Cancer patients' outcome prediction: novel techniques for estimating copy number alterations in childhood brain tumours using DNA methylation arrays.

Medulloblastoma is the most common malignant childhood brain tumour, comprising four risk groups (very high-, high-, low- and standard-risk). Survivors face debilitating late effects as a consequence of their treatment. Improved pre-treatment stratification could reduce late effects in low-risk patients by reducing their treatment intensity and reduce deaths in high-risk patients by intensifying treatment. Moreover, over 20% of patients currently defined as standard-risk still die of their disease in the absence of recognised high-risk markers. Therefore, it is important to characterise novel biomarkers that can better stratify this group of patients. Changes in chromosomal copy number (CN) have shown potential for risk stratification in cancer; chromosomal CN changes can result in oncogene gain and tumour suppressor gene loss, which are a mechanism for tumorigenesis.