Comparative Efficacy of Glucagon-Like Peptide 1 Receptor Agonists for Cardiovascular Outcomes in Asian Versus White Populations: Systematic Review and Meta-analysis of Randomized Trials of Populations With or Without Type 2 Diabetes and/or Overweight or Obesity

Results from meta-analyses of randomized cardiovascular outcome trials (CVOTs) have underscored the cardiovascular (CV) benefits of newer antihyperglycemic medications, such as glucagon-like peptide 1 receptor agonists (GLP-1RAs), which consistently reduce major adverse cardiovascular events (MACE) compared with placebo, in addition to having glycemic benefits in people with type 2 diabetes (T2D) (1).

Numerous epidemiologic studies have shown that Asian people, compared with other racial groups, exhibit a higher risk of T2D and atherosclerotic cardiovascular disease (ASCVD), even when the cardiometabolic risk profiles are similar between groups (2). For this reason, many Asian countries have more stringent guidelines to identify and manage high-risk individuals. Given this difference in ASCVD risk, it is possible that medications proven effective in overall trial cohorts may have differential efficacy for Asian people. For the GLP-1RAs, there are signals within the completed CVOTs that this may in fact be the case.

While Asian individuals comprise nearly 60% of the global population, they often account for ≤10% of participants in CVOTs. Therefore, meta-analyses focusing on Asian people with T2D have been particularly useful in understanding the potential differential CV benefit (3,4). Results from these studies highlighted a significant reduction in MACE with the use of GLP-1RAs compared with placebo among Asian people. One of these meta-analyses, which included six GLP-1RA trials, suggested that Asian people with T2D might derive greater MACE benefits from GLP-1RAs than White people, although these findings were of borderline statistical significance (3). However, prior meta-analyses did not consider absolute benefits or data on numbers needed to treat (NNTs) to help place results into clinical context (3,4).

The Semaglutide Effects on Cardiovascular Outcomes in People with Overweight or Obesity (SELECT) trial evaluated the CV efficacy of the GLP-1RA semaglutide versus placebo in individuals with overweight or obesity and ASCVD, but without diabetes, providing additional data across diverse racial groups (5). These analyses synthesized hazard ratio (HR), absolute risk reduction, and NNT estimates from CVOTs of GLP-1RAs, including SELECT, to investigate potential differences in CV efficacy between Asian and White people.

A systematic review, registered with PROSPERO (registration number CRD42024493010), was conducted to identify relevant trials published between 1 January 2015 and 1 November 2024. A search, executed in PubMed and ClinicalTrials.gov, targeted large, randomized placebo-controlled trials of GLP-1RAs with a sample size of 500 or more. The search strategy is illustrated in the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow diagram (Supplementary Fig. 1).

Eligible trials were CVOTs of GLP-1RAs reporting HRs with 95% CIs for MACE among Asian and White subpopulations. The search included trials independent of whether eligibility criteria required prevalent diabetes or living with overweight or obesity.

Data extraction focused on the HRs (95% CIs) for specified outcomes across the two racial subgroups of interest. This task was performed independently by two investigators (M.M.Y.L. and N.S.) according to PRISMA guidelines. The risk of bias within individual trials was assessed using the Cochrane Collaboration’s tool for assessing the risk of bias in randomized trials (RoB 2). The quality of evidence was assessed following the Grading of Recommendations Assessment, Development, and Evaluation (GRADE) approach.

The primary outcome of interest was three-point major adverse CV events (MACE, e.g., CV death, myocardial infarction, and/or stroke). Meta-analyses were performed using Stata/SE 18.0, using a random-effects model according to the DerSimonian and Laird method. The primary objective was to compare the effects of GLP-1RAs on MACE outcomes between Asian and White people, supplemented by tests for interaction to evaluate the significance of any differences in efficacy estimates observed between these groups (P_interaction). To account for the inclusion of trials with participants both with and without diabetes, an additional subgroup analysis was performed, stratified by diabetes status.

This systematic search identified 13 full-text articles from 11 GLP-1RA trials (Supplementary Fig. 1). The ELIXA (Evaluation of Lixisenatide in Acute Coronary Syndrome) trial was excluded due to unavailable HR data for Asian and White subgroups (6). Additionally, the FREEDOM-CVO (A Study to Evaluate Cardiovascular Outcomes in Patients With Type 2 Diabetes Treated With ITCA 650) trial was omitted for not reporting primary MACE outcomes in Asian people and using a unique delivery method (continuous subcutaneous infusion of exenatide [ITCA 650]), which yielded divergent results from other GLP-1RA CVOTs (7,8). The FLOW (Evaluate Renal Function with Semaglutide Once Weekly) trial, while reporting primary chronic kidney disease outcomes for Asian and White subgroups, did not report MACE outcomes stratified by these subgroups (9).

Consequently, the present analyses included eight GLP-1RA trials (5,10–16) (Table 1). These trials reported on MACE, contributing to comparative assessment of outcomes between Asian and White people. The trials were assessed as high quality with a low risk of bias (Supplementary Fig. 2).

These analyses encompassed data from 5,909 Asian and 55,855 White participants across the eight trials. Overall, the pooled estimate of efficacy for MACE was HR 0.83 (95% CI 0.78, 0.89; high-certainty evidence) (Supplementary Table 1). For MACE outcomes, the pooled HR was 0.69 (95% CI 0.58, 0.83) in Asian people compared with 0.85 (95% CI 0.79, 0.91) in White people, with statistically significant interaction of treatment by race (P_interaction = 0.045) (Fig. 1). Absolute risk reduction across trials analyzed ranged from 1.4% to 4.6% in Asian people compared with a range of 0.4% to 2.7% in White people (Supplementary Fig. 3). The pooled absolute risk reduction for MACE outcomes in Asian people was 2.9% (95% CI 1.5, 4.2) compared with 1.4% (95% CI 0.9, 1.9) in White people. The NNT was 35 (24–66) and 73 (54–112) in Asian and White people, respectively.

A subgroup analysis stratified by diabetes status revealed no statistically significant interaction between individuals with diabetes (HR 0.85; 95% CI 0.80, 0.90) and those without diabetes (HR 0.80; 95% CI 0.72, 0.89) (P_interaction = 0.345) (Supplementary Fig. 4).

The results of these meta-analyses add to those reported earlier by including results from the Effect of Efpeglenatide on Cardiovascular Outcomes (AMPLITUDE-O) and SELECT trials (3,4). The results of the present analyses suggest more pronounced CV benefits of GLP-1RAs for Asian people compared with White people in both relative and absolute risk reductions. GLP-1RAs had an approximately twofold greater relative hazard reduction and twofold greater absolute risk reduction in MACE risk in Asian people compared with White people. Potential explanations for these suggestive findings are unclear. One might speculate that, since Asian people tend to be smaller in overall body size on average than White people, the relative dose given, scaled to body size, is, on average, higher than that in White people. This is potentially relevant, as results from secondary analyses from the AMPLITUDE-O trial suggested that higher doses of efpeglenatide (6-mg vs. 4-mg arm) yielded greater MACE benefit relative to the placebo arm (17). Another potential explanation is that Asian participants in clinical trials tend to have greater medication concordance than is observed in White participants, at least as seen in the U.K. (18). A third possibility is that since weighting of risk factors for developing CV outcomes in South Asian individuals may be more related to ectopic fat gain and related metabolic perturbances, perhaps particularly diabetes, GLP-1RAs may better target pathways responsible for ASCVD in Asian compared with White individuals (19). Another possibility is that since Asian people tend to have lower baseline incretin levels, GLP-1RAs are correcting a greater incretin “deficiency” in Asian compared with White people (4). Asian people also tend to exhibit a higher postprandial glycemic response, driven by faster gastric emptying (20). The slowing of gastric emptying and reduction of postprandial hyperglycemia by GLP-1RAs may partially explain the more pronounced CV benefits seen in Asians. However, these potential explanations are speculative in the absence of supporting mechanistic data, and many other explanations may also be relevant.

This study’s limitations include analyses of trial-level summary data, preventing detailed subgroup analyses, and the lack of individual patient-level data for a finer adjustment of confounding factors. We also accept that subgroup analyses often are not reproducible, so there remains the chance of type 1 error. However, the directional consistency in the data for Asian participants in all eight trials with I² = 0 lends some confidence to the findings. Finally, the broad Asian category covers a diverse range of races and ethnicities, each with unique risk profiles for T2D, overweight or obesity, and CV disease, suggesting underlying biological differences might influence treatment efficacy, while grouping them together may mask important within-group variations in baseline risks for these conditions. However, data on specific Asian subgroups (e.g., groups from Central, East, South, Southeast, and Western Asia) were not available, limiting our ability to explore these differences in greater detail. Overall, the P_interaction value of 0.045 for MACE reduction between Asian and White subgroups should be interpreted with caution, given its modest significance and the study’s inherent limitations.

Further research should focus on mechanistic studies and individual participant data meta-analyses to refine understanding of GLP-1RA effects across races and ethnicities, aiming for more tailored treatment approaches. These findings advocate for more inclusive clinical trials with detailed racial categorization to better understand the efficacy differences across racial groups. This approach is vital for advancing personalized medicine and improving outcomes in diverse populations with diabetes and/or overweight or obesity.

Acknowledgments. The authors thank Liz Coyle, University of Glasgow, for assistance with the preparation of this article.

N.S. is an editor of Diabetes Care but was not involved in any of the decisions regarding review of the manuscript or its acceptance.

Funding. Y.M.K. is funded by a T32 postdoctoral training grant from the National Institute of Diabetes and Digestive and Kidney Diseases (5T32DK007529).

The funders had no role in the design and conduct of the study; collection, management, analyses, and interpretation of the data; preparation, review, or approval of the manuscript; or decision to submit the manuscript for publication.

Duality of Interest. M.M.Y.L. has received research grants through his institution, the University of Glasgow, from AstraZeneca, Boehringer Ingelheim, and Roche Diagnostics and is a member of a Trial Steering Committee for Cytokinetics and a Clinical Endpoints Committee for Bayer. N.G. has received speaker honoraria from Novo Nordisk (nonpromotional activity) and Boehringer Ingelheim within the past 3 years. A.M. reports consulting/speaker honoraria from Abbott Laboratories, AstraZeneca, Boehringer Ingelheim, Janssen, Sanofi, Lupin, US Vitamins, Cipla, and Glenmark. M.K.R. reports consulting honoraria from Eli Lilly. H.C.G. holds the McMaster-Sanofi Population Health Institute Chair in Diabetes Research and Care. He reports research grants from Eli Lilly, AstraZeneca, Novo Nordisk, Hanmi, and Merck; continuing education grants from Eli Lilly, Abbott, Sanofi, Novo Nordisk, and Boehringer Ingelheim; honoraria for speaking from AstraZeneca, Eli Lilly, Novo Nordisk, DKSH, Zuellig, Sanofi, Carbon Brand, and Jiangsu Hanson; and consulting fees from Abbott, Bayer, Eli Lilly, Novo Nordisk, Pfizer, Sanofi, Kowa, and Hanmi. D.K.M. has received honoraria for trial leadership from Boehringer Ingelheim, Pfizer, AstraZeneca, Novo Nordisk, Esperion, Lilly USA, CSL Behring, Eidos, and NewAmsterdam and honoraria for consultancy from Lilly USA, Boehringer Ingelheim, Novo Nordisk, Applied Therapeutics, Bayer, Altimmune, Intercept, Alnylam, Amgen, and Neurotronics. N.S. reports consulting/speaker honoraria from Abbott Laboratories, AbbVie, Amgen, AstraZeneca, Boehringer Ingelheim, Eli Lilly, Hanmi Pharmaceuticals, Janssen, Menarini-Ricerche, Novartis, Novo Nordisk, Pfizer, Roche Diagnostics, and Sanofi and grants from AstraZeneca, Boehringer Ingelheim, Novartis, and Roche Diagnostics outside the submitted work. No other potential conflicts of interest relevant to this article were reported.

Author Contributions. M.M.Y.L. and N.S. extracted the data, performed the statistical analyses, interpreted the data, and drafted the manuscript. N.S., M.M.Y.L., and D.K.M. conceptualized and designed the study. All authors critically edited the manuscript. M.M.Y.L. and N.S. are the guarantors of this work and, as such, had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analyses.

Handling Editors. The journal editor responsible for overseeing the review of the manuscript was Frank Hu.