938 adults. Type 2 diabetes and obesity, together in the same trial. Mounjaro's original approval was for glucose control — this is the study that measured both endpoints in one population.
Every earlier tirzepatide weight-loss trial had a catch: SURMOUNT-1 specifically excluded people with diabetes, so it couldn't say much about the population Mounjaro was originally approved to treat. SURMOUNT-2 closed that gap. It's the first tirzepatide trial designed with bodyweight change as a primary outcome in adults who actually have type 2 diabetes.
It was a phase 3, double-blind, randomized, multicenter, placebo-controlled trial run across sites in Argentina, Brazil, India, Japan, Russia, Taiwan, and the United States. Researchers enrolled 938 adults with a BMI of 27 or higher, type 2 diabetes, and an HbA1c between 7% and 10%, then randomized them 1:1:1 to once-weekly subcutaneous tirzepatide at 10mg, 15mg, or placebo for 72 weeks. Everyone in the trial, including the placebo group, also received a reduced-calorie diet and increased physical activity as background lifestyle intervention.
The co-primary endpoints were percent change in bodyweight and the proportion reaching at least 5% weight loss — the same weight-focused framework used in the non-diabetic SURMOUNT-1 trial, applied here to a population that also needed its blood sugar tracked as a key secondary outcome.
This matters for anyone using Mounjaro under its diabetes indication rather than Zepbound under its weight-management one: SURMOUNT-2 is the closest thing to a same-population comparison of what tirzepatide does to weight versus what it does in people without diabetes.
Weight loss was real but smaller than in people without diabetes. At the highest dose, participants lost 14.7% of body weight on average, versus 20.9% at the same 15mg dose in SURMOUNT-1's non-diabetic population. That gap isn't a red flag — it's a well-documented pattern across the entire GLP-1 and dual-agonist class, seen with semaglutide too. Type 2 diabetes tends to blunt the magnitude of drug-induced weight loss, likely tied to insulin resistance and beta-cell physiology, without changing the direction of the effect.
The glucose numbers were the more striking part of this particular trial. HbA1c fell by 2.1 percentage points on tirzepatide versus 0.5 on placebo, and nearly half of participants brought their A1c down into the non-diabetic range (under 5.7%) — something Garvey, the trial's principal investigator, called out as notable given how few participants on placebo achieved the same.
Beyond weight and A1c, tirzepatide groups saw larger drops in systolic blood pressure (−7.2 mmHg vs. −1.0) and triglycerides (−28.6% vs. −5.8%), and larger gains in HDL cholesterol, than placebo. Safety followed the same pattern as other trials in the program: gastrointestinal side effects — nausea, diarrhea, vomiting — were the most common adverse events, mostly mild to moderate, with fewer than 5% of participants stopping treatment because of them. No severe hypoglycemia signal was reported.
In adults with obesity and type 2 diabetes, tirzepatide produced 14.7% average weight loss and a 2.1-point HbA1c drop at 72 weeks — weight loss that ran smaller than in non-diabetic trials, but a glycemic effect strong enough to normalize A1c in nearly half of participants.
Sources: Garvey WT, Frias JP, Jastreboff AM, et al. Tirzepatide once weekly for the treatment of obesity in people with type 2 diabetes (SURMOUNT-2): a double-blind, randomised, multicentre, placebo-controlled, phase 3 trial. The Lancet 2023;402(10402):613-626 (doi:10.1016/S0140-6736(23)01200-X). Lancet abstract · NCT04657003.
This is the trial behind Mounjaro's original use case, not just Zepbound's. Mounjaro was approved for type 2 diabetes before tirzepatide was ever tested as a weight-loss drug on its own; SURMOUNT-2 is the study that shows people using it for that original indication also see substantial weight change, even if the average is a few points lower than in people without diabetes.
A smaller number on the scale doesn't mean the drug is working less hard. The same 15mg dose that produced 20.9% loss in SURMOUNT-1 produced 14.7% here. That's a difference in population physiology, not in dosing or adherence — a useful thing to know before assuming a plateau means something has gone wrong.
The A1c result is arguably the more clinically important number for this group. Bringing A1c under 5.7% — out of diabetic range entirely — happened in roughly half of tirzepatide-treated participants and almost none on placebo. For someone managing diabetes as the primary condition, that's the endpoint worth tracking alongside weight.
GI side effects showed up early and mostly settled. As in every other tirzepatide trial, nausea and diarrhea clustered during dose escalation. Logging symptoms against dose changes gives a prescriber something concrete to work with if a dose increase needs to be slowed rather than abandoned.
This article summarizes a published randomized controlled trial for general education. It is not medical advice. Individual results vary, and any decision about starting, switching, adjusting, or stopping any GLP-1 or GIP/GLP-1 medication should be made with a prescriber.