Previous DRP and CEP awardees


Extracellular vesicles were observed to have nanofilaments in GBM U87 cells but not in control normal human astrocyte (NHA), as imaged by atomic force microscopy (AFM). Image from Shivani Sharma’s CEP proposal “Nanoscale analysis of glioblastoma extracellular vesicles."

Below is a list of all previous DRP and CEP awardees. This page is generally updated annually as new awards are announced.

DRP awardees

Madhuri Wadehra
Associate Professor
Pathology and Lab Medicine

Targeting glioblastoma using anti-EMP2 antibodies

Currently, there are no effective treatments for glioblastoma, making the outcome for individuals afflicted with this malignancy extremely bleak.  The current  median survival time of patients with this disease is only 12 months, and therefore, completely novel approaches are desperately needed. Here, we describe and propose to study a new therapeutic target for glioblastoma. EMP2 is a tetraspan protein upregulated in glioblastoma with low expression in normal brain.  In disease, our central hypothesis is that EMP2 up-regulation in GBM orchestrates FAK-Src activation required for cell migration, invasion, and neoangiogenesis, and we predict that its expression can be therapeutically exploited. This proposal focuses on understanding the role of EMP2 in GBM tumor formation as well as testing novel methods to inhibit its expression.  New preliminary data achieved during the last cycle suggested that EMP2 is upregulated in three dimensional cultures and is necessary for efficient neurosphere formation.  As additional data suggests that EMP2 is a key contributor to GBM tumorigenicity in intracranial models, we hypothesize that EMP2 promotes invasion and micrometastases and that knockdown of its expression will inhibit these processes in primary cell lines.  Accordingly, to address these issues and achieve these goals, our Specific Aims test the following hypotheses:

  1. Investigate the mechanism of EMP2 on primary glioblastoma cells.Our past published data suggests that EMP2 modulates the repertoire of integrin heterodimers on the cell surface, stabilizes the FAK complex and signaling required for cell migration and neoangiogenesis.New preliminary data shows EMP2 contributes to neurosphere formation and VEGF expression in cell lines, and within this aim, we extend these findings to primary cell lines.As additional data suggests that EMP2 is a key contributor to GBM tumorigenicity in intracranial models, we hypothesize that EMP2 promotes invasion and micrometastases and that knockdown of its expression will inhibit these processes in primary cell lines.
  2. We translate the downregulation of EMP2 using inhibitory antibodies.We will determine if EMP2 can be targeted directly using either a minibody or IgG1 antibody. New preliminary data supports the targeting of these reagents into the intracranial tumor microenvironment with no observed toxicity.mPET imaging validates the localization of these agents within EMP2 positive tumors.Based on preliminary results, we hypothesize that treatment of GBM with anti-EMP2 antibodies or antibody fragments will disrupt the process of invasion and neoangiogenesis with the net effect of reducing tumor load.We will also pair our reagents with temozolomide to determine addictive or synergistic effects using established syngeneic models. We predict that successful completion of our Aims will validate the importance of EMP2 in brain tumors and consequently allow us to ultimately move anti-EMP2 therapy into the clinic for GBM.

Vivian Y. Chang
Assistant Professor, Department of Pediatrics

Genetic Susceptibility in Brain Tumor Development

Pediatric brain tumors are the second most common cancer in childhood and the leading cause of cancer death (1-3). Recent studies utilizing next generation sequencing have detected pathogenic mutations in patients with central nervous system tumors in 8.6-35% of patients (4-6). Therefore, the underlying molecular cause of brain tumor development remains unknown in the vast majority of patients. We hypothesize that germline de novo mutations may play a significant role in susceptibility to pediatric brain tumors.

Specific Aim 1 To identify effects of de novo mutations and overall diagnostic yield of novel germline variants in pediatric patients with brain tumors using a comprehensive diagnostic testing approach with germline trio exome and genome sequencing.

Specific Aim 2 To develop a high throughput approach to understanding pathogenicity of novel germline variants in their contribution to brain tumor susceptibility by introducing patient-specific mutations to create analogous tumors in mouse models.

Understanding genomic determinants of cancer susceptibility will ultimately revolutionize the way pediatric oncologists currently diagnose, treat, and screen patients with cancer, and will open the opportunity for cancer prevention. Therefore, a more thorough and broad scale germline analysis paired with high throughput in vivo modeling to understand pathogenicity of novel variants will allow us to translate findings from next-generation sequencing to clinically significant, actionable results.

Albert Lai
Professor in Residence, Department of Neurology

Investigation of RNA m6A demethylase inhibition as a novel treatment for glioma

The IDH mutation appears to be an early mutation triggering gliomagenesis of the IDH1MUT subset of gliomas through a mechanism involving generation of the oncometabolite, D-2-hydroxyglutarate (D-2HG).  D-2HG has the ability to inhibit members of the alpha-Ketoglutarate dependent dioxygenase family, and accumulating evidence supports the notion that gliomagenesis requires inhibition of the TET  DNA CpG demethylase and subsequent establishment of the G-CIMP hypermethylation phenotype.   Paradoxically, however, one of the clinical hallmarks of IDH1MUT gliomas is the relatively indolent phenotype compared to IDH1WT gliomas, indicating that in addition to its role in glioma initiation, that the IDH1MUT also occupies a role in maintaining and regulating glioma growth. Consistent with this notion, our preliminary data clearly indicate that IDH1MUT protein can slow glioma growth through generation of D2-HG.   How and why IDH1MUT gliomas adopt a less aggressive phenotype remains to be resolved, but is likely to depend on the balance of diverse effects resulting from ability of D-2HG to inhibit an array of alpha KG dependent dioxygenases, not only TET demethylase.  Recently, one of these  alpha-KG dependent dioxgenases, FTO (Fat Mass And Obesity Associated), a RNA m6A demethylase, has been shown to be sensitive to D-2HG inhibition.   In this proposal, we explore the hypothesis that IDH1 mutation maintains a slowly growing phenotype via alteration of the epitranscriptomic (m6A) landscape resulting from D-2HG inhibition of RNA m6A demethylases and that selective RNA demethylase inhibition may represent a novel strategy for glioma treatment that can easily translated into clinical trials.

Erina Vlashi
Assistant Professor, Department of Radiation Oncology

Targeting metabolic plasticity of GBM to overcome resistance to radiation therapy

Resistance to RT is a major contributor to treatment failure in GBM patients. Overcoming radiation resistance of these tumors is one of the major remaining frontiers in Radiation Oncology that, if resolved, could dramatically improve outcomes in this disease. GBM metabolism and its role in generating resistance to oxidative stress, such as during RT is a promising therapeutic angle that we will exploit in this proposal. Specifically, we have evidence that irradiated GBM cells reprogram their metabolism towards antioxidant pathways, by funneling glucose through the NADPH-producing pentose phosphate pathway (PPP). Such metabolic re-routing during RT is mediated in part by the glycolytic enzyme PKM2 and in part by the transcription factor Nrf2. Oxidative stress-dependent inactivation of PKM2 or activation of Nrf2, both result in rerouting of glycolytic intermediates into the PPP. We hypothesize that PKM2 and Nrf2 cooperate in driving an anti-oxidant metabolic response in irradiated GBM cells that promotes resistance to RT. Of importance, small molecule activators of PKM2 are available that exacerbate oxidative stress and have anti-tumor activity, although they have not been tested in GBM or with RT. These activators cross the blood brain barrier making them suitable for combining with RT to sensitize GBM tumors, and our in vivo preliminary data supports this premise. In summary, it is proposed that interfering with the Nrf2-PKM2-metabolism axis will limit the anti-oxidant, pro-survival metabolic reprogramming induced by radiation therapy and significantly improve the response for GBM patients to radiation therapy.

CEP awardees

Tom Davidson
Associate Professor, Department of Pediatrics

Neo-adjuvant PD-1 blockade and dendritic cell vaccination for recurrent pediatric high-grade glioma

For the past decade our group and others have been testing active vaccination strategies, such as dendritic cells (DC) pulsed with tumor lysates or synthetic peptides to induce antitumor immunity in high-grade glioma patients. We have previously demonstrated the effectiveness for DC vaccination in pre-clinical models; and early stage clinical trials have shown substantial promise, such that randomized Phase III clinical trials are ongoing. While the results strongly suggest DC vaccination is able to cure or significantly prolong the survival in prophylactic or small early-established intracranial gliomas, large tumors still often cannot be effectively cured in murine models. Thus the overall goals of this research project are to investigate mechanisms of immune evasion following treatment with DC vaccines, and to develop rational combinations of immunotherapeutic strategies to overcome the immunosuppressive milieu of the brain tumor microenvironment. 

Our recent work now strongly suggests that, in addition to inducing T-cell infiltration into brain tumors, DC vaccination may also create a pro-inflammatory environment within the tumor that induces the immigration of immunoregulatory antigen presenting cells expressing high levels of PD-L1 and IL-10. Thus, our recent published work illustrates how a personalized vaccine can be combined with PD-1 mAb blockade to generate enhanced anti-tumor immune responses and long-term survival. Effective anti-tumor immune-mediated tumor cell killing requires tumor antigen presentation by dendritic cells, T-cell trafficking to the tumor and effector T-cell function in the tumor microenvironment.

Furthermore, recent pre-clinical and clinical studies have revealed systemic anti-tumor immune responses after checkpoint inhibition prior to surgical resection of primary tumor in a number of tumor types/models. Our group recently ran a pilot surgical adult recurrent GBM study led by Drs. Timothy Cloughesy and Patrick Wen, through the IVY consortium, that compared patients who received one dose of Pembroluzimab (PD-1 monoclonal antibody) prior to surgical resection to patients who received no neo-adjuvant therapy. All patients in this pilot study then received adjuvant PD-1 mAb following surgery. The initial data demonstrated an increase in an inflammatory gene expression signature in the tumor after a single neoadjuvant dose of pembroluzimab. These patients also showed an expansion of specific peripheral T cell clones when compared to the group that did not receive neo-adjuvant pembroluzimab. Strikingly, the one year overall survival for the patients who received the neo-adjuvant pembroluzimab was 394 days compared to 179 days for the group that only received adjuvant pembroluzimab (<0.008).

Thus, we propose to study the effect of neo-adjuvant PD-1 blockade in combination with DC vaccination for pediatric recurrent high grade glioma patients in a phase I/II open-label randomized clinical trial with a lead-in cohort to test the safety and efficacy of neo-adjuvant Pembroluzimab administration, followed by tumor resection, autologous tumor lysate DC vaccination, and adjuvant Pembroluzimab in recurrent pediatric (≤ 21 years of age) HGG (WHO grade III and IV).

Aim 1 To evaluate the safety, immune responses, and potential clinical benefit of combining tumor-lysate pulsed DC vaccination (to induce TILs into the tumor) with checkpoint blockage (to enhance TIL function) in pediatric recurrent high grade glioma patients. We hypothesize that the combination of a personalized vaccine and PD-1 mAb blockade will be safe, result in enhanced tumor-specific T cell responses, and lead to extended survival.

Aim 2 To identify the cellular immune mechanisms by which timing of distinct checkpoint inhibitors might enhance effective anti-tumor immune responses and potential clinical benefit in pediatric recurrent high grade glioma patients. We hypothesize that neoadjuvant PD-1 mAb blockade will activate systemic anti-tumor immune responses that are potentiated after surgical debulking, when the immune-inhibitory tumor microenvironment is reduced, and will increase the inflammatory gene expression signature and improve event free survival.

Anthony C. Wang
Assistant Professor, Department of Neurosurgery

Activating adaptive immunity to target the H3.3G34 mutation in pediatric glioblastoma

Malignant brain tumors are the #1 cancer killer of children, with expected survival under two years, for pediatric glioblastoma (pGBM). The two main subtypes of pGBM are marked by ubiquitous somatic H3F3A gene mutations. The H3.3G34 pGBM occurs in children and young adults, whereas DIPG presents at a younger age, is typically lobar in location, and concurrent TP53, ATRX/DAXX, and PDGFRA alterations are frequent.

In silico algorithms predict strong HLA/MHC binding by H3.3G34R peptide sequences as well. However, H3.3G34 alterations also inhibit H3.3K36 methylation, which appears to result in alternative splicing events. First, we hypothesize that H3.3K36 methylation is altered in H3.3G34R/V pGBM, resulting in dysregulated KDM4 demethylase activity, METTL14-mediated m6A methyltransferase activity, and DNMT3a-mediated CpG methylation via NSD1. We then hypothesize that this epitranscriptomic dysregulation results in variants of alternative splicing that create neoantigens that are presented on MHC and can stimulate cytotoxic T cell responses.

In this proposal, we aim to: 1) test the accuracy with which the IRIS algorithm predicts splice variant-derived neoantigens in H3.3G34R pGBM by using mass spectrometry to confirm the presence of predicted neoantigens in patient-derived cell lines; 2) determine whether H3.3G34R pGBM neoantigens derived from the mutant histone peptide or from alternative splicing variants stimulate cytotoxic T lymphocyte activation; and 3) compare the safety and efficacy of dendritic cell (DC) vaccination targeting the H3.3G34R peptide, against splice variant-derived neoantigens. These experiments will yield valuable data designing targeted active vaccination modalities in H3.3G34R pGBM.

Shivani Sharma
Assistant Researcher IV, Department of Pathology & Lab Medicine

Nanoscale analysis of glioblastoma extracellular vesicles

Brain tumor biomarkers for diagnosis or monitoring disease progression on therapy are currently lacking. Recently, extracellular vesicles (EVs) secreted into bodily fluids have emerged as promising cancer diagnostic biomarkers. Based on current knowledge of EVs biomolecular cargo and our preliminary findings in glioblastoma EVs, we hypothesize that oncogenes (e.g., EGFRvIII) influence the molecular composition impacting nanoscale biomechanical properties of secreted EVs (such as altered morphology, presence of nanofilaments). We introduce quantitative atomic force microscopy (AFM) as a novel approach for in-depth assessment of nanomechanical variations in GBM derived EVs, for the first time.

Aim 1 will elucidate the link between oncogenic transformation pathways and biomechanical properties of GBM EVs using in vitro cell models, and glioma stem cells.

Aim 2 harnesses the biomechanical uniqueness of oncogene-driven EVs to assess their representation in biofluids and utility for diagnostic sensing. We will evaluate CSF EVs to test if EV structure-biomechanics traits under ultrahigh resolution (AFM) in patients with GBM are different than in cases of low-grade tumors or in tumor-free subjects. Also, CSF EVs in patients with distinct oncogenic alterations such as IDH1 or EGFRvIII mutations will be examined. Our long-term goal is to develop a blood-based EV assay (mechanodetection alternative to immunodetection) to reflect the molecular status and heterogeneity of brain tumors.

Extracellular vesicles were observed to have nanofilaments in GBM U87 cells but not in control normal human astrocyte (NHA), as imaged by atomic force microscopy (AFM). Image from Shivani Sharma’s CEP proposal “Nanoscale analysis of glioblastoma extracellular vesicles."