# Umbrella trial

Source: https://onco.cc/terms/umbrella-trial/  
OnCo record `umbrella-trial` (Term). Data CC BY-NC 4.0, attribute "Data from OnCo (onco.cc)"; commercial use needs a licence.

## TL;DR

An umbrella trial takes one cancer, tests every patient's tumour for a panel of markers, and routes each patient to the sub-study whose drug matches their marker, so several targeted drugs are tested at once under one roof.

## Summary

An umbrella trial is the mirror image of a basket trial. It is anchored in one disease, screens every patient's tumour with a common genomic panel, and assigns each patient to a sub-study defined by their alteration, each sub-study testing a matched drug and often carrying its own control. The shared screening infrastructure is the point: a patient is tested once and considered for many arms, sub-studies can open and close independently, and a drug that fails in its sub-study does not sink the rest of the trial. Umbrellas are most useful where a cancer splits into many small molecular subgroups, each too rare for a stand-alone trial, and where turnaround time matters.

myeloMATCH is the worked example in the corpus. Launched in 2024 across roughly 200 US sites, it sequences every newly diagnosed acute myeloid leukaemia or myelodysplastic syndrome centrally, returns results within about 72 hours, and routes the patient into tiered, biomarker-defined sub-studies (FLT3, TP53, NPM1, older-unfit cohorts) that run from induction through consolidation and maintenance, with measurable residual disease results moving patients between tiers. It is the infrastructure model for precision leukaemia trials. Lung-MAP in squamous lung cancer and the National Lung Matrix Trial in the United Kingdom are the older examples usually cited, and the platform design adds the ability to keep adding arms over years.

Umbrella trials inherit the problems of their parts. Screening a thousand patients to fill sub-studies of thirty each is slow and expensive, and many patients screen into no arm at all. Sub-studies are small and often single-arm, so they produce signals rather than proof. And the umbrella can lock in a set of biomarkers that the field moves past, which is why modern master protocols are written so that arms and markers can be amended without a new trial.

## Fields

- Kind: Term
- Last checked: 2026-09-17
- Also known as: umbrella trial; umbrella trials; umbrella study; umbrella design; umbrella protocol; one cancer many drugs; biomarker-matched sub-studies; sub-study; sub-studies; substudy; substudies; screening protocol; molecular screening protocol; genomic screening protocol; treatment arm assignment

## Sources

- Woodcock and LaVange, Master protocols to study multiple therapies, multiple diseases, or both (NEJM 2017): https://doi.org/10.1056/NEJMra1510062

## Connected records

- terms: [Basket trial](https://onco.cc/terms/basket-trial/), [Basket, umbrella, and platform trials](https://onco.cc/terms/basket-umbrella-platform/), [Enrichment and biomarker-stratified designs](https://onco.cc/terms/biomarker-stratified-design/), [Minimal / molecular residual disease (MRD)](https://onco.cc/terms/mrd/), [Phase 1, 2 and 3 trials](https://onco.cc/terms/trial-phases/), [Seamless, adaptive and Bayesian trial designs](https://onco.cc/terms/seamless-adaptive/), [Single-arm trial](https://onco.cc/terms/single-arm/), [Trial protocol and statistical analysis plan](https://onco.cc/terms/trial-protocol/)
- fronts: [Drug Discovery Platforms](https://onco.cc/fronts/drug-discovery/)
- trials: [myeloMATCH](https://onco.cc/trials/myelomatch/), [NCI-COG Pediatric MATCH (APEC1621)](https://onco.cc/trials/pediatric-match/), [STAMPEDE](https://onco.cc/trials/stampede/)
- bottlenecks: [Trial design, endpoints and cost](https://onco.cc/bottlenecks/b-trial-design/), [Trials enrol too few, too slowly](https://onco.cc/bottlenecks/b-trial-enrolment/)

---
JSON: https://onco.cc/api/v1/entities/umbrella-trial.json