Dose-response analysis of tirzepatide and acute pancreatitis: An international systematic review and quantitative meta-analysis of randomised trials.

BACKGROUND

Tirzepatide, a dual GIP/GLP-1 receptor agonist for type 2 diabetes and obesity, carries a theoretical risk of pancreatitis. Whether risk varies by dose is uncertain.

METHODS

Randomised controlled trials (RCTs) reporting outcomes from Tirzepatide 5, 10, and 15 mg once weekly were systematically reviewed. The primary outcome was acute pancreatitis, interpreted against the Revised Atlanta criteria where available. Pooled risk ratios (RRs) were calculated using Mantel-Haenszel random-effects models with a small continuity correction for single-zero trials; double-zero trials were excluded as non-informative. Sensitivity analyses excluded single-zero studies. Poisson generalised linear mixed models (GLMMs) with study-level random intercepts assessed robustness and explored age/sex as moderators. Risk of bias used RoB 2.0.

RESULTS

Nineteen RCTs (n = 15,471) met inclusion criteria. Acute pancreatitis was exceedingly rare (0.22%). Head-to-head dose comparisons showed no significant differences: 10 mg vs 5 mg (RR 0.78, 95% CI 0.29-2.09), 15 mg vs 5 mg (RR 0.70, 0.27-1.82), and 15 mg vs 10 mg (RR 1.13, 0.42-3.02); sensitivity analyses were concordant. GLMMs yielded similar inferences; age and sex did not materially modify risk. Most RoB 2.0 domains were low risk, with some concerns for missing data/selective reporting. All trials were industry-funded with short-to-moderate follow-up; funnel plots were underpowered.

CONCLUSIONS

Across RCTs, acute pancreatitis with Tirzepatide is rare and shows no demonstrable dose-response between 5 and 15 mg. Precision remains limited; clinicians should maintain vigilance, optimise modifiable risks, individualise dosing, and contribute to post-marketing surveillance.