for imaging biomarkers.
What would test it
Attach an avatar sub-study to three ongoing phase 2 trials in different tumour types with pre-registered analysis; report predictive values regardless of direction.
Maturity
early clinical
Who has to act
research
Cost to try
Medium ($1M to $50M)
Years to first evidence
4
Bottlenecks it attacks
- Preclinical models that do not predict people · Nine in ten cancer drugs that work in mice fail in humans. Our models are the reason.
- Biomarkers are not validated or standardised · Tests that decide who gets a drug are often not validated prospectively and are measured differently in every lab.