Program Co-Leaders Margaret Tempero, MD Martin McMahon, PhD

The Pancreas Cancer Program is comprised of a multi-disciplinary group of investigators focused on pancreatic cancer. These investigators include a team of clinicians and clinical scientists in endocrinology, epidemiology, gastroenterology, medical oncology, radiation oncology and surgical oncology who are qualified to address critical issues in pancreatic cancer, with full integration of faculty in basic and population sciences. The investigators in the program are not only interested in identifying better treatments but also in identifying persons at high risk for the disease and methods for screening and prevention.

Generous gifts from grateful patients and family have helped to expand the program and attract young doctors and scientists. Read about how you can help or donate to our program.

Choose below for our newsletter, Pancreas Cancer News
May 2005 (Volume 1, Number 1)
April 2007 (Volume 2, Number 1)

Major Scientific Goals

• Understand the genetic and environmental factors that influence risk for pancreatic adenocarcinoma and pancreatic neuroendocrine tumor development.
• Understand the genetic and biologic events occurring early in pancreas cancer, with a view to developing better diagnostic tools.
• Develop new paradigms and model systems to evaluate the role of specific genes in the initiation and progression of these cancers.
• Understand the factors that promote invasion of pancreas cancer, with a special emphasis on the supportive structure and environment of pancreatic adenocarcinoma cells and the biology of invasion and metastasis.
• Understand the basis of drug resistance and sensitivity in pancreatic adenocarcinoma and pancreatic neuroendocrine tumors.
• Use such model systems to evaluate new targeted therapeutics, either alone or in combination with conventional drugs, in the treatment of pancreatic adenocarcinoma and pancreatic neuroendocrine tumors.

Basic Science

The Pancreas Cancer program has an exceptionally strong and collaborative basic science component focusing on both pancreatic adenocarcinoma and neuroendocrine tumors.

Major strengths of the basic science component include:

• State-of-the-art mouse models of pancreatic cancers that are being used to explore basic mechanisms of tumor transformation and as platforms for pre-clinical evaluation of novel therapeutics to treat cancer.
• Analysis of signal transduction pathways that contribute to the abnormal proliferation, invasion and migration of pancreas cancer cells.
• Ability to derive primary cultures of primary human ductal epithelial cells (PDECs) for the analysis of early genetic and cell biological alterations in the cells that are believed to be the progenitors of pancreas cancer.

Clinical Research

The Pancreas Cancer Program also has a robust clinical research component and includes studies in both pancreatic adenocarcinoma as well as neuroendocrine tumors of the pancreas.

Program Research Themes

Development of Novel Therapeutics for Pancreatic Adenocarcinoma
Drs. Ko and Tempero and their colleagues at UCSF have been instrumental in improving chemotherapy treatment protocols, prolonging life and quality of life.

Dr. Tempero has focused on novel therapeutics for pancreatic cancer for most of her career. Her pioneering studies of gemcitabine have lead to better drug dosing protocols and combination therapies. Dr. Tempero is credited with the translational work for the clinical application of fixed dose rate (FDR) gemcitabine in pancreatic cancer patients, which is now commonly used in practice. This resulted from a randomized phase II trial comparing two dose intense regimens in patients with adenocarcinoma conducted by Dr. Tempero in collaboration with investigators at MD Anderson Cancer Center as well as other selected centers. Patients on a FDR gemcitabine demonstrated a 2-fold increase in intracellular activity of gemcitabine and also demonstrated a significant improvement in overall survival. This concept was transported to a large phase III trial, EGOG 6201, which compared FDR gemcitabine to standard infusion gemcitabine and to FDR gemcitabine plus oxaliplatin. Again, a modest survival advantage was seen with FDR gemcitabine compared to standard infusion gemcitabine. One problem with administering gemcitabine by FDR is a higher frequency of myelosuppression (lowered white blood cell counts). To address this, Drs. Ko and Tempero altered the schedule of FDR gemcitabine to alternate week dosing and added low dose cisplatin. By doing so, myelosuppression was reduced while a similar degree of clinical activity was retained.

Strategies such as fixed-dose infusion of gemcitabine and addition of other agents on a more logical pharmacokinetic basis have provided a consistent advantage over standard treatment strategies. Drs. Ko and Tempero also are collaborating with Kathleen Giacomini, chair of biopharmaceutical sciences at UCSF and an expert in pharmacogenetics, to study how individual genetic variations affect drug metabolism, effectiveness and side effects.

Because pancreatic cancers secrete vascular endothelial growth factor (VEGF) and express VEGF-R and because high VEGF levels are correlated with poor prognosis with this disease, Drs. Ko and Tempero subsequently incorporated bevacizumab, an anti-VEGF antibody, into their regimen of FDR gemcitabine/low-dose cisplatin. This has proved to be a highly active regimen and in their early report presented at ASCO 2006, the median survival in metastatic disease was shown to be 8 months with a one-year survivorship of 37%. The disease control rate (objective response plus stable disease) is about 70%. This may be the highest one-year survival rate that has ever been reported for a metastatic cohort. As this data matures, it may be important to explore this further in a phase III trial.

In addition to the above, Drs. Tempero and Ko have trials in progress evaluating erlotinib (Tarceva) and bevacizumab (Avastin) for patients with gemcitabine-refractory disease; as well as a randomized trial of cetuximab (Erbitux) and bevacizumab (Avastin) ) alone or in combination with fixed-dose-rate gemcitabine (Gemzar). Given the promising results of their phase II trial of FDR gemcitabine, cisplatin, and bevacizumab, the group has just opened a trial of fixed dose rate gemcitabine and bevacizumab as adjuvant therapy following resection.

Most recently Drs. Ko and Tempero have begun building a novel program to find out how genetic characteristics in pancreas tumors are linked to responsiveness to different treatment regimens. Dr. Ko is leading a multi-center study, in collaboration with colleagues at Johns Hopkins, whereby they are moving toward a more patient-tailored approach to pancreatic cancer treatment. The investigators are collecting tissue and genomic DNA from each patient prior to initiation of therapy. The study design incorporates several key elements: 1) a sequential 'folding in' of therapeutic agents to isolate the effects of each individual agent; 2) early therapeutic decision-making based on CA19-9 biomarker response; and 3) identifying potentially important gene-drug interactions. Taken in its entirety, this study should pave the way toward identifying both host and tumor characteristics that mediate responsiveness to gemcitabine and also to a panoply of other agents both studied and not yet studied in pancreatic cancer. This project will serve as the basis for future 'directed' therapy for patients with pancreatic cancer, in which selection of agents can be made in rational fashion.

Development of Biomarkers
Because of the treatment time required to achieve radiographic changes and difficulty measuring benefit from treatment in this disease with radiographic studies, our researchers have been interested in a more rapid assessment tool that can be used as a surrogate for survival. By retrospectively analyzing data from three clinical trials, they were able to show that progressive declines in CA19-9, a biomarker associated with pancreatic cancer, correlates with increasing time to progression as well as overall survival (Ko et al., British Journal of Cancer 93, 195-9, 2005). These studies have allowed them to incorporate early decision making based on CA19-9 response in a novel clinical trial design evaluating biomarkers for drug sensitivity.

Development of Novel Therapeutics for Neuroendocrine Tumors (NET)
Over the past several years, Drs. Emily Bergsland and Alan Venook have worked to develop a clinical program focused on patients with gastrointestinal neuroendocrine tumors (the second-most common tumor arising in the pancreas). Along with the Dana Farber Cancer Institute, they performed a phase II trial of recombinant human endostatin (an endogenous angiogenesis inhibitor) in patients with advanced NET (Kulke et al, Journal of Clinical Oncology 24, 3555-61, 2006). In addition, they explored the utility of SU11248 (a multi-targeted, oral receptor tyrosine kinase inhibitor) in the same patient population. More recently, Dr. Bergsland designed and implemented an investigator-initiated trial exploring oxaliplatin-based chemotherapy (mFOLFOX-6) plus bevacizumab in patients with NET.

[ selected research publications ]

Development of New Model Systems to Explore the Earliest Stages of Pancreas Cancer Development
To tackle the problem of pancreas cancer, we need more information about its cell of origin. We also need new and more accurate model systems for preclinical testing of promising new diagnostic and therapeutic tools. We are exploring the cell, developmental, and molecular biology of normal human and mouse pancreas cells. These studies are permitting an insight into the progenitor of pancreas cancer. In addition, using microarray technology, normal cells are being compared to pancreas cancer-derived cells to construct a catalogue of changes that take place in the transition of normal cells to a cancerous phenotype. Finally, we are deriving accurate mouse models that faithfully recapitulate the genetics and cell biology of the human disease in order to better understand its pathophysiology and to serve as a platform for the generation of new diagnostic and therapeutic agents.

Understanding Risk Factors and Identifying Diagnostic Markers Associated with Pancreas Cancer
Dr. Elizabeth Holly is heading data analyses and laboratory studies in a large population-based case-control study of pancreatic cancer. This large epidemiologic study, currently undergoing data analysis through an NCI grant, includes data and blood specimens from 532 patients diagnosed with adenocarcinoma of the pancreas and more than 1,700 control participants. Additional studies are being developed with Dr. Allan Balmain to translate molecular/genetic results from his mouse-model research to a human model using genetic results and biological specimens from Dr. Holly's study of pancreatic cancer. Under the aegis of a NCI-funded grant, data analyses have been conducted and recently published for anthropometric factors, exposures to endogenous and exogenous hormones, dietary factors and survival. Dr. Holly also has shown that a history of allergies is protective for pancreatic cancer. Results from analyses of women's exposures to hormones indicated that these exposures do not play a significant role in the development of pancreatic cancer in the Bay Area population that was included in this study. Dietary analyses of fruits and vegetables showed that participants with the highest total vegetable, total fruits and fruit juice consumption were about 50% less likely to have been diagnosed with pancreatic cancer compared with those with the least consumption of these foods. Other manuscripts have been submitted that describe the results of analyses on pancreatic cancer and history of pancreatitis. In the Bay Area study population, pancreatitis was more likely to be a manifestation of pancreatic cancer than a risk factor for its development. Similarly, in a recently published paper, increased risks for pancreatic cancer were observed only among those whose diabetes mellitus was diagnosed within 10 years of cancer diagnosis.

Dr. Holly recently was awarded a large grant by the NCI to conduct a clinic-based study of the molecular epidemiology of pancreatic cancer. Recruiting patients at the time of their initial clinic visit diminishes the high loss of pancreatic cancer patients due to death. Working with Dr. Ko, study personnel will recruit 600 patients and 600 control participants from UCSF clinics.

[ selected publications from Dr. Holly's group ]

Development of a Pancreas Tissue Bank
To date only two chemotherapeutic drugs, gemcitabine (Gemzar) and erlotinib (Tarceva), have been approved for patients with pancreas cancer. Although some patients display a good response to these drugs, others do not. The mechanism underlying this difference is not known. By exploring subtle genetic differences between tumors that respond to drugs or not, we hope to understand the genetic and biochemical basis of drug responsiveness. Our researchers have developed a tissue bank of pancreatic adenocarcinomas and neuroendocrine tumors, which are collected through the Cancer Center Tissue Bank facility. This rich repository of tissue enables our investigators the ability to begin these and other genetic tests. Information from these studies will permit us to design better therapeutic strategies to treat patients with drugs that are tailored to the specific genetic and biochemical features of their disease. Moreover, these efforts are proceeding in conjunction with efforts to identify new markers, which may permit earlier diagnosis in at-risk populations and thus allow earlier and more effective intervention.

Development of Informatics in Pancreatic Cancer
With the assistance of the Cancer Center's Informatics Facility, we have begun implementation of a new and powerful relational database that will house all clinical outcome data and any available molecular data from tissue specimens and analyses. Implementation of this database will serve as an important resource to investigators to identify new targets for therapy and ultimately help guide rational therapeutic decision-making in the future.