Mazdutide Cut Weight by 18.1% in a U.S. Trial. The 20% Dropout Rate Matters.

Mazdutide just produced one of the strongest mid-stage obesity readouts yet seen in a U.S. population: average weight loss of 18.1% at 32 weeks on the highest dose. That number deserves attention. It also needs two immediate qualifiers. This was a 179-person phase 2 trial, not a pivotal approval study, and one in five participants assigned to 16 mg stopped treatment because of adverse events.

The interesting story is not simply that the drug worked. It is how sharply efficacy and tolerability separated as the dose climbed. That tension may determine whether mazdutide becomes a genuine next-generation option or another impressive molecule whose best headline dose proves hard to use in everyday care.

What the new U.S. trial tested

Published online in The Lancet Diabetes & Endocrinology on August 21, 2026, the study enrolled adults at 24 U.S. centres. Participants were 18 to 75 years old, did not have type 2 diabetes, and had either a body-mass index of at least 30 or a BMI from 27 to under 30 with a weight-related condition. The design was randomised, double-blind and placebo-controlled, which is the right basic architecture for testing whether a drug itself drives weight change. Primary trial report on PubMed

The 179 participants were assigned to placebo or one of three once-weekly mazdutide regimens: a 3-to-6 mg schedule, 10 mg, or 16 mg. The lower-dose group stayed at 3 mg through week 32 before escalating to 6 mg, allowing investigators to examine 3 mg as a possible maintenance dose. The primary endpoint was percentage change in body weight at week 32. Treatment continued for 48 weeks.

The trial was completed and is registered as NCT06124807. The registry shows the stepwise escalation used for the higher doses: the 10 mg group reached its target after moving through 1.5, 3, 6 and 8 mg; the 16 mg group moved through 1.5, 3, 6, 9 and 12 mg before reaching 16 mg. That matters because tolerability is shaped not only by the destination dose but also by the speed and size of each step. ClinicalTrials.gov record

The dose-response curve was real

At week 32, least-squares mean weight change was 7.3% down in the 3-to-6 mg group, 15.6% down with 10 mg, and 18.1% down with 16 mg. Placebo participants lost 0.9%. The estimated differences versus placebo ranged from 6.5 to 17.2 percentage points, and every comparison met the reported threshold of p<0.0001.

Those are large differences for a 32-week study. They also show that “mazdutide” is not one result. The dose selected changed the average outcome by more than ten percentage points. The paper reports further weight reduction through week 48, but its abstract does not provide the final group values, so the 32-week numbers are the cleanest published figures to use here.

There is another reason the U.S. result matters. Earlier evidence came largely from China. In a 2023 phase 2 trial of 248 Chinese adults, 24-week losses were 6.7%, 10.4% and 11.3% with 3, 4.5 and 6 mg, while placebo gained 1.0%. The new study does not permit a neat country-to-country comparison: doses, duration, eligibility and baseline characteristics differed. It does show that the biological signal carried into a U.S. multicentre trial. Earlier peer-reviewed phase 2 trial

Why add glucagon to GLP-1?

Mazdutide, also called LY3305677 or IBI362, is a modified oxyntomodulin analogue. It activates both the GLP-1 receptor and the glucagon receptor. GLP-1 activity can reduce appetite, slow gastric emptying and support glucose-dependent insulin secretion. Glucagon biology is more complicated: it can raise glucose, but it can also affect energy expenditure, lipid metabolism and food intake. The development bet is that carefully balanced co-activation can deliver more weight reduction and broader metabolic effects than GLP-1 signalling alone.

That mechanism separates mazdutide from tirzepatide, which combines GIP and GLP-1 receptor activity, and from retatrutide, which adds glucagon activity to GIP and GLP-1. It does not tell us which drug is “best.” Only randomised head-to-head trials using comparable populations, titration and follow-up can answer that. Comparing headline percentages from unrelated programmes is marketing arithmetic, not clinical evidence.

The mechanism also creates questions that weight alone cannot settle. Investigators and regulators will want larger datasets on heart rate, blood pressure, liver markers, gallbladder and pancreatic events, glycaemic effects, lean mass and treatment persistence. A drug intended for chronic use has to be judged by what people can remain on, not just by what the most aggressive dose can produce under trial supervision.

The 16 mg result came with a tolerability warning

Gastrointestinal events were the most common adverse effects and were mostly mild to moderate, which is familiar for incretin-based therapies. The sharper signal was discontinuation: 20% of participants assigned to 16 mg stopped study treatment because of adverse events, primarily gastrointestinal disorders. In a group of 51 people, that is too prominent to dismiss as background noise.

The abstract does not provide every adverse-event rate or a complete rare-event analysis. The trial was far too small to establish uncommon safety risks, and 48 weeks is short relative to the years many patients may use obesity medication. A phase 3 programme will need to show whether slower escalation, a lower maintenance dose, or better patient selection can preserve much of the efficacy while reducing dropouts.

The lower doses deserve more attention than the headline may receive. A 15.6% mean reduction at 32 weeks with 10 mg is substantial, while the unusual 3-to-6 mg arm helps probe whether lower maintenance exposure can still deliver useful benefit. Dose optimisation is not an afterthought here. It may be the central development problem.

What this does and does not mean for patients

Mazdutide is not FDA-approved. In fact, the FDA has recently identified mazdutide products sold online in the United States as unapproved new drugs. A vial marketed for “research” is not the pharmaceutical product tested across controlled clinical sites, and a published trial does not make grey-market versions legal, authentic or safe. FDA warning letter dated March 31, 2026

The regulatory picture is different in China, where mazdutide received approval for chronic weight management in June 2025. That decision drew on a Chinese phase 3 programme, not this new U.S. phase 2 study. Lilly holds rights outside China, while Innovent commercialises the medicine in China. Approval in one jurisdiction does not transfer to another. Coverage of the Chinese approval and GLORY-1 data

For U.S. readers, the practical conclusion is narrower: mazdutide now has a credible, peer-reviewed proof-of-concept result in a U.S. population. It still needs larger phase 3 trials, a fuller safety database, regulatory review and an approved supply chain before it can become a prescribing option.

The Oria take

This study moves mazdutide forward because it answers a question earlier Chinese trials could not fully answer: can the dual GLP-1/glucagon concept produce a strong dose-dependent effect in a U.S. obesity trial? The answer is yes. The 10 mg and 16 mg arms produced weight reductions that justify late-stage development.

But the most useful number may be 20%, not 18.1%. The first is the share of the highest-dose group who discontinued because of adverse events; the second is their average week-32 weight loss. Read together, they describe the real challenge. More receptor activity and more dose can drive more weight loss, but a medicine only works in practice when people can tolerate and continue it.

The next evidence checkpoint should therefore be more demanding than another percentage on a slide. We need phase 3 confirmation, transparent treatment-policy analyses that count what happens after discontinuation, longer follow-up, body-composition data, and enough participants to characterise less common harms. Direct comparisons would be especially valuable, because clinicians will otherwise be left choosing among mazdutide, tirzepatide, retatrutide and other multi-agonists using separate trials that were never designed to rank them.

Sources

Hsia SH et al. Efficacy and safety of mazdutide in adults with obesity or overweight: a US-based, multicentre, phase 2, randomised, placebo-controlled clinical trial. The Lancet Diabetes & Endocrinology. Published online August 21, 2026. DOI: 10.1016/S2213-8587(26)00160-9

ClinicalTrials.gov. NCT06124807: A Study of LY3305677 Compared With Placebo in Adult Participants With Obesity or Overweight. View the completed study record

Ji L et al. A phase 2 randomised controlled trial of mazdutide in Chinese overweight adults or adults with obesity. Nature Communications. 2023;14:8289. DOI: 10.1038/s41467-023-44067-4

U.S. Food and Drug Administration. Prime Sciences warning letter, March 31, 2026. Read the FDA letter

Evidence grade: B. A peer-reviewed, randomised, double-blind, placebo-controlled phase 2 trial shows a large and statistically convincing dose-response signal. Confidence is limited by the small sample, 48-week duration, sponsor involvement, lack of an active comparator, and a 20% adverse-event discontinuation rate at 16 mg.

Medical disclaimer: This article is for educational purposes only and is not medical advice. Mazdutide is not approved by the U.S. FDA. Do not buy or use products sold as mazdutide outside a lawful clinical trial or approved medical system. Decisions about obesity treatment should be made with a qualified healthcare professional who can assess your medical history, medications, risks and goals.