New insights into the genetics of glioblastoma multiforme by familial exome sequencing

// Christina Backes 1,* , Christian Harz 2 , Ulrike Fischer 2 , Jana Schmitt 2 , Nicole Ludwig 2 , Britt-Sabina Petersen 3 , Sabine C. Mueller 1,2 , Yoo-Jin Kim 4 , Nadine M. Wolf 5 , Hugo A. Katus 5 , Benjamin Meder 5 , Rhoikos Furtwangler 6 , Andre Franke 3 , Rainer Bohle 4 , Wolfram Henn 2 , Norbert Graf 6 , Andreas Keller 1 and Eckart Meese 2 1 Clinical Bioinformatics, University of Saarland, Saarbrucken, Germany 2 Institute of Human Genetics, University of Saarland, Medical School, Homburg, Germany 3 Institute of Clinical Molecular Biology, Christian-Albrechts-University Kiel, Haus Niemannsweg, Kiel, Germany 4 Department of Pathology, University of Saarland, Medical School, Building, Homburg, Germany 5 Department of Internal Medicine III, University of Heidelberg, Heidelberg, Germany 6 Pediatric Hematology and Oncology, University of Saarland, Medical School, Homburg, Germany Correspondence: Christina Backes, email: // Keywords : glioblastoma multiforme, next generation sequencing, bioinformatics Received : October 02, 2014 Accepted : December 09, 2014 Published : December 10, 2014 Abstract Glioblastoma multiforme (GBM) is the most aggressive and malignant subtype of human brain tumors. While a family clustering of GBM has long been acknowledged, relevant hereditary factors still remained elusive. Exome sequencing of families offers the option to discover respective genetic factors. We sequenced blood samples of one of the rare affected families: while both parents were healthy, both children were diagnosed with GBM. We report 85 homozygous non-synonymous single nucleotide variations (SNVs) in both siblings that were heterozygous in the parents. Beyond known key players for GBM such as ERBB2, PMS2, or CHI3L1, we identified over 50 genes that have not been associated to GBM so far. We also discovered three accumulative effects potentially adding to the tumorigenesis in the siblings: a clustering of multiple variants in single genes (e.g. PTPRB, CROCC), the aggregation of affected genes on specific molecular pathways (e.g. Focal adhesion or ECM receptor interaction) and genomic proximity (e.g. chr22.q12.2, chr1.p36.33). We found a striking accumulation of SNVs in specific genes for the daughter, who developed not only a GBM at the age of 12 years but was subsequently diagnosed with a pilocytic astrocytoma, a common acute lymphatic leukemia and a diffuse pontine glioma. The reported variants underline the relevance of genetic predisposition and cancer development in this family and demonstrate that GBM has a complex and heterogeneous genetic background. Sequencing of other affected families will help to further narrow down the driving genetic causes for this disease.