A review by Australian experts in childhood brain cancer research has highlighted important gaps in our understanding of how brain cancers develop in children and how addressing these knowledge gaps could lead to new therapies to treat or even prevent such cancers in the future.
According to the review — authored by researchers at Children’s Cancer Institute, and one of a special collection titled ‘Origins of Childhood Cancer & Therapeutic Opportunities’ published in the journal Cancer and Metastasis Reviews in September — childhood brain cancer is a developmental disease, and understanding what happens at a molecular level as the brain is developing is key.
‘Brain cancers in children are often driven by what we call epigenetic factors,’ explains Associate Professor Fa Valdes Mora, leader of the Cancer Epigenetic Biology and Therapeutics Group at Children’s Cancer Institute. ‘People are familiar with the concept of genetics, but it is actually epigenetics that is responsible for how cells behave. To use a musical analogy, if DNA is the musical score, then epigenetics is the conductor. The conductor decides which part of the musical score is played, by which instruments, when, and how loudly. If something goes wrong with the conductor, the music is not going to be good.’
The review, which focuses on childhood brain cancers called gliomas — specifically, diffuse midline gliomas, diffuse hemispheric gliomas, and posterior fossa A ependymomas — points out that epigenetics plays a critical role in normal brain development. When things go wrong with epigenetics, cancer can result.
However, whether or not an epigenetic alteration leads to cancer is highly context dependent. Research has shown that for cancer to develop, the alteration must occur in a particular cell type in the brain (an immature cell state of cell differentiation), at a particular time point during brain development (a developmental window, pre or postnatally).
First author on the review, PhD student Afraah Cassim, explains: ‘It is known as the “three-event model” because it’s the combination of all three things that leads to tumourigenesis. For this reason, we called our review The Bermuda Triangle of paediatric brain cancers: epigenetics, developmental timing window and cell of origin. Like the real Bermuda Triangle, there is a very vulnerable state which can lead to things going badly wrong.’
The review points out that to fully understand the formation and progression of epigenetically-driven brain cancers, we need good laboratory models — patient-derived cell lines, mouse models and the like that can be used to study epigenomic alterations in different cell types at different time points in brain development.
‘It has been really hard to study these factors all together, and that is one of the key points we highlight in our review,’ commented Associate Professor Valdes Mora. ‘You need to take all three factors into consideration, according the three-event model, as studying them separately can lead to confusing or unexpected results. Fortunately, with new technology such as organoids becoming available, it is now possible to conduct these types of studies. We can simultaneously examine genes and epigenetics at the single-cell level. We have models to mimic neurodifferentiation in vitro, and we are actively working to establish these models in our lab.’
The researchers anticipate that studies using such models will assist with the development of new targeted therapies that, following successful clinical trials, may be used to treat children with brain cancer — the ultimate goal.
According to Associate Professor Valdes Mora, successfully treating aggressive brain cancers is likely to require combination treatment; that is, attacking the cancers from several different angles, looking at different causes, molecular drivers, and so on. However, she is optimistic about what may become possible with further research.
‘The use of this knowledge in preventing some brain cancers also holds great potential, and we hope to explore this area in the future.’
Read the full review here: https://link.springer.com/article/10.1007/s10555-025-10284-0
