Glioblastoma (GBM) is the most common primary brain tumor in adults and is highly lethal. A recent phase HI clinical trial demonstrated a benefit for temozolomide-chemoradiation (TMZ-CR) over radiation alone. While these results have changed the standard of care for newly diagnosed GBM patients, it is clear that only a fraction of patients derive significant benefit from this treatment, with overall two-year survival in the TMZ-CR treated patients only 26%. Since diagnosis and treatment decisions in GBM are currently based on histopathology alone, there is a need to: 1) develop sensitive and specific markers to prospectively distinguish those patients who will respond to standard TMZ-CR as initial treatment from those who will not respond;and 2) determine important molecular alterations that define the resistant tumors to design rational trials for patients who will not benefit from TMZ-CR alone. We have mined independent microarray datasets to identify an initial multimarker panel that robustly predicts outcome in GBM. We find that a molecular subtype, defined by overexpression of mesenchymal/angiogenic genes, is predictive of poor outcome.
In Aim 1, will refine this panel using 2 independent sets of tumor samples from TMZ-CR treated patients at The University of Texas M. D. Anderson Cancer Center (UTMDACC) and the Mayo Clinic College of Medicine. We will include a larger (-400) set of genes to ensure that an optimal gene panel can be identified that predicts progression-free survival. We will incorporate additional relevant markers, including MGMT and Akt signaling intermediates.
Aim 2 will rigorously test and validate this multimarker panel using patient samples from a large phase III clinical trial. The newly opened RTOG 05-25 trial, with a requirement of tissue submission, provides an unprecedented opportunity to study a large (n=834) cohort of uniformily-treated GBM patients using modern molecular techniques to identify a definitive set of predictive markers.
Aim 3 will focus on markers of response from novel agents targeted to recurrent GBM patients who have failed TMZ-CR. It is expected that tumors from most of these patients will display the mesenchymal/angiogenic phenotype, and we will therefore identify potential agents which target that phenotype in GBM. The results of this project will set the stage for future trials, in which newly diagnosed GBM patients will undergo a predictive test, and treatment will be individualized according to the molecular profile of the tumor to maximize the chances of benefit.
Showing the most recent 10 out of 232 publications
