Identifying Risk Factors for Pancreatic Cancer Screening

Screening and identifying pancreatic cancer at earlier stages is an often difficult task and one that does not always occur. CancerNetwork® spoke with Brian M. Wolpin, MD, MPH, director of the Gastrointestinal Cancer Center and co-director of the Pancreas and Biliary Tumor Center, and Robert T. and Judith B. Hale Chair in Pancreatic Cancer at Dana-Farber Cancer Institute, and professor of medicine at Harvard Medical School. Following his presentation at the 15th Annual Ruesch Center Symposium, Wolpin touched on the intricacies of early pancreatic cancer risk assessment, and how he hopes to see this integrated into frontline stages of care.

During the discussion, Wolpin spoke about the lethality of pancreatic cancer regarding the fact that more than 80% of the time, it presents as an advanced disease. Despite its rare nature, there are subgroups of patients—those with a strong family history of disease, those with pancreatic cysts, and those with certain genetic mutations—that possess higher levels of risk, where screening is more common.

Beyond those select few risk factors, very few, alone, are potent enough to warrant a screening from physicians. Wolpin suggested using the idea of “combinatorial risk scores” as a new risk assessment tool. Instead of only tracking the few, rare major factors, the tool could help locate patients with multiple characteristics that increase risk but aren’t significant enough to lead to more screening or treatment independently.

Wolpin emphasized that patients known to be at risk should be screened and that physicians need to be well-versed in pancreatic cancer symptoms because, although common, when they are present at the same time, there needs to be consideration of additional imaging or screening for pancreatic cancer. He also recommends that, when dealing with pancreatic cysts, patients and physicians seek out clinics that specialize in this diagnosis. In conclusion, Wolpin noted that it is also important for the goal to be curing pancreatic cancer, not prolonging treatment.

CancerNetwork: Your presentation at the Reusch Center Symposium highlighted the early detection of pancreatic cancer; what was the basis for that?

Wolpin: The presentation [I gave] at the Reusch Symposium was about the early detection of pancreatic cancer. The symposium was about many different facets of pancreatic cancer, but 1 of those [facets] was about how we can detect the disease earlier. A big problem in pancreatic cancer is that more than 80% of the time, [pancreatic cancer] presents as an advanced disease, which makes it hard to cure. A number of the sessions at the symposium were about how we can try to identify pancreatic cancers earlier, which would make them more curable.

How does the work you, and others, have done identifying parts of the genome associated with pancreatic ductal adenocarcinoma (PDAC) risk affect pancreatic cancer risk assessment?

Wolpin: There are a number of different ways that inherited risk relates to pancreatic cancer. There’s the subset of patients who have a strong family history or a known genetic mutation, like a BRCA mutation or an ATM mutation, and probably 10% to 15% of patients who develop pancreatic cancer have either that strong family history or 1 of these mutations that we know of. There are, however, a group of other studies being done trying to find more common variants in the genome that may predispose people to pancreatic cancer.

The Pancreatic Cancer Cohort Consortium and the Pancreatic Cancer Case-Control Consortium have teamed up to do a number of large studies where you look across tens of 1000s of people and try to identify areas of the genome that are related to the risk of pancreatic cancer but at a much lower rate. This means the relative risk of getting cancer for these different areas of the genome may have a 1.2-fold increase in risk, or a 1.3-fold, so [for] each individual variant in the genome, the relative risk is small. However, these areas of the genome can be inherited by much larger groups of patients, meaning 20% of the population could carry 1 of these variants. We have been working towards making this clinically useful by trying to put these variants together in a larger score that allows you to provide a readout of genetic risk beyond the more penetrating mutations like BRCA1 and BRCA2. It’s still in the research phase and not quite ready for clinic or patient care yet, but these studies are ongoing.

How do determinants like altered metabolism, inherited genetic variation, and somatic alterations affect impact survival in PDAC? How does that affect pancreatic cancer risk assessment?

Wolpin: One of the things that we have realized over time is that there aren’t that many risk factors for pancreatic cancer that, by themselves, confer a high enough risk that you could do something about them, [like] some screening approach. What we have come to realize over time is that we need to be thinking about how you combine factors of different kinds that may come together to then indicate a person has a high enough risk for pancreatic cancer that you would do something about it.

That can be metabolic alterations, like diabetes and others, combined with things like genetic risk or other risk factors for pancreatic cancer, like smoking. The goal of some of the work that a number of groups around the country have been doing—and part of what was presented at the Reusch symposium—was the idea of combinatorial risk scores, where you take these different features, look to combine them and create, essentially, a risk assessment tool. It would allow a primary care doctor to know that a patient in their clinic, out of the hundreds of patients they’re going to see over the next few months, is the 1 patient who’s at higher risk; where they may want to do something to look for pancreatic cancer, or certainly be more aware of it in case early symptoms start to develop.

The way we are thinking about this is [that] you can combine genetic determinants, metabolic determinants, and other risk factors to try to make a more clinically trackable group of patients. [Then] you could screen beyond the screening groups we’ve talked about already, like those who have a high risk because of family history or those that have a pancreatic cyst since, again, those are the minority of people, and most people don’t have those things.

What are some next steps for refining the ability to assess pancreatic cancer risk?

Wolpin: There’s been a lot of work thinking about vaccination and cancer. How do you design a vaccine that could prevent cancer or treat cancer?There’s a lot of work for pancreatic cancer going on in that space currently, both for patients who had an early tumor surgically removed and then got a vaccine after surgery to try to prevent a recurrence, but also in patients who have a high risk of pancreatic cancer [who] try to use a vaccine to prevent [themselves] from ever getting cancer in the first place. That’s the best kind of cancer; the one you never get. That’s another approach that’s being looked at now in the clinic, just in the past couple of years, that has a lot of promise as we think about how to try to tackle this difficult cancer.

What do you hope your colleagues were able to take away from your research on early pancreatic detection?

Wolpin: [The first thing] is that there are a couple groups of patients who should get screening now. It’s important that we are asking our patients about their family history. If they have a family history that includes multiple relatives with cancer, particularly pancreatic cancer, there are now clinics around the country that will see those patients and try to help them deal with their elevated risk of pancreatic cancer. That can be because they have multiple family members [who have had pancreatic cancer], or because they have an inherited mutation and pancreatic cancer in their family. That’s a key thing, and it’s talking to your patients and trying to get a sense of their family history and then referring them to an appropriate clinic to help with evaluation. There are now some data suggesting that if those patients—the patients with familial risk and genetic mutations—get regular screening, you can shift the stage at which the cancer is diagnosed to a much earlier stage.

[The] second [thing] I would say is that clinicians [should be] aware of the symptoms of pancreatic cancer. They’re somewhat nonspecific, but it’s something we should be thinking about, particularly with abdominal pain, weight loss, low appetite, and fatigue. When these things come together as a cluster of symptoms, it is worth having pancreatic cancer on the differential diagnosis list and thinking about whether an imaging test to look for would be appropriate.

A third [thing] would be that there are increasingly specialized clinics to help with pancreatic cysts. They, again, are complicated to manage because there are many of them out there, and most of them never become cancer, but some of them do. [If you’re] seeking out expertise [when] you have a patient in your clinic who has a pancreatic cyst, there are now clinics around the country that can help you manage that. If we have the appropriate management strategies around those cysts, it may allow us to diagnose the cancer early.

[One of] the last things would be that this relationship with diabetes and pancreatic cancer is important. In particular, the studies suggest that if you get diabetes at an older age, and your blood glucose is rising very quickly and you’re also losing weight, that’s a circumstance where thinking about pancreatic cancer is important. Diabetes is common and most patients with diabetes won’t have cancer, but if you have this cluster of [symptoms] coming together—rapid increase in someone’s glucose, new onset of diabetes, and a fall in somebody’s weight—those are things that are more predictive of if diabetes is related to pancreatic cancer. That is a group of patients [for whom] doing an imaging test to look for cancer would be appropriate.

The last thing I would say is there seems to be, finally, some advances coming in the treatment of disease. Sometimes we face some skepticism like “Oh, if we find it early, what are we going to do anyway?” Well, I would say that as treatments proceed and we have more advances in how we treat the disease, finding it earlier becomes more and more important, because it allows us to try to cure patients. That’s what we should all be shooting for; to try to cure the patients, not just treat them for a period of time.

Reference

Wolpin BM. Expanding the toolbox for pancreatic cancer risk assessment. Presented at the 15th Annual Ruesch Center Symposium; November 21-23; Washington, DC.