Improving the selection of brain tumours therapies

Background

Paediatric brain tumours that include ependymoma, medulloblastoma, and choroid plexus carcinoma (CPC) are the most common and lethal solid cancers to affect children. Even though these tumours are treated with aggressive surgery, radiation and chemotherapy, they fail to cure 30-50% of children. Importantly, ependymoma and CPC are insensitive to current chemotherapy; therefore there is a great need for effective new treatments.

 We and others have shown that childhood brain tumours comprise several distinct subtypes with different cellular origins and prognoses, suggesting that each will require different treatments. It has been demonstrated that accurate murine models can be used to develop new treatments that can be effectively translated to treat patients.

Ependymoma: Using a cross-species genomics approach, our lab has constructed the first models of ependymoma^1,2 and used these to conduct the first ever integrated high-throughput drug screen (HTDS) of a paediatric brain tumour and translate these findings into a clinical trial in children^3,4.

Medulloblastoma: While several groups have generated murine models of SHH and Group 3 type medulloblastoma, until recently there were no models of WNT-subtype. Since WNT-medulloblastoma is among the most curable brain tumours, murine models of this disease could be used to better understand brain tumour treatment sensitivity and how to reduce damaging radiation therapy in children. Using the cross-species genomics approach, we have developed the first and only murine model of WNT-medulloblastoma and shown that these tumours arise from a distinct population of hindbrain neural stem cells⁵. We are now using this model to better understand the biology and treatment of this and other brain tumours.

CPC: Until recently there were no validated murine models of this tumour type and there are no effective chemotherapies for children with this disease. We have developed the first murine model of CPC that accurately recapitulates the histology, transcriptome and behaviour of the human disease⁶. We have used this model to complete a 1.4 million compound screen of existing drugs and new chemical entities (NCEs) that might serve as new treatments for CPC.

Current research

Utilising these accurate brain tumour murine models we are performing preclinical therapeutic experiments in order to identify novel therapeutics for the clinic (Figure 3). We are performing combinatorial preclinical trials in ependymoma, medulloblastoma and CPC to advance lead compounds and NCEs that we have identified from our HTDS to combination preclinical trials in our preclinical models.

Figure 3: Active compounds from high-throughput drug screens that cluster according to major therapeutic indication and mechanism of action.

Further reading

1. Johnson RA et al. Cross-species genomics matches driver mutations and cell compartments to model ependymoma. Nature 466: 632–636, 2010.
2. Mohankumar M, et al., An in vivo screen identifies ependymoma oncogenes and tumor suppressor genes. Nature Genetics 8:878-87, 2015.
3. Atkinson J, et al., An integrated in vitro and in vivo high-throughput screen identifies treatment leads for ependymoma. Cancer Cell 20:384-99, 2011.
4. Wright, KM, Phase I study of 5-fluorouracil in children and young adults with recurrent ependymoma. Neuro Oncol. 12:1620-7, 2015.
5. Gibson P et al. Subtypes of medulloblastoma have distinct developmental origins. Nature 468:1095-9, 2010.
6. Tong Y et al. Cross-species genomics identifies TAF12, NFYC and RAD54Las choroid plexus carcinoma oncogenes. Cancer Cell. 27:712-27, 2015.