UBT251 phase 2: 205 participants, 24 weeks, 19.7% maximum mean weight loss. Read the published evidence and why cross-trial rankings mislead.
UBT251 is an investigational GLP-1, GIP and glucagon receptor agonist. A Chinese phase 2 trial in 205 adults without type 2 diabetes reported up to 19.7% mean weight loss at 24 weeks, versus 2.0% with placebo. That is a trial result, not a prediction for an individual or proof that UBT251 beats retatrutide.
Evidence checked October 21, 2026. This page explains human research, not purchasing, dosing or self-administration.
What is UBT251?
The three receptor targets put UBT251 in the same broad mechanism category as retatrutide. They do not make the molecules interchangeable. Our comparison below keeps each result attached to its own study rather than treating a receptor count as a performance score.
The published paper is UBT251, a novel triple GLP-1, GIP, and glucagon receptor agonist, in adults with overweight or obesity, in Cell Metabolism. Its publisher page is dated October 7, 2026. An earlier company announcement came out on February 24, 2026. The evidence is no longer just a press-release headline.
What did the 24-week trial find?
We separate the treatment-group result from the placebo-adjusted difference. Both are useful; they answer different questions.
| Measure | Published phase 2 result | How to read it |
|---|---|---|
| Participants | 205 adults in China, without type 2 diabetes | A defined trial population, not all people with obesity |
| Follow-up | 24 weeks | Does not establish multi-year durability |
| Mean weight change across UBT251 arms | −13.6% to −19.7% | 19.7% belongs to the best-performing arm |
| Placebo mean weight change | −2.0% | The within-study reference group |
| Greatest placebo-adjusted difference | −17.7 percentage points | 19.7 minus 2.0, not another treatment-arm result |
| 95% confidence interval for that difference | −20.8 to −14.6 percentage points | Statistical uncertainty around the estimate |
Source: published randomized phase 2 paper00434-1).
The word mean matters. Some participants lost more and some less. “Up to 19.7%” describes the largest group mean reported, not the maximum individual loss and not the average across every UBT251 arm.
UBT251 versus retatrutide: what can we compare?
Mechanisms and study designs can be compared. Superiority cannot be established by placing headline percentages beside each other.
Our retatrutide Phase 3 evidence card separates results for people who stayed on treatment from analyses counting everyone who started. Before comparing any UBT251 number with that card, check population, follow-up length and the analysis used. A 24-week placebo-controlled study and an 80-week study answer different questions.
The UBT251 trial did not randomize participants against retatrutide or tirzepatide. It therefore cannot establish which of those treatments performs better.
Safety evidence and the missing pieces
The paper identifies gastrointestinal symptoms as the most frequent adverse events, generally mild to moderate. We have not reproduced arm-specific adverse-event percentages here because a useful safety comparison needs the full denominators and discontinuation counts, not an isolated reassuring phrase.
A 205-person, 24-week study is not enough to settle rare harms, long-term outcomes or what happens after treatment stops. Participants without type 2 diabetes also do not establish the same result in people with diabetes.
The Chinese trial registry records actual enrollment of 205. A separate global phase 2 registry lists estimated enrollment of 333. That planned number is not an efficacy finding. We keep registrations separate from reported results.
Sources and related research
Frequently asked
Related Articles
New guides, by email
New compound guides and research notes when we publish them. No set schedule. Unsubscribe anytime.
No spam. Unsubscribe anytime.
Evidence notes from published research and trial registries. Educational information, not medical or dental advice.