Retatrutide dosage: the doses the trials actually used, and what is not known

By Retatrutide Risk editorial team. Published September 21, 2026. Sources at the end of the page.

Search interest in "retatrutide dosage" exceeds interest in its side effects, which says something about how the compound is being used outside trials. This page does one thing: it reports the doses and escalation schedules that were administered in registered clinical trials, with the outcome each produced, and it states plainly what the trials have not answered. It is not a dosing guide, and it should not be read as one.

The unit and the route

Every retatrutide trial has used a once-weekly subcutaneous injection. Doses are expressed in milligrams of peptide per weekly injection. The phase 1b trial established that the pharmacokinetics are dose proportional and that the half-life is approximately 6 days, which is what makes once-weekly dosing possible: after a weekly dose, roughly half of the peptide is still present when the next dose is given, so the concentration builds over the first weeks and then plateaus.

Phase 2 obesity trial: seven arms, five maintenance doses

The phase 2 obesity trial (Jastreboff et al., NEJM 2023; ClinicalTrials.gov NCT04881760) randomized 338 adults in a 2:1:1:1:1:2:2 ratio to placebo or one of six retatrutide regimens, for 48 weeks. The registry record describes each arm's escalation.

Figures are least-squares mean percentage changes in body weight from the published abstract. The abstract reports the two 4 mg arms and the two 8 mg arms combined, so the effect of the starting dose on weight cannot be read from the abstract alone. What the abstract does say is that gastrointestinal events were partially mitigated with a lower starting dose (2 mg versus 4 mg).

Phase 2 type 2 diabetes trial: a 0.5 mg arm and an active comparator

The type 2 diabetes trial (Rosenstock et al., Lancet 2023; NCT04867785) used a similar design over 36 weeks in 281 adults, with two differences: a 0.5 mg arm replaced the 1 mg arm, and dulaglutide 1.5 mg was included as an active comparator alongside placebo.

Two observations follow from this table. The 8 mg slow and fast escalation arms ended at nearly the same weight and HbA1c, so the slower schedule cost nothing in efficacy by 36 weeks. And the 8 mg and 12 mg arms produced almost identical results in this population, which raises the question of whether 12 mg adds anything in type 2 diabetes; the authors note that these data informed dose selection for phase 3.

Phase 3: the doses were changed

The TRIUMPH program did not carry the phase 2 doses forward unchanged. The registry records list the arms as "Dose 1", "Dose 2" and "Dose 3" without milligram values, but the company announcements of the results name them: 4 mg, 9 mg and 12 mg in TRIUMPH-1 and TRIUMPH-2, and 9 mg and 12 mg in TRIUMPH-3 and TRIUMPH-4. The 8 mg dose from phase 2 became 9 mg, and the 1 mg and 0.5 mg doses were dropped.

The escalation schedules used to reach these phase 3 doses have not been published in a peer-reviewed paper. The registry records and announcements do not state the starting dose or the interval between steps. Anyone claiming to know the "phase 3 titration schedule" is inferring it.

What "escalation" means and why it exists

Every phase 2 arm above 1 mg reached its maintenance dose through steps. The pattern in the registry is consistent: 2 mg, then 4 mg, then 8 mg, then 12 mg. The purpose is tolerability. Gastrointestinal side effects are strongest when the concentration rises, and they partly fade as the body adapts. The phase 2 obesity trial demonstrated this directly by comparing a 2 mg and a 4 mg start to the same maintenance dose.

What the published record does not include is the duration of each step. The abstracts give the sequence, not the calendar. Statements such as "four weeks per step" appear on forums and vendor pages but do not appear in any of the sources cited on this page.

Why a forum "dosing chart" is not a trial protocol

Forum dosing charts typically present a weekly milligram figure per week number, sometimes in fractions of a milligram, sometimes with a "cycle" length. They differ from trial protocols in ways that matter:

• Trial doses were fixed by a protocol approved by an ethics committee, administered with pharmaceutical-grade product of known concentration, and adjusted only within the protocol's rules.
• Trials excluded people with a range of conditions and monitored participants at scheduled visits with laboratory tests and electrocardiograms. A chart carries none of that.
• Trial participants received the compound from the manufacturer. A chart applied to a vial of unknown purity and concentration is not the same dose, even if the number matches. The reconstitution article explains how concentration errors arise.
• No trial has tested "cycles". Every trial administered the drug continuously for its full duration. What happens to weight after stopping retatrutide has not been published.
• The phase 1 single-dose study found that weight reduction persisted up to day 43 after a single dose, which shows that effects outlast any individual injection; it does not validate intermittent dosing.

Dose questions the trials have not answered

• The lowest effective maintenance dose for a given person. The trials tested groups at fixed doses; they did not titrate to individual response.
• Whether doses below 4 mg are worthwhile. The 1 mg arm produced -8.7% at 48 weeks, but 1 mg was dropped in phase 3.
• Whether 12 mg is better than 9 mg for most people. In TRIUMPH-1 the difference at 80 weeks was 2.4 percentage points of body weight, against a higher rate of nausea, vomiting, dysesthesia and dropout.
• What happens when a dose is missed, or when treatment stops.
• Any dosing in adolescents, in people over 75, in people with kidney or liver impairment, or during pregnancy.

What an eventual label would add

If retatrutide is approved, its prescribing information would define the starting dose, the escalation interval, the maintenance doses, dose adjustments for side effects and the populations in which it should not be used. Until then, the only defensible statements about retatrutide dosage are the ones in the tables above, and they describe what happened to trial participants, not what anyone should do.

See also what the trials measured as results and the side effects at each dose.