Semaglutide vs tirzepatide vs retatrutide: reading the trials
A popular chart lines up semaglutide, tirzepatide and retatrutide as a staircase of receptors and results. The underlying trials were never run against each other, and only one pair has been compared directly.
One receptor, two receptors, three. Semaglutide acts on GLP-1; tirzepatide on GIP and GLP-1; retatrutide on GIP, GLP-1 and glucagon. A chart pairing each compound with a body-weight figure has circulated widely, and it reads like a staircase. It is built from three trials that never tested these compounds against one another, so it is worth looking at what each trial actually measured.
What each trial measured
Each compound's figure comes from its own Phase 3 program: STEP for semaglutide, SURMOUNT for tirzepatide and TRIUMPH for retatrutide.
- Semaglutide (GLP-1 receptor): STEP 1, published in NEJM in 2021, enrolled 1,961 adults for 68 weeks. Mean body weight change was −14.9% versus −2.4% with placebo.
- Tirzepatide (GIP and GLP-1 receptors): SURMOUNT-1, published in NEJM in 2022, enrolled 2,539 adults for 72 weeks. The top arm showed −20.9% versus −3.1% with placebo.
- Retatrutide (GIP, GLP-1 and glucagon receptors): TRIUMPH-1, a Phase 3 topline release from May 2026, enrolled 2,339 adults for 80 weeks. The top arm showed −28.3%.
Why the chart is a collage, not a comparison
The three trials differ in length, participants and year. Lining up their results shows three separate measurements side by side, not a controlled comparison.
The numbers are also not calculated the same way. The 28.3% figure estimates the outcome if participants remained on the assigned compound throughout. The 14.9% and 20.9% figures count everyone, including participants who stopped. Calculated the adherence-based way, the SURMOUNT-1 figure for tirzepatide is 22.5%. A reader comparing 20.9% with 28.3% is therefore comparing two different estimates.
The one head-to-head pair
Only one pair has been tested directly. SURMOUNT-5, published in NEJM in 2025, ran tirzepatide against semaglutide in the same trial and reported −20.2% versus −13.7% at 72 weeks. Retatrutide has no completed head-to-head against either compound. TRIUMPH-5, which compares it with tirzepatide, is still running (ClinicalTrials.gov NCT06662383).
The third receptor and the limits of the evidence
Glucagon receptor activation is the least understood of the three mechanisms. It raises energy expenditure in animal models, but whether that explains the larger figure in human trials is a hypothesis, not a finding.
The larger number also came with more discontinuations. In TRIUMPH-1, discontinuation due to adverse events rose across the three retatrutide arms, at 4.1%, 6.9% and 11.3%, compared with 4.9% on placebo.
Regulatory and publication status differs as well. Semaglutide and tirzepatide are approved medicines. Retatrutide is investigational: its Phase 3 results so far are company topline data from Eli Lilly, not peer-reviewed papers, and Lilly has said it plans to file with the FDA in Q1 2027.
The staircase may hold up. TRIUMPH-5 is the trial designed to test it.
Sources: Wilding et al., NEJM 2021 (STEP 1); Jastreboff et al., NEJM 2022 (SURMOUNT-1); Aronne et al., NEJM 2025 (SURMOUNT-5); Eli Lilly topline release, May 21, 2026 (TRIUMPH-1); ClinicalTrials.gov NCT06662383 (TRIUMPH-5).