Integrative Subtype Discovery in Glioblastoma Using iCluster

Department of Epidemiology and Biostatistics, Memorial Sloan-Kettering Cancer Center, New York, New York, United States of America.
PLoS ONE (Impact Factor: 3.23). 04/2012; 7(4):e35236. DOI: 10.1371/journal.pone.0035236
Source: PubMed


Large-scale cancer genome projects, such as the Cancer Genome Atlas (TCGA) project, are comprehensive molecular characterization efforts to accelerate our understanding of cancer biology and the discovery of new therapeutic targets. The accumulating wealth of multidimensional data provides a new paradigm for important research problems including cancer subtype discovery. The current standard approach relies on separate clustering analyses followed by manual integration. Results can be highly data type dependent, restricting the ability to discover new insights from multidimensional data. In this study, we present an integrative subtype analysis of the TCGA glioblastoma (GBM) data set. Our analysis revealed new insights through integrated subtype characterization. We found three distinct integrated tumor subtypes. Subtype 1 lacks the classical GBM events of chr 7 gain and chr 10 loss. This subclass is enriched for the G-CIMP phenotype and shows hypermethylation of genes involved in brain development and neuronal differentiation. The tumors in this subclass display a Proneural expression profile. Subtype 2 is characterized by a near complete association with EGFR amplification, overrepresentation of promoter methylation of homeobox and G-protein signaling genes, and a Classical expression profile. Subtype 3 is characterized by NF1 and PTEN alterations and exhibits a Mesenchymal-like expression profile. The data analysis workflow we propose provides a unified and computationally scalable framework to harness the full potential of large-scale integrated cancer genomic data for integrative subtype discovery.

Download full-text


Available from: Nikolaus Schultz,
  • Source
    • "Secondary GBM are rare (∼5-10% of total GBM), progress from lower grade tumors, occur more frequently in younger patients with better prognosis and have a different molecular profile. Studies using gene expression, DNA copy number, miRNA, and DNA methylation show these molecular characteristics can divide GBM into subclasses, some with different clinical characteristics [3], [4], [5], [6], [7], [8], [9]. Three subtypes emerged in early studies of WHO grade IV GBM (studies that combine histological subtypes or grades of glioma and use molecular classification to distinguish them are excluded from this discussion). "
    [Show abstract] [Hide abstract]
    ABSTRACT: Activity of GFR/PI3K/AKT pathway inhibitors in glioblastoma clinical trials has not been robust. We hypothesized variations in the pathway between tumors contribute to poor response. We clustered GBM based on AKT pathway genes and discovered new subtypes then characterized their clinical and molecular features. There are at least 5 GBM AKT subtypes having distinct DNA copy number alterations, enrichment in oncogenes and tumor suppressor genes and patterns of expression for PI3K/AKT/mTOR signaling components. Gene Ontology terms indicate a different cell of origin or dominant phenotype for each subgroup. Evidence suggests one subtype is very sensitive to BCNU or CCNU (median survival 5.8 vs. 1.5 years; BCNU/CCNU vs other treatments; respectively). AKT subtyping advances previous approaches by revealing additional subgroups with unique clinical and molecular features. Evidence indicates it is a predictive marker for response to BCNU or CCNU and PI3K/AKT/mTOR pathway inhibitors. We anticipate Akt subtyping may help stratify patients for clinical trials and augment discovery of class-specific therapeutic targets.
    PLoS ONE 07/2014; 9(7):e100827. DOI:10.1371/journal.pone.0100827 · 3.23 Impact Factor
  • Source
    • "Recent work on genome wide profiling with help of the cancer genome atlas (TCGA) [18] database, using various parameters like copy number analysis, miRNA and mRNA analysis, mutational and methylation analysis, have all led to generation of GBM tumor subtype specific network profiles [19] [20] [21]. These sub-types are classical , mesenchymal, neural, and pro-neural. "
    [Show abstract] [Hide abstract]
    ABSTRACT: MiRNA-34a is considered as a potential prognostic marker for glioma, as studies suggest that its expression negatively correlates with patient survival in grade III and IV glial tumors. Here, we show that expression of miR-34a was decreased in a graded manner in glioma and glioma stem cell-lines as compared to normal brain tissues. Ectopic expression of miR-34a in glioma stem cell-lines HNGC-2 and NSG-K16 decreased the proliferative and migratory potential of these cells, induced cell cycle arrest and caused apoptosis. Notably, the miR-34a glioma cells formed significantly smaller xenografts in immuno-deficient mice as compared with control glioma stem cell-lines. Here, using a bioinformatics approach and various biological assays, we identify Rictor, as a novel target for miR-34a in glioma stem cells. Rictor, a defining component of mTORC2 complex, is involved in cell survival signalling. mTORC2 lays downstream of Akt, and thus is a direct activator of Akt. Our earlier studies have elaborated on role of Rictor in glioma invasion (Das Get al.; Mol Carcinog. 2011 Jun;50(6):412-23). Here, we demonstrate that miR34a over-expression in glioma stem cells profoundly decreased levels of p-AKT (Ser473), increased GSK-3β levels and targeted for degradation β-catenin, an important mediator of Wnt signaling pathway. This led to diminished levels of the Wnt effectors cyclin D1 and c-myc. Collectively, we show that the tumor suppressive function of miR-34a in glioblastoma is mediated via Rictor, which through its effects on AKT/mTOR pathway and Wnt signaling causes pronounced effects on glioma malignancy.
    FEBS Open Bio 05/2014; 4. DOI:10.1016/j.fob.2014.05.002 · 1.52 Impact Factor
  • Source
    • "TCGA provides a valuable resource of multiscale patient data for exploring integrative analysis methods with potential clinical impact for cancer diagnosis, prognosis, and therapy. The public availability of TCGA data has stimulated a wide variety of integrative data analysis methods, resulting in interesting cancer biomarker discoveries [112], [113] as well as new bioinformatics software tools [111], [114]. "
    [Show abstract] [Hide abstract]
    ABSTRACT: This paper reviews challenges and opportunities in multiscale data integration for biomedical informatics. Biomedical data can come from different biological origins, data acquisition technologies, and clinical applications. Integrating such data across multiple scales (e.g., molecular, cellular/tissue, and patient) can lead to more informed decisions for personalized, predictive, and preventive medicine. However, data heterogeneity, community standards in data acquisition, and computational complexity are big challenges for such decision making. This review describes genomic and proteomic (i.e., molecular), histopathological imaging (i.e., cellular/tissue), and clinical (i.e., patient) data; it includes case studies for single-scale (e.g., combining genomic or histopathological image data), multiscale (e.g., combining histopathological image and clinical data), and multiscale and multiplatform (e.g., the Human Protein Atlas and The Cancer Genome Atlas) data integration. Numerous opportunities exist in biomedical informatics research focusing on integration of multiscale and multiplatform data.
    12/2012; 5:74-87. DOI:10.1109/RBME.2012.2212427
Show more