GI SPORE Abstracts

Overall Abstract

UNC GI Cancer SPORE's unique goals emphasize multidisciplinary translational research that spans and links the population, clinical, and basic sciences. To decrease the burden of colorectal cancer on the patient population at large we will: (1) determine the clinical importance of new and existing targets on the colorectal cancer population (2) define the relationship of these molecular abnormalities to racial disparities (3) evaluate how best to manipulate these molecular targets for therapy, and (4)verify the target manipulation and effect in a group of colorectal cancer patients.To decrease colorectal cancer mortality a better understanding of the relationship of laboratory discoveries in signaling, oncogenesis and other molecular changes to the clinical setting is needed. We have developed projects that will allow us to accomplish these goals through the strong multi-disciplinary collaboration that already exists between clinical, laboratory and population-based scientists in the UNC Lineberger Comprehensive Cancer Center. The UNC GI Cancer SPORE consists of five projects, four core resources, a career development program, and a developmental research program.

Project 1: Prognostic and Predictive Factors in Outcomes of Patients with Colorectal Cancer: A Population-Based Study
Robert Sandler, M.D., M.P.H., Principal Investigator
Temitope Keku, Ph.D., Co-Principal Investigator
There is now evidence that molecular characteristics of colorectal cancer can influence prognosis and predict response to therapy. The proposed study is motivated by the belief that better information on prognostic and predictive factors will make it possible to tailor therapy to maximize benefits and reduce cost, and by a desire to understand racial disparities in colorectal cancer mortality. The specific aims of the study are: (1) To determine which patient, treatment and molecular characteristics of colon tumors are independent predictors of prognosis. (2) To determine interactions between tumor characteristics and treatment factors and the response to therapy. (3) To determine whether racial differences in tumor characteristics are responsible for the worse 5-year survival in colorectal cancer in blacks. The proposed study will take advantage of data collected in a prospective, population-based study that will obtain exceptionally detailed information on patient characteristics, treatment, outcomes as well as tumor blocks on 1000 newly-diagnosed colorectal cancer patients (600 whites and 400 blacks) drawn at random from a diverse, mixed-race, 22 county area in North Carolina. The specific molecular characteristics will include expression of protein products of genes involved in cell cycle, cell growth, apoptosis, and cellular adhesion such as p53, k-ras, calcium binding protein S100A4, cyclin D, IGFII, TGF-bRII, MLH1, and MSH2. These markers will be evaluated using multitumor tissue array blocks. In addition, the study will use DNA from microdissected blocks to evaluate loss of heterogeneity (LOH) on chromosomes 2p, 5q, 17p and 18q, and will assess microsatellite instability (MSI) using BAT25, BAT26 and D17S250, D5S346, D2S123.

Project 2: Molecular Changes in the NFkB Pathway in Response to Chemoradiation Therapy in Rectal Cancer
Al Baldwin, Ph.D., Principal Investigator
Carolyn Sartor, M.D., Co-Principal Investigator
Although early diagnosis for colorectal cancer has been significantly improved, the occurrence of locally advanced colorectal cancer is still a major medical problem since these patients have extremely poor prognosis. Clearly, new approaches for treatment of colorectal cancer are needed. Two major obstacles which exist relative to standard radiation and chemotherapy protocols are: (i) resistance of colorectal tumors to treatment and (ii) dose limitations of treatment due to fibrosis of normal intestinal tissue. Radio- and chemoresistance are characterized either as pre-existing resistance, based on constitutive expression of certain proteins such as MDR1 or anti-apoptotic proteins such as Bcl-2 and on loss of pro-apoptotic factors such as p53, or as inducible resistance. Inducible resistance is a relatively poorly characterized phenomenon in which tumors transiently induce resistance following exposure to treatment. In several studies, tumor response correlates with induction of apoptosis. Thus, it is likely that one major component of tumor chemo/radioresistance is suppression of apoptosis. We have reported that the transcription factor NF-kB provides a powerful anti-apoptotic mechanism through its ability to transcriptionally regulate genes encoding proteins which suppress apoptosis. Importantly, NF-kB activation is basally detected in colorectal tumor tissue (although expression may predominantly be localized in tumor-associated macrophages) and is strongly activated in tumor cells following radiation or chemotherapy exposure. Based on these findings, we have shown that inhibition of NF-kB through a gene delivery approach strongly potentiates chemotherapy-induced experimental colorectal tumor cytotoxicity through the induction of apoptosis. Furthermore, we have used an FDA-approved drug (PS-341, a proteasome inhibitor) which strongly blocks NF-kB activation to enhance colorectal tumor xenograft responses to both radiation and to CPT-11 based chemotherapy.

We hypothesize that radiation-induced NF-kB activation and associated downstream transcriptional responses will occur in colorectal tumors and that this will correlate with a decreased therapeutic response by providing an anti-apoptotic signal. To test our hypothesis, we propose to: (i) determine whether radiation-induced activation of NF-kB and associated induced genes occurs in pre-operative radiochemotherapy and determine whether this responses correlates with clinical response and (ii) determine the ability of PS-341 to modulate NF-kB-dependent responses and to measure the toxicity of PS-341 delivered with pre-operative radiochemotherapy.

Additionally, these studies will measure NF-kB activation and induction of transcription factor-dependent transcription responses in normal intestinal tissue following radiochemotherapy as a potential molecular correlate for cancer therapy-induced intestinal fibrosis. We also hypothesize that adjuvant radiochemotherapy responses will be improved and overall toxicity reduced through the use of more specific inhibitors of the NF-kB pathway. Thus, we will measure the efficacy of NF-kB inhibitors as an adjuvant approach with radiation or with the newest combination of chemotherapy regimens on different experimental tumors including liver metastases. These phase I clinical studies and basic translational experiments have the potential to significantly improve therapeutic approaches for advanced colorectal cancer.

Project 3: Investigation of ERBB signaling in Colorectal Cancers during Liver Metastasis
David Threadgill, Ph.D., Principal Investigator
Benjamin Calvo, M.D., Co-Principal Investigator
Colorectal cancer afflicts 135,000 Americans per year and 38% of these patients will die of disseminated disease most commonly to liver lung and bone. Of the patients that die of this disease, 70% have liver metastases and a significant 10% have liver-only disease. Even in those patients with metastases to multiple organ sites, the extent of liver disease remains the primary determinant of survival.

Over the last two decades we have empirically learned that patients with liver only metastases have improved survival when treated aggressively. Untreated, patients with hepatic only metastases have a median survival of only 12-21 months and the five year survival of patients with unresected metastasis is close to 0%. In sharp contrast, resection of metastases in patients with liver only disease yields five year survival rates of 20-40% and 10 year survival rates of 20%. Furthermore, the most common site of disease recurrence after resection is the liver. Consequently, liver metastases are a primary determinant of survival in patients with stage IV colorectal cancer.

Building upon preliminary data linking ERBB receptor activity to colorectal cancer progression and metastasis, we hypothesize that small molecule ERBB inhibitors, if optimally employed, will retard the growth and dissemination of metastatic colorectal cancer. We also hypothesize that colorectal cancer can also arise independently of ERBB and that an understanding of these mechanisms will allow us to design better therapies. Thus, using a combination of clinical smaples and pre-clinical mouse models, we propose to investigate the mechanism of how metastatic colon cancer uses EGFR and other ERBB receptor signaling to establish residency in the liver and to identify markers for response to dual EGFR/ERBB2 inhibitors during treatment of metastatic lesions. Experiments are planned to identify transcriptional profiles unique to EGFR independent colorectal cancer development.

Project 4: Targeting the RAS>ERK Pathway for Colorectal Cancer Treatment
Channing Der, Ph.D. Principal Investigator
H.J. Kim, M.D., Co-Principal Investigator
Mutations in Ras are associated with 50% of colorectal carcinomas, indicating the importance of aberrant Ras activation in tumor development and progression. Recently, mutations in B-Raf, the downstream target of Ras, have been identified in 10% of colorectal carcinomas. The presence of B-Raf mutations in tumors distinct from those with Ras mutations indicates that these mutations are genetically equivalent, such that either one confers a similar advantage. These data support the critical contribution of the Raf>MEK>ERK mitogen-activated protein kinase cascade in Ras-mediated oncogenesis. Currently, pharmacologic inhibitors of two kinases in this cascade, Raf and MEK, have been developed and are under evaluation in clinical trials. Such target-based drugs are believed to represent the key future direction for anti-cancer drug discovery. However, one major complication that has slowed the clinical development of target-based anti-cancer drugs (e.g., epidermal growth factor receptor inhibitors) is continued uncertainty regarding whether aberrant activation of the target alone is sufficient to define the patient population that will be responsive to these drugs. This uncertainty is based, in part, on the fact that the presence of an altered target may simply have a correlative, rather than a causal, role in oncogenesis. Based on observations in preclinical models, this will also be a concern for efforts to evaluate the clinical efficacy of anti-Ras therapies. It is likely that Ras mutation status alone will not be sufficient to define the subset of colorectal cancers that will be responsive to inhibitors of Ras signaling, specifically to inhibitors of the Raf and MEK protein kinases. Instead, we hypothesize that other approaches, such as microarray gene profiling, will be needed to determine the subsets of Ras mutation positive colorectal cancers that will be responsive to anti-Raf or anti-MEK therapy. Therefore, the broad goal of this project will be to determine whether a group of patients with colorectal carcinomas that harbor mutated Ras show gene expression profiles that may have clinical relevance in predicting sensitivity to anti-Ras and anti-Raf/MEK therapeutic strategies.

Project 5: Determination of the Role of Fucosyltransferases in Colorectal Cancer Initiation and Progression
Brent Weston, M.D. Principal Investigator
Robert Sandler, M.D., M.P.H., Co-Principal Investigator
Selectin-mediated cell adhesion has been implicated in the metastasis of colorectal carcinoma (CRC), and the carbohydrate ligand components sialyl Lewis x (sLex) and sialyl Lewis a (sLea) on tumor cells have long been considered markers for development and progression of human carcinoma. Recent data showing important roles for inflammation in CRC pathogenesis point to selectin ligands as potential translational targets. Although corresponding glycosyltransferase expression is complex in malignant cells and tumor specimens, sLex and sLea synthesis appears to be largely controlled by the human a(1,3)fucosyltransferase gene families. Two of these genes are highly expressed in CRC: FUT3 and FUT6. Our group has shown that inhibition of sLex/sLea expression in CRC cells by antisense FUT3 sequences results in markedly reduced CRC metastases in nu/nu mice. Furthermore, antisense inhibition of FUT6-- which is often co-expressed with FUT3 in CRC and is inducible with inflammatory cytokines-- results in decreased carcinoma proliferation invitro and in vivo. Mutations in FUT3 and FUT6 have been described in diverse human populations, but no information has been available to determine the effect(s), if any, of these null phenotypes on development and/or progression of CRC. Similarly, expression at the transcript level has been limited by lack of adequate human data sets and samples. Over the past several years at UNC, SPORE investigators have helped assemble large CRC patient data sets with appropriate specimens. We propose to: 1. Identify FUT3 and/or FUT6 mutations associated with polyp and/or CRC development and examine potential interactions with use of non-steroidal anti-inflammatory drugs (NSAIDs) and other clinical variables; 2. Examine FUT transcript levels in polyps and CRC lesions at various stages of progression; and 3. Combine FUT antisense oligodeoxynucleotides with NSAIDs for experimental therapy of CRC in vitro and in nu/nu mice models. Our long term goal is extend the pre-clinical use of these agents to appropriate patient populations as our understanding of selectin ligand function grows.

Core 1: Administration
Joel Tepper, M.D., SPORE Director
Richard Goldberg, M.D., SPORE Co-Director
Robert Sandler, M.D., M.P.H., SPORE Co-Director
David Threadgill, Ph.D., SPORE Co-Director
Under Dr. Joel Tepper, SPORE Director, the Administrative Core supports the UNC GI Cancer SPORE's overall scientific/translational goals by providing leadership and day-to-day operations/administration. The SPORE Director leads the core. An Administrative Director, a Program Coordinator, and an Administrative Assistant comprise the staff. The Core organizes: the intra and inter-SPORE interactions, administrative/scientific oversight of all research projects, cores, and developmental programs; and the activities of the external, internal, and advocate advisory committees. This Core also monitors SPORE expenditures and addresses grant management issues.

Core 2: Genomics
David Threadgill, Ph.D., Core Director
Chuck Perou, Ph.D., Core Co-Director
The Genomics Core will provide experimental planing, scientific services and computational support for SPORE projects using DNA microarrays. The facility is contained within the UNC Genomics Core & Microarray Facility housed in the Lineberger Comprehensive Cancer Center and adjacent to the core director’s lab. It maintains centralized equipment for microarray production, utilization, and analysis. The Core also produces customized microarrays and maintains libraries of long-oligos and cDNA for printing the custom microarrays. To maximize efficiency and generate consistent quality results, the Genomics Core will provide complete microarray research services to SPORE projects. Investigators will isolate RNA and submit samples. The facility will perform quality control on all samples and prepare fluorescent probes for hybridization with either in-house produced or commercial micorarrays. The Genomics Core will provides computer hardware and analysis programs to collect and pre-process raw data before being transmitted to the Bioinformatics Core for databasing and analysis. The Genomics Core will initially support three of the SPORE projects. Its use will most likely expand to other projects and development studies as needed.

Core 3: Biostatistics and Bioinformatics
Joseph Ibrahim, Ph.D, Core Director
Fred Wright, Ph.D., Core Co-Director
The UNC GI Cancer SPORE Biostatistics and Bioinformatics Core was developed from the newest core resources at the UNC Lineberger Comprehensive Cancer Center. Substantial expansion in faculty recruitment, with an emphasis on genetic analysis, and an investment in hardware and associated databases has prepared the Center to launch an integrated yet expanded core resource to support the GI SPORE projects. Senior faculty recruits Joe Ibrahim and Fred Wright will provide new intellectual leadership for this core. Both have substantial experience in biostatistics, new areas of statistical genetics, integration with clinical research, and Cancer Center programs for translational research. In addition to expanded leadership, involvement from the Departments of Statistics and Biostatistics have expanded the Cancer Center’s and thus the UNC GI Cancer SPORE’s capabilities in this area.

A substantial investment in databases to store gene expression microarray data from multiple platforms and to import data from other institutions is already leading to productive translational research at the Cancer Center. Continued development of parallel clinical and epidemiologic databases for clinical trials and population-based data will be linked to our microarray databases via an honest broker model that will be overseen by the GI SPORE and the UNC Lineberger Biostatistic cores.

The five projects in the GI SPORE present interesting challenges, both as translational research and as biostatistical/bioinformatics problems. Complementary skills possessed by the Core faculty will lead to new approaches to gene expression data as well as cross-platform analysis. Input from the Core Directors and other members of this core is described within both the project write-ups and in this section.

Top-notch senior and promising junior faculty, an emphasis on translational and genetic research at the Cancer Center, and continuing investment in data management will provide an excellent resource for the UNC GI Cancer SPORE’s projects with large data sets. The Core will provide appropriate input into design, management, and analysis as the most promising lines of translational research are pursued.

Core 4: Tissue Procurement and Analysis
William Funkhouser, M.D, Core Director
John Woosley, M.D., Core Co-Director
The Tissue Procurement and Analysis Core will build on existing services within the UNC Lineberger Comprehensive Cancer Center and within the UNC Department of Pathology and Lab Medicine. This core will provide centralized tissue procurement, tissue processing, tissue storage, and tissue distribution of normal and malignant colorectal tissues from UNC Hospitals as well as maintenance and enhancement of the existing GI tissue bank. This Core will support the listed GI SPORE projects, working with both clinical and research investigators to meet their unique research needs.

The Tissue Procurement component is responsible for collection and freezing of surplus fresh tissues from the UNC Division of Surgical Pathology. All Tissue Procurement frozen specimens will be reviewed by a board-certified Anatomic Pathologist to ensure that representative frozen tissue is banked and distributed. Potential downstream frozen tissue handling includes routine frozen section stains, frozen immunophenotyping, frozen section laser capture microdissection, and mRNA extraction.

The Tissue Analysis component is responsible for collection and analysis of diagnostic paraffin blocks and slides following Surgical Pathology case diagnosis finalization. Potential downstream fixed tissue handling includes paraffin tissue microarray manufacture, routine paraffin section stains, paraffin immunophenotyping, paraffin section laser capture microdissection, immunophenotypic analysis, and immunophenotype digital image collation.
In conjunction with the Biostatistics and Bioinformatics Core (Core #3), customized HIPAA-compliant databases will be used to track release, disposition, and return of both frozen and fixed specimens, including frozen tissue, paraffin blocks, tissue section slides, and patient reports, providing a coordinated system of quality control, specimen tracking, and efficient specimen distribution of specimens to appropriate investigators.

This Core will implement policies and procedures as necessary to support the above services, to address relevant technical, medical, and legal issues, and to comply with relevant ethical standards as defined by UNC School of Medicine Institutional Review Board (IRB) and Federal HIPAA policies.
This Core will implement policies and procedures as necessary to support the above services, to address relevant technical, medical, and legal issues, and to comply with relevant ethical standards as defined by UNC School of Medicine Institutional Review Board (IRB) and Federal HIPAA policies.

Developmental Research Program
The UNC GI Cancer SPORE’s Developmental Research Program will promote novel translational gastrointestinal cancer research in clinical/translational science, population sciences, gene/molecular discovery, and other relevant areas. The program will fund two types of investigators: those with GI cancer research experience whose projects are central to the SPORE’s translational mission; and, investigators (or teams) with limited GI cancer experience whom we wish to attract to the field. The Developmental Research Program includes mechanisms for stimulating grant applications, evaluating and selecting projects, and monitoring progress. The mechanisms include consultation with and/or evaluation by the GI Cancer SPORE;’s senior leadership (Executive Committee), the External Advisory Board, and GI cancer Advocates Advisory Board. This program will use mechanisms that have been successful for other developmental research endeavors at the UNC Lineberger Comprehensive Cancer Center and UNC Chapel Hill, including the UNC SPORE in Breast Cancer.

Career Development
The SPORE program is one of the nation’s primary avenues for career development in translational research. The long-term success of national efforts to reduce GI cancer incidence and mortality rests in part on the ability of the GI Cancer SPORE programs to attract and build the careers of talented young faculty and to redirect the more advanced faculty to translational GI cancer research. The Career Development Program offers unique opportunities to develop and support translational research scientists. The UNC GI Cancer SPORE will use a process with a proven track record to promote these goals.