Page 177 - Molecular features of low-grade developmental brain tumours

General discussion
Low grade gliomas are the most frequent CNS tumours in children and are considered WHO grades I/II 1,2. These tumours have a broad histological spectrum and classification based on histological criteria alone is difficult. An integrated diagnosis combining both histological and molecular features based on (epi)genetic and genomic hallmarks is needed. SEGAs are benign slow growing glioneuronal tumours accounting for ~2% of pediatric brain tumours and occur almost exclusively in in children and adolescents with TSC. Although there is evidence of second hit mutations in the TSC genes additional genetic events might be involved in the growth and progression of SEGAs. The aim of this thesis was to investigate the molecular mechanisms involved in SEGA development and growth on a (epi)genomic, transcriptomic and proteomic level and to investigate the molecular features of several GNTs.
Germline and somatic mutations in SEGA
TSC is a genetic disorder caused by loss of function mutations in TSC1 on chromosome 9q34 (encoding for hamartin) or TSC2 on chromosome 16p1 (encoding for tuberin) resulting in constitutive activation of the mTOR pathway 3,4. One third of TSC germline mutations are familial, whereas two-thirds are sporadic, de novo mutations 5,6. There is evidence that TSC follows the classic Knudson model, where second hit mutations in the same gene as the germline can be seen, resulting in complete loss of either TSC1 or TSC2 7-11. TSC lesions, such as renal angiomyolipomas and lymphangioleiomyomatosis generally have loss of heterozygosity (LOH) 12-15. Indeed, animal models with biallelic TSC mutations can recapitulate features of TSC, including SEN/SEGA-like lesions 16-21. However, LOH events seem to be more rare in TSC brain lesions and in SEGAs subsequent studies have produced contradictory results on the presence of second hit mutations 14,22-25. In chapter 2, we analysed 34 SEGA samples using massively parallel sequencing for evidence of LOH and identified ~80% of samples with copy neutral LOH (9/10 TSC1 mutated 14/19 TSC2 mutated). In accordance, more recent studies found copy neutral LOH in 76-85% of SEGA samples, suggesting that indeed the majority of SEGAs have TSC1/TSC2 biallelic inactivation 26,27. However, in ~20% of SEGAs second hit mutations are not found, indicating that either the second hit event is under detection level, or that these tumours only have a mono-allelic TSC mutation. Moreover, TSC1/TSC2 biallelic inactivation might already be present in SEN, suggesting that additional molecular mechanisms may play a role in SEGA progression and growth.
Approximately, 60% of patients have a TSC2 mutation, 30% of patients have a TSC1 mutation, while in 10-15% of TSC patients no mutation cannot be identified 28,29. In chapter 2, we found 19/34 (56%) SEGAs TSC2 mutated, 10/34 (29%) TSC1 mutated and 5/34 (15%) with no mutation identified, which is in line with previous research. TSC2 mutations are predictive of severe clinical manifestations in TSC infants including increased SEN/SEGA volume 30-33. In chapters 3 and 4 we investigated the epigenetic and transcriptomic profile of SEGAs and identified that methylation and gene expression changes in SEGA appear to be independent of the TSC1/TSC2 mutation. Although the overall difference in gene expression between TSC1/ TSC2 mutated SEGAs was low, more genes were differentially expressed in TSC2 compared to
GENERAL DISCUSSION
175
7