What Did the Retatrutide Phase 3 Trial Show?

It showed that retatrutide, an investigational once-weekly injection that activates three hormone receptors at once, reduced average body weight by roughly a fifth to a quarter over eighty weeks in adults with obesity or overweight who did not have diabetes, against a few per cent on placebo. The trial is called TRIUMPH-1. Lilly announced its results in May 2026: 2,339 participants, three active arms and a placebo arm, and every arm met its primary goal 1. On the company's more conservative way of counting, the average body-weight change was 17.6%, 23.7% and 25.0% across the low, middle and high arms, against 3.9% on placebo 1.

That is the answer. What it leaves out is almost as important: there are two sets of figures and the larger one is usually the one quoted; the widely shared 30% number comes from a smaller follow-up group, not from the trial as a whole; roughly one in nine people in the highest arm stopped because of side effects; and the result describes people without diabetes. Each of those gets its own question below.

What is retatrutide?

An investigational peptide that activates three hormone receptors — GIP, GLP-1 and glucagon — which is why it is often called a triple agonist. The earlier drugs in this family activate one receptor (semaglutide, GLP-1) or two (tirzepatide, GIP and GLP-1). Retatrutide adds the glucagon receptor, which is thought to raise energy expenditure and act on the liver, on top of the appetite and blood-sugar effects of the other two 4.

Its Phase 2 trial, published in 2023, was the reason Phase 3 attracted so much attention. In that forty-eight-week study, most participants in the higher arms reached a body-weight reduction of 15% or more, gastrointestinal side effects were the most common adverse events, and heart rate rose in a way that depended on the arm before settling 4. Phase 2 is designed to find out whether a compound is worth a large trial and at what exposure levels. Phase 3 is the large trial.

What exactly was TRIUMPH-1?

A randomised, placebo-controlled trial in adults with obesity, or with overweight and at least one weight-related health problem, none of whom had diabetes. Participants were allocated to one of three retatrutide arms or to placebo, and the exposure was stepped up every four weeks until each person reached their assigned arm. The main measurement was the percentage change in body weight at eighty weeks. The average starting body-mass index was 40, which is well inside the range usually classed as severe obesity 1.

Randomisation is what gives the result its weight. Because chance, not choice, decided who received what, the difference between the retatrutide arms and placebo can be attributed to the compound rather than to the kind of person who ended up in each group. The placebo arm matters just as much: people in weight trials receive lifestyle advice and regular contact, and their weight changes too. The placebo figure shows how much change the trial setting produces on its own.

Why are there two different sets of numbers?

Because the trial counted the people who stopped treatment in two different ways, and each way answers a different question. Lilly reported both 1.

The first column counts everyone who was randomised, whether or not they kept taking the treatment. It asks: if a group of people is started on this, what happens to that group on average? People who stopped early pull the average towards placebo, which is why these numbers are lower. The second column estimates what happens in people who stay on treatment as intended. It asks a narrower question — what does the compound do when it is actually being taken? — and gives larger numbers 1.

Neither is a trick, and neither is the "real" one. The first is closer to what a population would experience; the second is closer to what the molecule does. The problem comes only when a figure is quoted without saying which it is. Headlines have tended to use the second column. The plain answer at the top of this page uses the first.

Where does the 30% figure come from?

From a smaller, longer extension, not from the main trial. After the eighty-week measurement, 532 participants whose starting body-mass index was 35 or above continued to 104 weeks, all moved to the highest arm or the highest they could tolerate. That group reached an average body-weight change of 30.3% 1.

The number is genuine, but it describes a particular group: people who stayed in the trial for two years, started with a higher body-mass index, and had already tolerated treatment long enough to continue. A group chosen that way will usually look better than the full randomised population, because the people most likely to drop out or struggle with side effects are no longer in it. It is useful information about what longer exposure can reach. It is not the trial's headline result, and repeating it as though it were overstates what TRIUMPH-1 showed.

How many people stopped because of side effects?

About 4% in the low arm, 7% in the middle arm and 11% in the high arm, against about 5% on placebo 1. So the highest arm roughly doubled the placebo rate of stopping for adverse events, while the lowest arm did not differ much from placebo at all.

The side effects were mainly gastrointestinal, the same pattern as the rest of this drug family. In the high arm, 42.4% reported nausea against 14.8% on placebo, 32.0% diarrhoea against 13.5%, 26.1% constipation against 10.9%, and 25.3% vomiting against 4.8% 1. These are counts of people who reported the event at any point, not of people who had it throughout, and most such events in this class occur while the exposure is being stepped up.

One signal is specific to retatrutide. In the knee osteoarthritis trial, TRIUMPH-4, dysaesthesia — an abnormal or unpleasant skin sensation — was reported in 8.8% and 20.9% of participants in the middle and high arms, against 0.7% on placebo, and 18.2% of the high arm stopped because of adverse events 3. Lilly also reported it in TRIUMPH-2 and TRIUMPH-3 at lower rates 2. What causes it, and whether it resolves, is the kind of question the full peer-reviewed reports need to answer.

Who was actually studied?

Adults with obesity or overweight and no diabetes, with an average starting body-mass index of 40 1. That matters because this family of compounds produces smaller body-weight changes in people with type 2 diabetes, and the sister trial shows it. In TRIUMPH-2, which enrolled 1,152 adults with type 2 diabetes and obesity or overweight, the average change over eighty weeks was 12.7%, 19.1% and 20.8% across the three arms, against 4.0% on placebo 2.

The other two trials widen the picture without changing the main point. TRIUMPH-3 enrolled 1,949 adults with severe obesity and established cardiovascular disease and reported average changes of 21.6% and 22.6% in its two arms against 3.2% on placebo 2. TRIUMPH-4, announced in December 2025, enrolled 445 adults with obesity or overweight and knee osteoarthritis, and reported a 28.7% change in the high arm at sixty-eight weeks alongside a large reduction in knee pain scores 3. Each result belongs to its own population.

Does this mean retatrutide beats tirzepatide or semaglutide?

Not on this evidence, because nobody has tested them against each other. Tirzepatide's own obesity trial ran for seventy-two weeks in adults without diabetes and reported average reductions of roughly 15% to 21% across its arms 5. It is tempting to set that beside TRIUMPH-1 and subtract. The trials differed in length, in starting body-mass index, in where they ran and in how their placebo groups behaved, and each of those moves the number.

The honest statement is narrower. Retatrutide produced among the largest average body-weight changes yet reported in a Phase 3 obesity trial. Whether it outperforms an existing medicine in the same people under the same conditions is a question only a head-to-head trial can settle.

What does the trial not tell us?

Four things, and each is a normal limit of a trial at this stage rather than a flaw in it. First, the figures come from company announcements 12. A company release is a summary; the full methods, the handling of missing data and the complete adverse-event tables are what a peer-reviewed report sets out, and that is where the result can be checked by others.

• Duration. Eighty weeks is long for a weight trial and short for a lifetime. What happens over five or ten years, and what happens after treatment stops, is not measured here.
• Outcomes. Body weight is an intermediate measure. Whether the change reduces heart attacks, strokes or deaths is a separate question that needs a trial built to count those events.
• Body composition. A body-weight percentage does not say how much of the change was fat and how much was lean tissue.
• Population. The main result describes adults without diabetes, starting at a high body-mass index. It does not transfer automatically to anyone else.

Is retatrutide approved?

No. It remains investigational, and Lilly has said it plans to submit it to the FDA in the first quarter of 2027 2. Until a regulator has reviewed the full dossier — the trial data, the manufacturing, the safety record — there is no approved product, no approved label and no approved use. A successful Phase 3 trial is the evidence that goes into that review, not the review itself.

So what is the short answer?

In adults with obesity or overweight and no diabetes, retatrutide reduced average body weight by about 18% to 25% over eighty weeks, depending on the arm, against about 4% on placebo, counting everyone who started 1. Counting only those who stayed on treatment, the figures were higher, and a two-year subgroup reached about 30%. Side effects were mostly gastrointestinal, and stopping for adverse events rose with the arm.

What the trial does not show is that retatrutide is better than anything else, that the change lasts, or that it improves the outcomes body weight is meant to predict. It is a large, well-designed result on one endpoint in one population — a strong reason for the next questions, and not yet the answer to them.