Tirzepatide vs semaglutide: indirect data

A 2026 indirect treatment comparison found that tirzepatide 10 mg and 15 mg were associated with 5.10 kg and 6.50 kg more weight reduction than semaglutide 2.4 mg. The analysis compared results from SURMOUNT-1 and STEP 1 through their shared placebo comparator rather than assigning people to tirzepatide or semaglutide in a new trial^[1]. That distinction matters. The estimates support a difference at the two higher tirzepatide doses, but they do not carry the same protection against cross-trial differences as a direct randomized comparison.

What the analysis compared

The authors linked two placebo-controlled obesity trials. SURMOUNT-1 studied once-weekly tirzepatide 5 mg, 10 mg and 15 mg for 72 weeks. STEP 1 studied once-weekly semaglutide 2.4 mg for 68 weeks. Both enrolled adults without type 2 diabetes who had obesity, or overweight plus at least one weight-related complication. Treatment in both trials accompanied a reduced-calorie diet and increased physical activity^[1].

The main method was a Bucher indirect comparison. In plain language, the authors first measured how each active treatment performed against placebo in its own trial, then compared those active-versus-placebo differences. They also ran matching-adjusted indirect comparisons using participant-level SURMOUNT-1 data to make the tirzepatide population resemble the published STEP 1 population on sex, and then on sex plus Hispanic ethnicity^[1]. The adjusted estimates generally pointed in the same direction as the unadjusted estimates.

Baseline age, weight and body mass index were similar across the trials, but the populations were not identical. Women made up 67.5% of SURMOUNT-1 and 74.1% of STEP 1. Hispanic participants made up 47.8% and 12.0%, respectively. Placebo groups also differed in gastrointestinal events, nausea and all-cause discontinuation, an imbalance the authors said could bias the safety comparison in semaglutide's favour^[1]. Statistical adjustment can address measured differences. It cannot guarantee that every relevant difference was measured.

The weight and waist results

For the efficacy estimand, which describes outcomes if participants adhered to treatment, tirzepatide 10 mg was associated with 5.10 kg more weight reduction than semaglutide 2.4 mg (95% CI 3.57 to 6.63 kg). The 15 mg estimate was 6.50 kg (95% CI 4.97 to 8.03 kg). The 5 mg estimate was 0.90 kg in the other direction, with a confidence interval from 0.63 kg in favour of tirzepatide to 2.43 kg in favour of semaglutide. That 5 mg comparison was not statistically significant^[1].

Waist circumference followed a similar dose pattern. The estimated advantage over semaglutide was 5.25 cm for tirzepatide 10 mg and 5.75 cm for 15 mg. Tirzepatide 5 mg did not show a significant difference. The odds of reaching at least 10%, 15% or 20% weight reduction were also higher for the two larger tirzepatide doses in the unadjusted analysis^[1]. These are comparisons of trial-level treatment effects, not a forecast of how many kilograms one person will lose.

Body composition adds a narrower signal

The paper also compared dual-energy X-ray absorptiometry subgroups. For tirzepatide 15 mg, the estimated difference was 3.50 percentage points less body fat as a share of total mass and 3.40 percentage points more lean mass as a share of total mass than with semaglutide. Absolute lean mass still fell in every treatment group. The percentage result reflects the larger reduction in fat mass rather than evidence that tirzepatide built muscle^[1].

This part of the analysis is less secure than the overall weight comparison. The body-composition subgroups included 255 SURMOUNT-1 participants and 140 STEP 1 participants. Matching adjustment was not considered feasible because those samples were small and the STEP 1 subgroup included very few Hispanic participants. The authors called for future head-to-head validation^[1].

Safety looked similar, with important uncertainty

The analysis did not find statistically significant differences between the medicines in gastrointestinal adverse events, nausea or discontinuation due to adverse events. Tirzepatide 10 mg and 15 mg showed numerical trends toward higher odds for those outcomes in the unadjusted comparison, while 5 mg showed numerical trends toward lower odds. Matching-adjusted results varied, but none established a clear safety difference^[1]. Similar here means that this analysis did not detect a difference. It does not mean either medicine was free of adverse effects.

Safety is especially difficult to compare indirectly because STEP 1 and SURMOUNT-1 did not produce identical placebo event rates. Severe gastrointestinal adverse events could not be compared because STEP 1 did not report the required data. Readers considering either medicine should use the approved product information and discuss individual risks with a prescribing clinician. Our tirzepatide safety summary and semaglutide safety summary provide broader context.

What this adds beside SURMOUNT-5

A direct head-to-head trial already exists. SURMOUNT-5 compared participants assigned to the maximum tolerated dose of tirzepatide, either 10 mg or 15 mg, with the maximum tolerated dose of semaglutide, either 1.7 mg or 2.4 mg. This new paper instead estimates separate effects for each licensed tirzepatide dose, including 5 mg, and examines body composition. Its 10 mg and 15 mg weight estimates broadly aligned with the direct trial^[1]. It should therefore be read as a dose-specific extension, not a replacement for the randomized head-to-head evidence.

What we don't yet know

The evidence base was limited to two trials and to outcomes both reported. Their primary endpoints occurred four weeks apart. The analysis used one timepoint from each trial, so it could not test whether relative effects changed during treatment. It also could not compare progression to type 2 diabetes or severe gastrointestinal events. Even after adjustment, unmeasured differences in care, geography or participant characteristics may remain^[1].

The funding context belongs in the interpretation. Eli Lilly sponsored the analysis and provided open-access funding. Six authors were Lilly employees and shareholders. Two authors worked for Costello Medical, which Lilly paid for analytical services, and the acknowledgements credit Costello Medical staff with writing and statistical support^[1]. Those disclosures do not invalidate the work, but they make independent replication and the direct randomized evidence particularly valuable.

The useful conclusion is limited: this indirect analysis associated tirzepatide 10 mg and 15 mg with greater weight reduction than semaglutide 2.4 mg in adults without type 2 diabetes, while 5 mg did not show a significant difference. It did not test a switching strategy, choose a medicine for an individual or establish equivalent safety. Those decisions require a clinician who can weigh approved indications, contraindications and personal circumstances.

Medical disclaimer: This article is for educational and journalistic purposes only and does not constitute medical advice. Peptides discussed may be classified as prescription medicines or research chemicals depending on your jurisdiction. Always consult a qualified healthcare professional before using any peptide product. PeptideMethods.com does not sell, distribute, or facilitate the sale of any peptide product.