Trajectory and Architecture of Tumor Intrinsic Drug Resistance in AML

点击翻译按钮获取中文摘要

PROJECT SUMMARY/ABSTRACT: Project 1 Acute myeloid leukemia (AML) is a complex and genetically heterogenous disease and one of the most common hematologic malignancies. After 30-40 years without new therapies, several recent drugs have been approved for AML, including inhibitors of FLT3, IDH1/2, and BCL2. Despite improved initial response rates, none of these regimens lead to durable remissions. Acquired resistance to these agents develops due to diverse mechanisms that include tumor cell adaptation, often driven by microenvironmental signals. For the past decade, our collaborative team has employed numerous techniques, models, and analytical approaches to studying acquired drug resistance in AML – partly as a Center in the Drug Sensitivity and Resistance Network (DRSN) – the predecessor to ARTNet. We have developed the largest-to-date functional genomics platform of primary AML patient samples and implemented genome-wide CRISPR screening. Computational integration of these datasets has generated many predictions for mechanisms of drug resistance and nominated rationally selected drug combinations, some of which are in clinical trials. This analysis has led to a central hypothesis that tumor intrinsic biology can adapt in the face of therapeutic pressure, often with support from cell extrinsic signals, to undergo a multi-step process where early drug resistance is formed via cross-talk with immune and stromal cells that leads to an eventual late, cell autonomous resistant state with features of clonal evolution. For this project, our long-term goals are to optimize and translate the most effective drug combinations into the clinic for patients with AML. Our immediate goals are to understand tumor intrinsic mechanisms of acquired drug resistance. To accomplish these goals, three Aims are proposed: 1) Next-generation genome-wide interrogation of key acquired resistance scenarios – We created a panel of AML models of acquired drug resistance. These models have been generated with long-term drug exposure, sometimes with support from extrinsic cytokines. We will subject these drug resistant cells to genome-wide CRISPR screens with overlay of drugs or drug combinations. 2) Epigenomic evolution of acquired resistance – We will use protocols for expansion of myeloid progenitor cells from primary AML patient samples to study epigenetic adaptation. Using the same list of drugs and drug combinations as in Aim 1, we will profile shifts in epigenetic landscape using single-cell sequencing. 3) Atlas of intrinsic drug resistance in AML – We have expertise with broad data integration and modeling approaches. We will use these strategies to leverage our existing functional genomic dataset combined with the new data generated in Aims 1 and 2 to generate an Atlas of tumor intrinsic mechanisms of acquired drug resistance in AML. Cumulatively, we expect these innovative analyses to have a major impact on our understanding of acquired drug resistance in AML, leading to successful clinical translation of new, more effective drug combination strategies.