What the Clinical Trials Actually Show About Tirzepatide

Tirzepatide is the active ingredient in two US brands: Mounjaro, approved for type 2 diabetes, and Zepbound, approved for chronic weight management and for obstructive sleep apnea in adults with obesity. Same molecule, same manufacturer (Eli Lilly), different labels.

It works on two gut hormone receptors instead of one. Semaglutide targets the GLP-1 receptor. Tirzepatide targets GLP-1 and GIP, which is why it is often called a dual agonist. Whether the second receptor is the reason it produces more weight loss is still argued about. What is not argued about is the size of the trial program behind it.

This page walks through what those trials measured and what they found. It is not medical advice, and it contains no recommendation about whether anyone should take this drug or at what dose. Those are questions for a healthcare provider.

What are the two main tirzepatide trial programs?

SURPASS tested tirzepatide in type 2 diabetes. It produced the evidence for Mounjaro.

• SURPASS-1 (n=478, 40 weeks) tested tirzepatide alone in people not on any diabetes drug. HbA1c fell 1.87 to 2.07 percentage points depending on dose, versus a 0.04-point rise on placebo. Between 31% and 52% of participants reached an HbA1c below 5.7% [1].
• SURPASS-2 (n=1,879, 40 weeks) put tirzepatide against semaglutide 1 mg on top of metformin. All three tirzepatide doses beat it, by 0.15 to 0.45 HbA1c points and by 1.9 to 5.5 kg of weight [2]. The comparator was the 1 mg diabetes dose, not the 2.4 mg obesity dose. That distinction gets dropped constantly in secondary coverage.
• SURPASS-3 (n=1,444, 52 weeks) and SURPASS-4 (n=2,002) compared it with long-acting insulin. Tirzepatide lowered HbA1c more, caused far less low blood sugar, and produced weight loss where insulin produced weight gain [3].
• SURPASS-5 and SURPASS-6 tested it added to insulin regimens.
• SURPASS-CVOT (n=13,299) tested heart outcomes against dulaglutide. It gets its own article.

There are also regional trials — SURPASS J-mono and J-combo in Japan, SURPASS-AP-Combo across China, South Korea, Australia and India, and SURPASS-CN-MONO in China. All found broadly the same pattern [4][5].

SURMOUNT tested tirzepatide for weight. It produced the evidence for Zepbound.

• SURMOUNT-1 (n=2,539, 72 weeks) is the headline trial. Mean weight change was -15.0% at 5 mg, -19.5% at 10 mg and -20.9% at 15 mg, versus -3.1% on placebo. Half to 57% of the higher-dose groups lost at least 20% of their starting weight [6].
• SURMOUNT-2 (n=938) repeated it in people who also had type 2 diabetes. Weight loss was smaller — 12.8% and 14.7% — which is the same pattern semaglutide shows [7].
• SURMOUNT-3 (n=579) added tirzepatide after a 12-week intensive lifestyle program. Participants lost a further 18.4% while the placebo group started regaining [8].
• SURMOUNT-4 (n=670) is the withdrawal trial. It gets its own article.
• SURMOUNT-5 (n=751) is the head-to-head against semaglutide 2.4 mg. Tirzepatide produced 20.2% weight loss versus 13.7% at 72 weeks [9].

How big is the tirzepatide evidence base compared with semaglutide?

Smaller, and newer. Semaglutide has been in phase 3 trials since 2013 and has published outcome data in cardiovascular disease, kidney disease, peripheral artery disease and liver disease. Tirzepatide’s first phase 3 results appeared in 2021.

The practical gap is in outcome trials — the kind that measure heart attacks, strokes and deaths rather than weight and blood sugar. Semaglutide has SELECT, which showed a reduction in heart events in people with obesity and cardiovascular disease but without diabetes. Tirzepatide’s equivalent is SURMOUNT-MMO, and it has not reported. ClinicalTrials.gov lists an estimated primary completion date of October 2027 [10]. Secondary coverage that claims a late-2026 readout is out of date.

What tirzepatide does have is SURPASS-CVOT in type 2 diabetes, and it showed non-inferiority to dulaglutide rather than superiority [11]. That is a real result, but it is a different and narrower claim.

What did the sleep apnea trial find?

SURMOUNT-OSA ran two parallel 52-week trials in 469 adults with moderate-to-severe obstructive sleep apnea and obesity — one group not using a breathing machine, one group using one.

The apnea-hypopnea index, which counts breathing interruptions per hour of sleep, fell by 25.3 events per hour in the first trial and 29.3 in the second, versus 5.3 and 5.5 on placebo [12]. That supported the FDA approval of Zepbound for obstructive sleep apnea in December 2024 — an indication semaglutide does not have.

Follow-up analyses added two things. Participants also reported better sleep, less daytime impairment and higher quality-of-life scores, not just better numbers on the sleep study [13]. And a mediation analysis found that improvements in inflammation, insulin resistance and triglycerides tracked the sleep apnea improvement independently of weight loss, while blood pressure change tracked weight [14].

What about heart failure, liver and kidney disease?

Heart failure. The SUMMIT trial (n=731, median 104 weeks) enrolled people with heart failure with preserved ejection fraction and a BMI of 30 or higher. Cardiovascular death or worsening heart failure occurred in 9.9% on tirzepatide versus 15.3% on placebo, a hazard ratio of 0.62 (95% CI 0.41 to 0.95) [15]. That is one of very few trials in this drug class to reduce hard heart failure events rather than just symptoms. There is no FDA heart failure indication for tirzepatide.

Liver. SYNERGY-NASH (n=190) was a phase 2 trial in people with biopsy-confirmed MASH — a form of fatty liver disease with inflammation — and moderate or severe fibrosis. The disease resolved without worsening fibrosis in 44% to 62% of tirzepatide groups versus 10% on placebo [16]. That is a striking result, but it was phase 2, dose-finding, and the authors themselves called for larger trials. Tirzepatide has no FDA liver indication. Semaglutide received an accelerated approval for MASH in August 2025 on the strength of the larger ESSENCE trial. A phase 3 master protocol with a hard liver-outcomes endpoint, including tirzepatide, began recruiting in 2026 with an estimated completion in 2030 [17].

Kidney. Two post hoc analyses and one prespecified exploratory analysis point the same way. In SURPASS-4, kidney function declined more slowly on tirzepatide than on insulin glargine [18]. In SURMOUNT-1 and SURMOUNT-2 pooled, protein in the urine fell without any worsening of filtration [19]. And inside SURPASS-CVOT, the composite kidney outcome occurred in 6.0% on tirzepatide versus 7.6% on dulaglutide (hazard ratio 0.77; 95% CI 0.68 to 0.88) [20]. None of these is a dedicated kidney outcomes trial. The mechanistic study that was meant to inform whether one is worth running, TREASURE-CKD, completed in July 2026 and had not reported as of September 2026 [21].

Does tirzepatide cause muscle loss?

This is the most-asked question in the whole category, and the honest answer is: we have one small sub-study.

Within SURMOUNT-1, 160 participants had DXA body composition scans at baseline and week 72. In the pooled tirzepatide group, body weight fell 21.3%, fat mass fell 33.9% and lean mass fell 10.9%. In the placebo group the figures were 5.3%, 8.2% and 2.6% [22].

Do the arithmetic and roughly 75% of the weight lost was fat and 25% was lean tissue — in both groups. That matters. If the drug were uniquely destroying muscle, you would expect the ratio to be worse than placebo. It was not. The proportion also held up across sex, age and how much weight people lost.

Three caveats. Lean mass on a DXA scan includes water, organs and connective tissue, not just muscle. The sub-study was 160 people out of 2,539. And nobody has published a randomized trial showing whether strength training and extra protein change the picture — the one designed to answer it, LEAN-PREP, has published only its protocol [23].

What happened in the children’s trial?

SURPASS-PEDS (n=99) tested tirzepatide 5 mg and 10 mg against placebo in children and adolescents aged 10 to under 18 with youth-onset type 2 diabetes not controlled on metformin, insulin, or both. Mean age was 14.7 years.

Over 30 weeks, HbA1c fell 2.23 percentage points in the pooled tirzepatide group versus a 0.05-point rise on placebo — an estimated treatment difference of -2.28 points (95% CI -2.87 to -1.69). BMI fell 7.4% at 5 mg and 11.2% at 10 mg versus 0.4% on placebo. The benefit held through 52 weeks in an open-label extension. Side effects were gastrointestinal, all mild to moderate [24].

It is a small trial by adult standards, and it was in type 2 diabetes, not obesity alone.

What are the newest results?

SURMOUNT-MAINTAIN, published in The Lancet in May 2026, is the first randomized test of lowering the dose instead of stopping. After 60 weeks of open-label treatment, 378 participants were randomized to stay on their maximum tolerated dose, step down to 5 mg, or switch to placebo for another 52 weeks. At week 112, weight change from baseline was -21.9% on the higher dose, -16.6% on 5 mg and -9.9% on placebo [25]. It gets covered in detail in the article on stopping.

REDEFINE 4, announced by Novo Nordisk in February 2026, is the unusual case of a competitor running a registration-grade head-to-head trial against tirzepatide. CagriSema produced 23.0% weight loss at 84 weeks versus 25.5% for tirzepatide 15 mg, and missed its non-inferiority endpoint [26].

SURPASS-EARLY, published in 2026, tested starting tirzepatide soon after a type 2 diabetes diagnosis instead of working through the usual sequence of drugs. After two years, 60.2% of the tirzepatide group had an HbA1c below 5.7% versus 24.0% on conventional care [27].

Where does the evidence still run out?

Six honest gaps, as of September 2026:

1. No cardiovascular outcome result in obesity without diabetes. SURMOUNT-MMO is estimated to report no earlier than October 2027 [10].
2. No randomized tapering study. SURMOUNT-MAINTAIN tested a fixed step-down to 5 mg, not a gradual taper. Nobody has randomized a taper for any drug in this class.
3. No type 1 diabetes result. SURPASS-T1D-1 and SURPASS-T1D-2 are both active and not recruiting, with estimated primary completion in November 2026 [28]. Tirzepatide is not approved for type 1 diabetes.
4. No head-to-head against the newer semaglutide doses. SURMOUNT-5 used semaglutide 2.4 mg. It did not test the 7.2 mg dose (Wegovy HD), which the FDA approved on March 19, 2026. No trial has compared tirzepatide with that dose.
5. Contradictory real-world heart data. One large US claims analysis (STEER) found fewer heart events on semaglutide than tirzepatide in people with obesity and cardiovascular disease but no diabetes [29] — note that every author was a Novo Nordisk employee and shareholder, and Novo Nordisk makes semaglutide. Another, using the TriNetX network in older adults with type 2 diabetes, found the opposite and declared no conflicts [30]. Neither is randomized, and no head-to-head heart trial of the two exists.
6. Almost everything is manufacturer-funded. Eli Lilly paid for the entire phase 3 program. That is normal in drug development, and the trials were published in major journals with independent peer review — but it is worth knowing.

What should a reader take away?

Tirzepatide has a large, consistent and mostly recent evidence base. Across obesity trials it produces more weight loss than any approved comparator it has been tested against. Across diabetes trials it lowers HbA1c more than insulin, dulaglutide and semaglutide 1 mg. It has a sleep apnea indication nothing else in the class has, and a heart failure result in obesity-related HFpEF.

What it does not yet have is the four-year cardiovascular outcome evidence in obesity that semaglutide has. Whether it will is the question SURMOUNT-MMO is built to answer, and that answer is at least a year away.

Any decision about taking this drug, switching to it, or changing a dose belongs with a healthcare provider who knows your history.

Sources

1. SURPASS-1, The Lancet 2021 — https://pubmed.ncbi.nlm.nih.gov/34186022/
2. SURPASS-2, New England Journal of Medicine 2021 — https://pubmed.ncbi.nlm.nih.gov/34170647/
3. SURPASS-3, The Lancet 2021 — https://pubmed.ncbi.nlm.nih.gov/34370970/
4. SURPASS J-mono, The Lancet Diabetes & Endocrinology 2022 — https://pubmed.ncbi.nlm.nih.gov/35914543/
5. SURPASS-AP-Combo, Nature Medicine 2023 — https://pubmed.ncbi.nlm.nih.gov/37231074/
6. SURMOUNT-1, New England Journal of Medicine 2022 — https://pubmed.ncbi.nlm.nih.gov/35658024/
7. SURMOUNT-2, The Lancet 2023 — https://pubmed.ncbi.nlm.nih.gov/37385275/
8. SURMOUNT-3, Nature Medicine 2023 — https://pubmed.ncbi.nlm.nih.gov/37840095/
9. SURMOUNT-5, New England Journal of Medicine 2025 — https://pubmed.ncbi.nlm.nih.gov/40353578/
10. SURMOUNT-MMO registry record, ClinicalTrials.gov — https://clinicaltrials.gov/study/NCT05556512
11. SURPASS-CVOT, New England Journal of Medicine 2025 — https://pubmed.ncbi.nlm.nih.gov/41406444/
12. SURMOUNT-OSA, New England Journal of Medicine 2024 — https://pubmed.ncbi.nlm.nih.gov/38912654/
13. SURMOUNT-OSA patient-reported outcomes, Sleep Medicine 2025 — https://pubmed.ncbi.nlm.nih.gov/40774158/
14. SURMOUNT-OSA cardiometabolic secondary outcomes, Nature Medicine 2026 — https://pubmed.ncbi.nlm.nih.gov/41540105/
15. SUMMIT, New England Journal of Medicine 2025 — https://pubmed.ncbi.nlm.nih.gov/39555826/
16. SYNERGY-NASH, New England Journal of Medicine 2024 — https://pubmed.ncbi.nlm.nih.gov/38856224/
17. MASH outcomes master protocol, ClinicalTrials.gov — https://clinicaltrials.gov/study/NCT07165028
18. SURPASS-4 kidney post hoc, The Lancet Diabetes & Endocrinology 2022 — https://pubmed.ncbi.nlm.nih.gov/36152639/
19. Kidney parameters in SURMOUNT-1 and SURMOUNT-2, JASN 2025 — https://pubmed.ncbi.nlm.nih.gov/40512543/
20. SURPASS-CVOT kidney analysis, The Lancet Diabetes & Endocrinology 2026 — https://pubmed.ncbi.nlm.nih.gov/42114520/
21. TREASURE-CKD registry record, ClinicalTrials.gov — https://clinicaltrials.gov/study/NCT05536804
22. SURMOUNT-1 DXA body composition sub-study, Diabetes Obesity and Metabolism 2025 — https://pubmed.ncbi.nlm.nih.gov/39996356/
23. LEAN-PREP protocol, BMJ Open 2026 — https://pubmed.ncbi.nlm.nih.gov/42020128/
24. SURPASS-PEDS, The Lancet 2025 — https://pubmed.ncbi.nlm.nih.gov/40975112/
25. SURMOUNT-MAINTAIN, The Lancet 2026 — https://pubmed.ncbi.nlm.nih.gov/42119587/
26. REDEFINE 4 headline results, Novo Nordisk 2026-02-23 — https://www.globenewswire.com/de/news-release/2026/02/23/3242381/0/en/Novo-Nordisk-A-S-CagriSema-demonstrated-23-weight-loss-in-an-open-label-head-to-head-REDEFINE-4-trial-in-people-with-obesity-the-primary-endpoint-was-not-achieved.html
27. SURPASS-EARLY, Annals of Internal Medicine 2026 — https://pubmed.ncbi.nlm.nih.gov/42184419/
28. SURPASS-T1D-1 registry record, ClinicalTrials.gov — https://clinicaltrials.gov/study/NCT06914895
29. STEER real-world cardiovascular comparison, Diabetes Obesity and Metabolism 2026 — https://pubmed.ncbi.nlm.nih.gov/41491349/
30. TriNetX tirzepatide versus semaglutide cohort, Diabetes & Vascular Disease Research 2026 — https://pubmed.ncbi.nlm.nih.gov/42376874/