The American Association of Clinical Endocrinologists and the American Diabetes Association Respond to a Study Finding that GLP-1-Based Therapies Increase the Risk of Pancreatitis
A study in the Journal of the American Medical Association (JAMA) found that patients with type 2 diabetes who had been taking the GLP-1 agonists Bydureon (exenatide once-weekly) or Byetta (exenatide twice-daily) or the DPP-4 inhibitor Januvia (sitagliptin) within the last two years had higher odds of hospitalization for pancreatitis than those who never used these drugs. In response to the study, the American Association of Clinical Endocrinologists (AACE) and the American Diabetes Association (ADA) released a joint statement, which argued that the study “does not provide the basis for changing treatment in people with diabetes,” and that we need to wait for results from “prospective randomized controlled clinical trials (RCT), the gold standard for evaluating treatments.” Prospective RCTs are designed to investigate cause and effect – they are designed up front to measure the impact of an “intervention” (in this case GLP-1 agonists or DPP-4 inhibitors) on an “outcome” (like pancreatitis); in contrast, observational studies like the JAMA study look at databases to analyze patient data that were collected in “real life” and outside of a controlled trial setting. The latter method makes it more difficult to determine the effects of GLP-1 agonists on pancreatitis because the relationship is more likely to be confused by other risk factors, called confounding factors. The AACE and ADA rarely comment on clinical studies, and their joint statement suggests they are concerned the results could be misconstrued and lead patients to stop their medications.
In the JAMA study, researchers analyzed data from administrative insurance claims from seven Blue Cross Blue Shield Association Plans. They compared 1,260 people with type 2 diabetes who had pancreatitis and took the drugs with 1,260 others who did not (the control group). People from each group were matched based on age, sex, and diabetes severity. Their statistical model compared these groups and took into account possible confounding factors, which include previously taking metformin, high triglycerides, alcohol use, tobacco abuse, obesity, and pancreatic cancer among others.
Since the study is based on administrative claims data, it has limitations. For example, the researchers could not know when patient information was not coded correctly: information about patients’ tobacco or alcohol abuse or if they have obesity is not always reported. Dr. Philip Home (DM, PhD, Newcastle University, Newcastle upon Tyne, UK) noted an issue with the data: "Unfortunately the authors document that there are marked differences in the characteristics and findings between the treated and control groups. This represents a fundamental flaw in the study and means that any comparison is invalid and should not have been made." This study is aimed at identifying safety risks, and along with knowing the study’s results, it’s important to understand its limitations as well.
Past studies of GLP-1 agonists (Bydureon, Byetta, and Victoza) and DPP-4 inhibitors (Januvia, Onglyza, Tradjenta, Nesina) – and their association with pancreatitis – give conflicting results about their safety, and this JAMA study adds to that list. However, most clinical studies do not suggest a link between the drugs and pancreatitis. When we spoke with Dr. Daniel Drucker (Samuel Lunefield Research Institute, Ontario, Canada), he said there is not currently a biological mechanism for how these drugs could cause pancreatitis. He is waiting for RCTs before making his own decision, and characterized administrative-database focused research as “weak and misleading”. Currently, AACE and the ADA are waiting on the results of ongoing RCTs with over 65,000 participants to get more information on whether the drugs increase the risk of pancreatitis. The FDA is also investigating this question and has opened its own safety review. In the meantime, the AACE, ADA, and FDA recommend that people who are concerned about these drugs discuss the topic with their healthcare providers and not change their medications on their own. For background on GLP-1 agonists and DPP-4 inhibitors, see the learning curve from diaTribe #8).–MN/NR
