Retatrutide’s phase 2 trial: 24% weight loss at 48 weeks, in 338 people

In a 48-week phase 2 trial of 338 adults with obesity, weekly retatrutide at 12 mg cut body weight by 24.2%, against 2.1% on placebo [1]. At that dose, 83% of people lost at least 15% of their weight and a quarter lost 30% or more [1]. The authors note that no earlier randomised trial lasting a year or less had reported weight loss on this scale [1]. But this was a dose-finding study, and its headline numbers come from an analysis that assumes everyone kept taking their injections.

Why this study matters

Retatrutide is a single peptide that activates three hormone receptors: GIP, GLP-1 and glucagon [1]. Semaglutide hits one of those and tirzepatide two. The idea behind adding glucagon is that it may raise energy expenditure as well as cut appetite, though that's a mechanism theory the trial didn't directly test [1].

This NEJM paper is the reason retatrutide gets called the next step after tirzepatide. It was designed to find doses and map side effects before larger phase 3 trials; the paper points to a registered phase 3 trial, NCT05882045, as the next test [1].

What they did

The trial ran from May 2021 to November 2022 [1]. Participants were split across seven groups: placebo, 1 mg, two 4 mg groups (starting at 2 mg or 4 mg), two 8 mg groups (starting at 2 mg or 4 mg) and 12 mg (starting at 2 mg). Doses rose every four weeks for up to 12 weeks. Everyone got lifestyle counselling, but the protocol didn't require a set calorie deficit [1].

The primary outcome was weight change at 24 weeks. The 48-week figures that made the news were secondary outcomes [1].

Unusually, the trial deliberately enrolled roughly equal numbers of men and women, because women tend to lose more weight on these drugs [1].

What they found

Everyone on 8 mg or 12 mg lost at least 5% of their weight, and 26% of the 12 mg group lost 30% or more [1]. The weight curves were still falling at week 48; the authors note a plateau hadn't been reached [1].

The effect varied. On 12 mg, people with a BMI of 35 or more lost 26.4% on average, against 21.5% for those below 35. Women on 12 mg lost 26.6%, men 21.9% [1].

Blood pressure, HbA1c, fasting glucose, insulin and most lipids improved. Of those with prediabetes, 72% on retatrutide had normal blood sugar at 48 weeks, compared with 22% on placebo. These were exploratory outcomes [1].

How the numbers were counted

This is where retatrutide's figures differ from the headline figures in STEP 1 or SURMOUNT-1. Those trials led with a “treatment policy” analysis that counted everyone randomised, including people who stopped their injections. This paper's main results use an “efficacy estimand”: an estimate of the effect if everyone had taken the drug as intended [1]. That tends to produce larger numbers. SURMOUNT-1 reported both, and tirzepatide 15 mg came out at −20.9% under the stricter method and −22.5% under the as-intended one [3]. The trial also made no adjustment for multiple comparisons, and the authors say its confidence intervals shouldn't be used to infer definitive effects [1].

None of that makes the retatrutide result wrong. It means you can't put 24.2% next to 20.9% and call it a fair fight.

Side effects

Adverse events were reported in 73% to 94% of people on retatrutide, against 70% on placebo, rising with dose [1]. Nausea, diarrhoea, vomiting and constipation were the most common, mostly mild to moderate, and concentrated in the dose-escalation weeks. Starting at 2 mg instead of 4 mg partly reduced them [1].

Between 6% and 16% of people on retatrutide stopped because of side effects. Nobody on placebo did [1].

Other points worth knowing [1]:

• Heart rate rose in a dose-dependent way, peaked at 24 weeks and then eased.
• Skin sensitivity or heightened skin sensation (hyperaesthesia) was reported by 7% on retatrutide versus 1% on placebo; none of these cases was severe.
• Serious adverse events occurred in 4% of both groups. One person on retatrutide died by drowning, judged unrelated to the drug.
• One case of acute pancreatitis, and transient liver-enzyme rises above three times normal in 1%.
• Fourteen people's BMI fell to 22 or below (13 on retatrutide); eight had their dose reduced under the protocol.

Caveats

• Size. 338 people split seven ways leaves some arms with only a few dozen participants. Rare side effects won't show up at this scale.
• Completion. 81% completed the trial overall, ranging from 76% to 87% across the retatrutide groups and 71% on placebo [1].
• Population. US sites only and majority White; just 4% had a BMI between 27 and 30 [1].
• Sponsor. Eli Lilly, which makes retatrutide, designed the trial and analysed the data, and a Lilly employee co-wrote the first draft [1].
• Duration. 48 weeks tells you nothing about what happens over years, or after stopping.

A companion phase 2 trial in 281 people with type 2 diabetes found dose-dependent weight loss of up to 16.94% at 36 weeks on 12 mg, against 3.00% on placebo, along with HbA1c falls of up to 2.02 percentage points at 24 weeks [2].

What it means in practice

Retatrutide produced very large average weight loss in a well-run but modest phase 2 trial, with a side-effect pattern that looks like the GLP-1 drugs plus a heart-rate rise and some skin-sensation complaints. The results were promising enough to justify phase 3. They don't yet tell you how it compares with approved drugs, how safe it is over years, or what people experience outside a trial. At the time this paper was published, retatrutide was an investigational drug and not approved anywhere.

What is still open

• Whether the phase 3 programme reproduces these numbers in larger, more varied populations.
• How much of the weight lost is lean tissue; the abstract and results we read don't report body-composition scans.
• What the heart-rate increase means over the long term.
• What happens to weight after stopping.
• How it performs head to head against tirzepatide or semaglutide.

What this means

We think the phase 2 result is real, and we expect phase 3 to hold it directionally. A triple agonist that pushes average weight loss past the mid-20s in humans is a genuine step beyond tirzepatide, not a rounding error, and none of the trial's limits undercut that: the counting method and the absent body-composition scans change how you read the number, not whether the drug works.

The claim we would defend in print is narrower than the headline. Read 24.2% as the ceiling a person reaches on 12 mg for a year if they stay on it, not as what the average patient should expect. On the higher doses, roughly a third of people did not lose 15% of their weight at all, and 6% to 16% stopped injecting because of side effects. A figure framed as "if everyone took the drug as intended" is the right one for a dose-finding trial, but you and your doctor live in the world where people stop.

On the risk side, we would argue the weight number is not what decides retatrutide's fate; the heart-rate rise and the 7% reporting skin sensitivity are. If those stay mild and reversible at phase 3 scale, triple agonism becomes the new benchmark and the extra glucagon arm earns its place. If they do not, the weight figure stops mattering.

The comparison everyone wants, retatrutide against tirzepatide, is the one that will not arrive on its own. Retatrutide is an investigational, sponsor-owned molecule, and Lilly funds a head-to-head only when it serves the approval path, so a fair fight is not something the field should wait for. Until then, treat the cross-trial numbers as what they are: one upper bound under a generous method, set beside tirzepatide's own upper bound.

What readers report

First-hand accounts, shared with permission. They are self-reported, not evidence, and sit alongside the research above rather than in place of it.

Educational content only — not medical advice. Many peptides discussed on HPR are not approved for human use. Talk to a qualified clinician before making any decision about your health.