Risk of pancreatic cancer doubled among diabetes patients on incretin therapy in a recent retrospective cohort study, although it is unlikely that the drugs actually caused the increase, according to investigators.

The “straightforward interpretation” that incretin-based therapies cause pancreatic cancer is “not supported” by findings of the study, wrote Dr. Mathieu Boniol, vice president of statistics, International Prevention Research Institute, Lyon, France, and colleagues.

Instead, the reason for such an increase “could be the consequence of yet-undiagnosed pancreatic cancer that induced or aggravated diabetes,” Dr. Boniol and colleagues wrote in Diabetes Care (2017 Nov 16. doi: 10.2337/dc17-0280 ).

The study was sparked by concerns about a possible increased risk of cancer among patients prescribed incretin drugs, which include glucagonlike peptide receptor–1 (GLP-1R) agonists and dipeptidyl peptidase 4 (DPP-4) inhibitors, authors said.

Results of the retrospective cohort analysis are based on two European public health insurance databases, comprising 33,292 diabetes patients treated with incretin drugs and 525,733 treated with other noninsulin antidiabetic drugs.

The risk of pancreatic cancer doubled for patients receiving incretin treatment (adjusted hazard ratio, 2.14; 95% confidence interval, 1.71-2.67), compared with patients receiving other noninsulin drugs, after adjusting for age, sex, and whether or not patients subsequently received an insulin prescription, investigators reported.

However, risk of pancreatic cancer decreased over time in patients on incretin therapy, according to further analyses by Dr. Boniol and colleagues.

That finding of diminishing risk over time argues against concluding that incretin drugs cause pancreatic cancer, according to investigators.

“A direct causal effect could be suspected if a steadily increasing risk of pancreatic cancer was associated with a steadily longer exposure to incretin drugs,” they said.

In the first 3 months after the first incretin prescription, risk of pancreatic cancer was 3.35 times greater (95% CI, 2.32-4.84), which “gradually diminished” to 1.69 (95% CI, 1.12-2.55) a year after the first prescription, the investigators said in the report.

Richard Hellman, MD, FACP, FACE, of the University of Missouri–Kansas City, said the findings are “interesting” but limited because of the study’s retrospective cohort design, length of follow-up, and potential confounding factors.

As such, these new data do not help prove or disprove a causal link between incretin use and pancreatic cancer, continued Dr. Hellman, a past president of the American Association of Clinical Endocrinologists.

“I think, right now, we don’t have the data,” he said in an interview. “It’s a question that makes us nervous, because of what we know about the laboratory effects of the drugs, but we don’t have the answer yet. We need longer studies with larger numbers of people to answer that question.”

The study, funded by Sanofi, was conducted at the request of the European Medicines Agency to assess risk of pancreatic cancer associated with lixisenatide. However, the study covers a period prior to use of lixisenatide in the general population because the approval of the GLP-1R agonist was “too recent,” the study authors wrote.

The analyses in the study were conducted in 2015 and 2016 for two cohorts of subjects: one in Belgium, for which data had been collected from 2008 to 2013, and one in Italy, with data from 2000 to 2012.

Further studies are needed to assess how long-term use of incretin drugs may affect risk of pancreatic cancer, the authors said.

The analysis was conducted as part of a postauthorization safety study requested by the European Medicines Agency and funded by Sanofi. One study author reported receiving grants from Roche, Amgen, and Bristol-Myers Squibb outside the submitted work; the remaining investigators had no relevant financial disclosures.