Individualized treatment options for children with malignant brain cancer - Establishment of cellular, molecular and therapeutic tools

Project leader: Einar Osland Vik-Mo

Brain cancer is the most common and deadly form of solid tumor in children, and such cancers are responsible for the highest loss of quality-adjusted life years from any brain disease in the population. Seminal papers have identified subgroup characteristics and molecular aberrations for medulloblastoma (Pugh, Nature 2012), ATRT (Johann, Cancer Cell 2016) and high-grade gliomas (Mackay, Cancer Cell 2017). Through these detailed molecular analysis it is likely that all of the most important molecular changes found in these malignant pediatric tumors have been identified. Sadly, this molecular data has not made the therapeutic impact hoped. Clearly, we now can identify subgroups of patients with better prognosis, but for the most malignant tumor we have not been able to transform molecular knowledge into better treatments.

In principle, pediatric brain tumors should be ideal candidates for targeted therapies. These tumors have a lower number of mutations than many adult tumors (Alexandrov, Nature 2013). A lower mutational load should render the tumors more sensitive to targeted therapies. This would allow for therapies with a lower rate of off-target effects, and especially the avoidance of DNA-damaging drugs, which in surviving patients results in treatment induced neoplasia and developmental effects. However, efforts to use whole genome-, exome- or RNA sequencing can only identify druggable targets in a low frequency of patients (Letai, Nature Med 2017). This has been the findings both for pediatric tumors in general, and for pediatric brain tumors specifically (Reviewed in Mody, Ped Blood Cancer 2017). Most drugs have no identifiable biomarker, and even for drugs where such markers exist, they have been reported to give false positive and negative patient selection (Letai, Nature Med 2017). Even when a potential useful drug can be identified, the tumor clonal heterogeneity is likely to give any single drug only a limited effect. To target cancers effectively it is likely that more than one drug is needed. By using DNA/RNA sequencing the possibility of identifying more than one drug for any given patient is extremely low. No established strategy exist to use molecular data optimally in a clinical setting.

Department of Neurosurgery, Oslo University Hospital (OUH) operates 30 tumor cases each year. This high volume of patients will allow for building the know-how on cultivating cells from pediatric tumors. As some of these tumors could proliferate slow in vitro, it is imperative to collect as high a cell yield as possible. We have focused on this over more than a decade, and is thus very well positioned to perform this study. In addition – the establishment of pediatric brain tumor cell lines will be a part of the effort of generating a National Childhood Cancer Biobank. Our aim is to establish at least twelve cell lines from different pediatric brain tumors over the project period. These cell cultures will be validated according to state-of-the-art cellu-lar and molecular tools, including tumorigenicity in vivo (Fig. 1). Establishment of pediatric brain tumor cultures will allow for a strengthening of Norwegian research on this disease and support further international collaborations.