Tirzepatide weight loss results in the landmark obesity trial averaged 15.0% to 20.9% at 72 weeks, depending on dose, versus 3.1% with placebo. The change was not all at once: weight loss was greatest early, then slowed, and most participants in a later analysis reached a plateau by week 72. The first month is an initiation phase, so early change is usually modest and is not a reliable forecast of the final result. These are group averages from controlled trials, not a promise of what any one person will lose.
Tirzepatide weight loss results at 72 weeks
SURMOUNT-1 enrolled 2,539 adults with obesity, or overweight plus a weight-related complication, who did not have diabetes. Participants were assigned to a 5 mg, 10 mg, or 15 mg maintenance dose or placebo, alongside lifestyle intervention. The trial included a 20-week dose-escalation period.
At week 72, the treatment-regimen estimates in the 2022 New England Journal of Medicine report were:
| Assigned maintenance dose | Mean weight change | Lost at least 5% | Lost at least 20% |
|---|---|---|---|
| Tirzepatide 5 mg | -15.0% | 85.1% | 30.0% |
| Tirzepatide 10 mg | -19.5% | 88.9% | 50.1% |
| Tirzepatide 15 mg | -20.9% | 90.9% | 56.7% |
| Placebo | -3.1% | 34.5% | 3.1% |
The most honest reading is a range, not a single headline number. A 20.9% group average does not mean every participant lost 20.9%, and the threshold columns show that responses varied even within the same dose arm. The study population, follow-up, lifestyle support, and analysis method also matter. The tirzepatide clinical-trials overview maps the wider program; this page stays with the weight curve and its practical meaning.
What do tirzepatide results look like after one month?
One-month tirzepatide results are usually modest compared with the 72-week averages. The first four weeks in the studied schedule use 2.5 mg as an initiation dose, before any later increase a prescriber may choose. It is a tolerability runway, not the maintenance-dose phase that produced the trial's final numbers.
SURMOUNT-1's published paper does not tabulate a single formal average for week 4. Its plotted curve shows weight beginning to move down, but still far from the later result. Attaching a precise one-month percentage to every patient would create certainty the trial did not report. The relevant signal at that point is direction over several measurements, not whether a person has already matched an online anecdote.
The full tirzepatide dosage schedule explains how escalation decisions are made. The reviewing prescriber sets the actual dose and pace.
When the curve is steepest—and when it flattens
The curve generally falls faster early and becomes shallower with time. A 2025 post hoc analysis of SURMOUNT-1 and SURMOUNT-4 found that weight reduction was greatest during the first 24 weeks. It defined a plateau as less than 5% change across a 12-week interval and every later 12-week interval.
Among selected participants who adhered to treatment and had lost at least 5% by the trial endpoint, median time to plateau ranged from 24.3 to 36.1 weeks across starting BMI categories. By week 72, roughly 88% to 90% had reached that definition of plateau (Horn et al., Clinical Obesity 2025). That does not mean loss stops abruptly in month 6 or 9. It means the early slope eases, and by about a year the average curve is much flatter than it was at the start.
Tirzepatide versus semaglutide results
SURMOUNT-5 provides the cleanest head-to-head result because both medicines were tested in the same 72-week trial. Among 751 adults with obesity but without diabetes, the mean change was -20.2% with maximum-tolerated tirzepatide, compared with -13.7% with maximum-tolerated semaglutide, a 6.5-percentage-point difference (Aronne et al., NEJM 2025).
Those numbers apply to the trial's population and dose ranges, not to every person starting either medication. The broader tirzepatide-versus-semaglutide comparison owns the differences beyond weight outcomes.
What happens to the curve when tirzepatide stops?
SURMOUNT-4 tested withdrawal rather than asking participants to remember what happened. Everyone began with 36 weeks of tirzepatide at a maximum tolerated dose; the 670 participants who reached randomization had lost 20.9% on average. They then either continued tirzepatide or switched to placebo for 52 weeks.
From week 36 to week 88, the continuation group lost another 5.5% on average, while the placebo-switch group regained 14.0% from its week-36 weight. At week 88, 89.5% of those continuing tirzepatide had kept at least 80% of their initial loss, versus 16.6% after the switch (Aronne et al., JAMA 2024).
This design does not predict exactly what any individual will regain. It does show that the on-treatment curve should not be treated as a permanent reset after medication ends. Maintenance is a clinical planning question, not evidence that someone failed.
Why an individual result may differ
Starting weight, biology, tolerated dose, time on treatment, adherence, other medications, and whether diabetes is present can all move the result away from the trial mean. SURMOUNT-1 and SURMOUNT-5 excluded diabetes, so their headline percentages should not be casually transferred to every clinical population. A flat week or a sharp early drop also says little by itself; the trials measured trends over many months.
The evidence above came from branded products used under trial protocols. Through Promise, tirzepatide is dispensed as a compounded medication, which is different from the FDA-approved brand products Mounjaro and Zepbound: the formulation offered here is not FDA-approved. FDA review status is one fact in a clinical decision: at Promise, a licensed provider reviews every request and may prescribe a compounded formulation when medically appropriate, but not everyone qualifies; that decision belongs to the patient and reviewing provider.
The clearest personal comparison is a starting-weight percentage viewed across several weeks, with tolerability and the agreed clinical plan beside it. Trial averages are useful reference points. They are not deadlines or guarantees.