Abstracts

Department of Molecular, Cellular, and Developmental Biology, Yale University

1 R21 CA118631-01A1

A key part of determining the course of treatment for a specific cancer is the identification of the specific activated signaling pathways, which are causing the malignant growth. In fact the treatment for a given cancer can be dependent upon the activated signaling pathway; for example HER2/neu positive vs. negative breast cancers are treated differently. This personalized medicine approach is best exemplified by the development of the Abl tyrosine kinase inhibitor Gleevec, which has revolutionized the treatment of CML. As more drugs targeting specific signaling pathways are developed, it will be important to identify those oncogenic signaling pathways activated in a given tumor biopsy. Towards this end, our long-term goal is the development of a library of small molecules to be used as diagnostic tools for assessing primary cancerous tissue samples. We have recently developed a new technology known as PROteolysis TArgeting Chimera molecules (PROTACs) that can selectively knock down a specific protein in vivo . These cell permeable hetero- bifunctional molecules utilize the cells own ubiquitin/proteasome protein degradation pathway to selectively destroy a target protein of our choosing. We propose to adapt this technology so that proteins required for continued tumor growth are degraded only in those cells with a particular activated tyrosine kinase pathway. In this way, it will be possible to identify those signaling pathways upregulated in a particular tumor cell and which are required for its growth. Towards the goal of novel tumor diagnostic technology development, in the subsequent R33 application, we propose to develop a panel of PROTACs that can be used in identifying the activated cancerous cell signaling pathways. This panel will be tested for use as a diagnostic tool for determining the best course of drug treatment.

Department of Pathology, University of Utah School of Medicine

1 R33 CA112061-01A2

Non-Hodgkin’s lymphoma accounts for approximately 50,000 new cases of cancer annually. This figure represents an increase beyond that seen for most other forms of cancer. Among the non-Hodgkin’s lymphomas, follicular lymphoma represents the most common subtype of low-grade B-cell lymphoma in adults, and typically pursues an indolent clinical course. In a significant proportion of cases there is histologic transformation from a low-grade neoplasm to an aggressive diffuse large B-cell lymphoma with significantly decreased median survival. The recent advent of sophisticated mass spectrometry technology coupled with software algorithms that permit instantaneous protein identification, makes it feasible to study the pattern of protein deregulation that distinguish two biologic states. We propose to employ a combination of chromatographic techniques and tandem mass spectrometry in the identification of the alterations in protein expression that accompany histologic transformation. We shall be analyzing a cohort of matched pairs of follicular lymphoma and their transformed diffuse large B-cell lymphoma counterparts occurring in the same individual. Relevance: Comprehensive identification of the qualitative and quantitative changes in protein expression that are involved in follicular lymphoma transformation will permit the delineation of deregulated pathways, identify distinct prognostic subgroups of transformed lymphoma, and facilitate the development of novel therapies that target susceptible elements in the deregulated pathways.

Biological Engineering Division, Massachusetts Institute of Technology

1 R33 CA112151-01A2

Every time a cell divides, billions of base pairs of information must be accurately copied in the face of an onslaught of DNA damage. Homology directed repair (HDR) provides one of the most important mechanisms for coping with damaged DNA. If coding information is missing or corrupted, HDR can extract sequence information available elsewhere in the genome. Although HDR is generally beneficial, transfer of genetic information is risky, since misalignments can lead to tumorigenic rearrangements. To investigate the process of HDR in vivo , we have engineered the first mouse model in which HDR can be detected in somatic cells by the appearance of a fluorescent signal. In the fluorescent yellow direct repeat (FYDR) recombomice, recombination at an engineered substrate yields fluorescence. Recombination assays are simple and rapid, making it possible to do in days what used to take weeks. In addition, the FYDR mice overcome limitations of previous systems. For example, although APRT+/- mice can be used to detect loss of heterozygosity, technically demanding assays are necessary to identify HDR events; in the pun mice, only embryonic recombination events can be detected. In contrast, FYDR mice yield a fluorescent signal that is specific to HDR events, and the recombination rate can be readily measured in cells from both embryonic and adult tissues. Furthermore, fluorescence makes it possible to capture in situ images of recombined cells, making it possible to discern independent lineages of recombinant cells in vivo . Our Specific Aims are to I) Evaluate the frequency of recombinant cells in multiple tissues; II) Develop methodology for quantification of recombinant pancreatic cells in situ and reveal the relative frequency of recombinant cells among two different cell types within a normal tissue for the first time; III) Measure the effects of environmental factors on recombination in vivo ; and IV) Reveal how specific genes (Blm and p53) affect recombination susceptibility in vivo . The broad long term objectives of this work are to demonstrate the utility of this newly developed technology for studying recombination in mammals, to substantially expand the capabilities of the existing system, and to elucidate environmental and genetic factors that influence a person’s susceptibility to spontaneous, environmentally-induced, and cancer therapy-induced DNA rearrangements.

Biomedical Engineering Center, Ohio State University

1 R21 CA122864-01

This R21/R33 application entitled “Nanoparticles for Harvesting and Targeting Angiogenic Proteins” has as its hypothesis that development and refinement of surface characteristics of silica chips with nanocharacteristics can enhance sensitivity of mass spectrometry (MS) detection of the low molecular weight angiogenic proteins present in serum and tumors that produced at very early times of tumor development. In addition, refinement of conjugation methods of nanoporous particles will allow selective targeting of endothelial cells in vitro and tumor-associated blood vessels in vivo and in combination with refinement of loading strategies, cytotoxic agents loaded into nanoparticles can selectively destroy these vessels. Our experimental plan is based on our expertise in development and refinement of emerging nanotechnology approaches for protein capture, for selective targeting and loading of silicon nanoparticles. These studies also take advantage of our experience in identification of novel proteins within the vascular endothelial growth factor (VEGF) family of proteins that are essential in the process of tumor-associated angiogenesis. To achieve the goal of developing and refining tools for detection of angiogenic proteins and for selective targeting and destruction of tumor-associated blood vessels, the following Specific Aims are proposed: (1) Develop and refine silica chips with nanocharacteristics to enhance the sensitivity of LC-MS/MS identification VEGF proteins in serum and in skin tumors during skin tumor-associated angiogenesis in vivo ; (2) Refine conjugation of silicon nanoparticles to anti-VEGFR-2 receptor antibodies for selective targeting of endothelial cells in vitro and targeting tumor-associated blood vessels in vivo ; (3) Determine the ability of silicon nanoparticles conjugated with anti-VEGFR-2 antibodies to be loaded with and to deliver the cytotoxic agent melatin for destruction of endothelial cells in vitro and for destruction of tumor-associated blood vessels in vivo . These studies will provide sensitive nanotechnology tools that are critical in defining the proteome in serum and tumors related to tumor angiogenesis that is currently unexplored. These studies may also provide strategies to selectively target tumor vessels for destruction using nanotechnology approaches.

DAN Immunotherapy, Ltd.

1 R21 CA114160-01A1

Metastatic melanoma patients currently have a dismal diagnosis, and treatment at the metastatic stage is generally ineffective. The long-term objective of this research project is to develop a novel treatment and vaccination approach for cancer in general, and melanoma in particular, based on immunotherapy. Ex vivo antigen delivery for immunotherapy is laborious and expensive, and is thus not affordable to many of those in need. The investigators propose to develop an antigenic entity that can be applied on the skin, with direct antigen delivery to skin dendritic cells and without the need for in vitro cell manipulations. Thus, the major practical objective of this study is to establish the proof of principle that topically delivered tumor associated antigens can elicit effective anti-tumor responses, and can be used for cancer immunotherapy.

Specific Aims: The study will be based on two antigenic proteins derived from melanoma: the first is a hydrophilic recombinant gp100 protein, and the second is a multiepitope polypeptide that comprises 3’, repeats of 4 HLA-A2 melanoma peptides derived from 3 different melanoma proteins. In order to allow and to improve topical transdermal delivery, the antigens will be genetically fused to potential carrier molecules. One of these is E. coli heat labile enterotoxin, a molecule recently shown to act as carrier and adjuvant. Another is a novel haptotactic C-terminal fibrinopeptide (Haptide). During the first phase of the project, R21, the new antigenic entities will be cloned, expressed, and purified. Novel in vitro models using human skin will be used to evaluate transcutaneous passage of molecules, Langerhans cells activation and mobilization, and stimulation of specific cytotoxic T cells. The rationale for the milestones that will determine continuation to the second phase, R33, is based on the efficacy of antigen delivered transcutaneously to stimulate the immune system in human in vitro models, and will allow for the selection of the molecules that will be further evaluated in depth in vivo models. In the R33 phase, specific immune responses of splenic T cells from vaccinated mice will be evaluated, tumor models will be established in mice, and the response to vaccination will be determined. Finally, the most effective molecule/s will be produced under GMP or GMP-like conditions for phase I/II clinical trials in a subsequent study.

Public Health: The success of this project would allow topical application of an immunostimulant for treatment of melanoma and other cancers and would thus significantly simplify treatment, eliminating the need for hospitalization and even day-care and without the need for a specialized laboratory. As a result, one could treat a much larger number of patients, with the potential to clinically evaluate new antigens and immunotherapeutic modalities, improving the life quality and expectancy of metastatic melanoma patients.

Van Andel Institute

1 R21 CA122890-01

The development of methods to accurately detect early pancreatic cancer and to better differentiate benign from malignant disease could greatly improve the outcomes for pancreatic cancer patients. It is known that malignant transformation of epithelial cells of the pancreas results in alterations in the carbohydrate chains of certain proteins secreted or released by these cells. Glycosylated proteins form the basis for current biomarkers for detecting pancreatic cancer and other adenocarcinomas, and refinement of these tests are predicted to enable detection of early pancreatic cancer. Our preliminary data has shown that a novel antibody-microarray technology allows the efficient detection of glycans on distinct proteins and the identification of specific glycan structures associated with pancreatic cancer. The method uses antibody microarrays to capture specific proteins from serum samples, followed by the incubation of a glycan-binding protein (such as a lectin) to quantify specific glycans on the captured proteins. Two classes of glycoproteins, mucins and carcinoembryonic-antigen-related proteins, are particularly associated with cancer, both in altered expression patterns and in altered glycan structures on the proteins. In the R21 phase, we will determine the levels of multiple specific glycans on members of those protein classes to test the hypothesis that the measurement of specific cancer-associated glycans on specific proteins, as opposed to measuring just protein or just glycan levels, will yield improved sensitivities and specificities for cancer detection. The R33 phase of the project will expand and thoroughly test the approach. The sensitivity and specificity of detecting pancreatic cancer using measurements of glycans on mucins, CEA proteins, and proteins identified in the R33 phase will be characterized in a large set of serum samples from subjects with pancreatic cancer, benign pancreatic disease, other cancers, and no disease. We expect to characterize the value of these measurements for disease diagnostics and to gain insights into the generality and frequency of specific glycan alterations on secreted proteins. Relevance to public health: The ability to more accurately diagnose cancers at earlier stages could lead to improved outcomes for many patients. This research could lead to significantly improved blood tests for the detection of cancer, as well as a powerful, generally- applicable platform for studying carbohydrate alterations on multiple proteins.

Oncology Program, H. Lee Moffitt Cancer Center and Research Institute

1 R21 CA118632-01A1

Gene profiling technology has enabled analysis of the transcriptome and proteome of tumor cells, including multiple myeloma (MM). This information has provided useful information with regard to molecular mechanisms that define the enhanced survival and proliferation of MM cells. However, an equally, if not more important, goal is to define those proteins that participate in signaling pathways active in MM cells and their supporting stroma. Enzymes that phosphorylate tyrosine, serine and threonine residues on other proteins play a major role in signaling cascades that determine cell cycle entry and survival in MM and the stromal cells that support them. In particular, knowing the signaling pathways that are active in MM cells and their supporting stroma will provide critical information for understanding MM cell survival in the BM. We have developed and are applying to purified cells a novel array-based strategy that allows the simultaneous detection of phosphorylation for 1176 different kinase substrates. Here we propose to apply this emerging technology to the analysis of phosphorylation-based cell signaling pathways in MM and their supporting stroma. This R21/R33 Phased Innovation application will be pursued in two phases. In the R21 phase of this application Aims 1 and 2 will validate that PepChip technology can be applied to MM cells and their stroma to reveal signaling alterations in MM cells. In Aim 3 of the R33 phase we will use PepChip technology to identify kinome alterations within the MM patient population that are correlated with clinical parameters such as relapse and chromosomal abnormalities associated with poor prognosis. In Aim 4 we will utilize an in vivo model that supports the growth of primary patient isolates in human bone grafts to determine the effect of therapeutics on the kinome of MM cells. This study will be pursued in the following phased R21/R33 format: R21 Phase: Aim 1: Define the kinome of MM cells and normal plasma cells. Aim 2: Define the kinome of BM stroma from MM patients and normal controls. R33 Phase: Aim 3: Identify kinome alterations in MM correlated with clinical parameters of disease. Aim 4: Identify kinome alterations in MM cells in response to therapeutics in vivo .

Department of Molecular Pharmacology, Stanford University

1 R21 CA1207322-01

The development of human cancer is a multistep process in which future cancer cells acquire mutant alleles of proto-oncogenes, tumor-suppressor genes, and other regulatory genes. Many or most of these genes are signaling related proteins and we are focusing here on the design principles of signaling networks that control the cancer related processes of proliferation, migration and endocytosis. We will test the key questions of 1) whether these cancer related signaling networks have a modular structure and 2) whether cancer cells have missing or added signaling modules that cannot be observed in normal cells. We have made significant advances to answer these questions by developing a method to create 2304 in vitro Dicer generated siRNAs against a core set of human signaling proteins. Using these siRNAs, we have already discovered the function of STIM1, a Ca2+ sensor in the ER lumen that controls Ca2+ influx into cells, and which also acts as a tumor suppressor. We have also developed quantitative microscopy-based measurement tools to track signaling processes and cell functions. Phase 1 of the proposal will demonstrate the overall feasibility of using a microscopy-based siRNA strategy to investigate multiple cancer-related cell functions. Phase 2 will address the questions posed above using an expanded set of 6000 siRNAs and a focus on six cell-types, 3 non-transformed and three breast cancer epithelial cell lines. We will screen to identify signaling siRNAs that alter proliferation, cell migration or endocytosis and then utilize follow-up studies with live cell biosensors that we developed to measure the duration of different cell cycle phases, as well as migration velocity and other kinetic parameters. We will then link genes that alter these cell functions to a subset of cancer-relevant signaling pathways using secondary siRNA screens. Based on these functional and signaling datasets, we will create a modular map of signaling systems using clustering methods. We will experimentally test the predictive power of modular maps using perturbations with pairs of effective siRNAs. We will show if and how modularity in a signaling system can be used to predict how cell functions can be manipulated using combinations of siRNAs and learn if and what distinguishing features exist that define modularity of signaling systems in cancer versus noncancer cells. This will likely lead to the identification of new cancer drug targets and new therapeutic strategies.

College of Pharmacy, Ohio State University

1 R33 CA114304-01A2

MicroRNA is a newly discovered class of endogenous, small interfering RNA. MicroRNA binds to messenger RNA and translationally represses protein levels. While over 300 microRNAs have been discovered in humans alone, their biological function, targets, expression levels and role in disease remain largely unknown. A role between microRNA expression and carcinogenesis has been proposed. There is a lack of sensitive, high-throughput methodologies to monitor the expression of microRNAs. microRNA are challenging molecules to quantify because the microRNA precursors consists as a stable hairpin and the mature microRNA is only 22 nucleotides in length. We propose to evaluate sensitive and specific real-time PCR assays to quantify the expression of the mature and microRNA precursors. The microRNA expression will be analyzed in a number of important biological conditions relating to human cancer. The microRNA expression will be determined in specific sections of cancer and normal tissue isolated by laser-capture microdissection. The expression of mature and precursor microRNAs will be compared to using real-time PCR and a cDNA micro array. microRNA expression will be studied in clinical samples of human pancreatic cancer. The unparallel sensitivity and specificity of real-time PCR as applied to this new and exciting class of regulatory RNAs should propel the field into new directions not only in cancer but also in other areas of human health.

Sidney Kimmel Cancer Center

1 R33 CA118602-01

The molecular complexity and in vivo inaccessibility of most tumor cells within solid tumors can greatly limit genomic- and proteomic-based discovery of useful targets for tumor-specific imaging and therapeutic agents in vivo . To overcome endothelial cell (EC) barriers and achieve more effective targeting and penetration into solid tumors, we shift analytical focus from the tumor cell to the vascular EC surface and its caveolae in direct contact with the circulating blood. To reduce data complexity to a meaningful subset of targetable proteins expressed on the EC surface, we will use tissue sub-cellular fractionation, novel multimodal mass spectrometric analysis, in silico subtraction, and bioinformatics interrogation of structure and function to unmask, from the >100,000 proteins in the tissue, those few intravenously accessible proteins differentially expressed on vascular endothelium in human renal tumors. This technology and overall approach has been validated in rodent solid tumors whereby new vascular targets have been uncovered permitting tumor-specific imaging, penetration, and effective radio immunotherapy (Nature, 429:629-35, 2004). But, currently very little is known about the expression of proteins in tumor neovascular endothelium, especially in human tissue. We now wish to apply our new technology to map comprehensively the proteome of luminal EC surfaces and caveolae in human renal tumors in vivo . It is likely that human tumors will express a different constellation of proteins on tumor neovasculature not yet uncovered or induced in animal models. To this end, we propose the following specific aims: 1) To use novel tissue sub fractionation and proteomic analytical approaches to map comprehensively vascular EC surfaces and caveolae in human renal tumors vs. matched normal renal tissue to unmask candidate tumor-induced/associated vascular proteins. 2) To create new antibodies to newly-discovered human renal tumor EC targets and to use antibodies as probes to validate the expression of tumor-induced/associated proteins at the EC surface and its caveolae in human tissues and thereby to assess the degree of target specificity for the neovasculature of human solid tumors. Such mapping may also elucidate the effects of the tumor on the developing vascular endothelium and yield important tumor-specific vascular targets for improving noninvasive diagnostic imaging and therapy as well as yield new diagnostic and prognostic markers for the molecular classification of tumor biopsies.

Ellis Fischel Cancer Center, University of Missouri, Columbia

1 R21 CA123018-01

Hypermethylation of promoter CpG islands plays a prominent role in cancer. In partnership with alterations in histone acetylation/methylation, this epigenetic event establishes a repressive chromatin structure that leads to silencing of key cancer-related genes. The occurrence of DNA methylation within the genome is not random, but rather patterns of methylation are generated that are gene and tumor type specific. How DNA methylation patterns are established is still poorly understood. Since various transcriptional factors or regulators are found in association with DNA methyltransferases (DNMTs) in vivo , we hypothesize that: 1) Oncogenic transcription factors can recruit DNMTs to target gene promoters and define a unique epigenetic signature in tumor cells; 2) Dissecting such complex epigenetic hierarchy will identify novel molecular targets for diagnosis, prognosis and therapeutic intervention. To test our hypothesis, we developed a high throughput technique for genome wide analysis of DNA methylation associated with specific proteins such as histones, transcription factors or any DNA binding proteins. The new approach named ChlP-Chop-DMH will combine both genome wide location analysis (also known as ChlP-on-Chip) and Differential Methylation Hybridization (DMH) analysis, two emerging technologies used in epigenetic research. The proposed method has distinct advantages over current protocols: first, this method directly examines the in vivo interaction of specific proteins with methylated DNA throughout the genome; second, this method may uncover novel biological properties of transcription factors; third, this method can be applied to discover novel epigenetic biomarkers relevant to tumorigenesis. In preliminary studies, we have verified the utility of this method with methylated histone H3 at lysine 9 and lysine 4 in human cancer cells. In the R21 phase, we will continue minor refinement of the method and pursue three aims: 1) Improve and optimize the ChlP-Chop-DMH method for analyzing genome wide association of DNA methvlation with histone modification; 2) Utilize the proposed method to investigate the association of DNA methvlation with chromatin remodeling factors: 3) Show proof-of-concept using the array to examine primary non-Hodgkin’s lymphomas (NHLs). In this development phase we will focus on the sensitivity, reproducibility and accuracy of the proposed method. In the R33 phase, our goal is to utilize the technology to test biological hypotheses. We will fully implement the method and pursue these aims: 1) Discover epigenetic target genes associated with known oncogenic transcription factors c-Myc and BCL6: 2) Validate the identified epigenetic targets and investigate the regulatory role of the associated oncogenic transcription factors. This systematic approach will provide a powerful tool for future mechanistic studies as well as cancer diagnosis.

Chemistry Department, Georgia State University

1 R21 CA123329-01

Malignant transformation is often associated with alteration of cell surface carbohydrates. The expression or1 over-expression of certain carbohydrates, such as sialyl Lewis X (sLex), sialyl Lewis a (sLea), Lewis X (Lex) and Lewis Y (Ley), has been correlated with the development of certain cancers. These cell surface carbohydrates can be used for cell-specific identification and targeting of carcinoma cells. Recently, we have developed boronic acid-based small molecule lectin mimics (named boronolectins) that can recognize certain carbohydrates with selectivity. The same or similar methods can be used for the preparation of lectin mimics for a wide variety of carbohydrates. The long-term goal of this project is the development of conjugates of boronolectin-MRI contrast agents as biomarker-directed cancer imaging agents. Specifically, such conjugates can be used for the delivery of MRI contrast agents based on cell-surface carbohydrate biomarkers. In the R21 phase of this application, we plan to study the feasibility of this approach by (1) synthesizing boronolectin-MRI contrast agent conjugates using a boronolectin which is known selectively bind to sialyl Lewis X, (2) studying their ability to bind to cells with the target carbohydrate biomarkers, and (3) examining their ability to image implanted tumors in both an ex vivo and in vivo models. If the R21 phase is successful, in the R33 phase we plan to expand our biological evaluation to include tumors implanted at different positions, and to search for other lectin mimics that can bind specifically for other important carbohydrate-based cancer biomarkers. In addition, we also plan to examine the cytotoxicity of the boronolectin-MRI contrast agent conjugates. These small molecule-based recognition/delivery systems may have the following advantages over antibody-based systems: (1) greater stability during storage and in vivo ; (2) lower propensity to elicit undesirable immune responses, (3) easier conjugation chemistry, and (4) more desirable pharmaceutical properties.

Department of Internal Medicine, University of Texas Southwestern Medical Center at Dallas

1 R21 CA114157-01

Within the United States, 170,000 new cases of lung cancer are diagnosed per year. Over 60% of these patients will die within one year, making lung cancer the largest cancer killer of both men and women. The correct histopathological diagnosis of a tumor is critical in determining the appropriate treatment. However, precise classification of tumors remains a significant biomedical challenge. Furthermore, tumors of similar histology can have different clinical outcomes, stressing the need for more detailed molecular classifications. Generation of ligands specific to receptor(s) on a surface of a lung cancer cell will impact clinical issues including functional diagnosis. Our overall goal is to generate a panel of cell-specific molecules that could be used to classify tumor types and utilize these cancer specific reagents in a high throughput diagnostic assay. Using phage display technologies, my laboratory has developed platform methodologies to isolate peptides that bind to and mediate uptake into specific cell lines. We have identified cell-specific targeting peptides for 25 different cell types, including 4 lung cancer cell lines. The isolated peptides display remarkable cell specificities, even among similar cells, and are able to discriminate between normal and cancerous cells as well as different lung tumor cells. This high discriminating power suggests that peptides could be identified that selectively bind to different tumor types, even those with similar classifications. We propose to expand this panel of lung specific reagents by isolating cell targeting peptides for 4 different lung cancer lines and then utilize these peptides as diagnostic reagents. These peptides will be assayed for affinity and cell-specificity. We will remove the peptides from the phage backbone and synthesize peptide scaffolds that retain their affinity and cell-specificity. We will develop a high throughput fluorescent assay based on peptide binding that will allow for a more molecular classification of lung cancer samples. The assay can be multiplexed so that multiple binding events can be examined on a single sample. At the end of this pilot project we will have developed a novel diagnostic platform that can be expanded to clinical samples. Furthermore, the technologies developed here can be applied to other forms of cancer.

Department of Engineering, Yale University

1 R21 CA112144-01A1

This application proposes the evaluation for chemotherapy applications of a novel drug delivery technology based on injectable emulsions of perfluorocarbon droplets containing therapeutic agents. The micron-size droplets dispersed in the emulsions are superheated, i.e., they are kept slightly above their boiling point. The droplets can be vaporized by exposure to ultrasound, which permits the spatially and temporally controlled release of their drug content into a target region. The droplets are encapsulated with surfactants protecting them from the mechanical stresses due to inoculation and circulation in the bloodstream. Therefore they do not vaporize spontaneously, but only when triggered externally. The use of ultrasound as a triggering modality enables the integration of targeted delivery and imaging. Controlled and localized release is expected to increase the therapeutic index by minimizing systemic non-specific exposure to the drug. The project is a collaboration between Yale University, where the superheated emulsion technology was invented and drug delivery applications were proposed, and the University of Michigan, where the use of superheated emulsions has been proposed for occlusion therapy and proven in vivo . The two groups, with complementary interests and expertise, propose to explore the capabilities of the technology in a biologically relevant setting, i.e., the transport and release in vivo of paclitaxel, a highly cytotoxic chemotherapy agent. The project will comprise investigations in various animal models, to clarify different aspects of the delivery technology, as well as the utilization of various imaging techniques to confirm and quantify occurrence and location of droplet activation. Specific aims of the R21 phase are (1) To develop the emulsion manufacturing technology, investigating and optimizing droplet size, formulation, drug loading efficiency, ultrasound sensitivity, (2) To investigate the emulsions in vitro , simulating circulation in the bloodstream; determining the ultrasound intensity required to trigger the droplets and the triggering efficiency; and verifying that anticancer drugs retain their cytotoxicity after they have been released, and (3) To ascertain in vivo that droplets can circulate without undergoing spontaneous vaporization and yet be triggered to vaporize using ultrasound.

Mayo Clinic College of Medicine, Rochester

1 R33 CA112070-01A1

Antibodies provide superior targeting capabilities to a variety of therapeutic agents. Several technologies have greatly facilitated the initial identification of a variety of antibody reagents, including scFv and Fab antibodies, with virtually any possible specificity. However, lead antibodies often require further optimization to maximize their therapeutic performance: optimization of antibody expression and folding in relevant cells, and optimization of the affinity of the antibody for the target antigen. The development of promising targeting antibodies against cancer often languishes at this bottleneck. Therefore, technologies to facilitate antibody adaptation and optimization are urgently needed. Antibody optimization is best achieved by the randomization and subsequent selection of antibody mutants for the desired phenotypes since efficient rational design of antibodies is currently not feasible. Polypeptide display (e.g., phage display) is a powerful technology for the generation and screening of libraries of mutant polypeptides for a phenotype. A eukaryotic display technology that employs the efficient protein synthesis and quality control system of eukaryotic cells would best optimize the therapeutic parameters of targeting antibodies. We have recently demonstrated the feasibility of a retrovirus, avian leukosis virus (ALV), as a viral platform for the display of a variety of eukaryotic polypeptides including scFvs, and the efficient generation and selection of a peptide library in eukaryotic cells. The goal of this R33 application is to demonstrate the efficiency of using the ALV display technology for the optimization of the scFv scaffold for efficient folding and expression in eukaryotic cells and for generating a panel of scFvs with a range of affinities for their target antigen with an optimized scaffold. We will use the ALV display technology to optimize two scFvs with known specificity for tumor neovasculature: an anti-laminin scFv (L36) that inhibits angiogenesis in a variety of assays, presumably due to the exposure of laminin in the extracellular matrix during tumor neovessel formation; and a scFv that recognizes a VEGF:receptor complex (LL4) specific to endothelium in tumor neovessels. The ability of the nonoptimized and the optimized scFvs to target a therapeutic agent to tumor neovessels will be assessed using oncolytic measles viruses. Specifically, we aim to: 1. Create ALV display libraries of L36 and LL4 scFv mutants by error-prone PCR. 2. Screen the ALV display libraries of scFv mutants to generate a panel of L36 and LL4 scFv mutants with a range of known affinities (from fM to nM) for their target antigen and with an optimized scFv scaffold. 3. Generate recombinant measles viruses displaying nonoptimized and optimized targeting scFvs and compare them with respect to ease of production, efficiency of scFv display, particle to infectivity ratios, replication kinetics, and homing properties to tumor neovessels.

Department of Cellular and Molecular Medicine, University of California, San Diego

1 R33 CA114184-01

Alternative splicing is a permanent feature in higher eukaryotic cells and understanding of how alternative splicing alters the composition and function of the proteome represents a major challenge in the post-genome era. For cancer research, unique mRNA isoforms may provide a robust set of biomarkers for diagnosis and prognosis, and cancer-specific mRNA isoforms may serve as discriminating targets for effective therapeutic interventions. Furthermore, understanding of how splice choice is made and regulated in development and disease is a fundamental issue in cancer cell biology. An mRNA isoform-sensitive microarray technology would be ideally and timely suited for addressing a wide range of clinical and mechanistic questions regarding alternative splicing. In the past IMAT funding period, we have developed a unique and novel technology platform to attack the splicing problem. After a systematic and substantial effort in database construction and experimental development, the technology is now matured, and its superiority in reproducible measurement of mRNA isoforms under a variety of conditions has been demonstrated. Unique to the splicing array is the need to progressively enlarge the database for accurately annotated mRNA isoforms and preparation of corresponding oligo sets for measurement. We are therefore seeking IMAT support to put the technology in practical use and let the research community to take advantage of the technology development. We have three specific goals for the next phase in applying the emerging technology for molecular analysis of cancer. (1) We plan to use the technology to identify unique mRNA isoforms associated with prostate cancer. We will survey existing prostate cancer cell lines untreated or treated with androgen and estrogen as well as cancer tissues at different malignant stages to identify tumor-specific and hormonal regulated alternative splicing. (2) We propose to apply the technology to address mechanisms of splicing regulation by identifying direct targets for a large number of splicing regulators in knockdown and knockout cells. (3) Along with the proposed technology applications, we will progressively enlarge the high quantity databases coupled with the technology development and continue to improve and enlarge the database by adding new features and functions and develop linked software for splicing array data analysis.

Nimblegen Systems, Inc.

1 R21 CA116365-01

We propose to apply the emerging technology of chromatin immunoprecipitation on microarrays (ChIP chips) to identify the direct targets of transcription factors (TFs) known to be involved in breast and colon cancers. During the R21 phase of the project, we plan to identify these targets using a newly developed set of oligonucleotide microarrays that contain 15 million 50mer probes that tile through the non-repetitive sequence of the human genome at 100 bp resolution. We will validate that our ChIP chip protocols work with this new tiling array set. We will also develop and refine two new protocols, microarray reuse and 4-color hybridizations, that will allow economical use of this tiling array set so that it will be a more practical tool for research labs. In the R33 phase of this project, we will use this tiling array set to identify the direct targets of 9 TFs known to be involved in breast and colon cancers. Once we have collected the direct targets of these TFs, we will develop custom arrays that have probes that tile through all of the binding sites in the direct targets. We will use these TF focused screening arrays to study binding patterns in Icelandic breast and colon cancer samples to determine whether TF binding patterns could be a useful means of classifying tumors.

Department of Molecular Genetics, Cleveland Clinic Lerner College of Medicine, Case Western Reserve University

1 R21 CA116060-01

It is commonly accepted that the processes underlying the development of cancer are best modeled in vivo . A plethora of methods have been developed for genetic dissection of biological processes in somatic cells in culture; however, the adaptation of these techniques to in vivo use has been very limited. Our earlier experience indicates that reversible promoter-insertion mutagenesis offers a number of advantages over the currently used forward genetics techniques for cultured cells. The most important benefits are the dominant nature of the mutations, and the ease of identifying the affected gene and establishing the causal link between the mutation and the phenotype. The principal limitation for the adaptation of this technique to in vivo studies is the delivery of the insertional mutagen (typically, a retroviral vector) to the desired cell type. Based on our recent experience and published data from others, we propose a method to circumvent this limitation via the use of specially engineered transposon vectors. We propose to apply our technique to study the genetic events underlying melanoma development. We will screen for genetic events that cooperate with oncogenic Ras (commonly found in melanomas) in this process. We will confirm the properties of the proposed insertional mutagens in culture and then proceed to establish mouse strains suitable for mutagenesis. Upon verifying the properties of our constructs in vivo , we will selectively mutagenize melanocytes in transgenic animals and will test the functional link between tumorigenesis and the inserted promoter. Confirmed hits will be the subject of future investigation, while the technique itself will become available for dissection of various biological phenomena in general and aspects of tumorigenesis, in particular.

Pathology Department, Yale University School of Medicine

1 R21 CA116079-01

Head and neck cancer is a common disease worldwide, and more than 40,000 cases are diagnosed annually in the United States. The five-year survival rate is roughly 50% and has not improved over two decades. This research project focuses on the development of a new cancer biomarker paradigm based on the global epigenetic status of head and neck tissues. In many laboratories, DNA methylation is being examined as a means for early diagnosis of cancer. Recently DNA hypomethylation has been found to be a promising biomarker for the more advanced and aggressive stages of cancer, and there is an urgent need for analytical tools that will sample comprehensively the abnormal “methylome” hidden within the vast landscape of DNA repeats, in addition to the abnormal methylome of promoter-associated CpG islands. This grant application focuses on the validation of a novel microarray-based approach for epigenetic biomarker discovery and cancer classification that is complementary to other methods currently in use. The new method is relatively simple and has the unique capability to sample the methylation status of the majority of DNA repeats, as well as gene promoters, generating very large data sets. We will perform microarray-based methylation profiling using a statistically meaningful set of tumor samples from head and neck cancer patients. In addition, we will utilize the microarray-generated methylation profile information to identify clinically relevant subtypes of head and neck cancer, using a variety of analytical approaches. We will refine a subtype classification algorithm by inclusion of additional datasets available for the same samples, including allele gains and loses revealed by array-CGH and HPV infection status, as well as relevant patient clinical data. We will use this new information to address urgent diagnostic and prognostic clinical needs for improved classification of head and neck cancers in relation to diagnosis of early developmental stages, assessment of aggressiveness, likelihood of metastasis, and risk of recurrence after surgery.

Department of Cancer Genetics and Epigenetics, Burnham Institute

1 R33 CA112885-01

Distortion of the cell genome characterizes neoplastic transformation. Genetic alterations that occur in tumor cells lead to activation of positive regulators of cell growth or survival and inactivation of factors that suppress these processes. A particular type of genomic alteration, chromosomal segment copy number imbalance, plays a significant role in malignant transformation: chromosomal deletions may inactivate tumor suppressor genes, while chromosomal segment amplifications may increase the gene dosage of oncogenes. In this study, we propose to apply a new technique, Comparative Hybridization of AP-PCR Arrays (CHAPA), which was developed in our laboratory, for high-resolution profiling of breast tumors for DNA copy number alterations. This will allow the detection of single DNA copy number losses or gains at thousands of sites throughout the genome of the cancer cells (Specific Aim 1). We hypothesize that such genetic signatures may embrace the information on what cancer genes were responsible for the development and progression of each tumor and, consequently, the resulting pathologic behavior of tumor cells and their responsiveness to treatment. This general hypothesis will be tested by the analysis of genetic profiles to differentiate breast tumors according to their pathways of tumorigenesis (known or novel) and by the analysis of genetic profiles of breast tumors in association with their clinicopathologic characteristics, recurrence, and patient’s survival to reveal genetic markers for cancer diagnosis and prognosis. Once frequent (common for independent tumors) genomic alterations have been identified, they will be compared with the loci known to play a role in breast cancer development. The genetic aberrations in chromosomal regions that do not contain known cancer genes will be selected for further characterization with the ultimate goal to identify the underlying novel cancer genes (Specific Aim 3). These experiments will provide a comprehensive view on the role of genetic aberrations in breast tumorigenesis. They will also help to identify genetic markers for breast cancer diagnosis, development, and prognosis and facilitate the identification, mapping, and eventual isolation of novel cancer genes.

Department of Neurosurgery, Johns Hopkins University

1 R21 CA112148-01

Each year, approximately 185,000 people in the United States are diagnosed with a primary or metastatic brain tumor constituting the third leading cause of death in young adults ages 20-39. Among brain tumors, malignant gliomas are the most common and aggressive malignancies. Gliomas seem to be capable of inducing T-cell apoptosis through the Fas/Fas-ligand (Fas-L) pathway. This mechanism allows them to circumvent immune surveillance by decreasing cell mediated immunity. Brain tumor therapy using immune modulators such as interleukins (IL) increases the recruitment of active T lymphocyte populations and has proven to be an effective strategy in experimental models of the disease. However, due to tumor-secreted Fas-L, the peritumoral T-cells recruited by IL are activated through the trans-membrane Fas receptor, which initiates the caspase-3 mediated apoptotic cascade. Using RNA interference (RNAi) techniques, mRNA from tumor-derived Fas-L could be silenced and T cell apoptosis could be decreased, thus improving antitumor responses and potentiating the effect of interleukin therapy. RNAi sequences can be delivered by retroviruses, guaranteeing a constitutive transfection. In this proposal, the effect that treatment with Fas-L RNAi sequences delivered via retroviruses has on experimental gliomas will be investigated. The effect of this treatment modality will be studied alone and in combination with locally delivered IL incorporated into injectable microspheres. Treatment with Fas-L RNAi sequences is expected to decrease levels of tumor-derived Fas-L, therefore decreasing the rates of T cell apoptosis and increasing the populations of peritumoral T cells. Such an effect should allow a more consistent cell-mediated anti-tumor response able to prolong survival in animal models. Furthermore, the efficacy of locally delivered IL microspheres should be enhanced. The potential benefit of Fas-L RNAi delivered in this fashion could be applicable to several malignancies that have the Fas/Fas-L pathway among their immune privilege strategies.

Pathology Department, University of Southern California

1 R33 CA111940-01

Colorectal cancers are widely believed to develop through an adenoma-cancer sequence. By this paradigm, all cancers should be preventable with surveillance and polypectomy. However, in most surveillance studies, some cancers inevitably appear only a few years after negative clinical examinations. It is uncertain how such “interval” cancers appear, but either an adenoma was missed during the last “negative” examination, or there was an unexpected “rapid” mode of progression. A large number of such interval cancers have been found during an ongoing clinical surveillance program of high risk individuals with germline mutations in DNA mismatch repair genes (MMR), or hereditary nonpolyposis colorectal cancer (HNPCC). These interval cancers provide unique opportunities to rigorously understand why prevention fails. The key emerging technology is a new capability that infers time from cancer mutations. A molecular tumor clock can quantitatively infer times since MMR loss, and ages of final cancer expansions—the more mutations in a cancer, the greater these intervals. Distinguishing between failure from inadequate surveillance (“missed adenomas”), and failure due to the unique biology of HNPCC colorectal cancers (“rapid histologic or genetic progression”), is critical for the development of more effective strategies to prevent and treat colorectal cancer.

Sloan-Kettering Institute for Cancer Research

1 R21 CA111942-01

The information required for adequate diagnosis, treatment and monitoring of cancers is so complex that a panel of measurements, used in sum, may provide the best answers. The concept is embodied in SELDI-TOF mass spectrometric (MS) peptide profiling, an emerging technique for serum based cancer detection. Even though SELDI has thus far only produced low complexity spectra, the patterns, when analyzed as groups, have the potential to create learning algorithms with diagnostic accuracies as good as or better than conventional biomarkers. We have developed a system to capture peptides on magnetic reversed-phase beads, followed by MALDI-TOF MS, to yield increasingly complex, yet very reproducible patterns. This has clear advantages, as more displayed peptides provide more opportunity to select unique patterns (‘barcodes’) for cancer subtypes and stages, and to predict and monitor clinical outcome. Extreme care has also been taken to standardize specimen collection, handling and storage to avoid the introduction of artifact. Pilot projects at MSKCC with a variety of malignancies suggest that peptide patterns thus obtained appear to hold information that may have direct clinical utility. The goals of this project are to (i) automate our prototype serum peptide profiling platform and implement machine learning methods that use the resulting peptide patterns (‘barcodes’) for sample classification [R21]; and (ii) to test the ‘barcode diagnostic’ model in a high-throughput setting, using well defined and carefully observed groups of thyroid carcinoma patients [R33]. R21 aim one is to automate serum sample processing and analysis; aim two is to automate all data processing, to examine pattern selection and sample class prediction methods, and to integrate all software platforms; aim three is to develop routine MALDI-TOF/TOF tandem MS sequencing of ‘barcode’ peptides. R33 aim one is to define reproducibility of serum patterns in patients with thyroid disease; aim two is to determine barcodes that can distinguish patients with thyroid cancer from those with benign thyroid nodules; aim three is to assess if serum peptidome barcodes can identify occult metastasis in a large group of thyroid cancer survivors.

Massachusetts General Hospital

1 R21 CA114149-01

The long-term goal of this research is to develop a novel fibroblast activation protein (FAP) sensing near infrared fluorescence reporter for early tumor detection and tumor classification. FAP is a cell surface antigen of reactive tumor stromal fibroblasts founded in more than 90% of epithelial carcinomas, but it is absent from epithelial carcinoma cells, normal fibroblasts, and other normal human tissue. Supporting tumor stromal fibroblasts are generally localized close to tumor vasculature, which is essential for early tumor development and growth. Thus it has been chosen as a target for monoclonal antibody based tumor therapy. FAP is not only a membrane protein but also a dipeptidyl peptidase. An imaging probe to report enzymatic activity and location of FAP could be extremely useful for early tumor detection. In this application we will develop a small molecular probe with ultra-sensitivity based on a unique class of fluorogenic chromophore that has significant changes in emission at different chemical states. Specifically, the probes emit no fluorescence in their initial intact state but become brightly fluorescent after specific proteolytic reaction. The newly developed low molecular weight, well defined fluorogenic probes are expected to have several advantages for imaging: a) fast tissue distribution allowing earlier imaging after injection, b) fast clearance allowing repeated imaging, and c) high likelihood of developing key candidates into clinically useful agents. The choice of FAP is based on its importance in tumor growth, invasion, and other processes in oncogenesis. Together with recent developments in fluorescence imaging technologies, this research is expected to ultimately result in clinical imaging agents with specificity for targeted enzymes. We believe that the developed approach can be used as a platform to design a broad spectrum of activatable molecular probes to image other amino peptidases in vivo .