Skip to Content
Translational Research Program (TRP)
Contact CIP
Show menu
Search this site
Last Updated: 10/17/18

SPORE in Gastrointestinal Cancer

Vanderbilt University

Principal Investigator:
Robert J. Coffey Jr., M.D.


Robert L. Coffey, MD
Professor of Medicine and Cell and Developmental Biology
Vanderbilt University School of Medicine
2213 Garland Avenue
10415-G MRB IV
Nashville, TN 37232
Tel: (615) 343-6228
Fax: (615) 343-1591


This proposal represents the second competing renewal of Vanderbilt's GI SPORE. This SPORE continues to focus on colorectal cancer. the second leading cause of cancer deaths in the US, where it affects more men and women than all other gastrointestinal malignancies combined. Over the last four years, Vanderbilt’s GI SPORE has made discoveries and advances that hold great promise toward improvement of the management of individuals with colorectal neoplasia. These include 1) discovery of a novel Wnt antagonist (pyrvinium) and its target (casein kinase 1a), 2) discovery of a biologically-based, prognostic gene signature for colorectal cancer, 3) evidence that p120 acts as a tumor suppressor in colorectal cancer, 4) development, biological validation and clinical implementation of novel molecular imaging modalities to predict early response to treatment and 5) further development of a unique biorepository of colorectal adenomas, as well as matched normal rectal mucosa and bodily fluids (serum and urine).

Our potential for continued success is high based on 1) productivity during the past funding cycles, 2) strong and highly integrative institutional support, 3) recruitment of talented investigators to the field of GI cancer through career development and pilot project funding, 4) access to unparalleled resources for proteomics, drug discovery and small animal imaging, 5) a team of highly interactive clinical investigators and basic scientists working together in a collegial environment and 6) strong inter-SPORE, pharmaceutical, national and international collaborations.

After a rigorous internal and external review process followed by consultation with NCI SPORE administrators, we propose three new projects and continuation of one project.

  • Project 1. Multimodal Imaging & Targeted Therapeutics of Stem Cell-Derived Colon Cancer
  • Project 2. Targeting K-RAS in Colorectal Cancer
  • Project 3. Molecular Markers of Colorectal Cancer Recurrence
  • Project 4. Genetic & Epigenetic Markers of Colorectal Adenoma Recurrence
  • Core A. Administrative
  • Core B. Translational Pathology & Imaging
  • Core C. Biostatistics & Bioinformatics
  • Career Development Program
  • Developmental Research Program


Project Leaders:
Robert J. Coffey Jr., M.D. (clinical)
H. Charles Manning, Ph.D. (basic)

This is a new project that builds upon our combined expertise in ErbB signaling and molecular imaging. We have discovered that Lrig1, a pan-ErbB inhibitor, marks a novel, largely quiescent population of colonic stem cells. Using Lrig1-CreERT2;Apcflox/+ mice, we have developed a robust, highly tractable, clinically relevant, stem cell-derived mouse model of colon cancer. Through combined use of [18F]-FLT PET and [18F]-FDG PET, we discovered in vivo that mutant BRAF CRC cells resist BRAF inhibition through RAS-mediated activation of PI3K signaling. Combined inhibition of BRAF and PI3K is highly efficacious in this setting. We propose to evaluate an innovative suite of biologically orthogonal imaging biomarkers to monitor tumor initiation, progression and response to targeted therapeutics in our Lrig1-Cre driver mouse model.

We propose three Specific Aims to advance this work.

  • Aim 1. To utilize TSPO ligand PET imaging to assess the efficacy of a novel Wnt inhibitor in a clinically relevant, stem cell-derived mouse model of colon cancer. In colonoscopy-confirmed tumors in Lrig1- CreERT2/+;Apcflox/+ mice, TSPO ligand PET will be evaluated as a metric for predicting clinical and histopathological response to VU-WS113, an allosteric activator of CK1a. We predict that Wnt pathway inhibition will result in therapeutic efficacy that can be predicted quantitatively by non-invasive imaging.
  • Aim 2. To employ multimodal imaging to assess both colon cancer progression driven by cumulative mutations and responses to targeted therapeutics. The Lrig1-CreERT2/+;Apcflox/+ mouse will be used as a platform to introduce additional genetic events that commonly occur in human CRC (mutant BRAF, KRAS and loss of p53). Imaging metrics that report tumor cell fueling (glucose/glutamine uptake), steroidogenesis (TSPO expression), and DNA replication (thymidine kinase 1 (TK1) expression) will be compared over time and the results correlated to genetic events, tumor grade and stage, and responses to targeted therapeutics.
  • Aim 3. To utilize molecular imaging to predict therapeutic response in a clinical trial that combines BRAF and PI3K inhibitors in patients with mutant BRAF CRC. Patients will be identified by SNaPshot genotyping provided through Vanderbilt’s Personalized Cancer Medicine Initiative (PCMI). Serial [18F]-FLT PET will be evaluated prior to the initiation of therapy and day 15 of therapy, and these results will be compared to standard RECIST criteria.


Project Leaders:
Jordan Berlin, M.D. (clinical)
Stephen Fesik, Ph.D. (basic)

Effective therapies for CRC still remain a critical unmet medical need. One of the most important drivers of CRC is K-Ras, which is mutated in 40% of human CRCs. Indeed, Ras proteins play essential roles as molecular switches, controlling cell proliferation, growth, differentiation, and apoptosis. Deregulation of the Ras signaling pathway by activating mutations, overexpression, or upstream activation is common in many human tumors. Thus, K-Ras represents a highly validated and important target for CRC and a wide variety of other cancers. Although a number of different approaches have been attempted to target upstream or downstream proteins in the K-Ras signaling pathway, it would be ideal to target K-Ras itself. However, K- Ras is considered to be a poor drug target. In order to test whether K-Ras could be druggable with a small molecule, we cloned, expressed, isotopically labeled, and purified K-Ras (G12D) and conducted a fragment- based screen on GDP- and GTP-bound K-Ras using NMR. In these screens we identified over 100 small molecules that bind to K-Ras, suggesting that K-Ras may be a druggable target. In addition, we obtained several X-ray crystal structures of K-Ras bound to the hits identified in the screen, which we are currently using to guide the synthesis of K-Ras inhibitors. We propose to discover small molecules that potently bind to K-Ras, inhibit its functions, and are highly efficacious against in vivo CRC tumor models with the goal of discovering a compound that is suitable for entry into a CRC clinical trial. We hypothesize that a K-Ras inhibitor will be highly effective for treating CRC patients.

  • Aim 1. Generate lead compounds that bind K-RAS tightly (nM) from our fragment-based screens and recently determined three-dimensional structures of K-RAS/inhibitor complexes using iterative structure-based design.
  • Aim 2. Optimize lead K-RAS binders for their K-RAS inhibitory effects in biochemical assays and their cell-based activities against colon cancer cells.
  • Aim 3. Prioritize K-RAS inhibitors with excellent pharmaceutical properties that are efficacious in vivo using a new stem cell-derived, tamoxifen-inducible Cre driver (Lrig1-CreERT2) mouse to activate mutant KRAS in the mouse colon. Select a compound that is suitable for a clinical trial in colon cancer by the end of the granting period.


Project Leaders:
R. Daniel Beauchamp, M.D. (clinical)
Daniel C. Liebler, Ph.D. (basic)

The identification of colorectal cancer (CRC) patients that will benefit from adjuvant chemotherapy after surgical resection poses a major unmet need in providing their safest and most effective care. The current practice results in under-treatment of high-risk stage II patients and overtreatment of low-risk stage Ill patients. The core obstacle is the lack of a definitive diagnostic biomarker(s) to identify cancers with a high probability of metastasis and corresponding poor clinical outcome. Translation of microarray-based profiles into clinical diagnostics is complicated by their complexity, as well as by logistical, cost and regulatory barriers. Pathological assessment of solid tumors typically involves immunohistochemistry and other immunoassays to detect protein expression in formalin-fixed, paraffin-embedded (FFPE) tissue sections. Thus, there is a gap between an emerging body of genomic information and diagnostic application. We propose to fill this gap by combining emerging genomic and proteomic technologies to identify and validate new molecular markers of colorectal cancer recurrence using FFPE tissue samples. We hypothesize that molecular encoding of a recurrence-prone phenotype in CRC is reflected by both transcriptomic and proteomic features. We will combine high-dimensional network analysis and new, targeted analysis platforms for specific transcripts and proteins to develop and test new biomarkers in archival FFPE specimens. We will test this hypothesis and develop these approaches according to the following specific aims:

  • Aim 1: Develop our 34-gene nucleic acid-based colon cancer prognostic classifier for use in FFPE tissue samples and refine through a competitive evaluation of selected and published signature elements, using the novel nCounter multiplex expression analysis approach. We will use high-dimensional network models to predict protein biomarker candidates, which will be systematically evaluated with targeted proteomic analyses.
  • Aim 2: Identify candidate protein biomarkers by targeted proteomics analysis.
  • Aim 3: Test the protein-based and nucleic acid-based signature biomarkers in an independent set of archived colon cancer tissue samples annotated with patient outcomes.


Project Leaders:
Harvey J. Murff, M.D., M.P.H. (clinical)
Wei Zheng, M.D., Ph.D., M.P.H. (basic)

Most colorectal cancers arise from adenomatous polyps, and a large proportion of adenoma patients develop new (metachronous) adenomas after their initial polypectomy. There is considerable controversy regarding an appropriate surveillance interval for adenoma patients after removal of their initial adenomas. We propose to conduct a follow-up study to evaluate both genetic susceptibility risk variants and tumor markers in relation to the risk of metachronous adenomas. The proposed study will be conducted in approximately 1,500 patients diagnosed with either multiple adenomas or a pathologically advanced adenoma. These patients have already been recruited in our previous studies. In addition to clinical and epidemiologic data, we have already obtained germline DNA samples, fresh-frozen polyp tissues, and formalin-fixed, paraffin-embedded (FFPE) from a large proportion of study participants. In this study, we propose to:

  • follow up with study participants to collect information related to follow-up exams and adenoma recurrence and to obtain FFPE blocks of initial adenomas from the remaining patients whose samples have not yet been collected
  • evaluate the association of genetic and epigenetic tumor markers with recurrent adenomas
  • evaluate the association of adenoma recurrence with GWAS-identified genetic variants
  • establish a risk-assessment model and evaluate the utility of genetic susceptibility and tumor markers alone and in combination with known predictors (such as pathologic features of initial adenomas) in predicting the risk of adenoma recurrence.

This proposed study will provide critical information that is valuable to identify high-risk adenoma patients for intensive follow-up programs and chemoprevention.


Core Director:
Robert J. Coffey Jr., M.D.

Core Co-Director(s):
Jordan Berlin, M.D.

The GI SPORE Administrative Core supports projects and investigators by managing SPORE resources, communication and outreach, and by fostering the interaction among investigators, collaborators, other VICC SPOREs, other GI SPOREs, the patient and advocate community and the NCI. This management and support is accomplished by administrative and scientific meetings of SPORE investigators with oversight provided by the Internal and External Advisory Boards. Specific functions of the Core include:

  1. To coordinate all GI SPORE cancer-related research
  2. To administer the Developmental Research Program (DRP)
  3. To administer the Career Development Program (CDP)
  4. To create and prepare documents and reports to ensure compliance with federal regulations and reporting requirements
  5. To monitor and manage financial resources
  6. To serve as the point of contact for GI SPORE investigators and patient advocates
  7. To organize and schedule GI SPORE project meetings for oversight, communication, administration, evaluation, collaboration, outreach and education
  8. To coordinate communications with other GI SPOREs, VU SPOREs, VICC, VUMC, Epithelial Biology Center (EBC), Ayers Institute, Digestive Diseases Research Center (DDRDC) and collaborators
  9. To organize annual External and Internal Advisory Board (EAB & lAB) meetings
  10. To organize, along with the other Vanderbilt Lung and Breast SPORE administrative cores, the quarterly VICC SPORE PI meetings
  11. To proactively identify academic (Meharry Medical College, Melbourne University) and pharmaceutical (GE Healthcare, Novartis) interactions that will enhance the conduct of the GI SPORE program, taking special advantage of opportunities provided through Vanderbilt's Clinical & Translational & Science Award (CTSA)


Core Director:
M. Kay Washington, M.D., Ph.D.

Core Co-Director:
H. Charles Manning, Ph.D.

The Translational Pathology and Imaging (TPI) Core provides high quality biospecimens, research histology services, and imaging services to support the translational efforts of the SPORE. The Core banks and distribute human colorectal neoplasms and matched normal tissue samples to investigators in the Vanderbilt GI SPORE and other GI SPORES and banks portions of polyps and biopsies of grossly normal colorectal mucosa to facilitate research in adenoma recurrence (Project 4). Detailed biospecimen annotation, including histopathologic features, preanalytical variables, tissue quality metrics, and clinical outcome is recorded in relational databases supported and maintained by the Biostatistics and Bioinformatics Core. The Core follows a rigorous quality assurance program to ensure biospecimen and data quality. Other services include research immunohistochemistry, interpretation of mouse and human histopathologic findings, custom tissue microarray design, and consultative services for laser capture microscopy. The Core protects patient confidentiality through use of an explicit consent form that specifically addresses use of extraneous tissue for research purposes and through de-identification of specimens. The GI SPORE Tissue Core at VUMC has partnered with other mechanisms for tissue collection at VUMC under Dr. Washington's direction, the Tissue Morphology SubCore of the Digestive Disease Research Center, and the VUMC-Ied Western Division of the Cooperative Human Tissue Network to provide these services in a cost-effective manner.

This Core will provide pathology support to all four SPORE projects and will collaborate with the Biostatistics and Bioinformatics Core.


Core Director:
Yu Shyr, Ph.D.

The purpose of the Biostatistics & Bioinformatics Core is to provide professional expertise in biostatistics and bioinformatics for all GI Cancer SPORE projects, investigators, and participants. Functions provided by this core include development of experimental designs, power analysis, and sample size estimation; data quality control; statistical/bioinformatics analysis and interpretation of findings; and collaboration on presentation of results. To achieve these functions, the core director and core members are constantly available to investigators, and are in regular contact with the project and core leaders.

  1. The primary objectives of the Biostatistics and Bioinformatics Core are:
  2. To provide study design and review all laboratory, animal, and clinical studies including feasibility assessment, power analysis, and sample size estimation
  3. To collaborate in project data analysis, interpretation of results, and the writing of final study reports and manuscripts
  4. To work with the Translational Pathology and Imaging Core in the development of research project databases, to maintain data quality control and to ensure timely data capture
  5. To develop and evaluate statistical/bioinformatics methods for experimental design and data analysis

Biostatistics and Bioinformatics Core support is required in all GI Cancer SPORE studies. Core personnel have worked and will continue to work closely with project leaders to ensure the core provides state-of-the-art statistical/bioinformatics support.