Pancreatic cancer hereditary

When a 40-year-old man came to see gastroenterologist Teresa Brentnall, the University of Washington researcher had no idea the patient would ultimately shed new light on the genetics of pancreatic cancer, one of the deadliest malignancies, as well as change the course of her career.

The patient, recently diagnosed with diabetes, was greatly concerned because his father, uncles, and cousins had died from cancer of the pancreas. Familial pancreatic cancer, in which members are at high risk for developing the disease, accounts for about 10 percent of cases of the disease and is a robust area of research.

Five years after that initial visit, Dr. Brentnall's patient and his family are still helping investigators gain better insight into the changes that occur before pancreatic cancer develops--an understanding that will hopefully lead to earlier diagnosis, better treatments, and (potentially) the cure for the disease.

"The outlook on pancreatic cancer is exponentially improving, after not changing much in the past hundred years," Dr. Brentnall says. "Thanks to new breakthroughs in molecular biology, we can now gain a much better understanding of pancreatic cancer and how it forms. Understanding the mechanisms of how pancreatic cancer forms is critical to early diagnosis."

Q: Could you tell this about Family X?

A: Family X is a large family that inherits pancreatic cancer. Each family member has a 50-50 chance of getting pancreatic cancer. Many of the family members, not all, would get diabetes before they develop pancreatic cancer. Diabetes is a disease you get when your pancreas is not working. In this family, it was a tipoff that there was something up with the pancreas, and that cancerous or precancerous changes might be forming. I first met the family when one of the family members came to see me; he had just been diagnosed with diabetes and was worried that he was developing pancreatic cancer. He was about 40 years old at that time. The patient's father, uncles, and cousins had died of pancreatic cancer. Although some efforts had been made to perform early diagnosis for some of these other family members, the cancer would be missed every time. Usually, one of the tests that doctors use to look for pancreatic cancer is a CT scan. Some of the cousins had been followed by CT scan, and yet they still developed incurable pancreatic cancer. Amazingly, the cousins' CT scans were completely normal looking--so the CT scan wasn't sensitive enough to detect very early pancreatic cancer while it might be curable. That was a big learning experience for me. If you can't detect early pancreatic cancer with a CT scan, we had to figure out another way to do it.

Q: How did you do it?

A: We used two tests. One is called the endoscopic ultrasound. People are sedated and then swallow a flexible tube with an ultrasound probe on the end of it; the tube is passed down into the stomach. Using this probe, we get a very good ultrasonic look at the pancreas. It is very sensitive in telling us if there is something wrong in the pancreas. The changes that we see are not specific for pancreatic cancer or precancer because the changes can also be seen in chronic pancreatitis (chronic inflammation). But in the setting of a family that inherits pancreatic cancer, the abnormalities can tell us these patients are in trouble--for these families, it says, "OK, pay attention, because this person may be having a problem."

If the endoscopic ultrasound is abnormal, we do an endoscopic retrograde cholangiopancreatography (ERCP)--a test that looks at the pancreatic duct. We put contrast into the pancreatic duct and take an x-ray to see if the duct looks abnormal or not. I might add that it is very important to have these endoscopic tests performed by doctors who are experts in pancreatic cancer. If both the endoscopic ultrasound and ERCP look abnormal in these patients--with this family history of pancreatic cancer--then we recommend surgery if they are good surgical candidates.

Q: Is that a prophylactic measure?

A: Sort of. Pancreatic cancer is almost always lethal--most patients die within one year of diagnosis. On the other hand, we don't want to take the pancreas out unless we have to, because when the pancreas is surgically removed, patients get diabetes, and it can be very difficult to control. Therefore, the goal in these high-risk patients is to take the pancreas out before the cancer develops, but after the pancreas has developed precancer. It is hard to tell when that is going to happen. But using these tests (ERCP and endoscopic ultrasound), we can identify patients who may be developing the precancerous changes. The final step is to make sure that precancer is present before we take the pancreas out. A sample of the pancreas is removed in an operation to see whether precancer is present. A pathologist examines the tissue under the microscope; if precancer is present, we take the whole pancreas out.

Q: Don't needle biopsies show anything?

A: Needle biopsies are difficult because the precancer changes can be focal and the needle biopsy is a very small piece of tissue, so there is a reasonable chance that you might miss the precancer. Therefore, we get a nice-sized piece of tissue at an operation.

Q: Are you looking for pancreatic adenocarcinoma?

A: Pancreatic adenocarcinoma forms over time probably in a stepwise fashion. Initially, pancreatic ductal cells develop precancerous changes that would become increasingly abnormal until the cells begin to invade other tissues--that is, the moment that precancer turns into cancer. By studying families that inherit pancreatic cancer, we can understand how these precancer changes can turn into cancer over time and that endoscopic ultrasound and ERCP can help detect the family members who might be undergoing these changes. So far, we have followed ten familial pancreatic cancer patients who have had these abnormal changes on endoscopic ultrasound and ERCP, then had precancerous changes diagnosed histologically. We have taken the pancreas out of each person. All ten of these patients are alive to date, and none has developed pancreatic cancer. In the last five years, we have also followed the other family members who are just as likely to inherit pancreatic cancer. None of those patients have developed changes on ERCP and ultrasound, and none have developed pancreatic cancer under our watch. So far, so good. We are doing pretty well.

Q: What is unique about Family X?

A: The unique thing about Family X is as follows: It is a large family--there are so many members in the family at risk. From this family, we learned how to perform effective surveillance for the early diagnosis of pancreatic cancer while it is curable.

The other unique thing is that this family has allowed us to do genotyping to try to look for the pancreatic cancer gene. We have actually found the location for the gene but have not identified which gene it is. Finding the gene is like trying to find a house. It is somewhere in the world, and it is up to you to figure out where. We have actually figured out the continent, the country, the state, and the city, but not the exact location of the house.

Q: Was it a gene already suspected to be involved in the disease?

A: No. It is not any of those. It is a brand-new spot on a chromosome where no one is suspecting there would be a pancreatic cancer gene. We haven't identified the gene yet, but we know where it is located. I think within the next year we'll have it.

Q: That would have profound impact on the treatment of the disease.

A: Yes. There are scientific paradigms, or models, that we think about in science. Often it turns out that if you can find the gene that causes the hereditary or familial form of the disease, you can open up a whole new information base for looking at treatment, early diagnosis, and chemoprevention to protect people from getting the cancer. Frequently, the genes involved in family cancers turn out to be important in sporadic cancer. An example of this would be a familial colon cancer gene called APC. Once they discovered this gene was responsible for a familial type of colon cancer, it turned out that gene was important in sporadic colon cancer, too. And there are other examples like that in medicine.

We hope that if we can discover what this pancreatic cancer gene is, we can open a whole new set of doors and ask a whole new set of questions.

Q: And Family X has a very high hereditary risk for pancreatic cancer.

A: Just because you inherit a gene doesn't mean that you are going to get the disease. But in this family, the penetrance is high, which means that if they inherit the gene, they are extremely likely to get the disease--greater than 80 percent.