# Automated combination discovery: patient-sample screens feeding Bayesian platform trials

Source: https://onco.cc/ideas/idea-moon-automated-combination-discovery/  
OnCo record `idea-moon-automated-combination-discovery` (Idea). Data CC BY-NC 4.0, attribute "Data from OnCo (onco.cc)"; commercial use needs a licence.

## TL;DR

There are far more possible drug combinations than can ever be tried in patients. Test thousands on living samples of real tumours, then feed only the winners into adaptive trials.

## Summary

The combination space of approved and investigational cancer drugs is astronomically large; combinations reach the clinic through intuition and commercial convenience. Functional precision medicine (ex vivo drug testing on patient cells, EXALT trial in haematological malignancies) and high-throughput organoid screening make empirical combination discovery on patient material feasible. The proposal is an integrated pipeline: standardised ex vivo combination screens on fresh samples from trial participants, machine learning to prioritise synergistic and resistance-preventing pairs, and a standing Bayesian platform trial that adds and drops combination arms based on pipeline output.

## Fields

- Kind: Idea
- Last checked: 2026-09-08
- Hypothesis: Pipeline-prioritised combinations achieve response rates in the platform at least double those of historically selected combinations in the same settings.
- Rationale: Empirical screening beat rational design in antibiotics and in EXALT; connecting the screen directly to a trial closes the loop that has been missing.
- Proposed test: Run the pipeline in relapsed acute myeloid leukaemia and metastatic colorectal cancer for three years; compare platform response rates with contemporaneous conventional combination trials.
- Maturity: preclinical-evidence
- Actor: research

## Sources

- Bottleneck evidence (Too many combinations to test): Palmer & Sorger, Combination cancer therapy can confer benefit via patient-to-patient variability without drug additivity or synergy (Cell 2017): https://doi.org/10.1016/j.cell.2017.11.009

## Connected records

- ideas: [Patient-derived organoids to pick ADC payloads](https://onco.cc/ideas/idea-organoid-guided-adc/)
- cancers: [Acute myeloid leukaemia](https://onco.cc/cancers/aml/), [Colorectal cancer](https://onco.cc/cancers/colorectal/)
- technologies: [AI-driven drug & target discovery](https://onco.cc/technologies/ai-drug-design/), [Functional (ex vivo) drug testing](https://onco.cc/technologies/functional-drug-testing/), [Patient-derived organoids](https://onco.cc/technologies/organoids/)
- terms: [Basket, umbrella, and platform trials](https://onco.cc/terms/basket-umbrella-platform/)
- bottlenecks: [Lab models that fail to predict what happens in patients](https://onco.cc/bottlenecks/b-preclinical-models/), [Too many combinations to test](https://onco.cc/bottlenecks/b-combination-space/)
- key papers: [Combination Cancer Therapy Can Confer Benefit via Patient-to-Patient Variability without Drug Additivity or Synergy](https://onco.cc/key-papers/paper-palmer-cell/)

---
JSON: https://onco.cc/api/v1/entities/idea-moon-automated-combination-discovery.json