Diagnostics & Imaging Week

While several cancers are now touted as being curable, if not manageable, pancreatic cancer remains a veritable death sentence. The five-year survival rate is just 5%, according to the National Cancer Institute (Bethesda, Maryland). Researchers from Georgia Institute of Technology (Atlanta) and Saint Joseph's Hospital (Atlanta) have teamed up to challenge and alter that statistic.

"This is really an awful disease. The one thing people hear when they are told they have pancreatic cancer is Oh what a terrible disease, I'm sorry, you're going to die.' Those who treat it are faced with that reality. But I believe we have the tools now and we would really like to change that," George Daneker, MD, medical director of Oncology Research at Saint Joseph's Hospital and co-principal investigator of the new study told Diagnostics & Imaging Week.

"When I first was in fellowship, metastatic colon cancer was fatal and 20% survived for five years," he said. "Now we've changed that to a chronic disease. Sixty percent to 70% of people are alive at five years. We'd like to cure pancreatic cancer; if not we'd like to transfer it to a disease with longer survival."

The new study is focusing on the genetics and cell biology of pancreatic cancer. Tissue and serum samples from patients with cancer will be used to identify the differences in genetic and cellular features between normal and tumor cells. Findings from the research will be used to, first, develop tests for early diagnosis and then to identify specific, targeted therapies to treat each patient with pancreatic cancer.

The tools Daneker refers to, ones that are available from Georgia Tech, include a supercomputer, genomic profiling, proteomic analysis and microarray technology. This armamentarium, he said, will likely lead to a test for pancreatic cancer in three to five years.

"We're using genomic profiling as a main tool," he said. "We would like to expand genomic profiling using a standard genoarray into newer technology that will do sequencing. We're using advanced proteomic analysis with mass spectrometry. We'll also use some familiar molecular biology tools."

When Daneker was in the lab 15 years ago running experiments, it took four days to look at one protein or a couple of genes. With current gene chip and microarray technologies, it takes a technician or post-doctoral student a day to do an expression of 20,000 genes. "The Georgia Tech group is very good at looking at the huge mounds of data and making sense out of that data," he said.

Making sense of a lot of data comes via Georgia Tech's supercomputer, which will be used to tease out the chemical structures of pancreatic cancer molecules.

"We're looking at 10,000 peaks of glycoproteins off of a single serum sample," he said. "With nearly 100 patients with cancer, each had sera subjected to mass spectometry. That's a lot of data points. This is where you really require super-computational ability to manipulate the data to identify trends and do profiling."

Daneker admitted that the impetus for the study is long-term burnout in treating patients with pancreatic cancer.

"My background and training is in surgical oncology," he said. "Over the years one of my specialties has been surgical management of pancreatic cancer. We've seen several thousands with pancreatic cancer and we've participated in the miserable outcomes for this disease. It got to the point where I and my collaborators became involved with our patients and didn't want to see these bad outcomes continue. There hasn't been that much progress in a cure for pancreatic cancer. But now the promise of genomics and proteomics has the capability to make a major impact."

He explained that the anatomy of the pancreas and it's location in the body – it's small and situated in the posterior abdominal wall – make it difficult to if not impossible to inspect.

"If you have a lump in your breast, it's much more likely to be found," Daneker said. "Most people with pancreatic cancer don't present for evaluation until the cancer is well advanced and they've developed jaundice or a blockage of the bile duct that runs through the pancreas. Symptoms such as back pain or weight loss are non-specific. The anatomy is such that there are major blood vessels sitting very close to the pancreas, so even a small tumor can become unresectable because of the vessels. Some of the problems with early detection could be overcome if there was a reasonable screening test like what we have for prostate or other cancers."

The first goal, to develop a test for early diagnosis as well as recurrence, would be able to characterize the cancer compared with a normal pancreas. He's also aiming for a test that will help to identify the deranged genetic cellular pathways that can be exploited for drug therapy.

"We think personalized medicine needs to be brought to pancreatic cancer," he said. "The current treatment is that everyone gets the same thing. We're looking at a time when each person's tumor and tissue can be analyzed to identify pathways to be targeted with targeted small-molecule drugs. For each person, you could say, Here are the altered pathways and combination of drug therapies' and by looking at normal tissues you can identify patients at increased risk for complications from therapy."

Another enabling technology that will be incorporated into the study is nanotechnology, particularly for imaging. Nanoparticles combined with indicator molecules could be used for targeting and better quality imaging as well as for detection and monitoring for early occurrences.

So far the team has established a protocol for biomaterial collection and has already accumulated its first sets of specimens. Daneker hopes to have enough data amassed in three to four months for publication and then translate that into the clinic within a year or two.

"We need a major breakthrough in order to improve survival," he said. "We need a close look at the black box inside the cell and that will give us targets to exploit for therapy."