Reactive oxygen species promote leukaemogenesis in acute myeloid leukaemia

Acute myeloid leukaemia (AML) is characterised by the clonal proliferation of immature myeloid precursors that ultimately leads to death if untreated due to bone marrow failure. In Australia, there are approximately 900 new cases of AML per annum. The median age of diagnosis is 68 years and the overall survival remains poor, particularly in older adults. The backbone of treatment is intensive chemotherapy in those patients fit enough to tolerate it followed by consolidation with an allogeneic stem cell transplant in patients who attain a complete remission and with an appropriate donor. In older adults, the treatment consists of palliative chemotherapy, transfusion support and more recently the incorporation of novel targeted therapies. AML is a biologically heterogeneous disease and has been extensively genomically characterised which has afforded treating clinicians with a vast array of prognostic tools which primarily has guided transplant decision making. In addition, new ‘druggable’ mutations in the FMS-like tyrosine kinase 3 (FLT3) and the isocitrate dehydrogenase 1 and 2 (IDH1 and IDH2) genes have led to the development of novel targeted agents that have rapidly moved into the clinic. Despite this, the prospect for long-term survival in most patients with AML is limited, and the physical and emotional toll of treatment and the disease itself is immense. The first Chapter of this thesis characterises a cohort of local AML cases from the Hunter New England Haematology Department of the Calvary Mater Hospital Newcastle, and includes demographics, AML sub classification, cytogenetic and molecular characteristics as well as survival outcomes. This data has been compared to state-wide outcome data highlighting comparable results for the local regional setting. The data also highlights the huge unmet need for patients in our service treated for AML with overall poor survival statistics that correlate with other datasets both from Australia and globally.

Reactive oxygen species or ROS are a heterogeneous group of reactive molecules and free radicals that are derived from molecular oxygen. Our understanding of their role in both physiological and pathological cellular processes has increased dramatically in recent years. There has been particular interest in understanding the role of ROS in a variety of cancers, including AML. ROS were originally thought to be mere by-products of cellular metabolism that if not tightly regulated would lead to cellular damage triggering apoptosis. It is now understood that ROS, in fact, regulate intracellular signalling cascades through posttranslational modifications (PTM) of proteins that contain ‘ROS sensitive’ cysteine residues. In addition, high levels of ROS lead to DNA damage, which leads to genomic instability and further promoting oncogenesis. Chapter 2 of this thesis provides a literature review of the understood role of ROS in promoting leukaemogenesis in AML and following on from this we utilised a novel redox mass spectrometry approach to assess the proteome, phosphoproteome and reversible cysteine oxidation (‘oxidome’) comparing primary FLT3-ITD mutant AML with wild type FLT3 samples. Our data highlighted NOX2 and its associated regulatory subunits to be upregulated in the FLT3 mutant samples suggesting a novel therapeutic target in this subset of AML. Utilising novel NOX2 inhibitors and genetic knockdown of NOX2 in combination with FLT3 inhibitors we were able to demonstrate synergistic cytotoxicity using in vitro cell line models. We followed this up with a PDX model of FLT3-ITD AML and confirmed our results in vivo. Furthermore, we were able to demonstrate reduced activity of downstream growth and survival pathways of FLT3 including STAT5 and ERK, and restoration of the p38-MAPK apoptotic pathway.

Over the course of this thesis the BH3 mimetic Venetoclax has become the standard of care for elderly patients in combination with hypomethylating agents based on the success of the Viale A trial. Inhibition of anti-apoptotic BCL2 proteins leads to activation of the intrinsic apoptotic pathway and the generation of high levels of mitochondrial ROS. We observed upon NOX2 inhibition a significant rise in mitochondrial ROS levels, after an initial fall in cytoplasmic superoxide, suggesting activation of the intrinsic apoptotic pathway. This highlighted a possible therapeutic synergy using BCL2 and NOX2 inhibition in AML cells. In Chapter 7 of this thesis, we explored using our novel NOX2 inhibitors with Venetoclax using a number of in vitro and in vivo models, with the most promising results observed in a Sorafenib (FLT3 inhibitor) resistant model. This body of work in this thesis highlights the role that ROS play in promote leukaemogenesis in AML and identifies the therapeutic potential of targeting NOX2 as the primary driver of ROS, both alone and in combination with standard of care treatments.