Research Grants

2010 SFA Research Grant Recipients

Identification of causative mutations for Ewing sarcoma

• Mizuki Azuma, Ph.D. University of Kansas Center for Research
• Recipient of a $25,000 research award, Dominick Rizzi Memorial Research Award
• Abstract:  Ewing sarcoma is the second most common form of bone childhood cancer, and its genetic hallmark is an aberrant fusion gene known as EWS/FLI1.  It is unclear whether the expression of EWS/FLI1 alone is sufficient, or additional mutations are required for malignant transformation.  Identifying the common mutations among Ewing sarcoma cells is a significant process for patient treatment because of the following reasons: i) mutations can be utilized as biomarkers and ii) chemical compounds that target the mutations are strong candidates for treatment.  Tumor cells are not suitable for identifying the causative mutations because additional mutations are already accumulated.  And there is no Ewing sarcoma animal model.  Therefore, establishing Ewing sarcoma animal model allows us not only to identify the causative mutations required for malignant transformation, but also to screen for the candidate drugs for patient treatment. 

Specific aim for this proposal is to determine whether expression of EWS/FLI1 leads to aneuploidy and Ewing sarcoma in zebrafish.  In recent studies we reported that the EWS/FLI1 fusion protein leads to mitotic defects in human cells and zebrafish.  We subsequently demonstrated binding between EWS/FLI1 and EWS that leads to inhibition of EWS activity.  The hypothesis of this proposal is that the EWS/FLI1-induced mitotic defects leading to aneuploidy and to Ewing sarcoma formation.  To address the hypothesis, we generated a conditional transgenic zebrafish line expressing EWS/FLI1, and are in the process of characterizing the lines.  We will investigate the mutations caused by EWS/FLI1 or additional mutations if any required for Ewing sarcoma formation.  Zebrafish is a suitable model for this project because it contains the microenvironment for malignant formation, and high fecundity (200 eggs/female/week) ensures statistical reliability for genetics.  Zebrafish will provide an in vivo whole animal model amenable to genetic screening for second hit mutations, or chemical compounds for patient treatment.
• Final Report: Click to view PDF

---

Crosstalk between EGFR and IGF1R mediated by polymorphisms in the EGFR promoter as a mechanism for resistance to IGF1R directed therapy in osteosarcoma

• E. Anders Kolb, M.D. Alfred I. duPont Hospital for Children
• Recipient of a $25,000 research award, Dominick Rizzi Memorial Research Award
• Abstract:  Osteosarcoma is the most common primary malignant tumor of bone, with a pea incidence in the second decade of life. New therapies are needed to improve survival especially among patients with recurrent or metastatic disease. Insulin growth factor-I (IGF-I) is essential for growth in osteosarcoma. Several of the fully humanized monoclonal antibodies targeting the IGF-I receptor (IGF-IR) currently in clinical trials have clear evidence of growth inhibition in osteosarocoma tumor lines. However, some osteosarcoma lines remain resistant to IGF signal inhibition. Preliminary data presented using a fully humanized monoclonal anti-IGF-IR antibody, indicates that osteosarcoma tumors may employ salvage signaling through the epidermal growth factor receptor (EGFR) and mitogen-activated protein kinase (MAPK) pathways to permit growth in the presence of IGF-IR inhibition. In fact, preliminary data presented in this proposal demonstrate that single nucleotide polymorphism in the EGFR promoter predicts induction of EGFR in response to IGF1 R inhibiton. The hypothesis evaluated in Specific Aim 1 is that the activity of the EGFR pr6moter in osteosarcoma is influenced by the sequence of the Sp1 binding site of the EGFR promoter leading to differential activation of EGFR in response to IGF1R inhibition. Specifically, plasmid constructs representing the different -216 sequences of the EGFR promoter driving a luciferase reporter will be transiently transfected into osteosarcoma cells and human fibroblasts. Promoter activity in response to an antibody blocking IGF1R activity and IGF1R RNA interference (RNAi) will be measured.

The hypothesis evaluated in Specific Aim 2 is that the genotype at -216 of the EGFR promoter will serve a biomarker for response to IGF1 R inhibition. Peripheral blood mononuclear cells will be harvested from volunteer donors and EGFR expression evaluated
in response to an antibody blocking IGF1 R activity and IGF1 R RNAi. Results will be correlated with genotyping of buccal mucosal cells.
• Final Report: Click to view PDF

---

Functional analysis of Mirk/Dyrk 1B in osteosarcoma

• Zhenfeng Duan, M.D., Ph.D. Sarcoma Molecular Biology Laboratory, Center for Sarcoma and Connective Tissue Oncology, Massachusetts General Hospital
• Recipient of a $25,000 research award, Alexander Burdo/Ziopharm Research Award
• Abstract:  Osteosarcoma is the most common primary malignant tumor of bone.  Standard treatment is surgery and chemotherapy, which has significantly improved the survival rate from 11% with surgery alone to 60-70% when surgery is combined with chemotherapy.  Unfortunately, progress has been limited over the past 20 years, and the identification of new therapies is critical to improving the survival rate of osteosarcoma patients.  Protein kinases have recently been targeted by pharmaceutical agents to decrease tumor growth (i.e. Bcr-Abl, EGFR, Her2, c-kit).  One of the less known kinases, Mirk/Dyrk1B, is a unique, multifunctional kinase, that has been observed to be amplified, upregulated, or constitutively expressed in several different types of cancer. Previously, we utilized a lentiviral shRNA kinase library to screen osteosarcoma cells and showed that knockdown of Mirk expression inhibits cell growth and induces apoptosis in osteosarcoma cell lines.  We observed that Mirk is commonly overexpressed in osteosarcoma but not in normal osteoblasts. Mrik participates in multiple cellular pathways that enhance tumor growth and survival, and we observed that high levels of Mirk expression in osteosarcoma tissue are associated with more aggressive clinical behavior.  Therefore, our proposed project has the following goals: 1) Determine the impact of overexpression of Mirk in human osteoblast cell lines on cell differentiation, proliferation and neoplastic transformation; and 2) Analyze the molecular consequences of inhibiting Mirk in osteosarcoma cell lines.  Our long term goal is to elucidate the regulatory mechanisms controlling expression of Mirk and ultimately develop therapeutic strategies that can be used to improve the treatment of patients with osteosarcoma. 
• Final Report: Click to view PDF

---

A Quantitative Imaging-based Biomarker for Assessment of Therapy Response in Soft Tissue Sarcomas by Differential Volume Estimation of Viable and Non-viable Tumor Fractions

• Anand K. Singh, M.D. Massachusetts General Hospital
• Recipient of a $25,000 research award
• Abstract:  The rapid developments in imaging techniques and action of newer chemotherapeutic drugs have highlighted the limitations of response evaluation criteria in solid tumors (RECIST) for assessing treatment response in soft tissue sarcoma. Some chemotherapeutic agents may induce more tumor necrosis compared to another, causing enlargement of the total tumor size and thus leading to false positive interpretations of disease progression on RECIST of an otherwise stable or regressed disease state. Newer imaging techniques like tumor perfusion and positron emission tomography (PET) need standardization with regard to apparent diffusion coefficient values and tighter control on false positive detections respectively with added disadvantages of extra costs and scanning. Revisions in tumor response assessment criteria’s are therefore gaining importance.

In our preliminary study, excellent correlation was observed between the proposed MRI volumetry technique and histopathology for estimation of non-viable tumor fraction in treated and excised tumors. We thus hypothesize that non viable tissue fraction of soft tissue sarcomas can be accurately quantified by performing 3D segmentations on MRI datasets and such bio-estimates will serve as a better predictor of therapy response compared to existing RECIST criteria. We will obtain interval change in viable and non-viable tumor fractions by applying semi-automated 3D segmentation techniques on axial slices of contrast-enhanced pre-treatment and post-treatment T1weighted MRI datasets and compare them with the response obtained by RECIST criteria 1.0. We will also estimate one-dimensional tumor measurements on both MRI time-point datasets and assess therapy response based on RECIST criteria. Finally, we will compare statistical estimates of two methods generated by Kaplan-Meier survival curves for progression-free disease and overall (long-term) survival time. 

After successful testing of our hypothesis, this cost-effective and feasible innovation may have significant positive impact in clinical decision-making for treatment of sarcomas where similar principles can be applied on wider latitude for other body tumors. 
• Final Report: Click to view PDF

---

Targeting GLI1 in Ewing Sarcoma

• William A. May, M.D. Saban Research Institute, Children’s Hospital Los Angeles
• Recipient of a $25,000 research award, Mark Herzlich Research Award
• Abstract: The EWS/FLI1 oncogenic transcription factor is both unique to and indispensable to the development of Ewing’s Sarcoma and peripheral Primitive Neuroectodermal Tumor (collectively known as the Ewing Sarcoma Family Tumors or ESFT). It is imperative that we translate our understanding of EWS/FLI1 biology into improved therapeutics for Ewing Sarcoma.
We have recently published findings that EWS/FLI causes increased expression of Gli1 in ESFT in a Hedgehog independent fashion. We have further published data that inhibition of GLI1 expression by shRNA results in decreased tumorigenicity and that several known targets of EWS/FLI1 are diminished in their expression. This suggests that signaling through Gli1 may be a central and critical event in EWS/FLI1 downstream gene activation which is central to this disease. Our preliminary data demonstrates that two newly identified compounds termed GANT58 and GANT61, which inhibit this pathway at the level of GLI1, act to inhibit the HH-GLI1 pathway in Ewings cell lines in vitro. Preliminary data suggest that at least one of these compounds will also be effective in vivo. This proposal will utilize these novel compounds to fully test the potential of GLI1 inhibition in Ewing Sarcoma in vivo models. Both compounds will be tested at full doses in xenograft models of ESFT for their inhibition of tumor growth. We will assess experimental tumors for reduced expression of downstream targets of GLI1 in ESFT. We will also use in vivo imaging systems to assess the effect of the drugs on GLI1 transcriptional activity in vivo. Recently, several prominent tumor types have been shown to depend on GLI1 activation in a Hedgehog independent fashion. As these developments stimulate the development of pathway inhibitors effective in these circumstances, the data generated in this proposal will establish whether such compounds could benefit patients with Ewings Sarcoma in the near future.
• Final Report: Click to view PDF

---

Developing a MicroRNA-Based Strategy for Targeting Uterine Leiomyosarcoma

• Matthew L. Anderson M.D., Ph.D. Baylor College of Medicine
• Recipient of a $25,000 research award, Jay V. Jackson Memorial Research Award
• Abstract: MicroRNAs (miRNAs) are a novel class of small, non-coding RNA transcripts that broadly regulate patterns of gene expression. Recently, we used Next Generation Sequencing to dentify 49 individual miRNAs differentially expressed in uterine leiomyosarcoma (ULMS). We also identified potential targets for these miRNAs by using bioinformatic algorithms to screen gene products differentially expressed when genome-wide transcriptional profiling was used to examine our specimens. Based on these results, we hypothesize that miR-143 plays a central role in the pathogenesis of ULMS. To test this hypothesis, we intend to examine the function of miR-143 in uterine smooth muscle and smooth muscle tumors both n vitro and in vivo. Mimics and inhibitors for miR-143 will be transfected into an established LMS cell line (SK-LMS), primary cell strains derived from metastastic ULMS (285, 505 and 987) and telomerase-immortalized lines derived from human myometrium (HM) and eiomyoma (DD). Standard MTT and caspase assays will be used to compare rates of Droliferation and apoptosis. Flow cytometry will be used to measure the impact of altered miR-143 levels on the cell cycle. Scratch assays will be used to measure cell migration. Once biologic functions for miR-143 have been established, target genes will be validated using quantitative PCR and Western blotting. We also intend to test whether miR-143 sensitizes SK-LMS and primary ULMS cell strains to gemcitabine and docetexel. Lastly, we plan to examine whether miR-143 regulates the growth and metastasis of ULMS in vivo, using clones of SK-LMS stably transfected with miR-143 mimic, miR-143 inhibitor or scrambled control to create xenografts in nude mice.

We believe that the insight generated by the proposed work will establish a functional role for miR-143 in ULMS and set the stage for the clinical application of this insight. Future work would focus on how best to deliver miR-143 for therapeutic purposes.
• Final Report: Click to view PDF

---

Regulation of sarcomagenesis by the Piwi proteins and their interacting small RNAs (piRNAs)

• Igor Matushansky, M.D., Ph.D. Columbia University Medical Center, Irving Cancer Research Center
• Recipient of a $25,000 research award
• Abstract: Recently levels of HIWI (the human ortholog of Drosophila PIWI) HIWI, have been found to be expressed in sarcomas with higher HIWI levels correlating significantly with worse clinical outcomes. In lower organisms, PIWI family members silence mobile genomic elements (i.e., transposons) and thus help to maintain genomic integrity via regulating the production of a distinct class of small ‘silencing’ RNAs - aptly named Piwi-interacting RNAs or piRNAs. It is currently unclear as to why ‘higher’ levels of a gene that silences mobile genetic elements would correlate with worse clinical outcomes for sarcoma patients. Our studies have revealed that overexpression of HIWI in mesenchymal stem cells (MSCs; connective tissue progenitors) prevents their differentiation and generates sarcomas. Since Piwi works through generation of piRNAs we feel our models create the perfect environment to identify specific piRNAs that mediate the sarcomagenic process. Our working hypothesis is that overexpression of HIWI results in piRNA mediated suppression of genes that are essential to preventing tumorigenesis. The specific aims of this proposal are to (1) assay for the presence of piRNAs in Hiwi-mediated transformed MSCs and correlate their presence to piRNAs in sarcoma tissues; (2) identify target genes that are deregulated via Hiwi-mediated tumorigenicity; and (3) assess the role of HIWI in promoting sarcomagenesis in vivo and responsiveness to piRNA treatment.
Importantly, since Piwi/piRNA cycle is self-perpetuating, any tumorigenic process initiated by Piwi/piRNA may by permanently terminated via the one time (transient) introduction of the appropriate piRNA - and thus holds great promise as a curative cancer therapy.
• Final Report: Click to view PDF

---

Lipid Metabolism in Liposarcoma:  A Novel Target for Therapeutic Intervention

• Nancy B. Kuemmerle, Ph.D. Saban Research Institute, Dartmouth Medical School
• Recipient of a $25,000 research award, Richard and Kathy Lobo Research Award
• Abstract: There is no effective medical therapy for patients with liposarcoma. We find that liposarcomas are dependent upon a constant supply of fatty acids to fuel their growth. Tumors may potentially acquire these lipids by a) de novo synthesis using fatty acid synthase (FASN), b) extracellular hydrolysis of circulating fat using lipoprotein lipase (LPL), followed by cellular uptake using CD36, or c) endocytosis of triglyceride-rich particles using Syndecan-1. We find that liposarcomas express high levels of FASN, LPL, CD36, and Syndecan-1. Moreover, Spot 14, a key nuclear driver of the genes encoding enzymes involved in lipogenesis, is also abundant in liposarcomas. We have demonstrated that the growth of liposarcoma cells is impaired by inhibition of lipid synthesis and is promoted by increased lipid uptake. Our Aims are 1) Identify proteins which interact with Spot 14 in liposarcoma but not in normal adipose tissues. 2) Investigate the functional interactions among the three pathways of lipid acquisition in liposarcoma cells, and determine whether inhibition of one pathway results in upregulation of the others. This work will provide understanding of the mechanisms underlying the exquisite susceptibility of liposarcoma to metabolic manipulation, and will guide the development of novel interventions to control liposarcoma, which is currently curable only by successful surgical resection.
• Final Report: Click to view PDF

---

How does a sarcoma circumvent fusion oncoprotein-mediated toxicity?

• Frederic G. Barr, M.D., Ph.D. University of Pennsylvania School of Medicine
• Recipient of a $25,000 research award, ARIAD and Merck Research Award
• Abstract: Alveolar rhabdomyosarcoma (ARMS) is an aggressive soft tissue sarcoma with a 2;13 translocation that generates a PAX3-FKHR fusion oncoprotein. Introduction of PAX3-FKHR into multiple cell types showed that this oncoprotein exerts growth suppression and/or cell death when expressed at levels comparable to those found in ARMS cells. Subclones from a myoblast population expressing MYCN and an inducible form of PAX3-FKHR showed increasing growth suppression in a plating assay as PAX3-FKHR activity was induced. However, when fibroblasts were added for a focus formation assay, numerous transformed foci formed when PAX3-FKHR was induced. Multiple ARMS cell lines have a similar phenotype of higher focus formation than plating efficiency. In ARMS cells, siRNA-mediated suppression of FGFR4, a PAX3-FKHR downstream target, increases plating efficiency without affecting focus formation. We hypothesize that FGFR4 signaling contributes to the mechanism of PAX3-FKHR-induced toxicity and that stromal cells attenuate the toxic effects of high PAX3-FKHR activity and promote PAX3-FKHR-induced oncogenesis.  To investigate which stromal cells are capable of exerting this effect, we will use our cell culture system to study additional fibroblasts, myoblasts, and mesenchymal cell types associated with preferred sites of ARMS metastasis. We will study the role of FGFR4 signaling in PAX3-FKHR-induced toxicity and oncogenesis by extending our initial RNA interference studies. PCR and western blot approaches will then be used to analyze expression of FGFR4 and its ligands and the activity of downstream MAPK and AKT signaling pathways in these cells. Based on these findings, inhibitors will be selected to examine the role of specific components of the FGFR4 signaling pathway. These studies will provide a comprehensive approach to an important problem in sarcoma pathogenesis and lead to therapeutic strategies to undermine mechanisms by which sarcoma cells circumvent fusion and thereby become sensitized to this toxicity. 
• Final Report: Click to view PDF

---

Identification of the target genes of the EWS/NR4A3 fusion protein expressed in extraskeletal myxoid chondrosarcoma

• Yves Labelle, Ph.D. Centre Hospitalier Universitaire de Quebec
• Recipient of a $25,000 research award, Mandell/Kropp Run for a Sarcoma Cure Research Award
• Abstract: Extraskeletal myxoid chondrosarcoma (EMC) are soft tissue tumors occurring mainly in the extremities, most commonly the thigh and knee. In approximately 75% of these tumors, at (9;22) chromosome translocation is present. This translocation encodes a fusion protein named EWS/NR4A3 which consists of the amino-terminal domain of the EWS protein fused to the complete amino acid sequence of the NR4A3 nuclear receptor.  In my laboratory we have shown that EWS/NR4A3 is a highly potent transcriptional activator, suggesting that one way in which it may contribute to tumorigenesis is by activating specific target genes. There are no cell lines established from EMC tumors, so to generate a human cellular model to study EWS/NR4A3 we have used a human bone marrow mesenchymal stem cell line. This cell line, named hTERT, has retained the ability to differentiate into osteoblasts and chondrocytes, does not form foci in soft agar, has a normal karyotype, has retained inhibition of proliferation by cell-cell contact, and does not induce tumors in nude mice. We have chosen this cell line because an extensive immunohistochemical study of EMC tumors has shown that they most likely consist of primitive mesenchymal cells. We have transfected the cells with an EWS/NR.4A3 expression vector and isolated several cell lines stably expressing EWS/NR4A3. These cell lines display a loss of inhibition of proliferation by cell-cell contact. Since loss of contact inhibition is one characteristic of tumor cells, our hypothesis is that this loss of contact inhibition is due to the activation of specific target genes by EWS/NR4A3 that may also play a role in the development of EMC tumors. To identify these genes, we will perform microarray analyses of the cell lines to identify genes over-expressed in the presence of EWS/NR4A3. We will compare those genes to a list of genes which we have previously found to be specifically over-expressed in EMC tumors expressing EWS/NR4A3. Genes over-expressed in both will be considered potential targets of the fusion protein. Future experiments will involve immunohistochemical analyses of EMC tumors to confirm expression of the corresponding proteins in tumor cells. On the long-term, these proteins could be targeted to treat patients with EMC tumors expressing the fusion protein.
• Final Report: Click to view PDF

Programs & Resources