Tirzepatide Clinical Trial Results, Summarized

Tirzepatide is the active ingredient in two once-weekly injections: one approved for chronic weight management and obstructive sleep apnea, and one approved for type 2 diabetes. Its weight-loss evidence comes mainly from a series of large randomized trials called SURMOUNT, plus trials in sleep apnea (SURMOUNT-OSA), heart failure (SUMMIT) and heart disease in type 2 diabetes (SURPASS-CVOT). Headlines often quote one figure, usually “about 21%” or “22.5%,” without saying which dose, which people, which analysis or what happened afterward.

This guide summarizes the main trials in plain English: who took part, what they were given, the average results, the side effects and the limits of each study. It is meant to help you read the evidence and ask better questions, not to tell you which medicine or dose to use.

Educational only. This is not medical advice. Trial results are averages from carefully selected volunteers and do not predict what will happen to any one person. Tirzepatide is a prescription medicine; decisions about whether to start it, which dose to use, and when to change or stop it belong to you and your prescriber.

The short answer

In SURMOUNT-1, adults without diabetes who took weekly tirzepatide with lifestyle counseling lost an average of 15.0% (5 mg), 19.5% (10 mg) and 20.9% (15 mg) of their body weight over 72 weeks, compared with 3.1% on placebo. About half of people on the two higher doses lost at least a fifth of their starting weight.

A few other results matter just as much:

• People with type 2 diabetes lost less: 12.8% to 14.7% versus 3.2% in SURMOUNT-2.
• Results held while people kept taking it, including for three years in adults with prediabetes, but people switched to placebo regained much of the weight (SURMOUNT-4 and SURMOUNT-MAINTAIN).
• Head to head, tirzepatide beat semaglutide on average (20.2% versus 13.7% in SURMOUNT-5).
• Sleep apnea improved substantially, and heart-failure events were less frequent in adults with a specific type of heart failure and obesity.
• Digestive side effects such as nausea, diarrhea, vomiting and constipation were the most common problem in every trial.

If you are new to the medicine itself, start with What Is Tirzepatide? and then come back here for the details.

How to read these trials

A few features are shared by almost every trial below, and knowing them makes the numbers much easier to interpret.

• Everyone got lifestyle support, including the placebo group. Participants were counseled to follow a reduced-calorie diet (in later trials, about 500 calories a day below their needs) and to be more active (at least 150 minutes a week). So the placebo result shows roughly what that counseling achieved in that trial, and the gap between the arms is the extra effect of the medicine.
• Results are a percentage of starting weight. A 20% loss for someone who starts at 230 pounds is about 46 pounds. Most participants had a body mass index (BMI) of 30 or more, or 27 or more with a weight-related condition, and average starting weights were around 100 to 114 kg (about 220 to 250 pounds).
• The dose went up slowly. Tirzepatide starts at 2.5 mg and is raised in steps over up to 20 weeks, so side effects were most common early on and the full effect took months to appear.
• There are two ways to report the result. The main analysis in these trials (the “treatment-regimen estimand”) counts everyone who was randomized, including people who stopped the drug. A second analysis estimates what happened if everyone kept taking it as planned, and it gives a larger number. In a ReachMD interview, SURMOUNT-1’s lead author, Dr. Ania Jastreboff of Yale, described 20.9% for the 15 mg dose as “the more conservative” estimate; in a Yale magazine interview she cited 22.5%. Both are real, but they answer slightly different questions. This guide uses the main analysis unless it says otherwise.
• “Compared with placebo” versus “total.” The FDA’s 2023 approval announcement said people on 15 mg lost on average 18% of their body weight “compared to those randomized to placebo.” That is the difference between the arms (20.9% minus 3.1%, rounded), not the total loss. Both framings are accurate.
• Eli Lilly, which makes tirzepatide, funded all of these trials. They were still randomized, mostly double-blind and published in major peer-reviewed journals, but it is worth knowing who paid.

The main trials at a glance

Trial	Who took part	Length	Tirzepatide result	Placebo or comparison
SURMOUNT-1	2,539 adults without diabetes	72 weeks	−15.0% (5 mg); −19.5% (10 mg); −20.9% (15 mg)	−3.1%
SURMOUNT-1, 3 years	1,032 of those adults who had prediabetes	176 weeks	−12.3% (5 mg); −18.7% (10 mg); −19.7% (15 mg)	−1.3%
SURMOUNT-2	938 adults with type 2 diabetes	72 weeks	−12.8% (10 mg); −14.7% (15 mg)	−3.2%
SURMOUNT-3	579 adults who first lost 5% or more with intensive lifestyle	72 more weeks	−18.4% more	+2.5%
SURMOUNT-4	670 adults who had lost 20.9% in 36 weeks	52 more weeks	−5.5% more (kept taking it)	+14.0% (switched to placebo)
SURMOUNT-5	751 adults without diabetes	72 weeks	−20.2%	−13.7% (semaglutide)
SURMOUNT-OSA	469 adults with moderate to severe sleep apnea	52 weeks	Breathing events: −25.3 to −29.3 per hour	−5.3 to −5.5 per hour
SUMMIT	731 adults with heart failure (preserved ejection fraction)	Median 104 weeks	CV death or worsening heart failure: 9.9%	15.3%
SURPASS-CVOT	13,165 adults with type 2 diabetes and heart disease	About 4 years	Heart attack, stroke or CV death: 12.2%	13.1% (dulaglutide)

Figures are average changes in body weight from the start of each trial (from randomization for SURMOUNT-3 and SURMOUNT-4), using each paper’s main analysis. Different trials enrolled different people, so do not compare rows directly.

SURMOUNT-1: the headline trial

SURMOUNT-1, published in the New England Journal of Medicine in 2022, enrolled 2,539 adults without diabetes and randomly assigned them in equal numbers to weekly tirzepatide 5 mg, 10 mg or 15 mg, or placebo, for 72 weeks, including up to 20 weeks of dose increases. Their average weight was 104.8 kg (about 231 pounds), their average BMI was 38, about two-thirds were women and the average age was about 45.

Result at week 72	5 mg	10 mg	15 mg	Placebo
Average weight change	−15.0%	−19.5%	−20.9%	−3.1%
Lost at least 5%	85%	89%	91%	35%
Lost at least 10%	68.5%	78.1%	83.5%	18.8%
Lost at least 15%	48.0%	66.6%	70.6%	8.8%
Lost at least 20%	30.0%	50%	57%	3%

Percentages for the 10% and 15% thresholds, and for 20% on 5 mg, come from the Zepbound prescribing information, which reports the same trial (as “Study 1”).

Waist size fell by 14.0 to 18.5 cm on tirzepatide versus 4.0 cm on placebo, and systolic blood pressure fell by about 6.6 to 7.7 points versus 1.0. The authors reported improvements in all prespecified heart and metabolic measures.

Side effects leading to stopping the drug occurred in 4.3%, 7.1% and 6.2% of people on 5, 10 and 15 mg, versus 2.6% on placebo. Overall, fewer people on tirzepatide than on placebo stopped taking their study drug for any reason (about 14% to 16% versus 26.4%), which is common in weight-loss trials where placebo does not work as well.

The table also shows the other side of the averages: about one in ten people on the highest dose did not reach a 5% loss. A 2026 post hoc analysis of SURMOUNT-1 and SURMOUNT-2 found that people who had lost at least 5% by week 8 tended to lose more by week 72 than those who had not, although both groups saw meaningful improvements and had similar digestive side effects. Early response is a clue, not a verdict.

Three years: SURMOUNT-1 in people with prediabetes

About 1,032 of the SURMOUNT-1 participants also had prediabetes, and they stayed on their assigned treatment for 176 weeks (just over three years), followed by 17 weeks off treatment.

• Average weight change at three years: −12.3% (5 mg), −18.7% (10 mg) and −19.7% (15 mg), versus −1.3% on placebo.
• Diagnosed with type 2 diabetes during treatment: 1.3% on tirzepatide versus 13.3% on placebo.
• After 17 weeks off treatment: 2.4% versus 13.7%.

That is a large difference, and it is the kind of result clinicians care about beyond the scale. Two cautions: these were people who stayed in a long trial, and tirzepatide does not have an FDA-approved use specifically for preventing diabetes. Our GLP-1 weight-loss timeline explains how weight typically changes over the first year and beyond.

SURMOUNT-2: people with type 2 diabetes

SURMOUNT-2 enrolled 938 adults with type 2 diabetes and a BMI of 27 or more (average age 54, average weight 100.7 kg) and compared tirzepatide 10 mg, 15 mg and placebo for 72 weeks.

• Average weight change: −12.8% (10 mg), −14.7% (15 mg) and −3.2% (placebo).
• Lost at least 5%: 79% to 83% versus 32%.
• Lost at least 20%: 21.5% (10 mg) and 30.8% (15 mg) versus 1.0%, according to the Zepbound label.
• Fewer than 5% stopped because of side effects.
• Low blood sugar (below 54 mg/dL) was reported in 4.2% on tirzepatide versus 1.3% on placebo, according to the label; the risk is higher when tirzepatide is combined with some diabetes medicines.

As with semaglutide and other GLP-1 medicines, people with type 2 diabetes tend to lose less weight than people without it, even at the same dose. Our Zepbound vs Mounjaro guide explains why the same molecule has two brand names and how their trial numbers differ.

SURMOUNT-3: after an intensive lifestyle program

SURMOUNT-3 asked a practical question: if someone has already lost weight with diet and exercise, does tirzepatide add more?

All 806 participants first followed a 12-week intensive lifestyle program: eight sessions with a dietitian, a target of about 1,200 calories a day for women and 1,500 for men, and at least 150 minutes of exercise a week. The 579 who lost at least 5% (6.9% on average) were then randomly assigned to tirzepatide at the highest dose they could tolerate (10 or 15 mg) or placebo for 72 weeks.

• Additional weight change after randomization: −18.4% versus +2.5%.
• Lost at least another 5%: 87.5% versus 16.5%.
• Stopped because of side effects: 10% versus 2%, a higher rate than in SURMOUNT-1.

The placebo group is the striking part: after a successful lifestyle program, people on placebo slowly regained weight despite continued counseling, while people on tirzepatide kept losing. The lesson is not that diet and activity do not work. It is that biology pushes back after weight loss, as our guide to why weight-loss plateaus happen explains.

SURMOUNT-4 and SURMOUNT-MAINTAIN: what happened after stopping

These two trials matter most for anyone thinking about the long term.

SURMOUNT-4 started 783 adults without diabetes on open-label tirzepatide at the highest tolerated dose for 36 weeks. The 670 who completed that phase had lost an average of 20.9%. They were then randomly assigned, without knowing which, to keep taking tirzepatide or to switch to placebo for 52 more weeks.

• Kept taking tirzepatide: lost another 5.5%.
• Switched to placebo: regained 14.0%.
• Kept at least 80% of their earlier weight loss: 89.5% versus 16.6%.
• Total change from the very start (week 0 to 88): −25.3% versus −9.9%.

So the placebo group still weighed less than at the start, but they had regained a large share of what they lost. A 2026 post hoc analysis of 308 people switched to placebo found that waist size, blood pressure, non-HDL cholesterol and blood sugar worsened step by step with the amount of weight regained. About one in four (74 people) regained 75% or more of what they had lost, while a smaller group (54 people) regained less than a quarter.

SURMOUNT-MAINTAIN, published in The Lancet in 2026, tested whether a lower dose could be enough to hold the loss. After a 60-week weight-loss phase, 378 adults were assigned to keep their full dose, step down to 5 mg, or switch to placebo for 52 weeks.

• Average weight change from the start to week 112: −21.9% (full dose), −16.6% (5 mg) and −9.9% (placebo).
• People who regained at least half of their lost weight could restart full treatment as “rescue.” That happened to 8% on the full dose, 25% on 5 mg and 67% on placebo.

The authors concluded that long-term treatment is often needed to maintain weight reduction and that a lower dose “might provide a valuable alternative to discontinuation,” while noting that individual responses vary. Stopping, tapering or stepping down is a conversation to have with your prescriber, not something to plan from a trial average. Our guide on how to maintain weight loss covers what research shows about keeping weight off, with or without medicine.

SURMOUNT-5: tirzepatide versus semaglutide

Most tirzepatide trials compared it with placebo. SURMOUNT-5 is the one large head-to-head trial against semaglutide, the active ingredient in Wegovy. It enrolled 751 adults with obesity and no diabetes, who took the highest dose they could tolerate of either drug (tirzepatide 10 or 15 mg; semaglutide 1.7 or 2.4 mg) for 72 weeks.

• Average weight change: −20.2% with tirzepatide versus −13.7% with semaglutide.
• Waist size: −18.4 cm versus −13.0 cm.
• People on tirzepatide were more likely to reach every threshold tested: at least 10%, 15%, 20% and 25%.
• Digestive side effects were the most common problem with both, mostly mild to moderate and during dose increases.

Two caveats. The trial was open-label (people knew which drug they were taking) and was funded by Eli Lilly, which makes tirzepatide. And semaglutide’s result here was lower than its 14.9% in STEP 1, a reminder that numbers from separate trials should not be lined up side by side. Our Semaglutide Clinical Trial Results guide covers the other side, and the Semaglutide vs Tirzepatide roundup puts both in context.

SURMOUNT-OSA: sleep apnea

SURMOUNT-OSA was really two trials in 469 adults with obesity and moderate to severe obstructive sleep apnea, a condition in which breathing repeatedly stops or becomes shallow during sleep. One trial enrolled people who were not using a positive airway pressure (PAP) machine such as a CPAP; the other enrolled people who were. Participants took tirzepatide (10 or 15 mg) or placebo for 52 weeks. On average they started with about 50 breathing interruptions per hour.

Result at week 52	Not using PAP: tirzepatide	Placebo	Using PAP: tirzepatide	Placebo
Change in breathing events per hour	−25.3	−5.3	−29.3	−5.5
Remission or mild sleep apnea without daytime sleepiness	42.2%	15.9%	50.2%	14.3%
Weight change	−17.7%	−1.6%	−19.6%	−2.3%

Blood pressure, a marker of inflammation and sleep-related symptoms also improved. These results are the basis for Zepbound’s FDA-approved use to treat moderate to severe obstructive sleep apnea in adults with obesity. The label notes that the trials did not evaluate whether or when people who already use PAP should stop, so that decision belongs with a sleep specialist.

SUMMIT: heart failure with preserved ejection fraction

SUMMIT enrolled 731 adults with heart failure with preserved ejection fraction (a common type in which the heart pumps normally but is stiff) and a BMI of 30 or more. They took tirzepatide (up to 15 mg) or placebo, with a median follow-up of 104 weeks.

• Cardiovascular death or a worsening heart-failure event: 9.9% versus 15.3% (hazard ratio 0.62, a 38% lower relative risk).
• Worsening heart-failure events alone: 8.0% versus 14.2%.
• Cardiovascular death alone: 2.2% versus 1.4%, a small number of events that showed no clear difference either way.
• Quality-of-life score (on a 100-point heart-failure questionnaire) improved by 19.5 points versus 12.7.
• Stopped because of side effects (mainly digestive): 6.3% versus 1.4%.

SURPASS-CVOT: heart outcomes in type 2 diabetes

SURPASS-CVOT is the largest tirzepatide trial: 13,165 adults with type 2 diabetes and established heart or blood vessel disease (average age 64, 29% women). Instead of placebo, it compared tirzepatide (up to 15 mg) with dulaglutide 1.5 mg, an older weekly GLP-1 medicine already shown to reduce cardiovascular events.

• Heart attack, stroke or cardiovascular death: 12.2% versus 13.1% (hazard ratio 0.92).
• That met the trial’s test for non-inferiority (tirzepatide was not worse) but not its test for superiority (P = 0.09).
• Adverse events were broadly similar, with more digestive side effects on tirzepatide.

In short, tirzepatide performed at least as well as an established heart-protective drug in this group. This trial says nothing directly about heart outcomes in people with obesity who do not have diabetes. For how this compares with semaglutide’s SELECT trial, see What Is Tirzepatide? and the Semaglutide hub.

Side effects across the trials

The pattern matched other GLP-1 medicines: digestive side effects were common, usually mild to moderate, and mostly happened while the dose was being raised. The Zepbound label pools SURMOUNT-1 and SURMOUNT-2:

Side effect	Placebo	5 mg	10 mg	15 mg
Nausea	8%	25%	29%	28%
Diarrhea	8%	19%	21%	23%
Vomiting	2%	8%	11%	13%
Constipation	5%	17%	14%	11%
Stomach pain	5%	9%	9%	10%
Injection site reactions	2%	6%	8%	8%
Hair loss	1%	5%	4%	5%

• Any digestive side effect: 56% on each dose versus 30% on placebo. Between 1.9% and 4.3% stopped because of them, versus 0.5% on placebo.
• Hair loss was linked to weight reduction and reported far more often in women (7.1%) than men (0.5%) on tirzepatide. No one on tirzepatide stopped because of it.
• Gallbladder problems: gallstones in 1.1% versus 1%, gallbladder inflammation in 0.7% versus 0.2%.
• Pancreatitis confirmed by review: 0.2% in both groups in these two trials.
• Low blood pressure: 1.6% versus 0.1%, more often in people taking blood pressure medicines.
• Heart rate rose by an average of 1 to 3 beats per minute.

The label also warns that tirzepatide caused thyroid C-cell tumors in rats (it is not known whether this happens in humans) and lists other serious risks. For practical guidance, see our guides to GLP-1 nausea, constipation and diarrhea and GLP-1s and gallbladder problems, or the Side Effects hub.

What the trials do not tell you

• Your personal result. Averages include people who lost a third of their weight and people who lost almost nothing. Diabetes status, sex, age, dose and how consistently the medicine is taken all play a part. A post hoc analysis of SURMOUNT-1 and SURMOUNT-4 found that most people reached a plateau by week 72, with a median of roughly 24 to 36 weeks depending on starting BMI; women, younger people and those on higher doses tended to keep losing for longer.
• How muscle and fat change for each person. People lose some lean mass along with fat. A 2026 review concluded that lean-mass loss on these medicines is generally proportional to the amount of weight lost and should not be read as direct evidence of muscle damage, but that protein, resistance exercise and attention to older adults matter. See how much protein you need on a GLP-1 and strength training on GLP-1s.
• Very long-term effects. The longest weight data run to about three years. Effects over decades are still being studied.
• Results with compounded or unapproved products. The trials used the manufacturer’s medicine. They say nothing about the quality of products from other sources.
• Everyday care. Real-world patients are more varied, get less support and often face cost and insurance barriers. Our GLP-1 costs hub covers that side.

What people are saying online

On X, the tirzepatide conversation this month has centered on what happens after stopping. Several posts cited SURMOUNT-MAINTAIN’s finding that two-thirds of people switched to placebo needed rescue treatment, and one used it to push back on a commercial weight-loss program’s claim about keeping weight off after GLP-1s, pointing out that the claim came with no published method, sample size or study. A physician shared the three-year SURMOUNT-1 prediabetes results, summarizing them as new diabetes cases falling from about 13 in 100 people to about 1 in 100. Other posts compared tirzepatide’s numbers with figures from newer experimental drugs; the more careful replies noted that different trials and populations make such comparisons unreliable.

On YouTube, physicians such as Dr. Kathleen Hallinan explain tirzepatide as “two medications in one”: a GLP-1 receptor activator like older drugs in the class, plus activity at the GIP receptor, which is found in the brain, pancreas and fat tissue. Her video asks whether this is lifelong therapy, the same question SURMOUNT-4 and SURMOUNT-MAINTAIN were designed to answer. Scientists still debate exactly how the GIP part contributes; a 2026 review notes that the human evidence mainly supports reduced appetite and food intake as the drivers of weight loss. Personal stories online help you understand experiences, but for how well the medicine works on average, the trial numbers above are more reliable.

Questions to ask your prescriber

1. Which of these trials included people most like me (with or without diabetes, sleep apnea or heart disease)?
2. What would count as a good response for me, and when would we judge it?
3. How should I expect side effects to change as the dose goes up?
4. If I want to stop or lower the dose later, what does the evidence suggest, and how would we plan for it?
5. Would tirzepatide or semaglutide suit me better, considering side effects, coverage and my other conditions?
6. What should I do about protein, exercise and muscle while I am losing weight?

The bottom line

Tirzepatide’s trial record is large, consistent and, on average, impressive. In adults without diabetes, weekly tirzepatide with lifestyle counseling led to average weight loss of about 15% to 21% over 72 weeks depending on dose, versus about 3% on placebo, with smaller results in people with type 2 diabetes. Results held for up to three years while people stayed on it, but people switched to placebo regained much of the weight. In the only large head-to-head trial, it outperformed semaglutide. It also improved sleep apnea, reduced heart-failure events in one specific group and matched an established heart-protective drug in people with diabetes and heart disease. Digestive side effects were common throughout. The averages are a useful guide, but your own result and the right choice for you depend on your health and your prescriber’s advice.

Explore more in the Tirzepatide hub, compare options in the GLP-1 Guide hub, or see the Semaglutide hub for its trial results. You can also check where you stand with the BMI calculator.