Clinical trials of GLP-1 medications
Plain-English summaries of the most important clinical trials of GLP-1 medications, with their numbers, their limitations and a link to the original paper.
Here I summarize, one by one, the clinical trials that come up most often when people talk about semaglutide, tirzepatide and the drugs coming after them. Each summary explains what was studied, in whom, for how long, what was found (with confidence intervals when the paper gives them) and what its limitations are, including who funded it. I always link the original paper.
Before publishing a summary, I check every figure against the paper's abstract on PubMed and, when needed, against the full text. The trial doses are the ones in the papers; brand names such as Wegovy, Zepbound or Mounjaro are given only to help you place a drug. These summaries are for information: they do not replace the paper or your doctor's advice.
Semaglutide
| Study | Design | Participants | Duration | Main result | Publication |
|---|---|---|---|---|---|
| SUSTAIN-6Semaglutide | Randomized controlled trial | 3,297 | 104 weeks | With subcutaneous semaglutide (0.5 or 1 mg once weekly) over a median of 2.1 years, a major cardiovascular event occurred in 6.6% of patients, compared with 8.9% on placebo (hazard ratio 0.74; 95% CI, 0.58 to 0.95; p < 0.001 for noninferiority). | New England Journal of Medicine, 2016 |
| STEP 1Semaglutide | Randomized controlled trial | 1,961 | 68 weeks | Semaglutide 2.4 mg once weekly, together with a lifestyle intervention, reduced weight by an average of 14.9% at 68 weeks, compared with 2.4% on placebo; the estimated difference was −12.4 percentage points (95% CI, −13.4 to −11.5). | New England Journal of Medicine, 2021 |
| STEP 2Semaglutide | Randomized controlled trial | 1,210 | 68 weeks | In people with type 2 diabetes, semaglutide 2.4 mg once weekly reduced weight by 9.6% at 68 weeks, compared with 3.4% on placebo; the estimated difference was −6.2 percentage points (95% CI, −7.3 to −5.2). | The Lancet, 2021 |
| STEP 3Semaglutide | Randomized controlled trial | 611 | 68 weeks | At 68 weeks, weight changed by −16.0% with semaglutide and −5.7% with placebo; the estimated difference was −10.3 percentage points (95% CI, −12.0 to −8.6). | JAMA, 2021 |
| STEP 4Semaglutide | Randomized controlled trial | 803 | 68 weeks | From week 20 to week 68, weight changed by −7.9% in those who stayed on semaglutide and by +6.9% in those switched to placebo (difference of −14.8 percentage points; 95% CI, −16.0 to −13.5). | JAMA, 2021 |
| STEP 1 (extension)Semaglutide | Secondary analysis of a randomized trial | 327 | 120 weeks | From week 0 to week 68, weight fell by an average of 17.3% with semaglutide and 2.0% with placebo. | Diabetes, Obesity and Metabolism, 2022 |
| STEP 5Semaglutide | Randomized controlled trial | 304 | 104 weeks | At 104 weeks, weight changed by −15.2% with semaglutide 2.4 mg and −2.6% with placebo; the estimated difference was −12.6 percentage points (95% CI, −15.3 to −9.8). | Nature Medicine, 2022 |
| STEP 8Semaglutide and liraglutide | Randomized controlled trial | 338 | 68 weeks | At 68 weeks, weight changed by −15.8% with semaglutide and −6.4% with liraglutide (difference of −9.4 percentage points; 95% CI, −12.0 to −6.8); on placebo it changed by −1.9%. | JAMA, 2022 |
| OASIS 1Semaglutide | Randomized controlled trial | 667 | 68 weeks | At 68 weeks, weight changed by −15.1% with oral semaglutide 50 mg and −2.4% with placebo; the estimated difference was −12.7 percentage points (95% CI, −14.2 to −11.3). | The Lancet, 2023 |
| SELECTSemaglutide | Randomized controlled trial | 17,604 | Event-driven | With a mean follow-up of 39.8 months, a major cardiovascular event occurred in 6.5% of patients on semaglutide 2.4 mg once weekly and in 8.0% on placebo (hazard ratio 0.80; 95% CI, 0.72 to 0.90; p < 0.001). | New England Journal of Medicine, 2023 |
| STEP-HFpEFSemaglutide | Randomized controlled trial | 529 | 52 weeks | At 52 weeks, the KCCQ-CSS score improved by 16.6 points with semaglutide and 8.7 with placebo (difference of 7.8 points; 95% CI, 4.8 to 10.9), and weight changed by −13.3% versus −2.6% (difference of −10.7 percentage points; 95% CI, −11.9 to −9.4). | New England Journal of Medicine, 2023 |
| FLOWSemaglutide | Randomized controlled trial | 3,533 | Event-driven | Over a median follow-up of 3.4 years, the primary outcome occurred in 331 patients on semaglutide and 410 on placebo, a 24% lower risk (hazard ratio 0.76; 95% CI, 0.66 to 0.88). | New England Journal of Medicine, 2024 |
| ESSENCESemaglutide | Randomized controlled trial | 800 | 72 weeks | At 72 weeks, steatohepatitis resolved without worsening of fibrosis in 62.9% on semaglutide and 34.3% on placebo (difference of 28.7 percentage points; 95% CI, 21.1 to 36.2), and fibrosis was reduced without worsening of steatohepatitis in 36.8% versus 22.4% (difference of 14.4 percentage points; 95% CI, 7.5 to 21.3). | New England Journal of Medicine, 2025 |
| OASIS 2Semaglutide | Randomized controlled trial | 201 | 68 weeks | At 68 weeks, weight changed by −14.3% with oral semaglutide 50 mg and −1.3% with placebo; the estimated difference was −13.07 percentage points (95% CI, −15.61 to −10.52). | JAMA Internal Medicine, 2025 |
| OASIS 4Semaglutide | Randomized controlled trial | 307 | 64 weeks | Weight changed by −13.6% with oral semaglutide 25 mg and −2.2% with placebo at 64 weeks; the estimated difference was −11.4 percentage points (95% CI, −13.9 to −9.0). | New England Journal of Medicine, 2025 |
| Semaglutide and alcohol (phase 2)Semaglutide | Randomized controlled trial | 48 | 9 weeks | After 9 weeks, compared with placebo, low-dose semaglutide reduced the grams of alcohol consumed in a laboratory test (β −0.48; 95% CI, −0.85 to −0.11) and peak breath alcohol concentration (β −0.46; 95% CI, −0.87 to −0.06), a medium to large effect according to the authors. | JAMA Psychiatry, 2025 |
| SOULSemaglutide | Randomized controlled trial | 9,650 | Event-driven | Over a mean follow-up of 47.5 months, a major cardiovascular event occurred in 12.0% of participants on oral semaglutide and 13.8% on placebo (hazard ratio 0.86; 95% CI, 0.77 to 0.96; p = 0.006). | New England Journal of Medicine, 2025 |
| STEP UPSemaglutide | Randomized controlled trial | 1,407 | 72 weeks | At 72 weeks, weight changed by −18.7% with semaglutide 7.2 mg, −15.6% with 2.4 mg and −3.9% with placebo; the difference was −14.8 percentage points versus placebo (95% CI, −16.2 to −13.4) and −3.1 points versus 2.4 mg (95% CI, −4.7 to −1.6). | The Lancet Diabetes & Endocrinology, 2025 |
| STEP UP T2DSemaglutide | Randomized controlled trial | 512 | 72 weeks | At 72 weeks, weight had changed by −13.2% with semaglutide 7.2 mg and by −3.9% with placebo (difference of −9.3 percentage points; 95% CI, −11.0 to −7.7). | The Lancet Diabetes & Endocrinology, 2025 |
| evoke and evoke+Semaglutide | Randomized controlled trial | 3,808 | 104 weeks | At 104 weeks, CDR-SB worsened by 2.3 and 2.2 points on semaglutide in evoke and evoke+ and by 2.3 and 2.1 on placebo, with no significant differences. | The Lancet, 2026 |
Tirzepatide
| Study | Design | Participants | Duration | Main result | Publication |
|---|---|---|---|---|---|
| SURPASS-2Tirzepatide and semaglutide | Randomized controlled trial | 1,879 | 40 weeks | HbA1c fell by 2.01, 2.24 and 2.30 percentage points with tirzepatide 5, 10 and 15 mg, compared with 1.86 points with semaglutide 1 mg; the differences were −0.15 (95% CI, −0.28 to −0.03), −0.39 (95% CI, −0.51 to −0.26) and −0.45 points (95% CI, −0.57 to −0.32). | New England Journal of Medicine, 2021 |
| SURPASS-3Tirzepatide | Randomized controlled trial | 1,444 | 52 weeks | At 52 weeks, HbA1c fell by 1.93, 2.20 and 2.37 percentage points with tirzepatide 5, 10 and 15 mg, and by 1.34 points with insulin degludec; the difference versus degludec ranged from −0.59 to −1.04 points by dose, and all doses were superior. | The Lancet, 2021 |
| SURMOUNT-1Tirzepatide | Randomized controlled trial | 2,539 | 72 weeks | At 72 weeks, the mean change in weight was −15.0% with tirzepatide 5 mg (95% CI, −15.9 to −14.2), −19.5% with 10 mg (95% CI, −20.4 to −18.5) and −20.9% with 15 mg (95% CI, −21.8 to −19.9), compared with −3.1% on placebo (95% CI, −4.3 to −1.9). | New England Journal of Medicine, 2022 |
| SURMOUNT-2Tirzepatide | Randomized controlled trial | 938 | 72 weeks | In people with type 2 diabetes and obesity, tirzepatide reduced weight by 12.8% with 10 mg and 14.7% with 15 mg at 72 weeks, compared with 3.2% on placebo. | The Lancet, 2023 |
| SURMOUNT-3Tirzepatide | Randomized controlled trial | 579 | 72 weeks | From randomization to week 72, weight changed by an additional −18.4% with tirzepatide and +2.5% with placebo; the estimated difference was −20.8 percentage points (95% CI, −23.2 to −18.5). | Nature Medicine, 2023 |
| SURMOUNT-4Tirzepatide | Randomized controlled trial | 670 | 88 weeks | From week 36 to week 88, weight changed by −5.5% in those who stayed on tirzepatide and by +14.0% in those switched to placebo (difference of −19.4 percentage points; 95% CI, −21.2 to −17.7). | JAMA, 2024 |
| SURMOUNT-OSATirzepatide | Randomized controlled trial | 469 | 52 weeks | At 52 weeks, the AHI fell by 25.3 events per hour with tirzepatide and 5.3 with placebo in trial 1 (difference of −20.0; 95% CI, −25.8 to −14.2), and by 29.3 versus 5.5 in trial 2 (difference of −23.8; 95% CI, −29.6 to −17.9). | New England Journal of Medicine, 2024 |
| SUMMITTirzepatide | Randomized controlled trial | 731 | 52 weeks | Over a median follow-up of 104 weeks, cardiovascular death or worsening heart failure occurred in 36 patients on tirzepatide (9.9%) and 56 on placebo (15.3%) (hazard ratio 0.62; 95% CI, 0.41 to 0.95). | New England Journal of Medicine, 2025 |
| SURMOUNT-1 (body composition)Tirzepatide | Secondary analysis of a randomized trial | 160 | 72 weeks | At 72 weeks, weight changed by −21.3%, fat mass by −33.9% and lean mass by −10.9% with tirzepatide, compared with −5.3%, −8.2% and −2.6% on placebo (p < 0.001 for all comparisons). | Diabetes, Obesity and Metabolism, 2025 |
| SURMOUNT-5Tirzepatide and semaglutide | Randomized controlled trial | 751 | 72 weeks | At 72 weeks, weight fell by 20.2% with tirzepatide (95% CI, −21.4 to −19.1) and 13.7% with semaglutide (95% CI, −14.9 to −12.6); the estimated difference was −6.5 percentage points (95% CI, −8.1 to −4.9; p < 0.001). | New England Journal of Medicine, 2025 |
| SURPASS-CVOTTirzepatide and dulaglutide | Randomized controlled trial | 13,299 | Event-driven | Over a median follow-up of 4.0 years, a major cardiovascular event occurred in 12.2% of patients on tirzepatide and 13.1% on dulaglutide (hazard ratio 0.92; 95.3% CI, 0.83 to 1.01). | New England Journal of Medicine, 2025 |
| SURMOUNT-4 (post hoc analysis)Tirzepatide | Secondary analysis of a randomized trial | 308 | 88 weeks | After switching from tirzepatide to placebo, the more weight people regained between week 36 and week 88, the more their cardiometabolic parameters worsened: waist circumference increased by 0.8 cm (95% CI, −1.0 to 2.6) in those who regained less than 25% of the weight lost and by 14.7 cm (95% CI, 12.7 to 16.7) in those who regained 75% or more. | JAMA Internal Medicine, 2026 |
| SURMOUNT-MAINTAINTirzepatide | Randomized controlled trial | 378 | 112 weeks | At week 112, weight had changed by −21.9% on the maximum tolerated dose of tirzepatide, −16.6% on 5 mg and −9.9% on placebo; the differences versus placebo were −12.0 percentage points (95% CI, −13.8 to −10.1) and −6.6 points (95% CI, −8.3 to −5.0), both p < 0.0001. | The Lancet, 2026 |
Orforglipron
| Study | Design | Participants | Duration | Main result | Publication |
|---|---|---|---|---|---|
| ACHIEVE-1Orforglipron | Randomized controlled trial | 559 | 40 weeks | HbA1c fell by 1.24, 1.47 and 1.48 percentage points with orforglipron 3, 12 and 36 mg, compared with 0.41 points on placebo; the differences were −0.83 (95% CI, −1.10 to −0.56), −1.06 (95% CI, −1.33 to −0.79) and −1.07 points (95% CI, −1.33 to −0.81). | New England Journal of Medicine, 2025 |
| ATTAIN-1Orforglipron | Randomized controlled trial | 3,127 | 72 weeks | Orforglipron, an oral small-molecule GLP-1 receptor agonist, reduced weight at 72 weeks by 11.2% with 36 mg (95% CI, −12.0 to −10.4), 8.4% with 12 mg (95% CI, −9.1 to −7.7) and 7.5% with 6 mg (95% CI, −8.2 to −6.8), compared with 2.1% on placebo (95% CI, −2.8 to −1.4). | New England Journal of Medicine, 2025 |
| ATTAIN-2Orforglipron | Randomized controlled trial | 1,613 | 72 weeks | At 72 weeks, weight changed by −9.6% with orforglipron 36 mg, −7.0% with 12 mg and −5.1% with 6 mg, compared with −2.5% on placebo; the difference versus placebo was −7.1 percentage points with 36 mg (95% CI, −8.2 to −6.1), −4.5 with 12 mg (95% CI, −5.5 to −3.6) and −2.7 with 6 mg (95% CI, −3.7 to −1.6). | The Lancet, 2025 |
| ACHIEVE-2Orforglipron | Randomized controlled trial | 962 | 40 weeks | At 40 weeks, HbA1c fell by 1.23, 1.50 and 1.56 points with orforglipron 3, 12 and 36 mg, compared with 0.81 on dapagliflozin; the differences were −0.42 (95% CI, −0.62 to −0.23), −0.70 (95% CI, −0.90 to −0.49) and −0.75 points (95% CI, −0.96 to −0.55), and all three doses were noninferior and superior to dapagliflozin. | The Lancet, 2026 |
| ACHIEVE-3Orforglipron and semaglutide | Randomized controlled trial | 1,698 | 52 weeks | HbA1c fell by 1.71 and 1.91 points with orforglipron 12 and 36 mg, and by 1.23 and 1.47 points with oral semaglutide 7 and 14 mg. | The Lancet, 2026 |
| ATTAIN-MAINTAINOrforglipron | Randomized controlled trial | 376 | 52 weeks | On orforglipron, 74.7% of the weight lost on tirzepatide was maintained, versus 49.2% on placebo (difference of 25.5 points; 95% CI, 14.5 to 36.5); after semaglutide, 79.3% versus 37.6% (difference of 41.7 points; 95% CI, 24.4 to 59.0). | Nature Medicine, 2026 |
Liraglutide
| Study | Design | Participants | Duration | Main result | Publication |
|---|---|---|---|---|---|
| SCALE MaintenanceLiraglutide | Randomized controlled trial | 422 | 56 weeks | After losing 6.0% on the diet, weight fell by a further 6.2% with liraglutide and 0.2% with placebo between randomization and week 56 (estimated difference of −6.1 percentage points; 95% CI, −7.5 to −4.6). | International Journal of Obesity, 2013 |
| SCALE Obesity and PrediabetesLiraglutide | Randomized controlled trial | 3,731 | 56 weeks | With liraglutide 3.0 mg daily, plus diet and exercise, participants lost an average of 8.4 kg at 56 weeks, compared with 2.8 kg on placebo; the difference was −5.6 kg (95% CI, −6.0 to −5.1). | New England Journal of Medicine, 2015 |
| S-LiTE (off-treatment follow-up)Liraglutide | Secondary analysis of a randomized trial | 109 | 104 weeks | One year after treatment ended, those who had received exercise plus liraglutide weighed 5.1 kg less (95% CI, −10.0 to −0.2) and had a body fat percentage 2.3 points lower (95% CI, −4.3 to −0.3) than those who had received liraglutide alone. | eClinicalMedicine, 2024 |
CagriSema
| Study | Design | Participants | Duration | Main result | Publication |
|---|---|---|---|---|---|
| REDEFINE 1CagriSema and semaglutide | Randomized controlled trial | 3,417 | 68 weeks | CagriSema (cagrilintide 2.4 mg plus semaglutide 2.4 mg, once weekly) reduced weight by 20.4% at 68 weeks, compared with 3.0% on placebo; the estimated difference was −17.3 percentage points (95% CI, −18.1 to −16.6; p < 0.001). | New England Journal of Medicine, 2025 |
| REDEFINE 2CagriSema | Randomized controlled trial | 1,206 | 68 weeks | In people with type 2 diabetes, CagriSema reduced weight by 13.7% at 68 weeks, compared with 3.4% on placebo; the estimated difference was −10.4 percentage points (95% CI, −11.2 to −9.5; p < 0.001). | New England Journal of Medicine, 2025 |
Retatrutide
| Study | Design | Participants | Duration | Main result | Publication |
|---|---|---|---|---|---|
| Retatrutide phase 2 (obesity)Retatrutide | Randomized controlled trial | 338 | 48 weeks | In this phase 2 trial, retatrutide (a GIP, GLP-1 and glucagon receptor agonist) reduced weight at 48 weeks by 24.2% with 12 mg once weekly, 22.8% with 8 mg, 17.1% with 4 mg and 8.7% with 1 mg, compared with 2.1% on placebo. | New England Journal of Medicine, 2023 |
| TRANSCEND-T2D-1Retatrutide | Randomized controlled trial | 537 | 40 weeks | At 40 weeks, HbA1c fell by 1.69, 1.86 and 1.94 points with retatrutide 4, 9 and 12 mg, compared with 0.81 on placebo; the differences were −0.88 (95% CI, −1.18 to −0.59), −1.04 (95% CI, −1.32 to −0.76) and −1.12 points (95% CI, −1.39 to −0.85). | The Lancet, 2026 |
How to read these summaries
- Placebo
- An injection or pill with no active ingredient. Comparing with placebo separates the effect of the drug from the effect of taking part in the trial.
- Percentage points
- The difference between two percentages. If one group loses 15% of its weight and the other 3%, the difference is 12 percentage points.
- 95% CI
- The 95% confidence interval: the range of values compatible with the data. The narrower it is, the more precise the estimate.
- Secondary analysis
- A study that uses data from a trial that was already published (a subgroup, an extension or a question asked afterward). It adds information, but it carries less weight than the trial's main result.
- Funding
- Almost all of these trials are paid for by the drug's manufacturer. That does not invalidate the results, but I always note it under the limitations.
For the weight-loss figures in context, read how much weight you can lose on GLP-1 medications. The official information on each medication is on our drug pages.