Question 1

Is there a GLP-3 dosage chart?

Accepted Answer

A chart needs a molecule, and the GLP-3 label does not name one. Lay out the molecules it might cover and the problem is plain at once. Cotadutide, a dual GLP-1 and glucagon agonist, was given at 100, 300 or 600 μg in a kidney study. Survodutide, another dual agonist, is being tested at 3·6 or 6·0 mg once a week. Retatrutide, the triple agonist, was given at 4, 9 or 12 mg once a week in its phase 3 trial. The list does not stop at three. A 2025 review names further molecules in development, among them the dual agonists pemvidutide and mazdutide and a second triple agonist, efocipegtrutide, and the abstracts behind this page give no amounts for any of them. Micrograms and milligrams can share one chart only if every row names its molecule. Strip the names and print a single column headed GLP-3, and the chart would mix amounts that differ more than a hundredfold. That would be a chart that cannot be read rather than a simplified one, which is why this page lists the molecules instead of a GLP-3 amount.

Question 2

What amounts did the retatrutide trials assign?

Accepted Answer

The 2023 phase 2 trial gave weekly injections of placebo, 1·5 mg of dulaglutide, or retatrutide at maintenance doses of 0·5, 4, 8 or 12 mg, with some arms starting lower and climbing. A maintenance dose is the amount an arm settled at after any climb, and dulaglutide is an older GLP-1 medicine used as the comparison. The 2026 phase 3 trial narrowed the list to three amounts, assigning 537 adults with type 2 diabetes to retatrutide at 4, 9 or 12 mg, or placebo, for 40 weeks. Two things changed between the trials. The 0·5 mg arm was dropped, and a 9 mg arm appeared that the phase 2 trial had not tested. Neither change is a recommendation. Each is a design decision about where the next experiment should look, and the full retatrutide record, set against its weight results, is on the Retatrutide dosage page.

Question 3

What amounts did the dual agonists use?

Accepted Answer

The two-receptor molecules were each tested at a different stage, so their amounts answer different questions. The earliest, SAR425899, was first given to healthy volunteers as single-ascending doses (0·01-0·1 mg), meaning one injection at a time at rising amounts. A second trial then gave daily doses of SAR425899 or placebo over 21 or 28 days, to volunteers and to patients with type 2 diabetes. Cotadutide reached a kidney study in 247 participants, where people were randomised and titrated to 100, 300 or 600 micrograms, or to 1 mg of semaglutide. Survodutide is the furthest along, in a 76-week phase 3 trial in Japan comparing two weekly amounts with placebo, alongside a reduced-calorie diet and increased physical activity. A first-in-human test, a kidney study and an obesity trial are three different experiments, and their amounts belong to them rather than to each other.

Question 4

Who received those amounts, and alongside what?

Accepted Answer

Every amount above was given to a defined group, on top of something else, and the result belongs to that combination. The retatrutide phase 2 trial enrolled people who were treated with diet and exercise alone or with a stable dose of metformin (≥1000 mg once daily) for at least 3 months before the screening visit. The phase 3 trial tested retatrutide as a monotherapy, meaning on its own, in people whose type 2 diabetes was inadequately controlled by diet and exercise alone. The cotadutide amounts went to people with chronic kidney disease as well as type 2 diabetes, and its model found baseline blood pressure affected how far urine protein fell in those participants. So none of these amounts was given to a person with nothing else going on. A diet, a tablet, a kidney condition or an exercise plan sat underneath each one, and that background is part of what the number means rather than a footnote to it.

Question 5

How were these molecules given, and how often?

Accepted Answer

Under the skin every time, and weekly for retatrutide and survodutide, the two that reached phase 3 trials. The retatrutide phase 3 trial used once-weekly subcutaneous injection, subcutaneous meaning into the fat just under the skin. A 2023 commentary on the early retatrutide work notes that its pharmacokinetics, meaning how long it stays in the blood, support once-weekly dosing. Survodutide is also given once a week, while the earliest dual agonist was tested with subcutaneous administrations given daily over three or four weeks. The interval belongs to the molecule rather than to a preference. A weekly molecule and a daily one are not two settings of one dial, so moving an amount from one schedule to the other produces an exposure that no trial among these sources has measured rather than an equivalent one.

Question 6

What was the starting amount, and how fast did it climb?

Accepted Answer

The retatrutide phase 2 trial made the climb itself part of the experiment, which is unusual and useful. It ran two arms that both settled at 4 mg, one starting at 2 mg and one with no escalation at all, and two arms that both settled at 8 mg, one starting at 2 mg and one starting at 4 mg. Each of those four arms held about two dozen people, between 23 and 26 in the safety analysis. At 24 weeks, HbA1c, the three-month blood sugar test, had fallen by 1·39% in the 4 mg escalation group and 1·30% in the 4 mg group, and by 1·99% and 1·88% in the slow and fast 8 mg groups. Read that carefully. The pace of the climb barely moved the blood sugar result in those participants, though arms that small cannot rule out a modest difference. The reason trials climb is the gut rather than the sugar, which is set out on the GLP-3 safety page, and the abstract behind this page does not split the side effects by escalation arm, so that half of the question stays open.

Question 7

Did a larger amount do more?

Accepted Answer

On blood sugar, yes, up to a point. In the phase 2 trial, HbA1c fell by 0·43% at 24 weeks in participants on the 0·5 mg arm and by 2·02% on the 12 mg arm, against 0·01% on placebo. The dulaglutide comparison arm, at 1·5 mg weekly, brought HbA1c down by 1·41%, almost exactly where the 4 mg retatrutide arms landed. Cotadutide showed the same direction in its kidney study, where greater changes in the measured responses were observed with higher cotadutide doses. So bigger did more, and the steps flattened at the top: the gain from 8 to 12 mg was small beside the gain from 0·5 to 4 mg. That is a curve measured on blood sugar in people with diabetes at 24 weeks, which is a different thing from a reason to reach for the top of it.

Question 8

Why is a trial amount not a protocol?

Accepted Answer

Because it was chosen to answer a question rather than to suit a person. In the phase 2 retatrutide trial, amounts were handed out by an interactive web-response system, with stratification for baseline HbA1c and BMI, which means a computer assigned them and nobody tailored them. Stratification only keeps the groups balanced. Nobody's amount was raised or lowered according to how they were doing, because adjusting it would have spoiled the comparison the trial existed to make. The authors then used the results the way a study amount is meant to be used, writing that these phase 2 data also informed dose selection for the phase 3 programme. A trial amount is a setting held fixed so that a result can be read cleanly. Turning it into a schedule for one person drops the randomisation, the monitoring and the known material that gave the number its meaning.

Question 9

How many doses are in a vial labelled GLP-3?

Accepted Answer

That arithmetic needs two numbers, the amount in the vial and the amount per injection, and the label supplies neither with any certainty. The per-injection figure depends entirely on the molecule. Cotadutide's smallest tested amount was 100 μg, a tenth of a milligram. Retatrutide's largest was 12 mg a week. Take a vial printed with 10 mg. Read as cotadutide at that smallest amount it would hold a hundred injections, and read as retatrutide at its largest it would hold less than one. The same vial gives two answers that differ more than a hundredfold, and nothing on the label settles which applies. The reconstitution arithmetic itself is simple and is worked through on the reconstitution calculator, which takes your numbers and suggests none of its own. The calculation is only as good as the molecule name and the stated strength going into it, and for a vial sold as GLP-3 both are unverified rather than known.

Question 10

Is there a cycle or a break in the published work?

Accepted Answer

No trial among these sources tested an on-and-off pattern. Each one ran its arms continuously to a fixed endpoint and then stopped, which is a design choice rather than a finding about breaks. The lengths vary widely. The earliest dual-agonist work gave daily doses over 21 or 28 days, the retatrutide phase 2 trial measured outcomes at 24 and 36 weeks, and the phase 3 trial ran 40. The survodutide trial in Japan runs longest of all, with its main endpoints set at week 76. A fixed run that stops is not a cycle, and what happens after it stops is not reported in the abstracts behind this page. So a break has not been measured in either direction, which leaves it unknown rather than harmless or harmful.

Question 11

Does an approved GLP-1 dose apply to a GLP-3 vial?

Accepted Answer

No. The approved weight-loss labels belong to two named molecules, and a 2025 review describes liraglutide and semaglutide as the GLP-1 agonists approved by the FDA for long-term weight management in adults with obesity. Those labels carry their own amounts, with liraglutide at 3·0 mg daily and semaglutide at 2·4 mg once weekly in a 2025 review of the field. Equal effect does not mean equal milligrams either. A 2022 review found the effects of cotadutide equal those of liraglutide 1·8 mg, though cotadutide was tested in micrograms in the trials behind this page. A label is a regulatory fact about one product, and it does not stretch to a different molecule or to an unnamed one. A vial sold as GLP-3 has no label of its own, and borrowing a figure from someone else's is a guess carrying a citation rather than an amount anyone approved.

Retatrutide amount	2023 phase 2	2026 phase 3
0·5 mg	Assigned	Dropped
4 mg	Assigned	Assigned
8 mg	Assigned	Not tested
9 mg	Not tested	Assigned
12 mg	Assigned	Assigned

GLP-3 protocols in the published literature