The question arrives in a settled form — which one works better — and the published record answers a different one. No randomized trial has ever given cagrilintide-semaglutide to one group and tirzepatide to another. Everything written about the pair is built from separate trials, run by rival sponsors, against their own placebo arms, and the reasons that arithmetic goes wrong are more useful than the answer people want from it. What each regimen did on its own terms is set out in the CagriSema article and in the tirzepatide results article.
The two anchor trials, and how close they land
REDEFINE 1 randomized 3,417 adults without diabetes, with a body-mass index of 30 or higher or 27 or higher with an obesity-related complication, in a 21:3:3:7 ratio to cagrilintide-semaglutide, semaglutide alone, cagrilintide alone or placebo for 68 weeks. Mean body weight changed −20.4% against −3.0% on placebo — an estimated difference of −17.3 percentage points (95% CI, −18.1 to −16.6; P < 0.001), on the treatment-policy estimand.[1]
SURMOUNT-1 randomized 2,539 adults with the same entry criteria, diabetes excluded, 1:1:1:1 to tirzepatide 5, 10 or 15 mg or placebo for 72 weeks. Mean weight change was −15.0% (95% CI, −15.9 to −14.2) at 5 mg, −19.5% (−20.4 to −18.5) at 10 mg and −20.9% (−21.8 to −19.9) at 15 mg, against −3.1% (−4.3 to −1.9) on placebo — a gap of roughly 17.8 percentage points at the top dose, on the treatment-regimen estimand.[3] The trial itself is walked through in the SURMOUNT-1 explainer.
Two things are worth noticing before anyone subtracts. The placebo arms landed within a tenth of a point of each other, at −3.0% and −3.1%, which is the single strongest argument that the two populations were behaving similarly. And the entry criteria are the same sentence. That is as comparable as separate trials get, and it still leaves 17.3 against 17.8 — half a percentage point, from instruments that were never calibrated against each other.
In diabetes the gap widens, and the doses interleave
REDEFINE 2 randomized 1,206 adults with type 2 diabetes, a body-mass index of 27 or more and a glycated hemoglobin of 7% to 10%, 3:1 against placebo for 68 weeks. Mean weight change was −13.7% against −3.4%, an estimated difference of −10.4 percentage points (95% CI, −11.2 to −9.5; P < 0.001).[2]
SURMOUNT-2 randomized 938 adults with obesity and type 2 diabetes 1:1:1 to tirzepatide 10 mg, 15 mg or placebo for 72 weeks. Least-squares mean weight change was −12.8% at 10 mg and −14.7% at 15 mg against −3.2% on placebo, for estimated differences of −9.6 percentage points (95% CI, −11.1 to −8.1) and −11.6 (−13.0 to −10.1), both P < 0.0001.[4] Its design is covered in the SURMOUNT-2 explainer.
Line those up and the combination does not sit above or below the other drug. It sits inside it: −10.4 points falls between tirzepatide’s 10 mg result of −9.6 and its 15 mg result of −11.6, and the confidence interval around the combination (−11.2 to −9.5) overlaps both of them across most of its width. Whichever rung of the tirzepatide ladder gets quoted decides the direction of the comparison, and nothing in the evidence decides which rung to quote.
One fixed dose against a ladder
That is a structural asymmetry rather than an accident of reporting. Tirzepatide was developed as a titration with a measured dose-response: SURMOUNT-1 published three separate maintenance doses and showed roughly six percentage points of separation between the bottom and the top of them.[3] CagriSema was developed as a single fixed combination, 2.4 mg of each component, and the phase 3 program tested that one strength.[1][2]
A comparison between a fixed dose and a ladder has no natural pairing. Matched against the tirzepatide dose most people are actually prescribed rather than the maximum, the ordering can reverse without a single number changing. How dose maps onto result across this class is the subject of the second-receptor article, and the expected weight-loss tool works from published trial figures rather than from a maximum dose.
The one head-to-head in the program measured something else
There is exactly one randomized head-to-head anywhere in the CagriSema program, and it is against semaglutide rather than tirzepatide. REDEFINE 5 randomized 331 adults across 21 sites in Japan and one in Taiwan 1:1 to the fixed-dose combination or to semaglutide 2.4 mg for 68 weeks. Mean weight change was −18.4% (SE 0.7) against −11.9% (SE 0.7), an estimated treatment difference of −6.5 percentage points (95% CI, −8.4 to −4.6).[7]
Its primary estimand was the trial product estimand, which estimates the effect assuming the treatment was taken as intended.[7] The four placebo-controlled trials above all report effects regardless of whether participants stopped. Those are different questions: one asks what the drug does to a person who takes it, the other what it does to a randomized group including the people who quit. In a class where 10% of one arm and 6% of the other discontinued over 68 weeks,[7] the two answers are not interchangeable, and the head-to-head number that circulates most widely is the one computed under the more favorable assumption.
Where the pooled analyses put them, and where they disagree
Two 2026 network meta-analyses scored both drugs, and they are the closest thing to an answer that exists. A systematic review and network meta-analysis of 262 trials and 99,791 participants estimated one-year weight change against lifestyle modification alone of −14.9% (95% CI, −16.0 to −13.9) for tirzepatide and −14.8% (−16.9 to −12.7) for cagrilintide-semaglutide.[5] The point estimates are a tenth of a percentage point apart. The intervals are not comparable in the same way: tirzepatide’s spans 2.1 points and CagriSema’s spans 4.2, because one rests on a decade of trials and the other on a program that reported in 2025.
The same analysis placed cagrilintide-semaglutide among the agents with the highest discontinuation for adverse events, in a group carrying risk ratios from 1.9 to 4.2, and reported a fatigue risk ratio of 3.2 for it, an absolute increase of 92 per 1,000 people over a year.[5] Tolerability is where the two regimens are least similar, and it does not appear in a weight percentage at all.
The second analysis is the one that matters most, because it changes its mind depending on the question. Across 25 trials and 12 interventions, tirzepatide 15 mg produced the greatest mean percent weight reduction (mean difference −17.97%), with CagriSema second at −17.84%. At the threshold of a 20% or greater reduction, the order reverses: CagriSema led with a risk ratio of 27.82 against placebo, and tirzepatide 15 mg followed at 23.70.[6]
That reversal is the finding. On the average person in the trial, one drug is marginally ahead; on the share of people reaching the outcome that changes a life, the other is. A page that reports either half alone quotes a real number from a real analysis and leaves the reader with a false picture. A third 2026 living review, prepared for the American College of Physicians across 69 studies and 112,511 participants, declined to separate the leaders at all, concluding that semaglutide and tirzepatide showed the most favorable results across outcomes while noting that direct head-to-head comparisons were limited.[8]
The asymmetry that has nothing to do with weight
Tirzepatide is a drug a pharmacy can dispense. The Orange Book product file lists 48 tirzepatide products across two Lilly applications, 215866 and 217806, with new-chemical-entity exclusivity recorded against every one of them expiring May 13, 2027, and the earliest-expiring listed patent, US 9,474,780, running to May 13, 2036.[9] Cagrilintide has no row in that file, alone or in combination, and no record in Drugs@FDA under any name.
So one side of this comparison has an approved label, an approved indication, a dose ladder, a contraindication list and an adverse-reaction table, and the other side has trial results. That is not a small difference in degree; it is a difference in what kind of object is being discussed, and it is the distinction set out in the approval-status article. No preparation sold under the CagriSema name today is the trial regimen, because the trial regimen was a manufactured investigational product given under a 68-week protocol.
Compounded drugs generally are not FDA-approved, and the FDA does not review them for safety, efficacy or quality before a pharmacy dispenses them. A combination whose amylin component has no approved reference product anywhere in the United States is a further step out again, because there is no label against which a strength could be checked.
What would actually settle it, and why it has not
A randomized trial giving one group cagrilintide-semaglutide and another tirzepatide, at a stated dose, for a stated duration, on a stated estimand. No such trial has published. Until one does, the honest form of the answer is a range rather than a winner: across four separate phase 3 trials and three pooled analyses, the two regimens land within roughly one percentage point of each other on mean weight change, swap positions at the 20% threshold, and differ most on tolerability and on availability.
Two further gaps are worth naming. No cardiovascular outcome result has been published for cagrilintide-semaglutide, so the class benefit that carried the incretins beyond weight has not been demonstrated for it; tirzepatide’s outcome evidence is a separate literature. And nothing in either program compares the two in the population where the rankings were least stable — people already at a high dose of an incretin. The general problem of reading across trials is the subject of the tirzepatide and semaglutide article, and how the figures on this site are established before publication is described in the methodology.