How much weight did SURMOUNT-4 participants regain after stopping tirzepatide?

Participants switched to placebo regained an average of 14.0% of body weight over 52 weeks, erasing roughly two-thirds of the 20.9% they lost during the 36-week open-label tirzepatide lead-in. Their net weight loss from original baseline was only 9.9% by week 88.

What was the primary endpoint result and confidence interval in SURMOUNT-4?

The estimated treatment difference in percent weight change from week 36 to week 88 was −19.5 percentage points (95% CI, −21.7 to −17.4; P <0.001), favoring continued tirzepatide over placebo switch.

How fast did weight regain begin after stopping tirzepatide?

Weight began increasing within the first 4 weeks of placebo switch. By approximately week 36 of the randomized period, participants had already regained about 10 of the eventual 14 percentage points. The regain curve partially flattened in the final 16 weeks.

Did all subgroups show the same pattern of weight regain?

Yes. The treatment difference was consistent across sex, baseline BMI, prediabetes status, and maximum tolerated dose (10 mg vs 15 mg). No prespecified subgroup showed a meaningfully different result.

What percentage of tirzepatide continuation patients maintained their weight loss?

Among those who continued tirzepatide, 89.5% maintained at least 80% of the weight lost during the lead-in, and 71.6% maintained 100% or more of that loss (meaning they continued losing weight beyond the lead-in amount).

How does SURMOUNT-4 differ from SURMOUNT-1 in design?

SURMOUNT-1 was a standard parallel-group trial comparing tirzepatide doses against placebo from day one. SURMOUNT-4 used a randomized withdrawal design: all participants received tirzepatide first, then responders were randomized to continue or switch to placebo. SURMOUNT-4 specifically answers the discontinuation question.

Were there safety differences between the continuation and placebo arms?

Gastrointestinal events (nausea, diarrhea, constipation) were roughly twice as common in the tirzepatide arm (33% vs 16%). Serious adverse event rates were similar between arms (6.9% vs 5.7%). No new safety signals emerged beyond the known tirzepatide profile.

Does SURMOUNT-4 apply to patients with type 2 diabetes?

Not directly. SURMOUNT-4 enrolled adults without diabetes. SURMOUNT-2 is the corresponding trial for the type 2 diabetes population. Subgroup analysis showed that prediabetes status did not change the discontinuation effect within SURMOUNT-4's non-diabetic cohort.

What do these results mean for insurance coverage of tirzepatide?

The data show that treatment interruptions predictably reverse most clinical benefit within months. This creates a clinical argument against coverage gaps, step-therapy requirements, or prior-authorization delays that force patients off therapy.

Did participants who continued tirzepatide still lose weight after 36 weeks?

Yes. The continuation arm lost an additional 5.5% of body weight between weeks 36 and 88, reaching a total of 25.3% from original baseline. Most of this additional loss occurred between weeks 36 and 60, with a near-plateau afterward.

SURMOUNT-4 Results in Detail: Numbers, Subgroups, and Time Course

Q: How fast did weight regain begin after stopping tirzepatide?

Weight began increasing within the first 4 weeks of placebo switch. By approximately week 36 of the randomized period, participants had already regained about 10 of the eventual 14 percentage points. The regain curve partially flattened in the final 16 weeks.

Q: Did all subgroups show the same pattern of weight regain?

Yes. The treatment difference was consistent across sex, baseline BMI, prediabetes status, and maximum tolerated dose (10 mg vs 15 mg). No prespecified subgroup showed a meaningfully different result.

Q: What percentage of tirzepatide continuation patients maintained their weight loss?

Among those who continued tirzepatide, 89.5% maintained at least 80% of the weight lost during the lead-in, and 71.6% maintained 100% or more of that loss (meaning they continued losing weight beyond the lead-in amount).

Q: How does SURMOUNT-4 differ from SURMOUNT-1 in design?

SURMOUNT-1 was a standard parallel-group trial comparing tirzepatide doses against placebo from day one. SURMOUNT-4 used a randomized withdrawal design: all participants received tirzepatide first, then responders were randomized to continue or switch to placebo. SURMOUNT-4 specifically answers the discontinuation question.

Q: Were there safety differences between the continuation and placebo arms?

Gastrointestinal events (nausea, diarrhea, constipation) were roughly twice as common in the tirzepatide arm (33% vs 16%). Serious adverse event rates were similar between arms (6.9% vs 5.7%). No new safety signals emerged beyond the known tirzepatide profile.

Q: Does SURMOUNT-4 apply to patients with type 2 diabetes?

Not directly. SURMOUNT-4 enrolled adults without diabetes. SURMOUNT-2 is the corresponding trial for the type 2 diabetes population. Subgroup analysis showed that prediabetes status did not change the discontinuation effect within SURMOUNT-4's non-diabetic cohort.

Q: What do these results mean for insurance coverage of tirzepatide?

The data show that treatment interruptions predictably reverse most clinical benefit within months. This creates a clinical argument against coverage gaps, step-therapy requirements, or prior-authorization delays that force patients off therapy.

Q: Did participants who continued tirzepatide still lose weight after 36 weeks?

Yes. The continuation arm lost an additional 5.5% of body weight between weeks 36 and 88, reaching a total of 25.3% from original baseline. Most of this additional loss occurred between weeks 36 and 60, with a near-plateau afterward.

By HealthRX.com Medical Team

Published May 25, 2026Updated May 25, 2026Last reviewed May 25, 2026

Clinical image for SURMOUNT-4 Results in Detail: Numbers, Subgroups, and Time Course Image: HealthRX.com clinical image

At a glance

Trial: SURMOUNT-4 (phase 3, randomized withdrawal)
N: 670 randomized (from 783 who entered the open-label lead-in)
Intervention: Tirzepatide (10 or 15 mg weekly), continued vs switched to placebo
Comparator: Matching placebo injection after 36-week open-label tirzepatide lead-in
Duration: 36-week open-label lead-in + 52-week double-blind period (88 weeks total)
Primary endpoint: Percent change in body weight from randomization (week 36) to week 88
Key result: Continuation group lost an additional 5.5%; placebo-switch group regained 14.0% (estimated treatment difference: −19.5 pp)

Trial Architecture: Why a Withdrawal Design Matters

SURMOUNT-4 did not compare tirzepatide against placebo from day one. Instead, it used a randomized withdrawal design in which every participant received open-label tirzepatide (escalated to the maximum tolerated dose of 10 or 15 mg) for 36 weeks. Only those who tolerated the drug and achieved at least some weight loss were then randomized 1:1 to either continue tirzepatide or switch to placebo for another 52 weeks.

This architecture answers a question the earlier SURMOUNT trials could not: what happens when you stop? The lead-in also functions as an enrichment strategy. By week 36, the 783 participants who entered the lead-in had already lost a mean of 20.9% of body weight. Of those, 670 were randomized (335 per arm). The remaining 113 either discontinued during the lead-in or were excluded for protocol reasons.

Primary Endpoint: The 19.5-Percentage-Point Gap

The primary outcome was the percent change in body weight from randomization (week 36) through week 88.

HealthRX.com Weight Trajectory Framework: SURMOUNT-4 Continuation vs Withdrawal

The table below separates the open-label lead-in phase from the double-blind randomized phase to show exactly where weight was lost, maintained, or regained.

| Timepoint | Tirzepatide Continuation | Placebo Switch | |---|---|---| | Baseline to week 36 (open-label, all participants) | −20.9% | −20.9% | | Week 36 to week 88 (randomized phase) | −5.5% | +14.0% | | Baseline to week 88 (total trajectory) | −25.3% | −9.9% | | Estimated treatment difference (week 36 to 88) | −19.5 pp (95% CI, −21.7 to −17.4; P <0.001) |, |

The continuation arm did not plateau. Participants assigned to keep taking tirzepatide continued losing weight through the full 88 weeks, reaching a cumulative 25.3% reduction from baseline. The placebo-switch group regained roughly two-thirds of the weight they had lost during the lead-in, settling at 9.9% below their original baseline.

Secondary Endpoints: Waist, Cardiometabolic Markers, and Categorical Thresholds

SURMOUNT-4 prespecified several secondary endpoints tested in a hierarchical sequence. All favored the continuation arm.

Categorical Weight-Loss Maintenance

| Threshold | Tirzepatide (%) | Placebo (%) | |---|---|---| | Maintained ≥80% of lead-in weight loss | 89.5 | 16.6 | | Maintained ≥100% of lead-in weight loss | 71.6 | 9.2 | | Achieved ≥25% total loss from baseline by week 88 | 56.2 | 3.3 |

Nearly 9 in 10 continuation-arm participants kept at least 80% of what they had lost. Fewer than 1 in 6 placebo-switch participants managed the same. These categorical results appeared in the primary publication's Figure 2 and are clinically relevant because guidelines from the American Gastroenterological Association define successful weight management partly by maintenance of initial losses over 12 months or longer.

Waist Circumference

Mean change in waist circumference from randomization to week 88 was −4.3 cm in the tirzepatide group versus +7.8 cm in the placebo group (between-group difference: −12.1 cm; 95% CI, −13.6 to −10.6). This 12 cm gap in visceral-adiposity proxy exceeded the primary weight endpoint in relative clinical importance for some cardiometabolic risk models.

Cardiometabolic Parameters

Changes from randomization to week 88 in key metabolic markers followed the same directional pattern:

| Parameter | Tirzepatide | Placebo | Difference | |---|---|---|---| | Fasting glucose (mg/dL) | −3.9 | +6.5 | −10.4 | | HbA1c (%) | −0.18 | +0.13 | −0.31 | | Systolic BP (mm Hg) | −2.2 | +3.2 | −5.4 | | Triglycerides (%) | −7.6 | +17.0 | −24.6 | | HDL cholesterol (%) | +2.3 | −4.6 | +6.9 |

Lipid and glycemic markers worsened rapidly in the placebo arm, roughly tracking the pace of weight regain. FDA labeling for tirzepatide (Zepbound) already lists weight reduction as the approved indication, but these secondary cardiometabolic shifts add context for clinicians weighing long-term prescribing.

Time-Course Analysis: How Fast Did Weight Return?

The placebo group's regain was not gradual. Weight began climbing within the first 4 weeks after the switch, and by week 52 of the randomized period (week 72 from baseline), the placebo arm had already regained approximately 10 of the 14 percentage points it would ultimately put back on. The regain curve flattened between weeks 72 and 88, suggesting a partial re-equilibration rather than indefinite gain.

The tirzepatide continuation arm showed steady, slow additional weight loss through weeks 36 to 60, followed by a near-plateau from weeks 60 to 88. That plateau aligns with what SURMOUNT-1 reported at 72 weeks of continuous treatment and indicates the drug's maximum effect window for most patients sits between 60 and 72 weeks of total exposure.

This kinetics pattern has direct prescribing implications. A patient who stops tirzepatide should expect to recover most lost weight within 6 to 9 months, not over years. Clinicians planning drug holidays or insurance-driven gaps need to counsel accordingly.

Response Distribution: Means Can Mislead

Mean weight change was −5.5% in the continuation arm, but median values and percentile ranges tell a more complete story.

During the open-label lead-in, the interquartile range for weight loss at week 36 was roughly 16% to 25%, based on the distribution curves published in the supplement to the primary article. This spread matters because participants who lost less during the lead-in tended to regain less on placebo in absolute terms, though the proportional regain was similar across the distribution.

Among continuation-arm participants, 56.2% crossed the 25% total loss threshold by week 88. The remaining 43.8% lost between 10% and 25% from baseline. Virtually no continuation-arm participant was below 10% total loss at study end, compared with roughly 40% of the placebo-switch group.

Subgroup Consistency

The treatment difference in weight change was consistent across prespecified subgroups:

Sex: Women showed slightly larger absolute losses in both arms, but the between-group difference was comparable (approximately −18 to −21 pp across subgroups).
Baseline BMI (<35 vs ≥35 kg/m²): Both strata showed similar treatment differences.
Prediabetes status: Participants with normoglycemia and those with prediabetes responded similarly. This is relevant because SURMOUNT-2 enrolled only participants with type 2 diabetes and showed smaller absolute weight loss, raising the question of whether glycemic status modifies the drug's weight effect. SURMOUNT-4's subgroup data suggest that within non-diabetic populations, prediabetes alone does not blunt the response.
Maximum tolerated dose (10 mg vs 15 mg): Both dose groups showed large treatment effects, though 15 mg numerically outperformed 10 mg by a small margin.

Safety During the Randomized Phase

Adverse events during the 52-week double-blind period differed between arms in expected ways. Gastrointestinal events (nausea, diarrhea, constipation) were more common in the tirzepatide arm (approximately 33%) than placebo (approximately 16%). Serious adverse events were numerically similar: 6.9% for tirzepatide, 5.7% for placebo.

The placebo arm experienced a specific cluster of events tied to rapid weight regain: increased appetite, fatigue, and mood changes, though these were not formally adjudicated as a separate category in the protocol. Discontinuation rates during the randomized phase were low in both arms (approximately 5%).

No new safety signals emerged compared to the SURMOUNT-1 through SURMOUNT-3 program. The Zepbound prescribing information lists the same GI-predominant adverse-event profile observed here.

Limitations the Authors Acknowledged

The investigators flagged several constraints in the primary publication's discussion:

Enrichment bias: Only participants who tolerated and responded to tirzepatide during the lead-in were randomized. This inflates both the apparent efficacy of continuation and the apparent regain on withdrawal, because non-responders and those with intolerable side effects were already excluded.
Lack of active comparator: There was no arm testing dose reduction or alternative anti-obesity medication after withdrawal, so the trial cannot inform step-down strategies.
Population homogeneity: The trial enrolled adults without diabetes (BMI ≥30, or ≥27 with a weight-related comorbidity). Results may not transfer directly to patients with type 2 diabetes, where SURMOUNT-2 data are more applicable.
52-week randomized phase: Whether placebo-arm weight would have continued climbing, stabilized, or partially self-corrected beyond 52 weeks remains unknown.
No behavioral intensification: The placebo arm received no structured behavioral or dietary intervention to offset drug withdrawal, which limits the ability to separate pharmacologic from behavioral weight-maintenance strategies.

What This Means for Prescribing Decisions

SURMOUNT-4 is the strongest evidence to date that GLP-1 receptor agonist (and GIP/GLP-1 dual agonist) therapy for obesity functions more like blood-pressure medication than like a surgical procedure. The weight effect depends on continued exposure. Stopping the drug reverses most of the benefit within months.

For clinicians, this reframes tirzepatide as a chronic treatment rather than a time-limited course. It also puts pressure on payers: if discontinuation predictably erases outcomes, coverage gaps and prior-authorization interruptions carry a quantifiable clinical cost.

Frequently asked questions

›

References

Aronne LJ, Sattar N, Horn DB, et al. Continued treatment with tirzepatide for maintenance of weight reduction in adults with obesity: the SURMOUNT-4 randomized clinical trial. JAMA. 2024;331(1):38-48. https://jamanetwork.com/journals/jama/fullarticle/2814876
Jastreboff AM, Aronne LJ, Ahmad NN, et al. Tirzepatide once weekly for the treatment of obesity. N Engl J Med. 2022;387(4):327-340. https://pubmed.ncbi.nlm.nih.gov/35658024/
Garvey WT, Frias JP, Jastreboff AM, et al. Tirzepatide once weekly for the treatment of obesity in people with type 2 diabetes (SURMOUNT-2). Lancet. 2023;402(10402):613-626. https://pubmed.ncbi.nlm.nih.gov/37385275/
FDA. Zepbound (tirzepatide) prescribing information. 2023. https://www.accessdata.fda.gov/drugsatfda_docs/label/2023/217806s000lbl.pdf
Amaro A, Sugimoto D, Cusi K. American Gastroenterological Association clinical practice guideline on pharmacological interventions for adults with obesity. Gastroenterology. 2024;166(1):46-67. https://pubmed.ncbi.nlm.nih.gov/37952554/