Molecular mechanisms of drug resistance and disease progression in acute myeloid leukemia

点击翻译按钮获取中文摘要

Acute myeloid leukemia (AML) is a heterogeneous disease with highly variable clinical outcomes. Recurring chromosomal abnormalities permit assignment of some AML patients to favorable vs unfavorable prognostic groups. Adverse prognosis is also associated with increased expression of a set of homeodomain transcription factors (HoxA7-11 and Meis1) and their target genes. This expression profile is found in AML with translocation or partial duplication of the MLL1 gene (i.e. 11q23-AML), translocations involving the MYST3 and CREBBP genes, and a subset of cytogenetically normal (CN) AML. We previously identified a set of common HoxA9/10 target genes that was enriched for cytokine receptors and their pathways, including Fgf2, Tgfβ2 and β3 integrin. HoxA9 and 10 cooperate to activate these genes in hematopoietic and leukemia stem cells (HSC and LSC). Consistent with this, we found that treatment with Fgf- R inhibitors (Fgf-R blocking antibody or the small molecule inhibitor, nintedanib) decreased proliferation and survival of CD34+ bone marrow cells from subjects with “Hox-profile” AML compared to samples from those without. In preliminary studies, we found that adding nintedanib to standard induction chemotherapy significantly prolonged remission and survival vs chemotherapy alone in a murine model of MLL1 rearranged leukemia. HoxA9 and A10 also regulate the innate immune response, but in this case their activities are antagonistic. During the innate immune response, activation of emergency granulopoiesis requires HoxA9, but HoxA10 is required to terminate the process. Termination required activation of the Triad1 gene by HoxA10; a process that involved overcoming repression by HoxA9. Triad1 is an E3 ubiquitin ligase that degrades growth factor receptors and Mdm2. We found Triad1 knockdown accelerated leukemogenesis in a murine model of 11q23-AML. In this murine model, we found the Lin-ckit+ LSC transcriptome was enriched for pathways involved in cytokine production, receptor tyrosine kinase (RTK) signaling, regulation of protein kinase activity, and positive regulation of the immune response vs control cells. We also found activation of pathways not previously associated with 11q23-AML, including Rap1 signaling. We profiled gene expression in mice in chemotherapy- induced remission (destined to relapse) vs chemotherapy + nintedanib (with sustained remission). We found differences in cytokine receptor activity, Pi3k/Akt signaling, guanine nucleotide exchange factor activity, purine/pyrimidine metabolism, oxidative stress response, and glycosamino-glycan biosynthesis. We hypothesize that adverse prognosis in Hox-overexpressing AML is characterized by impaired regulation of cytokine stimulated pathways and an activated stress response. These pathways represent rationale therapeutic targets to decrease chemotherapy resistance. This hypothesis will be pursued through three aims: Aim 1: Define the role of receptor tyrosine kinase pathways in drug resistance in Hox-overexpressing AML. We will study RTK/PI3K signaling pathways identified in our studies in bone marrow from murine AML models, including activity of anti-apoptotic BH3 proteins and protein phosphatase 2a (PP2a). Aim 2: Determine the impact of Rap1 signaling on disease progression in Hox-overexpressing AML. We will determine the role of various guanine nucleotide exchange factors on Rap or Ras activity, disease progression, and LSC persistence during chemotherapy. Aim 3: Investigate these pathways in AML patient samples. CD34+ bone marrow cells from AML patients at presentation or with relapsed or refractory disease after standard chemotherapy +/- approved agents will be analyzed. Hox-overexpressing AML will be compared to AML without increased Hox protein expression. The goal of these studies is to identify therapeutic targets to abolish or suppress LSCs during or after chemotherapy in adverse prognosis AML. This will be especially relevant to design new approaches to patients who are not candidates for aggressive treatment due to refractory disease or the presence of co-morbidities.