Bofanglutide
Bofanglutide, long-acting GLP-1 receptor agonist (development code GZR18)
Written by Aaron CuhaReviewed Sep 2026
Also known as: GZR18, Bofanglutide injection, Oral bofanglutide
In a phase 2b trial in 272 Chinese adults with type 2 diabetes it lowered glycated haemoglobin more than semaglutide 1 mg, and gastrointestinal side effects hit 82 to 87 percent of people on it against 52 percent on semaglutide. No phase 3 has reported.
Overview
Bofanglutide is a long-acting GLP-1 receptor agonist from Gan and Lee Pharmaceuticals in Beijing, developed under the code GZR18 and designed to be given once every two weeks rather than once a week. It is worth separating from the compound it is routinely confused with: HEC88473, a dual GLP-1 and FGF21 agonist from a different company, which this site covers on its own page. They are not the same molecule.
What exists is two phase 2b trials, both published in 2026, plus earlier phase 1b/2a and phase 1 work. That is a substantial body of human data for a compound most Western readers have never heard of, and it is still short of a phase 3.
The diabetes trial is the interesting one because it was run against semaglutide rather than against placebo. Over 24 weeks, the biweekly and weekly bofanglutide arms lowered glycated haemoglobin by 1.87 to 2.32 percentage points against 1.60 for semaglutide 1 mg. The treatment differences favoured bofanglutide, and in two of the four arms the confidence interval excluded zero comfortably.
The tolerability figures are the part that should temper enthusiasm. Gastrointestinal adverse events occurred in 81.8 to 87.3 percent of bofanglutide participants against 51.9 percent on semaglutide. The trial was open-label, ran 24 weeks and enrolled only Chinese participants, all of which the authors list as limitations.
An oral formulation using the same absorption enhancer as oral semaglutide has completed a phase 1 study. Its relative bioavailability was measured at under one and a half percent for single doses, which is the ordinary arithmetic of oral peptides and the reason the oral doses are in tens of milligrams while the injected dose is under a milligram.
Mechanism of action
A GLP-1 receptor agonist engineered for a long circulating half-life, measured at approximately seven days in healthy American and Chinese adults, which is what allows a once every two weeks schedule rather than the weekly schedule of semaglutide and dulaglutide. Receptor pharmacology is conventional for the class: activation of the GLP-1 receptor on pancreatic beta cells increases glucose-dependent insulin secretion, suppresses glucagon, slows gastric emptying and reduces food intake through central pathways. The oral formulation is co-formulated with sodium N-(8-[2-hydroxybenzoyl] amino) caprylate, the same permeation enhancer used in oral semaglutide, which transiently raises gastric absorption of the peptide.
Human evidence
Two published phase 2b trials, one in type 2 diabetes against semaglutide and one in obesity against placebo, plus phase 1b/2a and phase 1 work including healthy American participants. No phase 3 has reported.
- Phase 2b diabetes, 272 participants, 24 weeks, open-label, against semaglutide 1 mg: glycated haemoglobin fell 1.87 to 2.32 percentage points on bofanglutide against 1.60 on semaglutide.
- Phase 2b obesity, 340 participants, 30 weeks, double-blind: weight loss of 9.75 to 16.69 percent against 1.15 percent on placebo.
- Gastrointestinal adverse events reached 81.8 to 87.3 percent in the diabetes trial against 51.9 percent on semaglutide, and 83.9 percent against 33.3 percent in the obesity trial.
- Phase 1 found a half-life of about seven days and no pharmacokinetic difference between healthy American and Chinese adults.
- An oral formulation has completed phase 1 with relative bioavailability under one and a half percent for single doses.
What this does not tell you: Every efficacy trial so far is phase 2b, conducted in China, and funded by the manufacturer with employees and a shareholder among the authors. The diabetes comparison was open-label, which matters for a drug class whose commonest adverse events are subjectively reported. The semaglutide comparator was dosed at 1 mg, not the 2 mg licensed for diabetes in several markets, so the comparison is against a submaximal dose. And there is no cardiovascular outcome data of any kind.
Reading the research record
A head to head against semaglutide at phase 2b is unusual and worth crediting. Most companies run their phase 2 against placebo, where a win is nearly guaranteed, and defer the comparison that actually informs prescribing.
But read the trial as the authors wrote it, not as the headline number. It was open-label, so both the participants and the investigators knew who was getting what, in a trial whose primary endpoint was a laboratory value but whose safety signal was self-reported nausea. The semaglutide arm was titrated to 1 mg. And the gastrointestinal burden on bofanglutide was substantially higher than on the comparator, which is the trade that anyone choosing between them would have to make.
The two phase 3 trials registered as of this review, one in type 2 diabetes and one comparing bofanglutide with semaglutide in Latin American adults with overweight or obesity, had not begun recruiting. Until one of those reports, this is a phase 2b compound.
One practical warning applies to every compound on this page. Chinese incretin drugs carry several names at once: a company development code, a Chinese trade name, an international nonproprietary name assigned later, and occasionally a partner's code after a licensing deal. Those names do not reliably appear together in any one source, so a search on the code can return only preclinical work while the clinical trials are indexed under the nonproprietary name, or the reverse. If a compound looks like it has thin evidence, check the other names before concluding that.
The evidence, charted
Fig. 1 · evidence composition
5of 5 citations (100%) are in people
Every citation cited here is Human work; no other study type is cited on this page. This count is of our own citation list and understates any literature larger than the sources we cite.
Fig. 2 · evidence over time
Citations here span just two years, 2025 and 2026.
Too few distinct publication years on this page to plot as a timeline.
Fig. 3 · legal status at a glance
US
Not approved
UK
Not approved
AU
Not approved
CA
Not approved
Not approved in any of the four jurisdictions shown. A jurisdiction's classification is a regulatory fact, not a verdict on the science; see Legal status below for the full text and any notes.
Fig. 4 · dose response
No human dose response curve exists
We draw this figure where the data supports it. For this compound in humans it does not, so the panel stays empty rather than borrowing an animal curve and implying it transfers.
Key studies & citations
- Human2026
Weekly and Biweekly Treatment With Bofanglutide Versus Semaglutide in Chinese Patients With Type 2 Diabetes: A Phase 2b Randomized Clinical Trial
Open-label phase 2b trial at 37 Chinese sites, 272 adults with type 2 diabetes, mean baseline glycated haemoglobin 8.35 percent. At week 24 the change from baseline was minus 1.87, minus 2.28 and minus 1.94 percentage points for bofanglutide 12, 18 and 24 mg every two weeks, minus 2.32 for 24 mg weekly, and minus 1.60 for semaglutide 1 mg weekly. Gastrointestinal adverse events, mostly grade 1 or 2, occurred in 81.8 to 87.3 percent of bofanglutide participants against 51.9 percent on semaglutide. Hypoglycaemia occurred in 0 to 3.8 percent versus 1.9 percent, none severe. The authors list open-label design, short duration and an all-Chinese population as limitations. Funded by Gan and Lee Pharmaceuticals.
Annals of Internal Medicine - Human2026
Efficacy and safety of bofanglutide, a GLP-1 receptor agonist, in Chinese adults with overweight or obesity: a randomized, double-blind, placebo-controlled phase 2b trial
Double-blind placebo-controlled phase 2b trial, 340 Chinese adults with overweight or obesity, mean age 33.1 years and mean weight 95.6 kg, randomised across five dose groups. Mean percentage change in body weight from baseline to week 30 ranged from minus 9.75 to minus 16.69 percent with bofanglutide against minus 1.15 percent with placebo. Adverse events occurred in 98.9 percent of the bofanglutide group against 86.4 percent of placebo and were mostly grade 1 to 2 gastrointestinal events, in 83.9 percent against 33.3 percent. 84.1 percent completed the trial. Several authors are Gan and Lee employees or shareholders.
Signal Transduction and Targeted Therapy - Human2026
GZR18, a GLP-1 analog with once-weekly or bi-weekly dosing for body weight management: A randomized, placebo-controlled, phase 1b/2a trial
The earlier and much smaller weight-management trial under the development code. Sixty participants enrolled, 46 completed. Least-squares mean change in body weight was minus 9.36 percent for GZR18 against 6.68 percent for placebo in part A, and minus 17.8 percent weekly and minus 12.8 percent every two weeks against 0.7 percent for placebo in part B. With 46 completers this is a dose-finding exercise, not an efficacy result, and the authors say as much in concluding that larger and longer trials are warranted.
Cell Reports Medicine - Human2025
The safety, tolerability, pharmacokinetics and pharmacodynamics of GZR18 in healthy American and Chinese adult subjects
Phase 1 dose-escalation studies in healthy American and Chinese adults. Half-life was approximately seven days in both populations and exposure was comparable between them, with geometric mean ratios for area under the curve and maximum concentration close to 1, so no ethnic pharmacokinetic difference was detected. Decreased appetite and nausea were the most frequent treatment-emergent events. This is the only study in the programme that included American participants at all.
Diabetes, Obesity and Metabolism - Human2026
Safety, tolerability, relative bioavailability, pharmacokinetics, and pharmacodynamics of single and multiple doses of the novel oral bofanglutide in healthy Chinese participants
Phase 1 study of an oral formulation co-formulated with the permeation enhancer SNAC, in 92 healthy Chinese participants across two parts. Single-dose relative bioavailability was 0.56 percent at 30 and 60 mg and 1.31 percent at 90 mg. On repeated daily dosing a 60 minute post-dose fast produced roughly twice the exposure of a 30 minute fast, 9.22 percent against 4.88 percent apparent relative bioavailability. Most adverse events were grade 1 gastrointestinal. This is a pharmacokinetic study; it reports no efficacy endpoint.
Diabetes Research and Clinical Practice
Frequently asked questions
Is bofanglutide better than semaglutide?
In one open-label phase 2b trial in 272 Chinese adults with type 2 diabetes it lowered glycated haemoglobin more than semaglutide 1 mg over 24 weeks. It also produced gastrointestinal adverse events in 82 to 87 percent of participants against 52 percent on semaglutide. That is a phase 2b result in one country against a submaximal comparator dose, and there is no phase 3 and no cardiovascular outcome data.
Is bofanglutide the same as HEC88473?
No. HEC88473 is a dual GLP-1 and FGF21 receptor agonist from a different company with only a first-in-human single ascending dose study behind it. Bofanglutide, development code GZR18, is a GLP-1 receptor agonist from Gan and Lee with two published phase 2b trials. The two are frequently conflated because both are Chinese GLP-1 programmes with an alphanumeric code.
Can I get it?
No. It is not approved in any country. Two phase 3 trials are registered and had not started recruiting as of this review.
Why every two weeks instead of weekly?
Its measured half-life in healthy adults is about seven days, roughly double that of semaglutide, which is what makes a fortnightly schedule pharmacologically plausible. The phase 2b trials tested weekly and fortnightly arms side by side; in the diabetes trial the weekly 24 mg arm produced the largest glycated haemoglobin reduction.
What about the oral version?
It has completed a phase 1 study in healthy Chinese participants using the same absorption enhancer as oral semaglutide. Relative bioavailability was under one and a half percent for single doses, and a longer post-dose fast roughly doubled exposure. No efficacy trial of the oral form has been published.