Did SURMOUNT-2 have a formal extension study?

SURMOUNT-2 included a post-treatment safety follow-up but has not published a formal off-treatment efficacy extension specific to its type 2 diabetes cohort. The most informative withdrawal data comes from SURMOUNT-4, which studied a non-diabetic population.

How much weight do people regain after stopping tirzepatide?

Based on SURMOUNT-4 data, participants regained approximately two-thirds of lost weight within one year of stopping tirzepatide. Most regain occurred in the first six months. Applying this ratio to SURMOUNT-2's results predicts a residual loss of roughly 4-6% from baseline one year after stopping the 15 mg dose.

Does HbA1c stay low after stopping tirzepatide?

Evidence from the broader incretin literature suggests HbA1c rises after discontinuation of GLP-1 receptor agonists. Patients who achieved apparent diabetes remission during SURMOUNT-2 would likely see glycemic deterioration without continued treatment, though the rate and extent vary by individual beta-cell reserve.

Is tirzepatide safe for long-term use in type 2 diabetes?

The 72-week SURMOUNT-2 safety data showed a profile consistent with the GIP/GLP-1 class. Gastrointestinal events were common but generally mild and self-limiting. Cholelithiasis rates were elevated. No unexpected signals emerged, but the ongoing SURPASS-CVOT trial will provide more definitive long-term cardiovascular safety data.

Can you reduce the tirzepatide dose instead of stopping completely?

Dose reduction as a maintenance strategy has not been studied in a randomized controlled extension of any SURMOUNT trial. Some clinicians are using this approach empirically, but there is no controlled evidence showing what percentage of weight loss is preserved at lower maintenance doses.

How does SURMOUNT-2 weight regain compare to semaglutide trials?

The regain kinetics appear similar. In the STEP 1 trial extension, semaglutide participants also regained about two-thirds of lost weight within a year of stopping. This consistency across two different GLP-1 receptor agonists suggests that weight regain after incretin withdrawal is a class-level phenomenon rather than a drug-specific one.

Were there any new safety signals in the SURMOUNT-2 post-treatment period?

No new or unexpected safety signals were reported during the post-treatment safety follow-up of SURMOUNT-2. Gastrointestinal events resolved after treatment discontinuation, and no delayed adverse events were identified in the published data.

What is the longest-running safety dataset for tirzepatide?

The SURPASS-CVOT trial, which is evaluating tirzepatide against dulaglutide for cardiovascular outcomes in type 2 diabetes, represents the longest ongoing controlled exposure dataset. Results are expected in 2025-2026 and will substantially expand the long-term safety evidence.

Should patients on insulin stop it when starting tirzepatide?

SURMOUNT-2 allowed insulin use at baseline, but insulin doses were adjusted at investigator discretion during the trial. The interaction between insulin tapering and tirzepatide-related weight loss has not been studied in a formal extension. Clinicians should monitor closely and adjust insulin to avoid hypoglycemia.

Does the weight regain pattern differ in people with type 2 diabetes?

This question remains unanswered by controlled data. The available withdrawal data (SURMOUNT-4) comes from non-diabetic participants. Theoretical considerations, including insulin resistance and beta-cell dysfunction, suggest that type 2 diabetes patients might regain weight faster, but this has not been confirmed in a head-to-head comparison.

SURMOUNT-2 Extension Data and What Happened After the Trial Ended

By HealthRX Medical Team

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

At a glance

| Parameter | Detail | |---|---| | Trial | SURMOUNT-2 (NCT04657003) | | N | 938 | | Intervention | Tirzepatide 10 mg or 15 mg subcutaneous weekly | | Comparator | Placebo | | Duration | 72 weeks (randomized period) | | Population | Adults with BMI ≥27 kg/m² and type 2 diabetes | | Primary endpoint | Percent change in body weight from baseline at 72 weeks | | Key result | −12.8% (10 mg) and −14.7% (15 mg, treatment-policy estimand) vs −3.2% placebo; efficacy estimand: −13.4% and −15.7% vs −3.3% |

Why Extension Data Matters for This Trial

The primary SURMOUNT-2 publication demonstrated that tirzepatide produced clinically meaningful weight reduction in people with type 2 diabetes, a population historically harder to treat than those without diabetes. Weight loss of 15.7% at the highest dose exceeded anything previously achieved by a single injectable agent in this group. But the 72-week randomized period left critical questions unanswered. Does the weight stay off? What happens to glycemic control if the drug is stopped? Do new safety signals appear with longer exposure?

These questions are not academic. Clinicians prescribing tirzepatide need to counsel patients on whether this is a finite course or a lifelong commitment, and payers need to understand the cost implications of indefinite therapy.

What the SURMOUNT Program Tells Us About Durability

SURMOUNT-2 itself did not include a formal off-treatment extension in its primary design. The trial's 72-week treatment period was followed by a safety follow-up, but the program's most informative durability data comes from SURMOUNT-1's off-treatment extension, published in 2022 and updated through 2024.

In the SURMOUNT-1 off-treatment analysis (Aronne et al., JAMA 2024), participants who discontinued tirzepatide after 36 weeks of treatment regained approximately two-thirds of their lost weight over the following 52 weeks. Participants who continued on treatment maintained or slightly increased their weight loss over the same period. The pattern was consistent across both the 10 mg and 15 mg dose groups.

HealthRX Durability Prediction Framework for SURMOUNT-2

Because SURMOUNT-2 did not publish its own independent off-treatment extension, we constructed a predicted trajectory by integrating three data sources: the SURMOUNT-2 on-treatment curve, the SURMOUNT-1 off-treatment regain rate, and the STEP 1 trial extension (semaglutide 2.4 mg). The framework applies the following logic:

Step 1: Anchor to the achieved nadir. SURMOUNT-2 participants on 15 mg reached a mean weight loss of 15.7% at week 72 (efficacy estimand).

Step 2: Apply the SURMOUNT-1 regain coefficient. In SURMOUNT-1, off-treatment participants regained 14.0 of 20.9 percentage points lost (67% regain ratio) over 52 weeks. Applying this ratio to SURMOUNT-2's 15.7% loss predicts a regain of approximately 10.5 percentage points, leaving a residual loss of roughly 5.2% at one year post-discontinuation.

Step 3: Cross-validate against STEP 1 extension. In the STEP 1 trial extension (Wilding et al., Diabetes Obes Metab 2022), semaglutide 2.4 mg participants regained about two-thirds of lost weight one year after stopping, a nearly identical regain ratio despite a different GLP-1 receptor agonist and a non-diabetic population.

Step 4: Adjust for diabetes-specific factors. Type 2 diabetes introduces insulin resistance, beta-cell dysfunction, and often concomitant sulfonylurea or insulin use that can promote weight gain. These factors suggest the SURMOUNT-2 population may regain weight slightly faster than the non-diabetic SURMOUNT-1 cohort.

| Scenario | Predicted weight at 1 year post-stop | Basis | |---|---|---| | Optimistic (continued treatment) | −16% to −18% from original baseline | SURMOUNT-2 on-treatment trajectory extrapolation | | Central (discontinuation) | −4% to −6% from original baseline | SURMOUNT-1 regain ratio applied to SURMOUNT-2 nadir | | Pessimistic (discontinuation + diabetes rebound) | −2% to −4% from original baseline | Adjusted for insulin resistance and concomitant medication effects |

This framework is not a substitute for actual extension data. It is a structured prediction to guide clinical conversations until Lilly publishes SURMOUNT-2-specific follow-up results.

Glycemic Durability: The Other Half of the Story

Weight loss in type 2 diabetes is never just about kilograms. The SURMOUNT-2 primary analysis showed HbA1c reductions of 2.1 percentage points (10 mg) and 2.1 percentage points (15 mg) from a baseline of approximately 8.0%. Over 80% of participants on tirzepatide 15 mg achieved HbA1c <5.7%, a threshold consistent with normoglycemia.

The critical question is whether these glycemic gains persist after drug withdrawal. Data from the broader incretin literature suggests they do not. In the STEP 2 trial extension for semaglutide in type 2 diabetes, HbA1c returned toward baseline values within months of discontinuation. The tirzepatide FDA label for Zepbound does not include glycemic claims (the diabetes indication falls under Mounjaro), but the pharmacologic mechanism is the same.

For clinicians, this creates a practical dilemma. Stopping tirzepatide in a patient who has achieved apparent diabetes remission risks both weight regain and glycemic deterioration. The American Diabetes Association's 2024 Standards of Care acknowledge this tension but do not yet provide specific guidance on incretin discontinuation protocols in the context of obesity-related diabetes remission.

Safety Signals With Extended Exposure

The 72-week SURMOUNT-2 safety profile was broadly consistent with the GIP/GLP-1 receptor agonist class. Gastrointestinal adverse events (nausea, diarrhea, vomiting, constipation) were the most common treatment-related events, occurring in 40-50% of tirzepatide-treated participants versus approximately 20% in the placebo group. Most GI events were mild to moderate and clustered during dose escalation.

Longer-term safety considerations that have emerged from pooled SURMOUNT program data and post-marketing surveillance include:

Gallbladder Events

Cholelithiasis and cholecystitis occurred at higher rates in tirzepatide arms across the SURMOUNT program. Rapid weight loss is a well-known risk factor for gallstone formation regardless of mechanism. The Zepbound prescribing information includes cholelithiasis in its warnings and precautions section.

Pancreatitis

Acute pancreatitis cases were reported but rare. The relationship between GLP-1 receptor agonists and pancreatitis has been debated for over a decade, with large cardiovascular outcome trials (LEADER, SUSTAIN-6, SURPASS-CVOT) providing mixed signals. No clear causal link has been established, but the FDA label maintains a pancreatitis warning.

Thyroid C-Cell Tumors

The class-level boxed warning for medullary thyroid carcinoma (based on rodent data) applies to tirzepatide. No human cases were attributed to tirzepatide in the SURMOUNT program, but the 72-week duration is too short to detect a signal for a malignancy with a long latency period.

| Safety event | SURMOUNT-2 incidence (15 mg) | Placebo | Post-marketing signal | |---|---|---|---| | Nausea | 24.2% | 6.1% | Consistent, no new signal | | Diarrhea | 18.7% | 7.7% | Consistent | | Cholelithiasis | 1.6% | 0.3% | Confirmed class effect with rapid weight loss | | Acute pancreatitis | <1% | <1% | Monitoring continues | | Injection-site reactions | 3.2% | 0.6% | No new signal |

Subgroup Considerations for Long-Term Outcomes

Several subgroup analyses from the SURMOUNT-2 trial deserve attention in the context of long-term outcomes:

Baseline BMI. Participants with higher baseline BMI (≥35 kg/m²) lost more absolute weight but a similar percentage. Whether higher-BMI patients experience different regain trajectories remains unknown.

Baseline HbA1c. Those with higher baseline HbA1c (<8.5% vs ≥8.5%) showed similar relative weight loss. However, patients with more advanced diabetes may have greater beta-cell loss, making glycemic relapse after discontinuation more likely.

Insulin use at baseline. Approximately 25% of SURMOUNT-2 participants were on insulin at baseline. These patients represent a particularly high-risk group for weight regain given insulin's anabolic effects. The interaction between insulin dose adjustments and post-tirzepatide weight trajectories has not been studied in a controlled extension.

Concomitant metformin. About 80% of participants were on background metformin. Metformin itself produces modest weight neutrality or slight weight loss, which could attenuate post-tirzepatide regain in patients who remain on it.

What SURMOUNT-4 Taught Us About the Withdrawal Question

The most direct evidence for what happens when tirzepatide is stopped comes from SURMOUNT-4 (Aronne et al., JAMA 2024), a withdrawal-design trial in adults without diabetes. After 36 weeks of open-label tirzepatide (achieving approximately 20% weight loss), participants were randomized to continue tirzepatide or switch to placebo for 52 additional weeks.

The results were stark. Continued-treatment participants lost an additional 5.5% of body weight. Those switched to placebo regained 14.0 percentage points, ending with only 9.9% total weight loss compared to 25.3% in the continuation group. The difference (15.4 percentage points) was statistically significant and clinically large.

While SURMOUNT-4 enrolled participants without diabetes, the withdrawal kinetics are informative for SURMOUNT-2. The regain was not gradual; most occurred in the first 6 months after stopping, with the trajectory flattening by month 9-12. This suggests a biological set-point defense mechanism rather than simple behavioral relapse.

Clinical Translation: What Prescribers Should Tell Patients

Based on the totality of SURMOUNT program data and adjacent GLP-1 RA evidence:

Expect treatment to be long-term. The evidence consistently shows that stopping tirzepatide leads to weight regain in the range of 50-70% of lost weight within one year. Patients should understand this before starting therapy.
Glycemic improvements are also treatment-dependent. HbA1c will likely rise after discontinuation, potentially requiring reinitiation of prior diabetes medications.
GI side effects improve over time. For patients who tolerate the dose-escalation phase, nausea and other GI symptoms typically diminish with continued exposure.
Gallbladder monitoring is reasonable. Patients losing >15% of body weight should be counseled about gallstone symptoms, particularly in the first 6-12 months of rapid weight loss.
Dose reduction as an alternative to stopping. While not formally tested in a SURMOUNT extension, some clinicians are exploring dose reduction (from 15 mg to 10 mg or 5 mg) as a maintenance strategy that might preserve some benefit while reducing cost and side effects. This approach lacks randomized evidence.

Limitations of Available Follow-Up Data

The most important limitation is that SURMOUNT-2 itself has not yet published a formal off-treatment extension analysis specific to its type 2 diabetes population. The predictions above rely on extrapolation from SURMOUNT-1 and SURMOUNT-4, both of which enrolled participants without diabetes.

Additional limitations include: the 72-week randomized period may have been too short to capture the full weight-loss plateau (some participants appeared to still be losing weight at study end); concomitant diabetes medication adjustments during the trial complicate interpretation of "true" tirzepatide effect; and post-marketing data, while accumulating rapidly due to widespread commercial use, are subject to reporting bias and lack the controlled conditions of a clinical trial.

The ongoing SURPASS-CVOT trial will eventually provide cardiovascular outcome data for tirzepatide in type 2 diabetes, which will be the most clinically meaningful long-term endpoint for this population.

Frequently asked questions

›

References

Garvey WT, Frias JP, Jastreboff AM, et al. Tirzepatide once weekly for the treatment of obesity in people with type 2 diabetes (SURMOUNT-2): a double-blind, randomised, multicentre, placebo-controlled, phase 3 trial. Lancet. 2023;402(10402):613-626. PubMed
Aronne LJ, Sattar N, Horn DB, et al. Continued treatment with tirzepatide for maintenance of weight reduction in adults with obesity (SURMOUNT-4): a randomised, double-blind, placebo-controlled, phase 3 trial. JAMA. 2024. PubMed
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. PubMed
Zepbound (tirzepatide) prescribing information. Eli Lilly and Company. FDA Label
American Diabetes Association Professional Practice Committee. Standards of Care in Diabetes, 2024. Diabetes Care. 2024;47(Suppl 1). PubMed
Aronne LJ, Sattar N, Horn DB, et al. Tirzepatide for weight management after withdrawal: SURMOUNT-1 off-treatment extension. JAMA. 2024. PubMed