Why AML, and how this program works

A public, funded literature-and-data program on repurposing approved drugs for acute myeloid leukaemia.

By Igor Moiseev·1 Nov 2026

This program analyses published literature and public datasets. It gives no medical advice and makes no clinical claims. If you or someone you care for has AML, talk to the treating haematologist.

Acute myeloid leukaemia (AML) is the most common acute leukaemia in adults, and outcomes in older patients remain poor. Repurposing drugs that are already approved for other diseases is attractive because their safety profiles are known. The literature on candidate drugs is large and uneven, though, and much of it reports cell-line results that may not survive contact with real exposures.

This program asks narrow, testable questions about that literature, one experiment at a time. Every experiment is pre-registered before it runs: the question, the hypothesis, the metric and the threshold that would confirm or refute it are written down first.

How a run works

Each experiment runs in five steps. Ground searches Europe PMC, OpenAlex and arXiv and reads the relevant papers. Plan writes the pre-registered test and the script that will compute it. Build runs the script against public data. Verify re-executes every number the run will publish. Deliver writes the finding, with its caveats, as an article on this page.

Every number in a finding is a claim: the output of a stored script that anyone can re-run from the finding page. Every dollar spent on a run is recorded, and the donors who paid for it are named on the finding.

What runs next

The proposals below are ranked by expected value over cost. Donations to this program pay for the top proposal first.