An independent expert panel found pancreatitis was rare with the GLP-1 drug albiglutide (0.3% of patients), but the rate was higher than placebo and most cases were judged possibly drug-related.
0.3% vs 0%Adjudicated pancreatitis rate with albiglutide was 0.3% compared to 0% with placebo across the HARMONY Phase III program
What the researchers found
An independent expert panel reviewed all suspected pancreatitis cases across the entire HARMONY Phase III clinical program for albiglutide (a GLP-1 receptor agonist). Of 4,895 patients studied, 43 had potential pancreatitis cases, of which 11 were adjudicated as definite or probable acute pancreatitis (8 on albiglutide, 3 on active comparators). The pancreatitis rate with albiglutide was 0.3% (6/2,365) compared to 0% with placebo (0/486) and 0.08% with non-GLP-1RA active comparators (2/2,062 — but both confirmed cases in the comparator group were actually in patients receiving a different GLP-1RA).
While pancreatitis was uncommon overall, the rate was numerically higher with albiglutide than placebo, and the independent committee judged most cases as at least possibly related to the drug.
Why it matters
Whether GLP-1 drugs cause pancreatitis has been one of the most debated safety questions in diabetes medicine for over a decade. This rigorous independent review of nearly 5,000 patients provides some of the most carefully adjudicated data on the topic. While absolute rates were low, the signal is there — and it matters because millions of people now take GLP-1 drugs for diabetes and weight loss.
The numbers in context
n=4,895 · 2,365 on albiglutide · 43 suspected cases reviewed · 11 adjudicated definite/probable · Albiglutide: 0.3% · Placebo: 0% · Active comparators: 0.08% · 6/8 albiglutide cases possibly drug-related
How the study worked
Independent pancreatitis adjudication committee (PAC) of gastroenterology and pancreatic disease experts prospectively reviewed all suspected acute pancreatitis cases from 8 HARMONY Phase III clinical trials. Cases were classified as definite, probable, possible, or unlikely pancreatitis, and relationship to study drug was assessed. Data came from 2,365 albiglutide patients and 2,530 comparator patients.
Who was studied
4,895 patients with type 2 diabetes across 8 HARMONY Phase III trials, average age 56 years, mean 8.3-year diabetes duration
What this study cannot tell us
The number of pancreatitis events was small (11 adjudicated cases), limiting statistical power for definitive conclusions. The placebo group (486 patients) was relatively small compared to active treatment groups. Albiglutide has since been withdrawn from the market (for commercial, not safety, reasons), so these data are most relevant as part of the broader GLP-1 class safety picture. The study could not determine whether pancreatitis risk is dose- or duration-dependent.
How to read the evidence
This is an independent expert adjudication of pancreatitis cases across 8 randomized Phase III clinical trials with nearly 5,000 patients. The prospective committee design and rigorous case review make this strong evidence, though the small number of events limits statistical conclusions.
When this study was published
Published in 2019. Albiglutide (Tanzeum) was withdrawn from the market in 2018 for commercial reasons, not safety concerns. The findings remain relevant to the broader GLP-1 drug class safety discussion.
The bigger picture
The pancreatitis question has followed the GLP-1 drug class since exenatide was first approved. This study, with its independent expert adjudication, is among the most methodologically rigorous assessments of the question. While albiglutide itself is no longer marketed, the findings are relevant to the entire GLP-1 class — including semaglutide and tirzepatide, which are now used by millions. The consistent finding across multiple GLP-1 drugs of a small but real pancreatitis signal means that monitoring for this complication remains clinically important.
Questions still open
- Is the pancreatitis risk shared equally across all GLP-1 agonists, or do some (like longer-acting drugs) carry more risk than others?
- Do patients with pre-existing risk factors for pancreatitis (gallstones, high triglycerides, alcohol use) face a disproportionately higher risk on GLP-1 drugs?
- With GLP-1 drugs now used in much larger non-diabetic populations for weight loss, will the absolute number of pancreatitis cases become a more visible concern?
Common questions
Should I worry about pancreatitis if I'm taking a GLP-1 drug?
Does this apply to drugs like Ozempic and Wegovy too?
Read the original research
PANCREATIC SAFETY IN STUDIES OF THE GLUCAGON-LIKE PEPTIDE-1 RECEPTOR AGONIST ALBIGLUTIDE.
Endocrine practice : official journal of the American College of Endocrinology and the American Association of Clinical Endocrinologists, 25(7), 698-716
Citation
Al-Kawas, Firas; Anderson, Michelle Ann; Enns, Robert; Wilson, Timothy H; Johnson, Susan; Mallory, Jason M. (2019). PANCREATIC SAFETY IN STUDIES OF THE GLUCAGON-LIKE PEPTIDE-1 RECEPTOR AGONIST ALBIGLUTIDE.. Endocrine practice : official journal of the American College of Endocrinology and the American Association of Clinical Endocrinologists, 25(7), 698-716. https://doi.org/10.4158/EP-2018-0507