Two drugs for type 2 diabetes seem to raise risk of acute pancreatitis, study shows

BMJ 2013; 346 doi: (Published 27 February 2013) Cite this as: BMJ 2013;346:f1304
  1. Deborah Cohen
  1. 1BMJ

People with type 2 diabetes who take exenatide and sitagliptin have double the risk of hospitalisation for acute pancreatitis as people who use other antidiabetes drugs, a study published in JAMA Internal Medicine concludes, though the absolute risk is still low.1

Both drugs are glucagon-like peptide-1 (GLP1) based treatments and have previously been shown to cause acute pancreatitis in rodents and also to lead to focal proliferation in the exocrine pancreas in mice.

Reports of acute pancreatitis in humans have previously been published, the researchers, from Johns Hopkins University, Baltimore, write. But “the strength of this association and causality cannot be inferred from these reports,” they say.

So the lead author, Sonal Singh, assistant professor at Johns Hopkins University School of Medicine, and colleagues, analysed data from seven BlueCross BlueShield health insurance plans in the United States between 1 February 2005 and 31 December 2008.

They identified 1269 people with type 2 diabetes who filled at least one prescription for any drug to treat the disease between 2005 and 2008 and matched them with 1269 people with type 2 diabetes who had not used one of the GLP-1 drugs.

After controlling for confounding variables such as gallstones, obesity, and heavy alcohol use, the researchers found that people who took one of the GLP-1 drugs were twice as likely to be hospitalised with acute pancreatitis within 60 days of first taking the drugs as those who had taken different drugs (adjusted odds ratio 2.02 (95% confidence interval 1.31 to 3.01; P=0.01)).

The authors also adjusted for metformin use, because evidence indicates that it decreases the risk of pancreatitis.

Singh said that the baseline risk of pancreatitis in people with type 2 diabetes was 0.3%. “Pancreatitis is a rare complication,” he told the BMJ. “But we found that the GLP-1 based drugs double the risk to 0.6%.”

The study was funded by Johns Hopkins University and the US National Institutes of Health. Previous studies that found no association between the drugs and acute pancreatitis in animal and clinical studies were carried out or were sponsored by the marketing companies. Little is yet known about long term adverse effects of the GLP-1 class of drugs on the exocrine pancreas.

However, in the US both drugs­­—and others in the class—already carry warnings on their label about pancreatitis. In the United Kingdom several drugs in the class, including exenatide and linagliptin, currently carry a black triangle warning, which is applied to drugs under intensive safety surveillance. Others, such as sitagliptin, saxagliptin, and liraglutide, have had their warnings recently removed in the UK.

A spokesman for the UK Medicines and Healthcare Products Regulatory Agency (MHRA) told the BMJ, “Black triangle status is maintained for a minimum of two years or until the MHRA is satisfied with the emerging safety profile of the product. When considering the removal of a black triangle from a product, a variety of factors are taking into consideration.” He added, “The black triangle symbol will not be removed until the MHRA is satisfied with the safety profile of the product.”

He said that pancreatitis was well recognised in association with both the drugs and that product information contained appropriate warnings.

Exenatide is currently marketed in the United Kingdom by Eli-Lilly (but from April will be marketed by Bristol-Myers Squibb and AstraZeneca), and sitagliptin is marketed by Merck.

In response to questions from the BMJ, Merck said that it had reviewed all the safety data on sitagliptin currently available to it and that, on the basis of that review, it had found no compelling evidence establishing a causal relation between use of sitagliptin and pancreatitis or pancreatic cancer.

A spokeswoman for Lilly told the BMJ that in its clinical studies the proportion of patients experiencing pancreatitis was similar among exenatide, placebo, and comparator controlled groups of participants.

She said that five postmarketing retrospective cohort studies, including a study conducted independently of Lilly or Amylin, which originally developed exenatide, have used a longitudinal design and data on insurance claims to evaluate the risk of acute pancreatitis in patients taking the drug. “These studies have not demonstrated significantly increased risk of acute pancreatitis associated with exenatide compared to other antidiabetic therapies,” she said.

The BMJ asked companies that make or market other GLP-1 based drugs if they agreed that the increased risk of pancreatitis was a class effect. The response was mixed.

Bristol-Myers Squibb and AstraZeneca, marketers of saxagliptin, declined to comment on the class effect. Others acknowledged a possible association, including Sanofi, which markets lixisenatide. It said, “Acute pancreatitis has been possibly associated with the use of incretin based therapies [GLP-1 based drugs].”

Boehringer Ingelheim acknowledged that various drugs in the class were associated with pancreatitis. “The number of reports of pancreatitis was higher in linagliptin treated patients compared to comparator treated patients. The data are reflected in the approved linagliptin labels, and pancreatitis is included in the side effect profile,” a spokesperson said.

Takeda, which makes alogliptin (marketed as Nesina), told the BMJ that the JAMA Internal Medicine study had not looked at its drug. But it said, “In the clinical trial program for Nesina, pancreatitis was reported in 11 of 5902 (0.2%) patients receiving Nesina 25 mg daily, compared to five of 5183 (0.1%) patients receiving all comparators.”

Novo Nordisk didn’t respond to questions.


Cite this as: BMJ 2013;346:f1304