COMMENTARY

Overdue but Welcome: New Reports Guide Screening for Pancreatic-Biliary Cancers

David A. Johnson, MD

Disclosures

September 13, 2019

This transcript has been edited for clarity.

Hello. I'm Dr David Johnson, professor of medicine and chief of gastroenterology at Eastern Virginia Medical School in Norfolk, Virginia.

Welcome back to another GI Common Concerns.

Today I want to talk about screening for pancreatic and biliary cancers in patients at risk for these diseases. Two new evidence-based reviews[1,2] offer important recommendations in these areas. I would invite you all to read and have these on hand to aid conversations with your patients about screening for these cancers. However, for now, I'd like to provide you with some key takeaways.

Screening for Pancreatic Cancer

The latest data on the prevalence and mortality of pancreatic cancer highlight how important it is to screen for this cancer. In 2019, it is estimated that there will be approximately 57,000 new diagnoses of pancreatic cancer.[3] Although this is not a huge number relative to something like colon cancer, the death rate is inordinately high, with approximately 46,000 expected to die from pancreatic cancer this year.[3]

How do we screen for this aggressive cancer with a high mortality rate?

In 2004, the US Preventive Services Task Force put forth some recommendations that said there is really no good screening for asymptomatic patients with otherwise undefined average risk, meaning those without familial risk or predisposing genetic syndrome. They assigned screening such asymptomatic patients a grade D value, meaning it is not recommended.

The Task Force has now released updated 2019 recommendations[1] after reviewing the evidence systematically and excluding certain studies. Interestingly, they excluded patients who had familial cancers and genetic mutations or syndromes, but included patients who you might otherwise consider as being high risk, such as those with new-onset diabetes or a history of chronic pancreatitis, smokers, and the obese.

Of all pancreatic cancers, 80%-90% have no familial or genetic syndromic association. A relatively low 5%-10% will have some familial risk, defined as having two first-degree kindreds with pancreatic cancer. Then 3%-5% will have genetic syndromes, a classic example being Peutz-Jeghers syndrome.[4]

The Task Force concluded that there was no evidence at present to change what they said in 2004. There are no asymptomatic screening recommendations for pancreatic cancer. Although it is a very unfortunate diagnosis, these patients should not be screened to look for problems.

In making this recommendation, they considered not only the downstream effects, such as whether patients had subsequent testing and biopsies, but they also looked at psychosocial effects.

Those with familial pancreatic cancer most likely should be surveyed in centers of excellence with experience in this. However, when patients ask about screening for asymptomatic pancreatic cancer, we simply do not have any good evidence on that.

Surveillance for Hepatobiliary Cancers

The second recent review of interest comes to us from the American Gastroenterological Association (AGA),[2] which surveyed the latest evidence for patients with primary sclerosing cholangitis (PSC). These patients have a recognized risk for hepatobiliary cancers, in particular cholangiocarcinoma and gallbladder cancer. This update offers several best-practice recommendations for screening.

In patients with PSC, the incidence of cholangiocarcinoma is around 6%-11% at 10 years and approaches 20% at 30 years, which represents a 400-fold increased risk over the general population.[5] These people clearly need regular screening.

According to these best-practice recommendations, all adult patients with PSC, regardless of disease stage, should be screened for cholangiocarcinoma and gallbladder cancer within the first year. That window of time is important, as around a third of cholangiocarcinomas are identified within the first year of a PSC diagnosis. In particular, patients with accompanying ulcerative colitis should be surveyed much more closely, as they may also have an increased risk for colon cancer.

The second recommendation is that the surveillance for cholangiocarcinoma and gallbladder cancer should include imaging by ultrasound, CT, or MRI, with or without serum carbohydrate antigen 19-9 (CA19-9), every 6-12 months. The obvious preference is MRI, although getting this paid for is not quite so easy.

You can adjust the level of CA19-9, although there are threshold values that appear somewhat more predictive as it relates to specificity. In this particular analysis, they used a level of 129 U/mL vs 20 U/mL. Sensitivity was not that great for either—78% for both thresholds—but the respective specificity variance was 98% vs 67%.

You should also recognize that CA19-9 has a potentially misleading element. For it to be of value, patients have to manifest the Lewis antigen, which is absent in 10% of the population. Some people who have cholangiocarcinoma may not manifest the CA19-9 elevation. It is not something that I use routinely in my practice at present, but nonetheless it is recommended to consider it at 6-12 months alongside the previously mentioned imaging technology.

Further Considerations

The other piece of advice was that you should use imaging, not endoscopic retrograde cholangiopancreatography (ERCP), for routine surveillance. ERCP obviously has an increased risk as it relates to infection, certainly behind strictures.

However, the use of ERCP is recommended for evaluating dominant strictures or changes in clinical features (eg, jaundice) that would warrant a more aggressive approach. In such instances, ERCP should be done with brush cytology and fluorescence in situ hybridization analysis. If there is a suggestion of some change in the perihilar biliary strictures, it is recommended not to perform transendoscopic biopsy, given the risk for tumor seeding in these patients which may exclude them from undergoing transplantation at a later point. Be conservative as it relates to your ERCP. If you use it, it should be with a very scrutinizing evaluation. Cholangioscopy is certainly recommended in centers that can do that. Fine-needle aspiration for biliary strictures should certainly be avoided if possible.

The recommendation was also not to begin surveillance in patients with PSC with small-duct cholangiopathy or in those younger than 20 years of age. There's never been a reported case of cholangiocarcinoma in these patients, so begin your surveillance in adults once they reach the age of 20 years.

The other recommendation relates to gallbladder cancer and polyps. Certainly, in average-risk patients, we recommend incidental cholecystectomy if a gallbladder polyp gets to be ≥ 10 mm. As it relates to gallbladder cancer in the PSC population, cholecystectomy is recommended for polyps > 8 mm. Although not a recommendation of this expert panel, I personally believe that this surgery should be done in a center of excellence. Should that cholecystectomy go awry, that center should also be transplant-capable.

The final recommendation relates to what happens when these patients also have cirrhosis. Interestingly, there is really no formal recommendation for PSC with cirrhosis. It defaults to the standard recommendations for surveilling hepatocellular carcinoma. Although it seems to be fairly rare in this population, they should fall into the same guidance and receive surveillance with ultrasound, CT, or MRI, with or without alpha-fetoprotein every 6 months.

In summary, we now have new best-practice recommendations from both the US Preventive Services Task Force in pancreatic cancer and the AGA for hepatobiliary cancers in patients with PSC. These are frustrating cancers to deal with, but knowing the latest evidence will help you in your discussions with patients about screening. These are a useful resource to have in your library.

I'm Dr David Johnson. Thanks again for listening, and see you next time.

Follow Medscape on Facebook, Twitter, Instagram, and YouTube

Comments

3090D553-9492-4563-8681-AD288FA52ACE
Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as:

processing....