SURMOUNT-2 Results in Detail: Numbers, Subgroups, and Time Course

By
Published Updated Last reviewed
Medical lab testing image for SURMOUNT-2 Results in Detail: Numbers, Subgroups, and Time Course

At a glance

  • Trial: SURMOUNT-2 (NCT04657003)
  • N: 938 randomized (1:1:1)
  • Population: Adults with BMI ≥27 kg/m² and type 2 diabetes
  • Intervention: Tirzepatide 10 mg or 15 mg subcutaneous weekly
  • Comparator: Matched placebo
  • Duration: 72-week treatment period
  • Primary endpoint: Percent change in body weight from baseline at week 72
  • Co-primary endpoint: Achieving ≥5% body weight reduction
  • Key result: 14.7% weight loss (15 mg, treatment-policy) vs 3.2% placebo
  • Source: Garvey et al., Lancet 2023

Why SURMOUNT-2 Matters for Clinicians

Weight loss in patients with type 2 diabetes (T2D) has always been harder to achieve pharmacologically than in people without diabetes. Insulin resistance, beta-cell dysfunction, and concomitant medications (sulfonylureas, insulin) all blunt the magnitude of weight reduction. The original SURMOUNT-1 trial enrolled participants without diabetes and reported up to 22.5% weight loss. The clinical question SURMOUNT-2 answers is direct: how much of that effect survives when the patient also has T2D?

The answer, as published by Garvey and colleagues in The Lancet (2023), is that tirzepatide still delivers weight reductions that were previously considered surgical territory, even in a metabolically resistant population. The numbers deserve a closer look than most trial summaries provide.

Primary Endpoint: Percent Change in Body Weight at Week 72

SURMOUNT-2 reported results using two pre-specified estimands, which is critical for interpreting the data correctly.

HealthRX Dual-Estimand Breakdown

The trial's two estimands answer different clinical questions. Readers comparing SURMOUNT-2 numbers to other obesity trials should confirm which estimand is being cited, because the gap between them can exceed two percentage points.

| Estimand | Tirzepatide 10 mg | Tirzepatide 15 mg | Placebo | Difference vs placebo (15 mg) | |---|---|---|---|---| | Treatment-policy (all randomized, regardless of adherence) | −12.8% | −14.7% | −3.2% | −11.6 pp (95% CI: −13.0 to −10.1) | | Efficacy (on-treatment, no intercurrent events) | −13.4% | −15.7% | −3.3% | −12.4 pp (95% CI: −13.9 to −10.9) |

The treatment-policy estimand includes everyone randomized, carrying forward data even after discontinuation. It reflects what happens in a real-world prescribing scenario where some patients stop the drug. The efficacy estimand restricts analysis to participants who stayed on treatment without intercurrent events (rescue medication, treatment discontinuation). Both were co-primary.

The between-group difference for tirzepatide 15 mg versus placebo was −11.6 percentage points by treatment-policy and −12.4 percentage points by efficacy estimand, with P <0.0001 for both comparisons. For the 10 mg dose, the placebo-subtracted difference was −9.6 pp (treatment-policy) and −10.1 pp (efficacy).

Co-Primary Endpoint: Categorical Weight Loss Thresholds

The proportion of participants reaching clinically meaningful weight loss thresholds tells a story that mean values alone cannot.

| Threshold | Tirzepatide 10 mg | Tirzepatide 15 mg | Placebo | |---|---|---|---| | ≥5% weight loss | 79% | 83% | 32% | | ≥10% weight loss | 57% | 65% | 9% | | ≥15% weight loss | 36% | 51% | 3% | | ≥20% weight loss | 18% | 31% | <1% |

These are efficacy-estimand figures. The 51% rate crossing the 15% threshold at the higher dose is clinically significant because multiple bariatric surgery outcome benchmarks use 15% total body weight loss as a marker of surgical-level efficacy. One in three patients on 15 mg lost at least 20% of their body weight, a result that was essentially absent in the placebo arm.

The treatment-policy percentages are modestly lower (for example, 48% reaching ≥15% on 15 mg), but the clinical message is the same: the majority of adherent patients in a T2D population achieved double-digit percent weight loss.

Secondary Endpoints: Glycemic and Cardiometabolic Parameters

HbA1c

Weight loss was not the only outcome that separated the groups. Tirzepatide is already FDA-approved as Mounjaro for glycemic control in T2D, and the prescribing information reflects HbA1c reductions from SURPASS trials. In SURMOUNT-2, mean baseline HbA1c was approximately 8.0%.

| Parameter | Tirzepatide 10 mg | Tirzepatide 15 mg | Placebo | |---|---|---|---| | HbA1c change (percentage points) | −2.1 | −2.1 | −0.5 | | Participants reaching HbA1c <5.7% | 48% | 55% | 4% | | Participants reaching HbA1c <6.5% | 82% | 87% | 17% |

More than half of participants on tirzepatide 15 mg reached an HbA1c below 5.7%, a level considered normoglycemic and below the diagnostic threshold for prediabetes. This co-benefit distinguishes tirzepatide from pure anti-obesity agents like orlistat or phentermine-topiramate, which have minimal glycemic effects in T2D.

Waist Circumference and Other Measures

Mean reductions in waist circumference were −10.2 cm (10 mg) and −10.8 cm (15 mg) versus −3.4 cm (placebo) by efficacy estimand. Fasting insulin levels decreased substantially in both tirzepatide groups. Systolic blood pressure reductions favored tirzepatide, though the trial was not powered for cardiovascular events. Lipid panels showed improvements in triglycerides and modest favorable shifts in LDL and HDL cholesterol, consistent with the metabolic benefits seen in SURPASS program data.

Time-Course Pattern: When Does Weight Loss Plateau?

The weekly weight-loss trajectory in SURMOUNT-2 followed a characteristic curve. Weight loss was steepest during weeks 0 through 36, corresponding to the dose-escalation phase and early maintenance. The rate of loss began to decelerate after week 36 but did not fully plateau by week 72. The published weight-change-over-time curves suggest the 15 mg group may have still been losing weight at the end of the treatment period, raising the question of whether a longer trial would have shown even greater maximal effect.

This contrasts with semaglutide 2.4 mg in the STEP trials, where weight loss largely plateaued by weeks 60 to 68. The absence of a clear plateau in SURMOUNT-2 at 72 weeks may reflect the dual GIP/GLP-1 mechanism. During dose escalation (weeks 0 to 20), participants titrated from 2.5 mg through 5.0 mg, 7.5 mg, 10.0 mg, and (in the higher-dose arm) to 15.0 mg. Meaningful separation from placebo was visible by week 12 and continued to widen throughout the study.

The placebo group showed modest weight loss of approximately 3% that stabilized by week 24, reflecting background lifestyle counseling provided to all arms.

Response Distribution: Means vs. Medians

Trial publications typically report means, but the distribution of individual responses matters for patient counseling. In SURMOUNT-2, the categorical threshold data allows reconstruction of the approximate response distribution.

For the 15 mg group (efficacy estimand):

  • 17% of participants lost <5% body weight (non-responders or modest responders)
  • 18% lost 5% to <10%
  • 14% lost 10% to <15%
  • 20% lost 15% to <20%
  • 31% lost ≥20%

This distribution is right-skewed: a substantial minority achieved very large weight reductions that pull the mean upward. The median weight loss was likely close to 14% to 15% for the 15 mg group, given that roughly half the participants exceeded 15%. Clinicians should communicate this spread to patients. Not everyone will lose 15%, but more than 80% will lose at least 5%, and nearly a third will exceed 20%.

Subgroup Considerations

SURMOUNT-2 enrolled a population with specific characteristics that affect generalizability.

Baseline BMI: Mean baseline BMI was approximately 36 kg/m², with a range from 27 to well above 40. Higher baseline BMI subgroups tended to lose more absolute weight (kilograms) but similar or slightly lower percentage weight.

Baseline HbA1c: Participants with higher baseline HbA1c (≥8.5%) showed larger absolute HbA1c reductions, as expected, though percentage weight loss was broadly consistent across glycemic strata.

Diabetes duration and background therapy: Participants were on stable oral antidiabetic therapy (typically metformin, with or without SGLT2 inhibitors or sulfonylureas). Insulin users were excluded. This means the results apply to early-to-mid-stage T2D, not to patients on basal-bolus insulin regimens.

Sex: Both sexes were well represented. Pre-specified subgroup analyses showed consistent treatment effects across men and women, though women numerically lost slightly more percentage body weight, a pattern consistent across the obesity pharmacotherapy literature.

Safety and Tolerability Relevant to Efficacy Interpretation

Discontinuation rates affect how we read the efficacy numbers. In SURMOUNT-2, approximately 14% of tirzepatide-treated participants discontinued treatment, compared to 12% on placebo. Gastrointestinal adverse events (nausea, diarrhea, constipation, vomiting) were the most common reasons for discontinuation in the active arms. These rates are comparable to those seen with semaglutide 2.4 mg in the STEP program.

The gap between treatment-policy and efficacy estimand results (roughly 1 to 1.5 percentage points) reflects this discontinuation: it is clinically meaningful but not so large that it undermines the overall treatment effect. In real-world practice, GI side effects are often managed with slower dose escalation, dietary modification, or anti-emetics. The efficacy-estimand results may therefore be achievable in motivated patients who tolerate the titration.

Limitations the Authors Acknowledged

The original publication noted several constraints. The 72-week duration, while longer than many obesity trials, may not capture the full weight-loss trajectory or durability after cessation. Exclusion of insulin-treated patients limits applicability to more advanced T2D. The trial population was predominantly White (approximately 60%) and from North America and Europe, raising questions about generalizability to other populations. Body composition data (fat mass versus lean mass changes) were not a pre-specified endpoint in SURMOUNT-2, though dual-energy X-ray absorptiometry sub-studies from the broader program suggest roughly 75% to 80% of lost mass is fat.

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
  • 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. PubMed
  • Davies M, Færch L, Jeppesen OK, et al. Semaglutide 2.4 mg once a week in adults with overweight or obesity, and type 2 diabetes (STEP 2): a randomised, double-blind, placebo-controlled, phase 3 trial. Lancet. 2021;397(10278):971-984. PubMed
  • Mounjaro (tirzepatide) prescribing information. Eli Lilly and Company. FDA Label
  • Mechanick JI, Apovian C, Brethauer S, et al. Clinical practice guidelines for the perioperative nutrition, metabolic, and nonsurgical support of patients undergoing bariatric procedures. Surg Obes Relat Dis. 2020;16(2):175-247. PubMed