Research

Project Leader: Michael Overman, M.D. (Clinical)
Co-leader: Gregory Lizee, Ph.D. (Basic)

Immunotherapy with antibodies targeting checkpoint blockade molecules PD-1 and CTLA-4 has not demonstrated clinical activity in the vast majority of colorectal cancers. By contrast, a rare subset of colorectal cancer tumors that show microsatellite instability (MSI-high) demonstrate marked responses to PD-1 therapy, similar to or exceeding that observed in other highly mutated cancers such as melanoma and lung adenocarcinoma. This correlation between mutational load and clinical response suggests that the anti-tumor immunity generated by checkpoint inhibitors is mediated largely through the activation of T lymphocytes recognizing mutated peptides presented by HLA class I molecules. While identification of mutated peptide epitope targets in individual patients remains a daunting technical challenge, recent advances in next generation sequencing (NGS) have provided a strong foundation on which to build these efforts. Revealing mutated target antigens in individual cancer patients would facilitate a "precision immunotherapy" strategy in which an immune response against multiple expressed tumor antigens could be generated in patients using multivalent peptide vaccines. The specific objective of this proposal is to generate an effective, personalized vaccine approach for treatment of metastatic colorectal cancer patients at high risk of recurrence following surgery. Specifically, we will test the hypothesis that a vaccination strategy targeting multiple mutated colorectal cancer tumor antigens along with specific combinations of immune adjuvants will be capable of preventing recurrence in minimal residual disease (MRD) setting in post-hepatectomy colorectal cancer patients. As the success of vaccination is likely predicated upon lower volume disease, we propose to investigate our personalized peptide vaccination approach in colorectal cancer patients in the adjuvant setting who have detectable mutations in circulating tumor-derived DNA (ctDNA) that can be monitored over time. Our Preliminary Data shows that the proposed personalized vaccination strategy is feasible, as several colorectal cancer patients have now undergone vaccination in ongoing clinical trial at MD Anderson. In addition, our pre-clinical mouse data has shown that Toll-like receptor ligands, anti-CD40, and checkpoint blockade can be highly effective combinations of adjuvants in vivo. However, it is critical to understand the optimal combination of agents to use for vaccination in order to translate these findings into more effective vaccine strategies for our colorectal cancer patients.

Project Leader: David Tweardy, M.D. (Basic)
Co-leader: Eduardo Vilar-Sanchez, M.D., Ph.D. (Clinical)

Evidence is increasing that signal transducer and activator of transcription (STAT) 3 contributes to sporadic and high-risk colorectal cancer carcinogenesis arising in a background of inflammation such as inflammatory bowel disease (IBD), and also hereditary populations such as familial adenomatous polyposis (FAP) and Lynch syndrome (LS). IBD presents as either ulcerative colitis (UC) or Crohn's disease (CD), constitutes attractive models to study the contribution of inflammation to colorectal carcinogenesis. STAT3α is proinflammatory and anti-apoptotic, while STAT3β antagonizes these effects of STAT3α. IBD in mice induced by either dextran sodium salt (DSS;UC) or trinitrobenzoic acid (TNBS;CD) was more severe in transgenic mice expressing only STAT3α compared to wild type mice. We developed a potent small-molecule STAT3 inhibitor, C188-9, that targets the phosphotyrosyl peptide-binding pocket within the STAT3 SH2 domain which blocks STAT3 activation [phosphorylation on tyrosine (Y) 705, pY-STAT3]. C188-9 administration prevented IBD caused by both DSS and TNBS. In studies by others, mice deficient in STAT3 in their intestinal epithelial cells demonstrated reduced tumor size and reduced tumor incidence in a model of colitis-associated colorectal cancer [azoxymethane (AM) plus DSS]. FAP is caused by germline mutations in APC. Genetically reducing levels of STAT3 in Apc GEMM decreased the number of intestinal polyps. STAT3 activation results in extra-nuclear sequestration of MSH3, which may further impair DNA mismatch repair (dMMR) in LS enterocytes bearing a mutation in one of the other dMMR enzymes resulting in increased risk of colorectal cancer. Our preliminary data using a CLIA-certified pY-STAT3 IHC stain and scoring system supports the activation of STAT3 signaling in normal-appearing mucosa that is further increased in colorectal cancer samples from IBD, FAP and LS patients. The long-term goal of Project 2 is to determine if C188-9 will be of benefit in the prevention of colorectal cancer. The central hypotheses are that STAT3 contributes to colorectal cancer development in patients at risk for colorectal cancer and can be targeted successfully with C188-9. The objectives are to determine the effects of targeting STAT3 with C188-9 for prevention of colorectal cancer in mouse models and the contribution of STAT3 signaling to colorectal cancer development in IBD, FAP and LS patients.

Project Leader: Shubham Pant, M.D. (Clinical)
Co-leader: Andrea Viale, M.D. (Basic)

Pancreatic ductal adenocarcinoma contributes to 6.9% of all cancer deaths in the US, and >1.5% of the U.S. population will be diagnosed with pancreatic ductal adenocarcinoma in their lifetime. Despite a better understanding of the genomic landscape and the importance of the tumor microenvironment, there has been no meaningful shift in the overall survival for this disease. Thus, new therapeutic strategies for this deadly disease represent an urgent unmet clinical need. Over 90% of pancreatic ductal adenocarcinoma harbor activating KRAS mutations, which are an early event in disease pathogenesis as they are present in pancreatic intraepithelial neoplasias (PanINs), the precursor lesion for pancreatic ductal adenocarcinoma. Increasingly, inter- and intra-tumoral heterogeneity in pancreatic ductal adenocarcinoma are appreciated, including KRAS dependency. Differential dependence on KRAS has been linked with altered metabolic dependencies. In one specific context, our group used an inducible oncogenic Kras-driven pancreatic ductal adenocarcinoma model to demonstrate that, upon extinction of oncogenic Kras, the persistent Kras-independent tumor-initiating cells (TICs) exhibited a metabolic profile very different from that of the Kras-dependent cancer cells of the bulk tumor. Whereas the Kras-dependent cancer cells of the bulk tumor exhibited high levels of glycolysis and metabolic dependencies, the Kras-independent TICs showed impaired glycolysis and increased mitochondrial respiration. Similar observations were made in TICs derived from human pancreatic ductal adenocarcinoma PDX models, which exhibited decreased glucose flux through glycolysis and elevated OXPHOS activity. These TICs harbor limited metabolic plasticity, rendering them particularly sensitive to inhibition of mitochondrial activity. Thus, heterogeneity of pancreatic ductal adenocarcinoma is not only defined on the genomic and cellular levels, but also defined by distinctive metabolism programs controlled by oncogenic signaling. However, to date, the documented dependency of some tumors or tumor cell subpopulations on OXPHOS has not yet been exploited therapeutically. We will explore the biology of response to treatment with metabolic inhibitors in these contexts, using a combination of ex vivo and in vivo studies, as well as evaluating patient response via clinical correlatives (transcriptomic signatures, hyperpolarized pyruvate-magnetic resonance imaging, quantitative CT scan) in planned phase 1b and phase 2a clinical studies in patients with treatment-naïve or refractory disease (phase 1b) or patients who have responded to prior standard-of-care chemotherapy (phase 2a)

Director: Scott Kopetz, M.D., Ph.D.
Co-Director: Anirban Maitra, M.B.B.S.

The Administrative Core (Core 1) will be responsible for the successful execution and management of all SPORE activities related to financial oversight and coordination, organization of all necessary meetings, and publicity and record keeping for all projects and the two other cores. This group will also provide regulatory oversight activities for clinical trials; ensure compliance with all institutional, federal and NCI-specific regulations; and oversee the peer-review and oversight processes of the Career Enhancement Program and Developmental Research Program. The core will be led by Drs. Scott Kopetz, Anirban Maitra and David Menter.

Director: Dipen Maru, M.D.
Co-Director: Huamin Wang, M.D., Ph.D.

Core 2 will coordinate efforts related to collection, processing, storage and distribution of annotated human and murine biospecimens for all of the SPORE projects, including the Career Enhancement Program (CEP) and Developmental Research Program (DRP). The Core will be co-led by two internationally reputed gastrointestinal/pancreatic pathologists, Drs. Dipen Maru and Huamin Wang. For human biospecimens, the core will interface with various Institutional systems (ePRTCL, PDMS, CORe, Electronic Health Record [EHR], or the Biospecimen Information Management System [BIMS]). Biospecimen resources from the lower gastrointestinal tissue bank include freshly collected/snap frozen and formalin fixed paraffin embedded tumor and normal specimens from more than 2,500 resected hepatic colorectal metastases (including TMAs), freshly collected and snap frozen adenomas from 334 patients and formalin fixed paraffin embedded specimens from 870 or more patients with sporadic adenoma or familial adenomatous polyposis. Existing biospecimen resources available in the pancreatic bank include freshly collected/snap frozen tumor and normal tissue sample from Whipple resection for pancreatic ductal adenocarcinoma from 232 patients, with formalin fixed paraffin embedded specimens and additional 672 patients, including tissue microarrays from pancreatic ductal adenocarcinoma and intraductal pancreatic mucinous neoplasms. The Core will support Project 1 by coordinating prospective blood collection, cryopreservation and transport with the ITB. Specifically, Core faculty will prospectively collect, process and distribute fresh tumor and normal samples from hepatic colorectal metastases after obtaining mirror image section for histology quality control. The Core will provide formalin fixed paraffin embedded samples of normal, adenoma and carcinoma to Project 2. The Core will also provide biospecimen qualification services, including but not limited to, histopathologic characterization of human and murine tissues treated with STAT3 inhibitor, and immunohistochemistry staining and interpretation, including validation of p-STAT3 staining by automated image analysis in a CLIA-certified facility. The Core will provide freshly resected pancreatic ductal adenocarcinoma samples for patient derived xenografts and ex vivo live tissue sensitivity assay (LTSA) for Project 3. In addition, the Core will provide histopathology characterization, immunohistochemistry services and interpretation guidelines for both preclinical samples from the ongoing co-clinical trials. The Core personnel, along with the ITB, will enter detailed information related to all processes of biospecimen collection, processing, qualification, distribution and analytes extraction into BIMS. The Core activities will lead to enhancement of these functionalities of Tissue Station and design a new interface specific for GI SPORE in the Tissue Station.

Director: Ying Yuan, M.D.
Co-Director: Ryan Sun, Ph.D.

The Biostatistics and Bioinformatics Core provides comprehensive service to guide design of experiments, to optimize quantitative data analysis and to maintain statistical justification and interpretation of results. Specifically, the core will implement sound experimental design principles that are tailored to address specific scientific questions for each project. The core will also carry out data analyses using suitable statistical methods and bioinformatics algorithms and will contribute to the interpretation of results through written reports and frequent interactions with project investigators. Whenever appropriate, the core will develop new analysis tools to address new challenges in the analysis of various data, especially high-throughput genomic and proteomic data. Members of the core will participate in monthly SPORE meetings with all project investigators, ensuring that statistical and data analysis/management issues are carefully considered during all phases of each SPORE experiment. Thus, from inception to reporting and publication, basic laboratory and translational experiments will benefit from the SPORE program that will be used to augment existing MD Anderson biostatistics resources.