Abstract
Glioblastoma (GBM) is the most common malignant primary tumor of the central nervous system. With no effective therapy, the prognosis for patients is terrible poor. It is highly heterogeneous and EGFR amplification is its most frequent molecular alteration. In this light, we aimed to examine the genetic heterogeneity of GBM and to correlate it with the clinical characteristics of the patients. For that purpose, we analyzed the status of EGFR and the somatic copy number alterations (CNAs) of a set of tumor suppressor genes and oncogenes. Thus, we found GBMs with high level of EGFR amplification, low level and with no EGFR amplification. Highly amplified tumors showed histological features of aggressiveness. Interestingly, accumulation of CNAs, as a measure of tumor mutational burden, was frequent and significantly associated to shortened survival. EGFR-amplified GBMs displayed both a higher number of concrete CNAs and a higher global tumor mutational burden than their no EGFR-amplified counterparts. In addition to genetic changes previously described in GBM, we found PARK2 and LARGE1 CNAs associated to EGFR amplification. The set of genes analyzed allowed us to explore relevant signaling pathways on GBM. Both PARK2 and LARGE1 are related to receptor tyrosine kinase/PI3K/PTEN/AKT/mTOR-signaling pathway. Finally, we found an association between the molecular pathways altered, EGFR amplification and a poor outcome. Our results underline the potential interest of categorizing GBM according to their EGFR amplification level and the usefulness of assessing the tumor mutational burden. These approaches would open new knowledge possibilities related to GBM biology and therapy.