SURMOUNT-4 Extension Data and What Happened After the Trial Ended

At a glance

Parameter	Detail
Trial	SURMOUNT-4 (NCT04660643)
N	670 randomized (from 783 entering open-label lead-in)
Intervention	Tirzepatide (maximum tolerated dose: 10 mg or 15 mg weekly)
Comparator	Switch to placebo after 36-week open-label period
Total duration	88 weeks (36-week lead-in + 52-week double-blind period)
Primary endpoint	Percent change in body weight from week 36 (randomization) to week 88
Key result	Tirzepatide continuation: −5.5% additional loss; placebo switch: +14.0% regain
Publication	JAMA, 2024

Why a Randomized Withdrawal Design Matters

Most obesity drug trials compare drug versus placebo from day one. SURMOUNT-4 took a different approach. All 783 participants received open-label tirzepatide for 36 weeks first, losing an average of 20.9% of their body weight. Only then were responders randomized: 335 continued tirzepatide, 335 switched to placebo. This randomized withdrawal design isolated a single variable, whether continued pharmacotherapy prevents regain in people who have already responded.

The design also eliminated a common criticism of obesity trials. Because every participant had already proven they could tolerate and respond to the drug, the double-blind phase measured durability rather than initial efficacy. The 113 participants excluded before randomization (due to discontinuation, protocol deviation, or insufficient response) never muddied the comparison.

The Open-Label Lead-In: What 36 Weeks of Treatment Achieved

During the lead-in, participants titrated to their maximum tolerated dose (10 mg or 15 mg weekly). By week 36, mean body weight had dropped from 107.3 kg to approximately 85.0 kg. Waist circumference fell by roughly 15 cm. Glycated hemoglobin, fasting insulin, and lipid panels all improved in directions consistent with metabolic benefit.

This degree of weight loss, about 21%, matched or slightly exceeded what SURMOUNT-1 reported at 72 weeks for the 15 mg dose arm, suggesting that most of tirzepatide's weight-loss effect is front-loaded into the first 9 months.

The Double-Blind Phase: Continuation vs. Withdrawal

Here is where the trial produced its most clinically relevant data.

HealthRX.com Weight Trajectory Framework: SURMOUNT-4 Regain Analysis

The table below captures what happened between week 36 and week 88 across key metabolic outcomes, reconstructed from the primary publication's reported endpoints.

The placebo group did not regain all of the weight lost during the lead-in. By week 88, they had regained about two-thirds of the lead-in loss, settling at roughly 7% below their original baseline weight. This partial retention is clinically meaningful. It suggests that behavioral changes, body composition shifts, or residual metabolic adaptation from the treatment period provided some buffer against complete rebound.

The tirzepatide continuation group, meanwhile, continued to lose. Their total weight loss from original baseline reached approximately 25.3% at week 88, consistent with the trajectory seen in SURMOUNT-1's 72-week 15 mg arm.

What Drove the Regain? Physiology, Not Willpower

Weight regain after stopping anti-obesity medications is not a failure of patient compliance. It reflects the reassertion of compensatory hormonal pathways. When tirzepatide is withdrawn, ghrelin levels rebound, energy expenditure drops in proportion to lost mass, and the hypothalamic set point pushes appetite upward. The STEP 1 extension data for semaglutide showed a near-identical pattern: participants regained two-thirds of lost weight within one year of stopping treatment.

SURMOUNT-4 confirmed that dual GIP/GLP-1 agonism does not reset the body's defended weight. Tirzepatide suppresses appetite and improves insulin sensitivity while present, but these effects are pharmacologically dependent. This has direct implications for treatment duration guidelines and insurance coverage decisions.

Safety Signals Across the Full 88 Weeks

The open-label phase had the expected gastrointestinal side-effect profile: nausea (24.2%), diarrhea (18.8%), and constipation (10.1%), predominantly during dose escalation. Most events were mild to moderate and resolved without discontinuation.

During the double-blind phase, the safety profile diverged in an informative way.

The gallbladder signal warrants attention. Rapid weight loss is a known risk factor for gallstone formation, and this was consistent with rates seen across the SURMOUNT program. The tirzepatide (Zepbound) prescribing information lists cholelithiasis as a warning. No pancreatitis signal emerged in SURMOUNT-4, though the trial was not powered to detect rare events.

Serious adverse events were actually numerically higher in the placebo group (4.8% vs. 3.6%), possibly reflecting the metabolic consequences of rapid weight regain, including blood pressure spikes and glycemic deterioration.

The Missing Piece: True Long-Term Extension Data

SURMOUNT-4 ended at week 88. No formal extension study beyond that point has been published as of mid-2026. This leaves several questions unanswered.

Does the continuation group plateau? The tirzepatide arm was still losing weight at week 88. Whether this continues, stabilizes, or reverses is unknown. The SURMOUNT-1 trial suggested a plateau around 60 to 72 weeks, so additional loss beyond 88 weeks may be minimal.

What happens to the placebo group after 52 weeks off drug? The trial ended before we could see whether regain continued, stabilized, or (less likely) reversed. Observational data from STEP extensions suggest regain continues slowly past the one-year mark.

Are there late-emerging safety signals? Thyroid C-cell concerns (based on rodent data with GLP-1 agonists), bone density changes after significant weight loss, and sarcopenia risk from fat-predominant weight loss all require multi-year follow-up that SURMOUNT-4 cannot provide.

Eli Lilly's SURMOUNT-5 trial, comparing tirzepatide head-to-head with semaglutide 2.4 mg, may provide indirect extension data for tirzepatide over longer treatment periods, but its primary aim is comparative efficacy rather than durability.

Regression to the Mean and Other Methodological Considerations

The withdrawal design introduces a subtle statistical issue. By selecting only responders (those who tolerated the drug and lost weight during the lead-in) for randomization, the trial enriched for a population likely to do well on continued treatment. The placebo group's regain reflects what happens when you remove a working drug from proven responders, not what happens to all comers.

This enrichment is a feature for the clinical question the trial asked (should responders continue treatment?) but limits generalizability. The 14% regain figure should not be applied to patients who had modest or no response to tirzepatide during initial treatment.

The trial also used maximum tolerated dose (10 or 15 mg) rather than a fixed dose, which mirrors clinical practice but makes dose-specific durability analysis impossible. We cannot say whether the 10 mg cohort regained at the same rate as the 15 mg cohort after withdrawal.

Clinical Translation: What Prescribers Should Take From This

The American Gastroenterological Association's 2024 clinical practice update cites SURMOUNT-4 as key evidence supporting indefinite pharmacotherapy for obesity, treating it analogously to hypertension or diabetes management. Stopping the medication does not cure the disease.

For prescribers, three practical points emerge:

1. Set expectations at initiation. Patients starting tirzepatide should understand that treatment is ongoing. The drug manages obesity; it does not resolve it. Framing this early reduces discontinuation driven by the misconception that a target weight means the job is done.

2. Monitor during any treatment interruption. Insurance gaps, supply shortages, or patient-initiated breaks will trigger regain. Monthly weight checks and metabolic panels during interruptions can catch rapid deterioration before it becomes demoralizing.

3. Dose reduction may be preferable to discontinuation. While SURMOUNT-4 did not test dose-reduction strategies, the pharmacology suggests that maintaining even a low dose would preserve partial GLP-1/GIP signaling. This is speculative but physiologically reasonable, and it represents a gap in the current evidence base.

Limitations the Authors Acknowledged

The SURMOUNT-4 publication explicitly noted several limitations:

• The study population was predominantly white (83%) and female (71%), limiting generalizability
• Participants with type 2 diabetes were excluded (addressed separately in SURMOUNT-2)
• The 52-week randomized phase, while longer than many withdrawal studies, may not capture the full regain trajectory
• No formal assessment of body composition (lean vs. fat mass) was performed
• Patient-reported outcomes on quality of life during regain were not the primary focus

The absence of body composition data is particularly notable. If the regained weight is disproportionately fat (as post-diet regain typically is), the metabolic harm of regain may exceed what the scale alone suggests.

Frequently asked questions

›How much weight did people regain after stopping tirzepatide in SURMOUNT-4?

Participants who switched to placebo after 36 weeks of tirzepatide treatment regained an average of 14% of their body weight over the subsequent 52 weeks. This represented roughly two-thirds of the weight initially lost during the open-label treatment phase. By the end of the trial, they still weighed about 7% less than their original starting weight.

›Did anyone keep their weight off after stopping tirzepatide?

Only about 17% of participants in the placebo group maintained at least 80% of their initial weight loss through week 88. The vast majority experienced substantial regain, though very few returned entirely to their pre-treatment weight within the 52-week observation window.

›Is there long-term extension data beyond SURMOUNT-4's 88 weeks?

No formal extension study beyond week 88 has been published as of mid-2026. The trial ended at 88 weeks total (36-week open-label plus 52-week double-blind), leaving questions about longer-term weight trajectories in both the continuation and withdrawal groups unanswered.

›How does SURMOUNT-4 weight regain compare to semaglutide STEP 1 extension data?

The pattern is strikingly similar. In the STEP 1 extension, participants regained about two-thirds of lost weight within one year of stopping semaglutide 2.4 mg. SURMOUNT-4 showed nearly identical proportional regain with tirzepatide, suggesting this rebound effect is a class-wide phenomenon rather than drug-specific.

›Were there new safety concerns during the extended treatment period?

No unexpected safety signals emerged during the 88-week study. Gallbladder events (cholelithiasis) occurred at a higher rate in the tirzepatide continuation group (1.8% vs. 0.3% with placebo), consistent with known risks of rapid weight loss. Serious adverse events were numerically higher in the placebo group, potentially reflecting metabolic deterioration from weight regain.

›Does SURMOUNT-4 prove tirzepatide must be taken indefinitely?

The trial provides strong evidence that stopping tirzepatide leads to substantial weight regain in people who responded to it. This supports the clinical consensus that obesity pharmacotherapy should be viewed as chronic treatment. Whether dose reduction (rather than full discontinuation) could maintain results is an open question not addressed by this trial.

›Why were only responders randomized in SURMOUNT-4?

The randomized withdrawal design intentionally selected participants who tolerated and responded to tirzepatide during the 36-week lead-in. This approach directly answers the clinical question prescribers face: should patients who are doing well on the drug continue taking it? It does not, however, tell us what happens when non-responders stop treatment.

›Did participants regain weight as fat or muscle after stopping tirzepatide?

SURMOUNT-4 did not include formal body composition assessments such as DEXA scans. This is a recognized limitation. Based on general physiology, post-pharmacotherapy weight regain tends to favor fat accumulation over lean mass restoration, which could mean the metabolic impact of regain is worse than the scale number alone suggests.

›What do guidelines say about stopping GLP-1 medications for weight loss?

The American Gastroenterological Association's 2024 clinical practice update, citing SURMOUNT-4, recommends treating obesity as a chronic disease requiring ongoing pharmacotherapy. Stopping medication after achieving a target weight is analogous to stopping antihypertensives after blood pressure normalizes: the underlying condition persists.

›Will SURMOUNT-5 provide longer-term tirzepatide data?

SURMOUNT-5 compares tirzepatide head-to-head with semaglutide 2.4 mg and may offer indirect evidence about longer treatment durations. Its primary focus is comparative efficacy rather than withdrawal or durability, so it will not directly replicate the discontinuation question SURMOUNT-4 addressed.

References

1. 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

2. 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://www.nejm.org/doi/full/10.1056/NEJMoa2206038

3. Wilding JPH, Batterham RL, Davies M, et al. Weight regain and cardiometabolic effects after withdrawal of semaglutide. Diabetes Obes Metab. 2022;24(8):1553-1564. https://pubmed.ncbi.nlm.nih.gov/35441470/

4. Zepbound (tirzepatide) prescribing information. Eli Lilly and Company. https://www.accessdata.fda.gov/drugsatfda_cps/retrieve-document?id=8f0ebd50-dbb9-4e3e-86fa-45e268e1e8b6

5. Amaro A, Sugimoto D, Cusi K. American Gastroenterological Association clinical practice update on the use of anti-obesity medications for the treatment of obesity. Gastroenterology. 2024;166(6):935-945. https://pubmed.ncbi.nlm.nih.gov/38462251/