Stem Cells, Differentiation and Therapeutic Resistance in AML

描述（由申请人提供）：与大多数癌症一样，急性髓性白血病（AML）治疗的主要问题是治疗耐药性。虽然>85%的患者对化疗有反应，但超过一半的患者会复发。耐药性意味着癌细胞之间的异质性。存在两种假说来解释这种异质性。干细胞假说提出了一个有序结构内的功能异质性，其中大部分白血病细胞从白血病干细胞（LSC）分化。在该模型中，LSC本质上对化疗具有抗性。因此，复发性肿瘤来自LSC，其遗传多样性将反映原始肿瘤的遗传多样性。因此，化疗药物靶向LSC内发现的遗传病变。克隆进化模型提出细胞之间存在可遗传的变异，其中部分或全部具有形成新肿瘤的潜力。外部压力源，如化疗药物，对整个种群施加进化压力，驱动克隆进化并选择抗性克隆。在这个模型中，原始肿瘤中的（epi）遗传多样性将增加化学抗性的可能性，因为更广泛的突变库可以提供选择性优势。此外，随着肿瘤细胞的进化和发展新的耐药机制，复发肿瘤中遗传和表观遗传病变的特征继续改变。因此，最佳治疗将需要具有不同机制基础的治疗剂的混合。在这里，我们将测试我们的假设，即AML中的化疗耐药性最好通过克隆进化来建模，并利用我们的结果更好地理解AML中化疗耐药性的机制。在具体目标1中，我们将使用一种新的AML异种移植模型来确定化疗后LSC是否富集。在具体目标2中，我们将使用创新的单细胞方法来表征30个AML配对样本（从头诊断与复发）的克隆多样性，以确定治疗是否选择克隆。在特定目标3中，我们将进行一项探索性研究，以确定克隆多样性（通过新型乳剂亚硫酸氢盐测序（BBS）测定）是否可预测AML患者的复发概率。具体目标4将通过进行RNA测序和SNP阵列来比较来自30对匹配患者样本的新发和复发AML之间的表达水平、表达基因的序列和拷贝数改变，从而鉴定和确认通过化疗选择的基因突变。通过这些不同的方法，我们将测试化疗耐药性的LSC与克隆进化模型的几个特征。这些结果将对AML新疗法的开发、耐药疾病的治疗和复发的预防具有重要意义。

DESCRIPTION (provided by applicant): The major problem in therapy of acute myeloid leukemia (AML), like most cancers, is therapeutic resistance. Although >85% of patients respond to chemotherapy, over half will relapse. Chemotherapeutic resistance implies heterogeneity among cancer cells. Two hypotheses exist to explain this heterogeneity. The stem cell hypothesis proposes a functional heterogeneity within an ordered structure in which the bulk of leukemic cells differentiate from leukemic stem cells (LSC). In this model, LSCs are intrinsically resistant to chemotherapy. Consequently, relapsed tumors arise from LSCs and their genetic diversity will reflect that of the original tumor. As such, chemotherapeutics are targeted toward the genetic lesion found within the LSC. The clonal evolution model proposes that there is heritable variation between cells, some or all of which have the potential to form a new tumor. External stressors, such as chemotherapeutics, apply evolutionary stress on the entire population, driving clonal evolution and selecting for resistant clones. In this model, (epi)genetic diversity in the original tumor will increase the probability of chemoresistance, due to the wider repertoire of mutations that could provide a selective advantage. Moreover, as tumor cells evolve and develop new mechanisms of resistance, the profile of genetic and epigenetic lesions in the relapsed tumor continues to change. Accordingly, optimal therapy will require a mix of therapeutics with diverse mechanistic bases. Here, we will test our hypothesis that chemotherapy resistance in AML is best modeled by clonal evolution and utilize our results to better understand the mechanism(s) of chemotherapy resistance in AML. In Specific Aim 1, we will use a novel xenotransplantation model of AML to determine empiricially if LSCs are enriched after chemotherapy. In Specific Aim 2, we will use innovative single cell approaches to characterize a cohort of 30 AML paired samples (de novo diagnosis vs. relapse) for clonal diversity to determine if therapy selects for clones. In Specific Aim 3, we will perform an exploratory study to determine if clonal diversity (as measured by a novel emulsion bisulfite sequencing (BBS) assay) predicts the probability of relapse in AML patients. Specific Aim 4 will identify and confirm gene mutations that are selected by chemotherapy, by performing RNA sequencing and SNP arrays to compare the level of expression, sequence of expressed genes and copy number alterations between the de novo and relapsed AML from 30 matched pairs of patient samples. Through these diverse approaches, we will test the several features of the LSC vs. clonal evolution models of chemotherapy resistance. These results will have important implications for the development of novel therapeutics for AML, the treatment of chemoresistant disease and the prevention of relapse.