Acute Myeloid Leukemia (AML) is the most fatal hematologic malignancy in adults, largely due to high rates of relapse following chemotherapy. Residual disease, the persistence of immunophenotypically, morphologically, or genetically defined leukemia cells following frontline chemotherapy, is a poor prognostic factor for AML. Residual AML cells have been shown to localize to endosteal bone marrow regions in murine models, where they may utilize microanatomical niches specialized to support long-term quiescent hematopoietic stem cells to become dormant and gain resistance to chemotherapy, which targets dividing cells. However, these niches are ill-defined in human bone marrow, and it remains unclear whether human AML cells actually interact with canonical stem cell niches as a mechanism to evade chemotherapy. Our central hypothesis is that AML cells interact with cells in protective niches within bone marrow by overexpressing stem cell niche retention receptors, which allows them to acquire chemoresistance. In Aim 1, we will employ Co-Indexing by Epitopes (CODEX), a recently developed method to visualize dozens of antibody markers on a single-cell level from fixed human tissue, to dissect the exact composition and spatial organization of human bone marrow niches. For the first time, we will be able to determine where chemoresistant cells localize in clinical bone marrow biopsies from AML patients. We will assess whether or not non-dividing residual AML cells localize preferentially to endosteal stem cell niches in the bone marrow in response to chemotherapy. In Aim 2, we will determine whether stem cell niche retention receptors are overexpressed on residual AML cells and promote chemoresistance. To achieve this, in Aim 2A we will first perform single-cell RNA sequencing of AML bone marrow aspirate samples from diagnosis and time of residual disease determination, as well as healthy controls, providing an expression profile of AML cells, AML-associated microenvironmental cells, and their healthy counterparts. We will determine whether receptors mediating niche retention are overexpressed. In Aim 2B, we will assess whether they are functionally important for chemoresistance using CRISPR-Cas9 mediated knockout of overexpressed receptor-ligand pairs which mediate AML-stromal interaction in an AML- stormal co-culture system that will be treated with cytotoxic chemotherapy. If AML cells indeed localize to these endosteal niches and gain chemoresistance through receptor-ligand mediated interaction in the stem cell niche, these receptor/ligand pairs would represent an attractive target for chemosensitization, which may help reduce AML relapse risk. Together, this project is poised to evaluate whether leukemic cells hijack healthy stem cell niches to evade standard-of-care chemotherapy, and this work has the potential to reveal new therapeutic opportunities to better treat this deadly disease.