Key takeaways
- SURMOUNT-5 randomized 751 adults with obesity without diabetes to maximum-tolerated tirzepatide or semaglutide for 72 weeks; mean weight loss was 20.2% versus 13.7%, a 6.5-point absolute and 47% relative advantage for tirzepatide.
- Every responder threshold favored tirzepatide: 31.6% of tirzepatide participants lost at least 25% of body weight versus 16.1% on semaglutide.
- Gastrointestinal side effects were similar in frequency and mostly mild to moderate; discontinuation for adverse events was low in both arms.
- The trial was open-label, excluded people with diabetes, and did not test cardiovascular outcomes; semaglutide's SELECT trial remains the only outcomes evidence in this population.
- For most telehealth patients whose goal is weight loss, tirzepatide's extra $20 to $30 per month at the same provider buys roughly six and a half percentage points of additional average loss.
Why one trial matters this much
Until 2025, every comparison of tirzepatide and semaglutide relied on putting two separate trials side by side. SURMOUNT-1 gave tirzepatide 20.9% at 15 mg over 72 weeks; STEP 1 gave semaglutide 14.9% at 2.4 mg over 68 weeks. The populations were similar, but cross-trial comparisons are unreliable, and Novo Nordisk could reasonably say that no direct evidence showed tirzepatide was better. SURMOUNT-5, published in the New England Journal of Medicine in 2025, ended that argument. It is the one trial in which the same patients, recruited under the same criteria, were randomized to one drug or the other and followed the same way.
Design
SURMOUNT-5 enrolled 751 adults in the United States and Puerto Rico with a BMI of 30 or higher, or 27 or higher with at least one weight-related condition, and without diabetes. They were randomized one-to-one to once-weekly tirzepatide or once-weekly semaglutide for 72 weeks. Both arms used a treat-to-maximum-tolerated-dose design: tirzepatide was escalated to 10 or 15 mg and semaglutide to 1.7 or 2.4 mg, with clinicians permitted to hold a lower dose for tolerability. Both arms received the same lifestyle counselling. The trial was open-label, meaning participants and investigators knew which drug was being given, a limitation discussed below. The primary endpoint was percentage change in body weight at week 72.
The population was typical of obesity trials: mean age about 45, about 65% women, mean BMI about 39, mean weight around 113 kg. The exclusion of people with diabetes matters because diabetes blunts weight loss on both drugs; the result applies to the population most telehealth patients belong to.
The primary result
At 72 weeks, mean weight change was −20.2% on tirzepatide and −13.7% on semaglutide. The absolute difference was 6.5 percentage points, and the relative difference about 47%. For the average participant, that was roughly 22.8 kg (about 50 pounds) lost on tirzepatide versus 15.0 kg (about 33 pounds) on semaglutide. The confidence interval for the difference excluded zero by a wide margin, and the result was consistent across sex, age, baseline BMI and race subgroups.
The numbers line up closely with the separate trials: 20.2% here versus 20.9% in SURMOUNT-1 for tirzepatide; 13.7% here versus 14.9% in STEP 1 for semaglutide. That consistency is the strongest reason to trust the head-to-head result. The small shortfall relative to the single-drug trials reflects that not every participant reached the top dose.
Responder thresholds
Averages hide distributions, and the distribution is where SURMOUNT-5 is most informative for an individual patient. The share of participants losing at least 10% of body weight was about 82% on tirzepatide and 61% on semaglutide. At least 15%: 65% versus 40%. At least 20%: about 48% versus 27%. At least 25%: 31.6% versus 16.1%. Read another way, a patient on tirzepatide was roughly twice as likely to reach the 25% threshold, which is bariatric-surgery territory, and about half as likely to be a sub-10% responder.
Secondary endpoints
Waist circumference fell by about 18.4 cm on tirzepatide versus 13.0 cm on semaglutide. Improvements in blood pressure, triglycerides, non-HDL cholesterol and fasting insulin favored tirzepatide, in proportion to the greater weight loss rather than as an independent effect. HbA1c fell in both arms in a population that was, by design, non-diabetic. Body-composition data were not a primary focus of the trial; the SURMOUNT-1 sub-study had previously shown roughly 75% of tirzepatide weight loss was fat mass.
Side effects and dose reached
Gastrointestinal adverse events were the most common in both arms, at similar frequencies: nausea, diarrhea, constipation and vomiting, mostly mild to moderate and concentrated in the escalation period. Discontinuation because of adverse events was low and similar between arms. Serious adverse events were uncommon and balanced. The proportion of participants who reached and stayed on the top dose was somewhat higher in the tirzepatide arm, which may reflect GIP's apparent effect of moderating nausea, although the trial was not designed to test that.
No new safety signals appeared. Both drugs carry the same boxed warning for thyroid C-cell tumors, the same contraindication in medullary thyroid carcinoma and MEN 2, and the same class warnings for pancreatitis, gallbladder disease, hypoglycemia with insulin or sulfonylureas, kidney injury from dehydration, and gastroparesis.
What the trial did not show
It did not test cardiovascular outcomes. Semaglutide's SELECT trial, with 17,604 participants, showed a 20% reduction in major adverse cardiovascular events in people with obesity and established cardiovascular disease; tirzepatide's equivalent trial, SURMOUNT-MMO, is ongoing. For a patient with a prior heart attack or stroke, that asymmetry can outweigh 6.5 percentage points of weight loss.
It did not include people with type 2 diabetes. In that population, the earlier SURPASS-2 trial compared tirzepatide with semaglutide 1 mg (not the 2.4 mg obesity dose) and found greater HbA1c and weight reductions with tirzepatide; SURMOUNT-2 gave tirzepatide 12.8% to 14.7% in diabetes. A direct comparison at the obesity doses in diabetes has not been done.
It was open-label. Participants knew their drug, which can influence adherence and reporting of subjective side effects. Weight is an objective endpoint and is less vulnerable to that bias than symptoms are, but the limitation is real.
It did not test the semaglutide 7.2 mg dose studied in STEP UP (20.7% at 72 weeks), which was approved in Europe in 2025 and is under FDA review; a head-to-head at that dose has not been reported. It also did not test the drugs beyond 72 weeks, although STEP 5 and SURMOUNT-4 show both drugs maintain loss with continued use.
What it means for a telehealth patient
For a patient whose goal is weight loss and who can tolerate either drug, SURMOUNT-5 makes tirzepatide the evidence-based first choice. The cost difference through telehealth is small: about $20 to $30 per month at the same provider on the verified prices in our database, or $240 to $360 a year, for about 6.5 percentage points of additional average loss and a doubled chance of reaching 25%. The cost-per-percent article works that through for different body weights.
For a patient with established cardiovascular disease, semaglutide's outcome data are a reason to prefer it, or at least to discuss the choice with a clinician who knows the history. For a patient already on semaglutide and doing well, switching for the sake of the trial is unnecessary; for a patient who has plateaued on semaglutide, switching is common practice with a reasonable pharmacological rationale, and the plateau article covers how it is done.
For every patient, the trial is a reminder that dose matters. SURMOUNT-5's design escalated to the maximum tolerated dose; providers that hold patients at low doses, or sell "microdose" protocols, are delivering something the trials did not test. The microdosing article explains why that matters.
The regulatory footnote
Compounded tirzepatide and semaglutide, which most telehealth patients receive, are not the products SURMOUNT-5 tested. The trial used Lilly's tirzepatide and Novo's semaglutide. A well-compounded product from a licensed pharmacy contains the same peptide at the prescribed dose and would be expected to behave the same way; a poorly compounded, sub-potent or salt-form product would not. The trial's results apply to the molecule, and the compounding pharmacy guide explains how to make sure the molecule is what you are getting.
Frequently asked questions
Which is better for weight loss, tirzepatide or semaglutide?
Tirzepatide. In SURMOUNT-5 it produced 20.2% mean weight loss versus 13.7% for semaglutide 2.4 mg over 72 weeks, and more participants reached every weight-loss threshold. The difference was consistent across subgroups.
Is tirzepatide safer than semaglutide?
The side-effect profiles are similar. In SURMOUNT-5, gastrointestinal adverse events occurred at comparable rates in both arms and were mostly mild to moderate. Both drugs carry the same boxed warning and the same class warnings. Semaglutide has a completed cardiovascular outcomes trial in obesity (SELECT); tirzepatide's is ongoing.
Was SURMOUNT-5 a fair comparison?
It was randomized with a treat-to-maximum-tolerated-dose design in both arms, which is the fairest available comparison. It was open-label (participants knew their drug) and funded by Lilly, and its primary endpoint was weight, not cardiovascular events. Independent analyses have found the weight-loss result consistent with the separate SURMOUNT-1 and STEP 1 trials.
Sources
- Aronne LJ, Horn DB, le Roux CW, et al. Tirzepatide as Compared with Semaglutide for the Treatment of Obesity. N Engl J Med 2025;393:26-36.
- Jastreboff AM et al. Tirzepatide Once Weekly for the Treatment of Obesity. N Engl J Med 2022;387:205-216 (SURMOUNT-1).
- Wilding JPH et al. Once-Weekly Semaglutide in Adults with Overweight or Obesity. N Engl J Med 2021;384:989-1002 (STEP 1).
- Lincoff AM et al. Semaglutide and Cardiovascular Outcomes in Obesity without Diabetes. N Engl J Med 2023;389:2221-2232 (SELECT).
- Frías JP et al. Tirzepatide versus Semaglutide Once Weekly in Patients with Type 2 Diabetes. N Engl J Med 2021;385:503-515 (SURPASS-2).
- Zepbound (tirzepatide) prescribing information, Eli Lilly, 2025.
- Wegovy (semaglutide) prescribing information, Novo Nordisk, 2025.
Citations are to primary sources (peer-reviewed trials, FDA labeling and announcements, and provider pricing pages). See our sources policy and corrections log.