Cagrilintide’s headline number was never cagrilintide’s.

Why this note exists

Cagrilintide is unusual on this site. Most compounds pepmg tracks have thin human evidence and loud marketing. This one has a large, recent, well-run phase 3 program published in The New England Journal of Medicine and The Lancet Diabetes & Endocrinology — and the marketing still misattributes it, because the trials tested a pairing and the market sells a single vial.[1][2][14]

In the index generated on August 15, 2026, 65 of 112 vendors carried a cagrilintide listing — 120 listings, twenty-second of 83 compounds by vendor coverage.[11] Nine of those 120 name a cagrilintide-and-semaglutide combination in the product title; 108 name cagrilintide with no second compound alongside it.[11] Where a size is stated, 10 mg is the most common vial at 50 of 115, with 5 mg next at 33.[11]

A note on what kind of evidence this is: every efficacy and safety figure below is human data, carrying its design, estimand and participant count. One paragraph reports rodent and in vitro work and is labeled as such. Sponsor-announced results that have not yet been published in a peer-reviewed journal are identified as sponsor announcements in the same sentence as the number, and pepmg does not convert a trial dose into a protocol for a research vial.

The pivotal trial

Four arms, and only one of them was the question

REDEFINE 1 is a phase 3a, 68-week, multicentre, double-blind, placebo-controlled and active-controlled trial in adults without diabetes with a BMI of 30 or higher, or 27 or higher with at least one obesity-related complication.[1] A total of 3,417 participants were randomized in a ratio of 21:3:3:7 — 2,108 to cagrilintide-semaglutide, 302 to semaglutide alone, 302 to cagrilintide alone and 705 to placebo, all with lifestyle intervention.[1] Its registry record, NCT05567796, was listed as active and not recruiting when queried on August 25, 2026.[12]

That ratio is the whole story of this note. The coprimary endpoints were the change in body weight and a reduction of 5% or more, in the combination arm as compared with placebo.[1] The two monotherapy arms exist to characterise the contribution of each component; they were not the trial's question, and they are each about a seventh the size of the combination arm.[1]

Mean weight reduction at week 68 as reported by the sponsor in its headline announcement of December 20, 2024; the leading figure is the trial product estimand, which assumes adherence to treatment, and the second is the treatment policy estimand, which is consistent with the intention-to-treat principle.[7][1]

The peer-reviewed report in the New England Journal of Medicine gives the primary result as −20.4% with cagrilintide-semaglutide against −3.0% with placebo, an estimated difference of −17.3 percentage points with a 95% confidence interval of −18.1 to −16.6 and P<0.001.[1] Gastrointestinal adverse events affected 79.6% of the combination group against 39.9% on placebo, and are described as mainly transient and mild-to-moderate.[1] The trial was funded by Novo Nordisk.[1]

A secondary, post hoc analysis of the same trial makes the arm gap concrete in a different currency. The proportion of participants reaching both a BMI under 27 and a waist-to-height ratio under 0.53 at week 68 was 30.3% on the combination, 19.1% on semaglutide, 9.0% on cagrilintide and 3.3% on placebo.[4] Post hoc analyses generate hypotheses rather than test them, and this one is labeled as such by its authors.[4]

The monotherapy record

One dedicated trial, and it was a phase 2

Strip out the combination trials and the reported evidence for cagrilintide on its own comes down to a single dedicated study: a phase 2 dose-finding trial published in The Lancet in 2021, run at 57 sites in ten countries, which randomized its participants between March and August 2019.[13] Seven hundred and six adults without diabetes were assigned across five once-weekly cagrilintide doses — 0.3, 0.6, 1.2, 2.4 and 4.5 mg — against 99 on liraglutide 3.0 mg and 101 on placebo, over a 26-week treatment period with a six-week untreated follow-up.[13]

Its results are modest and were reported as such. Mean weight reductions were 6.0% to 10.8% across the dose range against 3.0% on placebo, and at the top dose of 4.5 mg cagrilintide beat liraglutide 3.0 mg by 10.8% to 9.0%, with an estimated treatment difference of 1.8 percentage points at p=0.03.[13] Gastrointestinal adverse events occurred in 41% to 63% of cagrilintide groups against 32% on placebo, nausea in 20% to 47% against 18%.[13] The authors' own conclusion is a development statement, not an efficacy claim: the findings "support the development of molecules with novel mechanisms of action for weight management."[13]

Everything else is a reference arm. In REIMAGINE 2, a phase 3 study in type 2 diabetes reported in 2026, 2,713 people were randomized across six groups and the cagrilintide 2.4 mg arm was 152 of them — while the primary endpoint was the change in HbA1c with the combination versus semaglutide 2.4 mg, which the combination won by 0.16 percentage points (−1.91 against −1.75, 95% CI −0.27 to −0.05, p=0.0035).[14] A 0.16-point difference on the trial's primary endpoint is the kind of result that gets rounded away in a product description, and the cagrilintide arm is not what that endpoint measured at all.[14]

The pattern repeats where the combination is tested against semaglutide without any monotherapy arm. REDEFINE 5, in 331 participants across 21 sites in Japan and one in Taiwan, reported −18.4% with the combination against −11.9% with semaglutide at week 68.[3] That trial contains no cagrilintide-alone group and therefore says nothing about cagrilintide alone.[3]

The trial that will answer this

Cagrilintide alone is finally in phase 3. The result is due in 2027.

The most useful thing in the registry is not a published result — it is two trials that started in November 2025 and are designed to ask exactly the question this note is about.[15][20]

NCT07220642 and NCT07220759 are randomized, quadruple-masked, placebo-controlled phase 3 trials of cagrilintide against placebo — one in adults with overweight or obesity, estimated enrollment 300, the other in adults who also have type 2 diabetes, estimated enrollment 330.[15][20] Both list a single primary outcome, the relative change in body weight from baseline to end of treatment at week 64, both began on November 5, 2025, and both list an estimated primary completion in May 2027.[15][20] A third phase 3 monotherapy trial, NCT07745504, comparing two formulations of injectable cagrilintide and placebo over 26 weeks with an estimated enrollment of 285, was registered as not yet recruiting.[21]

Neither trial has posted results, and both were listed as active and not recruiting when the registry was queried on August 25, 2026.[15][20] Across the whole registry that search returned 43 registered studies of cagrilintide, the great majority of them phase 1 pharmacology or phase 2 and 3 studies of the combination.[19] Until those two trials report, the strongest evidence about cagrilintide on its own remains a 2021 phase 2 and two reference arms — and a registered trial with no reported results is not a result.[13][19]

Independent reads

Two analyses nobody at the sponsor wrote

A 2026 systematic review and meta-analysis in Diabetes, Obesity and Metabolism pooled three randomized trials totalling 3,545 participants to compare CagriSema and cagrilintide monotherapy against semaglutide.[16] It found the combination produced significantly greater weight loss than semaglutide — a mean difference of −7.47 percentage points, 95% CI −10.58 to −4.36 — and reported that cagrilintide monotherapy weight loss was comparable to semaglutide.[16]

That last finding does not sit flush with REDEFINE 1's own arm figures of 11.8% against 16.1%, and it should not be smoothed over. A pooled estimate across three trials and a single trial's arm-level report are different comparisons drawing on different data, and neither one supersedes the other here. pepmg reports both.[7][16]

The same meta-analysis carries a safety result that runs the other way from the usual "gentler than a GLP-1" framing of amylin analogues: cagrilintide monotherapy had a higher risk of serious adverse events than semaglutide, at a risk ratio of 1.83 with a 95% confidence interval of 1.03 to 3.24.[16] That interval only just clears 1.0, so the finding is suggestive rather than settled — but it is the opposite of the direction the market assumes.[16]

The largest independent placement of the combination comes from a network meta-analysis published in The BMJ in July 2026, covering 262 trials and 99,791 participants across 19 drugs, using GRADE and Cochrane Risk of Bias 2.[17] At one year, against lifestyle modification alone, it puts cagrilintide-semaglutide at −14.8% (95% CI −16.9 to −12.7) with moderate to high certainty, just behind tirzepatide at −14.9%.[17] The same review names CagriSema among the drugs with the highest discontinuation because of adverse events, in a group with risk ratios from 1.9 to 4.2, and reports increased fatigue risk with a risk ratio of 3.2, an absolute increase of 92 more per 1,000 people over one year.[17]

Regulatory status

Filed, not approved — and filed as a pair

No FDA approval for cagrilintide was identified in the sources checked for this note.[10] On December 18, 2025 Novo Nordisk submitted a New Drug Application for once-weekly CagriSema — cagrilintide 2.4 mg and semaglutide 2.4 mg — for weight management, based on REDEFINE 1 and REDEFINE 2, and said the FDA "is expected to review the CagriSema application in 2026."[5] No FDA approval was identified when rechecked on September 18, 2026.[10]

Submission is not approval, and the application is not for the compound in the vial. The cited filing is for CagriSema. It does not establish an approved dose, indication or population for cagrilintide monotherapy.[10][5]

The result that changed the odds

REDEFINE 4 missed its primary endpoint

On February 23, 2026 the sponsor announced headline results from REDEFINE 4, an 84-week open-label phase 3 trial of 809 randomized participants with obesity and one or more comorbidities, comparing CagriSema head-to-head against tirzepatide 15 mg.[8][9] The registry record confirms the design: phase 3, randomized, parallel assignment, masking none, actual enrollment 809, primary completion December 8, 2025.[9]

The trial "did not achieve its primary endpoint of demonstrating non-inferiority on weight loss for CagriSema compared to tirzepatide after 84 weeks."[8] Reported weight loss was 23.0% for CagriSema against 25.5% for tirzepatide on the efficacy estimand, and 20.2% against 23.6% on the treatment-regimen estimand, from a mean baseline body weight of 114.2 kg.[8] The sponsor described the safety profile as safe and well tolerated, with gastrointestinal adverse events most common and the vast majority mild to moderate and diminishing over time.[8]

These are sponsor-announced headline figures. As of August 25, 2026 no peer-reviewed report of REDEFINE 4 had appeared in the published literature, so the numbers above have not been through external review, and no subgroup or sensitivity tables are available to check them against.[8][9] An open-label design with a weight endpoint is also worth naming plainly: neither participants nor investigators were masked.[9]

Mechanism

What amylin is, and what the rodent work does and does not establish

Cagrilintide is a long-acting analogue of amylin, a pancreatic hormone co-secreted with insulin that induces satiety; the rationale for pairing it with a GLP-1 receptor agonist is that the two act on complementary pathways.[13][14] That is a mechanistic rationale, and mechanistic rationales are the cheapest thing in this field.

Animal and in vitro data, labeled as such: a 2026 study from a preventive-doping research centre characterised the metabolism of pramlintide, cagrilintide and KBP-066 using human skin and kidney S9 fractions and biological fluids, and confirmed predicted cagrilintide metabolites in authentic post-administration rat plasma samples, in order to develop an LC-MS/MS detection method for sports drug testing.[18] The same paper notes that semaglutide has been on the World Anti-Doping Agency's monitoring programme since 2024, and frames amylin receptor agonists as warranting consideration for the same reason.[18] None of that is efficacy evidence, none of it is human outcome data, and it is included here only because a detection method existing is a fact worth knowing for anyone in a tested sport.[18]

The market

A single vial for a two-drug result

The mismatch here is simpler than the usual one on this site. Nobody is misnaming the peptide — 108 of 120 listings say cagrilintide plainly, and nine say cagrilintide and semaglutide together.[11] The gap is that the number attached to the name in general circulation was produced by two molecules dosed together for 68 weeks under a protocol, and the single-compound arm of that same protocol produced a materially smaller one.[7][1]

The published dose is worth stating precisely, because it is the one piece of this that transfers cleanly. In the phase 3 program the studied dose is 2.4 mg once weekly, subcutaneous, in adults with overweight or obesity, escalated over an initial period and given with lifestyle intervention.[1][14] The phase 2 dose-finding trial tested 0.3, 0.6, 1.2, 2.4 and 4.5 mg once weekly for 26 weeks in adults without diabetes.[13] pepmg reports those figures as their sources published them, with species, route, population and study phase attached, and does not convert a trial dose into a protocol for a research vial or make any claim about what is inside one.[11]

One adherence detail from the sponsor's own announcement belongs here too, because it bounds every figure above: in REDEFINE 1, 57.3% of patients treated with CagriSema were on the highest dose at the end of the trial.[7] That is the gap between the two estimands in a single number — the trial product estimand asks what happens if people take the drug as intended, and rather more than a third of the combination arm did not end up at the top dose.[7]

Three things this note is not saying

First, it is not saying cagrilintide does not work. A phase 2 trial in 706 adults reported 6.0% to 10.8% mean weight reduction across its dose range against 3.0% on placebo, and a 302-person arm of a phase 3 trial reported 11.8%.[13][7] Those are real human results, and they are larger than most compounds in pepmg's index can show at all.

Second, it is not saying the combination is a failure. REDEFINE 1 met its coprimary endpoints with a −17.3 percentage-point difference against placebo, an independent BMJ network meta-analysis rates it at −14.8% with moderate to high certainty, and missing non-inferiority against the current market leader is a commercial problem before it is a clinical one.[1][17][8]

Third, it is not saying the amylin route is a dead end. REIMAGINE 2 showed the combination beating semaglutide on HbA1c, REDEFINE 5 showed it beating semaglutide on weight in an east Asian population, and the class is under active development.[14][3]

What it is saying is narrower: the figure that travels with the word cagrilintide was measured on cagrilintide plus semaglutide, the same trial's cagrilintide-only arm reported 11.8% against semaglutide-only's 16.1%, and no FDA approval was identified, while the cited application is for the pair.[7][5][10]

Questions people are asking