Evidence for dulaglutide reducing major adverse cardiovascular events in type 2 diabetes (MACE outcomes)

Dulaglutide 1.5 mg weekly reduces major adverse cardiovascular events by 12% in type 2 diabetes patients, with particularly strong benefits for stroke reduction (24%) but no significant effect on myocardial infarction, and these cardiovascular benefits extend across patients with and without established cardiovascular disease.

Abstract

Dulaglutide 1.5 mg weekly reduces major adverse cardiovascular events in type 2 diabetes by 12% compared to placebo (HR 0.88, 95% CI 0.79-0.99, p=0.026), based on the REWIND trial of 9,901 participants followed for a median 5.4 years. This benefit was driven primarily by a 24% reduction in stroke (HR 0.76, p=0.010), particularly ischemic stroke (HR 0.75, p=0.012), with no significant effect on myocardial infarction (HR 0.96). The cardiovascular benefit remained consistent across multiple analytic approaches, including total cardiovascular event burden (HR 0.90) and atherosclerosis-related outcomes (HR 0.91), and was evident regardless of baseline kidney function (interaction p=0.545 for eGFR, p=0.374 for albuminuria). Mediation analysis revealed that improvements in HbA1c and albuminuria account for approximately 65% of the MACE reduction, with the remaining benefit likely attributable to direct vascular effects including improvements in endothelial function and anti-inflammatory mechanisms. Notably, the REWIND population had relatively low baseline cardiovascular risk, with only 31% having established cardiovascular disease, yet benefits extended across the spectrum from primary to secondary prevention. In SURPASS-CVOT, tirzepatide demonstrated noninferiority to dulaglutide for MACE (HR 0.92, p=0.003 for noninferiority) despite achieving greater improvements in metabolic parameters, reinforcing that cardiovascular benefits involve mechanisms beyond glucose lowering and weight reduction alone.

Methods

We analyzed 10 sources from an initial pool of 200, using 8 screening criteria. Each paper was reviewed for 8 key aspects that mattered most to the research question. More on methods

Records from Elicit search

n = 200

Papers screened using: Population, Intervention, Outcomes, Study Design, Follow-up Duration, Treatment Isolation, Diabetes Type, Publication Type

n = 200

Papers screened out

n = 190

Papers included for extraction

n = 10

Paper search

We performed a semantic search across over 138 million academic papers from the Elicit search engine, which includes all of Semantic Scholar and OpenAlex.

We ran this query: "Evidence for dulaglutide reducing major adverse cardiovascular events in type 2 diabetes (MACE outcomes)"

The search returned 200 total results from Elicit.

We retrieved 200 papers most relevant to the query for screening.

Screening

We screened in sources based on their abstracts that met these criteria:

• Population: Does the study include adults (≥18 years) with type 2 diabetes mellitus?
• Intervention: Does the study include dulaglutide as a primary treatment or comparison arm?
• Outcomes: Does the study report major adverse cardiovascular events (MACE) as primary or secondary outcomes?
• Study Design: Is the study a randomized controlled trial (RCT) or a systematic review/meta-analysis?
• Follow-up Duration: Does the study have a minimum follow-up period of 12 weeks or longer?
• Treatment Isolation: Can the effects of dulaglutide be isolated from other treatments in this study (i.e., are dulaglutide effects distinguishable from combination therapies)?
• Diabetes Type: Does the study avoid focusing exclusively on type 1 diabetes or gestational diabetes?
• Publication Type: Is the study NOT a case report, case series, editorial, conference abstract, animal study, or in vitro study?

We considered all screening questions together and made a holistic judgement about whether to screen in each paper.

Data extraction

We asked a large language model to extract each data column below from each paper. We gave the model the extraction instructions shown below for each column.

• Study Design:

Extract study design and key methodological details for studies evaluating dulaglutide’s effect on MACE in type 2 diabetes, including:

• Study type (RCT, observational, post-hoc analysis, etc.)

• Blinding status

• Follow-up duration

• Primary vs secondary/exploratory analysis designation

• Sample size and statistical power for MACE outcomes

• T2DM Population:

Extract characteristics of type 2 diabetes participants relevant to cardiovascular risk and MACE outcomes, including:

• Sample size

• Mean age and age range

• Diabetes duration

• Baseline HbA1c

• Cardiovascular disease status (established CVD vs risk factors only)

• Key comorbidities (kidney disease, hypertension, etc.)

• Prior cardiovascular events

• Dulaglutide Intervention:

Extract details about dulaglutide treatment regimen for MACE prevention in type 2 diabetes, including:

• Dose (mg)

• Frequency of administration

• Duration of treatment

• Route of administration

• Adherence/compliance rates

• Concomitant diabetes medications allowed

• Comparator Group:

Extract details about the control/comparison group for evaluating dulaglutide’s MACE effects, including:

• Type of comparator (placebo, active medication, standard care)

• If active comparator: specific medication name and dose

• Duration of comparison treatment

• Background therapy allowed in both groups

• Adherence rates in comparator group

• MACE Definition:

Extract the specific definition and components of major adverse cardiovascular events (MACE) used in studies of dulaglutide in type 2 diabetes, including:

• Primary MACE definition (3-point, 4-point, etc.)

• Specific components included (CV death, MI, stroke, hospitalization for unstable angina, etc.)

• How each component was defined/diagnosed

• Whether fatal and non-fatal events were distinguished

• Any expanded MACE definitions used

• MACE Outcomes:

Extract all major adverse cardiovascular event outcomes for dulaglutide vs comparator in type 2 diabetes patients, including:

• Primary MACE composite: event rates, hazard ratio with 95% CI, p-values

• Individual MACE components: CV death, MI, stroke with separate effect estimates

• Time-to-first-event vs total cardiovascular burden analyses

• Absolute risk reduction and number needed to treat if reported

• Statistical significance for superiority vs non-inferiority

• Subgroup Analyses:

Extract subgroup analyses of dulaglutide’s MACE effects in different type 2 diabetes populations, including:

• Subgroups analyzed (by age, sex, diabetes duration, baseline CVD, kidney function, etc.)

• MACE outcomes within each subgroup (hazard ratios with CIs)

• Interaction p-values between treatment and subgroup

• Whether effects were consistent across subgroups

• Any subgroups with enhanced or diminished benefit

• Mechanism Analysis:

Extract evidence about potential mechanisms explaining dulaglutide’s MACE reduction in type 2 diabetes, including:

• Changes in cardiovascular risk factors (HbA1c, weight, BP, lipids, albuminuria)
• Mediation analyses showing which factors explain MACE benefits
• Degree of mediation (% reduction in hazard ratio when accounting for mediator)
• Direct vs indirect effects on cardiovascular outcomes
• Proposed biological mechanisms beyond traditional risk factors

Results

Characteristics of Included Studies

The evidence base comprised 10 sources examining dulaglutide’s effects on MACE outcomes in type 2 diabetes. Of these, 9 sources reported on the REWIND trial (Researching cardiovascular Events with a Weekly INcretin in Diabetes), while one source examined the SURPASS-CVOT trial comparing tirzepatide to dulaglutide. Full text was available for 3 of the 10 sources.

The REWIND trial enrolled 9,901 participants with type 2 diabetes, mean age 66 years, of whom 46% were women. Participants had mean diabetes duration of 10 years and baseline HbA1c of 7.3%. Notably, only 31% had established cardiovascular disease, while the remainder had cardiovascular risk factors only. Key comorbidities included hypertension in 93% and prior heart failure in 9%. The trial used a double-blind design with dulaglutide 1.5 mg administered subcutaneously once weekly compared to matched placebo, both in addition to standard care. Adherence was 82% for dulaglutide, with 97% study retention.

The SURPASS-CVOT trial enrolled 13,299 patients with mean age 64.1 years and 29% women. These participants had longer diabetes duration (mean 14.7 years), higher baseline HbA1c (8.4%), and all had established atherosclerotic cardiovascular disease, including coronary artery disease in 65%, history of stroke in 19.2%, peripheral artery disease in 25.3%, and heart failure in 20.3%. The trial compared tirzepatide (up to 15 mg weekly) to dulaglutide (1.5 mg weekly), both administered subcutaneously.

MACE Definition and Components

The primary MACE definition across studies was predominantly a 3-point composite. In REWIND, this comprised cardiovascular death, nonfatal myocardial infarction, and nonfatal stroke. SURPASS-CVOT used an identical 3-point definition. Fatal and nonfatal events were distinguished in the composite definitions.

Several post-hoc analyses examined expanded MACE definitions. One analysis added unstable angina, heart failure, and revascularization to the standard MACE components. Ferrannini et al. examined an atherosclerosis-related composite that included cardiovascular death, nonfatal myocardial infarction, nonfatal ischemic stroke, and any revascularization (coronary, peripheral, or carotid). SURPASS-CVOT also evaluated an expanded definition including death from cardiovascular causes, myocardial infarction, stroke, or coronary revascularization, as well as a composite of cardiovascular death or heart failure hospitalization.

Effects on Primary MACE Outcomes

In REWIND, dulaglutide reduced the primary 3-point MACE composite by 12% compared to placebo (HR 0.88, 95% CI 0.79-0.99, p=0.026). The absolute risk reduction was 4.1 events per 1000 person-years for first occurrence of MACE, translating to a number needed to treat of 67 (95% CI 38 to 803) over a median 5.4 years. Event rates were 2.35 per 100 person-years with dulaglutide versus 2.66 per 100 person-years with placebo.

In SURPASS-CVOT, tirzepatide demonstrated noninferiority to dulaglutide for the primary MACE composite (HR 0.92, 95.3% CI 0.83 to 1.01, p=0.003 for noninferiority), with event rates of 12.2% versus 13.1%. However, statistical superiority was not achieved (p=0.09).

Individual MACE Components

Dulaglutide demonstrated particularly robust effects on stroke outcomes in REWIND. Overall stroke risk was reduced by 24% (HR 0.76, 95% CI 0.62-0.94, p=0.010), driven primarily by a 25% reduction in ischemic stroke (HR 0.75, 95% CI 0.59-0.94, p=0.012), with no effect on hemorrhagic stroke (HR 1.05, 95% CI 0.55-1.99, p=0.89). Dulaglutide also reduced disabling stroke by 26% (HR 0.74, 95% CI 0.56-0.99, p=0.042) and the composite of nonfatal stroke or all-cause death by 12% (HR 0.88, 95% CI 0.79-0.98, p=0.017). The number needed to treat to prevent one nonfatal stroke was 121 (95% CI 75 to 582).

For myocardial infarction, dulaglutide showed no significant benefit in REWIND (HR 0.96, 95% CI 0.79 to 1.16). In SURPASS-CVOT, tirzepatide showed numerically lower MI rates than dulaglutide (HR 0.86, 95% CI 0.74 to 1.00), though statistical significance was not reported.

Total Cardiovascular Burden Analysis

Beyond time-to-first-event analyses, REWIND examined total cardiovascular event burden. A total of 1,972 MACE outcomes or deaths occurred, with an additional 1,701 expanded cardiovascular events (unstable angina, heart failure, revascularization). The incidence of total MACE or deaths was 35.8 per 1000 person-years with dulaglutide versus 40.3 per 1000 person-years with placebo, representing an absolute reduction of 4.5 events per 1000 person-years. Using conditional time gap modeling, dulaglutide reduced total MACE or deaths by 10% (HR 0.90, 95% CI 0.82-0.98, p=0.020). For the expanded definition including all cardiovascular events or deaths, dulaglutide reduced incidence from 74.7 to 67.1 per 1000 person-years (HR 0.93, 95% CI 0.87-0.99, p=0.023). Notably, similar effect sizes were observed for both first occurrence and total events (absolute risk reduction 4.1 vs 4.5 per 1000 person-years), suggesting consistent benefit across recurrent events.

Atherosclerosis-Related Outcomes

Ferrannini et al. examined whether dulaglutide impacts clinical manifestations of atherosclerosis through a composite endpoint including cardiovascular death, nonfatal myocardial infarction, nonfatal ischemic stroke, and any revascularization. This outcome occurred in 16.1% of dulaglutide-treated patients versus 17.6% of placebo patients, representing a 9% risk reduction (HR 0.91, 95% CI 0.83-1.00, p=0.05). Incidence rates were 3.25 versus 3.58 per 100 person-years. This finding was consistent across subgroups regardless of sex, body mass index, previous cardiovascular disease status, or diabetes duration.

Subgroup Analyses by Kidney Function

At baseline, 22.2% of REWIND participants had eGFR <60 mL/min/1.73m², 27% had microalbuminuria, and 8% had macroalbuminuria. Dulaglutide reduced MACE-3 consistently regardless of baseline kidney function, with hazard ratios of 0.93 in those with reduced eGFR and 0.86 in those with preserved eGFR (interaction p=0.545). Similarly, benefits were consistent between those with micro-/macroalbuminuria (HR 0.84) and normoalbuminuria (HR 0.93, interaction p=0.374). The lack of significant treatment-by-subgroup interactions indicates dulaglutide’s cardiovascular benefits extend across the spectrum of diabetic kidney disease.

Mechanisms of MACE Reduction

Mediation analysis from the REWIND trial revealed that improvements in HbA1c and albuminuria partially explain dulaglutide’s cardiovascular benefits, while changes in weight, blood pressure, lipids, and waist-to-hip ratio did not mediate the MACE reduction despite being significantly improved by treatment. Updated mean HbA1c during follow-up accounted for 36.1% of the hazard ratio reduction, while updated mean UACR accounted for 28.5%. When both factors were included simultaneously, they explained 65.4% of the MACE effect (interaction p=0.75, indicating independent contributions). Change from baseline in these parameters showed similar patterns, with HbA1c and UACR explaining 16.7% and 25.4% respectively, or 41.7% combined.

In SURPASS-CVOT, tirzepatide achieved greater improvements than dulaglutide in multiple risk factors: HbA1c decreased by 1.66 versus 0.88 percentage points, body weight decreased by 11.6% versus 4.8%, systolic blood pressure decreased by 6.2 versus 4.1 mm Hg, and triglycerides decreased by 24.2% versus 10.2%. Despite these more favorable risk factor changes, tirzepatide did not demonstrate superiority for MACE outcomes, suggesting mechanisms beyond glycemic control and weight reduction are important for cardiovascular benefit.

The finding that 35-65% of dulaglutide’s MACE benefit remains unexplained by traditional risk factors aligns with proposed direct cardiovascular mechanisms. Dulaglutide may improve endothelial function, endothelial cell responses to ischemia, and platelet function. Additionally, dulaglutide may retard atherosclerosis progression, as evidenced by the 9% reduction in the atherosclerosis-related composite outcome. GLP-1 receptors expressed in the heart have anti-atherosclerotic and anti-inflammatory effects that could contribute to cardiovascular protection.

Synthesis

The evidence demonstrates consistent cardiovascular benefit from dulaglutide across multiple analytic approaches and patient subgroups in type 2 diabetes. The primary REWIND trial, comprising 9,901 participants with a median 5.4-year follow-up, showed a 12% reduction in the primary 3-point MACE composite (HR 0.88). This finding was robust whether examining first events or total cardiovascular burden, and extended to broader atherosclerosis-related outcomes.

The stroke benefit appears particularly pronounced and clinically meaningful. The 24% reduction in overall stroke was driven specifically by ischemic events, with notable reductions in disabling stroke, translating to prevention of one stroke per 121 patients treated for 5.4 years. In contrast, myocardial infarction showed no significant benefit, creating an apparent discrepancy in vascular outcomes.

This divergence between stroke and MI effects can be explained by several mechanisms. First, the mediation analysis revealed that glycemic control and albuminuria reduction—both improved by dulaglutide—account for approximately two-thirds of the MACE benefit. These pathways may differentially affect cerebrovascular versus coronary disease progression. Second, ischemic stroke is particularly sensitive to blood pressure, which dulaglutide reduces, while MI prevention may depend more on lipid profiles, where dulaglutide shows minimal effects. Third, the unexplained one-third of cardiovascular benefit likely involves direct vascular effects—improvements in endothelial function and anti-inflammatory mechanisms—which may have greater impact in cerebrovascular beds.

The REWIND population’s lower baseline cardiovascular risk (only 31% with established CVD) contrasts sharply with SURPASS-CVOT (100% with established CVD), yet benefits were consistent across this spectrum. In REWIND subgroup analyses, effects were uniform regardless of baseline cardiovascular disease status or kidney function, with no significant treatment-by-subgroup interactions (p=0.545 for eGFR, p=0.374 for albuminuria). This suggests dulaglutide provides cardiovascular protection across the continuum from primary to secondary prevention in type 2 diabetes.

The SURPASS-CVOT comparison of tirzepatide to dulaglutide provides context for interpreting dulaglutide’s absolute cardiovascular benefit. Despite tirzepatide achieving substantially greater improvements in HbA1c (1.66 vs 0.88 percentage point reduction), weight loss (11.6% vs 4.8%), and blood pressure (6.2 vs 4.1 mm Hg reduction), it demonstrated only noninferiority (HR 0.92), not superiority, for MACE outcomes. This disconnect between metabolic benefits and cardiovascular outcomes reinforces that dulaglutide’s MACE reduction involves mechanisms beyond glucose lowering and weight loss alone. The mediation analysis showing that HbA1c and albuminuria explain only 65% of benefit supports direct cardiovascular effects as critical contributors.

Across all analyses examining dulaglutide versus placebo, hazard ratios for MACE ranged narrowly from 0.88 to 0.93, demonstrating remarkable consistency. The atherosclerosis-focused composite showed similar magnitude benefit (HR 0.91), as did total cardiovascular event burden (HR 0.90-0.93). This consistency across first events, recurrent events, and different composite definitions indicates a genuine underlying cardiovascular protective effect rather than statistical artifact or selective outcome reporting.