Acute myeloid leukemia (AML) is a deadly blood cancer with poor long-term survival, and current treatments often fail due to relapse and limited model accuracy. To better capture disease heterogeneity, we developed a diverse panel of patient-derived AML models. By integrating proteomics and drug-response data, this approach enabled the creation of a deep learning–based map of therapeutic sensitivity that guides treatment positioning for upcoming clinical trials.

www.hupo.org