PopulationHARMONY Outcomes: N=9463; T2D with established CVD; double-blind placebo-controlled RCT; median 1.6 years (short).
Fundingindustry - GlaxoSmithKline
Scope limitsADDED. MI-driven superiority, contrasting the stroke-driven SUSTAIN-6/REWIND; recorded as observation. Albiglutide later withdrawn from market for commercial (not safety) reasons.
Comparatorsplacebo
GLP-1 RA classMARKETED
Cardiovascular outcomesDecrease
Moderate evidence
Class-level meta-analysis of 8 GLP-1RA CVOTs in T2D: ~14% MACE reduction, with consistent reductions in each MACE component, all-cause mortality, HF hospitalisation...
SourceSattar N, Lee MMY, Kristensen SL et al., Lancet Diabetes Endocrinol, 2021Source
Full findingClass-level meta-analysis of 8 GLP-1RA CVOTs in T2D: ~14% MACE reduction, with consistent reductions in each MACE component, all-cause mortality, HF hospitalisation and a kidney composite; no significant heterogeneity by structural homology or eight other subgroups.
PopulationSystematic review and random-effects meta-analysis of 8 placebo-controlled GLP-1RA CVOTs (ELIXA, LEADER, SUSTAIN-6, EXSCEL, HARMONY, REWIND, PIONEER-6, AMPLITUDE-O); PROSPERO CRD42021259711.
Fundingacademic/independent - no funding (Funding: None)
Scope limitsClass anchor: the ~14% MACE RRR headline. Graded MODERATE per the meta-analysis default rule even though constituents are pivotal CVOTs. STALE-ANCHOR caveat (Council): this is a 2021-vintage pool that pre-dates SELECT (2023), FLOW (2024) and SOUL (2025), so it excludes the non-diabetic-obesity trial most relevant to retatrutide; the 0.86 should NOT be cited as the current class figure. LUMP-NOT-SPLIT caveat: 'no significant heterogeneity by structural homology' is a power statement over sparse-event trials, NOT proof the two null constituents (ELIXA, EXSCEL) are noise; pooling is a value choice. TRANSFER caveat: homology homogeneity is WITHIN the GLP-1RA class and says nothing about molecules adding GIP or glucagon agonism; do not use the 14% RRR as a retatrutide expectation. Per-study COI not individually audited.
Comparatorsplacebo (pooled across constituent trials)
DulaglutideMARKETED
Cardiovascular outcomesDecrease
High evidence
In T2D with a majority WITHOUT established CVD (primary-prevention-weighted), dulaglutide reduced MACE versus placebo, driven notably by reduced non-fatal stroke.
SourceGerstein HC et al. (REWIND), Lancet, 2019Source
PopulationREWIND: N=9901; T2D, broad CV-risk (majority primary-prevention); double-blind placebo-controlled RCT; median 5.4 years (longest GLP-1 CVOT).
Fundingindustry - Eli Lilly
Scope limitsidentifier not fully verified
Comparatorsplacebo
DulaglutideMARKETED
Cardiovascular outcomesDecrease
Moderate evidence
In a dedicated REWIND analysis, dulaglutide reduced total stroke, driven by ischaemic stroke, with no effect on haemorrhagic stroke and no change in post-stroke...
SourceGerstein HC, Hart R, Colhoun HM, et al. The effect of dulaglutide on stroke: an explorato...Source
Full findingIn a dedicated REWIND analysis, dulaglutide reduced total stroke, driven by ischaemic stroke, with no effect on haemorrhagic stroke and no change in post-stroke disability severity.
PopulationREWIND stroke analysis: N=9901 type 2 diabetes, broad CV risk; double-blind placebo-controlled randomised trial; median follow-up 5.4 years; NCT01394952.
Fundingindustry - Eli Lilly (trial sponsor; inferred from registration trial)
Scope limitssurrogate/exploratory endpoint
Comparatorsplacebo
EfpeglenatideMARKETED
Cardiovascular outcomesDecrease
Moderate evidence
In T2D with CVD or kidney disease plus another risk factor, efpeglenatide reduced MACE versus placebo, with a dose-dependent trend, and reduced a kidney composite. An...
SourceGerstein HC et al., N Engl J Med, 2021Source
Full findingIn T2D with CVD or kidney disease plus another risk factor, efpeglenatide reduced MACE versus placebo, with a dose-dependent trend, and reduced a kidney composite. An exendin-4-based agent showing hard CV benefit.
PopulationAMPLITUDE-O: N=4076; T2D with CVD or kidney disease; double-blind placebo-controlled RCT; median 1.8 years.
Scope limitsADDED. Key because efpeglenatide is exendin-4-BASED yet showed clear CV benefit (HR 0.73), countering the idea that exendin-based agents are inert on MACE; it is the trial folded into the Sattar 2021 class meta-analysis. High background SGLT2i use enabled interaction analyses.
Comparatorsplacebo
ExenatideMARKETED
Cardiovascular outcomesNo change
High evidence
In T2D (most with prior CVD), once-weekly exenatide was non-inferior to placebo for MACE; superiority was not met (HR favoured exenatide, CI crossed 1).
Scope limitsADDED (not in prior rows). Largest GLP-1 CVOT; non-inferior with a near-miss on superiority (P=0.06). Sits between the neutral (ELIXA) and superior (LEADER/SUSTAIN-6/REWIND) trials. Exendin-4-based, yet the class meta found no heterogeneity by structural homology.
Comparatorsplacebo
SurvodutideINVESTIGATIONAL
Cardiovascular outcomesNot directional
Moderate evidence
ABSENCE OF EVIDENCE: as of the 2026-06-21 (non-exhaustive) sweep, no powered hard-endpoint CV outcome trial has REPORTED for any GLP-1/glucagon dual agonist; the class...
SourceAbsence established via PubMed sweep (2026-06-21): no dual-agonist CV-outcome-trial recor...Source
Full findingABSENCE OF EVIDENCE: as of the 2026-06-21 (non-exhaustive) sweep, no powered hard-endpoint CV outcome trial has REPORTED for any GLP-1/glucagon dual agonist; the class sits in the phase-2/early-phase-3 era with weight, glycaemic, hepatic (MASH) and CV-risk-factor readouts only. This absence is NOT reassurance: the glucagon arm's CV-risk hypothesis has never been tested at hard-endpoint level in any agent, so the net effect is UNMEASURED, not measured-and-null. It is an evidence GAP, not a demonstration of CV neutrality or harm, and does not exclude ongoing/unreported trials.
PopulationClass-level absence statement (cotadutide, survodutide, mazdutide, pemvidutide, efinopegdutide); efficacy/surrogate data only; no dedicated CVOT reported.
Fundingacademic / non-commercial (De Fano GIP review; no industry sponsor; disclosed)
Scope limitsanimal data; human relevance uncertain; surrogate/exploratory endpoint
LiraglutideMARKETED
Cardiovascular outcomesDecrease
High evidence
In T2D at high CV risk, liraglutide reduced MACE versus placebo; unlike SUSTAIN-6, CV death and all-cause death were also significantly reduced.
In T2D after a recent acute coronary syndrome, lixisenatide was non-inferior but NOT superior to placebo for the CV composite: a fully NEUTRAL CV outcome.
PopulationELIXA: N=6068; T2D with ACS within 180 days; double-blind placebo-controlled RCT; median 25 months.
Fundingindustry - Sanofi (disclosed in publication)
Scope limitsidentifier not fully verified
Comparatorsplacebo
RetatrutideINVESTIGATIONAL
Cardiovascular outcomesNot directional
Very low evidence
The dedicated retatrutide CV outcomes trial (TRIUMPH-Outcomes) is ONGOING; no hard-outcome MACE or CV-death data have been reported. A registry promissory note, NOT...
Full findingThe dedicated retatrutide CV outcomes trial (TRIUMPH-Outcomes) is ONGOING; no hard-outcome MACE or CV-death data have been reported. A registry promissory note, NOT evidence: retatrutide currently has ZERO hard cardiovascular outcome data.
PopulationTRIUMPH-Outcomes (NCT06383390): adults age >=45, BMI >=27, with established ASCVD and/or CKD (HbA1c <=10% if T2D); retatrutide s.c. once weekly (escalated dose) vs placebo; double-blind, event-driven; ~5-year study.
Fundingindustry - Eli Lilly (trial sponsor; inferred from registration trial)
Scope limitsno outcome data yet (ongoing)
Comparatorsplacebo
RetatrutideINVESTIGATIONAL
Cardiovascular outcomesNot directional
Moderate evidence
OPEN QUESTION: no outcome trial isolates a glucagon (GCGR) arm, so the net CV-OUTCOME consequence of glucagon agonism is unresolved. The concern, whether...
SourceAbsence established via PubMed sweep (2026-06-21): no dual-agonist CV-outcome-trial recor...Source
Full findingOPEN QUESTION: no outcome trial isolates a glucagon (GCGR) arm, so the net CV-OUTCOME consequence of glucagon agonism is unresolved. The concern, whether glucagon-driven sustained heart-rate elevation or increased cardiac workload could ERODE the CV benefit established for GLP-1 (with the GIP-containing dual agonist shown CV-safe at whole-molecule level via SURPASS-CVOT, GIP not isolated), is recorded as a tension to be settled by retatrutide's outcome programme, not as a quantified finding.
PopulationNo isolating trial exists. The only dedicated triple-agonist CVOT is TRIUMPH-Outcomes (NCT06383390), hard endpoints ~2029.
Fundingacademic / non-commercial (De Fano GIP review; no industry sponsor; disclosed)
Scope limitsanimal data; human relevance uncertain; surrogate/exploratory endpoint
RetatrutideINVESTIGATIONAL
Cardiovascular outcomesNot directional
Moderate evidence
The retatrutide phase-2 trial produced NO hard CV-outcome data; MACE and CV death were not endpoints. Its heart-rate rise is a CV-mechanism signal held in the...
SourceJastreboff AM et al. (retatrutide phase 2), NEJM, 2023;389:514-526Source
Full findingThe retatrutide phase-2 trial produced NO hard CV-outcome data; MACE and CV death were not endpoints. Its heart-rate rise is a CV-mechanism signal held in the heart-rate / CV-mechanism thread, deliberately not counted in the cv-outcomes domain to avoid double-counting.
PopulationJastreboff AM et al. phase-2 (NCT04881760): N=338 adults with obesity (entry BMI-based: >=30, or 27-30 with a weight-related condition); retatrutide vs placebo, 48 weeks; double-blind RCT.
CLARIFYING NOTE: TRIUMPH-3 (NCT05882045) enrols adults with severe obesity (BMI >=35) AND established CVD, but its primary endpoint is percent change in body weight,...
SourceEli Lilly. TRIUMPH-3: Phase 3 study of LY3437943 once weekly vs placebo in severe obesity...Source
Full findingCLARIFYING NOTE: TRIUMPH-3 (NCT05882045) enrols adults with severe obesity (BMI >=35) AND established CVD, but its primary endpoint is percent change in body weight, not MACE. It is a weight-loss trial in a CVD population, NOT a CV outcomes trial; the only dedicated reta CV outcomes (MACE) trial is TRIUMPH-Outcomes (NCT06383390).
PopulationTRIUMPH-3: N=1,946; severe obesity (BMI >=35) with established CVD (prior MI, stroke, or symptomatic PAD); retatrutide s.c. once weekly (two doses) vs placebo; Phase 3 double-blind.
Fundingindustry - Eli Lilly (trial sponsor; inferred from registration trial)
Scope limitsADDED to disambiguate TRIUMPH-3 (weight-loss primary in a CVD population) from the true CVOT NCT06383390. No other registered reta trial with a hard CV/MACE primary endpoint was found. VALIDATOR: confirm NCT05882045 primary endpoint is body weight, not MACE.
Comparatorsplacebo
SemaglutideMARKETED
Cardiovascular outcomesDecrease
High evidence
In adults aged >=45 with pre-existing CVD and overweight/obesity (BMI >=27) but WITHOUT diabetes, once-weekly s.c. semaglutide 2.4 mg reduced 3-point MACE (CV death,...
SourceLincoff AM et al. (SELECT), NEJM, 2023Source
Full findingIn adults aged >=45 with pre-existing CVD and overweight/obesity (BMI >=27) but WITHOUT diabetes, once-weekly s.c. semaglutide 2.4 mg reduced 3-point MACE (CV death, non-fatal MI, non-fatal stroke) versus placebo.
PopulationSELECT: N=17,604; established CVD + overweight/obesity, no diabetes; multicentre double-blind placebo-controlled event-driven superiority RCT; mean follow-up 39.8 months.
Fundingindustry-Novo Nordisk
Scope limitsDEEPENS the prior one-line SELECT row with the HR and event rates. First GLP-1RA CVOT in a non-diabetic obesity population. Weight-independence framing carried in companion analysis (C-SEMA-CARDIO-05). Cross-ref the T2D sema CVOTs.
Comparatorsplacebo
SemaglutideMARKETED
Cardiovascular outcomesDecrease
High evidence
In T2D at high CV risk, s.c. semaglutide reduced first MACE versus placebo, meeting non-inferiority then superiority; reduction driven mainly by non-fatal stroke and...
SourceMarso SP et al. (SUSTAIN-6), NEJM, 2016Source
Full findingIn T2D at high CV risk, s.c. semaglutide reduced first MACE versus placebo, meeting non-inferiority then superiority; reduction driven mainly by non-fatal stroke and MI, with CV death unchanged.
PopulationSUSTAIN-6: N=3297; T2D high CV risk; double-blind placebo-controlled pre-approval CV-safety RCT; median 2.1 years.
In T2D at high CV risk, once-daily oral semaglutide was non-inferior to placebo for MACE but did NOT demonstrate superiority (CV-safety trial, underpowered for...
SourceHusain M et al. (PIONEER 6), NEJM, 2019Source
Full findingIn T2D at high CV risk, once-daily oral semaglutide was non-inferior to placebo for MACE but did NOT demonstrate superiority (CV-safety trial, underpowered for superiority).
PopulationPIONEER 6: N=3183; T2D high CV risk; double-blind placebo-controlled CV-safety RCT; median 15.9 months.
Fundingindustry-Novo Nordisk
Scope limitsBOTH-POLES: a non-inferiority (NOT superiority) oral-sema result; the design was CV-safety. Definitive oral-sema superiority came later in SOUL (C-SEMA-CARDIO-04). DEEPENS prior row with HR and CI-crossing-1 detail.
Comparatorsplacebo
SemaglutideMARKETED
Cardiovascular outcomesDecrease
High evidence
In T2D with established ASCVD, CKD, or both, oral semaglutide 14 mg reduced MACE versus placebo in a dedicated superiority CVOT; the kidney composite did not differ.
SourceMcGuire DK, Marx N, Buse JB et al., N Engl J Med, 2025Source
PopulationSOUL: N=9650; T2D age >=50, HbA1c 6.5-10.0%, with ASCVD and/or CKD; double-blind placebo-controlled event-driven superiority RCT; mean follow-up 47.5 months.
Fundingindustry - Novo Nordisk
Scope limitskidney confirmatory-secondary not met
Comparatorsplacebo
SemaglutideMARKETED
Cardiovascular outcomesDecrease
Moderate evidence
Prespecified SELECT analysis: semaglutide produced sustained weight loss across every baseline BMI category to 4 years (208 weeks). The trial's headline 20% MACE...
SourceRyan DH, Lingvay I, Deanfield J et al., Nat Med, 2024Source
Full findingPrespecified SELECT analysis: semaglutide produced sustained weight loss across every baseline BMI category to 4 years (208 weeks). The trial's headline 20% MACE reduction is restated as context; this paper reports weight-by-BMI, not a per-subgroup MACE analysis.
PopulationSELECT prespecified weight/BMI secondary analysis (prespecified secondary of the N=17,604 double-blind RCT); up to 208 weeks.
Fundingindustry - Novo Nordisk (SELECT sponsor)
Scope limitssurrogate/exploratory endpoint
Comparatorsplacebo
SemaglutideMARKETED
Cardiovascular outcomesDecrease
Moderate evidence
CROSS-DOMAIN (kidney trial with CV components): in T2D with CKD, s.c. semaglutide 1.0 mg reduced the major-kidney-disease composite (which includes CV death) and...
Full findingCROSS-DOMAIN (kidney trial with CV components): in T2D with CKD, s.c. semaglutide 1.0 mg reduced the major-kidney-disease composite (which includes CV death) and separately reduced CV death, MACE and all-cause death.
PopulationFLOW: N=3533; T2D with CKD (eGFR 25-75, raised UACR); double-blind placebo-controlled RCT; median 3.4 years (stopped early at interim).
Fundingindustry - Novo Nordisk (FLOW sponsor)
Scope limitsstopped early for efficacy (may inflate estimate)
Comparatorsplacebo
SemaglutideMARKETED
Cardiovascular outcomesDecrease
Moderate evidence
In the prespecified SELECT mediation analysis, semaglutide's cardiovascular benefit was largely INDEPENDENT of baseline adiposity and of the amount of weight lost -...
SourceDeanfield J, Lincoff AM, Kahn SE et al. Semaglutide and cardiovascular outcomes by baseli...Source
Full findingIn the prespecified SELECT mediation analysis, semaglutide's cardiovascular benefit was largely INDEPENDENT of baseline adiposity and of the amount of weight lost - roughly two-thirds of the MACE reduction was not explained by waist-circumference change - indicating mechanisms of CV benefit beyond adiposity reduction. CRITICAL: this is the evidence the popular 'works even without weight loss' claim rests on, and it is SEMAGLUTIDE-SPECIFIC; there is NO equivalent tirzepatide mediation analysis, so the weight-independent CV claim must NOT be transferred to tirzepatide (the SURMOUNT-MMO obesity CV trial is ongoing).
PopulationPrespecified analysis of SELECT (NCT03574597): n=17,604 adults >=45 y, BMI >=27, with established CVD and WITHOUT diabetes; semaglutide 2.4 mg weekly vs placebo; 41 countries; primary outcome time-to-first MACE (CV death, non-fatal MI, non-fatal stroke). Mediation via time-varying adjustment for waist/weight change + week-20 and week-104 landmarks.
Fundingindustry - Novo Nordisk
Scope limitsno outcome data yet (ongoing); possible confounding by indication
Comparatorsplacebo
SemaglutideMARKETED
Cardiovascular outcomesNo change
High evidence
In SELECT, semaglutide's cardiovascular-death component did not independently reach significance, despite a positive composite MACE and a significant...
SourceLincoff AM et al. (SELECT), NEJM, 2023Source
Full findingIn SELECT, semaglutide's cardiovascular-death component did not independently reach significance, despite a positive composite MACE and a significant all-cause-mortality reduction - a component null, NOT evidence that semaglutide fails to reduce CV death.
PopulationPeer-reviewed RCT (measured null)
Fundingindustry-Novo Nordisk
Scope limitsBOTH-POLED: the point estimate (0.85) is protective-leaning within a positive trial. The CV-death HR is from the SELECT supplementary appendix, not the primary abstract.
Comparatorsplacebo
SemaglutideMARKETED
Cardiovascular outcomesNo change
High evidence
In PIONEER-6, oral semaglutide showed no significant effect on the individual non-fatal MI or stroke components; the MACE 'non-superiority' is a design feature of a...
SourceHusain M et al. (PIONEER 6), NEJM, 2019Source
Full findingIn PIONEER-6, oral semaglutide showed no significant effect on the individual non-fatal MI or stroke components; the MACE 'non-superiority' is a design feature of a non-inferiority safety trial, not a measured efficacy null.
PopulationPeer-reviewed RCT (measured null)
Fundingindustry-Novo Nordisk
Scope limitsMEASURED component nulls (wide CIs, both-poled). The MACE-superiority is NEVER-MEASURED-AT-POWER and is already recorded in C-SEMA-CARDIO-03 - this row adds only the component detail.
Comparatorsplacebo
TirzepatideMARKETED
Cardiovascular outcomesNo change
High evidence
In T2D with established ASCVD, tirzepatide was NON-INFERIOR to dulaglutide for 3-point MACE but did NOT meet superiority. The comparator is a CV-proven GLP-1RA, so...
SourceNicholls SJ et al. Cardiovascular Outcomes with Tirzepatide versus Dulaglutide in Type 2 ...Source
Full findingIn T2D with established ASCVD, tirzepatide was NON-INFERIOR to dulaglutide for 3-point MACE but did NOT meet superiority. The comparator is a CV-proven GLP-1RA, so this establishes that the GIP-containing dual-agonist MOLECULE is CV-safe at the whole-molecule level; it does NOT isolate the GIP receptor's own contribution, because tirzepatide's GLP-1 moiety differs from dulaglutide's (cross-ref thread 11), so the null delta cannot be read as 'GIP adds nothing' rather than as offsetting differences between the two GLP-1 components. Despite materially greater HbA1c and weight reduction, the second incretin arm bought no incremental hard MACE.
PopulationSURPASS-CVOT: N=13,299 randomised (NCT04255433); T2D + established ASCVD, age >=40, HbA1c 7.0-10.5%, BMI >=25; double-blind active-comparator; tirzepatide up to 15 mg vs dulaglutide 1.5 mg; median follow-up ~4.0 years.
Fundingindustry - Eli Lilly (trial sponsor)
Scope limitsactive-comparator (dulaglutide) non-inferiority design - not placebo-controlled; superiority not met (P=0.09)
Comparatorsdulaglutide
TirzepatideMARKETED
Cardiovascular outcomesNo change
Moderate evidence
Pre-specified pre-registration pooled CV meta-analysis of the SURPASS phase-3 programme found tirzepatide did NOT increase MACE versus controls; the point estimate was...
SourceSattar N, McGuire DK, Pavo I et al., Nat Med, 2022Source
Full findingPre-specified pre-registration pooled CV meta-analysis of the SURPASS phase-3 programme found tirzepatide did NOT increase MACE versus controls; the point estimate was favourable but the CI crossed 1 (exploratory CV-safety evidence, not powered superiority).
PopulationPre-specified pooled meta-analysis of 7 RCTs (>=26 weeks) in the SURPASS T2D programme; 4,887 tirzepatide vs 2,328 control; controls = placebo and active comparators.
Dedicated morbidity-mortality OUTCOME trial of tirzepatide in OBESITY WITHOUT diabetes (primary and secondary prevention) is ONGOING with no results; the first...
SourceLam CSP, Rodriguez A, Aminian A et al. SURMOUNT-MMO rationale and design. Obesity (Silver...Source
Full findingDedicated morbidity-mortality OUTCOME trial of tirzepatide in OBESITY WITHOUT diabetes (primary and secondary prevention) is ONGOING with no results; the first incretin outcome trial to span both primary and secondary CV prevention.
PopulationSURMOUNT-MMO (NCT05556512): randomised double-blind event-driven; BMI >=27, obesity, T1D/T2D excluded; adults >=40 with established CVD OR adults with CV risk factors; once-weekly tirzepatide vs placebo; confirmed on clinicaltrials.gov 2026-06-21.
Fundingindustry - Eli Lilly (trial sponsor; inferred from registration trial)
Scope limitsADDED. PMID 40545827 is the published rationale-and-design paper, NOT a results paper, hence very-low (promissory note). Will give the first incretin CV-outcome read in obesity-without-diabetes spanning both prevention settings.
The Echo Compendium