are GLP1's associated with pancreatitis??

 A recent study found that GLP-1 receptor agonists were not associated with more acute pancreatitis than sulfonylureas. There was a small increase in GLP-1 related pancreatitis that was offset by decreased alcohol- and triglyceride-related pancreatitis attributable to the GLP-1s: see GLP-1 pancreatitisBMJ2026 in dropbox, or doi:10.1136/ bmjmed-2025-002183

Details:
-- 333,687 patients were studied who had type 2 diabetes and antidiabetic therapy was initiated by a GLP-1 receptor agonist (n=132,551) or a sulfonylurea (n= 201,136), between January 2017 and January 2024, from the Veterans Affairs healthcare system databases

-- mean age 66, 75% white/21% black
-- smoking: never 43%/former 38%/current 20%; alcohol use disorder 6%
-- area deprivation index 57 (a composite of socioeconomic disadvantage based on education, income, employment, and other poverty measures, measured at the participants' location, with values ranging from low to high of 0 to 100)
-- BMI 34, A1c 8.6%, eGFR 76
-- LDL 87 mg/dL, triglycerides 204
-- serum calcium 9, AST 25, ALT 32, alkaline phosphatase 84, bilirubin 0.6
-- systolic BP 134 mmHg, diastolic 77 mmHg
-- health care utilization in the prior year: inpatient visits 0.2; outpatient visits 3; blood panel visits 3; number of prescriptions 12; number of HbA1c measurements 2
-- medications:
    -- metformin 67% (duration 684 days), insulin 38% (duration 938 days), DPP4 14% (duration 408 days), TZDs 3% (duration 403 days), SGLT-2 19% (duration 360 days)
    -- ACE/ARB 60%, calcium channel blocker 29%, b-blocker 42%, diuretics 40% (thiazides 25%), statins 77%, fibrates 4%
-- comorbidities:
    -- stroke 3%, acute coronary disease 3%, MI 1%, ischemic cardiomyopathy 2%, angina 2%, heart failure 9%, non-ischemic cardiomyopathy 3%
    -- other comorbidities having prevalence of >4%: abdominal pain 5%, GERD 23%, NAFLD 4%, acute kidney injury 8%, cancer 9%
-- for baseline characteristics after statistical weighting to equalize above differences, see their supplementary table S5
    -- adjustments in statistical analysis (by far the most exhaustive list i have ever seen...):
        -- sociodemographic variables: age, race (white, black, and other), sex, rurality (urban, rural, highly rural, and isolated island), and area social deprivation index (defined as above)
        -- covariates included systolic and diastolic blood pressure and body mass index
        -- laboratory measurements: eGFR, albuminuria, serum albumin, low density lipoprotein, triglycerides, serum calcium, aspartate aminotransferase and alanine aminotransferase, alkaline phosphatase, and total bilirubin
        -- diabetes status and antihyperglycemic initiation: hemoglobin A1c at baseline, average hemoglobin A1c within one year before beginning of study and separately within five years before study initiation, duration of diabetes from October 1999 until beginning of study, use of metformin, sulfonylureas, thiazolidinediones, dipeptidyl peptidase-4 inhibitors (DPP4i), sodium-glucose cotransporter-2 inhibitors (SGLT2i), and insulin, as well as the duration of taking metformin, sulfonylureas, thiazolidinediones, DPP4i, SGLT2i, and insulin use before initiation of study
        -- circulatory comorbidities: stroke, transient ischemic attack, atrial fibrillation, arrhythmias, acute coronary disease, myocardial infarction, ischemic cardiomyopathy, angina, heart failure, and non-ischemic cardiomyopathy
        -- gastrointestinal comorbidities including nausea, diarrhea, abdominal pain, dyspepsia, constipation, gastroesophageal reflux disease, gastroparesis, gastritis, non-alcoholic fatty liver disease, ulcerative colitis, Crohn's disease, intestinal obstruction and ileus, diverticulosis and diverticulitis, biliary tract disease, hepatic failure, gastrointestinal hemorrhage, gallstone and cholecystitis, cholecystectomy, primary sclerosing cholangitis, other cholestatic liver disease, liver cancer, pancreatic cancer, pancreatic cysts and other pancreatic anatomic abnormalities, cancer of the gallbladder or biliary system, endoscopic retrograde cholangiopancreatography, bariatric surgery, and other abdominal surgeries.
        -- other conditions: acute kidney injury, urinary tract infections, fluid and electrolyte disorders, hypoglycemia, diabetic ketoacidosis, HIV, primary hyperparathyroidism, cystic fibrosis, other cancers, peripheral artery disease, peripheral neuropathy, and autoimmune disorders
        -- health behavior and healthcare use: smoking status and alcohol use disorder; long term care, number of outpatient encounters, number of hospital admissions, number of prescriptions, number of blood panel tests, number of hemoglobin A1c measurements, and number of outpatient encounters and hospital admissions (accessed through Medicare) The calendar week of treatment assignment was also adjusted for.
        -- medications: statins, other antihyperlipidemic agents, angiotensin converting enzyme inhibitors, angiotensin receptor blockers, beta blockers, diuretics, calcium channel blockers, anti-obesity medications including bupropion, naltrexone, orlistat, phentermine, topiramate, and medications that may increase risk of pancreatitis (valproate, azathioprine, 6-mercaptopurine, isotretinoin, tetracyclines, trimethoprim/sulfamethoxazole, thiazides, and fibrates)

--MAIN OUTCOME MEASURES Risks of acute pancreatitis during 12 months’ follow-up including all-cause acute pancreatitis and cause-specific acute pancreatitis: eg, pancreatitis that is suspected to be drug induced, alcohol-induced, hypertriglyceridemia (triglycerides >500mg/dL) likely causing the pancreatitis, biliary, and idiopathic or other cause. The risks were measured through discrete time survival models after application of inverse probability weighting to statistically equalize the 2 groups

-- Follow-up of mean of 359.74 days

Results:

--all-cause pancreatitis at one year, in intention-to-treat analyses (all statistically adjusted as above):

    -- GLP-1 initiators vs sulfonylurea initiators: overall rate difference −3.64 (−30.76 to 23.48) per 100,000 persons)

       -- no statistically significant difference between the 2 groups

    -- GLP-1 was associated with an increased risk of suspected drug-induced acute pancreatitis at 23.45 (14.27 to 33.85) per 100,000 persons

        -- but GLP-1 was associated with:

              -- decreased risk of hypertriglyceridemia-associated pancreatitis: −16.96 (−27.41 to −7.34) per 100,000 persons

              -- decreased risk of alcohol-induced pancreatitis: −10.32 (−18.12 to −3.17) per 100,000 persons

        -- in sum, the overall null finding for the GLP-1's was the net result of these countervailing effects: GLP-1 associated pancreatitis increases, but then GLP-1 associated decreases in alcohol consumption and hypertriglyceridemia that both decrease acute pancreatitis incidence (see graph below)

-- idiopathic biliary causes of acute pancreatitis (ie not related to meds), comparing GLP-1 and sulfonylurea: -2.66 (-12.82 to 7.40), not significant

-- and other causes of acute pancreatitis not associated with GLP-1 or sulfonylurea were not significant 2.85 (-1956 to 26.44)

-- the graphs below reinforce the above, including the timing of the pancreatitis

-- in per protocol analyses (ie those patients actually taking the assigned drug through follow-up), there were similar results to the intention-to-treat analyses above

-- semaglutide (the most widely prescribed GLP-1 globally) was assessed specifically as a prespecified GLP-1: no difference from the above results

-- temporal distribution of the various causes of acute pancreatitis, evaluating the monthly risk over 12 months of follow- up.

    -- the risk of suspected drug-induced pancreatitis was highest in the first three months of the first year of therapy, when over 40% occurred (15.8%, 14.1%, and 12.1% during the first, second, and third months, respectively).

    -- conversely, the risk of alcohol induced pancreatitis showed its steepest decline during months 4- 6, with these months accounting for approximately 33% of the total one-year risk reduction

    -- there was a similar trend for hypertriglyceridemia associated pancreatitis, where the risk declined most sharply in the final quarter (months 10-12)

    -- in contrast, and as expected for causes with no significant association with GLP-1RA use, the risks for biliary pancreatitis and idiopathic pancreatitis remained stable throughout the follow- up period

        -- hence, these divergent temporal trends resulted in a net increase in all cause pancreatitis risk during the first two months of treatment, but with lower monthly risk subsequently

--summary graph, highlighting incidence of different causes of pancreatitis along with the differences in when in the course of the study the pancreatitis typically occurred:

Commentary:

-- the concern about GLP-1 use and pancreatitis is increasing in the setting of GLP-1 prescriptions increasingly for varied potential indications for use, including for weight loss, decreases in smoking and alcohol use, metabolic dysfunction-associated with steatohepatitis, chronic kidney disease, cardiovascular disease, lipid improvement, obstructive sleep apnea, and even potential benefit for Parkinson's disease, and Alzheimer's (GLP-1 is produced in the brain and there are GLP-1 receptors there with evidence of GLP-1s decreasing neuroinflammation): https://www.pharmacytimes.com/view/five-unexpected-new-uses-for-glp-1-receptor-agonists. there are animal studies and some observational data that cocaine use is decreased with GLP1's as well

-- both sulfonylureas and insulin do seem to have some increased risk of acute pancreatitis, in several studies:

    -- several large observational studies have found about a 2.5-fold higher risk of pancreatitis with sulfonylureas. a recent comparative study of 388,262 individuals in a claims-based study found that sulfonylureas had a higher risk of pancreatitis than GLP-1s: https://pubmed.ncbi.nlm.nih.gov/40945659/

    -- studies do suggest an increased risk of pancreatitis with insulin and increased risk of progression/recurrence of acute pancreatitis (https://pubmed.ncbi.nlm.nih.gov/31553055/), but there is likely confounding since diabetes itself is associated with pancreatitis, and insulin is more often used in patients with more advanced diabetes

-- this current study found a null association between GLP-1s and all-cause pancreatitis, and this represents the net effect of a small increased risk of drug-induced pancreatitis in the first few months of initiating GLP-1 use, but that being offset by subsequent reductions in alcohol-induced pancreatitis and hypertriglyceridemia-associated pancreatitis

-- as a point in reference, the reported incidence of pancreatitis that seems to be associated with GLP-1 use is quite small

    -- for example. the LEADER trial of liraglutide did find that in 9340 people with type 2 diabetes and high cardiovascular risk followed 3.4 years, there was a significant 28% increase in serum lipase and 7% increase in amylase; these increased in the first 6 months of use and then leveled off; this was clinically associated with 18 cases of acute pancreatitis, similar in those on liraglutide (0.4%, 1.1 events/1,000 patient-years of observation) vs in the placebo group 23 in placebo group (0.5%, 1.7 events/1,000 patient-years of observation): https://pubmed.ncbi.nlm.nih.gov/28476871/

        -- these results highlight the rarity of pancreatitis associated with GLP-1s, perhaps being less frequent than with sulfonylureas as found in the study noted above. A few, but not all, prior observational studies on GLP-1s have found an increased risk of pancreatitis with the use of GLP-1s. these studies assessed all-cause pancreatitis, which, per the above study, misses the different pancreatitis incidences associated with different cause-specific pancreatitis (ie GLP-1s appear to be associated with more drug-induced pancreatitis but less alcohol and triglyceride associated pancreatitis)

Limitations:

-- Since sulfonylureas and insulin do seem to have some increased risk of acute pancreatitis in several studies, it would have been useful to have a placebo control instead of sulfonylureas to assess the added risk of pancreatitis specifically associated with GLP-1 receptor agonists

-- as a VA study, this one has the advantages of being able to access large numbers of patients' clinical notes, diagnoses, medications, laboratory and radiology results, health care encounter at diverse settings, etc

    -- that being said, this group of patients in the VA system are different from the broad population (skewed to older men who were veterans), which might limit broader generalization of the results

-- there was also a bias to patients who had more severe cases of pancreatitis, since those with milder cases may have been seen only in outpatient settings and would not be included in the above databases

-- the data also relies on the accuracy of the coding: did the problem list/notes in the medical records accurately reflect the pancreatitis diagnosis? there could be a misclassification bias in some cases, affecting the results

    -- for example, some patients may have been classified as "pancreatitis" by their clinicians but not one specific to GLP1 use

-- the above results were controlled for many covariates, but there is still the possibility of residual confounding because of uncontrolled covariates that were not identified

so,

-- this study documents the timing of pancreatitis after initiation of GLP-1s, noting it was typically early after beginning the GLP-1. But there was a time shift to later for the effects of the pancreatitis-protective roles of GLP1's for alcohol and triglycerides associated pancreatitis

    -- clinicians, pharmacists etc should discuss this small early increased risk of pancreatitis when starting a GLP1 (on the order of 23 per 100,000 persons), and that there was protective effects against other causes of pancreatitis (alcohol, lipids) later. and, an added positive for the overall the GLP1's is in decreasing cardiac and kidney causes of death as well as all-cause mortality: eg see https://pmc.ncbi.nlm.nih.gov/articles/PMC12664052/

--given the many important clinical uses of GLP-1's, it is obscene that the cost in the US hovers in the $1000-$1500 range/month, though in India it is about $14/month: https://www.vox.com/future-perfect/484767/india-generic-semaglutide-ozempic

-- by the way, there is concern that GLP1's may be associated with muscle wasting, perhaps even more than the other means of weight loss. however, there was an impressive study using many different modalities to assess changes in body habitus finding that GLP1s actually improved skeletal muscle: https://gmodestmedblogs.blogspot.com/2024/11/glp-1-agonists-improve-skeletal-muscle.html

geoff

-----------------------------------

If you would like to be on the regular email list for upcoming blogs, please contact me at gmodest@bidmc.harvard.edu

to get access to all of the blogs:  go to http://gmodestmedblogs.blogspot.com/ to see the blogs in reverse chronological order

to get access to the dropbox: https://www.dropbox.com/scl/fo/pcf4vymtpo77f49wd2ljr/ANyJqB40CuBd-DMEr67XDoI?rlkey=5dsalmuhf8ekd0svsgvn93xm1&dl=0

or you can just click on the magnifying glass on top right, then type in a name in the search box and get all the blogs with that name in them