News Release

Pancreatic cancer genes ID'd in most accurate study to date

Research may improve diagnosis and treatment for #4 cancer killer

Peer-Reviewed Publication

Michigan Medicine - University of Michigan

ANN ARBOR, MI – Being diagnosed with pancreatic cancer is like receiving a death sentence -- one that, for many patients, is carried out within weeks or months of their cancer's discovery. Less than 20 percent of patients are diagnosed in time to qualify for the only known cure – an arduous operation – and only 3 percent of all patients live even five years.

But a team of researchers from the University of Michigan Comprehensive Cancer Center hopes to overturn that death sentence, through studies that zero in on the genes and proteins that help pancreatic cancer grow and spread.

In a new paper in the journal Cancer Research, and through work funded by a $2.36 million grant received last week, as well as a new patient care program, the team aims to find new ways to improve diagnosis and treatment of the deadly disease.

In the paper, published May 15, the team reports finding 158 genes specific to pancreatic cancer – the most accurate list to date. They used tissue samples that had been swiftly processed after being removed from patients, to preserve fragile molecules that act as telltale signatures for genes that are "turned on" or expressed.

Unlike other scientists, the U-M team was able to distinguish genes involved in cancer from those involved in a chronic inflammatory disease, pancreatitis, that's often mistaken for cancer. Both diseases produce similar scar tissue around the pear-shaped pancreas gland, which produces insulin, hormones and digestive juices. The team narrowed the list of genes down to 80 that were expressed three times more often in pancreatic cancer cells than in non-cancerous or pancreatitis cells.

As a result, the U-M team believes its results will be more applicable to making specific diagnostic tests and effective treatments. The new paper even reports their success in identifying and detecting four of the proteins – 14-3-3-sigma, S100P, S100A6, Beta-4 integrin – that are encoded by the cancer-specific genes.

The new grant, from the Michigan Life Sciences Corridor, will let the team continue cataloging proteins associated with the genes, and determining which of them might be "biomarkers" that could be used in diagnostic tests or treatment planning.

"Pancreatic cancer is one of the swiftest and surest cancer killers, and not nearly enough has been learned about what, at the molecular level, makes it so deadly," says Craig Logsdon, Ph.D., lead author of the new paper and principal investigator on the new grant. "Others have made lists of genes that might be involved, but ours provides a much more accurate accounting, and reveals dozens of novel genes."

Logsdon, a professor of physiology at the U-M Medical School, has special expertise in studying the pancreas. He co-leads the team with Diane Simeone, M.D., who removes pancreatic tumors from patients as an associate professor in the U-M Department of Surgery's Gastrointestinal Surgery division, and directs studies of the cancer in her basic research laboratory.

The newly published paper is based on tissue from ten pancreatic tumors, five samples from pancreatitis patients and five samples of normal pancreas, as well as seven commercially available pancreatic cancer cell lines. The researchers screened all the cell types for expression of more than 6,800 genes, and focused in on those expressed more often in cancer cells.

They were able to overcome limitations of past studies by other teams, which did not distinguish between genes expressed by cells in the scar tissue (called stroma), which could be associated with either cancer or pancreatitis, and those expressed only by cancer cells.

Late diagnosis, a lack of effective treatment options, rapid metastasis, and a dearth of research on how pancreatic cancer works combine to make the nation's 10th most common cancer into its 4th most deadly cancer. More than 29,000 Americans are diagnosed with it each year, most of them already too far advanced to be treated.

"We truly need a better understanding of pancreatic cancer, to help improve the terrible odds that these patients face because of late diagnosis and treatment resistance, and to spare pancreatitis patients unnecessary surgery because of misdiagnosis," says Simeone.

"Protein biomarkers, which we hope will be easier to find now that we have our gene list, may help us find ways to make diagnosis more accurate, and to tailor treatment to individual patients."

More than 200 pancreatic cancer patients a year receive treatment at the U-M Cancer Center, taking part in clinical trials that combine surgery, radiation therapy and chemotherapy in an effort to improve the cancer's response to treatment.

Simeone and her colleagues have established a new comprehensive pancreatic cancer clinic that will give patients access to a broad range of clinical and supportive care, and help speed the clinical trials process.

She notes that statistics show patients do best at centers that treat many cases each year, such as U-M, and that even patients who face no chance of survival often want to help research by enrolling in clinical trials or donating tissue.

U-M pancreatic cancer patients are already being added to a database that will track their treatment experience, and allow the research team to study their tumor tissue, including metastases, and their blood and saliva for further clues to the cancer's molecular workings.

As was done in the newly published study, tissue samples will be hand-carried straight from the operating room to U-M pathology laboratories, to preserve the messenger RNA that carries instructions for protein construction from the cell nucleus after a gene is expressed. Then, sophisticated genetic microarrays allow the researchers to determine which genes are "turned on" in different kinds of cells, and compare gene expression via statistical analysis.

Chronic pancreatitis patients treated in gastroenterology clinics of the U-M Department of Internal Medicine will also be asked to donate tissue removed during fine-needle aspiration biopsies, so further studies can be done to distinguish cancer from inflammatory disease.

In order to expand the study, the U-M team has enlisted the cooperation of the Karmanos Cancer Center in Detroit, which will collect samples of tissue from its patients, and of Creighton University in Nebraska, which studies families with the rare inherited form of pancreatic cancer.

Now, using the gene list published in Cancer Research as a roadmap, and the Life Sciences Corridor Funding, the U-M researchers plan to catalog the proteins encoded by the genes, and create antibodies to them that will allow their detection. This will give them as systematic way to determine if they can detect certain proteins in the blood or saliva of cancer patients, and to test treatments in animal models and, eventually, human patients.

To identify cancer related proteins these researchers will utilize an approach known as proteomics, in collaboration with Samir Hanash, M.D., Ph.D., a U-M professor of pediatrics and oncology specialist who was senior author on the new paper and who is an internationally recognized expert in this area. Hanash is president and chair of the Human Proteome Organization, an international effort to catalog the human proteome in much the same way as the human genome was recently completed.

Logsdon and Simeone says their first targets will be genes that are expressed early in the development of cancer, and those that encode proteins found on the surfaces of cancer cells. This approach will speed the development of potential diagnostic tests, and of immunotherapies that target cancer cells while sparing normal tissue.

"We're focusing the U-M's resources on pancreatic cancer, combining clinical and basic science in a way that we feel will help accelerate the progress on this disease," says Logsdon. "This gene list is a very promising start, but there is still far to go."

In addition to the new Life Sciences Corridor grant, the team has support from the Lustgarten Foundation for Pancreatic Cancer Research, the U-M Comprehensive Cancer Center, and the National Cancer Institute. The U-M has applied for a patent on the entire gene list published in Cancer Research.

Besides Logsdon, Simeone and Hanash, the new paper's authors include surgery resident Charles Binkley, M.D.; physiology research fellow Thiruvengadam Arumugam, Ph.D.; associate pathology professor Joel Greenson, M.D.; associate clinical pathology professor Thomas Giordano, M.D., Ph.D.; and senior pediatrics research associates David Misek, Ph.D. and Rork Kuick, Ph.D.

###

Reference: Cancer Research 63, 2649-2657, May 15, 2003

Pancreatic cancer patients and their families who wish to learn more about pancreatic cancer treatment and research at the U-M Comprehensive Cancer Center, including clinical trials, may call the U-M Cancer AnswerLine toll-free at 800-865-1125, or visit www.cancer.med.umich.edu/clinic/pancreaticclinic.htm .

Note: Images relating to the study, including a multicolor representation of gene expression in the various cell types, and tissue staining images from the four initial proteins, are available upon request.


Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.