LEADER Results in Detail: Numbers, Subgroups, and Time Course

At a glance

Why the Primary Endpoint Matters Beyond the Headline

Before LEADER published in June 2016, no GLP-1 receptor agonist had demonstrated cardiovascular superiority in a dedicated outcomes trial. The FDA's 2008 guidance for diabetes drugs required proof of cardiovascular safety (ruling out a hazard ratio upper bound above 1.3), but LEADER went further by showing actual benefit. This distinction shifted clinical practice: liraglutide became the first GLP-1 RA to carry an indication for reducing cardiovascular events in its FDA prescribing label.

Primary Endpoint: Dissecting the 13% MACE Reduction

The three-point MACE composite occurred in 608 of 4,668 patients (13.0%) in the liraglutide group and 694 of 4,672 patients (14.9%) in the placebo group. The event rates per 100 patient-years were 3.4 and 3.9, respectively.

The primary publication shows that the MACE benefit was driven most strongly by the cardiovascular death component (HR 0.78, 95% CI 0.66 to 0.93), which reached statistical significance on its own. Nonfatal MI and nonfatal stroke trended favorably but did not individually cross the significance threshold. This pattern is unusual among cardiovascular outcomes trials, where reductions in nonfatal events (especially MI) more commonly drive the composite.

Time-Course Pattern: When Did the Curves Separate?

The Kaplan-Meier curves for the primary endpoint did not diverge immediately. Visual inspection of the published survival curves reveals three distinct phases:

Phase 1 (0 to 12 months): The curves overlapped almost entirely. Event rates were nearly identical between groups during this early period. This lag is consistent with the hypothesis that liraglutide's cardiovascular benefit stems from chronic metabolic improvements (weight loss, glycemic control, blood pressure reduction, lipid changes) rather than an acute anti-ischemic or antithrombotic mechanism.

Phase 2 (12 to 30 months): Gradual separation began. The liraglutide curve started to pull below placebo, with the gap widening progressively. By 24 months, a consistent and visible difference had emerged.

Phase 3 (30 to 60 months): The separation stabilized and persisted through end of follow-up. The benefit did not appear to accelerate or attenuate over time during this later period, suggesting a sustained rather than time-limited effect.

This time-course pattern has practical implications for treatment decisions. Clinicians should not expect rapid cardiovascular risk reduction from liraglutide in the first year. The 2019 ADA/EASD consensus report incorporated this evidence, recommending GLP-1 RAs with proven benefit for patients with established atherosclerotic cardiovascular disease regardless of baseline HbA1c.

Secondary and Exploratory Endpoints

Beyond the primary composite, LEADER pre-specified several secondary endpoints tested in a hierarchical sequence.

All-cause mortality reached significance (HR 0.85, 95% CI 0.74 to 0.97), making LEADER one of the few diabetes drug trials to show a mortality signal. Heart failure hospitalization did not reach significance, a finding that contrasts with the SGLT2 inhibitor trials like EMPA-REG OUTCOME where heart failure reduction was a prominent benefit.

Metabolic Changes: Mediators or Markers?

The trial recorded several metabolic parameters that may partially explain the cardiovascular findings.

The HbA1c difference was modest and narrowed over time as investigators could intensify background therapy in both arms. Post hoc mediation analyses published subsequently suggested that HbA1c reduction explained only a fraction of the MACE benefit, pointing toward direct vascular or anti-inflammatory effects of GLP-1 receptor agonism. The heart rate increase of approximately 3 bpm was a consistent finding across GLP-1 RA trials and did not appear to translate into arrhythmic risk within the trial's follow-up window.

Subgroup Analyses: Who Benefited Most?

LEADER pre-specified subgroup analyses across 17 baseline characteristics. The primary publication's forest plot showed no statistically significant treatment-by-subgroup interactions, but several patterns merit attention.

By cardiovascular disease history: Patients with established cardiovascular disease at baseline (about 81% of participants) had a numerically larger benefit (HR 0.83) compared to those with cardiovascular risk factors only (HR 1.20). This contrast did not reach formal interaction significance (p = 0.10), but it influenced guideline recommendations to prioritize GLP-1 RAs for patients with established disease.

By baseline HbA1c: The MACE reduction appeared consistent across HbA1c categories (above and below 8.3%), suggesting the benefit was not simply a glucose-lowering effect.

By baseline BMI: The benefit did not vary significantly by BMI tertile, arguing against weight loss as the sole mediator.

By age: Patients aged 60 years and older showed consistent benefit, an important finding given that this demographic carries the highest absolute cardiovascular risk.

By renal function: Patients with eGFR <60 mL/min/1.73 m² had a numerically larger MACE reduction. This finding was later supported by the dedicated LEADER renal sub-study, which demonstrated that liraglutide slowed nephropathy progression (HR 0.78, 95% CI 0.67 to 0.92).

Limitations the Authors Acknowledged

The LEADER investigators and subsequent commentators identified several constraints worth noting.

Enriched population. Over 80% of participants had established cardiovascular disease. Extrapolating the 13% MACE reduction to lower-risk type 2 diabetes populations requires caution, and the point estimate for the risk-factor-only subgroup did not favor liraglutide.

Concomitant therapy changes. Investigators could adjust background glucose-lowering medications throughout the trial. Fewer liraglutide-arm patients required insulin initiation or addition of other agents, which complicates attribution of the cardiovascular benefit to liraglutide alone.

Discontinuation rates. Approximately 25% of patients in the liraglutide arm and 24% in the placebo arm permanently discontinued the study drug. In the liraglutide group, gastrointestinal adverse events (nausea, vomiting, diarrhea) were the most common reason for early discontinuation.

No active comparator. The trial compared liraglutide to placebo, not to another active glucose-lowering agent. Whether the cardiovascular benefit would persist against, say, an SGLT2 inhibitor has not been tested head-to-head.

Event adjudication. All suspected cardiovascular events were adjudicated by an independent committee blinded to treatment assignment. While this is standard practice, the modest absolute risk reduction (1.9 percentage points for MACE over 3.8 years) means that small changes in adjudication could shift results.

LEADER in Context: How It Compares to Later GLP-1 CVOTs

LEADER was the first positive GLP-1 RA cardiovascular outcomes trial, but it was not the last. The SUSTAIN-6 trial (semaglutide, 2016) showed a 26% MACE reduction over just 2.1 years with a smaller sample of 3,297 patients, though that trial was powered only for noninferiority. The 2019 REWIND trial (dulaglutide) showed a 12% MACE reduction in a broader population with only 31% having established cardiovascular disease.

Taken together, these trials established a class effect for GLP-1 RAs on atherosclerotic cardiovascular events. LEADER's contribution was foundational: it provided the first proof of concept and remains the largest completed GLP-1 RA cardiovascular outcomes trial by enrollment.

Practical Takeaways for Clinicians

The LEADER data support initiating liraglutide (or another GLP-1 RA with proven benefit) in adults with type 2 diabetes and established atherosclerotic cardiovascular disease who need additional glucose lowering or cardiovascular risk reduction. The 2024 ADA Standards of Care recommend GLP-1 RAs with demonstrated cardiovascular benefit in this population independent of HbA1c level or metformin use.

Clinicians should set patient expectations: cardiovascular benefit builds over months, not weeks. The 12-to-18-month lag before curve separation means that short treatment courses are unlikely to capture the protective effect seen in the trial.

Frequently asked questions

›What was the exact hazard ratio for the LEADER trial's primary endpoint?

The primary three-point MACE endpoint had a hazard ratio of 0.87 with a 95% confidence interval of 0.78 to 0.97 (p = 0.01 for superiority). This translates to a 13% relative risk reduction favoring liraglutide 1.8 mg over placebo when added to standard of care.

›Did LEADER show a mortality benefit for liraglutide?

Yes. All-cause mortality was significantly lower in the liraglutide group (8.2% vs. 9.6%, HR 0.85, 95% CI 0.74 to 0.97). Cardiovascular death specifically was reduced by 22% (HR 0.78, 95% CI 0.66 to 0.93). This makes LEADER one of relatively few diabetes drug trials to demonstrate a statistically significant mortality reduction.

›How long did it take to see cardiovascular benefit in LEADER?

The Kaplan-Meier curves for MACE began separating at approximately 12 to 18 months. No meaningful difference was visible during the first year. The benefit grew progressively and was sustained through the maximum follow-up of 5 years.

›Did liraglutide reduce heart failure hospitalizations in LEADER?

The point estimate favored liraglutide (HR 0.87, 95% CI 0.73 to 1.05), but this did not reach statistical significance. GLP-1 RAs have not shown the same consistent heart failure benefit observed with SGLT2 inhibitors in trials like EMPA-REG OUTCOME and DAPA-HF.

›What dose of liraglutide was used in LEADER?

Participants were titrated to liraglutide 1.8 mg once daily (or the maximum tolerated dose). The starting dose was 0.6 mg, increased to 1.2 mg after one week, and then to 1.8 mg after another week. This is the same target dose used for type 2 diabetes treatment in clinical practice.

›Were the LEADER results consistent across subgroups?

Broadly, yes. No statistically significant treatment-by-subgroup interactions were detected across 17 pre-specified subgroups. The benefit appeared consistent regardless of age, sex, BMI, baseline HbA1c, and renal function. Patients with established cardiovascular disease showed a numerically greater benefit than those with risk factors only, though this interaction was not statistically significant.

›How does LEADER compare to SUSTAIN-6 for semaglutide?

SUSTAIN-6 showed a larger relative MACE reduction (26% vs. 13%) over a shorter follow-up (2.1 vs. 3.8 years) with fewer patients (3,297 vs. 9,340). SUSTAIN-6 was designed as a noninferiority trial and was not formally powered for superiority, while LEADER prospectively tested and confirmed superiority. Direct comparison between the two agents has not been performed in a head-to-head cardiovascular outcomes trial.

›What were the main side effects in the LEADER liraglutide group?

Gastrointestinal events were more common with liraglutide: nausea (reported in approximately 14% more liraglutide patients than placebo during the trial), vomiting, diarrhea, and decreased appetite. These side effects were most pronounced during dose titration and tended to diminish over time. Approximately 9.5% of liraglutide patients discontinued due to adverse events versus 7.3% on placebo.

›Did the LEADER trial include patients without established heart disease?

Yes. About 19% of participants (1,748 patients) were enrolled based on cardiovascular risk factors alone (age 60 or older with at least one risk factor) rather than established cardiovascular disease. In this subgroup, the point estimate for MACE did not favor liraglutide (HR 1.20), though the interaction test was not significant (p = 0.10). Guidelines generally reserve the strongest recommendation for patients with established disease.

›Is liraglutide FDA-approved for cardiovascular risk reduction?

Yes. Based on LEADER data, the FDA updated liraglutide's prescribing information in 2017 to include an indication for reducing the risk of major adverse cardiovascular events in adults with type 2 diabetes and established cardiovascular disease. This was the first such indication for a GLP-1 receptor agonist.

References

• Marso SP, Daniels GH, Tanaka-Poulsen K, et al. Liraglutide and cardiovascular outcomes in type 2 diabetes. N Engl J Med. 2016;375(4):311-322. PubMed
• 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. PubMed
• Mann JFE, Orsted DD, Brown-Frandsen K, et al. Liraglutide and renal outcomes in type 2 diabetes. N Engl J Med. 2017;377(9):839-848. PubMed
• Gerstein HC, Colhoun HM, Dagenais GR, et al. Dulaglutide and cardiovascular outcomes in type 2 diabetes (REWIND): a double-blind, randomised placebo-controlled trial. Lancet. 2019;394(10193):121-130. PubMed
• Davies MJ, D'Alessio DA, Fradkin J, et al. Management of hyperglycemia in type 2 diabetes, 2018. A consensus report by the ADA and EASD. Diabetes Care. 2018;41(12):2669-2701. PubMed
• Victoza (liraglutide) prescribing information. FDA. AccessData