Molecular Diagnostics

This program is utilising molecular genetic techniques to improve the diagnosis and risk classification of childhood cancers including leukaemia and neuroblastoma. Major areas within the program include molecular detection of residual disease following chemotherapy and the use of functional genomics and high-throughput screening strategies to detect novel molecular targets and relevant inhibitors.

    • ABCs in adult cancers


      Amplification of the MYCN oncogene is now a well-known prognostic indicator in childhood neuroblastoma patients. The closely related c-myc oncogene is commonly amplified and/or over-expressed in a number of adult cancer types; however, the prognostic significance of c-myc aberrations in such cancers is not well-established.

      We have previously shown that high level expression of ABCC1/MRP1 and ABCC4/MRP4 are independent, powerful predictors of poor outcome in neuroblastoma and we have recently reported that the oncogenes MYCN and c-myc drive expression of ABCC1/MRP1 and ABCC4/MRP4 to high levels in cancer cells. Since c-myc is known to be amplified and/or dysregulated in a number of adult cancers including ovarian, colon, lung, prostate, breast and melanoma, the aim of this project is to examine the level of c-myc gene amplification and expression of c-myc and ABCC/MRP family genes in these diseases and correlate these parameters with clinical outcome.

      We have obtained high quality RNA specimens from cohorts of several cancers known for frequent c-myc aberration, including cancers of the lung, ovary and colon and we are currently investigating the relationship between ABCC/MRP transporter gene expression and clinical outcome in each cohort. Also, we have found that both MYCN and c-myc regulate a much larger set of genes, closely related to the ABCC/MRP genes, which belong to the ABC transporter gene family.

      While several transporters of the ABC family are established as drug transporters, many also mediate transport of cellular substances that control important processes associated with tumour progression, independent of drug efflux. Therefore we are also examining expression of the wider family of ABC transporter genes in relation to clinical outcome in the c-myc driven cancers. Knowledge of molecular factors that may predict response to treatment is critical in order to develop new biomarkers and ultimately novel therapies. This study promises to yield powerful prognostic markers for this disease and to identify important targets for potential therapeutic intervention.

      External Collaborators: Anna DeFazio (Westmead Millennium Institute, NSW), David Bowtell (Peter MacCallum Cancer Centre, VIC), Georgia Chenevix-Trench (Queensland Institute for Medical Research, QLD), Robyn Ward (Prince of Wales Hospital, NSW), Kwun Fong (Prince Charles Hospital, QLD).

    • MRP4 as a target in neuroblastoma


      We have previously shown that the MYCN oncogene and the Multidrug Resistance-associated Protein 1 (MRP1/ABCC1) gene are both critically important in neuroblastoma and are powerful prognostic markers of poor outcome. As such, they represent valuable targets for the development of novel therapeutics through the generation of specific small-molecule inhibitors.

      We have recently demonstrated that another ABCC/MRP family member, ABCC4/MRP4, is also an extremely powerful independent predictor of neuroblastoma outcome. Although this drug transporter is able to expel chemotherapeutics from tumour cells, our data suggest that ABCC4/MRP4 makes a fundamental contribution to the malignant phenotype of neuroblastoma independent of drug exposure. We are currently investigating in more detail precisely how ABCC4/MRP4 contributes to highly malignant neuroblastoma using a range of experimental approaches.

      Since siRNA-mediated silencing of ABCC4/MRP4 dampens neuroblastoma cell proliferation and induces a more differentiated cellular morphology, we expect that therapeutic targeting of ABCC4/MRP4 will be of clinical benefit for this disease, as well as for other cancers expressing high ABCC4/MRP4 levels. Therefore we are using high-throughput screening of chemical libraries to generate novel inhibitors of ABCC4/MRP4.

      External collaborators: Andrei Gudkov (Roswell Park Cancer Institute, USA), Catherine Burkhart (Cleveland BioLabs Inc, USA).

    • Improving treatment for leukaemia


      Gene-targeted therapy for infant ALL

      Despite the success in treating childhood cancer, leukaemia remains one of the most common causes of death from disease in children, and accounts for the greatest number of deaths from childhood cancer overall. Particularly poor survival rates are found in both children and adults whose leukaemias display abnormalities of the MLL gene. Upon diagnosis of leukaemia with MLL abnormality, the high-risk classification of their disease currently leaves these patients with no alternative but to endure intensified treatment. Unfortunately, although MLL-leukaemia patients show an initial response, almost half of the patients will eventually succumb to their disease or die from side effects associated with the highly intense treatment that they receive. Hence there is a desperate need to develop less intensive, more specific treatments for this disease.

      Our goal has been to identify new molecules which specifically target the abnormal MLL gene, and thereby develop a successful drug that will only kill the cancer cells, leaving developing organs of the young children who are diagnosed with this poor prognosis disease relatively unharmed. One of the most exciting new approaches to the identification of potential new pharmaceutical agents is the high-throughput screening of small-molecule chemical libraries. Such libraries are increasingly being used to identify new therapeutic compounds and a number of these novel compounds have shown great promise in the clinic for the treatment of adult cancer. Identification of molecular targets in cancer cells and the development of small-molecule inhibitors to these targets provides the opportunity to devise therapies that are specific in action, have an irreversible effect on cancer cells, and are effective at low concentrations. This research is being undertaken in collaboration with the Cleveland Clinic Foundation, Ohio USA and Cleveland Biolabs Inc.

      We have screened a 30,000-compound library of chemical small-molecules and isolated a number of compounds with specificity against infant leukaemia cells. One particularly exciting compound, named SM-7, is effective at killing leukaemia cells with MLL gene abnormality, while it will not harm normal cells, nor other leukaemia types with a normal MLL gene. SM-7 is now being intensively investigated to determine its precise mechanism of action and the core structure responsible for the differential effects observed. Further screening is also being undertaken to generate highly effective inhibitors with broader application to other childhood and adult leukaemias.

      The next steps in getting SM-7 to clinical trial have been to undertake optimisation for pre-clinical testing as well as further delineation of the anti-MLL activity of SM-7. To do this we are collaborating with Roswell Park Cancer Institute in the USA to undertake a medicinal chemistry program to synthesise additional candidate molecules that are structurally similar to SM-7, and hence are likely to have its anti-MLL activity but which have much more favourable properties than SM-7 as a candidate drug. As such, the medicinal chemistry program will deliver a number of optimised and improved anti-MLL drugs that will then each need to undergo rigorous preclinical testing, to determine which is the safest and most effective.

      In preparation for this stage, and in collaboration with Professor Richard Lock (Head, Leukaemia Biology Program), we have developed an experimental system that closely mimics the behaviour of human leukaemia in the child and which will be used to predict how effectively the new molecule(s) inhibit the growth of MLL-leukaemia cells in the patient.

      These chemical small-molecules offer a potentially exciting new treatment for infants with leukaemia because this molecular targeted approach is likely to be more specific and less toxic than conventional therapy. Successful completion of these studies will allow for the development of pharmaceuticals for eventual clinical trial.

      External collaborators: Andrei Gudkov (Roswell Park Cancer Institute, USA), Catherine Burkhart (Cleveland BioLabs Inc., USA), Olga Chernova (Cleveland BioLabs Inc., USA), Huw Davies (Emory University, USA).

      Mechanism of relapse in ALL

      Despite significant improvements in treatment regimes, relapse remains a barrier to survival in approximately 20 per cent of children suffering from acute lymphoblastic leukaemia (ALL). Advances in our understanding of relapse coupled with evaluation of new treatments in relevant pre-clinical models are critical for further improvements in outcome for these children.

      Relapse occurs as a result of small numbers of cancer cells surviving treatment. In the past, it has not been known whether the cells that cause relapse evolve from other leukaemia cells during treatment (i.e. mutate to become treatment-resistant), or whether the relapse-causing cells existed from the beginning.

      We recently made a breakthrough in our understanding of relapse when we used sensitive molecular techniques to analyse samples of bone marrow taken from children at the time of their diagnosis and compared these with samples taken from the same children at the time of relapse. We found that, in most cases, the type of leukaemia cells found at relapse (‘treatment-resistant’ clones) in fact already existed in exceedingly small numbers at the time of diagnosis, and were living in parallel with ‘treatment-sensitive’ clones. Our studies have also shown that the higher the number of treatment-resistant clones present at diagnosis, the faster relapse will occur.

      The clinical implications of these findings are disturbing. Using current treatment protocols, relapse appears inevitable for those children who have treatment-resistant leukaemia cells in their bone marrow at the time of diagnosis. This research highlights the need to identify such patients early during treatment, so that alternate treatments can be explored. To further characterise the cells responsible for relapse, we are carrying out experiments to expand the treatment-sensitive and treatment-resistant sub-populations of individual leukaemias. By studying the molecular differences between these sub-populations we aim to uncover the critical features of leukaemias that are refractory to treatment, thus highlighting potential new therapeutic avenues for high-risk leukaemia.

      External Collaborators: Ursula Kees (Telethon Institute for Child Health Research, WA), David Baker (Princess Margaret Hospital, WA), Luciano Dalla Pozza (The Children’s Hospital, Westmead, NSW).

    • Genetic suppressors of neuroblastoma


      The majority of patients with neuroblastoma present with widely disseminated disease at diagnosis and despite highly intensive treatment regimens, the prognosis for such patients is dismal. A number of prognostic markers have been identified for this disease and one of the most powerful is MYCN oncogene amplification, which can be demonstrated in 25–30 per cent of primary untreated neuroblastomas.

      Numerous reports have confirmed the association between MYCN amplification and rapid tumour progression, advanced clinical stage and poor outcome. MYCN regulates numerous genes mediating cell behaviour; however those genes necessary or sufficient for tumourigenesis have remained elusive.

      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.

      N-ethyl-N-nitrosurea (ENU) was discovered in the 1970s to be a powerful compound that causes mutations in DNA. ENU works by creating DNA adducts that cause heritable mutations during DNA replication and mutations occur at a rate of approximately one base every 1000 bases. ENU screening processes employ a phenotype-driven methodology that assumes no bias towards any gene or pathway, and as such are a powerful tool for elucidating new molecules that are important to the disease of interest.

      In collaboration with the Walter and Eliza Hall Institute (Victoria), we have undertaken a large ENU mutagenesis screen in order to identify co-factors that are responsible for mediating the oncogenic effects of MYCN in neuroblastoma tumourigenesis. We have generated a founder line with a significant heritable increase in tumour latency and we will use this, and other lines showing heritable delays in tumorigenesis, to identify the responsible genes by a combination of mapping and next-generation sequencing.

      This forward genetic screen offers the potential to discover critical cellular genes whose loss of function results in a reduction of neuroblastoma tumourigenesis, and as such may provide novel pathways for therapeutic targeting in this disease.

      External Collaborators: Doug Hilton and Ben Kile (Walter and Eliza Hall Institute, VIC).

Staff List


Professor Murray Norris


Dr Chengyuan Xue

Dr Ruby Pandher

Dr Emanuele Valli

Dr Mawar Karsa


Dr Jayne Murray