Most drug candidates fail in clinical trials not because the science is wrong, but because the models used to predict patient response are outdated and treat every patient as the same.
CellStories Bio links organoid biology, AI, and patient data into a single model — starting with women's health, one of the least funded, least studied areas in medicine.
90% of drugs still fail in human trials. The bottleneck has shifted from chemistry to biology — the disease models used to test efficacy haven't changed in decades.
Wong et al. 2019 · DiMasi 2016AI has made drug design dramatically better — 80–90% of AI-designed drugs pass Phase I. But most AI is trained on generic cell data, not real patients — which is why only 40% survive Phase II. Better AI needs better data.
ScienceDirect 2024 · Nature Biotechnology 2025The EU's December 2024 roadmap and the FDA Modernization Act 2.0 are pushing pharma toward human-relevant models. Patient-derived organoids are the credible, regulator-recognized alternative.
EU Commission Dec 2024 · FDA Mod. Act 2.0Regulators in both the US and EU are moving pharma away from animal testing and toward human-relevant, patient-derived models for predicting drug efficacy.
Summary of the EU Commission's roadmap to phase out animal testing (December 2024) and the FDA Modernization Act 2.0 See our sources →We link organoid phenotypes to clinical metadata and patient-reported outcomes — identifying which patient subgroups are likely to respond to a candidate therapy before it reaches a trial.
We built our AI on organoid biology and real patient data — not just molecules — so it can finally predict how an actual patient will respond, not just how a generic cell reacts in a dish.
Conditions that affect half the population remain some of the least funded, least studied areas in medicine. Until 1993, women were routinely excluded from clinical trials in the US altogether.
Fibroids, endometriosis, and other reproductive conditions affect millions — yet the research base behind them is a fraction of what conditions of comparable prevalence receive elsewhere in medicine.
If you invest in precision medicine, drug discovery infrastructure, or trial de-risking — we'd like to talk.
Founding scientific and clinical partnerships are also open to a small number of pharma collaborators.