Vorschau

PDF, Englisch Download (4MB) | Nutzungsbedingungen

Abstract

Gliomas are the most common paediatric brain tumours, accounting for about 50% of all brain tumours in children. They are typically classified by the putative cell type they arise from based on morphology, the location they are found and malignancy grade. The most common glioma found in adults is glioblastoma, which is a WHO grade IV tumour with an average survival of less than 15 months. Extensive work has gone into characterising, modelling and treating this tumour in vivo and in vitro, but it is now clear that glioblastoma in children is biologically very different from that in adults. Paediatric high-grade gliomas do, however, share their aggressiveness with their adult counterparts, with few patients achieving long-term survival - dictating an urgency to find more precise therapies for these patients. Additionally, relatively little work has focused on low-grade gliomas to date. Nevertheless, these tumours also deserve a lot of attention, because low-grade gliomas make up 30-50% of all paediatric brain tumours and more than half of all paediatric gliomas. Although the tumour itself does not necessarily lead to a massively reduced life span and should rather be considered a chronic disease, the impact on the patients’ and families’ lives caused by the therapy load and possible recurrences remains a major clinical burden. Unlike high-grade gliomas, which are very heterogeneous with multiple oncogenic drivers, most low-grade gliomas are driven by alterations in the mitogen-activated protein kinase (MAPK) pathway through different mechanisms. Examples are BRAF alterations, FGFR1 mutations, and NTRK or MYB fusions. After analysing around 200 paediatric glioma samples by whole-genome/whole-exome and RNA sequencing within the ICGC PedBrain Tumour Project and the INFORM personalised medicine study, fusions involving an additional candidate gene (ALK) were found to be of interest in terms of their pattern of occurrence and availability of targeted inhibitors. Clinically these fusions arose in an interesting patient population, affecting infants (<2 years old) with histologically malignant tumours that however showed outcomes more similar to low-grade glioma than glioblastoma. PPP1CB:ALK was the most common gene fusion found in this context, which was then modelled in vivo using two different state-of-the-art methodologies: in utero electroporation and p0 injection using the RCAS-tva system. Through in utero electroporation, a novel mouse model was generated that nicely recapitulates the human tumours. Different ALK-specific inhibitors already used in clinical trials for other tumours, like non-small cell lung cancer or neuroblastoma, were tested on the tumour cells in an in vitro sphere culture setting and in addition also in vivo on allografted tumour cells. The results show a promising effect of the third-line, blood-brain-barrier penetrant ALK inhibitor lorlatinib, with IC50 values below 1.3nM and a significant increase in lifespan with a decrease in tumour signal, respectively. Thus, the project led from the genomic discovery of a novel driving event in paediatric glioma through to its modelling and identification of a promising new option for therapy. Further proof-of-concept application with other oncogenic combinations leads to the conclusion that the mouse model strategy and the methodology behind this can be further applied to test other candidate genes and specific inhibitor therapies. The overall aim is thus to accelerate the approval of targeted drugs by authorities after running a stratified clinical trial on a small infant patient population carrying the gene of interest, to enable patients to get the most potent therapy with the fewest side effects.