Pilot study to evaluate validity of pancreatic cancer biomarkers

Despite improvements in overall survival for most cancers, survival for patients with pancreatic adenocarcinoma has remained dismally low for decades. Relative one- and five-year survival rates are 20 percent and less than 5 percent, respectively.

The high mortality rate is due to late diagnosis, early metastasis and poor response to chemo- and radiotherapy. Patients with resectable tumors have an average survival of just 18 to 24 months, even when novel neoadjuvant therapies such as FOLFIRINOX are used.

"Pancreatic cancer is one of the most untreatable, if not the most untreatable, cancer; it's generally a death sentence when diagnosed," says Michael B. Wallace, M.D., a gastroenterologist at Mayo Clinic's campus in Jacksonville, Florida. "The one area where we can really improve the cure rate is early detection, so great need exists for biomarkers for pancreatic cancer."

But existing markers such as CA 19-9 are inadequate because of low sensitivity and specificity, and identification of other molecular markers in serum, bile, stool and urine have proved challenging and, for the most part, disappointing. "The pancreas is very enigmatic and difficult to work with in every respect," says Massimo Raimondo, M.D., whose Mayo research in Jacksonville focuses on pancreatitis and pancreatic cancer.

Recently, Dr. Raimondo investigated a method for detecting pancreatic adenocarcinoma — and potentially discriminating it from pancreatitis — using aberrantly methylated DNA markers in pancreatic juice. This approach has several advantages. Pancreatic fluid consists of biomolecules secreted by pancreatic exocrine cells as well as sloughed ductal cells, allowing investigation of disease within the entire ductal system of the pancreas. And pancreatic juice emptied into the duodenum can be collected during routine upper endoscopy after intravenous administration of Secretin — a safe and relatively noninvasive method of collection.

In collaboration with colleagues from Mayo Clinic's Rochester, Minnesota, campus who were involved in development of a next-generation stool DNA test for screen-detection of colorectal cancer, molecular assays were performed on archived pancreatic juice Dr. Raimondo had collected from nearly 170 patients. Included were samples from patients with known pancreatic cancer, chronic pancreatitis (which can mimic pancreatic cancer) and pancreatic cystic neoplasms as well as healthy controls.

"Molecular markers are highly expressed in pancreatic cancer," Dr. Raimondo says. "From thousands of potential markers, my Rochester colleagues John B. Kisiel, M.D., and David A. Ahlquist, M.D., have identified 30 or 40 of the most promising. The markers were initially tested in archived tissue from patients with pancreatic cancer, and the results are now used to guide analysis of pancreatic fluid. It was a very Mayo effort, where people with similar ideas and interests collaborate and come up with something very encouraging."

Results of the study appeared in a 2013 Gastroenterology abstract.

The collaborative effort will extend further as the study's next phase expands to all three Mayo sites. The aim is to enroll 100 patients at each site for the purpose of assessing the validity of Dr. Raimondo's data and the reproducibility of the pancreas juice collection methods by other gastroenterologists in the Mayo system.

Once the markers are identified, the goal is to test the most discriminating in patients who have early-stage pancreatic cancer or are at high risk of the disease compared with the general population. Those risk factors include:

  • Family history of pancreatic cancer
  • Chronic pancreatitis
  • A premalignant cystic lesion of the pancreas

"We hope to determine which markers, alone or in combination, are the most promising for a future serum-, stool- or urine-based test that will be easier and less expensive to screen a large population," Dr. Raimondo says. "We'll run the study for a year. A one-year pilot will be very good if we can confirm the promising initial results."

For more information

Raimondo, M et al. Methylated DNA markers in pancreatic juice discriminate pancreatic cancer from chronic pancreatitis and normal controls. Gastroenterology. 2013;144:S-90.

June 04, 2014