Comparative Efficacy and Safety of Tirzepatide in Asian and Non-Asian Adults With Obesity Without Diabetes: A Systematic Review and Meta-Analysis.

INTRODUCTION

Tirzepatide, a dual GIP/GLP-1 receptor agonist, produces substantial weight loss in adults with obesity. Given ethnic differences in obesity phenotype and cardiometabolic risk, this systematic review and meta-analysis compared tirzepatide's efficacy and safety between Asian and non-Asian adults without diabetes.

METHODS

This PROSPERO-registered review (CRD420261361351) searched PubMed, Scopus, Web of Science and ClinicalTrials.gov through 15 March 2026. Phase 3 SURMOUNT randomised, placebo-controlled trials were classified as Asian (100% Asian participants) or predominantly non-Asian (< 25% Asian participants). The primary outcome was the percentage change in body weight. Secondary outcomes included absolute weight change, body mass index (BMI), waist circumference, weight loss thresholds (≥ 5%, ≥ 10%, ≥ 15%) and adverse events. Random-effects meta-analyses and meta-regression were performed.

RESULTS

Five trials (six comparisons; N = 4022) were included: two Asian (SURMOUNT-CN, SURMOUNT-J; n = 148 tirzepatide-treated) and four predominantly non-Asian. Tirzepatide 15 mg/maximum tolerated dose produced an 18.04% greater reduction in body weight versus placebo (95% CI -19.86 to -16.22; p < 0.00001, I2 = 77%). Percentage weight loss appeared comparable between Asian (-18.16%) and non-Asian (-17.92%) trials (p = 0.94), but this underpowered comparison cannot confirm equivalence. Absolute weight loss was greater in non-Asians (-23.46 vs. -16.72 kg; p = 0.03), largely reflecting higher baseline weight. Safety profiles were similar, with no statistically significant interactions by ethnic subgroup.

CONCLUSIONS

Tirzepatide provides clinically meaningful weight reduction, with broadly similar relative percentage efficacy across Asian and non-Asian trial populations. These exploratory, indirect findings do not establish ethnic equivalence and require confirmation in adequately powered, ethnicity-stratified trials.

REGISTRATION: The meta-analysis was registered in PROSPERO under the registration number CRD420261361351. The review protocol summary is available on the PROSPERO website.