The number everyone repeats about retatrutide is 24.2 percent — the mean body-weight reduction at 48 weeks in the highest-dose arm of the Phase 2 trial Eli Lilly published in the New England Journal of Medicine in June 2023. It is a genuinely startling figure, the largest ever recorded for a drug in obesity. And I think it is the least interesting number in the dataset.
Here is my thesis. Retatrutide is not "tirzepatide, but more." It is a structurally different molecule — a triple agonist that adds the glucagon receptor on top of the GIP and GLP-1 receptors that Mounjaro and Zepbound already cover. The evidence that the third receptor actually does something is not in the headline weight figure. It is in five quieter numbers, most of which got a single line in the coverage and then disappeared.
As the Phase 3 TRIUMPH program reads out across 2026 and 2027, those five are what I will be watching — not the leaderboard. Let me take them in order.
First, a thirty-second mechanism
Semaglutide (Wegovy, Ozempic) is a single agonist: GLP-1 only. Tirzepatide (Zepbound, Mounjaro) is a dual agonist: GIP plus GLP-1. Retatrutide adds a third arm, the glucagon receptor. Glucagon is the hormone most people file under "raises blood sugar," which is exactly why putting an agonist for it into a metabolic drug sounds backwards. The bet is that at the right receptor balance, glucagon agonism raises energy expenditure and mobilizes fat out of the liver while the GLP-1 and GIP arms suppress appetite. Phase 2 was the first real test of whether that bet pays. It mostly did — but for reasons the 24 percent number hides.
Number one: the curve that never bent
The 24.2 percent figure conceals the more important fact. At 48 weeks, the weight-loss curve had not plateaued. In the 8 mg and 12 mg arms the line was still descending when the trial stopped. Every other obesity drug we have long data on flattens somewhere — semaglutide around the 60-week mark in the STEP trials, tirzepatide a little later in SURMOUNT. Retatrutide's Phase 2 simply ran out of road before it ran out of effect.
That single observation is why the longer Phase 3 trials matter more here than for any prior molecule. We do not know retatrutide's ceiling. Nobody does. The number everyone is ignoring is not 24.2 — it is the absence of a plateau underneath it.
Number two: one hundred percent
In the two highest-dose arms, every single participant — 100 percent — lost at least 5 percent of body weight. There were no non-responders. For contrast, in semaglutide's STEP-1 trial roughly 14 percent of patients never cleared that 5 percent bar. Non-response is the quiet failure mode of this drug class, the thing clinicians see in the room and trials bury in a subgroup table. A zero-non-responder arm, even in a few hundred people, is worth more to me than another point on the mean.
The interesting question retatrutide poses is not "how much weight." It is "what is the glucagon receptor doing that the other two cannot?" Four of these five numbers are attempts to answer that.
Number three: eighty-six percent (the liver)
This is the figure I think is most underrated. In a substudy of participants with metabolic-associated fatty liver disease, liver-fat content fell sharply, and normalization — liver fat under 5 percent — was reached in roughly 86 percent of participants in the higher-dose groups at 48 weeks. Glucagon agonism is the most plausible mechanistic driver; mobilizing hepatic fat is, after all, part of what glucagon does.
That matters commercially as much as clinically. MASH — metabolic dysfunction-associated steatohepatitis — is a market every large pharma is sprinting toward. Retatrutide's liver data hints that the glucagon arm may make it a MASH drug that also causes weight loss, rather than a weight-loss drug that happens to help the liver. A dedicated liver-outcomes study, not the obesity leaderboard, will settle that. The number to watch is the liver-fat number.
Number four: minus 2.02 (the diabetes read)
In the separate Phase 2 trial in type 2 diabetes, published in The Lancet in 2023, the 12 mg dose produced an HbA1c reduction of about 2.02 percentage points, alongside weight loss near 17 percent. The reason this matters: weight response is usually blunted in people with type 2 diabetes — it is one of the most reliable findings in the field, visible in every SURPASS and STEP diabetes arm. Retatrutide's diabetes weight numbers stayed high anyway. That is unusual, and it points again at the glucagon arm doing distinct metabolic work rather than just stacking appetite suppression.
Number five: the heart rate the bulls skip
Glucagon agonism does not come free. Retatrutide raised resting heart rate — dose-dependent increases reported on the order of several beats per minute in the higher-dose arms during Phase 2, peaking around the middle of the trial before attenuating. This is the number the most enthusiastic coverage waves past, and I think it is the single largest open question for the drug.
A few beats per minute is not alarming on its own. But it is exactly the kind of signal that only a large, long, cardiovascular-outcomes-powered trial can characterize properly — which is to say, exactly what Phase 3 is for. If retatrutide is going to be used by tens of millions of people for years, the heart-rate question is not a footnote. It is a gating item.
The counter-argument I take seriously
Here is the case against my own enthusiasm. Phase 2 was 338 people. Gastrointestinal side effects — nausea, vomiting, diarrhea — were common and dose-related, as they are across the whole class, and a non-trivial share of dropouts in any GLP-class trial trace back to them. A molecule that is still pushing weight down at 48 weeks is also a molecule whose long-term tolerability we have barely sampled. The plateau that never arrived in Phase 2 cuts both ways: it is exciting, and it means we have not yet seen where the body pushes back.
And the glucagon bet, elegant as it is on paper, has buried a lot of drugs before. Several earlier glucagon-containing candidates from other companies stalled on exactly the safety margins — heart rate, glucose handling — that retatrutide will now have to clear at scale. I think retatrutide is the most likely molecule to leapfrog tirzepatide. I do not think that is settled, and anyone telling you it is settled is selling the 24 percent headline, not reading the data.
What the TRIUMPH readouts will actually decide
So watch these, not the leaderboard. Does the curve finally bend in a longer trial, and where? Does the liver-fat result hold up in a properly powered MASH study? Does the heart-rate signal stay small and stable across years and a far larger population? Does the zero-non-responder result survive contact with thousands of patients instead of a few hundred? Those four questions decide whether retatrutide is a genuinely new category or simply the next, slightly larger number on a chart we have all started to find boring.
The 24 percent will get the headlines either way. The five numbers underneath it are where the actual story is — and, I would argue, where the next two years of this field will be won or lost.
Ozemback — May 2026