Skip to Content
Translational Research Program (TRP)
Contact CIP
Show menu
Search this site
Last Updated: 10/10/18

SPORE: Targeted Therapies for Glioma

Dana-Farber Harvard Cancer Institute

Principal Investigator:
Tracy T. Batchelor, M.D.


Tracy T. Batchelor, M.D., MPH
Center for Neuro-Oncology
Mailstop: Yawkey Center 9E
Massachusetts General Hospital
55 Fruit Street
Boston, MA 02114
Phone: (617) 724-8770
Fax: (617) 724-8769


The primary objective of the Dana-Farber/Harvard Cancer Center (DF/HCC) brain cancer SPORE is to improve the standard of care for gliomas through the use of targeted therapies. Towards this end, basic scientists from Harvard Medical School have joined with clinical/translational investigators from Brigham and Women’s Hospital, Dana-Farber Cancer Institute and Massachusetts General Hospital. This initiative is supported by central cores for Pathology, Biostatistics and Administration and includes career and developmental programs. The research program uses clinical materials and exploits a number of clinical trials to attack glioblastoma on four distinct fronts:


Project Co-Leaders:
Tracy T. Batchelor, M.D.
Rakesh K. Jain, Ph.D.

Project one is designed to improve upon the outcome of anti-angiogenic therapy in glioblastoma. The Food and Drug Administration approved bevacizumab, an anti-VEGF monoclonal antibody, as monotherapy for recurrent glioblastoma in 2009. However, the precise mechanism(s) of action of this drug in glioblastoma is not fully understood. Previously, the project co-leaders showed that blocking VEGF-signaling with DC101, a murine anti-VEGF antibody, or AZD2171, a pan-VEGF tyrosine kinase inhibitor, normalizes glioblastoma blood vessels and increases overall survival in animal models. However, the effects of anti-VEGF therapies on overall survival appear modest in glioblastoma patients as the tumor eventually escapes from normalization by the up regulation of other pro-angiogenic signal transduction pathways. One pro-angiogenic molecule of particular interest in this regard is angiopoietin-2 (Ang2). The project co-leaders have demonstrated that overexpression of Ang2 compromises the survival benefit from DC101 treatment. Moreover, Ang2 expression decreased during the vascular normalization window but increased as the tumor vessels began to become abnormal. This pattern of dynamic Ang2 expression has also been observed in the plasma of recurrent glioblastoma patients treated with AZD2171. Based on these pre-clinical and clinical data, the project co-leaders hypothesize that the inhibition of Ang2 may prolong the normalization window, thereby improving the clinical benefit induced by VEGF-blockade. The investigators will test this hypothesis by blocking both Ang2 and VEGF signaling in different schedules to determine if this inhibits tumor growth and increases overall survival in different murine glioblastoma models (Aim One). Subsequently, it will be determined whether blocking both of these pathways can increase survival by prolonging the normalization window over that observed with VEGF inhibition alone (Aim Two). Finally, in a clinical trial (Aim Three) the investigators will assess the impact of selective Ang2 inhibition on the vascular normalization window in recurrent glioblastoma patients as well as the safety and potential efficacy of anti-Ang2 therapy. This clinical trial will provide the foundation for future combination trials of Ang2 and VEGF inhibitors, the latter informed by the results of Aims One and Two.


Project Co-Leaders:
Thomas M. Roberts, Ph.D.
Patrick Y. C. Wen, M.D.

The phosphatidylinositol 3 kinase (PI3K) signaling axis is aberrantly activated in the majority of adult high-grade gliomas. Activation in glioblastoma occurs via one of four mechanisms: 1) Loss of function mutations in the PTEN tumor suppressor; 2) Amplification/gain of function mutations in the receptors for EGF or PDGF; 3) Activating mutations in the PIK3CA gene that encodes p110a, a catalytic subunit of PI3K, or; 4) mutations in the gene PIK3R1 that encodes one of the PI3K regulatory subunits, p85a. A number of PI3K inhibitors are in the early stages of clinical trials. One of these, BKM 120, is being developed by Novartis and has been shown to pass through the blood brain barrier, making it an excellent candidate for glioblastoma therapy. Project 2 will be centered on a trial of BKM in patients with recurrent glioblastoma. The broad goal of Project 2 is to use the data and clinical materials from patients on the BKM120 trial- in concert with genetically defined mouse models – to address important unresolved questions involving PI3 kinase inhibitors as glioblastoma therapeutics. In addition to the key data on the impact of genetic modifiers on response to BKM120 (if any) coming from the human trial, cell culture and animal studies will address optimization of, and the potential benefits from, combination therapies using BKM120 in concert with standard of care, as well as a number of rationally targeted therapies. Finally, great promise has been seen with inhibitors targeting a single catalytic isoform of PI3K. To prepare clinical testing of this new class of inhibitors, preclinical experiments will be carried out determining the relative importance of the individual PI3K isoforms in disease driven by PTEN loss.


Project Co-Leaders:
Daniel Cahill, M.D., Ph.D.
William Kaelin, M.D.

Targeted therapeutics designed against specific oncogenic genomic alterations have had a large clinical impact. Recently, large-scale sequencing studies have identified recurrent, gain-of-function IDH gene mutations in a significant subset of glioblastomas, with particular enrichment in malignant gliomas of younger adults. The mutant enzyme catalyzes the production of the novel oncometabolite 2- hydroxyglutarate (2-HG). Increased levels of 2-HG inhibits the 2-oxoglutarate dependent dioxygenase class of enzymes in cells that impact a range of cellular functions including chromatin structure and the epigenetic control of gene expression, which are thought to promote tumorigenesis. Because 2-HG is not found at appreciable quantities in normal cells, where basal levels are cleared via 2-HG dehydrogenase, the accumulation to millimolar levels in human gliomas suggests that it could be an ideal biomarker for mutant enzyme activity. Understanding the requirements for mutant IDH1 activity in existing tumors, and whether 2- HG levels can serve as a surrogate for mutant enzyme activity in patients are critical issues for the development of new targeted therapies in this disease. In preliminary studies, the project co-leaders have characterized the biological correlates and potentially actionable avenues for inducing therapeutic response in IDH mutant gliomas. In Project 3, the investigators will use clinical material to test the hypotheses that non-invasive measurement of 2-HG levels can serve as surrogate for IDH mutant enzyme activity, and that targeting of IDH mutation and 2-HG may be a novel therapeutic strategy for malignant glioma patients.Prior investigations by the Dr. Kaelin have helped define the functional metabolic consequences of IDH1 mutation and 2-HG production on the epigenome of cancer cells, and he was the first to show that mutant IDH1 transforms human astrocytes in vitro, and was the first to demonstrate that a potential therapeutic intervention (EglN inhibition) can selectively target the abnormal biochemical environment within IDH1 mutant tumors. Dr. Cahill’s lab performed IDH stratification of the recent national RTOG-0525 trial in glioblastoma, and with his colleagues, has established IDH1-mutantorthotopic xenograft glioma models derived from freshly resected patient tumor samples. We believe that the successful execution of Project 3 will support the future development of clinical trials for IDH1 mutant gliomas.


Project Co-Leaders:
Jay S. Loeffler, M.D.
Charles D. Stiles, Ph.D.

Glioblastomas are notoriously insensitive to radiation and genotoxic drugs. Paradoxically, the p53 gene is structurally intact in the majority (~65-75%) of these tumors. Resistance to genotoxic modalities in p53-intact gliomas has been attributed to attenuation of p53 functions by other mutations within a p53 signaling axis that includes CDKN2A(p14Arf), MDM2 and ATM. In preliminary studies, project investigators have generated an alternative and potentially actionable resolution to the p53 paradox. Put briefly, the project leaders have shown that the gliogenictranscription factor OLIG2 suppresses p53-mediated responses to genotoxic damage in glioblastoma cells. Against this backdrop, the broad objectiveof studies proposed in this project is to use clinical materials to test the hypothesis that small molecule inhibitors of OLIG2 could serve as targeted therapeutics for glioblastoma – either as stand alone modalities or (more likely) as adjuvants to radiotherapy and genotoxic drugs. This hypothesis makes four testable predictions. The first specific aimis to test the prediction that current standard of care (radiation and Temozolomide) actually enriches for OLIG2-positive cells within p53-positive glioblastomas. The second specific aimis to test the prediction that one current class of radiosensitizing drugs – the HDAC inhibitors – actually work by suppressing OLIG2 expression in cancer patients. The third specific aimis to test the prediction that genetic suppression of OLIG2 can sensitize p53-positive human gliomas to radiotherapy in vivo. The fourth specific aimis to test the prediction that shRNA-mediated knockdown of genes essential to OLIG2 function (e.g. HDACs) will be synthetic lethal to irradiation in p53 positive gliomas.Dr. Stiles and his colleagues initially cloned the OLIG genes and defined their biological functions in brain development and malignant glioma. The work proposed in this project will be supported by dedicated SPORE core facilities for Pathology and Biostatistics. If the work described here supports the view that OLIG2 is a viable target for glioma therapeutics, clinical trials of OLIG2 antagonists (e.g. HDAC inhibitors) as an adjuvant to radiotherapy can be initiated.


Core Co-Leaders:
Keith L. Ligon, M.D., Ph.D.
David N. Louis, M.D.

The overall goal of this SPORE application is to develop more effective targeted molecular therapies and biomarkers for glioblastoma. The four proposed projects focus on 1) Overcoming resistance to VEGF inhibition by targeting Angiopoietin-2; 2) Inhibiting the phosphatidylinositol 3-kinase (PI3 kinase) pathway; 3) designing novel strategies for imaging and targeting IDH mutant gliomas; and 4) Increasing the radiosensitivity of glioblastomas by inhibiting the bHLH transcription factor Olig2. One of the innovative themes of this SPORE proposal is that biospecimens collected by the Pathology Core will be comprehensively molecularly profiled using multiplexed technologies so projects may effectively evaluate the interaction of targeted therapies and their targets. The Pathology Core will support the goals of the SPORE by providing expert neuropathological review, biospecimen banking and clinical trial support. In addition, the Pathology Core will be a centralized resource for clinically and molecularly annotated glioblastoma tissues and primary glioblastoma cell cultures that will be essential to the success of the proposed projects. By centralizing these activities, the Pathology Core will ensure the safe and effective use of finite glioblastoma tissue resources without limiting the scope of the translational research planned in this proposal. The SPORE support, combined with significant institutional support for these goals, will ensure that over the next 5 years the Pathology Core will provide the critical research infrastructure required for successful, collaborative translational research of the SPORE program.


Core Leader:
Dianne M. Finkelstein, Ph.D.

The Biostatistics Core provides statistical support for the research conducted in the brain cancer SPORE program, including related database and computing requirements. The Core provides consultation and expertise on the development and implementation of plans for data collection, particularly the coding convention and variable format required for statistical analysis. The Core contributes to the data management plan, including form design and database requirements, during development of the SPORE related clinical trials. The Core provides expertise and resources for data transfer, merging, sharing and security while maintaining the confidentiality of all patient-related protected health information. The Core provides ready and dedicated access to statistical expertise for the design, planning and conduct of preclinical experiments, clinical trials and correlative tissue and biomarker studies. Analysis of the clinical trials will include interim monitoring, if necessary, including the stopping rules for safety and futility. The Core also provides statistical expertise for data analysis and interpretation of research studies, including manuscript preparation.


Core Co-Leaders:
Tracy T. Batchelor, M.D.
Charles D. Stiles, Ph.D.

This SPORE grant is intended to support multi-project, interdisciplinary and multi-institutional translational research in glioblastoma. The governance structure of this DF/HCC SPORE grant provides the foundation for the implementation, execution and ultimate success of all projects and cores. The Administration Core serves as the “hub” for this governance structure and aims to achieve a number of specific objectives as defined below. The SPORE Director and Co-Director are senior administrators and researchers who have worked together on prior DF/HCC initiatives and who provide strong, complementary leadership for the grant. The trans-institutional administrative team consists of senior personnel at DF/HCC institutions who have worked together effectively on multiple brain cancer projects. There are two senior clinical and imaging scientists into the Administration Core to supervise brain cancer SPORE-specific clinical trials and imaging studies, respectively, and to enable this SPORE to capitalize on existing DF/HCC Cores to support these types of studies. There is an effective internal and external committee structure to provide expertise, advice and oversight of the program. Each of the committees consists of collaborative, complementary members who have already worked together on prior collaborative projects. A series of regular SPORE meetings involving both administrative and scientific SPORE serve to facilitate close collaboration, troubleshooting and monitoring of the SPORE program. There is an effective internal and external communications program, which includes a brain cancer SPORE-specific component. An established and extensive DF/HCC communications infrastructure is utilized to promote open, regular, trans-institutional communication regarding SPORE opportunities and activities. There is an active SPORE program to enhance participation by underrepresented minorities and women.