The study is described as small and preliminary, and investigators cautioned that the initial findings will need to be confirmed in larger trials.
"Pancreas cancer is the fourth leading cause of cancer death in the United States," said study coauthor Dr. Nita Ahuja, an associate professor of surgery in the department of oncology and urology at the Johns Hopkins University School of Medicine, in Baltimore. "There have been minimal to no improvements in the survival from this disease in the last 40 years. There are over 40,000 people diagnosed every year and about that many deaths."
"One of the main reasons for the lethal nature of this cancer is that most cancers are diagnosed too late once they have spread to other organs," Ahuja said. "Around 8 percent have spread to distant organs such as the liver or lungs, while another 10 percent have locally spread to major blood vessels. However, in the patients where cancer can be detected early and has not spread, a long-term cure is possible with surgical removal of the cancer with the surrounding lymph."
Any means of spotting the cancer early would therefore be crucial, Ahuja added. "We have mammograms to screen for breast cancer and colonoscopies for colon cancer, but we have had nothing to help us screen for pancreatic cancer," she said.
Ahuja said the new study sought to find blood "markers" for pancreatic cancer "in patients who are at increased risk for developing this cancer, such as [those with a] family history or heavy smokers."
Ahuja's team had previously identified mutations in two genes, called BNC1 and ADAMST1, that typically occurred in the presence of pancreatic cancer. Since both mutations are found in 97 percent of early stage pancreatic cancer tissues, the researchers developed tests to search for signs of the mutations in blood samples collected from 42 people already diagnosed with early stage pancreatic cancer.
Reporting in the current online edition of the journal Clinical Cancer Research, Ahuja's team said both genetic markers were found in 81 percent of the tested blood samples, but not in samples taken from patients who either did not have pancreatic cancer or had a history of pancreatitis (an inflamed pancreas).
The researchers said the results are much more impressive than, for example, the prostate-specific antigen (PSA) test used to screen for prostate cancer, which has roughly a 20 percent success rate.
Still, an 81 percent accuracy rate is "far from perfect," Ahuja said. The test also had a false-positive rate of 15 percent, meaning that 15 percent of people who get the test initially will be told they might have pancreatic cancer when that is not the case.
And Ahuja stressed that the test is not designed as a screen for the population as a whole -- only for those already deemed to be at high risk for the disease.
"The eventual goal is to develop a cost-effective test to test patients who are at high risk," she said. "The beauty of this test is that it can be repeated every year as you go for your annual physical."
Dr. Smitha Krishnamurthi is an associate professor of medicine in the division of hematology and oncology with University Hospitals Case Medical Center & Case Western Reserve University School of Medicine, in Cleveland. She applauded the research, saying that "if pancreatic cancer could be detected at an early stage, more patients would be cured."
"This study presents an encouraging step in the right direction," Krishnamurthi said. "The authors have developed a blood test that detected the earliest stage of pancreatic cancer and correctly identified most of the healthy individuals tested. However, this was a very small study. The blood test must be studied in many more patients with early stage pancreatic cancer and healthy individuals to really know if it will be an accurate and reliable screening test for pancreatic cancer."
Find out more about pancreatic cancer at the Pancreatic Cancer Action Network.