Fortunately for the neuro-oncology community, glioblastoma represented the first human cancer selected as a model to study in the TCGA. The first manuscript describing the molecular characterization resulting from this effort was published in 2008 [1]. In this interim analysis, DNA copy number, gene expression, and DNA methylation were studied in a set of 206 glioblastomas, as well as selected gene sequencing (~600 genes) in a subset of tumors. Many genes previously known to be relevant to gliomagenesis were confirmed by this effort including oncogenes (EGFR, CDK4, PDGFRA, MDM4, MET) and tumor suppressor genes (CDKN2A/B, PTEN, RB1) which were altered at varying levels. An important outcome of this analysis is that three major core pathways (RB, p53, and receptor tyrosine kinase) were altered in the majority of the tumors studied. Furthermore, NF1 emerged as an important tumor suppressor gene altered in a subset of glioblastoma (14 %), which highlighted its relevance not only as a cause of an inherited tumor syndrome but also to sporadic glioblastoma in general.

At the same time this study was published, Parsons et al. reported the results of whole-exome sequencing studies in glioblastoma [2]. The remarkable observation of IDH (1 or 2) mutations in a subset of glioblastoma highlighted the power and potential of this approach to uncover previously unsuspected robust biomarkers using high-throughput unbiased approaches. Since then, IDH mutations have emerged as one of the most frequent early alterations in diffuse gliomas [3, 4] and almost definitional for the long-elusive “secondary glioblastoma” with a better prognosis.

In a follow-up, more comprehensive report of the TCGA, integrative molecular data was obtained from 543 glioblastomas [5]. In this more comprehensive analysis, whole-genome and whole-exome sequencing were added, as well as methylome and proteome analysis. This more refined analysis confirmed prior observations, but also provided additional data on mutations in LZTR1, a putative transcriptional regulator, and the TERT promoter, as well as novel genomic rearrangements.

This component of the TCGA focuses on a comprehensive genomic study of diffuse gliomas in adults, other than glioblastoma, grouped together under the designation of “lower-grade glioma.” These include grade II and III tumors, histologically diffuse astrocytomas, oligoastrocytomas, or oligodendrogliomas. The first observations on a cohort of 293, previously untreated tumors were published [6]. Multidimensional analysis of the data included the concurrent study of DNA methylation, gene expression, DNA/mRNA sequence, and DNA copy number analysis.

The main molecular alteration, which is a recurrent feature in these lower-grade gliomas and previously highlighted by a number of studies, is the high frequency of IDH (1 or 2) mutations [2–4, 7, 8]. The main overall result of this study was the delineation of three different molecular subtypes of lower-grade gliomas with clinical implications, more robust than schemes based on histopathologic analysis alone. These three subgroups were tumors with concurrent IDH mutations and 1p19q co-deletion (best prognosis, closely associated with oligodendroglioma histology, and also containing CIC, FUBP1, NOTCH1, and TERT promoter mutations at various rates), concurrent IDH and TP53 mutations (intermediate prognosis, closely associated with astrocytic histology), and IDH wild-type tumors (worse prognosis, similar to glioblastoma).