Single-Cell Analysis of the Transcriptional Response in AML Patients Treated with BET-Inhibitors

Acute Myeloid Leukemia (AML) is the most common myeloid leukemia in adults. Although substantial progress has been made in recent years, the long-term prognosis for patients remains poor which is mainly due to the dated treatments that consist of cytotoxic drugs with low specificity. AML is a clonal disease with multiple co-existing clones in each patient. Often, patients that initially respond to treatment may develop resistance due to lingering leukemic stem cells (LSC), or sub-clones that survive the treatment and cause a relapse. Therefore, novel therapeutic strategies are needed to fully eradicate all leukemic cells. AML has a strong epigenetic component meaning mutations in genes encoding epigenetic regulators are frequently acquired during early AML development, and are present in the initiating clones. Thus, targeting the epigenetic machinery may offer a new avenue for AML treatment. Among the newer epigenetic drugs are BET inhibitors, which bind reversibly to bromodomains of BRD proteins and prevent protein-protein interactions with acetylated histones and transcriptions factors. One of the most promising BET inhibitors is OTX015, which has already been in Phase II clinical trials for AML in the U.S. (Braun & Gardin, Expert Opinion on Investigational Drugs, 2017).

We aim to analyze the heterogeneous response to OTX015 in AML, and normal stem/progenitor, cells in order to dissect the BET-inhibitor response. The main focus is the specific transcriptional signatures at promoters and enhancers as enhancers, and especially super-enhancers, have previously been shown to be sensitive to BET-inhibitors (Loven et al, Cell, 2013). To this effect, we have established a protocol that allowed for the transcriptional profiling of single cells from AML patients that were at different differentiation stages, using FACS- sorting. The patients were obtained from the Swedish Acute Leukemia Registry. To decrease population heterogeneity, the project focused on distinct subgroups of AML that previously has been shown to be sensitive for BET inhibitors. The different isolated AML, and normal progenitor populations, were exposed to OTX015 for 48hrs, and processed with both bulk transcriptional profiling of the general cell population response, and single cell profiling to analyze cell heterogeneity, and single cell response. For the transcriptional profiling, we utilized a unique technique called Cap Analysis of Gene Expression (CAGE), a powerful 5' start profiling technology, that allows for the identification of the transcription start site at base pair resolution, and determination of enhancer activity based on enhancer RNA expression. The single cell profiling was performed using C1 CAGE, which is a single-cell implementation CAGE (Kouno et al, bioRxiv 330845, 2018).We envision that the heterogenic transcriptional drug response, on the single-cell level, in AML and normal stem/progenitor cells will lead to the identification of key genes and pathways involved in the differential drug response. Additionally, the application of CAGE technology will lead to discovery of specific transcriptional signatures at promoters and enhancers that may be predictive of drug resistance.

Clinical significance: Leukemic cell heterogeneity remains the main problem in AML, as chemotherapy often fails to completely eradicate all AML sub-clones including LSC, leading to relapses and high mortality of the disease. This study will shed light to the unique features of AML cell heterogeneity and how their drug response differs, not only between AML cells, but also between AML cells and their normal counterparts, on the single-cell level, based on the response to OTX015. The significance will be two-fold: the in-depth characterization of the features in AML populations and normal cells, and the potential this study will provide for novel, more targeted, combination treatments in AML.