Most people ask the same question once Mounjaro starts working: can I stop when I hit goal weight? SURMOUNT-4 tested it directly — and a 2025 follow-up traced exactly how much cardiometabolic benefit comes undone as the weight returns.
Obesity medicine has a persistent misconception: that once someone reaches their goal weight on a GLP-1 or dual agonist, the drug has "done its job" and can be stopped. SURMOUNT-4 was designed as a randomized withdrawal trial specifically to test what happens next — do the results hold, or fade, once tirzepatide (Mounjaro, Zepbound) is taken away?
The original results, published in JAMA in December 2023, already answered the headline question. A newer post hoc analysis, published in JAMA Internal Medicine in November 2025, goes further: it breaks down exactly how blood pressure, cholesterol, waist size, and blood sugar move depending on how much of the lost weight actually comes back.
Adults with obesity, or overweight plus a weight-related complication, first completed 36 weeks of open-label tirzepatide at the maximum tolerated dose (10 or 15 mg weekly). The 670 who finished that lead-in — mean age 48, 71% women, mean weight 107.3 kg — were then randomized 1:1 to either continue tirzepatide or switch to placebo for another 52 weeks, unaware which arm they were in.
That second phase is the withdrawal test: same people, same starting point, one group kept on the drug and one switched cold to placebo — with nothing else about their care standardized or restricted.
From week 36 to week 88, the placebo group regained a mean 14.0% of their body weight, while the group that stayed on tirzepatide lost an additional 6.7%. Counting from the very start of the trial, total weight change came out to −25.3% for tirzepatide versus −9.9% for placebo at week 88. Among those who kept taking the drug, 89.5% held onto at least 80% of the weight they'd lost during the lead-in — versus just 16.6% of those switched to placebo.
The November 2025 analysis sorted the 308 placebo-arm participants by how much weight they regained in the year after stopping. Among those who regained ≥75% of what they'd lost, waist circumference grew back 14.7 cm and systolic blood pressure rose 10.4 mmHg — nearly reversing the gains entirely. Among those who kept regain under 25%, waist circumference, non-HDL cholesterol, and fasting insulin were statistically no different from where they stood the day withdrawal began.
Sources: Horn et al., JAMA Internal Medicine, published online Nov. 24, 2025 (post hoc SURMOUNT-4 analysis, doi:10.1001/jamainternmed.2025.6112) — full analysis. Original trial: Aronne et al., JAMA, Dec. 11, 2023 — SURMOUNT-4 on PubMed.
Of the 308 people in the placebo arm, 82% regained more than a quarter of the weight they'd lost within a year — most of them well more than that. Sorted into four regain bands (under 25%, 25–50%, 50–75%, and 75%-plus), every cardiometabolic marker the researchers tracked got worse as the regain band went up: waist circumference (0.8 cm → 5.4 cm → 10.1 cm → 14.7 cm), systolic blood pressure (6.8 → 7.3 → 9.6 → 10.4 mmHg), non-HDL cholesterol (−0.4% → 1.6% → 8.4% → 10.8%), and HbA1c (0.14% → 0.15% → 0.27% → 0.35%), all with p-values below 0.002.
The people who kept regain small kept the benefit. That's the part easy to miss in headlines about "everyone regains weight" — this data shows regain isn't all-or-nothing, and neither is the loss of benefit. It's a dose-response relationship: less regain, less reversal.
None of this is an argument that stopping is always wrong — plenty of people taper off deliberately with their prescriber's guidance. But it is a reason not to stop the shot quietly on your own once the number on the scale looks good. Staying consistent with dosing, or working out a monitored plan for cutting back, matters more than the trial's topline "14% regain" headline suggests.
This article summarizes published trial data for general education. It is not medical advice and does not replace guidance from your prescriber. Decisions about starting, continuing, or stopping GLP-1 or dual-agonist therapy should be made with clinical supervision.