ideasIdea
A tumour board assistant that cites its evidence and tracks outcomes
Give every tumour board a tool that pulls up the relevant trials and guideline lines for each case with citations, records what was decided, and later shows how the patient did.
Molecular tumour boards spend much of their time on literature search and produce recommendations that are rarely tracked to outcomes. The proposal is an assistant built on the evidence graph and computable guidelines: it prepares a cited evidence brief per case, records the board's recommendation in structured form, and links it to the patient's subsequent treatment and outcome, creating a learning loop across boards. Precedents include MSK's OncoKB-driven annotation and academic MTB decision-support pilots.
Hypothesis
Boards using the assistant will spend less time per case, reach recommendations more concordant with current guidelines, and will generate an outcome-linked dataset that reveals which recommendation types improve survival.
Rationale
Studies of tumour boards show variable concordance with guidelines and almost no outcome feedback; the missing ingredients are retrieval and structured recording, both now tractable.
What would test it
Deploy at five boards for one year; measure preparation time, guideline concordance, and completeness of outcome linkage versus the prior year.
Maturity
early clinical
Who has to act
clinic
Cost to try
Small (under $1M)
Years to first evidence
2
Bottlenecks it attacks
- Knowledge reaches practice too slowly · Knowledge diffusion is slow: it takes years for a proven result to change what most patients receive, and no one can keep up with the literature.
- Weak real-world evidence and registries · We do not reliably know what happens to patients after approval, so we cannot tell which drugs deliver in practice.