Tirzepatide vs Semaglutide for Diabetes: SURPASS-2
Before they were weight-loss rivals, these two drugs were diabetes drugs — and one trial put them head-to-head for blood sugar. Here is what it found, and the dose asterisk it still carries.
Abstract: SURPASS-2 is the only large head-to-head trial of tirzepatide versus semaglutide in type 2 diabetes, and tirzepatide lowered A1C and body weight further at every dose. This guide breaks down the glycemic numbers, the weight numbers, the GIP-receptor mechanism, and the single biggest caveat — that semaglutide was tested at its 1 mg diabetes dose, not the 2.0 mg dose later approved.
Most people now meet tirzepatide vs semaglutide as a weight-loss question. But both molecules were diabetes drugs first, and the cleanest comparison of them for blood-sugar control is older than the obesity debate: SURPASS-2, a Phase 3 trial that randomized 1,879 adults with type 2 diabetes to one drug or the other and measured A1C as the primary endpoint. It remains the only large, direct, same-trial comparison of these two for diabetes — which makes it far more decision-relevant than lining up their separate trials, and worth reading carefully, including the part of the result that gets quietly dropped from the headlines.
This is an educational, research-focused breakdown, not medical advice. Tirzepatide, semaglutide and retatrutide are sold on this site for laboratory research use only; the branded medicines (Mounjaro, Ozempic) are prescription pharmaceuticals referenced here only to describe the published evidence.
Tirzepatide vs semaglutide for diabetes: why a head-to-head matters
For a diabetes decision, the question is not “which drug posts a bigger number in its own trial” but “which drug does more in the same patients, measured the same way.” Cross-trial comparison is unreliable because populations, background therapy, trial era and statistical methods all move the result independently of the molecule. A head-to-head cancels those out: identical patients, identical conditions, two drugs, one readout. Whatever gap remains is attributable to the drugs themselves.
SURPASS-2 is that trial for the tirzepatide vs semaglutide diabetes question. It is the diabetes counterpart to the obesity comparison we cover separately in the SURMOUNT-5 head-to-head — and reading the two together is more informative than either alone, because they were run in different populations, at different doses, with different primary endpoints.
Inside SURPASS-2: the design
SURPASS-2 was a 40-week, randomized, open-label, Phase 3 trial that added the study drug on top of metformin in adults whose type 2 diabetes was inadequately controlled.
- Who: 1,879 adults with type 2 diabetes (mean baseline A1C ~8.3%, mean body weight ~94 kg, mean diabetes duration ~8.6 years), all already on metformin.
- What: once-weekly subcutaneous tirzepatide at 5 mg, 10 mg, or 15 mg versus once-weekly subcutaneous semaglutide 1 mg, randomized across the four arms.
- How long: 40 weeks.
- Primary endpoint: change in A1C from baseline. Body weight was a key secondary endpoint.
Two design facts shape how you read the result. First, the primary endpoint here is glycemic control, not weight — this is a diabetes trial, and A1C is what it was powered to answer. Second, the semaglutide comparator was 1 mg, the maximum approved Ozempic dose for type 2 diabetes at the time the trial was designed. Hold onto that number; it is the crux of the fairness debate later.
The glycemic result: tirzepatide lowered A1C further
Tirzepatide reduced A1C more than semaglutide at all three doses, with a clear dose-response. At the top dose, the gap was about half a percentage point of A1C — a meaningful margin in diabetes, where guideline targets are measured in tenths.
| Outcome at 40 weeks | Tirzepatide 5 mg | Tirzepatide 15 mg | Semaglutide 1 mg |
|---|---|---|---|
| Mean A1C reduction | −2.09% | −2.46% | −1.86% |
| Mean body-weight change | −7.8 kg (−8.5%) | −12.4 kg (−13.1%) | −6.2 kg (−6.7%) |
| A1C ≤6.5% + ≥10% weight loss, no severe hypoglycemia | — | ~60% | ~22% |
| Reached A1C below 5.7% (non-diabetic range) | — | ~51% | ~20% |
The 10 mg arm landed between the 5 mg and 15 mg results on both A1C and weight, as the dose-response would predict. The most striking figure is the composite: at the 15 mg dose roughly 60% of participants hit a stringent target of A1C ≤6.5% plus at least 10% weight loss with no severe hypoglycemia, versus about 22% on semaglutide — and around half the 15 mg group pushed A1C below 5.7%, the threshold normally seen in people without diabetes. That is a glycemic response unusual for a non-insulin therapy.
The weight result: a secondary endpoint, but a large gap
Although weight was secondary here, the separation echoed what later showed up in the obesity trials. Tirzepatide 15 mg produced about −12.4 kg (−13.1%) versus −6.2 kg (−6.7%) for semaglutide 1 mg — roughly double the body-weight loss. For people whose diabetes is entangled with obesity, that secondary endpoint is often the one that drives the real-world choice, which is exactly why these molecules migrated into the weight-loss arena. For the broader ranking of where each agent sits on pure weight loss, see our GLP-1 weight-loss class ranking.
On tolerability, the trial was reassuring and unsurprising: the dominant adverse events in every arm were mild-to-moderate gastrointestinal effects (nausea, diarrhea, vomiting) concentrated during dose escalation, at broadly similar rates across tirzepatide and semaglutide. Discontinuations for adverse events were modestly higher on the upper tirzepatide doses (around 8%) than on semaglutide (around 4%) — a small tolerability tax for the larger effect.
The dose question: was semaglutide underdosed?
This is the caveat that separates an honest reading from a marketing one. SURPASS-2 compared tirzepatide’s full dose range (up to 15 mg) against a single, lower semaglutide dose of 1 mg. After SURPASS-2 was designed, a 2.0 mg dose of semaglutide was studied and approved for type 2 diabetes — the SUSTAIN FORTE trial showed 2.0 mg lowered A1C and weight more than 1.0 mg. So the head-to-head did not pit tirzepatide against semaglutide’s highest available diabetes dose; it beat the 1 mg dose, and the 2.0 mg comparison has never been run head-to-head.
That does not erase the result — tirzepatide 15 mg’s margin over semaglutide 1 mg is large enough that closing the dose gap would likely narrow, not reverse it — but it is a real asterisk, and anyone quoting SURPASS-2 should quote it with the dose attached.
It is worth noting this asterisk runs the opposite direction from the obesity trial. In SURPASS-2, semaglutide was arguably under-dosed (1 mg). In SURMOUNT-5, the obesity head-to-head fixed exactly that by running semaglutide at its full 2.4 mg obesity dose against tirzepatide 15 mg — and tirzepatide still won, 20.2% vs 13.7% on weight. Read together, the two trials make a more durable case than either does alone: tirzepatide came out ahead both when semaglutide was at a lower dose (diabetes) and when it was at its maximum (obesity).
Why the GIP arm helps in diabetes too
The mechanism behind the gap is the same one that explains tirzepatide’s obesity edge. Semaglutide is a single-receptor agonist — it activates the GLP-1 receptor only. Tirzepatide is a dual agonist, hitting GLP-1 and GIP (glucose-dependent insulinotropic polypeptide). Both drugs share the core GLP-1 actions that help diabetes: glucose-dependent insulin secretion, suppressed glucagon, slowed gastric emptying and reduced appetite. Adding the GIP arm appears to amplify the insulinotropic and weight effects rather than simply duplicating the GLP-1 signal, which is the leading explanation for why tirzepatide reached both lower A1C and greater weight loss in the same trial. For the full account of how these gut-brain-pancreas circuits work, see how GLP-1 works.
This stacking logic also points up the class ladder: adding still more receptor arms tends to raise the ceiling further, which is why the triple agonist retatrutide — GLP-1 + GIP + glucagon — posts even larger numbers than tirzepatide in its own trials, a comparison we cover in retatrutide vs tirzepatide.
What SURPASS-2 does not settle
A clean win on A1C and weight is not a verdict on everything that matters in a diabetes decision.
Hard outcomes, not just lab values. SURPASS-2 measured A1C and weight — not heart attacks, strokes, kidney decline or deaths. This is where semaglutide still leads the class: dedicated cardiovascular and kidney outcome trials have shown injectable semaglutide reduces those events, evidence tirzepatide’s own outcome programs are still maturing. Lowers A1C more and has more proof it prevents complications are different claims that currently point at different drugs.
The 2.0 mg gap. As above, the strongest semaglutide diabetes dose was never in this trial. Until a tirzepatide-vs-semaglutide-2.0 mg head-to-head exists, the precise size of the glycemic gap at maximum doses is an estimate, not a measurement.
Open-label and industry-funded. Participants and investigators knew which drug was given, and the trial was funded by tirzepatide’s manufacturer. A pre-specified, objectively measured endpoint like A1C resists expectation bias more than a subjective one, but both facts are honest caveats to carry.
Estimands and durability. Trials report effects two ways — across everyone randomized, or only in those who stayed on the drug — and the headline numbers shift depending on which is quoted, which is part of why different sources cite slightly different SURPASS-2 figures. The trial also ran 40 weeks; it does not tell you how the gap holds at three or five years.
Diabetes vs obesity: how to read the two head-to-heads together
The honest summary is narrow and strong. For type 2 diabetes, SURPASS-2 shows tirzepatide delivered greater A1C reduction and greater weight loss than semaglutide 1 mg in the same trial — but at a semaglutide dose that has since been surpassed. For obesity, SURMOUNT-5 shows tirzepatide still wins at semaglutide’s maximum dose. And for proven prevention of cardiovascular and kidney events, semaglutide still owns the deepest evidence base.
Put plainly:
- If the priority is glycemic control and weight together in type 2 diabetes: tirzepatide posted the larger effect on both in the only diabetes head-to-head.
- If the priority is the strongest proof a drug prevents downstream complications: injectable semaglutide still leads on hard outcomes, which can outweigh a half-point of A1C for higher-risk individuals.
- If raw magnitude is the whole question and investigational agents are on the table: the ceiling is higher one tier up in the triple agonist retatrutide vs tirzepatide — not approved, and with no head-to-head of its own yet.
SURPASS-2 settled which lowers A1C more in this trial. It did not settle which is right for whom — and conflating those two is the most common mistake in the coverage.
FAQ
Is tirzepatide better than semaglutide for type 2 diabetes? On the trial’s endpoints, yes. SURPASS-2 — the only large head-to-head in diabetes — found tirzepatide lowered A1C more at every dose (about −2.46% vs −1.86% at the top dose) and produced roughly double the weight loss. The important caveat is that semaglutide was tested at 1 mg, not the 2.0 mg diabetes dose approved later.
By how much did tirzepatide beat semaglutide on A1C? At the 15 mg dose, about half a percentage point of A1C (−2.46% vs −1.86%). Around 60% of the 15 mg group hit a stringent composite of A1C ≤6.5% plus ≥10% weight loss with no severe hypoglycemia, versus about 22% on semaglutide.
Was semaglutide underdosed in SURPASS-2? Partly. It used semaglutide 1 mg, the maximum approved diabetes dose when the trial was designed. A 2.0 mg dose was later approved (SUSTAIN FORTE) and gives more A1C and weight reduction, but it has never been compared head-to-head against tirzepatide. The 15 mg margin is large enough that the higher dose would likely narrow rather than reverse it.
Why does tirzepatide work better in diabetes? Tirzepatide activates two incretin receptors (GLP-1 and GIP); semaglutide activates only GLP-1. The added GIP arm appears to amplify insulin secretion and weight loss rather than just duplicating the GLP-1 effect.
Does this mean semaglutide is the worse diabetes drug overall? No. Semaglutide loses on A1C and weight magnitude in this trial but still leads the class in proven cardiovascular and kidney outcome reductions. “More glucose lowering” and “more outcome proof” currently describe different drugs.
How is this different from the SURMOUNT-5 trial? SURPASS-2 is the diabetes head-to-head (A1C primary, semaglutide 1 mg, on metformin). SURMOUNT-5 is the obesity head-to-head (weight primary, semaglutide 2.4 mg, no diabetes). Tirzepatide won both.
Are these the same as the research peptides sold online? Tirzepatide, semaglutide and retatrutide are sold for laboratory research use only and are not approved for human use in that form. The branded medicines (Mounjaro, Ozempic) are prescription pharmaceuticals. This article describes published clinical evidence, not a protocol for personal use.
References
- Frías JP, Davies MJ, Rosenstock J, et al. Tirzepatide versus Semaglutide Once Weekly in Patients with Type 2 Diabetes (SURPASS-2). N Engl J Med. 2021;385(6):503-515. doi:10.1056/NEJMoa2107519 · NCT03987919.
- Frías JP, Auerbach P, Bajaj HS, et al. Efficacy and safety of once-weekly semaglutide 2.0 mg versus 1.0 mg in patients with type 2 diabetes (SUSTAIN FORTE): a double-blind, randomised, phase 3B trial. Lancet Diabetes Endocrinol. 2021;9(9):563-574. doi:10.1016/S2213-8587(21)00174-1.
- Aronne LJ, Horn DB, le Roux CW, et al. Tirzepatide as Compared with Semaglutide for the Treatment of Obesity (SURMOUNT-5). N Engl J Med. 2025. doi:10.1056/NEJMoa2410819 · NCT05822830.
Research and educational use only. This article summarizes published clinical research and is not medical advice, a treatment recommendation, or a substitute for a licensed clinician. SURPASS-2 was an open-label, manufacturer-funded trial, and the figures cited come from that and other industry-sponsored studies. Tirzepatide, semaglutide and retatrutide referenced here are sold for laboratory research purposes only and are not intended to diagnose, treat, cure, or prevent any disease.