The most useful comparison between semaglutide and tirzepatide now comes from a trial that put them in the same study. That is a substantial improvement over placing two unrelated trial averages beside each other. It also creates a new reading challenge: the headline result belongs to specific regimens and a specific population, rather than to every product containing either molecule.
Semaglutide activates the GLP-1 receptor. Tirzepatide activates both GIP and GLP-1 receptors. These mechanisms help explain why both are studied for metabolic conditions, but counting receptors does not establish how much benefit an individual will obtain. A clinical comparison needs outcomes, tolerability, approved uses and a realistic plan for ongoing care.
What SURMOUNT-5 actually compared
SURMOUNT-5 randomly assigned 751 adults with obesity without type 2 diabetes to once-weekly tirzepatide or semaglutide for 72 weeks. It was open label: participants and investigators knew which treatment was assigned. The trial compared maximum tolerated tirzepatide doses of 10 or 15 mg with semaglutide doses of 1.7 or 2.4 mg. These are research descriptions, not instructions for selecting or switching a dose. The original trial abstract sets out those boundaries.
Mean weight change was minus 20.2% with tirzepatide and minus 13.7% with semaglutide. The difference is 6.5 percentage points of starting weight, not a promise of an extra 6.5 kilograms for every person. The trial also found a greater average waist reduction with tirzepatide. Both groups experienced gastrointestinal adverse events, mainly mild to moderate and concentrated during escalation. A better group average establishes a treatment effect in this comparison; it does not identify the best treatment for everyone who reads it.
Randomization matters because it makes the two groups more comparable at the start. Shared follow-up and outcome definitions remove many problems that arise when comparing separate studies. Open labeling remains a limitation for experiences and decisions influenced by knowing the treatment, even when measured body weight is relatively objective. Funding by Eli Lilly should also remain visible alongside the trial design rather than treated as either proof of invalidity or a reason to ignore possible bias.
The result does not cover every semaglutide formulation
The comparator in SURMOUNT-5 was injectable semaglutide at the studied doses. It was not oral Wegovy, Ozempic used for diabetes, or the newer Wegovy HD presentation. A headline that turns the study into a verdict on all semaglutide products quietly changes the question. Those other formulations or indications need their own evidence.
The current US Wegovy label, reviewed on 3 October 2026, lists injection and tablet presentations with formulation-specific indications. Injection includes defined cardiovascular, weight-management and MASH uses; tablet indications are not identical to the injection list. Our Wegovy and Ozempic comparison explains why a shared molecule does not erase differences between approved products. The existence of a higher-dose presentation also does not retroactively change which dose was tested in the direct comparison.
The same discipline applies to tirzepatide. The US Zepbound label supports adult weight management and moderate-to-severe obstructive sleep apnea in adults with obesity. That does not mean SURMOUNT-5 measured sleep-apnea outcomes or established superiority for cardiovascular events. Weight, sleep breathing, kidney outcomes and heart events are different endpoints. A medicine can have strong evidence for one and a different level of evidence for another.
Diabetes status and follow-up change the question
The trial excluded type 2 diabetes. A person with diabetes may have different treatment goals, background medicines and safety considerations. Results from diabetes trials can inform that discussion, but they should not be folded into a single average without explaining the population. Similarly, the 72-week result is not a lifetime maintenance estimate or evidence about what happens after withdrawal.
Averages conceal variation. Two people starting at similar weights may have different responses, adverse effects or interruptions. A confidence interval around a group mean describes uncertainty about that mean; it is not the range in which each person’s eventual weight loss will fall. Threshold outcomes, withdrawals and longer follow-up help describe the distribution, but none supplies a personal forecast. Our trial-reading guide provides a way to check those details before relying on a headline.
Practical differences can outweigh a ranking
A comparison in the clinic starts with the intended indication and medical history. It should include pregnancy plans, other medicines, previous gastrointestinal problems, contraindications and the ability to obtain consistent follow-up. These are not administrative extras added after choosing the largest percentage. They determine whether a treatment plan can be used safely and sustained.
Access is another source of uncertainty. A list price, a pharmacy’s current stock or a promotional offer does not establish what someone will pay over a year. Ask about coverage requirements, expected ongoing costs and what happens if coverage changes. This article does not rank current prices because those figures vary by jurisdiction, insurance and date. Treatment interruptions should be discussed with the prescribing team; do not use a comparison article to invent a conversion between products.
Tolerability also needs more than a single event percentage. Ask whether symptoms were transient, led to dose changes, caused treatment discontinuation or required medical care. The number reporting nausea and the number stopping because of nausea answer different questions. A trial can report frequent manageable symptoms alongside a lower withdrawal rate, or the reverse. Discontinuation data explain one part of that experience.
A better way to use the direct comparison
Bring the trial to a consultation as a defined piece of evidence: in adults without diabetes, these tirzepatide regimens produced greater average weight reduction than these semaglutide regimens over 72 weeks. Then ask which goal the proposed prescription is intended to address and what evidence supports that goal for your situation. This avoids both dismissing a strong trial and asking it to answer questions it never tested.
It can help to write down three separate decisions: the outcome that matters, the safety factors that affect treatment, and the practical conditions needed to continue it. Weight reduction may matter alongside mobility, sleep symptoms or diabetes management. Follow-up should assess benefits and burdens together. The strongest comparison is useful because it narrows uncertainty; it should not replace that broader review.
US and EU indications remain separate. EMA’s Mounjaro record includes adult weight management as well as diabetes use, while US brands divide indications differently. Always check the country, brand and current label. The right conclusion from SURMOUNT-5 is a precise comparative finding, followed by a more informed care conversation.
