Same drug class, same weekly injection, same diabetes diagnosis. 1,879 people, randomized head-to-head — one arm pulled ahead on almost every measure.
SURPASS-2 was a 40-week, randomized, open-label, active-comparator trial. It enrolled 1,879 adults with type 2 diabetes who were already on metformin but not adequately controlled. Average baseline HbA1c was 8.28%, and average baseline weight was about 93.7kg (roughly 207 lb).
Participants were randomized to one of four weekly injections: tirzepatide (Mounjaro) at 5mg, 10mg, or 15mg, or semaglutide (Ozempic) at 1mg — the highest dose approved for diabetes at the time. Everyone stayed on their background metformin. Because the trial was open-label, both patients and investigators knew which drug they were on, which is worth keeping in mind when reading patient-reported outcomes.
This is a diabetes trial, not a weight-loss trial. The primary endpoint was change in HbA1c at 40 weeks. Weight change was a secondary endpoint, tracked because both drugs are known to lower body weight as a side effect of glycemic control. The trial is formally titled "Tirzepatide versus Semaglutide Once Weekly in Patients with Type 2 Diabetes" and ran under ClinicalTrials.gov identifier NCT03987919, sponsored by Eli Lilly.
The comparator matters here too. Semaglutide 1mg was, at the time of the trial, the top-of-label dose for Ozempic — not an artificially weak comparison. This was tirzepatide against semaglutide's best approved diabetes dose, not a starter dose stacked against a maximum one.
Every tirzepatide dose beat semaglutide 1mg on HbA1c, and the gap widened with dose. Even the lowest tirzepatide dose (5mg) edged out semaglutide's top dose, and the difference was statistically significant across all three comparisons — this was the trial's primary result, and it was unambiguous.
The weight difference was larger than the glucose difference. At the highest dose, tirzepatide produced almost twice the weight loss of semaglutide 1mg. That is a substantial gap for a trial not designed around weight as its main outcome, and it is the data point most often cited when tirzepatide is discussed as a weight-management option under its own brand, Zepbound.
Tirzepatide also worked faster. People on the two higher tirzepatide doses reached a 5% weight-loss milestone in about half the time it took people on semaglutide. Reaching a glycemic target sooner isn't the same as a better long-term outcome, but it does mean fewer weeks of elevated blood sugar along the way.
Side effects were similar in kind, not identical in degree. Gastrointestinal symptoms — nausea, diarrhea, vomiting — were the dominant adverse events in both arms and mostly mild to moderate. Discontinuation due to GI side effects ran 3% to 7% across the tirzepatide doses versus 3% on semaglutide, and serious adverse events were reported in 5% to 7% of the tirzepatide group versus 3% on semaglutide.
Over 40 weeks in adults with type 2 diabetes on metformin, every dose of tirzepatide lowered HbA1c more than semaglutide 1mg, and the highest tirzepatide dose produced roughly twice the weight loss — 13.1% versus 6.7% — with a similar, slightly higher, rate of GI side effects.
Sources: Frías JP, Davies MJ, Rosenstock J, et al. Tirzepatide versus Semaglutide Once Weekly in Patients with Type 2 Diabetes. New England Journal of Medicine, Aug 5 2021;385:503-515 (doi:10.1056/NEJMoa2107519) · Eli Lilly, SURPASS-2 topline results release, June 2021. NEJM full text · ClinicalTrials.gov NCT03987919.
A head-to-head trial is rarer than it sounds. Most drugs get compared to placebo, not to each other. SURPASS-2 is one of the few large randomized trials where two GLP-1-class drugs were put in the same room, so to speak, under the same protocol. That makes its comparative numbers more trustworthy than cross-trial comparisons pieced together from separate studies.
Dose matters when comparing drugs across classes. This trial compared tirzepatide against semaglutide's diabetes ceiling dose of 1mg — not the 2mg dose later approved, and not the higher doses used for weight management under the Wegovy brand. A head-to-head trial with newer dosing might narrow, or fail to reproduce, this gap.
More weight loss and faster results also mean a steeper ramp. The doses that outperformed most also carried the higher discontinuation-for-side-effects rate. If your prescriber has you titrating up, tracking which week symptoms spiked can help distinguish a rough patch that settles from one that means the current dose isn't tolerable.
This is a diabetes population, not a weight-loss-only population. Everyone enrolled had type 2 diabetes and was already on metformin. Mounjaro is the tirzepatide brand approved for type 2 diabetes; these results describe that indication and shouldn't be read as a direct stand-in for trials in people without diabetes.
This article summarizes a published, peer-reviewed clinical 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.