Mazdutide (IBI362)
Mazdutide acts at the GLP-1 and glucagon receptors. The phase 3 GLORY-1 trial put 610 adults into three groups by chance. At 48 weeks, the two Mazdutide groups had mean weight changes of -11.00% and -14.01%. The placebo group had a 0.30% gain (PMID 40421736; read 25 Aug 2026). Innovent Biologics paid for GLORY-1. The register keeps evidence tier B and EU status unclear. No direct official approval record was opened in the source set checked 25 Aug 2026.
Which names and identifiers belong to Mazdutide?
GSRS uses MAZDUTIDE as its main name (GSRS; checked 25 Aug 2026). It lists IBI362, IBI-362, LY3305677, and LY-3305677 for the same drug. The UNII is MB76Z4IBZ5. The CAS number is 2259884-03-0.
What did the 610-person GLORY-1 trial find?
GLORY-1 put 610 adults in China into two Mazdutide groups or a placebo group by chance (PMID 40421736; read 25 Aug 2026). The trial was blind to both the staff and the people in it. It ran for 48 weeks. Innovent Biologics paid for it. Seven authors said they worked for Innovent.
| GLORY-1 result at week 48 | First Mazdutide group | Second Mazdutide group | Placebo | Source |
|---|---|---|---|---|
| Mean change in body weight | -11.00% | -14.01% | +0.30% | PMID 40421736 · read 25 Aug 2026 |
| Participants with at least 15% weight loss | 35.7% | 49.5% | 2.0% | PMID 40421736 · read 25 Aug 2026 |
The trial reported P < 0.001 for each test against placebo. The open abstract did not give a full table of unwanted events.
What did GLORY-2 and the US phase 2 trial find?
GLORY-2 had 461 adults in China with obesity (PMID 42251595; read 25 Aug 2026). Innovent Biologics paid for the phase 3 trial. The NCBI record has a full abstract but no article body.
| GLORY-2 result at week 60 | Mazdutide | Placebo | Source |
|---|---|---|---|
| Mean change in body weight | -16.65% | -1.50% | PMID 42251595 · read 25 Aug 2026 |
| Between-group difference | -15.15% (95% CI -17.22 to -13.09) | — | PMID 42251595 · read 25 Aug 2026 |
| At least 5% weight loss | 84.3% | 33.1% | PMID 42251595 · read 25 Aug 2026 |
| Difference in 5% responders | 51.6% (95% CI 43.0 to 60.1) | — | PMID 42251595 · read 25 Aug 2026 |
Both GLORY-2 tests had P < 0.001. Six authors said they worked for Innovent. The paper says Innovent helped plan and run the study. It says Innovent also worked on the data and the paper.
A US phase 2 trial put 179 adults into four groups by chance (PMID 42628555; read 25 Aug 2026). The trial took place at 24 sites. The people did not have type 2 diabetes. Mean weight changes at week 32 were -7.3%, -15.6%, and -18.1% across three Mazdutide groups. The placebo group had a -0.9% change. The gaps from placebo ranged from -6.5% to -17.2%. Each test had P < 0.0001. Eli Lilly paid for the trial. Nine authors said they worked for Lilly and held its stock.
What did the diabetes phase 3 trials find?
One phase 3 trial put 320 adults in China into three groups by chance (PMID 41407859; read 25 Aug 2026). All had type 2 diabetes. Innovent Biologics paid for the trial.
| Result at week 24 | First Mazdutide group | Second Mazdutide group | Placebo | Source |
|---|---|---|---|---|
| Mean HbA1c change | -1.57% | -2.15% | -0.14% | PMID 41407859 · read 25 Aug 2026 |
| Mean body-weight change | -5.61% | -7.81% | -1.26% | PMID 41407859 · read 25 Aug 2026 |
Each test in that table had P < 0.0001. The most common unwanted events were diarrhoea, less hunger, and nausea. The abstract did not give their rates.
A second phase 3 trial put 731 adults with type 2 diabetes into three groups by chance (PMID 41407860; read 25 Aug 2026). Innovent Biologics paid for it. The other drug was dulaglutide.
The two Mazdutide groups cut HbA1c by 0.24 and 0.30 percentage points more than dulaglutide. The results had P = 0.0032 and P = 0.0003. Weight loss was 3.78 and 5.76 percentage points greater than with dulaglutide. Both weight tests had P < 0.0001.
The most common events were diarrhoea, nausea, and vomiting. Stomach and bowel events were more common with Mazdutide than with dulaglutide. The abstract did not give their rates.
Which unwanted events did the trials record?
In GLORY-1, an event made 1.5%, 0.5%, and 1.0% of the two Mazdutide groups and placebo stop treatment (PMID 40421736; read 25 Aug 2026). Stomach and bowel events were the most common. Most were mild or moderate.
In GLORY-2, an event made 2.9% with Mazdutide and 0% with placebo stop treatment (PMID 42251595; read 25 Aug 2026). Vomiting was seen in 53.1% and 1.3%. Nausea was seen in 46.9% and 3.2%. Diarrhoea was seen in 39.4% and 6.5%.
In the US phase 2 trial, stomach and bowel events were the most common (PMID 42628555; read 25 Aug 2026). Most were mild or moderate. The highest study group had the highest event-linked stop rate at 20%.
An earlier phase 2 trial has a full open safety table for 248 people (PMCID PMC10719339; read 25 Aug 2026). Innovent Biologics paid for it.
| Phase 2 safety finding | First Mazdutide group | Second Mazdutide group | Third Mazdutide group | Placebo |
|---|---|---|---|---|
| Any event after treatment began | 58/62 (93.5%) | 60/63 (95.2%) | 59/61 (96.7%) | 50/62 (80.6%) |
| Serious event after treatment began | 1/62 (1.6%) | 4/63 (6.3%) | 4/61 (6.6%) | 0/62 |
| Event that made a person stop treatment | 0/62 | 1/63 (1.6%) | 0/61 | 0/62 |
| Nausea | 13/62 (21.0%) | 15/63 (23.8%) | 25/61 (41.0%) | 3/62 (4.8%) |
| Vomiting | 8/62 (12.9%) | 13/63 (20.6%) | 17/61 (27.9%) | 2/62 (3.2%) |
| Diarrhoea | 12/62 (19.4%) | 19/63 (30.2%) | 19/61 (31.1%) | 9/62 (14.5%) |
The full paper reported no severe low blood sugar. Eight authors said they worked for Innovent. The paper says Innovent helped plan the trial. It says Innovent also took part in the data work and writing.
What status is established?
The live register keeps evidence tier B (field checked 25 Aug 2026). Phase 3 human results exist. No direct official approval record was opened in the source set. That gap does not prove that no approval exists.
Exact-name EudraCT searches found no records for Mazdutide, IBI362, or LY3305677 (EudraCT; checked 25 Aug 2026). This is a gap in that one register. It does not settle current EU sale status. The EU field stays unclear.
No dated vendor listing was established. The first_observed field remains empty rather than turning a forum comment into a vendor record.
What did one forum participant report?
One original Reddit comment was posted on 10 October 2024 (original comment; archive checked 25 Aug 2026). It is self-reported, self-selected, and free of anyone's money. The author wrote:
Yes. But has only helped with energy that the Tirz is taking away. Have noticed any more weight loss benefits from it
The wording is reproduced without correction. One comment is not a count or a consensus.
What does this register entry cover?
This entry covers the drug's names, human results, unwanted events, funding, status, and one marked forum account. It does not show what is in a given vial.
→ Weight-loss research and compounds → What can be in a vial → All register entries