Drug facts

Not losing weight on tirzepatide?

By David Chen·Editor·Reviewed September 2026

How often it did not work¹

SURMOUNT-1’s second co-primary endpoint was the share of participants who lost 5% or more of their body weight over 72 weeks. The share who did not is simply the rest:

The first column is the trial’s own figure; the second is 100 minus it. These count all 2,539 people randomized, whether or not they kept taking the drug. The placebo row is worth reading too: about a third of people lost 5% on placebo, with the reduced-calorie diet and activity counseling every participant received.⁵

At a higher bar the spread is wider. The trial reported that 57% of the 15 mg group lost 20% or more, which is also to say that 43% did not. An average of 20.9% sits inside a wide spread of individual results. How the loss built up week by week →

A slow start is not no response²

A post hoc analysis of SURMOUNT-1 followed 1,545 participants who took at least 75% of their assigned doses. At week 12, 278 (18.0%) had lost less than 5%. The authors called them late responders, and tracked what happened next:

• 70% of them had lost 5% or more by week 24, four weeks after the escalation period ended;
• 90% had by week 72;
• among those who reached it, the average time to 5% was 24.8 weeks;
• at week 72 they averaged an 11.0% reduction, against 22.5% for those who had already lost 5% at week 12. By dose, the late responders averaged 8.3%, 11.4% and 13.6% at 5, 10 and 15 mg.

The paper’s own conclusion is that extending treatment well beyond 12 weeks may allow additional patients to reach a clinically meaningful reduction. Why the first weeks are slow by design is on how long tirzepatide takes to work.

Who stayed under 5%²

In the same analysis, 28 of the 1,545 participants (1.8%) still had less than a 5% reduction at week 72. Thirteen of the 28 were in the 5 mg group, and 4 had missed a run of consecutive doses and gone back through dose escalation, which the authors note lowered their exposure to the assigned dose.

The paper also reports how the late responders differed at the start. Compared with early responders they were more often male (45% against 30%), heavier (110.2 kg against 103.6 kg) and more likely to have hypertension (45% against 30%). The authors present these as differences seen after the fact, call the p-values nominal, and say further research is needed; the analysis could not assess diet or physical activity.

The plateau³

A separate SURMOUNT-1 analysis looked at when weight loss levels off. It defined a plateau as a change of less than 5% over a 12-week interval and every interval after it, and studied 1,438 participants who took at least 75% of their doses and lost at least 5% by week 72.

• Weight reduction was greatest in the first 24 weeks.
• The median time to a plateau was 24.3, 26.0, 36.1 and 36.1 weeks for participants who started in the overweight range and in obesity classes I, II and III.
• By week 72, 87.6% to 90.2% of participants in each group had reached a plateau.
• Higher doses, younger age and female sex went with a later plateau: on average 4.4 and 6.7 weeks later at 10 and 15 mg than at 5 mg, and 4.2 weeks later in women than in men.
• A later plateau went with more weight lost: participants who plateaued at 24 weeks had lost 12.1% at that point, and those who plateaued between 36 and 60 weeks had lost 24.3%.

The authors point out that a plateau is often misread as the end of the treatment’s benefit, or as the drug losing its effect. The trial that tested that directly is SURMOUNT-4.⁴ After 36 weeks on tirzepatide, adults randomized to keep taking it lost a further 5.5% over the next 52 weeks, while those switched to placebo regained 14.0%. The trial is summarized on SURMOUNT-4.

Type 2 diabetes changes the numbers⁵

SURMOUNT-1 excluded diabetes. The Zepbound label reports a second 72-week trial, Study 2, in adults with type 2 diabetes, and the results were lower than SURMOUNT-1’s at the same doses. Mounjaro, the same molecule under a type 2 diabetes label⁶, is used in exactly this population:

A person with type 2 diabetes comparing themselves with the SURMOUNT-1 averages is comparing against the wrong trial.

Reasons that are about the vial

Everything above describes the approved product at a known dose. With a compounded vial there are further reasons a result can fall short that have nothing to do with how a body responds: a concentration that makes the syringe mark mean a different dose, a vial that arrived warm, or a gap in supply. Those are set out, with FDA’s own words on dosing errors, in why compounded tirzepatide can seem not to work, and the arithmetic is in the dose-in-units calculator.

What none of this covers⁵

The label leaves the next step to the prescriber, in its own words: “Consider treatment response and tolerability when selecting the maintenance dosage.” If price is what is interrupting a course, the price index compares what each seller really charges per month.

Nothing here is medical advice, and this page deliberately suggests no change to dose, diet or treatment. Whether a result is a slow start, a plateau or a reason to change course is a question for the prescriber who knows your history. See the medical disclaimer.