SURMOUNT-4: what happens when tirzepatide is stopped

If you take one number away from this site, this is a candidate. SURMOUNT-4 is the trial that measured what happens to weight when tirzepatide is withdrawn.

What the trial asked

Whether continuing tirzepatide, compared with switching to placebo, maintains the weight reduction achieved during an initial treatment period.

Design

Phase 3 randomised withdrawal trial at 70 sites in four countries. Everyone first received open-label tirzepatide at maximum tolerated dose (10 or 15 mg) for 36 weeks. At week 36, participants were randomised 1:1 to continue tirzepatide or switch to placebo for 52 more weeks, double-blind. Diet and physical activity counselling continued throughout.

Who was enrolled

783 adults with a BMI of 30 or more, or 27 or more with a weight-related complication, excluding diabetes, entered the lead-in. 670 were randomised: 335 to continue tirzepatide, 335 to placebo. Mean age 48; 71 percent women. The abstract gives a mean weight of 107.3 kg for the randomised participants; it does not say whether this refers to week 0 or week 36, so treat it as reported rather than as a baseline you can do arithmetic on.

Primary endpoint and result

Primary endpoint: mean percent change in weight from week 36 (randomisation) to week 88.

Difference for the primary endpoint: -19.4 percent (95% CI -21.2 to -17.7; P less than 0.001).

Key secondary results

At week 88, 300 participants (89.5 percent) who continued tirzepatide had kept at least 80 percent of the weight they lost during the lead-in, compared with 16.6 percent of those switched to placebo (P less than 0.001).

Adverse events and discontinuation

The abstract states that the most common adverse events were mostly mild to moderate gastrointestinal events, more common with tirzepatide than placebo. It does not give discontinuation percentages, so none are reported here.

Funding and registration

The abstract does not name the funder, so this digest does not either. The trial is registered under ClinicalTrials.gov NCT04660643, where the sponsor is listed.

What this trial does not show

• Withdrawal was abrupt and to placebo. The trial did not test a gradual taper, a reduced dose, or a switch to another agent.
• The 52-week placebo period was long enough to show substantial regain, but not long enough to say whether weight returns all the way to baseline. At week 88 the placebo group was still 9.9 percent below their week 0 weight.
• Everyone randomised had tolerated 36 weeks of tirzepatide. People who stopped during the lead-in are not represented.
• It does not report whether the regained weight was fat or lean mass.

Where the numbers come from

PubMed abstract of the primary paper (PubMed 38078870), fetched 4 September 2026. The semaglutide equivalent, with a shorter run-in, is STEP 4. For what withdrawal data mean in practice, the sibling site Weight Loss on GLP-1s covers maintenance planning.