Direct comparison
In SURMOUNT-5, 751 adults without diabetes were randomly assigned to tirzepatide or semaglutide, each at the highest tolerated dose (tirzepatide 10 or 15 mg, semaglutide 1.7 or 2.4 mg). Both were injected under the skin once a week, and there was counselling on diet and physical activity. After 72 weeks, participants had lost a mean of 20.2 % with tirzepatide and 13.7 % with semaglutide [1]. The trial was not blinded, ran only in the United States and Puerto Rico, and was funded by the manufacturer of tirzepatide [1].
Semaglutide 7.2 mg was not studied in this comparison. For this dose there are separate randomised data from the STEP UP trial [8]. Results from different trials can be compared with one another only to a limited extent.
Gastrointestinal symptoms were the most common, mostly mild to moderate, and above all during the dose increase [1]. Because of side effects, 6.1 % on tirzepatide and 8.0 % on semaglutide stopped the study medication (not withdrawal from participation in the trial). Superiority on safety cannot be inferred from these figures alone [1].
Body composition
In a substudy of SURMOUNT-1 with 160 people, body composition was measured with DXA. With tirzepatide, 74 % of the weight loss was fat and 26 % was lean soft tissue (excluding bone mineral). With placebo the split was similar (75 % and 25 %), but the total loss was much smaller [2]. Lean soft tissue is not the same as muscle mass. Muscle strength was not measured, and the study does not allow a comparison with semaglutide [2].
Heart and circulation
SELECT included people without diabetes who had overweight or obesity and established cardiovascular disease. With semaglutide 2.4 mg, the combined endpoint of cardiovascular death, non-fatal heart attack or non-fatal stroke occurred less often than with placebo (6.5 % versus 8.0 %, mean observation 39.8 months) [3]. The advantage was mainly due to fewer non-fatal heart attacks. For cardiovascular death the difference was not statistically established, and for non-fatal stroke there was hardly any difference [3]. The trial does not show whether the same benefit also occurs without established cardiovascular disease.
SURPASS-CVOT studied people with type 2 diabetes and established atherosclerotic cardiovascular disease. For the combined endpoint of cardiovascular death, heart attack or stroke, tirzepatide was not inferior to dulaglutide, which is already effective. Superiority was not demonstrated statistically [4]. The comparison here was with an effective treatment, whereas in SELECT it was with placebo. No ranking of the benefit for the heart and circulation can be derived from the two trials.
In SUMMIT, the participants had obesity and heart failure with preserved ejection fraction. Compared with placebo, tirzepatide reduced the combined endpoint of cardiovascular death or worsening of heart failure (9.9 % versus 15.3 %). The effect was due to fewer worsenings of heart failure. For cardiovascular death there was no advantage (2.2 % versus 1.4 %, not statistically established) [5]. Participants with and without type 2 diabetes benefited to a similar extent [5, 13].
Long term, stopping and open questions
Data over several years already exist: with semaglutide, the weight loss in SELECT lasted for up to 4 years [9]. Tirzepatide was studied for 176 weeks in people with obesity and prediabetes [10]. The effect seems tied to continued treatment. In a subgroup of the STEP 1 trial, in which the lifestyle programme also ended, participants regained about two thirds of the lost weight in the year after stopping semaglutide [11]. In SURMOUNT-4, weight rose by a mean of 14.0 % within 52 weeks after the switch from tirzepatide to placebo [12].
Further research is needed, among other things on an even longer duration of treatment and on keeping the weight off permanently after treatment ends. The results of SURMOUNT-MMO are still awaited. The trial studies disease and mortality in overweight or obesity without diabetes, in a broader population, with established cardiovascular disease or several risk factors [6]. The end of the trial is estimated for October 2027 [7]. These results should not be confused with the data already available on heart failure.
Conclusion
More weight loss does not automatically mean greater protection against events of the heart and circulation. The available studies do not allow a blanket ranking of the two substances with regard to their overall health benefit. What matters is the treatment goal, existing accompanying conditions, tolerability and the evidence available for the particular situation. The choice belongs in a conversation with a doctor.