Bottom line up front
TRIUMPH-3 met its weight-loss goal decisively: 22.6% average weight loss at 12 mg after 80 weeks in people with severe obesity and established cardiovascular disease. Its cardiovascular-event analyses were inconclusive, not positive or negative. The study was not designed with enough events to prove whether retatrutide reduces MACE.
What TRIUMPH-3 actually measured
TRIUMPH-3 (NCT05882045) was a randomized Phase 3 study of retatrutide in adults with a body-mass index of at least 35 kg/m² and established cardiovascular disease. Participants could have type 2 diabetes. The study’s central efficacy question was weight loss at week 80; cardiovascular events were evaluated in pre-specified analyses, but the trial was not powered as a dedicated MACE study. ClinicalTrials.gov lists the completed trial and its weight and cardiometabolic outcome measures.
This distinction matters. A weight-loss study can enroll people who have cardiovascular disease and still be unable to answer a cardiovascular-outcomes question. To prove a lower event rate, a trial needs enough heart attacks, strokes, cardiovascular deaths, and other qualifying outcomes for a precise comparison. TRIUMPH-3 ran for 80 weeks and included 1,949 people; it was designed to establish efficacy in a high-risk obesity population, not to replicate an event-driven outcomes program.
The 22.6% headline
At 80 weeks, people assigned to 12 mg retatrutide lost an average of 22.6% of baseline body weight, compared with 3.2% for placebo. The 9 mg arm lost 21.6%. Lilly described the 12 mg result as 55.8 pounds on average. The reported top-line trial data place that result in a clinically important context: these were people with obesity and established cardiovascular disease, not a lower-risk general obesity sample.
The result also arrived alongside favorable marker changes at 12 mg: triglycerides −37.0%, non-HDL cholesterol −16.5%, systolic blood pressure −9.3 mmHg, waist circumference −7.5 inches, and hsCRP −51.2%. Each direction is encouraging, especially in a cardiovascular-disease population. Still, favorable risk factors do not prove a reduction in clinical events. That question is deliberately being tested in a larger trial.
MACE-5 and MACE-3: the numbers that need context
| Composite | Events | HR (95% CI) | Interpretation |
|---|---|---|---|
| MACE-5 All-cause death, MI, stroke, heart-failure event, coronary revascularization | 44 retatrutide 52 placebo | 0.82 (0.55–1.22) | Numerically lower with retatrutide; CI crosses 1, so not statistically significant. |
| MACE-3 CV death, MI, stroke | 27 retatrutide 23 placebo | 1.12 (0.64–1.96) | Numerically higher with retatrutide; CI crosses 1, so not statistically significant. |
The table captures why headlines must be precise. MACE-5’s HR of 0.82 numerically favors retatrutide. MACE-3’s HR of 1.12 numerically favors placebo. Neither establishes a real difference because each confidence interval includes 1. It would be incorrect to present the first as cardiovascular protection or the second as evidence of harm. HCPLive’s summary reports the event counts and CIs; the key statistical implication is uncertainty.
Why the cardiovascular data were underpowered
Lilly said cardiovascular events occurred less frequently than expected in both trial arms. That is welcome for individual participants but challenging for statistical inference: fewer total events mean wider confidence intervals and less ability to distinguish a true treatment effect from chance. STAT’s reporting described this as the reason the study did not resolve the heart-benefit question.
On-treatment sensitivity estimates were HR 0.73 for MACE-5 and 0.92 for MACE-3. These figures can be useful for hypothesis generation, but they do not repair the original limitation. Analyses based on people who remained on treatment can differ from analyses that retain follow-up regardless of discontinuation, and neither can create events that never occurred. The in-study estimates remain the appropriate primary reference point for what TRIUMPH-3 did and did not show.
What this means for regulatory approval
The obesity case and the cardiovascular-benefit case are separate. Positive TRIUMPH-3 weight loss supports Lilly’s proposed obesity filing, and Lilly has said it plans a U.S. BLA submission in Q1 2027. Reuters reported the plan; it is a company filing plan, not an FDA approval or decision date. The current evidence does not support assuming that any initial label would include a cardiovascular risk-reduction indication.
That sequencing is normal in drug development. Weight-loss efficacy can be strong enough to support an obesity application while an ongoing outcomes program gathers longer-term evidence for a different claim. The evidence threshold for “helps people lose weight” differs from the evidence threshold for “reduces heart attacks or strokes.”
Why Wegovy’s SELECT trial is not a fair head-to-head comparison
Semaglutide’s SELECT trial showed a 20% reduction in three-point MACE in people with established cardiovascular disease and overweight or obesity. SELECT’s NEJM report describes an event-driven study with 17,604 participants and a cardiovascular primary endpoint. That is why SELECT could establish a cardiovascular-benefit claim.
TRIUMPH-3 had different size, duration, endpoints, event counts, medicine, dosing, and trial purpose. Comparing the 22.6% weight figure with SELECT’s MACE benefit answers no scientific question. Retatrutide might ultimately show cardiovascular benefit, no effect, or a different effect profile; TRIUMPH-3 cannot decide among those possibilities. The fair future comparison is between dedicated outcomes trials, not between an underpowered supportive analysis and SELECT.
A practical way to read the table
Start with the outcome definition, then the event count, then the interval around the estimate. MACE-5 is broader than MACE-3, so it asks a different question and has more total events. In this study, 44 versus 52 events within MACE-5 produced an HR below 1, while 27 versus 23 within MACE-3 produced an HR above 1. It is tempting to select the more favorable composite or to treat the two results as a contradiction. Neither approach is sound. Both were reported, both have wide intervals that include no difference, and neither was sized to provide a final clinical answer.
For a patient, this means the result should not be used to choose retatrutide over an approved medicine specifically for heart protection. For a clinician, it means the data can be discussed as supportive safety and efficacy information while acknowledging the unanswered event question. For a researcher, it is a reminder that an endpoint hierarchy and statistical power are as important as a dramatic weight-loss percentage. The forthcoming outcomes study is designed to turn this uncertainty into a more definitive answer.
TRIUMPH-Outcomes: the real cardiovascular trial
TRIUMPH-Outcomes (NCT06383390) is Lilly’s separate cardiovascular and kidney outcomes study. It is expected to include about 10,000 participants in public company descriptions and is designed around time to first event composites, including cardiovascular and kidney outcomes. Results are expected in 2028–2029. That event-driven design is intended to supply the statistical power that TRIUMPH-3 did not have.
Expert reactions: strong efficacy, unresolved outcomes
The reporting consensus is notably consistent. Fierce Biotech called the cardiovascular impact less clear, while STAT emphasized the absence of a statistically significant lower event risk. BMO Capital Markets analysts, quoted by Reuters, described the event dataset as limited and difficult to interpret, while noting the on-treatment directional signal. That is a reasonable reading: the data are worth following, but they are not a cardiovascular verdict.
What patients and clinicians should take away
For now, the most accurate summary is simple. Retatrutide produced very large average weight loss in TRIUMPH-3, including in people with established cardiovascular disease. It also improved several risk markers. But neither MACE analysis proved it prevents cardiovascular events. Patients should not delay established evidence-based cardiovascular or obesity care in anticipation of an investigational drug, and clinicians should separate the compelling weight-loss result from the unproven cardiovascular claim.
For a broader interpretation of the current evidence, see our cardiovascular outcomes guide. For the distinct, tightly limited pre-approval route Lilly announced, see our expanded-access update.