Skip to main content
NCBI home page
Wiley Open Access Collection logoLink to Wiley Open Access Collection
. 2026 Jan 21;28(4):2806–2814. doi: 10.1111/dom.70458

GLP‐1 receptor agonists and cardiovascular outcomes in Asian, Black or African American, and White populations: An updated meta‐analysis including the SOUL trial

Masashi Hasebe 1,2,3,4,✉, Chen‐Yang Su 1,5, Hisashi Kamido 6, Daisuke Yabe 3, Satoshi Yoshiji 1,2,5,✉
PMCID: PMC12992182  PMID: 41565576

Abstract

Aims

To evaluate the cardiovascular efficacy of glucagon‐like peptide‐1 receptor agonists (GLP‐1RAs) in Asian, Black or African American, and White populations, and to assess whether the magnitude of cardiovascular risk reduction differs across these populations.

Materials and Methods

PubMed and EMBASE were searched to 11 November 2025 for randomized placebo‐controlled GLP‐1RA trials in adults with type 2 diabetes or overweight/obesity that reported race‐stratified major adverse cardiovascular events (MACE; cardiovascular death, non‐fatal myocardial infarction, or non‐fatal stroke). Hazard ratios (HRs) for MACE were extracted for Asian, Black or African American, and White populations. Random‐effects meta‐analyses were used to obtain pooled HRs and ratios of HRs (RHRs) comparing treatment effects between populations.

Results

Nine trials, including the recent SOUL trial, were included, comprising 8164 Asian, 4036 Black or African American, and 62 503 White participants. GLP‐1RAs reduced MACE risk in Asian (HR 0.73; 95% CI 0.63–0.85; p < 0.001) and White populations (HR 0.86; 95% CI 0.81–0.91; p < 0.001). In Black or African American populations, the effect was similar to that in White populations (HR 0.88; 95% CI 0.67–1.15; p = 0.34) but did not reach statistical significance. The pooled RHR for Asian versus White populations was 0.84 (95% CI 0.71–0.98; p = 0.027), indicating a significantly greater risk reduction in Asian populations. The RHR for Asian versus Black or African American populations was 0.81 (95% CI 0.57–1.16; p = 0.25), with point estimates favouring Asian populations.

Conclusions

GLP‐1RAs reduced MACE risk across populations, with greater relative risk reduction in Asian populations and broadly similar benefits in Black or African American and White populations.

Keywords: cardiovascular disease, GLP‐1 receptor agonists, meta‐analysis, race

1. INTRODUCTION

Glucagon‐like peptide‐1 receptor agonists (GLP‐1RAs) reduce the risk of cardiovascular diseases among people with type 2 diabetes 1 and are recommended for individuals with type 2 diabetes who have elevated cardiovascular risk. 2 Beyond diabetes, GLP‐1RAs have recently demonstrated cardiovascular benefits in obese individuals without diabetes, 3 and benefits have also been observed in other cardiometabolic conditions. 4 , 5 However, the current body of evidence is largely derived from cardiovascular outcome trials (CVOTs) conducted in predominantly White populations, with limited representation of non‐White populations. In this context, “populations” refers to self‐reported or investigator‐assigned race in each study, reflecting socially defined rather than genetic categories. Given the rising burden of type 2 diabetes and cardiovascular disease across diverse underrepresented groups, it is important to determine whether the cardiovascular benefits of GLP‐1RAs are consistent across these populations. 6 , 7 , 8 , 9

Previous meta‐analyses of GLP‐1RA CVOTs have suggested possible heterogeneity in the cardiovascular efficacy of GLP‐1RAs across groups defined by race. However, earlier studies have not consistently evaluated treatment effects across Asian, Black or African American, and White participants, and a formal statistical assessment of differences in treatment effect between race groups has not been provided. 10 , 11 , 12

Recently, the SOUL trial, a large CVOT of oral semaglutide, enrolled a large number of participants,  including a particularly substantial number of Asian individuals. 13 Building on this new evidence, we performed an updated race‐stratified meta‐analysis of nine GLP‐1RA CVOTs, including SOUL, to compare the cardiovascular efficacy of GLP‐1RAs among Asian, Black or African American, and White populations and to formally assess differences in treatment effect between race groups.

2. METHODS

2.1. Search strategies and selection criteria

This study followed Preferred Reporting Items for Systematic Reviews and Meta‐Analyses (PRISMA) guidelines and was registered in PROSPERO (CRD420251163333). 14 Trials were eligible if they: (i) were randomized placebo‐controlled CVOTs enrolling ≥1000 adults (age ≥ 18 years); (ii) evaluated a GLP‐1RA administered as a subcutaneous injection or oral formulation, compared with placebo; (iii) included participants with type 2 diabetes or with overweight/obesity; (iv) reported major adverse cardiovascular events (MACE) as an endpoint with a follow‐up ≥1 year; and (v) provided effect estimates for MACE stratified by race. We excluded trials involving device‐based continuous‐infusion systems or combination products and small glycemic‐efficacy studies not designed as CVOTs. Across the CVOTs assessed, MACE was commonly defined as a composite of cardiovascular death, non‐fatal myocardial infarction, or non‐fatal stroke.

PubMed and EMBASE were systematically searched from database inception through 11 November 2025, without language restrictions. The detailed search strategies for each database are provided in Figure S1, and the process of screening and study selection is summarized in the PRISMA flow diagram (Figure S2). Two investigators (CYS and MH) conducted the literature search and study selection independently, resolving any discrepancies through discussion until agreement was reached.

2.2. Data extraction and quality assessment

We extracted hazard ratios (HRs) and 95% confidence intervals (CIs) for MACE in Asian, Black or African American, and White populations from each trial's publication or supplementary materials. Data extraction and assessment of risk of bias were carried out independently by CYS and MH, using Version 2 of the Cochrane Risk of Bias Tool. 15 Race categories were used as reported in the original trials and reflected self‐reported or investigator‐assigned race, rather than genetically inferred ancestry.

2.3. Data integration and statistical analysis

For each population defined by race, we performed random‐effects meta‐analyses of HRs to obtain pooled estimates of cardiovascular efficacy in Asian populations, Black or African American populations, and White populations, respectively. 16 To assess whether the treatment effect differed across populations, we then conducted random‐effects meta‐analyses of the ratio of HRs (RHR) between population pairs. For each trial, we first calculated a trial‐specific RHR by taking the ratio of the HRs reported for two populations (for example, HR in an Asian population divided by HR in a White population). This trial‐level contrast provides a standardized comparison within each study, allowing the relative treatment effect across populations to be evaluated consistently across trials. 17 , 18 RHRs less than 1.0 indicate a relatively greater risk reduction in the numerator population. We examined three pairwise comparisons: Asian population versus White population, Black or African American population versus White population, and Asian population versus Black or African American population.

For both HR and RHR meta‐analyses, we pooled log‐transformed effect estimates with 95% CIs using random‐effects inverse‐variance weighting (DerSimonian–Laird estimator) to account for between‐trial heterogeneity. 19 For RHRs, each trial‐specific log(RHR) was calculated as the difference between the two log(HR) estimates, and its variance was defined as the sum of the sampling variances of the corresponding log(HR) values, with standard errors derived from the reported 95% CIs. 18 , 20 , 21 Heterogeneity was quantified using Cochran's Q statistic and Higgins's I 2 , categorized as low (I 2  ≤ 25%), moderate (26%–50%), or high (>50%). 22 To assess the robustness of the results when the number of trials is limited, we repeated these random‐effects meta‐analyses using the Hartung–Knapp method, which yields more conservative 95% CIs for the pooled HRs and RHRs. 23 , 24

As a prespecified sensitivity analysis, we repeated the RHR meta‐analyses restricted to trials that enrolled only participants with type 2 diabetes. Two‐sided p values <0.05 were considered statistically significant for all analyses, including Cochran's Q test for heterogeneity. All statistical analyses were conducted using R version 4.4.1 (R Foundation for Statistical Computing, Vienna, Austria), primarily with the meta (v 8.0–1) and metafor (v 4.8–0) packages. All race‐stratified results are presented in alphabetical order (Asian population, Black or African American population, White population).

3. RESULTS

3.1. Study identification and baseline study characteristics

After removing duplicates, 779 records were identified. Title and abstract screening yielded 12 studies for full‐text review, of which three were excluded for not meeting inclusion criteria (Figure S2, Table S1). The nine included trials consisted of LEADER (liraglutide), SUSTAIN‐6 (subcutaneous semaglutide), EXSCEL (once‐weekly exenatide), Harmony Outcomes (albiglutide), PIONEER 6 (oral semaglutide), REWIND (dulaglutide), AMPLITUDE‐O (efpeglenatide), SELECT (subcutaneous semaglutide), and SOUL (oral semaglutide). 3 , 13 , 25 , 26 , 27 , 28 , 29 , 30 , 31 Across these trials, race‐stratified analyses included 8164 participants in Asian populations, 4036 participants in Black or African American populations, and 62 503 participants in White populations. SELECT enrolled participants with overweight/obesity without history of diabetes, whereas the other eight trials exclusively included individuals with type 2 diabetes. Across trials, the median follow‐up spanned 1.3 to 5.4 years. Details regarding trial characteristics are provided in Table 1. MACE definitions were broadly similar across trials (Table S2). Risk of bias assessment showed low risk of bias across all assessed domains (Table S3).

TABLE 1.

Key characteristics of the included randomized controlled trials.

Trial Year Asian population, n Black or African American population, n White population, n Drug Follow‐up, years Participant
LEADER 2016 936 777 7238 Liraglutide (SC) 1.8 mg/day 3.8 Type 2 diabetes and either atherosclerotic CVD, CKD, HF (NYHA class II‐III), or multiple cardiovascular risk factors
SUSTAIN‐6 2016 273 221 2736 Semaglutide (SC) 0.5 or 1.0 mg/week 2.1 Type 2 diabetes and either atherosclerotic CVD, CKD, HF (NYHA class II‐III), or multiple cardiovascular risk factors
EXSCEL 2017 1452 878 11 175 Exenatide (SC) 2 mg/week 3.2 Type 2 diabetes either with or without a previous atherosclerotic CVD event
HARMONY Outcomes 2018 470 225 6583 Albiglutide (SC) 30 or 50 mg/week 1.6 Type 2 diabetes and atherosclerotic CVD
REWIND 2019 434 677 7498 Dulaglutide (SC) 1.5 mg/week 5.4 Type 2 diabetes and either atherosclerotic CVD, left ventricular hypertrophy, CKD, albuminuria, or multiple cardiovascular risk factors
PIONEER 6 2019 630 192 2300 Oral semaglutide 14 mg/day 1.3 Type 2 diabetes and either atherosclerotic CVD, CKD, or multiple cardiovascular risk factors
AMPLITUDE‐O 2021 267 143 3534 Efpeglenatide (SC) 4 or 6 mg/week 1.8 Type 2 diabetes and either established atherosclerotic CVD or CKD plus at least one other cardiovascular risk factor
SELECT 2023 1447 671 14 791 Semaglutide (SC) 2.4 mg/week 3.3 BMI ≥27 kg/m2 and established atherosclerotic CVD without history of diabetes
SOUL 2025 2255 252 6648 Oral semaglutide 14 mg/day 4.1 Type 2 diabetes and either atherosclerotic CVD or CKD
Open in a new tab

Note: Follow‐up duration is presented as the median for all trials except SELECT. For SELECT, mean follow‐up duration is shown because median follow‐up duration was not reported.

Abbreviations: BMI, body mass index; CKD, chronic kidney disease; CVD, cardiovascular disease; HF, heart failure; NYHA, New York Heart Association; SC, subcutaneous.

3.2. Meta‐analyses of MACE outcomes stratified by race

During follow‐up in the placebo groups, MACE occurred in 10.2% (413/4056) of Asian populations, 12.4% (250/2020) of Black or African American populations, and 11.1% (3415/30666) of White populations. In the meta‐analysis of nine trials, GLP‐1RA therapy reduced MACE incidence in both Asian and White populations (Figure 1). In Asian populations, the pooled HR for MACE was 0.73 (95% CI 0.63–0.85; p < 0.001; I 2  = 0.0%). In White populations, the pooled HR was 0.86 (95% CI 0.81–0.91; p < 0.001; I 2  = 30.5%). In Black or African American populations, GLP‐1RA also reduced the risk of MACE (HR 0.88; 95% CI 0.67–1.15; p = 0.34; I 2  = 41. 4%), with a point estimate similar to that observed in White populations. The wider CI, compared with the other populations, likely reflected the smaller number of participants and events in Black or African American populations. Using the Hartung–Knapp method, pooled HR point estimates were unchanged; CIs were narrower for Asian populations, minimally different for White populations, and wider for Black or African American populations (Table S4).

FIGURE 1.

FIGURE 1

Open in a new tab

Effects of GLP‐1 receptor agonists (GLP‐1RAs) on major adverse cardiovascular events (MACE) stratified by race. Forest plots show hazard ratios (HRs) and 95% confidence intervals (CIs) for GLP‐1RAs versus placebo in Asian populations (top), Black or African American populations (middle), and White populations (bottom) across cardiovascular outcome trials. Diamonds represent pooled random‐effects estimates (inverse‐variance weighting).

3.3. Comparison of GLP‐1RA effects across populations

Comparisons of GLP‐1RA effects across populations defined by race are shown in Figure 2. For Asian populations versus White populations, all trial‐specific RHRs were below 1.0, and the pooled RHR was 0.84 (95% CI 0.71–0.98; p = 0.027; I 2  = 0.0%), providing evidence of effect modification by race between Asian and White populations. For Black or African American populations versus White populations, trial‐specific RHRs varied in direction, and the pooled RHR was 1.03 (95% CI 0.75–1.43; p = 0.81; I 2  = 53.0%), indicating no clear difference in treatment effect between these populations and high between‐trial heterogeneity. For Asian populations versus Black or African American populations, most trial‐specific RHRs were below 1.0 and the pooled RHR was 0.81 (95% CI 0.57–1.16; p = 0.25; I 2  = 36.9%). The direction of the effect was similar to that seen for Asian versus White populations, but this comparison was not statistically significant. Taken together, these results suggest that GLP‐1RA therapy exerted a consistently greater relative reduction in MACE risk in Asian populations, with pooled RHRs below 1.0 when compared with both White and Black or African American populations, although the latter comparison did not reach statistical significance. Using the Hartung–Knapp method, pooled RHR point estimates were unchanged; CIs were narrower for Asian populations versus White populations and wider for comparisons involving Black or African American populations (Table S5).

FIGURE 2.

FIGURE 2

Open in a new tab

Ratios of hazard ratios (RHRs) for major adverse cardiovascular events (MACE) comparing GLP‐1 receptor agonist (GLP‐1RA) effects across populations defined by race. Forest plots show trial‐specific and pooled RHRs comparing Asian populations and White populations (top), Black or African American populations and White populations (middle), and Asian populations and Black or African American populations (bottom). Diamonds represent pooled random‐effects estimates (inverse‐variance weighting). RHRs <1.0 indicate that the relative treatment effect is greater in the first‐listed population of each comparison.

In a sensitivity analysis excluding the SELECT trial, which exclusively included participants without diabetes, pooled RHRs for all three comparisons were similar to those in the main analyses, although CIs were wider and the evidence for a difference between Asian and White populations was weaker (Figure S3).

4. DISCUSSION

In this updated race‐stratified meta‐analysis of nine GLP‐1RA CVOTs, including the recently reported SOUL trial, GLP‐1RA therapy reduced the risk of MACE across Asian, Black or African American, and White populations. Across trials, Asian populations consistently showed greater relative risk reduction than the other populations, while the effect observed in Black or African American populations was similar to that in White populations but less precise, likely due to the limited number of participants. Overall, these findings indicate that the cardiovascular benefits of GLP‐1RAs are present across populations defined by race and that differences in effect size may exist between populations.

To our knowledge, this is the first race‐stratified meta‐analysis of GLP‐1RA CVOTs to include the SOUL trial, which enrolled the largest number of Asian participants (n = 2255) to date, and to evaluate treatment effects concurrently in Asian, Black or African American, and White populations while formally testing for effect modification by race. Although the number of Black or African American participants was smallest, leading to limited statistical power and wide CIs, all three populations showed a risk reduction consistent with the overall cardiovascular efficacy of GLP‐1RAs. 1 , 32 , 33 Several previous analyses have examined differences in GLP‐1RA cardiovascular efficacy across groups defined by race, but most were limited to comparisons involving only Asian and White populations or did not formally assess effect modification by meta‐analysing trial‐level RHRs. 10 , 11 , 12 By integrating groups defined by race and incorporating all available GLP‐1RA CVOTs, including SELECT—which enrolled individuals without diabetes—and the recent SOUL trial, 3 , 13 our analysis provides the most extensive evaluation to date of GLP‐1RA efficacy across populations defined by race and a quantitative assessment of differences in treatment effect between populations.

Several factors may underlie the greater cardiovascular benefit of GLP‐1RAs observed in Asian populations. Asian individuals often exhibit a distinct cardiometabolic profile, characterized by lower body mass index (BMI) but greater visceral adiposity, different lipid profiles, and higher stroke risk. 34 Because routine cardiovascular assessment frequently relies on BMI and related measures, absolute risk in Asian populations may be underestimated and background cardiometabolic therapy may be less intensive, which could make the incremental benefit of GLP‐1RAs appear larger. In addition, several studies have reported that GLP‐1RAs reduce stroke risk more than other MACE components, 35 which may amplify overall MACE reduction in Asian populations with a greater stroke burden. Differences in standard measures such as glycemia and BMI could also contribute, but reported responses to GLP‐1RAs for these measures are broadly similar in Asian and White populations. 36 These explanations remain speculative, and more granular data on individual‐level characteristics, concomitant treatments, and specific cardiovascular outcomes are needed to clarify mechanisms underlying the observed differences.

Among Black or African American participants, GLP‐1RAs tended to reduce MACE risk (HR 0.88; 95% CI 0.67–1.15), and the overall magnitude of benefit appeared similar to that in White populations (HR 0.86; 95% CI 0.81–0.91). However, Black or African American participants constituted the smallest population in our analysis, and trial‐specific estimates were imprecise with wide CIs, so modest differences in treatment effect between these populations cannot be ruled out. Consistent with this limited precision, CIs were wider with the Hartung–Knapp method than with the DerSimonian–Laird method for both HR and RHR meta‐analyses involving this group (Tables S4 and S5). This highlights an important limitation of the current evidence: most CVOTs have enrolled predominantly White populations, with relatively few participants from other populations. 37 Given the possibility that medication responses may vary across populations, 38 future cardiovascular outcome and pragmatic trials should enrol more diverse populations and include larger sample sizes in each population to allow reliable assessment of treatment effects.

An important conceptual consideration in our analysis concerns the use of race. In this study, “populations” refer to race categories as reported in the original trials, based on self‐reported or investigator‐assigned race rather than genetically inferred ancestry. Race is a socially defined construct that only partially overlaps with genetic ancestry and is closely linked to structural and environmental factors such as access to care and socioeconomic conditions. 39 , 40 Therefore, differences in treatment effect between populations in our meta‐analysis should not be taken as evidence of inherent biological differences. One limitation is that none of the included CVOTs reported outcomes stratified by genetically inferred ancestry. Future studies that incorporate ancestry information alongside social and environmental determinants of health will be important to clarify the mechanisms underlying any apparent differences across populations.

Our study also has several other limitations. First, we used trial‐level aggregate data rather than individual participant data, which prevented adjustment for individual‐level covariates. Accordingly, we could not disentangle the contribution of social and structural determinants of health, such as environmental exposures, socioeconomic conditions, and access to care, that may be correlated with the race categories reported in each trial. 41 In addition, the trial‐level nature of the available data limited exploration of potential effect modifiers within race categories, including sex and geographic region. Second, the race categories used in the trials were broad and do not capture within‐population heterogeneity. For example, Asian participants were treated as a single aggregated category, with no stratification into East Asian, South Asian, or other Asian subgroups, despite evidence that cardiometabolic risk profiles and cardiovascular disease risk differ substantially across these subgroups. 42 , 43 Third, we were unable to assess GLP‐1RA efficacy in other populations because race‐stratified results were not consistently reported. Fourth, the present study focused on relative treatment effects (HRs and RHRs) and therefore does not quantify absolute effects. Because follow‐up duration and censoring differed across trials and only race‐stratified subgroup‐level data were available, time‐specific absolute risk reductions or numbers needed to treat could not be quantified.

In summary, this race‐stratified meta‐analysis of nine GLP‐1RA CVOTs showed that GLP‐1RAs reduced MACE risk in Asian, Black or African American, and White populations. Asian populations consistently demonstrated a larger relative risk reduction than the other populations, while the benefit in Black or African American populations was similar to that in White populations, although estimates for Black or African American populations were less precise because of the smaller sample size. Taken together, these findings support the cardiovascular efficacy of GLP‐1RAs across populations and highlight the need for future trials with more diverse enrollment and larger sample sizes in each population to reliably evaluate treatment effects.

AUTHOR CONTRIBUTIONS

Masashi Hasebe conceived and designed the study, performed analyses, and interpreted the data. Chen‐Yang Su and Masashi Hasebe conducted the literature review and collected the data. Chen‐Yang Su and Masashi Hasebe drafted the initial manuscript. Hisashi Kamido, Daisuke Yabe, and Satoshi Yoshiji critically revised the manuscript and provided intellectual input. Satoshi Yoshiji supervised the study. All authors reviewed and approved the final version of the manuscript.

FUNDING INFORMATION

The Yoshiji Lab is supported by the Canada Research Chairs Program (CRC‐2025‐00097), the Canadian Institutes of Health Research (183596), the DNA to RNA (D2R) Foundational Program, Japan Society for the Promotion of Science, and McGill University. The funders had no role in the study design, conduct, analysis, or reporting.

CONFLICT OF INTEREST STATEMENT

Daisuke Yabe received clinically commissioned or joint research grants from Novo Nordisk, Ono Pharmaceutical, Taisho Pharmaceutical, Terumo, and Arklay. Daisuke Yabe also received consulting or speaker fees from Sumitomo Dainippon Pharma, Boehringer Ingelheim, Astellas Pharma, MSD, Novo Nordisk, Ono Pharmaceutical, Eli Lilly, and Takeda Pharmaceutical. Satoshi Yoshiji serves as a consultant to the Broad Institute of MIT and Harvard through Precision Global Consulting and to PriveBio, Inc., unrelated to this project. All other authors declare no conflicts of interest.

Supporting information

Data S1. Supporting Information.

Table S1. Excluded studies through detailed full‐text assessment.

Table S2. Definition of major adverse cardiovascular events (MACE) of the included trials.

Table S3. Risk of bias of the included trials.

Table S4. Random‐effects meta‐analyses of MACE stratified by race: DerSimonian–Laird vs. Hartung–Knapp methods.

Table S5. Random‐effects meta‐analyses of ratios of hazard ratios (RHRs) for MACE across populations defined by race: DerSimonian–Laird vs. Hartung–Knapp methods.

Figure S1. Full electronic search strategies for PubMed and EMBASE.

Figure S2. Preferred Reporting Items for Systematic Reviews and Meta‐Analysis (PRISMA) flow diagram for study selection.

Figure S3. Sensitivity meta‐analysis of ratios of hazard ratios (RHRs) for major adverse cardiovascular events (MACE) with GLP‐1 receptor agonists across subgroups defined by race, excluding the SELECT trial.

DOM-28-2806-s001.pdf (537.3KB, pdf)

ACKNOWLEDGEMENTS

Masashi Hasebe is supported by the Japan Student Services Organization (JASSO; Graduate Scholarship for Degree‐Seeking Study Abroad) and the Watanabe Foundation (6th Toshizo Watanabe International Scholarship). Chen‐Yang Su is supported by a CIHR Canada Graduate Scholarship‐Doctoral Award, an FRQS Doctoral Training Scholarship, and a Lady Davis Institute/TD Bank Scholarship. These scholarships had no role in the study design, conduct, analysis, or reporting.

Contributor Information

Masashi Hasebe, Email: masashi.hasebe@mail.mcgill.ca.

Satoshi Yoshiji, Email: satoshi.yoshiji@mcgill.ca.

DATA AVAILABILITY STATEMENT

All data were extracted from publicly available sources and are included in this published article and its Supporting Information file.

REFERENCES

  • 1. Hasebe M, Su CY, Keidai Y, et al. Cardiovascular risk reduction with glucagon‐like peptide‐1 receptor agonists is proportional to HbA1c lowering in type 2 diabetes: an updated meta‐regression analysis incorporating FLOW and SOUL trials. Diabetes Obes Metab. 2025;27(12):7210‐7220. doi: 10.1111/dom.70121 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2. American Diabetes Association Professional Practice C . 9. Pharmacologic approaches to glycemic treatment: standards of Care in Diabetes‐2025. Diabetes Care. 2025;48(1 Suppl 1):S181‐S206. doi: 10.2337/dc25-S009 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3. Lincoff AM, Brown‐Frandsen K, Colhoun HM, et al. Semaglutide and cardiovascular outcomes in obesity without diabetes. N Engl J Med. 2023;389(24):2221‐2232. doi: 10.1056/NEJMoa2307563 [DOI] [PubMed] [Google Scholar]
  • 4. Siddiqi TJ, Khan MS, Waqas SA, et al. Effect of glucagon‐like peptide‐1 receptor agonists on heart failure outcomes and cardiovascular death across varying cardiovascular‐kidney‐metabolic comorbidity. Eur J Heart Fail. 2025. doi: 10.1002/ejhf.70048 [DOI] [PubMed] [Google Scholar]
  • 5. Waqas SA, Sohail MU, Saad M, et al. Efficacy of GLP‐1 receptor agonists in patients with heart failure and mildly reduced or preserved ejection fraction: a systematic review and meta‐analysis. J Card Fail. 2025;31(7):1076‐1080. doi: 10.1016/j.cardfail.2025.01.022 [DOI] [PubMed] [Google Scholar]
  • 6. Collaboration NCDRF . Worldwide trends in diabetes prevalence and treatment from 1990 to 2022: a pooled analysis of 1108 population‐representative studies with 141 million participants. Lancet. 2024;404(10467):2077‐2093. doi: 10.1016/S0140-6736(24)02317-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7. Dalakoti M, Lin NHY, Yap J, et al. Primary prevention of cardiovascular disease in Asia: challenges: a narrative review. JACC Adv. 2025;4(4):101670. doi: 10.1016/j.jacadv.2025.101670 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8. Javed Z, Haisum Maqsood M, Yahya T, et al. Race, racism, and cardiovascular health: applying a social determinants of health framework to racial/ethnic disparities in cardiovascular disease. Circ Cardiovasc Qual Outcomes. 2022;15(1):e007917. doi: 10.1161/CIRCOUTCOMES.121.007917 [DOI] [PubMed] [Google Scholar]
  • 9. Ali MR, Nacer H, Lawson CA, Khunti K. Racial and ethnic disparities in primary prevention of cardiovascular disease. Can J Cardiol. 2024;40(6):1016‐1030. doi: 10.1016/j.cjca.2024.01.028 [DOI] [PubMed] [Google Scholar]
  • 10. Rivera FB, Bantayan NRB, Aparece JP, et al. Sex, racial, ethnic, and geographical disparities in major adverse cardiovascular outcome of glucagon‐like peptide‐1 receptor agonists among patients with and without diabetes mellitus: a meta‐analysis of placebo‐controlled randomized controlled trials. J Clin Lipidol. 2024;18(4):e588‐e601. doi: 10.1016/j.jacl.2024.03.011 [DOI] [PubMed] [Google Scholar]
  • 11. Lee MMY, Ghouri N, Misra A, et al. 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. Diabetes Care. 2025;48(3):489‐493. doi: 10.2337/dc24-1533 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12. Kunutsor SK, Khunti K, Seidu S. Racial, ethnic and regional differences in the effect of sodium‐glucose co‐transporter 2 inhibitors and glucagon‐like peptide 1 receptor agonists on cardiovascular and renal outcomes: a systematic review and meta‐analysis of cardiovascular outcome trials. J R Soc Med. 2024;117(8):267‐283. doi: 10.1177/01410768231198442 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13. McGuire DK, Marx N, Mulvagh SL, et al. Oral Semaglutide and cardiovascular outcomes in high‐risk type 2 diabetes. N Engl J Med. 2025;392(20):2001‐2012. doi: 10.1056/NEJMoa2501006 [DOI] [PubMed] [Google Scholar]
  • 14. Page MJ, McKenzie JE, Bossuyt PM, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ. 2021;372:n71. doi: 10.1136/bmj.n71 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15. Sterne JAC, Savovic J, Page MJ, et al. RoB 2: a revised tool for assessing risk of bias in randomised trials. BMJ. 2019;366:l4898. doi: 10.1136/bmj.l4898 [DOI] [PubMed] [Google Scholar]
  • 16. Borenstein M, Hedges LV, Higgins JP, Rothstein HR. A basic introduction to fixed‐effect and random‐effects models for meta‐analysis. Res Synth Methods. 2010;1(2):97‐111. doi: 10.1002/jrsm.12 [DOI] [PubMed] [Google Scholar]
  • 17. Song F, Bachmann MO. Cumulative subgroup analysis to reduce waste in clinical research for individualised medicine. BMC Med. 2016;14(1):197. doi: 10.1186/s12916-016-0744-x [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18. Riley RD, Debray TPA, Fisher D, et al. Individual participant data meta‐analysis to examine interactions between treatment effect and participant‐level covariates: statistical recommendations for conduct and planning. Stat Med. 2020;39(15):2115‐2137. doi: 10.1002/sim.8516 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19. DerSimonian R, Laird N. Meta‐analysis in clinical trials revisited. Contemp Clin Trials. 2015;45(Pt A):139‐145. doi: 10.1016/j.cct.2015.09.002 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20. Altman DG, Bland JM. Interaction revisited: the difference between two estimates. BMJ. 2003;326(7382):219. doi: 10.1136/bmj.326.7382.219 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21. Diallo A, Carlos‐Bolumbu M, Renard PE, Galtier F. Larger effect size in composite kidney outcomes than in major cardiovascular events associated with sodium‐glucose cotransporter‐2 (SGLT2) inhibitors compared with glucagon‐like peptide‐1 receptor agonists (GLP‐1RAs): a pooled analysis of type 2 diabetes trials. Diabetes Obes Metab. 2023;25(1):166‐176. doi: 10.1111/dom.14859 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22. Higgins JP, Thompson SG. Quantifying heterogeneity in a meta‐analysis. Stat Med. 2002;21(11):1539‐1558. doi: 10.1002/sim.1186 [DOI] [PubMed] [Google Scholar]
  • 23. Hartung J, Knapp G. A refined method for the meta‐analysis of controlled clinical trials with binary outcome. Stat Med. 2001;20(24):3875‐3889. doi: 10.1002/sim.1009 [DOI] [PubMed] [Google Scholar]
  • 24. IntHout J, Ioannidis JP, Borm GF. The Hartung‐Knapp‐Sidik‐Jonkman method for random effects meta‐analysis is straightforward and considerably outperforms the standard DerSimonian‐Laird method. BMC Med Res Methodol. 2014;14:25. doi: 10.1186/1471-2288-14-25 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25. Marso SP, Daniels GH, Brown‐Frandsen K, et al. Liraglutide and cardiovascular outcomes in type 2 diabetes. N Engl J Med. 2016;375(4):311‐322. doi: 10.1056/NEJMoa1603827 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26. Marso SP, Bain SC, Consoli A, et al. Semaglutide and cardiovascular outcomes in patients with type 2 diabetes. N Engl J Med. 2016;375(19):1834‐1844. doi: 10.1056/NEJMoa1607141 [DOI] [PubMed] [Google Scholar]
  • 27. Holman RR, Bethel MA, Mentz RJ, et al. Effects of once‐weekly exenatide on cardiovascular outcomes in type 2 diabetes. N Engl J Med. 2017;377(13):1228‐1239. doi: 10.1056/NEJMoa1612917 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28. Hernandez AF, Green JB, Janmohamed S, et al. Albiglutide and cardiovascular outcomes in patients with type 2 diabetes and cardiovascular disease (Harmony outcomes): a double‐blind, randomised placebo‐controlled trial. Lancet. 2018;392(10157):1519‐1529. doi: 10.1016/S0140-6736(18)32261-X [DOI] [PubMed] [Google Scholar]
  • 29. Husain M, Birkenfeld AL, Donsmark M, et al. Oral Semaglutide and cardiovascular outcomes in patients with type 2 diabetes. N Engl J Med. 2019;381(9):841‐851. doi: 10.1056/NEJMoa1901118 [DOI] [PubMed] [Google Scholar]
  • 30. Gerstein HC, Colhoun HM, Dagenais GR, et al. Dulaglutide and renal outcomes in type 2 diabetes: an exploratory analysis of the REWIND randomised, placebo‐controlled trial. Lancet. 2019;394(10193):131‐138. doi: 10.1016/S0140-6736(19)31150-X [DOI] [PubMed] [Google Scholar]
  • 31. Gerstein HC, Sattar N, Rosenstock J, et al. Cardiovascular and renal outcomes with Efpeglenatide in type 2 diabetes. N Engl J Med. 2021;385(10):896‐907. doi: 10.1056/NEJMoa2108269 [DOI] [PubMed] [Google Scholar]
  • 32. Lee MMY, Sattar N, Pop‐Busui R, et al. Cardiovascular and kidney outcomes and mortality with long‐acting injectable and oral glucagon‐like peptide 1 receptor agonists in individuals with type 2 diabetes: a systematic review and meta‐analysis of randomized trials. Diabetes Care. 2025;48(5):846‐859. doi: 10.2337/dc25-0241 [DOI] [PubMed] [Google Scholar]
  • 33. Hasebe M, Yoshiji S, Keidai Y, et al. Efficacy of antihyperglycemic therapies on cardiovascular and heart failure outcomes: an updated meta‐analysis and meta‐regression analysis of 35 randomized cardiovascular outcome trials. Cardiovasc Diabetol. 2023;22(1):62. doi: 10.1186/s12933-023-01773-z [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34. Kwan TW, Wong SS, Hong Y, et al. Epidemiology of diabetes and atherosclerotic cardiovascular disease among Asian American adults: implications, management, and future directions: a scientific statement from the American Heart Association. Circulation. 2023;148(1):74‐94. doi: 10.1161/CIR.0000000000001145 [DOI] [PubMed] [Google Scholar]
  • 35. Husain M, Bain SC, Jeppesen OK, et al. Semaglutide (SUSTAIN and PIONEER) reduces cardiovascular events in type 2 diabetes across varying cardiovascular risk. Diabetes Obes Metab. 2020;22(3):442‐451. doi: 10.1111/dom.13955 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36. Gan S, Dawed AY, Donnelly LA, et al. Efficacy of modern diabetes treatments DPP‐4i, SGLT‐2i, and GLP‐1RA in White and Asian patients with diabetes: a systematic review and meta‐analysis of randomized controlled trials. Diabetes Care. 2020;43(8):1948‐1957. doi: 10.2337/dc19-2419 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37. Dixon DD, Wilkins CH. Strategies to cultivate diversity and achieve equity in cardiovascular clinical trials. Circulation. 2023;148(3):204‐206. doi: 10.1161/CIRCULATIONAHA.123.065476 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38. Tamargo J, Kaski JC, Kimura T, et al. Racial and ethnic differences in pharmacotherapy to prevent coronary artery disease and thrombotic events. Eur Heart J Cardiovasc Pharmacother. 2022;8(7):738‐751. doi: 10.1093/ehjcvp/pvac040 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39. Flanagin A, Frey T, Christiansen SL, Committee AMAMoS . Updated guidance on the reporting of race and ethnicity in medical and science journals. JAMA. 2021;326(7):621‐627. doi: 10.1001/jama.2021.13304 [DOI] [PubMed] [Google Scholar]
  • 40. Cooper RS, Kaufman JS, Ward R. Race and genomics. N Engl J Med. 2003;348(12):1166‐1170. doi: 10.1056/NEJMsb022863 [DOI] [PubMed] [Google Scholar]
  • 41. Jain A, Brooks JR, Alford CC, et al. Awareness of racial and ethnic bias and potential solutions to address bias with use of health care algorithms. JAMA Health Forum. 2023;4(6):e231197. doi: 10.1001/jamahealthforum.2023.1197 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42. Satish P, Sadaf MI, Valero‐Elizondo J, et al. Heterogeneity in cardio‐metabolic risk factors and atherosclerotic cardiovascular disease among Asian groups in the United States. Am J Prev Cardiol. 2021;7:100219. doi: 10.1016/j.ajpc.2021.100219 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43. Lim A, Elias S, Benjasirisan C, et al. Heterogeneity in the prevalence of cardiovascular risk factors by ethnicity and birthplace among Asian subgroups: evidence from the 2010 to 2018 National Health Interview Survey. J Am Heart Assoc. 2024;13(5):e031886. doi: 10.1161/JAHA.123.031886 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Data S1. Supporting Information.

Table S1. Excluded studies through detailed full‐text assessment.

Table S2. Definition of major adverse cardiovascular events (MACE) of the included trials.

Table S3. Risk of bias of the included trials.

Table S4. Random‐effects meta‐analyses of MACE stratified by race: DerSimonian–Laird vs. Hartung–Knapp methods.

Table S5. Random‐effects meta‐analyses of ratios of hazard ratios (RHRs) for MACE across populations defined by race: DerSimonian–Laird vs. Hartung–Knapp methods.

Figure S1. Full electronic search strategies for PubMed and EMBASE.

Figure S2. Preferred Reporting Items for Systematic Reviews and Meta‐Analysis (PRISMA) flow diagram for study selection.

Figure S3. Sensitivity meta‐analysis of ratios of hazard ratios (RHRs) for major adverse cardiovascular events (MACE) with GLP‐1 receptor agonists across subgroups defined by race, excluding the SELECT trial.

DOM-28-2806-s001.pdf (537.3KB, pdf)

Data Availability Statement

All data were extracted from publicly available sources and are included in this published article and its Supporting Information file.

Articles from Diabetes, Obesity & Metabolism are provided here courtesy of Wiley

RESOURCES