Experimental Therapeutics & Molecular Oncology 

Contact: Professor Michelle Haber AM, MHaber@ccia.org.au; Professor Murray Norris AM, mnorris@ccia.org.au; Dr Jayne Murray, JMurray@ccia.org.au

Neuroblastoma is the quintessential Myc-driven cancer. Patients often present with high-level amplification (>100-fold) of the MYCN (N-myc) oncogene in their tumours, which in turn confers a particularly poor prognosis. We and others have shown that high expression of the MYCN oncoprotein is an important causative factor in neuroblastoma. The MYCN oncogene regulates a large number of genes that influence neuroblastoma growth and development. One of the best characterised is the ornithine decarboxylase (ODC) gene which codes for the rate-limiting enzyme in polyamine synthesis.

High polyamine levels are essential for neuroblastoma initiation and maintenance, and high ODC gene expression is associated with rapid cell proliferation and malignant transformation. Our group has shown that depletion of polyamines by treatment with the ODC inhibitor difluoromethylornithine (DFMO) is sufficient to inhibit tumour formation in the TH-MYCN transgenic mouse model of neuroblastoma. More importantly, DFMO given in combination with current chemotherapeutic drugs produces prolonged tumour-free survival, with no overt toxicity in either TH-MYCN or human xenograft mouse models.

Based on this research, a joint Australia–US Phase I/II clinical trial using a combination of DFMO and standard chemotherapy for relapsed neuroblastoma patients was undertaken. A number of patients who completed the study had a positive response, with their tumours shrinking, leading to extended survival. This clinical trial provided strong evidence supporting polyamine depletion, in combination with conventional chemotherapy, as a powerful therapeutic strategy and triggered a Phase 2 clinical trial by the Children’s Oncology Group (COG, USA), which is ongoing.

While some positive results were achieved by inclusion of DFMO in the treatment scheme for neuroblastoma, the drug was not effective in all cases. We have therefore continued to work to further enhance polyamine inhibition therapy. We have shown that while DFMO prevents neuroblastoma and leukaemia cells from creating their own polyamines, it simultaneously increases the ability of the cells to take up polyamines from their surroundings by upregulating the expression of polyamine transporters, including SLC3A2 and the P5B-type ATPase ATP13A3, thus mediating resistance to DFMO.  Based on these findings, we combined DFMO with the polyamine transport inhibitor, AMXT 1501, and demonstrated that polyamine inhibition using DFMO/AMXT1501, which simultaneously inhibits polyamine synthesis and transport, is a highly effective treatment strategy in laboratory models of both neuroblastoma and the deadly brain cancer, DIPG.

With the combination of DFMO/AMXT1501 having entered adult Phase 1 clinical trials, we are now working closely with our clinical colleague, Prof David Ziegler, and Dr Mark Burns, the chemist who developed AMXT1501, in leading an international consortium involving the UK, USA and Australia, to develop a follow-up clinical trial using DFMO/AMXT-1501 in combination with standard-of-care chemotherapy/immunotherapy for refractory neuroblastoma patients. Current research focuses on identifying markers of response to DFMO/AMXT 1501, optimising the scheduling of DFMO/AMXT1501 in preparation for the clinical trial, characterising mechanisms of resistance to polyamine depletion, and elucidating the mechanisms of polyamine transport in cancer cells to allow the development of targeted inhibitors.

Contact: Professor Michelle Haber AM, mhaber@ccia.org.au;  Professor Murray Norris AM, mnorris@ccia.org.au; Klaartje Somers, ksomers@ccia.org.au

The strands of DNA inside each cell are wrapped around proteins called histones, to form tightly packed structures also known as nucleosomes. Histone chaperone proteins unpack these structures to facilitate DNA replication, transcription and repair through direct, dynamic interaction with histone proteins. We identified a Myc target gene called FACT (Facilitates chromatin transcription), for which elevated expression  is highly predictive of poor prognosis in children with neuroblastoma.

Our work has shown that FACT and MYCN expression have a strong dependent relationship, functioning in a feedback loop that forces very high expression of MYCN — beyond that achieved even by MYCN amplification in neuroblastoma cells.

We also found that a chemical inhibitor of FACT called CBL0137 profoundly inhibits cancer progression in a neuroblastoma model. CBL0137 also synergises with existing chemotherapeutic drugs used to treat neuroblastoma through blocking DNA repair.

Based on our findings, which we have extended to brain tumours and leukaemias, a joint Australian−US Phase I trial has opened for all refractory childhood solid cancers, including neuroblastoma. Using our in vitro and in vivo models, we plan to determine the best CBL0137 combination treatment for a subsequent Phase II combination trial, and the best timing for delivering this treatment.

Contact: Professor Murray Norris AM, mnorris@ccia.org.au; Professor Michelle Haber AM, mhaber@ccia.org.au; Dr Jayne Murray, JMurray@ccia.org.au

Many children with neuroblastoma present with widely disseminated disease at diagnosis and the prognosis in these cases is dismal. Several prognostic markers have been identified for this disease and one of the most powerful is MYCN oncogene amplification, demonstrated in 25–30% of primary untreated neuroblastomas. We have sought to identify pathways downstream of MYCN that are required for neuroblastoma initiation and maintenance, as these represent potential candidates for therapeutic intervention.

Using a mouse ENU mutagenesis genetic screen and whole genome sequencing, we have identified a specific gene, RUNX1T1, with a single point mutation that prevents neuroblastoma tumorigenesis in Th-MYCN transgenic mice. Our results show that MYCN does not transcriptionally regulate RUNX1T1, but rather drives increased protein translation in MYCN-amplified tumours. In addition, we have shown that loss of this protein through mutation or gene knockout prevents tumorigenesis by reversal of neuroblast hyperplasia (increased cell number) in mouse ganglia, which we have previously shown is a prerequisite for neuroblastoma initiation.

RUNX1T1 forms part of a transcriptional repressive complex recruited by HAND2 to regulate chromatin accessibility, controlling neuron-specific pathway genes and maintaining an undifferentiated state in MYCN-amplified neuroblastoma cells. Our research suggests that inhibiting RUNX1T1 can lead to otherwise immune ‘cold’ embryonal cancers becoming visible to the immune system and susceptible to immunotherapeutics. We are now working on developing inhibitors targeting this protein that can be further developed as therapeutic or preventative drugs in MYCN-driven cancers. These studies have the potential to elucidate an entirely novel approach to the treatment and, ultimately, prevention of this refractory childhood malignancy.

Contacts: 

Associate Professor Jamie Fletcher, JFletcher@ccia.org.au; Dr Alvin Kamili, AKamili@ccia.org.au

Despite intensive and prolonged treatment, half of all children diagnosed with high-risk neuroblastoma either do not respond, or subsequently relapse. There are no curative therapies for these patients. Current approaches are clearly inadequate for these children, and experimental therapies are being introduced too late. This study aims to change this paradigm through comprehensive pre-clinical evidence supporting earlier intervention with targeted agents.

Children diagnosed with high-risk neuroblastoma (HR-NB) undergo intensive and prolonged treatment with conventional high-dose, multi-agent chemotherapy, surgery, radiation therapy and immunotherapy. However, nearly 50% have refractory disease or experience relapse, and despite numerous early phase clinical trials, their outcomes remain especially poor, with survival <10%. Recurrent mutations in a restricted number of survival pathways occur in >50% relapsed or refractory HR-NB, suggesting that activation of these pathways allows tumour cells to persist through conventional chemotherapy. We propose that combination treatment targeting these pathways would be more effective than conventional therapy at diagnosis.

We seek to: 1) understand the relevance of drug resistance mechanisms for response to chemotherapy, including the role of multidrug transporters; 2) develop new combinations that target resistance mechanisms, and 3) assess the potential for pre-emptive treatment at diagnosis, based on knowledge of conserved pathway activation at relapse.

Contacts: Associate Professor Jamie Fletcher, JFletcher@ccia.org.au; Dr Alvin Kamili, AKamili@ccia.org.au; Dr Klaartje Somers, KSomers@ccia.org.au; Dr Mawar Karsa, MKarsa@ccia.org.au

Preclinical testing of new agents and drug combinations requires realistic models representing the heterogeneity of the disease and both diagnosis and relapse models. Success in this area also requires more efficient development of models, particularly for personalised medicine applications.

High quality preclinical testing is essential to prioritize agents for early phase trials. We have currently built up a library of over 40 high-risk neuroblastoma patient-derived xenograft models and are seeking to expand and characterise these models further. In addition, we are developing immune-competent models for paediatric cancers to allow efficacy studies with immunotherapies and agents that exert their anticancer effects through modulating anti-tumour immunity.

Contacts:

Dr Klaartje Somers, ksomers@ccia.org.au; Dr Mawar Karsa, mkarsa@ccia.org.au

Leukaemia accounts for the second greatest number of deaths from childhood cancer after brain cancer. Particularly poor survival rates are found in infants whose leukaemias display abnormalities of the MLL gene (MLLr leukaemia), relapsed patients, and children diagnosed with T-ALL or AML.

We have established a pipeline for the development and characterisation of novel agents for acute leukaemia, including high-throughput screening, cellular and molecular characterisation in a large and diverse panel of leukaemia cell lines, ex vivo and in vivo testing in paediatric leukaemia patient-derived xenograft (PDX or avatar) models, and efficacy testing of potential new agents in combination with the leukaemia drugs currently in use.

One of the strategies under investigation is focused on inhibiting production of nicotinamide adenine dinucleotide (NAD, a compound essential for energy production in cancer cells), using the small molecule OT-82, an exciting new anti-cancer drug which we have shown to be extremely potent in PDX models of very poor outcome paediatric leukaemias including MLLr leukaemia, T-ALL and AML.  Moreover, reduction of intracellular levels of polyamines (which are essential for cancer cell growth and division) by combining the polyamine synthesis inhibitor DFMO with the polyamine uptake inhibitor AMXT1501, has shown promise in inhibiting progression of MLLr leukaemia in laboratory models of this disease. Combining these metabolism-targeting agents with an exciting new class of drugs known as BH3 mimetics, exemplified by the new clinically approved drug, venetoclax, shows promising effects in several animal models for poor outcome high-risk paediatric cancers. These combinations  are under further investigation for progression into the clinic, with studies aimed at identifying markers of response.

Contacts: Dr Klaartje Somers, ksomers@ccia.org.au; Dr Mawar Karsa, mkarsa@ccia.org.au

It is now recognized that the tumour microenvironment (TME), involving immune cells that can modulate the response of the body to a tumour, plays a crucial role in adult solid tumour development, progression and response to current treatments, thus heavily impacting outcome. Metabolic cross-talk between tumour cells and associated immune cells has been shown to be a key factor in promoting tumour progression. This knowledge has resulted in the realisation that novel therapeutic strategies should encompass methods to target the tumour-promoting processes mediated by the TME, and should not just focus on inhibiting the cancer cells. However, whether and how the TME contributes to the progression of paediatric cancers is less well understood.

This project aims to provide an in-depth characterisation of the TME and resident tumour-promoting immune cells in animal models of childhood cancer and patient samples by utilising novel high-throughput spatial imaging technologies.We have previously shown that several novel anti-cancer agents currently moving into clinical trial for children (e.g. CBL0137, DFMO/AMXT 1501) are able to re-program the TME of paediatric cancers to make them more susceptible to both intrinsic and external immune modulating processes or agents, hence enhancing the therapeutic outcome. We are now assessing the potential of these agents to potentiate immunotherapies, including CAR-T cells, for the treatment of paediatric cancers.

Contacts: 

Professor Michelle Haber AM, MHaber@ccia.org.au; Professor Murray Norris AM, mnorris@ccia.org.au; Associate Professor Jamie Fletcher, JFletcher@ccia.org.au; Dr Jayne Murray, JMurray@ccia.org.au

The high genetic and epigenetic plasticity of tumour cells as tumours grow and progress is known as “tumour adaptation”, and although preventing tumour cell adaptation is key to better outcomes, the mechanisms of adaptation are currently undruggable. Exciting new discoveries show that adaptation can be driven by re-activation of retrotransposons, ancient virus-like genetic elements in our DNA. In tumours, mechanisms keeping retrotransposons inactive become dysfunctional, allowing them to replicate by reverse transcription, enter new sites in the tumour DNA, and drive adaptation.

These discoveries have striking clinical implications. Rapid adaptation driven by retrotransposon expansion can be prevented by inhibiting reverse transcription, and this can be achieved with existing antiviral drugs developed to treat HIV or hepatitis B. We have already confirmed that antiviral drugs can prolong survival in mouse cancer models after chemotherapy.

We intend to progress toward clinical application by demonstrating retrotransposon activity in human neuroblastomas, developing approaches to detect this activity in neuroblastoma patients, and identifying the optimal antiviral from among those.