RNA Polymerase II (RNAPII) transcription is a discontinuous and unidirectional process, control of which is necessary for normal gene expression and cellular functioning. Regulation occurs at discrete checkpoints throughout transcription and is in part mediated by Cyclin Dependent Kinases (CDKs). This includes CDK7, where it phosphorylates RNAPII among other transcriptional substrates to enable transcriptional initiation and coordination of co-transcriptional processes, such as splicing. However, the genome-wide role of CDK7 in nascent RNA synthesis and splicing, in addition to whether its function is negatively regulated, has not been robustly established. The loss of RNAPII transcription control invariably leads to disease, including cancer. Mutations in cancer can affect several proteins involved in regulating RNAPII and engenders a reliance on sustained functioning of the core RNAPII machinery for continued malignant cell survival and proliferation. This dependence is termed ‘transcription addiction’ and is the basis for targeting RNAPII core components for therapeutic benefit. Small molecule perturbation of transcription in these cancers results in selective gene expression changes attributed to the target gene’s specific chromatin context, which includes the occupancy of oncogenic transcription factors (TFs) and association with genomic elements such as super enhancers (SEs). Whether selective transcriptional changes are also influenced by RNA stability however, is largely unexplored. In this thesis, we demonstrated that RNA decay rates largely determined gene expression changes to clinically relevant transcriptional and epigenetic compounds. Transcripts that were the most down-regulated on the total RNA level had both short half-lives and high production rates and often encoded TFs, including the proto-oncogene c- MYC. Moreover, genetically engineering the three prime untranslated region (3’UTR) of c-MYC to prolong its RNA half-life rendered c-MYC transcripts resistant to transcriptional targeting on the total RNA level. Therefore, these finding highlight an under-appreciated role of RNA stability in gene expression changes to therapeutic RNAPII perturbation in cancer. Among the most promising therapeutic transcriptional targets is CDK7, and inhibitors targeting this protein are currently clinical trials and have been expedited for Food and Drug Administration (FDA) approval. Proteomic and transcriptomic characterization of the CDK7 inhibitor, YKL-5-124, in the THP-1 Acute Myeloid Leukaemia (AML) cell line revealed that it perturbed the phosphorylation of several proteins involved in cell cycling, transcription and co-transcriptional processes, namely splicing. This was paralleled with a global decrease in de novo RNA synthesis, most prominently towards the 5’ ends of genes, and total RNA levels. In addition, CDK7 inhibition increased alternative isoform expression, most strikingly related to exon skipping and intron retention. On a phenotypic level, these changes were also associated with a decrease in THP-1 cell proliferation. Taken together, these findings improve our understanding of the significance of CDK7 kinase activity in RNAPII transcription and splicing. To identify factors that functionally antagonize CDK7 and may confer resistance to sustained CDK7 inhibition, CRISPR-Cas9 knockout positive enrichment screening in THP-1 cells was performed using YKL-5-124 treatment as a selective pressure. The screen showed that knockout of histone acetyltransferase (HAT) and core structural components of the Spt-Ada-Gcn5 acetyltransferase (SAGA) complex mediated resistance to CDK7 inhibition. Further investigation into the resistance mechanism revealed that cells knocked out for the SAGA HAT subunit, TADA2B, were less sensitive to nascent RNA down-regulation with YKL-5-124. Moreover, this was concomitant with maintained H2BK120 mono-ubiquitin levels across gene bodies genome-wide. Therefore, these data highlight SAGA complex components as novel negative regulators of CDK7 and how cancers might overcome the effect CDK7 inhibitors in the clinic. Overall, the results presented herein improve our understanding of CDK7 in RNAPII transcription and the importance of RNA decay in defining the anti-cancer effects of therapeutic perturbation of the core RNAPII machinery.

The advancement of cancer immunotherapy has revolutionized cancer treatment. In particular, the success of immune checkpoint blockade, including anti-PD-1/L1 and anti-CTLA-4 in the clinic has changed guidelines for treating multiple solid tumours such as melanoma, non-small-cell-lung cancer (NSCLC), renal cell carcinoma and head and neck squamous cell cancers. Adoptive T cell therapy against haematological malignancies has also achieved remarkable success. Two anti-CD19 CAR T cell products have been approved by Food and Drug Administration (FDA) for the treatment of B cell Acute Lymphoblastic Leukaemia (ALL) and non-Hodgkin lymphoma. However, the efficacy of adoptive CAR T cell therapy against solid tumours has been less impressive. Even though numerous preclinical or clinical trials have been conducted to identify novel neoantigens or to improve CAR T cell anti-tumour efficacy against solid tumours, all attempts have so far not matched the extraordinary clinical responses observed in patients with haematological malignancies. There is an urgent need to 1) understand CAR signalling in T cells, 2) identify the key mechanisms intracellularly and extracellularly limiting CAR T cell function, and 3) to develop strategies to overcome the immunosuppressive tumour microenvironment to improve the efficacy of CAR T cells in the context of solid tumours. Recent studies have highlighted the potential of targeting negative regulators of T cell signalling to enhance the efficacy and extend the utility of CAR T cells to solid tumours. Protein tyrosine phosphatases (PTPs) including PTPN2 and PTPN22 are key regulators of T cell signalling and contribute to the maintenance of immune tolerance. PTPN22 is known as a critical autoimmunity-associated PTP that suppresses T cell receptor (TCR) signalling by dephosphorylating substrates such as the Src family kinase LCK and the Syk family kinase ZAP70 in effector/memory T cells. Loss of function SNPs in PTPN2 have also been associated with the development of autoimmunity. Studies from our group have shown that PTPN2 plays pivotal role in negatively regulating T cell receptor (TCR) signalling by dephosphorylating and inactivating the Src family protein tyrosine kinase LCK to limit T cell proliferation and function. PTPN2 also attenuates cytokine signalling by dephosphorylating JAK-1, JAK-3 and their target substrates STAT-1, -3 and -5 in a cell context-dependent manner. Beyond PTPN22 and PTPN2, early studies established that PTP1B could directly dephosphorylate and inactivate JAK-2 and Tyk-2 to attenuate cytokine-induced JAK/STAT signalling. There is also evidence that PTP1B can dephosphorylate STAT family members. In this thesis, we found that basal STAT-5 Y694 phosphorylation and IL-2/IL-7/IL-15 induced STAT-5 Y694 phosphorylation were enhanced by the deletion of PTP1B in T cells. Since CARs signal via LCK, and cytokines can affect CAR T cell function, differentiation and memory, we postulated that inhibiting PTPN2, PTP1B and PTPN22 individually or combinatorically might bolster the anti-tumour activity of CAR T cells. Work in this thesis characterized the role of PTPN22, PTPN2 and PTP1B in adoptive T cell therapy and demonstrated that the deletion or inhibition of PTPN2 or PTP1B resulted in enhanced T cell signalling and promoted CAR T cell anti-tumour efficacy in vitro and in vivo. In addition, we developed a CRISPR/Cas9 RNP gene editing based method to target PTPs in CAR T cells for translational purposes.

Multiple myeloma (MM) is a haematological malignancy of plasma cells with disease present in the bone marrow. Despite the success of new therapies in extending life expectancy, MM is still considered incurable and requires further investigation into new treatments. In the context of refractory/relapsed MM (RRMM), the combination of the two immunotherapies Elotuzumab (Elo, anti-SLAMF7 mAb) and Lenalidomide (Len, IMiD) has resulted in an 82% objective response rate and 16% complete remission in patients. This success led to FDA approval of Elo plus Len treatment for patients with RRMM and has warranted further investigations into the exact mechanism of action. Elo targets SLAMF7 antigen which is highly expressed on MM tumour cells and activates NK cells for antibody dependent cell-mediated cytotoxicity (ADCC) resulting in MM tumour cell death. Len is an immunomodulatory agent commonly used to treat a wide range of haematological cancers. Len has been shown to activate various cells of the immune system as well as directly suppress tumour progression. This thesis aims to identify how the mechanisms of Len intersect with, and enhance, Elo activated NK cell control of tumours. The work from this thesis revealed that to induce ADCC, Elo required both expression of SLAMF7 on MM target cells and NK cell CD16 engagement. Elo-activated NK cells had increased expression of CD107a and CD69, as well as loss of CD16 expression which was a result of ADAM17 induced cleavage. Elo activation of NK cells also secreted increased levels of cytokines and chemokines associated with recruitment of effector immune cells. In both in-vitro and in-vivo studies, Len significantly enhanced Elo-induced healthy donor NK cell killing of MM tumour cells. In contrast, only 4/12 RRMM patient (responders) NK cells induced increased MM cell killing in the context of Elo plus Len. Important for this increased cytotoxicity was the expression of CD54 on NK cells that allowed differentiation between these responder and non-responder patients. CD54 was also more highly expressed on NK cells activated by Elo plus Len compared to Elo activation alone. However, further phenotypic studies and RNA sequencing did not reveal any further mechanisms in NK cells by which Len enhances Elo activation. Further studies revealed that MM tumour (myeloma) cells specifically upregulated CD54 and CD11a expression in response to Elo plus Len activation of NK cells and this did not occur with Elo treatment alone. This increase in CD54/CD11a expression on myeloma cells was also dependent on the presence of CD14+ monocytes. RNA sequencing also revealed enrichment of cytokine and chemokine signalling/secretion pathways in MM cells specifically in the context of Elo and Len activated NK cells and monocytes. In conclusion, these data indicate that a complex network of direct and indirect mechanisms between NK cells, monocytes and myeloma cells contribute to the Elo plus Len treatment induced response in RRMM patients. Elo alone activates NK cells and monocytes, whereas the Len effect is largely on myeloma cells. Collectively the two treatments led to robust myeloma antibody dependent cellular cytotoxicity (ADCC). There was a strong correlation between increased myeloma cell CD54/CD11a expression and myeloma cell ADCC suggesting that stabilisation of the immune synapse with strong adhesion may be a key factor for this improved myeloma cell killing. Finally, this study revealed that upregulation of MM cell CD54 expression on MM cells may be a useful predictive biomarker to stratify RRMM patients for Elo plus Len therapy.

Mucinous ovarian carcinoma (MOC) is a rare subtype of ovarian cancer (OC) distinct from all other subtypes. MOC represents 3% of OC cases; however, it responds very poorly to conventional chemotherapies contributing to the poor outcome of OC. Moreover, there is a lack of effective therapies for MOC since treatments with platinum-based chemotherapy or PARP-inhibitors are unlikely to be effective. The recent description of the molecular landscape of MOC, including contributions by work in this thesis, allowed the identification of novel therapies that target the most recurrent genetic features of the disease. Nevertheless, there is still the need to develop accurate pre-clinical models in order to test the efficacy of these new therapies. This thesis evaluates the transcription factor ELF3 as a new MOC candidate driver gene. We performed a comprehensive analysis of 210 human mucinous ovarian tumours and identified that ELF3 was mutated in 8.2% of our sample cohort. To assess whether ELF3 acts as a tumour suppressor or oncogene in MOC and how its mutated function is expressed in cancer, we evaluated the ability of MOC cells to proliferate, migrate and invade, by introducing truncating mutations using CRISPR-Cas9 technology. It was found that ELF3 could possibly act as a tumour suppressor in MOC; however, its role might depend on coexisting mutations in the cells. This thesis also describes a novel 3D patient-derived therapeutic approach to pre-test the effectiveness of drugs, as monotherapy or in combination, before exposing patients to ineffective therapies or side effects. A mucinous ovarian tumour 3D platform that enables the correlation of patient genetic data with drug testing was established in the first result chapter of this thesis. MOC organoid development was based on the MOC genetic profile and characteristics as well as optimization that adapted human colorectal and pancreatic organoid culture systems. It was found that Wnt-3a is essential for mucinous ovarian tumour organoid long-term derivation. Moreover, MOC organoids faithfully recapitulate MOC primary tumours at the genomic level and retain histological characteristics of their corresponding primary tumours. The development of MOC organoids and 3D spheroids from cell lines together with the description of the genetic profile of MOC allowed the development and testing of a panel of novel therapies that targets the most frequent mutations found in MOC tumours, including a mutant TP53 activator, a Wnt inhibitor, RAF/EGFR/RAS/CDK4 and CDK6 inhibitors, and two antitumour compounds for HER2-overexpressing tumour cells (APR-246, OMP-18R5, BGB-283, BKM120, PD-0332991, PLX4032, Trastuzumab and Lapatinib). The organoids and spheroids showed different responses to the individual therapies tested, depending on the individual genetic features of the models. Spheroids showed sensitivity to PD-0332991, Trastuzumab, PLX4032 , and BKM-120 . MOC organoids also showed different responses: MOC organoid MOCOR-16 showed sensitivity to APR-246, and resistance to Trastuzumab; however, MOCOR-38 was sensitive to Trastuzumab but resistant to APR-246. This chapter also shows the first attempt of drug synergy scoring using automatic robots to create 3D MOC models. Synergistic combinations were identified for one of the spheroid lines when APR-246 and BKM-120 were combined. Interestingly, the spheroids were resistant to APR-246 in monotherapy. If validated independently, these results could potentially represent a more effective treatment for MOC. Antagonistic interactions were also found, with the strongest antagonism occurring between PLX4032 and BGB-283. The findings of this study substantially increase our understanding of mucinous ovarian tumours together with potential personalized therapeutic targets to improve the survival outcomes for women with this deadly disease.

Massively parallel sequencing (MPS) technology has revolutionised the genomic exploration of human disease. This is especially true in the case of cancer, which is primarily driven by the development of acquired genomic aberrations. The body of work described within this thesis represents a broad yet in-depth array of novel applications of MPS technology in the evaluation of haematological malignancies. This field is currently surging in relevance and clinical utility as the ongoing movement of MPS technology from the research to the routine diagnostic setting continues to facilitate the development of increasingly personalised medicine. High impact contributions have been made in a number of areas encompassing myeloid and lymphoid malignancies as well as haematological malignancies as a collective. Key achievements include: quantifying the risk of incidentally detecting germline variants of potential clinical significance during unpaired MPS testing of cancer samples; definitively proving that ASXL1 NM_015338.5:c.1934dup;p.Gly646Trpfs*12 is a true somatic alteration and developing an accurate and sensitive assay for its detection; exploring the pathogenesis of and mechanisms of resistance to histone deacetylase inhibitors in cutaneous T-cell lymphomas as well as defining the clinical features, outcomes and genomic landscape of transformed marginal zone lymphoma. This thesis represents a diverse portfolio of novel research with a strong translational focus. Despite the wide scope of the individual lines of inquiry described herein there is a common thread that binds the narrative together: the pursuit of innovative yet practical ways of utilising the powerful technology now available to improve the genomic characterisation of haematological malignancies and ultimately the lives of the patients and families they affect.

The tumour suppressor gene TP53 is mutated in over half of all cancers, resulting in mutant p53 protein accumulation and poor patient survival. Decades after the discovery that p53 plays a central role in the human body’s defence against cancer, the development of effective therapeutic strategies to target mutant p53 cancers still remains an unmet need in oncology. Eprenetapopt (also known as APR-246 and PRIMA-1MET) is currently in clinical development as a mutant p53 reactivator. The reported mechanism of action of eprenetapopt is to restore wild-type p53 functionality to mutant p53 in cancer cells, triggering p53-dependent cell cycle arrest and caspase- dependent apoptosis. As a result, clinical investigation of eprenetapopt has focused on TP53-mutated cancers, relying solely on TP53 mutation status for patient selection into clinical trials. However, the mechanism of action of eprenetapopt remains contentious, as evidence demonstrates that eprenetapopt maintains strong potency in cancer cells with no mutant p53. Herein, utilising novel unbiased approaches, including CRISPR perturbation screening, metabolomics, proteomics and pan-cancer cellular-feature profiling, this thesis demonstrates two major innovations regarding the determinants of sensitivity and mechanisms of action of eprenetapopt. First, the expression of SLC7A11, which encodes the functional subunit of the cystine- glutamate antiporter, not TP53-mutation status, is the major determinant of sensitivity to eprenetapopt. Second, eprenetapopt triggers a caspase-independent, iron- dependent form of cell death known as ferroptosis, following glutathione depletion and inhibition of iron-sulfur cluster biogenesis. Together, this thesis provides a roadmap for broadening the clinical utility of eprenetapopt as a ferroptosis inducer in haematological and solid malignancies, beyond TP53-mutant cancers.

Chimeric antigen receptor (CAR) T cell therapies have been highly effective and clinically approved for treating haematological malignancies, however trials in solid cancers have shown limited efficacy, likely due in part to the increased complexity of the immunosuppressive tumour microenvironment (TME) in solid cancers. CAR T cells are inhibited by immunosuppressive proteins, cytokines or physical barriers deployed by the tumour to evade and avoid destruction by anti-tumour immunity. One such process involves the accumulation of extracellular adenosine (eADO), in the TME which has potent immunosuppressive effects on T cells and other immune cells. eADO has four known G protein coupled receptors, the A1R, A2AR, A2BR and A3R, of which the A2AR is primarily responsible for suppressing T cell function. Our previous studies highlighted a major impediment to pharmacological blockade of the A2AR which was predicted to be hindered by poor solubility and suboptimal in vivo pharmacokinetic profile [1]. This became apparent when comparing the effectiveness with genetic deletion of A2AR in CAR T cells to pharmacological blockade, in which the CAR T cells generated from A2AR-/- mice elicited comparatively greater efficacy in vivo when combined with anti-PD-1 blockade [1, 2]. This thesis therefore investigated multiple gene editing strategies to modulate adenosine receptor signalling, firstly by overexpressing the alternative signalling A1R or A3R in human or mouse CAR T cells. A1R or A3R have been shown to act by the opposing downstream signalling pathway to A2AR, and thus it is hypothesised that A1R or A3R overexpression can reverse suppression and supercharge CAR T cells in the presence of eADO. Interestingly, A1R or A3R overexpression did not confer protection to suppression by eADO in both mouse and human models, but A1R expression instead enhanced effector and terminal differentiation, activation, and baseline cytokine production of CAR T cells. This however translated to higher expression of exhaustion markers, loss of memory associated gene expression and reduced stem-like memory fraction in the CAR T cell product, ultimately leading to reduced persistence in vivo, and limiting the therapeutic efficacy of this approach. Alternatively, a previous publication from our lab briefly examined short-hairpin RNA (shRNA) mediated silencing of A2AR expression [1]. While shRNA-mediated silencing of the A2AR was able to partially reverse suppression by eADO, much like A1R expression, it also led to effector differentiation, activation, and increased baseline cytokine production. Importantly, while shRNA-mediated silencing of the A2AR also resulted in reduced persistence in vivo, it was able to mediate modest anti-tumour efficacy leading to reduced tumour growth and increased mouse survival. Both overexpression and knockdown approaches are limited by sub-optimal persistence in vivo which limited their overall therapeutic efficacies. Yet these results contradicted our prior observations of CAR T cells derived from A2AR-/- mice and from studies in the Lymphocytic choriomeningitis virus (LCMV) setting, whereby A2AR deletion was linked to increased T cell numbers [1, 3]. Therefore, the final gene-editing approach examined in this thesis utilised CRISPR/Cas9 protocols to achieve full deletion of the A2AR in CAR T cells. CRISPR/Cas9 methodologies are currently being used in clinical trials and therefore deleting the A2AR in CAR T cells using this approach is highly novel and clinically translatable. To reasons unknown, CRISPR/Cas9 mediated deletion of A2AR had minimal effects on CAR T cell memory phenotypes and no adverse effects on engraftment or persistence in vivo. Furthermore, CRISPR/Cas9-mediated deletion of A2AR in CAR T cells led to enhanced therapeutic efficacy in both mouse and human models, thus representing a potent approach to targeting the A2AR. In conclusion, future studies comparing full A2AR deletion to A2AR silencing/ pharmacological blockade or A1R overexpression may be of interest to fully elucidate the mechanisms of adenosine receptor signalling on T cell persistence and memory.

Infection is a near-universal human experience and is responsible for substantial child mortality across the globe, despite impressive reductions in child mortality and morbidity since the twentieth century. Antibiotics and other antimicrobial drugs have transformed our ability to prevent and treat infection. In general, these drugs are so safe, effective and widely available that overuse and inappropriate use are common. This is a cause of real problems in hospitals and the community, with unintended consequences of antimicrobial use including rising antimicrobial resistance. Antimicrobial stewardship (AMS) is aimed at improving the safety and efficacy of prescribing and has received growing attention in recent years. However, evidence to support and improve AMS for Australian children in hospitals is lacking. Australian hospitals are mandated to implement AMS programs and provide access to appropriate national and/or local prescribing guidelines. However, hospitals are under no current obligation to provide appropriately targeted AMS for the children in their care. Prior to mid-2019, national antimicrobial guidelines contained little paediatric and no neonatal advice. Since 2013, the voluntary National Antimicrobial Prescribing Survey (NAPS) has provided national reports on prescribing. However, until now, paediatric-specific data have not been reported. Compared with the literature on adult AMS, research on paediatric AMS is lacking, with few high-quality studies on interventions to improve care. This situation creates challenges for child healthcare providers and paediatric AMS program leaders, and more evidence is required to prioritise and improve care. The overall aims of this thesis are to improve the understanding of current antimicrobial use and stewardship for children in Australian hospitals and determine priorities to improve antimicrobial use now and in the future. This is achieved by analysing antimicrobial prescribing epidemiology and quality using national datasets, including national point prevalence survey and cohort study data. Chapter 1 reviews antimicrobial prescribing to children in hospitals, including in Australia. Chapter 2 presents the first analysis of paediatric antimicrobial prescribing to children in hospitals throughout Australia using NAPS data. Chapter 3 turns to high-risk groups, presenting the first nationwide analysis of prescribing for neonatal sepsis and fungal infections, again using NAPS data. Chapter 4 presents an analysis of antimicrobial prescribing in a contemporary cohort of immunocompromised children with fever and neutropenia, including prescribing quality and outcomes. Chapter 5 presents an interventional study evaluating the implementation of Australian guidelines on antibiotic duration and intravenous-to-oral switch. This is an example of the evidence translation and implementation approach needed for sustainable AMS improvement. Chapter 6 concludes the thesis, discussing the implications of the research and the paediatric AMS horizon in Australia. The analyses reported here reveal unnecessary variations in care and systemic inequities, which have implications for policy and guidelines. Non-metropolitan and non-tertiary hospitals in general provide lower-quality antimicrobial prescribing to children. This is likely to reflect decreased access to high-quality AMS resources, including guidelines and personnel, suggesting the need for systemic improvements. Neonates in Australian hospitals receive highly structured care in terms of antimicrobial choice and indications, but variations in dosing are substantial and undesirable, reflecting the lack of use of national guidelines. Prescribing for febrile neutropenia is highly diverse and often includes empiric aminoglycosides, which this research reveals are associated with real harm, suggesting the need for national guidelines to optimise care. Finally, the standard management of infections in hospitals involves excessive intravenous therapy, which is associated with unnecessarily increased hospital length of stay. As demonstrated, this can be improved with a structured AMS program, which should be available wherever children are treated in hospital. The information generated by this thesis provides new evidence on current antimicrobial prescribing practice and priorities and demonstrates the importance of utilising routinely collected data for the surveillance and improvement of paediatric AMS. Since this body of research began, national guidelines and paediatric-specific resources are now being developed, establishing new benchmarks. Along with continuous surveillance, these must be implemented appropriately to improve care. The research collaborations and networks developed during the production of this thesis will be used to support future surveillance and implementation work, which is needed to address AMS priorities in Australia and support the research and development of paediatric AMS across the globe.

Over the past decade, outcomes for patients with advanced melanoma have been transformed with the introduction of immune checkpoint inhibitors and molecular targeted therapies directed against the MAPK pathway. Prior to the introduction of these treatments, the median overall survival (OS) of patients with melanoma was 9-12 months. In contrast, patients treated with combination anti-CTLA-4 and anti-PD1 immune checkpoint inhibitors (ipilimumab-nivolumab) now exhibit OS rates at the 5-year landmark that exceed 50%. This stunning revolution in systemic therapeutics for melanoma has paved the way for the use of anti-PD1 immunotherapy in various other solid cancer types. The central aim of this thesis is to provide additional knowledge that may improve outcomes for patients with advanced melanoma through the refinement and development of new approaches to the use of immunotherapy. The first chapter addresses the sequencing of systemic treatment in patients with melanoma brain metastases where previously outcomes were dismal with a median OS of 6 months or less. Phase II studies show that the first line combination dabrafenib-trametinib (BRAF-MEKi) or ipilimumab-nivolumab both display intracranial response rates of 50-55%. Notably responses to targeted therapy are relatively shortly lived at approximately 6 months whereas immunotherapy displays potentially durable intracranial activity. Critically, the real-world outcomes and the sequencing of targeted therapy and immunotherapy is currently unknown. This thesis shows that ipilimumab-nivolumab given after progression of BRAF-MEKi has modest intracranial activity with objective response rates of less than 5%. Given this unexpected result, whole transcriptome sequencing of BRAF V600 melanoma brain metastases naive to systemic treatment and those excised after progression of BRAF-MEKi treatment was performed to identify potential mechanisms of resistance to both targeted and immunotherapy. Gene set testing of this cohort showed enrichment of the Innate PD-1 Resistance Signature (IPRES) with upregulation of endothelial and myeloid cell activation genes. Furthermore, macrophage infiltration of the BRAF-MEKi resistant brain metastases was higher than lesions naive to treatment which suggests that this cell type is involved with resistance to immunotherapy particularly in the central nervous system (CNS). The second chapter of this thesis investigates whether the addition of cyclin-dependent kinase 4/6 inhibitors (CDK4/6i) can augment adoptive cell transfer (ACT) immunotherapy in combination with BRAF-MEKi. Other laboratories have shown that the addition of BRAF-MEKi with pmel-1 ACT can augment anti-tumour activity against the murine melanoma cell line SM1 in vivo. The addition of BRAF-MEKi not only enhanced anti-tumour activity by blockade of the MAPK pathway, but also exerted immunomodulatory effects including upregulation of melanoma differentiation antigens and antigen presentation pathways. The addition of CDK4/6i might enhance anti-tumour activity of combination BRAF-MEKi with ACT as melanoma exhibits a high rate of molecular aberrations that impact upon the cell cycle pathways and are an emerging therapeutic target. CDK4/6i such as palbociclib might further enhance ACT by inducing cell cycle arrest; moreover, several papers also show immunomodulatory effects such as increased tumour expression of major histocompatibility complex (MHC) class I and inhibition of T regulatory cells in the tumour microenvironment. The data contained herein shows that the CDK4/6 inhibitor, palbociclib did not further enhance expression of MHC Class I on SM1 or YOVAL1.1 melanoma cell lines by itself or in combination with BRAF-MEKi. However, combination BRAF-MEK-CDK4/6 inhibitor with OT-1 adoptive cell transfer leads to robust anti-tumour activity in vivo against YOVAL1.1. The third chapter of the thesis explores whether the addition of denosumab to standard immune checkpoint inhibitor regimens improves outcomes for patients with advanced melanoma. Denosumab is an anti-RANKL antibody and is an approved agent for osteoporosis, prevention of bone metastases and giant cell carcinoma. Recent pre-clinical studies show denosumab can enhance in vivo antitumour activity in pre-clinical models of melanoma, renal cell and prostate carcinoma. Given there are no known overlapping toxicities of denosumab with nivolumab or with the combination ipilimumab-nivolumab, we developed a Phase Ib/II trial of Ipilimumab-Nivolumab-Denosumab or Nivolumab-Denosumab for Unresectable or Metastatic Melanoma (NCT03161756). This multi-centre, investigator-initiated study was opened in December 2017. Available data of the combination of nivolumab-denosumab shows no concerning safety signals. Overall, this thesis sought to identify optimal treatment sequencing strategies for brain metastases, the development of novel immunotherapy strategies involving CDK4/6i with adoptive cell transfer and the safety of combining denosumab with nivolumab and ipilimumab-nivolumab.

Background Neural stem cells are cells that can self-renew and differentiate to give rise to specialized cell types, such as neurons. The appropriate regulation of neural stem cell activities during development guarantees that the correct number of functional neurons are generated to form the adult nervous system. Dysregulation of this process can result in a number of neurological diseases. Too few neural stem cells (and their neuronal progeny) can lead to microcephaly, and too many neural stem cells (and their neuronal progeny) can lead to brain tumour formation. Therefore, understanding the mechanisms regulating neural stem cell activities may shed light on novel therapeutics for neurological diseases. Drosophila neural stem cells, the neuroblasts (NBs), have been an advantageous model for the study of neural stem cell biology during development and in disease contexts. In the developing larval Central Nervous System (CNS), NB activities are maintained by intrinsic regulatory machinery as well as extrinsic cues from glial cells, which form a microenvironment enwrapping NBs. Dysregulation of either mechanism can have severe consequences. Deregulation of NB intrinsic regulators such as a homeobox transcription factor called Prospero (Pros) can result in decreased differentiation and an expansion of NBs, resulting in increased brain size. The disruption of glial niche or glial derived signals can lead to an alteration of NB proliferation rate, thus affecting the number of neurons in the adult CNS. While multiple intrinsic factors regulating NB activities during development have been intensively studied, it remained poorly understood how NB intrinsic metabolic pathways are important for NB proliferation under physiological and disease context. Furthermore, the mechanisms of how NBs respond to extrinsic cues and how the glial niche relay extrinsic signals to control NB activities is still not well understood. Hence, the aim of my study is to expand our knowledge on how intrinsic metabolic adaptations and extrinsic cues regulate the progenitor pool of NBs and the rate at which neuronal progeny is produced. Aims, methods and results: In this work, I have addressed two key questions: 1) what metabolic rewiring events occur as a result of NB tumour formation caused by dedifferentiation? 2) How do glial niche and glial derived signals non-autonomously regulate NB proliferative activities? 1) Metabolic rewiring is known to accompany aberrant proliferation and has recently been shown to drive tumour growth. Utilising the Drosophila NB tumour model caused by deregulation of pros (a homeobox transcription factor), we conducted transcriptomics and metabolomics studies to examine metabolic changes that occur as a result of defective NB differentiation. I found that glutathione metabolism is altered as a result of pros loss-of-function (LOF) tumour formation. It is known that the Pros homologue PROX1 is deregulated in neuroblastoma. Therefore, findings from our work may inform the search of novel metabolite-based targeted therapies for the treatment of neuroblastoma. 2) In the larval CNS, NBs re-enter into cell cycle from a quiescent state in response to extrinsic signals from their microenvironment made up of glial cells. In my study, I have uncovered several new regulators of glia-NB interaction using genetic analyses. I found that lipid metabolism and Hedgehog (Hh) in the glial niche are involved in regulating NB proliferation. Glial Hh is autonomously required for the growth of glial niche, which is key in maintaining NB proliferation during development. Upon glial overgrowth mediated by FGF activation, excessive Hh from the glial niche triggers a delay in NB cell cycle progression. In this glial overgrowth context, Hh function is modulated by two lipid regulators, Fatty acid synthase 1 (Fasn1) and Lipid droplet storage 2 (Lsd-2). I demonstrated that Fasn1 can modulate Hh palmitoylation. However, it remains unclear how Lsd-2 modulates Hh function, despite the observation that Lsd-2 and Hh both localize to the surface of lipid droplets localized to glial cells. I also found that glial derived chitinase like proteins, called imaginal disc growth factors (IDGFs) are important non-autonomous regulators of NB proliferation. I showed that IDGF2 is necessary for NB exit from quiescence and is required to sustain NB proliferation at later larval stages. Collectively, these findings revealed novel mechanism by which Drosophila neural stem cell activities are regulated during development and during tumorigenesis.

Metastasis is the lethal aspect of cancer for most patients. Remodelling of lymphatic vessels associated with a tumour is a key initial step in metastasis because it facilitates the entry of cancer cells into the lymphatic vasculature and their spread to lymph nodes and distant organs. Although it is clear that vascular endothelial growth factors (VEGFs), such as VEGFC and VEGFD, are key drivers of lymphatic remodelling in cancer, the means by which many signalling pathways in endothelial cells are co-ordinately regulated to drive growth and remodelling of lymphatics in cancer is not understood. In this thesis, I seek to understand the broader molecular mechanisms that control cancer metastasis through the analysis of microRNAs which act to co-ordinately regulate signalling pathways involved in complex biological responses, such as lymphatic remodelling, in health and disease. Here, using high-throughput small RNA sequencing, I found that a specific microRNA, miR-132, is up-regulated in expression in lymphatic endothelial cells (LECs) in response to stimulation with VEGFC and VEGFD. Interestingly, inhibiting the effects of miR-132 in LECs in vitro blocked proliferation and tube formation of these cells induced by VEGFC and VEGFD - LEC proliferation and tube formation are key steps in lymphatic remodelling. Moreover, I demonstrated that miR-132 is expressed in the lymphatic vessels of a subset of human breast tumours which were previously found to express high levels of VEGFD. In order to dissect the complexity of molecular regulation by miR-132 in lymphatic biology, my collaborators and I identified miR-132 target mRNAs in LECs, using high-throughput sequencing after RNA-protein cross-linking and immunoprecipitation of the Argonaute protein (Argonaute HITS-CLIP), which led us to define the miR-132-mRNA interactome in LECs. We found that this microRNA in LECs is involved in the control of many different molecular pathways mainly involved in cell proliferation and regulation of the extracellular matrix and cell-cell junctions. It is logical that miR-132 regulates such pathways given they are involved in the processes of LEC proliferation and tube formation, which I showed are dependent on miR-132 in my in vitro studies. Finally, I demonstrated that inhibiting the effects of miR-132 in a mouse ear model of lymphangiogenesis, using an antagomiR inhibitor of miR-132 coupled to cholesterol, blocked the complex remodelling of lymphatic vessels stimulated by VEGFC, in vivo. It was noteworthy that all aspects of lymphatic remodelling induced by VEGFC were restricted by inhibition of miR-132, including the enlargement, branching and sprouting of lymphatic vessels. Thus the inhibitory effect of targeting this microRNA on lymphangiogenesis and lymphatic remodelling can be considered comprehensive. The research described in this thesis identified miR-132 as a critical regulator of lymphangiogenesis and lymphatic remodelling, and delineated molecular mechanisms by which this microRNA influences these important biological processes. This work also identified new molecular pathways which are involved in modifying the lymphatic vasculature in response to key lymphangiogenic growth factors. In-so-doing, these studies identified potential therapeutic targets for drugs designed to block the growth and remodelling of tumour lymphatics, and thereby restrict the metastatic spread of cancer.

Global initiatives applying next generation sequencing to thousands of primary and metastatic tumour samples have detailed the genomic landscape of breast cancer. These unique insights have transformed our understanding of the biological processes underlying these tumours. Despite this, the clinical relevance of these findings remains poorly understood. This is because these datasets are frequently heterogeneous, include patients that may not have received contemporary treatments, often have limited clinical annotation, and commonly lack long-term follow-up for patient survival outcomes. In order to better assess the clinical relevance of genomic driver alterations in breast cancer, this research project performed genomic sequencing using primary tumour samples from patients enrolled in two pivotal phase 3 clinical trials – the Breast International Group 1-98 (BIG 1-98) study and the Suppression of Ovarian Function Trial (SOFT). This provided a well-defined population with hormone receptor-positive, human epidermal growth factor receptor (HER) 2-negative early breast cancer in both postmenopausal and premenopausal populations, as well as long-term survival outcome data collection. Crucially, this allowed for further investigation of late distant recurrence, a unique feature of hormone receptor-positive, HER2-negative breast cancers, as well as the ability to focus on distinct high-risk subpopulations, such as breast cancer arising in young women. In the postmenopausal BIG 1-98 study, comprehensive panel sequencing was performed using 538 tumour samples from patients with hormone receptor-positive, HER2-negative early breast cancer. PIK3CA mutations were the most common driver (49%) and were associated with a reduction in the risk of distant recurrence. Contrastingly, TP53 mutations, amplifications on 11q13 and 8p11, and increasing number of driver alterations were associated with an increase in the risk of distant recurrence. Additionally, PIK3CA mutations were predictive of a greater magnitude of benefit from letrozole compared with tamoxifen. Using the same dataset, we further investigated the association of oncogenic drivers with late distant recurrence (5 years or greater from the time of randomisation). PIK3CA mutations were associated with reduced risk of late distant recurrence, whereas amplifications on 8p11 and BRCA2 mutations were associated with increased risk of late distant recurrence. In the premenopausal SOFT study, comprehensive panel sequencing was performed using 1258 tumour samples from the entire study cohort, and whole exome sequencing was performed using 82 tumour samples from very young patients with high-risk, hormone receptor-positive, HER2-negative early breast cancers. Subgroups of tumours with distinct copy number-amplifications had a higher risk of distant recurrence than tumours that were amplification-devoid, validating previously reported data. Importantly, the study further defined unique molecular characteristics that were enriched in very young women and may explain their poor prognosis. This includes subgroups of tumours with homologous recombination deficiency, a high-risk PIK3CA mutated subgroup, amplification-enriched subsets, and a low estrogen receptor expressing subtype with immune infiltration. Taken together, these results highlight the clinical relevance of genomic sequencing of hormone receptor-positive, HER2-negative early breast cancers. They provide a new prognostic framework using genomic stratification, and highlight molecular targets that should be prioritised for future clinical trial research.