Diabetes drug does not increase short-term risk for pancreatic cancer, study finds

Aug. 29, 2014

A team of UNC researchers has found that drugs widely used to treat diabetes do not increase the short-term risk for pancreatic cancer.

Mugdha Gokhale

Mugdha Gokhale

The report, “Dipeptidyl-peptidase-4 inhibitors and pancreatic cancer: a cohort study,” published online Aug. 11 and in the September print issue of Diabetes, Obesity and Metabolism, was prepared by Mugdha Gokhale, MS, lead author and research assistant in the epidemiology department at UNC Gillings School of Global Public Health, along with John B. Buse MD, PhD, of the UNC School of Medicine; Christine L. Gray, MPH, Til Sturmer, MD, PhD, and Virginia Pate, MS, all of the Gillings School’s epidemiology department; and M. Allison Marquis, MSTAT, of the Collaborative Studies Coordinating Center in the Gillings School’s biostatistics department.

The study focuses on DPP-4 inhibitors, a relatively new class of diabetes medications first marketed in 2006. DPP-4 inhibitors have become among the most commonly prescribed diabetes treatments based on a favorable side-effect profile. They are expected to account for more than $6 billion in sales in 2014.

In 2009, the Food and Drug Administration issued a safety warning following reports of acute pancreatitis in patients using DPP-4 inhibitors. Subsequently, analysis of the FDA Adverse Events Reporting database and autopsy studies of people who died with diabetes suggested an increased risk of pancreatic cancer with DPP-4 inhibitors. However, the studies were criticized as limited by study design, small numbers of patients and absence of other key information.

Using a state-of-the-art new-user cohort design, the UNC study compared the rate of pancreatic cancer diagnosis among a random sample of Medicare beneficiaries who had begun using DPP-4 inhibitors with those starting two other commonly used alternative diabetes medications — sulfonylureas and thiazolidinediones. The study examined real-world treatment patterns and used a variety of well-validated techniques to minimize confounding problems that can occur in such observational studies.

The UNC study was funded by a pilot award from the N.C. Translational and Clinical Sciences Institute, the UNC home of the NIH’s Clinical and Translational Science Award. Matching funds came from the Center for Pharmacoepidemiology, based in the Gillings School’s epidemiology department, and the School of Medicine’s Diabetes Center.

The researchers found no evidence of increased short-term risk of pancreatic cancer with DPP-4 inhibitors as compared to the other classes examined. The results contradict previous evidence that suggested an increased pancreatic cancer risk.

This population-based study, which mimicked clinical treatment decisions, indicated that cancer risk among older diabetic patients starting DPP-4i does not increase even in the second year of drug exposure. However, longer treatment periods have not been studied, nor have long-term risks. That said, the results should provide some reassurance to patients and providers regarding the safety of DPP-4 inhibitors.

The UNC researchers hope to follow up with studies addressing important public health questions of diabetes drug safety.


Gillings School of Global Public Health contact: David Pesci, director of communications, (919) 962-2600 or dpesci@unc.edu.