The SURMOUNT-2 results answer a narrow but useful question. After 72 weeks, average weight change was -12.8% with tirzepatide 10 mg and -14.7% with 15 mg, compared with -3.2% on placebo. The 938 participants all had type 2 diabetes as well as obesity or overweight. These are group averages from a controlled trial, not a forecast for one person.
What question did the SURMOUNT-2 trial ask?
SURMOUNT-2 asked whether tirzepatide plus lifestyle support would produce a larger change in body weight than lifestyle support plus placebo in adults with type 2 diabetes. A placebo is a look-alike injection without tirzepatide.
It was a phase 3 trial, a large late-stage study designed to confirm benefits and track harms. Assignment happened by chance, and participants and investigators did not know who received which injection. Eli Lilly funded the study. The Lancet paper by Garvey and colleagues appeared in 2023, and the completed ClinicalTrials.gov record is NCT04657003.
Who was enrolled, and for how long?
The trial enrolled 938 adults across seven countries. Their average age was 54.2 years, 51% were women, and their average starting weight was 100.7 kg, or about 222 pounds. Average HbA1c was 8.02%; HbA1c is a blood test that reflects roughly three months of blood sugar.
Everyone had a body mass index, or BMI, of at least 27 and an HbA1c between 7% and 10%. Diabetes treatment had to be stable for at least three months. The study excluded people with type 1 diabetes, recent large weight changes, a history of pancreatitis, or certain serious eye, thyroid, and psychiatric conditions.
Participants were assigned to 10 mg, 15 mg, or placebo for 72 weeks. Doses increased on a fixed study schedule, and every group received guidance built around a calorie deficit and physical activity. Those assigned doses describe the experiment; they are not dosing instructions. A prescriber sets the actual dose.
How did the study measure the result?
The two primary endpoints, meaning the main outcomes chosen before the trial began, were average percentage change in weight and the share of people who lost at least 5%.
The headline analysis used a treatment-regimen estimand, a statistical approach meant to count outcomes regardless of whether someone stopped the assigned treatment or needed rescue diabetes medication. That makes the result more practical than a calculation limited to people who stayed on every dose.
Weight, HbA1c, side effects, and treatment discontinuations were all tracked. The comparison was against placebo, not semaglutide and not usual care outside a trial.
SURMOUNT-2 results at week 72
| Assigned group | Average weight change | Lost at least 5% | Lost at least 10% | Lost at least 15% |
|---|---|---|---|---|
| Tirzepatide 10 mg | -12.8% | 79% | 61% | 40% |
| Tirzepatide 15 mg | -14.7% | 83% | 65% | 48% |
| Placebo | -3.2% | 32% | 9% | 3% |
Put plainly, about 4 in 5 people assigned tirzepatide lost at least 5% of their starting weight, compared with about 1 in 3 on placebo. Nearly half of the 15 mg group lost at least 15%. The averages still hide a wide range of individual outcomes.
Blood sugar changed too. From a starting HbA1c near 8.0%, the average drop was about 2.1 percentage points in both tirzepatide groups and about 0.5 points with placebo. The separate guide to tirzepatide for type 2 diabetes puts that glucose result beside the wider diabetes trial program.
Side effects reported in the paper
Stomach and bowel problems were the most common side effects. Most were mild or moderate.
| Event | 10 mg | 15 mg | Placebo |
|---|---|---|---|
| Nausea | 20.2% | 21.9% | 6.3% |
| Diarrhea | 19.9% | 21.5% | 8.9% |
| Vomiting | 10.9% | 13.2% | 3.2% |
| Constipation | 8.0% | 9.0% | 4.1% |
| Stopped because of an adverse event | 3.8% | 7.4% | 3.8% |
Low blood sugar below 54 mg/dL occurred in 4% of the 10 mg group, 5% of the 15 mg group, and 1% of the placebo group. Background diabetes medication matters here: insulin and sulfonylureas can change that risk, which is one reason trial averages cannot replace a medication review.
Why the diabetes group lost less than SURMOUNT-1
The difference is real, but the comparison needs care. SURMOUNT-1 results showed average weight change of -19.5% and -20.9% with the same 10 mg and 15 mg assignments in adults without diabetes. SURMOUNT-2 reported -12.8% and -14.7% in adults with diabetes. These were separate trials, not a head-to-head test of diabetes itself.
A 2025 analysis of SURMOUNT-2 found that lower starting HbA1c was associated with greater weight reduction. That helps explain variation inside the diabetes group, but it was an after-the-fact analysis and cannot prove why one person responds differently from another (Sattar et al., Diabetes Care 2025).
As of September 18, 2026, the day this article was written, a June 25, 2026 post hoc analysis had added one practical caution. Among 609 tirzepatide-treated SURMOUNT-2 participants with weight data at week 8, those below 5% early weight reduction still averaged -10.8% at week 72, versus -20.0% among early responders. The study grouped people after the fact, excluded placebo, and included only people with the required measurements, so week 8 was not tested as a stop-or-continue rule (Kokkinos et al., Diabetes, Obesity and Metabolism 2026).
SURMOUNT-2 also did not compare tirzepatide with semaglutide. A direct comparison requires a different trial and population, not two numbers lifted from separate studies.
What SURMOUNT-2 cannot say about a compounded version
SURMOUNT-2 studied the manufacturer's tirzepatide under a fixed research protocol. It did not test Promise's compounded formulation, and its averages do not establish an equivalent result for a compounded preparation.
Through Promise, tirzepatide is dispensed as a compounded medication, which is different from an FDA-approved product: the formulation offered here is not FDA-approved. The FDA explains that it does not review a compounded drug for safety, effectiveness, or quality before marketing.
A licensed provider may still prescribe a compounded formulation; that decision is between the patient and the doctor. At Promise, a licensed provider reviews every request, and not everyone qualifies.
The useful way to read this trial
Keep the population, time, and comparison attached to the headline number: adults with type 2 diabetes, 72 weeks, fixed trial doses, lifestyle support, and placebo. The study supports a conversation about expected ranges and tradeoffs. It cannot name one person's result or supply a self-directed plan.