LAS VEGAS – New results from a single center, retrospective analysis suggest that individuals with diabetes and pancreatic cysts have larger cyst sizes at diagnosis, and a faster subsequent cyst growth rate. Smoking was independently associated with faster growth rate.
Most pancreatic cancer patients were previously diagnosed with hyperglycemia and diabetes, and pancreatic cancer can cause diabetes. “This sort of dual causality raises questions as to whether or not hyperglycemia, or the new diagnosis of diabetes itself, could be a harbinger of cancer or precancer. And should these patients be more closely monitored?” David Robbins, MD, said in an interview.
Dr. Robbins, associate professor of medicine and program director in gastroenterology in the Northwell Health System, New York, presented the study at the annual meeting of the American College of Gastroenterology.
Faster growth rates of pancreatic cysts in the presence of diabetes are important because they represent a potential mark for cyst aggressiveness. “So the question really is, in the setting of diabetes, are there factors perhaps circulating in the bloodstream, or other intrinsic factors, that make these cysts more dangerous and require a different surveillance approach than someone who doesn’t have diabetes? We have (surveillance) guidelines that address the average population, but they don’t really hone in on what do you do with (individuals with diabetes),” Dr. Robbins said during the presentation.
The study could have implications for screening, said session moderator Dayna Early, MD, professor of medicine at Washington University and director of endoscopy at Barnes Jewish Hospital, both in St. Louis. “I think this is important information to guide us to look more closely at patients with diabetes who do have pancreatic cysts,” she said in an interview.
The study included 177 adults with pancreatic cysts or abnormal imaging results between 2013 and 2020. Sixty-five percent were female, and the mean age was 65.4 years; 64% were White, 10% were Black, and 8.5% were Asian. Among the participants, 24.8% were smokers and 32.2% had type 2 diabetes.
Patients with diabetes had larger cyst sizes (2.23 cm versus 2.76 cm), as well as a higher annual cyst growth rate (1.90 cm versus 1.30 cm). Cyst size and growth rate were similar between patients with controlled and uncontrolled diabetes. Smoking was associated with a larger cyst size overall (2.2 cm versus 1.81 cm), and were larger still among patients with diabetes who smoked (2.35 cm).
Seventy-one patients went on to have pathologic confirmation by endoscopic ultrasound-guided fine needle aspiration. “In the diabetic group, two developed adenocarcinoma, six of the nondiabetics developed adenocarcinoma, and there was no difference in CEA or serum CA 19-9,” Dr. Robbins said during his presentation.
Of 28 patients diagnosed with pancreatic cancer, 13 had type 2 diabetes.
There remains uncertainty about what cyst growth rate is most dangerous. Somerecommend that individuals with new-onset or worsening diabetes and intraductal papillary mucinous neoplasm or mucinous cystic neoplasm cysts, or cysts alone that are growing faster than 3 mm per year, may be at significantly increased risk of pancreatic cancer. These guidelines recommend that they be screened with short-interval magnetic resonance imaging or endoscopic ultrasound (EUS) fine needle aspiration. However, this recommendation is conditional and is backed by a very low level of evidence.
Other reports have shown varying risks at different growth rates. “It’s not really clear at this point. And that’s why I think, while our study is small and exploratory, this is a particular area that is relatively easy to evaluate. We have huge databases of pancreatic cyst evolution, and we know that 30 million Americans have diabetes. So, the next obvious study is to do a more systematic look at that, and work towards refining and making sense of these divergent guidelines, all of which are saying the same thing but using different threshold numbers,” said Dr. Robbins.
The next step is do larger, multicenter studies in the context of other risk factors such as family history and smoking, but the current finding represents an opportunity to catch at least some pancreatic cancers earlier, according to Dr. Robbins. He suggested that individuals with diabetes who are diagnosed with a pancreatic cyst should be referred to a gastroenterologist or another specialist to track cyst growth. “That is going to miss a lot of folks who didn’t get imaging for whatever reason (and so don’t have a cyst identified), but it is an early opportunity, and it’s better than what we’re doing now.”
During the talk, Dr. Robbins said, “Given the ease, availability and low cost of diabetes screening in the general clinic population, we encourage the inclusion of HbA1c and fasting glucose in algorithms for pancreatic cyst surveillance.”
Dr. Early found the suggestion intriguing, but wasn’t ready to lend full support. “I think looking at the suggestion of possibly monitoring hemoglobin A1c levels was novel. I don’t know that we’ll necessarily adopt that as standard practice, but that’s something I think that could be looked at in the future as a way to help risk stratify whether patients need to be surveyed more frequently,” she said.
Dr. Robbins and Dr. Early have no relevant financial disclosures.