Retatrutide Phase 2 trial: what did the paper find?

What did the retatrutide Phase 2 trial find?

The Phase 2 retatrutide trial found greater model-adjusted body-weight change in every trial group than with placebo (Jastreboff et al., New England Journal of Medicine, 2023). It was a double-blind, randomised, placebo-controlled study that followed 338 adults for 48 weeks on a once-weekly schedule (Jastreboff et al., NEJM, 2023). The primary end point was the least-squares mean percentage change in body weight at week 24: -17.5% in the highest trial arm against -1.6% with placebo (Jastreboff et al., NEJM, 2023). A least-squares mean is a model average that adjusts for people who missed measurements. The week-48 figures in the same paper were secondary results (Jastreboff et al., NEJM, 2023).

The order matters. The paper asked its main question at the earlier time point and treated the later reading as supporting evidence. Starting with the biggest number would tell the story backwards.

A placebo-controlled study gives the trial a reference group that follows the same study structure without the active trial drug. That reference group is the whole point. Without it, nobody can tell how much of a change belongs to the compound and how much belongs to being watched, weighed, and asked about your week for a year.

Evidence limits

What has this Phase 2 paper not shown?

The distinction is easy to lose in a headline. This paper is Phase 2, while how long retatrutide has been studied follows the later Phase 3 record (ClinicalTrials.gov NCT04881760). The trial also ran against placebo, so how retatrutide, Mounjaro, and Ozempic differ is the better place for a careful comparison of separate records.

It is worth saying plainly where this page stops. The accessible record is the PubMed abstract and the ClinicalTrials.gov entry, and the full NEJM text sits behind a paywall. Confidence intervals, baseline mean weight, discontinuation rates, and the detailed adverse-event tables are not in that record. So they are not here either, and nothing on this page estimates them.

How was the retatrutide Phase 2 trial designed?

The retatrutide Phase 2 trial assigned 338 adults across seven groups under a double-blind design, meaning neither participants nor investigators knew the assignment (Jastreboff et al., NEJM, 2023). Participants were randomised, or assigned by chance, in a 2:1:1:1:1:2:2 ratio across four ascending trial-arm levels and placebo (Jastreboff et al., NEJM, 2023). Two of those four levels were each reached by two separate groups, which is why seven groups report as five columns. Entry required a body-mass index of 30 or above, or 27 to under 30 with a weight-related condition, and ClinicalTrials.gov NCT04881760 sets the upper bound at 50.

ClinicalTrials.gov NCT04881760 records adults aged 18 to 75 at 28 sites in the United States and Puerto Rico, with the trial running from 20 May 2021 to 22 November 2022. Eli Lilly and Company funded the study (ClinicalTrials.gov NCT04881760; Jastreboff et al., NEJM, 2023).

Why build a trial with two roads to the same level? Because how a group arrives somewhere can change what happens along the way, and the researchers wanted that on the record. The abstract then combines each of those pairs when it reports results (Jastreboff et al., NEJM, 2023). The weekly rhythm of the study is a design choice too, and retatrutide's published pharmacokinetics explains why a weekly interval was the interval worth studying.

Is the 24.2% figure the trial's main result?

No. The retatrutide Phase 2 trial's primary end point was the percentage body-weight change recorded at week 24 (Jastreboff et al., NEJM, 2023; ClinicalTrials.gov NCT04881760). The least-squares means at that point ran -7.2% in the lowest trial arm, -12.9% and -17.3% in the two combined middle levels, and -17.5% in the highest, against -1.6% with placebo (Jastreboff et al., NEJM, 2023). The week-48 values across those same groups were -8.7%, -17.1%, -22.8%, and -24.2%, against -2.1% with placebo (Jastreboff et al., NEJM, 2023). Those later figures were registered as a secondary end point (ClinicalTrials.gov NCT04881760).

A primary end point is the main question a trial is designed and powered to answer. A secondary end point adds context. It does not get promoted to the main question simply because its number is larger, and the -24.2% figure travels furthest online for exactly that reason.

The paper reports combined results for each pair of groups that arrived at the same level by different routes. The accessible record does not publish a separate result for each of those groups (Jastreboff et al., NEJM, 2023).

How many participants reached 5%, 10%, or 15%?

In the retatrutide Phase 2 trial at 48 weeks, the proportions reaching body-weight changes of at least 5%, 10%, and 15% were 92%, 75%, and 60% in the second trial-arm level; 100%, 91%, and 75% in the third; and 100%, 93%, and 83% in the highest, against 27%, 9%, and 2% with placebo (Jastreboff et al., NEJM, 2023). A responder threshold is a fixed bar counted person by person, rather than an average across a whole group. The accessible abstract does not publish a responder row for the lowest trial arm (Jastreboff et al., NEJM, 2023), so this page does not infer one.

An average and a responder percentage answer different questions. The average says where the group sat as a whole, and it can hide both the people who barely moved and the people who moved a great deal. The responder rows count how many individual people crossed each line. Neither reading is complete on its own, which is why the paper prints both.

The placebo column belongs in the answer too. Some participants crossed the first threshold without the active trial drug at all, and that is exactly the sort of thing a reference group exists to reveal.

What did the Phase 2 trial report about side effects?

The retatrutide Phase 2 trial reported that the most common adverse events were gastrointestinal, related to the trial-arm level, and mostly mild to moderate (Jastreboff et al., NEJM, 2023). The paper says those events were partly mitigated in the groups that began from the lower of the two starting points (Jastreboff et al., NEJM, 2023). It also reports increases in heart rate that tracked the trial-arm level, peaked at 24 weeks, and declined afterwards (Jastreboff et al., NEJM, 2023). Those lines are the full safety detail in the accessible abstract, and they do not support a broader cardiovascular conclusion in either direction.

Tracking the trial-arm level means the pattern changed as the assigned label changed. That is the kind of gradient researchers watch for, because a signal that scales with the arm is harder to explain away as coincidence than one that shows up in a single group.

The detailed adverse-event tables and the discontinuation rates sit behind the journal paywall and were not retrieved. This page therefore keeps the paper's own wording and stops where the accessible record stops.

Is research-use retatrutide the same material this trial used?

No. The retatrutide figures on this page came from a trial drug product used under clinical trial conditions, with 338 adults under medical supervision (Jastreboff et al., NEJM, 2023). Research-use reference material is a laboratory compound, not that trial product. The distinction matters because a result belongs to the exact material, people, and protocol that produced it. Research material is not a medicine, not a substitute for one, and not for human or animal use. Nothing in this paper turns a trial-arm label into an instruction, or a group result into an outcome any reader should expect.

Laboratory buyers can inspect availability and lot paperwork for retatrutide reference material. That is a catalogue and documentation question, not a clinical one. The quality and verification hub explains what a certificate records, what a third-party report can establish, and how a batch stays tied to its own documents.