Stopping tirzepatide: what SURMOUNT-4 measured

The label's indication wording changed the way people should think about stopping before any trial did: Zepbound is indicated "to reduce excess body weight and maintain weight reduction long term". Maintenance is in the indication. SURMOUNT-4 is the trial that shows why.

The design

Seven hundred and eighty-three adults with obesity, or overweight plus a complication, and no diabetes took open-label tirzepatide for 36 weeks, escalated to the maximum tolerated dose of 10 or 15 mg. Everyone also got a standard lifestyle programme: a roughly 500 kcal/day deficit and at least 150 minutes of activity a week, per label section 14. At week 36, 670 people who were still on treatment were randomised, double-blind, to continue tirzepatide or switch to placebo for another 52 weeks. PubMed 38078870.

Two details from the label matter for interpretation. At randomisation 93% were on 15 mg and 7% on 10 mg, so this is a trial of stopping from the top dose. And 14.4% of the original 783 had already dropped out before week 36, 6.8% for adverse events, so the randomised group is people who tolerated the drug well enough to reach week 36.

The numbers

During the lead-in, mean weight fell 20.9%, from 107.3 kg to 85.2 kg.

From week 36 to week 88, people switched to placebo regained a mean of 14.0% of their week-36 weight. People who continued lost a further 5.5%. The difference between the arms was 19.4 percentage points (95% CI 17.7 to 21.2).

Over the whole 88 weeks, the placebo-switch group finished 9.9% below their starting weight; the continuing group finished 25.3% below.

At week 88, 89.5% of those who continued had kept at least 80% of their lead-in loss. Among those switched to placebo, 16.6% had.

What it means, and what it does not

The drug acts while it is present. Tirzepatide's half-life is about 5 days (label 12.3), so drug levels are low within a few weeks of the last dose, and the biology that produced the weight loss reverts. That is consistent with the trial: the regain curve in the placebo arm starts at the switch, not months later.

It does not mean regain is total. The placebo group's 88-week result, 9.9% below baseline, is comparable to what people achieve on long-term lifestyle programmes alone, and it came with the lifestyle programme still running. Nor does it mean regain is uniform; the 16.6% who kept most of their loss without the drug exist, and the trial does not tell us who they were in advance.

It does not tell you about tapering. The switch was abrupt, full dose to nothing. There is no arm that stepped down to 5 mg, or to every other week, so the data neither support nor refute those approaches. Anyone who tells you SURMOUNT-4 proves a taper works, or proves it does not, has gone beyond the trial.

It does not tell you about people who stopped early. Everyone randomised had done 36 weeks and reached a high dose. Stopping at week 10 on 5 mg is a different situation with no trial behind it.

What the label offers instead of a stopping plan

The label's tools are dose choices, not exit plans. Section 2.2 names 5, 10 and 15 mg as maintenance doses and says to consider response and tolerability in choosing one, and to consider a lower maintenance dose if a higher one is not tolerated. A person who wants less drug has, in label terms, the option of a lower maintenance dose; the label does not describe intermittent dosing or discontinuation schedules. The titration planner will not model a step-down for that reason.

If you are stopping anyway

Cost and supply are the usual reasons, and both have pages here: the self-pay price page tracks Lilly's dated prices and the savings-card page covers the commercial and Medicare programmes. If the reason is side effects, the label's own suggestion is a lower maintenance dose before discontinuation. Whatever the reason, the SURMOUNT-4 numbers are the realistic baseline for a conversation with your prescriber, and the lifestyle programme that ran through the whole trial is the part you keep.