Paediatrics (RCH) - Theses

Permanent URI for this collection

Search Results

Now showing 1 - 1 of 1
  • Item
    Thumbnail Image
    A tool for risk profiling and accurate prognostication in paediatric glioma integrating clinical features with epigenetics: it is time to move on from the binary classification
    Dodgshun, Andrew John ( 2016)
    Paediatric glioma, the most common group of brain tumours in children, encompasses a wide range of entities with highly variable prognoses. Gliomas are grouped by histopathological features into high and low grade glioma but this classification does not take into account many established and emerging risk factors in this disease. Research into the molecular features of these lesions has shown that histology does not always correlate with biology and, where they differ, molecular features are usually superior at predicting outcome. Risk classifications have been developed for other paediatric malignancies which combine clinical, radiological, pathological and molecular factors to predict disease risk and prognosis. A comprehensive risk classification has not been published for paediatric glioma despite many risk factors being established in this disease. Using a clinical cohort of all paediatric glioma treated at a single institution (Royal Children’s Hospital, Melbourne, Australia) over an 18 year period a database was developed incorporating clinical, radiological, pathological and treatment factors. Where sufficient tumour tissue was available genome-wide methylation analysis was performed. The results of this were processed and evaluated by an established cluster analysis algorithm. Breaking the cohort into clinically appropriate subgroups, risk factors for disease progression and death were determined and prognosis estimated for distinct groups. A tool for robust risk profiling and prognostication was established with 5 main risk groups and 10 subgroups. Patients in the very low risk group have a predicted 100% overall survival and the majority require no treatment. In contrast there was a group of patients with 100% mortality within a short space of time where attempts at curative therapy are futile and may be deleterious to quality of life. The intermediate risk group contained a number of entities with a prognosis genuinely falling between that of low grade glioma and that of high grade glioma. Histopathological diagnosis retained prognostic importance for some, but not all, groups and methylation analysis was shown to have a significant role to play, particularly in high grade lesions or where diagnosis was unclear. This is likely to become part of routine care in the future and its place is strongly supported by the analysis presented here. Risk profiling is possible in paediatric glioma with far more accuracy than the current binary classification provides. A combination of established clinical factors and emerging molecular features provides an accurate and nuanced assessment of risk and prognosis.