SURMOUNT-5: Semaglutide vs Tirzepatide Head-to-Head

Quick Answer

• Tirzepatide beat semaglutide: -20.2% vs -13.7% weight loss at 72 weeks
• 36.5% on tirzepatide lost ≥25% of body weight vs 18.6% on semaglutide
• GI side effects were similar; tirzepatide caused more injection-site reactions
• Trial was open-label and didn't test heart attack or stroke risk

For most of the GLP-1 era, patients and doctors have had to guess how tirzepatide (Zepbound, Mounjaro) stacks up against semaglutide (Wegovy, Ozempic) for weight loss. Every comparison came from lining up separate trials that used different people, different timelines, and different measurement methods. That changed in 2025, when the New England Journal of Medicine published SURMOUNT-5, the first randomized trial to put both drugs in the same study, at the same time, in the same patients. This article walks through exactly what the trial found, using the numbers from the primary publication and the official trial results record, plus the caveats that matter before you use this data to make a decision.

What Was SURMOUNT-5, and Why Does It Matter?

SURMOUNT-5 was a phase 3b, randomized, controlled trial funded by Eli Lilly, the maker of tirzepatide, and run at 32 sites across the United States and Puerto Rico between April 2023 and November 2024 (Aronne et al., 2025). It randomly assigned 751 adults with obesity, but without type 2 diabetes, to receive either tirzepatide or semaglutide, injected once a week, for 72 weeks. The primary question was simple: when you give both drugs a fair, identical shot at the maximum dose each person could tolerate, which one takes off more weight?

That question matters because before SURMOUNT-5, "tirzepatide beats semaglutide" was an inference, not a finding. Tirzepatide's own approval trial, SURMOUNT-1, showed roughly 20.9% weight loss at the top dose (Jastreboff et al., 2022). Semaglutide's approval trial, STEP 1, showed roughly 14.9% weight loss (Wilding et al., 2021). Those numbers looked different, but the trials enrolled different people at different times using different placebo comparisons, so cross-trial comparisons are always a little shaky. SURMOUNT-5 was built to settle that specific question with a direct, apples-to-apples comparison. It's registered on ClinicalTrials.gov as NCT05822830, and the full results record there matches the published paper's design and headline numbers.

How Was the Trial Designed?

Understanding what SURMOUNT-5 can and can't tell you starts with understanding exactly how it was built.

Who Was Enrolled?

The trial enrolled 751 adults with a body mass index of 30 or higher, or 27 or higher if they had at least one weight-related complication such as high blood pressure or joint problems. People with type 2 diabetes were excluded, which matters — this trial tells you nothing directly about how the two drugs compare in people managing diabetes alongside their weight. The average participant was 44.7 years old, and about 65% were women. Participants were randomly assigned 1:1 to tirzepatide or semaglutide, and 750 of the 751 actually received at least one dose of their assigned drug.

What Doses Did Each Group Take?

This is a detail that gets glossed over in headlines, so it's worth spelling out. Neither group received a single fixed dose. Both were titrated up over several months and then treated at whatever dose each individual could tolerate, capped at a ceiling:

• Tirzepatide group: started at 2.5 mg once weekly, then increased by 2.5 mg every 4 weeks (2.5 → 5 → 7.5 → 10 → 12.5 → 15 mg) toward a target of 15 mg weekly, or the maximum tolerated dose, which could land at 10 mg or 15 mg.
• Semaglutide group: started at 0.25 mg once weekly, then increased roughly every 4 weeks (0.25 → 0.5 → 1.0 → 1.7 → 2.4 mg) toward a target of 2.4 mg weekly, or the maximum tolerated dose, which could land at 1.7 mg or 2.4 mg.

Both schedules mirror the FDA-approved titration for each drug's obesity indication, so this wasn't a case of stacking the deck by under-dosing one arm. It does mean, however, that "SURMOUNT-5" isn't strictly a "15 mg vs. 2.4 mg" trial — some participants in each arm settled at the lower of the two top doses because the highest dose didn't agree with them. If you're curious how that titration process plays out in practice, our semaglutide dosing and titration guide covers the same step pattern used here.

The primary endpoint was the percent change in body weight from baseline to week 72. Key secondary endpoints included the percentage of participants achieving weight reductions of at least 10%, 15%, 20%, and 25%, plus the change in waist circumference.

What Were the Actual Weight-Loss Results?

Here's where the trial delivers a clear answer. At 72 weeks, tirzepatide produced significantly more weight loss than semaglutide, and the gap wasn't small.

The least-squares mean percent change in body weight was -20.2% with tirzepatide (95% CI, -21.4 to -19.1) versus -13.7% with semaglutide (95% CI, -14.9 to -12.6), a statistically significant difference (P<0.001). In absolute terms, that worked out to about 22.8 kg (roughly 50 pounds) lost on tirzepatide versus 15.0 kg (roughly 33 pounds) on semaglutide. Waist circumference told the same story: -18.4 cm with tirzepatide versus -13.0 cm with semaglutide (P<0.001).

The table below pulls together the full results picture, combining the numbers reported in the NEJM publication with the more granular figures posted to the trial's official results record on ClinicalTrials.gov.

Arm	Target dose (weekly, SC)	Mean % weight loss (72 wk)	≥10% weight loss	≥15% weight loss	≥20% weight loss	≥25% weight loss	Key GI adverse events (nausea / diarrhea / vomiting / constipation)	Discontinued due to any AE
Tirzepatide	15 mg or MTD (10 or 15 mg)	-20.2% (-22.8 kg)	87.7%	71.7%	55.0%	36.5%	43.6% / 23.5% / 15.0% / 27.0%	1.6%
Semaglutide	2.4 mg or MTD (1.7 or 2.4 mg)	-13.7% (-15.0 kg)	66.7%	45.0%	31.1%	18.6%	44.4% / 23.4% / 21.3% / 28.5%	1.6%

Weight-loss and milestone percentages are from the trial's official results record on ClinicalTrials.gov (NCT05822830), cross-checked against the mean weight-loss and waist-circumference figures reported in the primary NEJM publication (Aronne et al., 2025). Adverse-event percentages are calculated from the same registry record, based on 374 tirzepatide and 376 semaglutide participants who received at least one dose. "Discontinued due to any AE" reflects participants who left the study early for that reason, not participants who simply stopped the injections but stayed in follow-up.

A few things jump out. Tirzepatide won on every single weight-loss measure — average percent lost, absolute kilograms, waist circumference, and every response-rate threshold. At the far end, 23.0% of the tirzepatide group lost 30% or more of their body weight, compared with 8.2% of the semaglutide group.

How Many People Hit Major Weight-Loss Milestones?

The mean numbers matter, but averages can hide a lot. Two people can average the same weight loss while one drug produces consistent moderate results and the other produces a mix of huge responders and non-responders. That's why SURMOUNT-5's "milestone" data is arguably more useful for an individual patient trying to picture their own odds.

Nearly 9 in 10 people on tirzepatide (87.7%) lost at least 10% of their body weight, a threshold generally associated with meaningful improvement in blood pressure, blood sugar, and joint pain. About two-thirds of the semaglutide group (66.7%) hit that same mark. At the higher end, more than a third of tirzepatide patients (36.5%) reached the 25% mark, sometimes described informally as "surgery-adjacent" weight loss, versus roughly 1 in 5 (18.6%) on semaglutide.

None of this means semaglutide "doesn't work." Two-thirds of people on semaglutide still lost a clinically meaningful amount of weight, and nearly half lost 15% or more. What the trial shows is a consistent tilt toward larger, more frequent high-magnitude responses on tirzepatide across the entire distribution of outcomes, not just the average.

Why Might Tirzepatide Outperform Semaglutide? The Mechanism

SURMOUNT-5 tells you that tirzepatide produced more weight loss, but it's worth understanding why that's biologically plausible, because it changes how much weight you should put on the result.

Semaglutide is a single-receptor drug. It activates the GLP-1 receptor, which mimics a gut hormone released after eating. GLP-1 receptor activation slows stomach emptying, dials down appetite signals in the brain, and improves how the body handles blood sugar. That's the entire mechanism, and it's a well-proven one — it's also the basis for liraglutide (Saxenda) and dulaglutide (Trulicity), just at different potencies.

Tirzepatide is a dual-receptor drug. It activates the GLP-1 receptor the same way semaglutide does, but it also activates a second receptor, GIP (glucose-dependent insulinotropic polypeptide). For years, GIP's role in weight regulation was debated — some early animal research even suggested blocking GIP might help with weight loss, the opposite of what tirzepatide does. But the clinical trial data settled the practical question: combining GIP agonism with GLP-1 agonism produces a larger effect on appetite suppression and food intake than GLP-1 agonism alone, and it appears to do so partly through an independent pathway rather than simply "more of the same signal." GIP receptors are also present in fat tissue itself, where activation may improve how fat cells store and process energy, a mechanism GLP-1-only drugs don't touch directly.

None of this mechanism talk is a substitute for trial evidence, and mechanistic plausibility has misled drug development before. But it does mean SURMOUNT-5's result isn't a surprising fluke — it lines up with a specific, testable biological hypothesis (two appetite-and-metabolism pathways beat one) that researchers had already proposed before this trial ran. It's also why the industry's next wave of drugs, like the triple-agonist retatrutide, is chasing even more receptor targets rather than fewer — the SURMOUNT-5 result is treated as evidence for that broader strategy, not just a one-off finding about these two specific drugs.

How Did Side Effects Compare Between the Two Drugs?

This is the part people often assume works against tirzepatide, since it activates an extra hormone receptor (GIP, on top of GLP-1). SURMOUNT-5's actual safety data is more nuanced than "more mechanisms, more side effects."

Gastrointestinal Side Effects

Both drugs caused gastrointestinal side effects at similar rates, and both were worst during the dose-escalation months and eased over time. Nausea affected 43.6% of the tirzepatide group and 44.4% of the semaglutide group — essentially a tie. Diarrhea was nearly identical too, at 23.5% versus 23.4%. Constipation was a touch more common with semaglutide (28.5% versus 27.0%). The one meaningful gap ran the opposite direction from what people often expect: vomiting was more common with semaglutide (21.3%) than tirzepatide (15.0%).

Put simply, the "gut" side effects that make people quit GLP-1 drugs in real life were comparable between the two, with a slight edge to tirzepatide on the vomiting front. If GI tolerability specifically is your top concern, this trial doesn't give a strong reason to pick one over the other on that basis alone. For a broader rundown of how these side effects show up and how to manage them, see our complete GLP-1 side effects guide.

Injection-Site Reactions and Serious Adverse Events

Where the two drugs did diverge clearly was injection-site reactions — redness, itching, and irritation at the injection spot. These occurred in 8.6% of the tirzepatide group but only 0.3% of the semaglutide group, a meaningful and consistent gap. It's a minor annoyance for most people, but worth knowing if you're needle-sensitive.

Serious adverse events (the trial's term for events requiring hospitalization or judged medically significant) were reported in 4.8% of the tirzepatide group and 3.5% of the semaglutide group. Both figures are low, the confidence intervals almost certainly overlap given the modest sample size, and no consistent pattern of a specific serious harm stood out in either arm in the primary analysis. There were no deaths in either group during the trial.

Who Stopped Treatment, and Why?

Overall, the two drugs were tolerated similarly well at the level of finishing the study. Per the trial's official results record, only 1.6% of participants in each arm left the study altogether because of an adverse event — an identical rate. About 85% of both groups completed the full 72 weeks (85.1% tirzepatide, 84.8% semaglutide), with the rest split between lost-to-follow-up, personal withdrawal, and a handful of protocol-related reasons.

Separately, in coverage of the trial's more granular safety tables, a gap has been reported specifically in stopping the study drug itself because of GI adverse events — around 2.7% of the tirzepatide group versus 5.6% of the semaglutide group. That's a different, narrower metric than leaving the study altogether (someone can stop their injections due to nausea but stay in the trial for follow-up visits), and it points the same direction as the vomiting data above: semaglutide's GI burden pushed slightly more people off the drug itself, even though it didn't push more people out of the study entirely.

Does This Mean Tirzepatide Is Simply "Better"?

For the specific question SURMOUNT-5 asked — which drug produces more weight loss, at maximum tolerated doses, over 72 weeks, in adults with obesity but without diabetes — the answer is unambiguous: tirzepatide did. That's not a marginal statistical footnote; a roughly 6.5 percentage-point gap in mean weight loss, replicated across every response threshold, is a real and clinically meaningful difference.

But "better" is doing a lot of work in that sentence, and it's worth resisting the urge to over-read a single efficacy comparison into a global verdict. Weight loss magnitude is one axis. Cost, insurance coverage, injection-site tolerability, vomiting frequency, needle/pen mechanics, and — critically — the depth of long-term outcome data are separate axes, and SURMOUNT-5 didn't measure most of them. It also didn't measure anything about diabetes management, since people with diabetes were excluded, and it didn't measure cardiovascular outcomes at all. For the fuller picture beyond this one trial, our general semaglutide vs. tirzepatide comparison covers cost, availability, and day-to-day differences that SURMOUNT-5 wasn't designed to touch.

Which Drug Might Suit Whom?

None of this is medical advice, and the right choice always runs through a conversation with a prescriber who knows your health history. But a few evidence-grounded patterns are worth laying out plainly.

Tirzepatide may be the stronger fit if: you and your doctor have set a large weight-loss target, previous GLP-1-class treatment underperformed and you want the drug with the bigger average effect, and injection-site irritation or a slightly higher chance of a serious-adverse-event report (still low in absolute terms) isn't a dealbreaker for you.

Semaglutide may be the stronger fit if: you have established cardiovascular disease, since semaglutide (not tirzepatide) has a completed, dedicated outcomes trial showing it reduces heart attacks, strokes, and cardiovascular death in adults with obesity and existing heart disease — the SELECT trial (Lincoff et al., 2023). Semaglutide may also make sense if your insurance formulary or budget specifically favors it, if you're needle-averse to injection-site reactions, or if a somewhat smaller but still substantial weight-loss target is your goal.

Either can be reasonable if: you're earlier in treatment and want to start with whichever your insurance actually covers, since real-world access often decides this question before efficacy differences do. A Cochrane systematic review of tirzepatide trials for obesity, published in 2025, reinforces that both drug classes produce clinically meaningful weight loss versus placebo — SURMOUNT-5's contribution is telling you the relative size of that effect, not whether either drug "works."

It's also worth thinking about this less as "pick a winner" and more as "pick a starting point." Plenty of people switch between the two drug classes over the course of treatment, whether because of side effects, cost changes, insurance formulary shifts, or a plateau in results. If you're on semaglutide and considering a change, our guide on switching from Ozempic to Mounjaro walks through what that transition actually looks like in practice, including how clinicians typically re-titrate the new drug rather than jumping straight to a high dose.

What Are the Important Caveats?

A trial this well-designed still has real limits, and a citation-grade breakdown means naming them clearly rather than burying them.

It Was Open-Label, Not Blinded

SURMOUNT-5 was explicitly an open-label trial — both participants and their study teams knew which drug each person was taking (Aronne et al., 2025). That's a real limitation for a self-reported side-effect and subjective-symptom trial, though less of a concern for a primarily objective endpoint like body weight measured on a scale. Still, it's not nothing: knowing you're on the drug widely reported as "more effective" could plausibly shape adherence, effort on diet and exercise, or how symptoms get reported. This exact concern was raised directly in published correspondence responding to the trial — a letter from independent researchers argued the open-label design, combined with the fact that some participants may have been unblinded expectation-wise from prior media coverage of tirzepatide, could have modestly inflated the observed difference, a critique the trial's lead author addressed in a published reply (Muskiet et al., 2025). It's a genuine scientific debate, not a fatal flaw, but it's a reason not to treat the exact percentage-point gap as gospel-precise.

This Isn't a Cardiovascular-Outcomes Trial

SURMOUNT-5 measured weight, waist circumference, and standard safety events over 72 weeks. It did not measure — and was not statistically powered to measure — heart attacks, strokes, or cardiovascular death. Semaglutide already has that evidence from a dedicated trial, SELECT (Lincoff et al., 2023), which enrolled people with obesity and existing cardiovascular disease and showed a reduction in major cardiovascular events. Tirzepatide does not yet have a completed, published trial of that kind in this population. That doesn't mean tirzepatide lacks cardiovascular benefit — it very plausibly has one, given the weight loss alone — it means the head-to-head, gold-standard proof doesn't exist yet for tirzepatide the way it does for semaglutide. Don't let "tirzepatide loses more weight" quietly get translated in your head into "tirzepatide is proven to protect your heart more." That specific claim isn't supported by SURMOUNT-5 or by any completed trial yet.

Funding and Sponsorship

SURMOUNT-5 was funded by Eli Lilly, which manufactures tirzepatide and did not manufacture the semaglutide comparator. That's disclosed transparently in the trial publication and doesn't automatically bias a well-run randomized trial — the methodology, blinding-of-outcome-assessors for the weight measurement, and statistical analysis all follow standard practices regardless of sponsor. But it's a detail worth knowing, particularly alongside the open-label critique above, since sponsor-funded head-to-head trials against a competitor's drug are worth reading with a bit more scrutiny than an independent comparison would warrant.

Cost and Access Aren't Part of the Science

Nothing in SURMOUNT-5 tells you what either drug costs, whether your insurance covers it, or how easy either one is to actually obtain given ongoing supply constraints in some regions. Those practical factors often matter more day-to-day than a 6-percentage-point efficacy gap. If cost is the deciding factor for you, our GLP-1 medication cost guide breaks down current pricing and savings-program options for both drugs.

How Do These Numbers Compare to Each Drug's Own Pivotal Trial?

One useful sanity check on any head-to-head trial is whether its results line up with what each drug already showed on its own, in a different trial, against placebo. Here, they largely do.

In SURMOUNT-1, the placebo-controlled, double-blind trial that led to tirzepatide's approval, the 15 mg dose produced -20.9% mean weight loss at 72 weeks in a much larger group of 2,539 participants (Jastreboff et al., 2022). SURMOUNT-5's tirzepatide arm landed at -20.2% — very close.

In STEP 1, the placebo-controlled, double-blind trial behind semaglutide's approval, the 2.4 mg dose produced -14.9% mean weight loss at 68 weeks (a slightly shorter trial) in 1,961 participants, with 69.1% reaching 10% or more weight loss and 50.5% reaching 15% or more (Wilding et al., 2021). SURMOUNT-5's semaglutide arm came in at -13.7%, with 66.7% and 45.0% hitting those same thresholds — again, reasonably close, on the slightly lower side, which is plausible given the somewhat different population and the shorter placebo-trial duration used for comparison.

That consistency matters. It means SURMOUNT-5 isn't some outlier result driven by an unusual sample — it's a direct replication, in the same patients under the same conditions, of a gap that each drug's separate pivotal trial had already hinted at independently. A broader 2026 systematic review and network meta-analysis of GLP-1 agonist weight loss in nondiabetic adults reached a similar overall ranking across the class, reinforcing that this isn't a one-trial fluke (systematic review and network meta-analysis, 2026). Separately, real-world data outside any trial setting — a 2025 analysis of GLP-1 prescription patterns — found that discontinuation and reinitiation of these drugs is common in routine practice regardless of which one a patient starts on, a reminder that trial-setting adherence (SURMOUNT-5's ~85% completion rate) doesn't always translate directly to the real world, where cost and access disruptions play a bigger role (Discontinuation and Reinitiation of Dual-Labeled GLP-1 Receptor Agonists, 2025).

Medical disclaimer: This article is for general educational purposes and is not medical advice. It summarizes published clinical trial data but does not replace a consultation with a licensed healthcare provider. Do not start, stop, switch, or adjust the dose of any medication, including tirzepatide or semaglutide, based on this article. Individual response to these medications varies, and only a qualified clinician who knows your full medical history can determine which treatment, if any, is appropriate for you.

Frequently Asked Questions

What is SURMOUNT-5, and what did it actually test?

SURMOUNT-5 is a phase 3b randomized clinical trial published in the New England Journal of Medicine in 2025 that directly compared tirzepatide (Zepbound/Mounjaro) against semaglutide (Wegovy/Ozempic) in 751 adults with obesity but without type 2 diabetes. Both groups took their assigned drug once weekly, at the maximum dose they could tolerate, for 72 weeks. It's the first published head-to-head trial putting these two drugs against each other rather than each against a placebo separately.

Did tirzepatide really produce significantly more weight loss than semaglutide?

Yes. Tirzepatide produced a mean weight loss of 20.2% versus 13.7% for semaglutide at 72 weeks, a statistically significant difference (P<0.001). Tirzepatide also produced higher rates of reaching every major weight-loss milestone tested, from 10% up through 30% of body weight.

Was SURMOUNT-5 a blinded, placebo-controlled trial?

No, and this matters. It was open-label — both participants and researchers knew which drug each person was taking — and it compared the two active drugs directly rather than against a placebo. Open-label design is a real limitation for subjective measures, though less so for an objective outcome like weight on a scale. Published correspondence following the trial specifically debated whether the open-label design may have modestly inflated the observed gap between the drugs.

Does this trial prove tirzepatide is safer than semaglutide?

No. It shows the two drugs had broadly similar rates of the most common side effects (nausea, diarrhea, constipation), with semaglutide causing more vomiting and tirzepatide causing more injection-site reactions. Serious adverse events were low and similar in both groups, and neither trial arm had any deaths. Overall study discontinuation due to any adverse event was identical between groups (1.6% each). This trial does not address longer-term safety questions or cardiovascular outcomes, where only semaglutide currently has completed, dedicated outcome-trial evidence.

Does this mean semaglutide is now the "second-choice" GLP-1 drug?

Not necessarily, and that framing oversimplifies a more nuanced picture. Tirzepatide produced more weight loss in this specific trial, but semaglutide is the only one of the two with a completed, published cardiovascular-outcomes trial (SELECT) showing a reduction in heart attacks and strokes in people with obesity and existing heart disease. Cost, insurance coverage, injection-site tolerability, and vomiting frequency also factor into a real decision. The right drug for any individual depends on their specific health priorities, which a prescriber is best positioned to weigh alongside this evidence.

Related Reading

• The Complete Guide to Tirzepatide: Mounjaro and Zepbound
• Mounjaro vs. Ozempic: Head-to-Head Clinical Comparison
• GLP-1 Costs in 2026: Insurance, Medicare & Savings

---

— The GLP-1 Daily Team