GLP-1 Heart Benefits Are Real. The RAS Mechanism Isn’t.

GLP-1 receptor agonists appear to lower cardiovascular risk in people with type 2 diabetes and coronary artery disease. The proposed explanation involving the renin-angiotensin system, however, remains a computational hypothesis—not a demonstrated mechanism in patients.

That distinction matters. Drugs such as semaglutide and dulaglutide do not need a fully mapped molecular story to provide clinical benefit. Randomized outcome trials already support their use in appropriately selected high-risk patients. But claiming that GLP-1 drugs protect the heart by acting through a specific renin-angiotensin pathway gets ahead of the evidence.

A clinical benefit with an unfinished explanation

A 2023 meta-analysis by Kan et al. pooled 13 randomized controlled trials involving 35,563 people with type 2 diabetes complicated by coronary artery disease. The trials compared once-weekly dulaglutide, extended-release exenatide, or semaglutide with placebo.

The review reported a statistically significant reduction in nonfatal stroke, with p<0.001. It also found statistically significant improvements in HbA1c, fasting blood glucose, body weight, systolic blood pressure, total cholesterol, and LDL cholesterol; each result had p<0.001.

Those metabolic changes are clinically relevant because each is tied to cardiovascular risk. They also make the mechanism harder to isolate. If a treatment improves glucose, weight, blood pressure, and lipids at the same time, a pooled analysis cannot tell us how much of its cardiovascular benefit comes from each change—or whether a separate direct cardiac effect accounts for part of the result.

Kan et al. reported p-values for the cardiovascular findings but did not provide hazard ratios or 95% confidence intervals for major adverse cardiovascular events, cardiovascular death, or nonfatal myocardial infarction. A p-value can indicate that an observed difference is unlikely under a null hypothesis; it does not show the size of that difference. Without effect estimates and confidence intervals, readers cannot directly compare the clinical magnitude of the pooled result with the landmark cardiovascular-outcome trials.

The broader trial record supports a class effect, with caveats

The clinical signal does not depend on one review. A 2025 network meta-analysis by Rai et al., drawing on cardiovascular-outcome trials including LEADER, SUSTAIN-6, REWIND, and EXSCEL, estimated that injectable GLP-1 receptor agonists reduced major adverse cardiovascular events by about 12% to 14% versus placebo.

That is a relative effect. The absolute benefit for an individual depends on baseline risk: someone with established coronary disease has more potential events to prevent than someone at lower risk. The supplied analysis does not provide an absolute risk reduction or number needed to treat, so neither should be invented.

Rai and colleagues also found meaningful differences in cardiovascular efficacy among agents in the class. That heterogeneity complicates a single, uniform mechanism. Dulaglutide, extended-release exenatide, and semaglutide all activate the GLP-1 receptor, but pooled evidence cannot prove that they generate equal cardiovascular benefit through the same downstream biology.

The result from SELECT strengthens the case that glucose lowering is not the whole explanation. In the 2023 SELECT randomized trial, semaglutide reduced major adverse cardiovascular events versus placebo in 17,604 participants with obesity but without diabetes. Because the population did not have diabetes, improved glycemic control cannot explain the entire cardiovascular signal. SELECT does not, however, identify renin, angiotensin, MMP2, or any other single pathway as the cause.

Why the renin-angiotensin story is still a hypothesis

Kan et al. added a network-pharmacology analysis to their clinical meta-analysis. It identified matrix metalloproteinase-2, or MMP2, as a putative hub target linking GLP-1 receptor agonists to renin-angiotensin-system modulation.

This is useful for generating research questions. It is not proof that GLP-1 treatment changes this pathway in a way that prevents heart attacks or strokes. Network pharmacology draws connections from drug-target and protein-interaction databases. Those databases can suggest plausible relationships, but they cannot establish causation, measure the pathway's contribution to an outcome, or distinguish a meaningful target from a well-annotated bystander.

The key missing evidence is direct validation in the relevant setting. The available record does not include an animal experiment, cell study, or clinical study in people with type 2 diabetes and coronary artery disease showing that GLP-1 treatment changes plasma renin activity, angiotensin II, ACE expression, aldosterone, or an MMP2-RAS pathway—and then links that change to cardiovascular outcomes.

A 2022 prospective trial by Vukajlovic et al. did measure vasoactive hormones after three weeks of GLP-1 receptor agonist treatment. The investigators observed measurable hormonal changes and proposed vasoactive-hormone modulation as one possible route to lower blood pressure. But the participants were euvolemic and did not have diabetes, and the study lasted three weeks. That design can identify a short-term biological signal; it cannot validate a mechanism for multi-year cardiovascular outcomes in people with diabetes and coronary artery disease.

Metabolic improvement explains some benefit, but not necessarily all of it

A 2025 post-hoc mediation analysis by Peng et al. examined LEADER and SUSTAIN-6. The analysis found that the proportion of three-point MACE benefit statistically attributable to metabolic changes varied by follow-up time and patient subgroup.

That result argues against a simple story in either direction. Weight loss, lower blood pressure, and improved glycemic control likely account for some cardiovascular benefit, and their contribution may differ between patients and over time. The remaining signal is compatible with direct effects, but mediation analysis cannot name the responsible pathway. It does not validate the MMP2-RAS proposal.

Rai et al. describe anti-inflammatory, anti-atherosclerotic, and direct vascular effects as candidate mechanisms with emerging experimental support. These remain active areas of research, not settled explanations. The most accurate account is that GLP-1 cardiovascular protection probably has several contributors whose relative importance differs by drug, patient, and time on treatment.

Benefits belong in a safety conversation

Cardiovascular efficacy does not erase adverse effects. A 2025 systematic review by Sillassen et al. identified gastrointestinal adverse events as the dominant safety concern with semaglutide in patients at increased cardiovascular risk. Nausea, vomiting, and other gastrointestinal effects can determine whether a patient can continue treatment.

Metformin deserves practical attention because it is common in type 2 diabetes. A 2024 post-hoc analysis by Klein et al. assessed gastrointestinal adverse events during GLP-1 receptor agonist initiation and titration among participants with and without concomitant metformin across four trials. The abstract reports that metformin use did not increase the percentage of participants who developed gastrointestinal adverse events. That does not make initiation symptom-free; it means the available analysis does not support a simple claim that metformin automatically amplifies the rate of those events.

Eye safety also warrants attention in diabetes care. A 2025 systematic review and meta-analysis by Natividade et al. evaluated ocular adverse events with semaglutide, including diabetic retinopathy and possible nonarteritic anterior ischemic optic neuropathy. The review describes an unresolved signal with apparently low absolute risk. Patients with existing diabetic retinopathy should discuss monitoring with the clinician managing their diabetes and eye care rather than treating cardiovascular benefit as the only relevant outcome.

The useful conclusion

For people with type 2 diabetes and high cardiovascular risk, weekly injectable GLP-1 receptor agonists have credible randomized-trial evidence for reducing major cardiovascular events. The 13-trial review supports that direction of benefit, while broader cardiovascular-outcome evidence places the class-level relative MACE reduction around 12% to 14% versus placebo.

The renin-angiotensin explanation does not yet meet the same standard. MMP2 and RAS modulation are reasonable targets for experiments, but no direct evidence in the relevant cardiac population establishes them as the main reason these drugs reduce events. The evidence supports the treatment effect. The molecular headline still needs a laboratory receipt.