Current Projects

Show All

Using the experimental design previously published (Cancer Cell, 2011, 19:305) we have developed additional GEMMs with different survival characteristics and tumor genetics. The properties and similarities of these models to human disease are being investigated at the histological as well as at the molecular levels. Importantly, our GEMM tumors recapitulate both the proneural and mesenchymal gene expression subgroups noted in human high-grade gliomas. We are interested in uncovering the molecular determinants of these subgroup differences as well as their signaling pathway characteristics. High-grade gliomas are highly infiltrative and aggressive tumors, a feature which is correlated with the lethality of these tumors and is replicated in our GEMMs. We are investigating the characteristics of infiltrating tumor cells in our models to better understand these processes and to determine if the biology of these highly active cells is different from the primary tumor mass.

Pseudopalisading Necrosis in Mouse High-Grade Gliomas




Pseudopalisading Necrosis in Mouse High-Grade Gliomas. Hematoxylin and eosin staining of mouse brains reveals tumors with pseudopalisading necrosis, a feature of grade 4 tumors.



Analysis of genomic aberrations occurring in our GEMM tumors and comparisons to data from human high-grade gliomas indicates the involvement of several novel genes in the genesis of these tumors. We are actively characterizing the specific mutations affecting these novel genes and will investigate the mechanisms by which they lead to tumor formation using both mouse genetic and cellular genetic studies. We anticipate that these studies will result in new hypotheses on the biology of certain high-grade glioma subgroups and may lead to novel strategies to combat these tumors.

Copy Number Aberrations in Mouse High-Grade Gliomas.




Copy Number Aberrations in Mouse High-Grade Gliomas. Array comparative genomic hybridization (aCGH) can reveal copy number aberrations in mouse tumors, including regions of homozygous loss.


The PI3K/AKT/mTOR signaling pathway is one of the three core pathways dysregulated in all high-grade gliomas. It therefore represents an attractive target for therapy for these tumors. However, early stage clinical trials utilizing inhibitors of this pathway have been disappointing to date. Using our GEMMs, we are characterizing tumor response to various inhibitors of the pathway in an effort to understand the adaptive mechanisms that allow tumors to escape inhibition of this critical pathway. We will test the hypothesis that combination therapies that block the adaptive mechanisms as well as the PI3K pathway will be synergistic or at least additive in effect.

Pi3k/Akt/mTor Pathway.


Pi3k/Akt/mTor Pathway. This important signaling pathway is a core pathway in glioma biology and includes many potential targets for pharmacologic intervention.

Recent large scale studies of human high-grade glioma tumor samples have led to the identification of many novel tumor suppressor genes and oncogenes that potentially play a role in gliomagenesis. We have an ideal mouse genetic model which we can use to test these candidates for cooperation in tumor formation. We are currently investigating one intriguing candidate gene in our model. This project will serve as a template for testing other driver mutations.

Cooperative Mutations in Glioma Biology.



Multiple lines of evidence suggest that the biology of pediatric high-grade glioma is significantly different from that of their adult counterpart. Therefore, simply treating these patients based on therapy developed in adult trials or from models of the adult disease is not an effective strategy. In order to expand on the modest number of pediatric samples with genomic and gene expression data, we are collecting through various avenues, a large cohort of pediatric high-grade glioma samples from which we can obtain genomic, gene expression and micro-RNA expression profiles for integrative analyses. A critical problem that patients and clinicians face is the ability to follow tumor growth and recurrence in patients. MRI scanning is the gold standard but is expensive, time consuming and occasionally ambiguous. The availability of a serum biomarker to use instead of or in conjunction with MRI would improve the diagnostic tools available to the clinician. We have collected high-grade glioma patient sera from a subset of individuals at various timepoints in their treatment course. We will analyze these samples at both the protein and nucleic acid level to identify potential biomarkers of disease.