STEP-1 vs SURMOUNT-1: Semaglutide and Tirzepatide for Weight Loss, A Cross-Trial Synthesis

By
Published Updated Last reviewed
GLP-1 medication and metabolic health image for STEP-1 vs SURMOUNT-1: Semaglutide and Tirzepatide for Weight Loss, A Cross-Trial Synthesis

At a Glance

At a glance

| Feature | STEP-1 (Wilding 2021) | SURMOUNT-1 (Jastreboff 2022) | STEP-8 (Rubino 2022) | |---|---|---|---| | Drug / dose | Semaglutide 2.4 mg SC weekly | Tirzepatide 5, 10 to 15 mg SC weekly | Semaglutide 2.4 mg vs liraglutide 3.0 mg SC | | N (randomized) | 1,961 | 2,539 | 338 | | Randomization ratio | 2:1 (drug:placebo) | 1:1:1:1 (three doses + placebo) | 2:1 (sema:lira) | | Run-in design | No formal run-in | No formal run-in | No formal run-in | | Comparator | Placebo | Placebo | Active (liraglutide 3.0 mg) | | Follow-up / primary timepoint | 68 weeks | 72 weeks | 68 weeks | | Primary endpoint | % change in body weight + proportion ≥5% loss | % change in body weight + proportion ≥5% loss | % change in body weight + proportion ≥5% loss | | Primary result (mean % weight change) | −14.9% (sema) vs −2.4% (placebo) | −15.0% / −19.5% / −20.9% (5/10/15 mg) vs −3.1% (placebo) | −15.8% (sema) vs −6.4% (lira) | | Lifestyle intervention | Reduced-calorie diet + exercise counseling | Reduced-calorie diet + exercise counseling | Reduced-calorie diet + exercise counseling | | Dropout (any cause) | ~14% (sema arm) | ~16% (tirzepatide arms pooled) | ~13% (sema arm) | | Key GI adverse events | Nausea 44%, vomiting 24%, diarrhea 30% | Nausea 31 to 43%, vomiting 16 to 25%, diarrhea 19 to 30% | Nausea 44% (sema) vs 40% (lira) | | Discontinuation due to AEs | ~7% | ~4 to 6% (dose-dependent) | ~8% (sema) vs ~13% (lira) |

Sources: STEP-1 (Wilding et al., NEJM 2021), SURMOUNT-1 (Jastreboff et al., NEJM 2022), STEP-8 (Rubino et al., JAMA 2022).

Population Differences

The two flagship trials enrolled people with obesity or overweight-plus-comorbidity, but the populations were not identical, and those differences carry real implications.

Baseline body weight. Participants in SURMOUNT-1 had a mean baseline weight of approximately 231 lbs (104.8 kg), compared with roughly 232 lbs (105.4 kg) in STEP-1. On the surface this looks matched. But the variance around that mean was somewhat wider in SURMOUNT-1, meaning more participants at the extreme ends of the weight spectrum were enrolled. Absolute kilogram losses appear larger in SURMOUNT-1 partly because a higher share of participants had more weight to lose.

Diabetes exclusion. Both trials excluded participants with type 2 diabetes. This is important context: the weight-loss effect of GLP-1 receptor agonists is attenuated in people with diabetes, as shown in the STEP-2 trial of semaglutide in T2D (mean loss ~9.6%). SURMOUNT-2, the tirzepatide trial in T2D, showed mean losses of roughly 13 to 15%. Neither STEP-1 nor SURMOUNT-1 results generalize directly to that population.

Comorbidity thresholds. STEP-1 required a BMI ≥30, or ≥27 with at least one weight-related comorbidity (hypertension, dyslipidemia, obstructive sleep apnea, or cardiovascular disease). SURMOUNT-1 used essentially the same criteria. Prediabetes was permitted in both; in SURMOUNT-1, approximately 40% of participants had prediabetes at baseline, a detail that matters when interpreting glycemic secondary endpoints.

Geographic and racial distribution. SURMOUNT-1 enrolled participants across 19 countries; STEP-1 spanned 16. Both were majority-White cohorts. Subgroup analyses by race and ethnicity in both trials are underpowered for firm conclusions, which limits generalizability to populations with higher rates of visceral adiposity at lower BMI thresholds, such as East Asian individuals. The WHO expert consultation on BMI cutoffs has long flagged this gap.

Methodology Differences

Understanding what these numbers mean requires understanding how each trial was built.

Estimand framework. STEP-1 used a treatment-policy estimand as its primary analysis. This means weight at the final visit was used regardless of whether participants had stopped the drug. A separate "efficacy estimand" (on-treatment only, with multiple imputation) showed larger effects for semaglutide but was secondary. SURMOUNT-1 similarly used a treatment-policy estimand as primary, with an efficacy estimand pre-specified as a key secondary analysis. Under the efficacy estimand, tirzepatide 15 mg produced a mean weight loss of approximately 22.4%. This alignment in estimand choice is one of the few genuine methodological points of comparability between the two trials.

Blinding. Both trials used double-dummy, double-blind designs with matching placebo injections. This is a meaningful similarity. STEP-8, by contrast, was an active-comparator trial comparing semaglutide with liraglutide, which required separate injection schedules but maintained blinding through identical-looking devices and volumes.

Primary endpoint definition. Both STEP-1 and SURMOUNT-1 used a co-primary endpoint: mean percent change in body weight AND proportion of participants achieving ≥5% weight loss. This dual structure was required by FDA guidance on obesity drug development. The similarity makes the threshold-response comparisons (≥5%, ≥10%, etc.) more directly informative than the continuous mean alone.

Lifestyle background. All three trials provided a standardized lifestyle intervention including reduced-calorie diet counseling and a physical activity recommendation of roughly 150 minutes per week. The interventions were described comparably in methods sections but were not independently audited for fidelity, which is a shared limitation. Real-world adherence to lifestyle components likely varied across sites and countries.

Dose-escalation schedules. Semaglutide in STEP-1 followed a fixed 16-week escalation to the 2.4 mg maintenance dose. Tirzepatide in SURMOUNT-1 escalated over 20 weeks to 5 mg, 10 mg, or 15 mg. The longer escalation for higher tirzepatide doses means participants spent fewer weeks at full therapeutic exposure within the same overall trial duration, a point that may slightly compress the observed weight loss in SURMOUNT-1.

Results, Matched

Mean weight loss at primary endpoint

| Outcome | Semaglutide 2.4 mg (STEP-1) | Tirzepatide 5 mg (SURMOUNT-1) | Tirzepatide 10 mg (SURMOUNT-1) | Tirzepatide 15 mg (SURMOUNT-1) | |---|---|---|---|---| | Mean % weight change (treatment-policy) | −14.9% | −15.0% | −19.5% | −20.9% | | Mean % weight change (efficacy estimand) | ~−17.4% | ~−17.8% | ~−21.1% | ~−22.4% | | Placebo-subtracted difference | −12.4 pp | ~−11.9 pp | ~−16.4 pp | ~−17.8 pp |

At the lowest tirzepatide dose (5 mg), results are nearly identical to semaglutide 2.4 mg. The gap widens substantially at 10 mg and 15 mg. Both agents far exceed the FDA's minimum efficacy threshold of ≥5% placebo-subtracted weight loss.

Proportion reaching weight-loss thresholds

| Threshold | Semaglutide 2.4 mg (STEP-1) | Tirzepatide 5 mg | Tirzepatide 10 mg | Tirzepatide 15 mg | |---|---|---|---|---| | ≥5% loss | 86% | 85% | 89% | 91% | | ≥10% loss | 69% | 69% | 79% | 83% | | ≥15% loss | 50% | 51% | 67% | 72% | | ≥20% loss | 32% | 30% | 49% | 57% |

At the ≥20% threshold, tirzepatide 15 mg was approximately 25 percentage points ahead of semaglutide. That is a clinically meaningful difference. One in two participants on tirzepatide 15 mg lost at least one-fifth of their body weight. The number needed to treat calculations for that threshold differ substantially between the two agents. Jastreboff et al. pre-specified all four thresholds; Wilding et al. reported them as secondary or exploratory outcomes.

Gastrointestinal adverse events

GI events were the most common adverse events in both trials, and rates were broadly comparable at equivalent points in the dose-escalation schedule. Nausea was slightly lower in SURMOUNT-1 tirzepatide arms than in the STEP-1 semaglutide arm (31 to 43% vs 44%), though cross-trial rate comparisons are sensitive to differences in how symptoms were elicited and coded. Discontinuation due to adverse events was modestly lower with tirzepatide (4 to 6%) than semaglutide in STEP-1 (~7%) or liraglutide in STEP-8 (~13%).

STEP-8 is particularly informative here. Rubino et al. showed semaglutide 2.4 mg was dramatically better-tolerated than liraglutide 3.0 mg, with a discontinuation-due-to-AE rate of ~8% vs ~13%, and produced roughly 2.5 times the weight loss. This positions semaglutide clearly ahead of earlier-generation GLP-1 agents, and frames the comparisons with tirzepatide in appropriate historical context.

Cardiometabolic secondary endpoints

Both trials reported improvements in blood pressure, waist circumference, fasting glucose, and lipid panels. In SURMOUNT-1, systolic blood pressure fell by 6.2 to 8.3 mmHg (dose-dependent) vs 3.5 mmHg in the placebo group. In STEP-1, systolic blood pressure fell by approximately 6.2 mmHg. Improvements in HbA1c were numerically larger in SURMOUNT-1, which may partly reflect the higher baseline prediabetes prevalence. Neither trial was powered for cardiovascular events. The SELECT trial of semaglutide, published in NEJM in 2023, subsequently demonstrated a 20% reduction in major adverse cardiovascular events in adults with established CVD and obesity but without diabetes. An equivalent cardiovascular outcomes trial for tirzepatide (SURPASS-CVOT) is ongoing.

Quality-of-life measures

STEP-1 used the Impact of Weight on Quality of Life-Lite Clinical Trials (IWQOL-Lite-CT) and the SF-36, reporting meaningful improvements in physical functioning scores with semaglutide vs placebo. SURMOUNT-1 also used the IWQOL-Lite-CT, reporting comparable or slightly larger improvements, consistent with the greater absolute weight loss. Because different instruments were used in some subscales and the trials enrolled slightly different comorbidity burdens, QoL comparisons carry meaningful uncertainty.

What the Trials Together Do and Do Not Establish

What they establish:

Both drugs produce weight loss well beyond what lifestyle intervention alone achieves. The evidence is high-quality, from large, well-blinded, multi-center RCTs with pre-specified co-primary endpoints meeting FDA standards. Tirzepatide at 10 mg and 15 mg achieves weight loss in the 20% range that previously required bariatric surgery for most patients. Semaglutide 2.4 mg produces ~15% loss. For many patients, either outcome represents a clinically significant change in metabolic risk.

Tolerability for both drugs is acceptable. GI events are common but rarely cause permanent discontinuation. Slower dose escalation reduces early GI burden for both agents.

What they do not establish:

The trials do not establish which drug is superior in a given patient because there has been no randomized head-to-head comparison of semaglutide 2.4 mg vs tirzepatide at any dose for weight loss as a primary endpoint. Cross-trial comparisons of continuous outcomes, even from near-identical designs, are confounded by site effects, regional dietary backgrounds, different randomization ratios, and subtle differences in lifestyle intervention delivery. The apparent dose-response advantage of tirzepatide 10 mg and 15 mg over semaglutide 2.4 mg is consistent across many subgroups, but it remains an indirect comparison.

The trials do not establish long-term durability beyond 68 to 72 weeks. Extension and withdrawal studies (including STEP-4 for semaglutide) suggest rapid weight regain upon discontinuation, which reframes both agents as chronic treatments rather than courses.

Cardiovascular outcome data exist for semaglutide (SELECT trial) but not yet for tirzepatide. This is a real clinical asymmetry. Clinicians managing patients with established CVD have a stronger evidence base for semaglutide at this time.

Outstanding Questions for the Next Trial

  1. Direct head-to-head at equivalent doses. A four-arm trial (semaglutide 2.4 mg, tirzepatide 5 mg, 10 mg, 15 mg, and placebo) with a cardiovascular event endpoint would resolve the indirect-comparison problem and establish the benefit hierarchy on hard outcomes simultaneously.

  2. Responder prediction. Neither trial identified reliable baseline predictors of who achieves ≥20% loss vs <5% loss. Pharmacogenomic work on GIP receptor variants and GLP-1 receptor expression may identify subpopulations where one mechanism predominates. The Obesity Society's 2023 Clinical Practice Statement flagged this as a priority research gap.

  3. Long-term safety signal for non-GI events. Both drugs carry FDA labeling noting potential thyroid C-cell tumor risk based on rodent data, though human evidence does not currently support a causal link. Registries with >5-year follow-up are needed.

  4. Subpopulations excluded from both trials. People with type 2 diabetes, BMI <27, or prior bariatric surgery were not studied in STEP-1 or SURMOUNT-1. Each of these groups has high clinical relevance.

  5. Oral formulations. Oral semaglutide for obesity is under evaluation. If tirzepatide moves to oral formulation, the relative tolerability and dose-response profiles may shift in ways the current injectable trials cannot predict.

  6. Combination and sequencing strategies. If a patient loses 12% on semaglutide and plateaus, does switching to or adding tirzepatide confer additional benefit? No RCT has addressed this.

Frequently asked questions

›
›
›
›

References

  1. Wilding JPH, et al. Once-Weekly Semaglutide in Adults with Overweight or Obesity (STEP-1). N Engl J Med. 2021;384:989-1002. https://pubmed.ncbi.nlm.nih.gov/33567185/

  2. Jastreboff AM, et al. Tirzepatide Once Weekly for the Treatment of Obesity (SURMOUNT-1). N Engl J Med. 2022;387:205-216. https://pubmed.ncbi.nlm.nih.gov/35658024/

  3. Rubino DM, et al. Effect of Weekly Subcutaneous Semaglutide vs Daily Liraglutide on Body Weight in Adults With Overweight or Obesity Without Diabetes (STEP-8). JAMA. 2022;327:138-150. https://pubmed.ncbi.nlm.nih.gov/35089258/

  4. Davies M, et al. Semaglutide 2.4 mg Once a Week in Adults with Overweight or Obesity, and Type 2 Diabetes (STEP-2). Lancet. 2021;397:971-984. https://pubmed.ncbi.nlm.nih.gov/33545554/

  5. Rubino D, et al. Effect of Continued Weekly Subcutaneous Semaglutide vs Placebo on Weight Loss Maintenance in Adults With Overweight or Obesity (STEP-4). JAMA. 2021;325:1414-1425. https://pubmed.ncbi.nlm.nih.gov/33940583/

  6. Lincoff AM, et al. Semaglutide and Cardiovascular Outcomes in Obesity without Diabetes (SELECT). N Engl J Med. 2023;389:2221-2232. https://pubmed.ncbi.nlm.nih.gov/37952131/

  7. Obesity Medicine Association / The Obesity Society Clinical Practice Statement 2023. https://pubmed.ncbi.nlm.nih.gov/37899513/

  8. FDA Guidance for Industry: Developing Products for Weight Management. U.S. Food and Drug Administration. https://www.fda.gov/media/71252/download

  9. WHO Expert Consultation. Appropriate body-mass index for Asian populations and its implications for policy and intervention strategies. Lancet. 2004;363:157-163. https://www.who.int/publications/i/item/WHO-TRS-916