Retatrutide Dose in Type 2 Diabetes: What 0.5 to 12 mg Did to HbA1c and Weight

Why the diabetes doses are a separate question

Retatrutide was developed for type 2 diabetes and obesity in parallel, and the two populations were given their own trials from phase 1b onward. The diabetes trials use a different primary endpoint (HbA1c rather than weight), include an active comparator (dulaglutide), enrol people with a lower average BMI, and add a 0.5 mg dose that the obesity trial did not test.

So while the ladder is the same, the numbers each rung produced are specific to this population and should be read separately from the obesity results on the results page.

Phase 2: 0.5 to 12 mg in 281 adults

The Lancet 2023 trial (Rosenstock and colleagues) enrolled adults aged 18 to 75 with type 2 diabetes, HbA1c 7.0 to 10.5% and BMI 25 to 50, treated with diet and exercise alone or a stable dose of metformin of at least 1,000 mg a day.

They were randomised in a 2:2:2:1:1:1:1:2 ratio to placebo, dulaglutide 1.5 mg, or retatrutide maintenance doses of 0.5 mg, 4 mg (2 mg start), 4 mg (no escalation), 8 mg (2 mg start), 8 mg (4 mg start) or 12 mg (2 mg start). The primary endpoint was the change in HbA1c at 24 weeks.

Every dose from 4 mg up beat placebo. Against dulaglutide 1.5 mg, the 4 mg groups landed level with it (-1.39 and -1.30 versus -1.41), and the 8 mg slow-escalation and 12 mg groups beat it with statistical significance (p=0.0019 and p=0.0002). The findings were consistent at 36 weeks. In HbA1c terms, 8 mg and 12 mg were within 0.03 points of each other, an early sign of the plateau above 8 mg that shows up across the program.

Weight in the diabetes trial

Weight at 36 weeks fell in a dose-dependent way: 3.19% at 0.5 mg, 7.92% in the 4 mg escalation group, 10.37% in the 4 mg no-escalation group, 16.81% and 16.34% in the two 8 mg groups, and 16.94% at 12 mg, against 3.00% with placebo and 2.02% with dulaglutide.

For doses of 4 mg and above the difference from both comparators was statistically significant. These are smaller losses than the obesity trial produced at the same doses over 48 weeks (-17.1% at 4 mg, -24.2% at 12 mg), a gap that is usual for incretin drugs in people with type 2 diabetes and that the shorter duration only partly explains.

Phase 3: 4, 9 and 12 mg in TRANSCEND-T2D-1

The first phase 3 retatrutide paper, published in The Lancet in June 2026 (Bajaj and colleagues), enrolled 537 adults at 48 sites in the USA, Mexico and India with type 2 diabetes inadequately controlled by diet and exercise alone, HbA1c 7.0 to 9.5% and BMI of at least 23. They were randomised 1:1:1:1 to 4, 9 or 12 mg once weekly or placebo for 40 weeks. Mean baseline HbA1c was 7.9%, mean BMI 35.8, mean diabetes duration 2.5 years.

HbA1c fell by 1.69 points at 4 mg, 1.86 at 9 mg and 1.94 at 12 mg, against 0.81 with placebo, giving placebo-adjusted reductions of 0.88, 1.04 and 1.12 points (all p<0.0001). Weight fell 11.5%, 13.9% and 15.3%, against 2.6% with placebo. Discontinuation of the study drug for adverse events was 2 to 5% on retatrutide and 0% on placebo, and there was no severe hypoglycaemia. Two deaths occurred, both in the 4 mg group and both judged unrelated to the drug.

Phase 2 and phase 3 side by side

The phase 3 placebo group improved more than the phase 2 one (HbA1c -0.81 versus -0.01), which is common when a trial enrols people with shorter diabetes duration and no background medication; the placebo-adjusted effect is the fairer comparison, and at 12 mg it is 1.12 points in phase 3 against about 2.0 in phase 2. The phase 3 escalation was also different in design, a change discussed in the titration guide.

Fat mass and lean mass at each dose

A DXA substudy of the phase 2 trial, published in The Lancet Diabetes and Endocrinology in 2025, measured body composition at 36 weeks. Total fat mass fell 4.9% at 0.5 mg, 15.2% at 4 mg (pooled), 26.1% at 8 mg (pooled) and 23.2% at 12 mg, against 2.6% with dulaglutide and 4.5% with placebo.

The authors report that the proportion of lean mass lost relative to total weight lost was similar to other obesity treatments. The 8 mg and 12 mg fat-mass figures overlapping is one more instance of the plateau above 8 mg in this population.

Hypoglycaemia and tolerability by dose

Neither published diabetes trial reported severe hypoglycaemia, and neither enrolled people on insulin or sulfonylureas, which is the context that finding belongs in. Tolerability followed dose and escalation speed: in phase 2, mild-to-moderate gastrointestinal events were reported by 35% of retatrutide participants overall, from 13% at 0.5 mg to 50% in the 8 mg fast-escalation group, against 13% on placebo and 35% on dulaglutide.

In phase 3, gastrointestinal events were again the most frequent, mild to moderate, and subsided over time. The complete dose ladder across all indications is on the retatrutide dosage guide; the phase 3 program for people with type 2 diabetes and obesity, TRIUMPH-2, is described in the TRIUMPH and TRANSCEND guide.

Back to the retatrutide dosage guide