Cagrilintide Trial by Trial and What REDEFINE Actually Showed
Back to Blog
Metabolic research

Cagrilintide Trial by Trial and What REDEFINE Actually Showed

Primetime AdminOctober 14, 202618 min read

Key Takeaways

  • Cagrilintide is an investigational long-acting amylin and calcitonin receptor agonist from Novo Nordisk. It is not a GLP-1 receptor agonist and it is not FDA approved.
  • CagriSema is cagrilintide combined with semaglutide. Most large published results describe the combination, not cagrilintide alone, and the two are quoted interchangeably far too often.
  • REDEFINE 1's 20.4% and 22.7% come from the same 3,417 participants analysed under two estimands. The treatment policy estimand gives 20.4% and counts everyone randomised; the trial product estimand gives 22.7% and assumes protocol adherence.
  • Inside REDEFINE 1, semaglutide alone reached 14.9% and cagrilintide alone 11.5% under the treatment policy estimand, so the combination outperformed both of its own components.
  • REDEFINE 2 reported 13.7% in a type 2 diabetes population. The lower figure reflects a different population, not a weaker result.
  • REDEFINE 4, the only head-to-head against tirzepatide 15 mg, reported 23.0% against 25.5% on the efficacy estimand and 20.2% against 23.6% on the treatment-regimen estimand. Non-inferiority was not met.
  • Gastrointestinal events affected 79.6% of the CagriSema group in REDEFINE 1 against 39.9% on placebo, with nausea at 55%. Discontinuation due to adverse events was 6% against 3.7%.
  • Novo Nordisk states it filed a New Drug Application with the FDA and Health Canada in December 2025, with a decision expected in Q4 2026. No PDUFA date is public and filing is not approval.
  • Cagrilintide is ineligible for 503A compounding because it has no USP monograph, is not a component of an approved drug, and is not on the FDA's 503A bulks list.
  • RENEW, a dedicated cagrilintide monotherapy Phase 3 programme, exists and was announced in September 2025. It has not reported results.

Two numbers follow cagrilintide everywhere it is written about. One is 20.4%. The other is 22.7%. Both come from the same trial, the same participants, and the same 68 weeks. Most pages quote whichever one suits them and move on.

They are not competing estimates of one thing. They answer two different questions, and the difference is the single most useful thing to understand about this compound's evidence.

This article works through what each cagrilintide study actually found. It covers the mechanism, the early-phase work, the Phase 2 monotherapy trial, the combination trials, the REDEFINE and REIMAGINE programmes, the adverse events reported in the largest trial, and where the regulatory process currently sits.

Every study below carries a label. Peer-reviewed publications are marked as such. Company announcements are marked as company announcements, because a press release and a published paper are not the same kind of evidence. Trials still running are marked as running, with no results invented to fill the gap.

Cagrilintide is an investigational compound. Nothing here is a recommendation, a protocol, or guidance for use in people.

The short answer on cagrilintide

Cagrilintide is an investigational long-acting amylin analogue from Novo Nordisk, developed under the code AM833. It is not a GLP-1 receptor agonist. CagriSema is the investigational combination of cagrilintide with semaglutide, and most of the large published evidence describes that combination rather than cagrilintide alone.

REDEFINE 1 reported two weight-loss figures for CagriSema at week 68. The treatment policy estimand gave 20.4%, measuring everyone randomised regardless of whether they stayed on treatment. The trial product estimand gave 22.7%, estimating the effect had participants followed the protocol throughout. Different questions, not conflicting answers.

On regulation, Novo Nordisk has stated that it submitted a New Drug Application for obesity to the FDA and to Health Canada in December 2025, and that it expects a decision in the fourth quarter of 2026. A submission is not an approval. The FDA has not approved cagrilintide or CagriSema for any use, and no PDUFA date has been made public.

Verified 14 October 2026 against the primary publications in the New England Journal of Medicine, The Lancet and Lancet Diabetes & Endocrinology, Novo Nordisk company announcements, and the FDA's bulk drug substances pages. Trial figures, publication status and regulatory status were each checked against the source rather than carried over from secondary coverage.

What is cagrilintide and what is CagriSema

Cagrilintide is an investigational long-acting amylin analogue developed by Novo Nordisk. It activates amylin and calcitonin receptors, pathways involved in satiety and meal-related appetite signalling. It is not a GLP-1 receptor agonist. The FDA has not approved cagrilintide for any use, alone or in the CagriSema combination.

Amylin is a hormone released by the pancreas alongside insulin when a meal is eaten. It is one of the signals that tells the brain a meal has arrived. Natural amylin clumps into fibrils very readily, which is what made it difficult to turn into a usable long-acting medicine for decades.

CagriSema is the name for cagrilintide and semaglutide given together, each at 2.4 mg in the main Phase 3 trials. Semaglutide is an established GLP-1 receptor agonist. The combination pairs two separate appetite-signalling routes rather than pushing harder on one.

Three names get used interchangeably online, and they should not be. Cagrilintide alone is one investigational compound. Semaglutide alone is an approved medicine. CagriSema is a third thing, an investigational combination, and results from any one of the three do not describe the other two. A good deal of the confusion in cagrilintide coverage comes from quoting a CagriSema figure as though it were a cagrilintide figure.

Pairing the two mechanisms is a reasonable research idea. It is not, on its own, evidence that the combination works better, and the trials below are what settle that question.

How cagrilintide works in the body

Cagrilintide is described as a dual amylin and calcitonin receptor agonist. Amylin receptors are not single proteins. They are the calcitonin receptor paired with one of three accessory proteins called RAMP1, RAMP2 and RAMP3, producing the receptor complexes known as AMY1, AMY2 and AMY3. Cagrilintide activates these complexes and the calcitonin receptor itself.

A 2025 structural study in Nature Communications mapped how cagrilintide sits in each of those receptors. According to PubMed, the authors found it binds in an amylin-like way, while producing conformational dynamics at calcitonin-family receptors that differ from rat amylin or salmon calcitonin. They suggest that difference could contribute to its clinical effect. The word "could" is theirs, and it belongs in any honest summary. A structure explains how a molecule engages a receptor. It does not establish what happens to a person.

The receptors that matter sit mostly in the brainstem. The area postrema and the nucleus of the solitary tract are small regions that receive signals about what has been eaten and feed into the circuits governing fullness and nausea. That anatomical overlap is worth noting, because it is the same neighbourhood that produces the gastrointestinal side effects described later.

What makes cagrilintide long-acting is chemistry rather than biology. The development paper in the Journal of Medicinal Chemistry describes building a stable, lipidated analogue resistant to the fibril formation that defeats natural amylin. A C20 fatty diacid side chain lets the molecule bind albumin in the blood, which slows clearance considerably.

The measured consequence is a half-life of 159 to 195 hours, reported in the Phase 1b trial across cagrilintide doses from 0.16 mg to 4.5 mg. That is roughly a week. It is a pharmacokinetic measurement from a trial population, not a schedule, and it says nothing about how any material should be handled outside one.

Semaglutide reaches appetite by a different road, the GLP-1 receptor. Having two distinct mechanisms is the stated rationale for combining them. Rationale is not proof, and the comparison in REDEFINE 4 later in this article is a useful reminder of why.

Why amylin research moved beyond pramlintide

Amylin is not a new drug target. Pramlintide, sold as Symlin, was approved by the FDA on 16 March 2005 as an injectable adjunct in diabetes. It proved the pathway could be engaged in people.

It also showed the limits of a short-acting amylin analogue. Pramlintide required dosing with meals rather than weekly, and the weight effects reported historically were modest next to what later compounds produced. All Symlin products are now listed as discontinued in the FDA's records, which says something about how the field moved on.

The development paper for cagrilintide makes the design problem explicit. Natural amylin forms amyloid fibrils so readily that turning it into a stable long-acting medicine took sustained structure-activity work. Pramlintide solved the stability problem in one way and accepted frequent administration as the cost. Cagrilintide solved it differently, using lipidation and albumin binding to stretch exposure across about a week.

Two things follow, and only one of them is about effectiveness. A longer-acting molecule is a different pharmacological proposition from a short-acting one, which is why the research programme restarted rather than continuing. Whether that translates into better outcomes for people is a question for trials, not for molecular design. A compound being newer, longer-acting, or more elegantly engineered than its predecessor establishes none of those outcomes by itself.

Cagrilintide remains investigational. Pramlintide's approval does not extend to it.

What every major cagrilintide trial actually found

Method note. Every figure in the table below was checked on 14 October 2026 against its primary source. Peer-reviewed results come from the published papers in the New England Journal of Medicine, The Lancet, Lancet Diabetes & Endocrinology, and the American Heart Journal, retrieved through PubMed. Results that exist only as company statements come from Novo Nordisk announcements and are labelled as such. Where a trial reported both the treatment policy estimand and the trial product estimand, both appear, because quoting one without naming it is how most reporting on this compound goes wrong. Where a detail could not be confirmed in a primary source, it is marked unavailable rather than estimated. Registry identifiers were taken from publications and company statements; they were not independently confirmed against ClinicalTrials.gov, which could not be reached during this review.

Two terms carry the table. The treatment policy estimand measures outcomes across everyone randomised, whether or not they kept taking the drug. It is consistent with the intention-to-treat principle and it is the more conservative, more realistic figure. The trial product estimand estimates what would have happened had everyone followed the protocol throughout. Loose coverage calls these "ITT" and "on-treatment"; those are approximations of the terms the publications and the sponsor actually use.

  • Trial or programme: Phase 1b (Enebo) — Evidence status: Peer-reviewed primary publication — Size, population, duration: n=96 randomised, 95 exposed; adults with BMI 27.0 to 39.9, no diabetes; 20 weeks — Regimen studied and comparator: Cagrilintide 0.16 to 4.5 mg with semaglutide 2.4 mg, vs placebo with semaglutide 2.4 mg — Main result: Weight reduction 17.1% (cagrilintide 2.4 mg arm) vs 9.8% pooled placebo, both on semaglutide. Half-life 159 to 195 hours — Which figure to quote: Quote 17.1% only with its comparator. Both arms received semaglutide, so this is combination vs semaglutide alone, not cagrilintide vs semaglutide. Safety and pharmacokinetics were the primary endpoints, weight was exploratory — Source: Lancet 2021, PMID 33894838
  • Trial or programme: Phase 2 monotherapy (Lau) — Evidence status: Peer-reviewed primary publication — Size, population, duration: n=706 cagrilintide, 99 liraglutide, 101 placebo; adults without diabetes; 26 weeks — Regimen studied and comparator: Cagrilintide 0.3 to 4.5 mg vs liraglutide 3.0 mg vs placebo — Main result: Trial product estimand 6.0% to 10.8% across doses vs 3.0% placebo. Cagrilintide 4.5 mg 10.8% vs liraglutide 9.0%, difference 1.8%, p=0.03 — Which figure to quote: Quote the trial product estimand and say so. The paper notes treatment policy results were similar but reports the dose-response under the trial product estimand — Source: Lancet 2021, PMID 34798060
  • Trial or programme: Phase 2 combination, type 2 diabetes (Frias) — Evidence status: Peer-reviewed primary publication — Size, population, duration: n=92 total (31 CagriSema, 31 semaglutide, 30 cagrilintide); type 2 diabetes on metformin; 32 weeks — Regimen studied and comparator: All escalated to 2.4 mg — Main result: Weight 15.6% vs 5.1% semaglutide vs 8.1% cagrilintide. HbA1c −2.2, −1.8, −0.9 percentage points — Which figure to quote: Quote weight results, which were significant against both comparators. Do not quote the HbA1c advantage over semaglutide as a finding; it was not significant (p=0.075). With 92 participants this is a small, exploratory trial — Source: Lancet 2023, PMID 37364590
  • Trial or programme: REDEFINE 1 (Garvey) — Evidence status: Peer-reviewed primary publication — Size, population, duration: n=3,417; overweight or obesity without diabetes; 68 weeks — Regimen studied and comparator: CagriSema 2.4/2.4 vs semaglutide 2.4 vs cagrilintide 2.4 vs placebo — Main result: Treatment policy: CagriSema 20.4%, semaglutide 14.9%, cagrilintide 11.5%, placebo 3.0%. Trial product: 22.7%, 16.1%, 11.8%, 2.3% — Which figure to quote: Quote 20.4% as the headline. It is the treatment policy estimand and the paper's primary analysis, covering everyone randomised. Quote 22.7% only when naming it as the trial product estimand. Never present them as rival estimates of the same quantity — Source: NEJM 2025, PMID 40544433; NCT05567796
  • Trial or programme: REDEFINE 2 (Davies) — Evidence status: Peer-reviewed primary publication — Size, population, duration: n=1,206 (904 CagriSema, 302 placebo); overweight or obesity with type 2 diabetes; 68 weeks — Regimen studied and comparator: CagriSema 2.4/2.4 vs placebo — Main result: Treatment policy weight 13.7% vs 3.4%. HbA1c at or below 6.5% in 73.5% vs 15.9% — Which figure to quote: Quote 13.7% with its population attached. This trial enrolled people with type 2 diabetes and is not comparable to REDEFINE 1 — Source: NEJM 2025, PMID 40544432; NCT05394519
  • Trial or programme: REDEFINE 5 — Evidence status: Peer-reviewed primary publication — Size, population, duration: n=331; adults in Japan and Taiwan; 68 weeks — Regimen studied and comparator: CagriSema vs semaglutide 2.4 mg — Main result: Weight 18.4% vs 11.9%, difference 6.5 percentage points, p<0.0001, trial product estimand — Which figure to quote: Quote 18.4% with the East Asian population named. Results from one regional population do not transfer automatically to another — Source: Lancet Diabetes Endocrinol 2026;14(6):450-462; NCT05813925
  • Trial or programme: REIMAGINE 2 — Evidence status: Peer-reviewed primary publication — Size, population, duration: n=2,713; type 2 diabetes; 68 weeks — Regimen studied and comparator: CagriSema vs semaglutide — Main result: HbA1c −1.91 vs −1.75 percentage points, difference −0.16, p=0.0035 — Which figure to quote: Quote the HbA1c result, which was the primary endpoint. Weight-loss figures circulating for this trial are not supported by the publication's primary analysis — Source: Lancet Diabetes Endocrinol 2026;14(8):662-677; NCT06065540
  • Trial or programme: REDEFINE 3 — Evidence status: Design paper published; results not yet reported — Size, population, duration: n=7,076; established cardiovascular disease and BMI 25 or above, with and without type 2 diabetes; event-driven — Regimen studied and comparator: CagriSema vs placebo — Main result: No results. Safety endpoint is non-inferiority on three-point MACE; efficacy endpoint is superiority on four-point MACE — Which figure to quote: Quote nothing as a result. This is a cardiovascular outcomes trial that has not reported — Source: Am Heart J 2026; NCT05669755
  • Trial or programme: REDEFINE 4 — Evidence status: Company-announced result, not peer-reviewed — Size, population, duration: n=809; obesity; 84 weeks; open-label — Regimen studied and comparator: CagriSema vs tirzepatide 15 mg — Main result: Efficacy estimand: 23.0% vs 25.5%. Treatment-regimen estimand: 20.2% vs 23.6%. Non-inferiority was not met — Which figure to quote: Quote both estimands or neither. Non-inferiority was not demonstrated, so this trial does not establish that CagriSema matches tirzepatide. It also does not establish that tirzepatide is superior across other outcomes — Source: Novo Nordisk announcement, 23 February 2026. Registry identifier not located
  • Trial or programme: REDEFINE 9 — Evidence status: Company-announced result, not peer-reviewed — Size, population, duration: Sample size not stated in the announcement; adults with overweight or obesity; 68 weeks — Regimen studied and comparator: CagriSema 1.7/1.7 and 1.0/1.0 vs placebo — Main result: 21.0% vs 2.0% placebo, efficacy estimand, for the 1.0/1.0 arm — Which figure to quote: Quote 21.0% as a company-announced result, naming the 1.0/1.0 arm and the efficacy estimand. The trial tested two dose arms; results for the 1.7/1.7 arm were not reported, and no peer-reviewed publication exists — Source: Novo Nordisk announcement, 21 September 2026. Registry identifier not located
  • Trial or programme: REIMAGINE 5 — Evidence status: Company-announced result, not peer-reviewed — Size, population, duration: Type 2 diabetes; 60 weeks — Regimen studied and comparator: CagriSema 1.0/1.0 vs tirzepatide 5 mg — Main result: Weight 12.4% vs 9.1%. HbA1c reduction 1.71 vs 1.67 percentage points, non-inferiority met — Which figure to quote: Quote with the comparator dose named. Tirzepatide 5 mg is not its maximum dose. Superiority on weight does not imply superiority on glycaemic outcomes, where only non-inferiority was shown — Source: Novo Nordisk announcement, 21 September 2026
  • Trial or programme: REIMAGINE 1 — Evidence status: Ongoing study, no results reported — Size, population, duration: n=189; type 2 diabetes managed with diet and exercise; 40 weeks — Regimen studied and comparator: CagriSema 2.4/2.4 and 1.0/1.0 vs placebo — Main result: No results reported — Which figure to quote: Quote nothing as a result. Includes body composition and MRI sub-studies. Registry identifier omitted here because two conflicting candidates were found — Source: Novo Nordisk trial listing
  • Trial or programme: REDEFINE 8 — Evidence status: Ongoing study, no results reported — Size, population, duration: n=400; 104 weeks plus a 52-week extension — Regimen studied and comparator: CagriSema 2.4/2.4 vs placebo — Main result: No results reported. Includes body composition assessments — Which figure to quote: Quote nothing as a result. No primary source specifies the measurement method, so do not describe it as a DXA study — Source: Novo Nordisk trial listing
  • Trial or programme: REDEFINE 11 — Evidence status: Ongoing study, no results reported — Size, population, duration: n=600; 80 weeks plus an 80-week extension — Regimen studied and comparator: CagriSema 2.4/2.4 vs placebo — Main result: No results reported — Which figure to quote: Quote nothing as a result. An extended-duration study is a fair description. No primary source gives a readout date — Source: Novo Nordisk trial listing
  • Trial or programme: RENEW — Evidence status: Programme announced; no results reported — Size, population, duration: Cagrilintide monotherapy Phase 3 programme — Regimen studied and comparator: Cagrilintide alone — Main result: No results. Programme announced as due to start in Q4 2025, following a REDEFINE 1 sub-analysis showing cagrilintide 2.4 mg monotherapy at 11.8% vs 2.3% placebo — Which figure to quote: Quote the programme's existence, not results. RENEW corrects the common claim that no monotherapy development exists. A programme starting is not a programme reporting — Source: Novo Nordisk release, 16 September 2025

The pattern across that table is worth stating once. The large, peer-reviewed, placebo-controlled evidence is strong and consistent. The head-to-head evidence against the strongest approved comparator is weaker, newer, open-label, and did not meet its non-inferiority target. Those two facts sit side by side, and pages that report only the first are not describing the evidence.

What REDEFINE 1 and REDEFINE 2 tell us

REDEFINE 1 is the largest cagrilintide trial published. It randomised 3,417 adults with overweight or obesity and without diabetes across four groups, and ran 68 weeks.

The two figures everyone quotes come from the same participants analysed two ways. Under the treatment policy estimand, which the publication identifies as consistent with the intention-to-treat principle, mean weight change was −20.4% for CagriSema against −3.0% for placebo. That analysis counts everyone randomised, including those who stopped treatment, reduced their dose, or started something else. Under the trial product estimand, which estimates the effect assuming adherence to the protocol throughout, the figure was −22.7% against −2.3%.

Neither is wrong. They measure different things. The treatment policy figure answers what happened to the people who entered this trial. The trial product figure answers what the drug does when taken as designed. The first is the appropriate headline for describing a randomised population, which is why the paper leads with it.

The gap between them has a concrete cause. Only 57.3% of the CagriSema group was on the highest dose at week 68. When a meaningful share of participants reduce or stop treatment, an analysis that ignores that will report a larger effect than one that counts it.

The comparator arms matter as much as the headline. Under the treatment policy estimand, semaglutide alone reached 14.9% and cagrilintide alone 11.5%. The combination outperformed both of its own components in the same trial, under the same analysis, which is a genuinely useful piece of evidence and better than any cross-trial comparison.

REDEFINE 2 enrolled a different population and should never be read as a weaker version of REDEFINE 1. Its 1,206 participants all had type 2 diabetes alongside overweight or obesity. Weight reduction was 13.7% against 3.4% for placebo under the treatment policy estimand.

People with type 2 diabetes consistently lose less weight on incretin-based treatment than people without it, across every programme in this field. The lower figure reflects the population, not a less effective drug. Setting 20.4% beside 13.7% as though one trial outperformed the other compares two different groups of people.

On glycaemic outcomes, 73.5% of the CagriSema group reached an HbA1c at or below 6.5%, against 15.9% on placebo. That is a separate endpoint from weight and belongs in its own sentence.

Cagrilintide vs semaglutide vs tirzepatide

Cagrilintide works through amylin and calcitonin receptors, which carry meal-related satiety signals to the brainstem. Semaglutide works through the GLP-1 receptor, an incretin pathway affecting insulin release, gastric emptying and appetite. The two routes are separate, which is why they were combined. Different mechanisms alone prove nothing about outcomes.

The question "is cagrilintide better than semaglutide" needs splitting before it can be answered. Cagrilintide alone against semaglutide alone was measured inside REDEFINE 1, and semaglutide won that comparison: 14.9% against 11.5% under the treatment policy estimand. The combination beat both, at 20.4%. Most pages that claim cagrilintide outperforms semaglutide are quoting the combination figure.

Against tirzepatide, one direct comparison exists and its result is frequently misreported. REDEFINE 4 randomised 809 adults to CagriSema or tirzepatide 15 mg over 84 weeks, open-label. Under the efficacy estimand the figures were 23.0% against 25.5%. Under the treatment-regimen estimand they were 20.2% against 23.6%. Non-inferiority was not met.

That last sentence is the finding. A trial designed to show CagriSema was not meaningfully worse than tirzepatide did not show it. Whether tirzepatide is better across other endpoints, populations or durations is a separate question this trial was not built to settle. Our tirzepatide evidence review covers that compound's own trial record in detail.

REIMAGINE 5 compared CagriSema 1.0/1.0 against tirzepatide 5 mg in type 2 diabetes over 60 weeks, reporting 12.4% against 9.1% on weight with HbA1c non-inferiority met. Tirzepatide 5 mg is not its maximum dose, and that result exists only as a company announcement.

  • Compound: Cagrilintide alone — Mechanism: Amylin and calcitonin receptor agonist — Evidence being compared: REDEFINE 1 comparator arm; Phase 2 monotherapy — Result context: 11.5% at 68 weeks, treatment policy — Main limitation: Smallest published evidence base of the three; Phase 3 monotherapy programme has not reported
  • Compound: Semaglutide — Mechanism: GLP-1 receptor agonist — Evidence being compared: REDEFINE 1 comparator arm — Result context: 14.9% at 68 weeks, treatment policy — Main limitation: Approved and marketed, so its evidence base is far larger than the comparator arm shown here
  • Compound: CagriSema — Mechanism: Both of the above combined — Evidence being compared: REDEFINE 1, 2, 5, 9; REIMAGINE 2, 5; REDEFINE 4 — Result context: 20.4% at 68 weeks, treatment policy — Main limitation: Investigational; head-to-head against tirzepatide did not meet non-inferiority
  • Compound: Tirzepatide — Mechanism: GIP and GLP-1 receptor agonist — Evidence being compared: REDEFINE 4 head-to-head — Result context: 25.5% efficacy estimand, 23.6% treatment-regimen, 84 weeks — Main limitation: Single open-label comparison; different trial duration from the REDEFINE 1 figures above

No published trial compares cagrilintide monotherapy against tirzepatide. Comparisons between the 20.4% from REDEFINE 1 and figures from separate tirzepatide trials are cross-trial guesswork, with different populations, durations and analyses on each side.

What cagrilintide side-effect data show

The adverse-event figures below come from Novo Nordisk's company release reporting the REDEFINE 1 results published in the New England Journal of Medicine. They describe CagriSema, not cagrilintide alone.

  • Event: Gastrointestinal adverse events — CagriSema: 79.6% — Placebo: 39.9%
  • Event: Nausea — CagriSema: 55% — Placebo: 12.6%
  • Event: Constipation — CagriSema: 30.7% — Placebo: 11.6%
  • Event: Vomiting — CagriSema: 26.1% — Placebo: 4.1%
  • Event: Discontinuation due to adverse events — CagriSema: 6% — Placebo: 3.7%

The publication describes these events as mainly transient and mild to moderate. REDEFINE 2, in the type 2 diabetes population, reported gastrointestinal events in 72.5% against 34.4%, with discontinuation due to adverse events at 8.4% against 3%.

Three things are worth drawing out of those numbers.

Reporting an event during a trial does not establish that the drug caused it. Nearly 40% of the placebo group reported gastrointestinal events too. The meaningful quantity is the difference between arms, not the headline percentage.

Discontinuation is a blunter measure of tolerability than any single symptom, and it cuts in an unexpected direction here. Only 6% of the CagriSema group stopped because of adverse events, against 3.7% on placebo, despite nausea affecting more than half of them. People tolerated considerable discomfort and stayed in the trial. That also means the 57.3% who reached the highest dose is a better guide to real-world tolerability than the discontinuation figure alone.

Rates depend on the population, the escalation schedule, the duration and the monitoring. A screened, supervised trial population is not a general one.

On the symptoms people search for most, the evidence is thinner than the questions deserve. Fatigue, depression, hair loss and heart-rate changes were not reported as notable findings in the REDEFINE 1 or REDEFINE 2 summaries available. That is not the same as showing they do not occur. Trials of this size and length detect common events well and rare or slow-developing ones poorly, and outcomes that were not specifically measured cannot be ruled in or out afterwards. Anyone claiming cagrilintide definitely does or does not cause hair loss is going beyond the published evidence in both directions.

A prespecified body composition analysis in 252 REDEFINE 1 participants reported fat mass down 35.7% against 5.7% on placebo, and lean soft tissue mass down 14.4% against 4.3%. No primary source names the measurement method, and no functional outcomes such as strength were reported, so what that lean tissue change means in practice is unresolved.

Is cagrilintide FDA approved and when could CagriSema be available

No. The FDA has not approved cagrilintide, and it has not approved CagriSema.

Novo Nordisk has stated that it submitted a New Drug Application for obesity to the FDA and to Health Canada in December 2025, and that it expects a regulatory decision in the fourth quarter of 2026. No PDUFA date has been made public, and a company's stated expectation is not an agency commitment.

The distinction between the molecule and the product matters here. The application covers CagriSema, the combination. Cagrilintide as a single agent is in a separate and earlier stage of development through the RENEW programme, which has not reported Phase 3 results.

Three steps are routinely collapsed into one in coverage of this compound. Filing means a sponsor has submitted its dossier. Review means the agency is examining it, a process that can end in approval, rejection, or a request for more data. Approval means the agency has cleared a specific product for a specific use. Only the first has happened.

On compounding, the accurate statement is narrower than what circulates. Under section 503A, a bulk drug substance is eligible for compounding only if it has a USP or NF monograph, is a component of an FDA-approved drug, or appears on the FDA's 503A bulks list. Cagrilintide meets none of those three conditions. It is unapproved, so it is not a component of an approved drug, and it does not appear on the list. It is therefore ineligible, not because the FDA issued a ruling against it, but because it does not satisfy the statutory route in the first place. Our guide to peptide legality covers the 503A category system and recent enforcement activity in more detail, though it does not address cagrilintide specifically.

This article gives no guidance on obtaining cagrilintide, and research-use-only labelling does not resolve legal questions in every jurisdiction or transaction.

What comes next for cagrilintide and amylin research

Three CagriSema trials are running without published results. REDEFINE 3 is the cardiovascular outcomes trial, with 7,076 participants who have established cardiovascular disease, testing non-inferiority on three-point MACE for safety and superiority on four-point MACE for efficacy. It is event-driven and has not reported. REDEFINE 8 runs 104 weeks with body composition assessments in 400 participants. REDEFINE 11 runs 80 weeks with an extension in 600 participants. None has produced a result, and no public readout date was found for any of them.

RENEW deserves separate mention because it corrects a claim repeated across much of the cagrilintide coverage. A dedicated Phase 3 programme for cagrilintide as a monotherapy does exist. Novo Nordisk announced it in September 2025, on the strength of the REDEFINE 1 sub-analysis showing cagrilintide alone at 11.8% against 2.3% for placebo. Saying no monotherapy development exists is wrong. Saying monotherapy Phase 3 results exist would be equally wrong.

Two other investigational amylin-related compounds come up alongside cagrilintide and are frequently confused with it. Zenagamtide, the name adopted for the compound previously called amycretin, is Novo Nordisk's unimolecular amylin and GLP-1 receptor co-agonist, meaning one molecule engaging both pathways rather than two compounds given together. It has reported early-phase results in both oral and subcutaneous forms and has not completed Phase 3. Eloralintide is Eli Lilly's selective amylin receptor agonist, which reported Phase 2 results in 263 participants over 48 weeks, with fixed doses producing 9.5% to 20.1% against 0.4% for placebo.

Those compounds sit at earlier stages than CagriSema, in different trials with different populations, durations and endpoints. Lining their percentages up against the REDEFINE figures produces a ranking that looks informative and is not. Our retatrutide evidence review makes the same point about a differently targeted compound.

How to read cagrilintide research without confusing the evidence

Evidence about this compound arrives in at least five forms, and they are not interchangeable. A randomised controlled trial tests a prespecified question in people. A mechanistic study explains how a molecule engages a receptor. A meta-analysis pools results across trials and inherits every limitation of what it pooled. A registry entry records that a trial exists. A company announcement reports what a sponsor chose to disclose before peer review. Treating the last as equivalent to the first is the most common error in this area.

Five questions settle most confusion about a headline figure. Which population was enrolled, since diabetes status alone moves results by several percentage points. How long did the trial run. What was the comparator, and at what dose. Which estimand produced the number. How many participants stayed on treatment, since that is what separates 20.4% from 22.7%.

Separately, a certificate of analysis answers a narrower question than it is usually asked. It can document that a tested sample matches an expected identity and report measured purity by a stated method. It cannot establish safety, effectiveness, regulatory approval, or that every vial sold under a label matches the one tested. Our guide to reading a certificate of analysis covers what each section supports, and our supplier due-diligence checklist covers documentation practices generally.

Material sold online under the cagrilintide name may differ in identity, purity and potency from the compound studied in these trials. Nothing in the REDEFINE or REIMAGINE programmes describes any material other than what Novo Nordisk manufactured to its own specification and administered under protocol.

Research-grade status is not approval for human use, and no purity figure changes that.

PrimeTime BioLabs sells research peptides, including Metabolic X, a research vial combining retatrutide and cagrilintide. We have a direct commercial interest in that product, and this article is not an independent endorsement of it. No published clinical trial has evaluated a retatrutide and cagrilintide combination. Every result described above concerns cagrilintide alone or cagrilintide with semaglutide, and none of it transfers to a different pairing.

The bottom line on cagrilintide

Cagrilintide's evidence is real and growing, and it is routinely quoted badly. The 20.4% and 22.7% figures from REDEFINE 1 both describe the same trial under different estimands. Results from the combination do not describe cagrilintide alone. Results from a type 2 diabetes population do not describe one without diabetes. Company announcements are not peer-reviewed findings, and the one head-to-head comparison against tirzepatide did not meet its non-inferiority target.

CagriSema is under FDA review. Cagrilintide monotherapy is earlier still. The primary publications and the regulatory record are where any specific claim should be checked.

References

  1. Enebo LB, Berthelsen KK, Kankam M, et al. Safety, tolerability, pharmacokinetics, and pharmacodynamics of concomitant administration of multiple doses of cagrilintide with semaglutide 2.4 mg for weight management: a randomised, controlled, phase 1b trial. Lancet. 2021;397(10286):1736-1748. PMID 33894838.
  2. Lau DCW, Erichsen L, Francisco AM, et al. Once-weekly cagrilintide for weight management in people with overweight and obesity: a multicentre, randomised, double-blind, placebo-controlled and active-controlled, dose-finding phase 2 trial. Lancet. 2021;398(10317):2160-2172. PMID 34798060.
  3. Frias JP, Deenadayalan S, Erichsen L, et al. Efficacy and safety of co-administered once-weekly cagrilintide 2.4 mg with once-weekly semaglutide 2.4 mg in type 2 diabetes: a multicentre, randomised, double-blind, active-controlled, phase 2 trial. Lancet. 2023;402(10403):720-730. PMID 37364590.
  4. Garvey WT, Blüher M, Osorto Contreras CK, et al. Coadministered Cagrilintide and Semaglutide in Adults with Overweight or Obesity (REDEFINE 1). N Engl J Med. 2025;393(7):635-647. PMID 40544433.
  5. Davies MJ, Bajaj HS, Broholm C, et al. Cagrilintide-Semaglutide in Adults with Overweight or Obesity and Type 2 Diabetes (REDEFINE 2). N Engl J Med. 2025;393(7):648-659. PMID 40544432.
  6. Kruse T, Hansen JL, Dahl K, et al. Development of Cagrilintide, a Long-Acting Amylin Analogue. J Med Chem. 2021;64(15):11183-11194. PMID 34288673.
  7. Cao J, Belousoff MJ, Johnson RM, et al. Structural and dynamic features of cagrilintide binding to calcitonin and amylin receptors. Nat Commun. 2025;16(1):3389. PMID 40204768.
  8. REDEFINE 5. Lancet Diabetes Endocrinol. 2026;14(6):450-462.
  9. REIMAGINE 2. Lancet Diabetes Endocrinol. 2026;14(8):662-677.
  10. REDEFINE 3 trial design. Am Heart J. 2026.
  11. Novo Nordisk A/S. CagriSema 2.4 mg / 2.4 mg demonstrated 22.7% mean weight reduction in adults with overweight or obesity in REDEFINE 1, published in New England Journal of Medicine. 22 June 2025.
  12. Novo Nordisk A/S. CagriSema demonstrated superior weight loss in REDEFINE 1. Company announcement No 99/2024, 20 December 2024.
  13. Novo Nordisk A/S. Novo Nordisk presents phase 3 data for next-generation amylin cagrilintide, leading to advancement into dedicated clinical programme (RENEW). 16 September 2025.
  14. US Food and Drug Administration. Bulk Drug Substances Used in Compounding Under Section 503A of the FD&C Act.
  15. US Food and Drug Administration. Drugs@FDA: SYMLIN (pramlintide acetate), NDA 021332, approved 16 March 2005.

Frequently Asked Questions

This needs splitting into two questions. Cagrilintide alone against semaglutide alone was measured directly inside REDEFINE 1, and semaglutide produced more weight reduction: 14.9% against 11.5% under the treatment policy estimand. CagriSema, the combination of both, reached 20.4% against semaglutide's 14.9% in the same trial. So the combination outperformed semaglutide, while cagrilintide by itself did not. Claims that cagrilintide beats semaglutide almost always quote the combination figure while naming the single compound. Both comparisons come from the same 68-week trial under the same analysis, which makes them considerably more reliable than comparisons drawn across separate studies.

They act on entirely different receptor systems. Cagrilintide is an amylin and calcitonin receptor agonist. Tirzepatide activates the GIP and GLP-1 receptors, and it is an approved medicine rather than an investigational one. The only direct comparison is REDEFINE 4, which randomised 809 adults to CagriSema or tirzepatide 15 mg over 84 weeks in an open-label design. It reported 23.0% against 25.5% under the efficacy estimand, and 20.2% against 23.6% under the treatment-regimen estimand. Non-inferiority was not met, meaning the trial did not establish that CagriSema performs comparably. It also exists only as a company announcement, without peer review.

On weight reduction in trials, the published evidence is substantial, though most of it describes the combination. CagriSema produced 20.4% mean weight reduction against 3.0% for placebo across 3,417 participants over 68 weeks in REDEFINE 1, and 13.7% against 3.4% in a type 2 diabetes population in REDEFINE 2. Cagrilintide alone produced 11.5% in the REDEFINE 1 comparator arm, and 6.0% to 10.8% across doses in the earlier Phase 2 monotherapy trial. All of that describes the compound Novo Nordisk manufactured, administered under trial protocol with lifestyle support. It does not establish long-term safety, and the one head-to-head against tirzepatide did not meet its non-inferiority target.

No. Cagrilintide is an amylin analogue. It activates amylin receptors, which are the calcitonin receptor paired with accessory proteins called RAMP1, RAMP2 and RAMP3 to form the AMY1, AMY2 and AMY3 complexes, along with the calcitonin receptor itself. GLP-1 receptor agonists such as semaglutide act on a completely separate receptor in the incretin system. The confusion arises because cagrilintide's best-known evidence comes from CagriSema, where it is paired with semaglutide, which genuinely is a GLP-1 receptor agonist. Combining two separate appetite-signalling routes was the whole rationale for the pairing.

No. Cagrilintide is investigational and has not been approved by the FDA for any use, alone or as part of CagriSema. Novo Nordisk has stated it submitted a New Drug Application for obesity to the FDA and Health Canada in December 2025, with a decision expected in the fourth quarter of 2026. Submitting an application is not the same as receiving approval, and the agency may approve, reject, or request further data. Until a regulatory decision is made, there is no approved product for a prescriber to prescribe. This is general regulatory information, not legal or medical advice about any individual situation.

No availability date exists, because no approval decision has been made. Novo Nordisk has said it expects a regulatory decision in the fourth quarter of 2026, following its December 2025 submission to the FDA and Health Canada. A company's stated expectation is a forecast, not a commitment from the agency, and no PDUFA date has been made public. Regulatory timelines in this field move regularly. Even after a decision, availability depends on manufacturing, distribution and market factors that sit outside the review process. Any specific availability date circulating now is speculation.

The fullest picture comes from REDEFINE 1, and it describes CagriSema rather than cagrilintide alone. Gastrointestinal adverse events affected 79.6% of the CagriSema group against 39.9% on placebo. Nausea was reported by 55% against 12.6%, constipation by 30.7% against 11.6%, and vomiting by 26.1% against 4.1%. Most were described as transient and mild to moderate. Discontinuation due to adverse events was 6% against 3.7%. In REDEFINE 2, gastrointestinal events affected 72.5% against 34.4%, with discontinuation at 8.4% against 3%. Reporting an event in a trial does not establish the drug caused it, and the placebo rates show why the between-group difference matters more than the headline number.

The available trial summaries do not report fatigue, depression or hair loss as notable findings in REDEFINE 1 or REDEFINE 2. That is a genuinely limited answer, and it should not be read in either direction. An outcome that was not specifically measured and reported cannot be confirmed or ruled out afterwards. Trials of this size detect common events well, and detect rare, slow-developing or subjective outcomes much less reliably. What the evidence does document clearly is the gastrointestinal profile. Anyone stating confidently that cagrilintide definitely causes or definitely does not cause these effects is going beyond what has been published.

The Phase 1b trial measured a half-life of 159 to 195 hours across cagrilintide doses from 0.16 mg to 4.5 mg, which is roughly seven days. Half-life describes the time for blood concentration to fall by half, so a compound remains measurable for considerably longer than one half-life, and reaching a steady concentration takes several. The long duration comes from a C20 fatty diacid side chain that binds albumin in the blood and slows clearance. This is a pharmacokinetic measurement from a supervised trial population. It is not a schedule and carries no implication for how any material should be used.

No trial establishes that any compound is safe for every person in every context, and cagrilintide is investigational with no approval anywhere. On monotherapy specifically, the published evidence is the REDEFINE 1 comparator arm at 11.5% over 68 weeks and the Phase 2 dose-finding trial at 6.0% to 10.8% over 26 weeks. The RENEW Phase 3 monotherapy programme was announced in September 2025 and has not reported results, so the largest monotherapy evidence does not yet exist. Everything published describes material manufactured to a defined specification and given under medical supervision with screening and monitoring. None of it transfers to use outside that setting.

No published clinical trial has evaluated a retatrutide and cagrilintide combination. Each compound has its own separate research record, and neither record says anything about the pair. Retatrutide's evidence comes from Eli Lilly's TRIUMPH programme, covered in our [retatrutide evidence review](https://www.primetimebiolabs.com/blog/retatrutide-peptide-explained). Cagrilintide's combination evidence comes from pairing with semaglutide, which is a different molecule acting on a different receptor. Results from two compounds studied separately do not establish that combining them is safe or effective; that inference does not follow, and it is the sort of reasoning trials exist to test. This article gives no guidance on combining compounds.

Cagrilintide is a single investigational compound, a long-acting amylin analogue. CagriSema is cagrilintide and semaglutide given together, each at 2.4 mg in the main Phase 3 trials. They are not interchangeable terms and their results are not interchangeable either. REDEFINE 1 measured both in the same trial: CagriSema reached 20.4% weight reduction under the treatment policy estimand while cagrilintide alone reached 11.5%. Quoting the 20.4% figure as a cagrilintide result roughly doubles what the single compound actually produced. The FDA submission covers the combination, while cagrilintide as a monotherapy sits at an earlier development stage.

Cagrilintide is an unapproved investigational compound, which shapes its legal position in several ways at once. Under section 503A, a bulk substance is eligible for compounding only if it has a USP or NF monograph, is a component of an FDA-approved drug, or appears on the FDA's 503A bulks list. Cagrilintide meets none of those conditions, so it is ineligible, not through a specific ruling against it but because it does not satisfy the statutory route. Research-use-only labelling describes an intended use for laboratory work; it is not a determination of legality for every transaction or jurisdiction, and it makes no claim about what a vial contains. Our [guide to peptide legality](https://www.primetimebiolabs.com/blog/are-peptides-legal) covers the wider framework. This is not legal advice.

Both figures come from REDEFINE 1, from the same 3,417 participants over the same 68 weeks, analysed under two different estimands. The treatment policy estimand gives 20.4% and measures outcomes across everyone randomised, whether or not they continued treatment, reduced their dose, or started something else. It is consistent with the intention-to-treat principle and is the paper's primary analysis. The trial product estimand gives 22.7% and estimates what would have happened had participants followed the protocol throughout. The gap exists because only 57.3% of the CagriSema group was on the highest dose at week 68. Neither number is wrong; quoting either without naming which one it is misleads the reader.

Zenagamtide is the name adopted for the compound previously called amycretin. It is Novo Nordisk's unimolecular amylin and GLP-1 receptor co-agonist, meaning one molecule engaging both pathways rather than two compounds combined. It has reported early-phase results in oral and subcutaneous forms and has not completed Phase 3. Eloralintide is Eli Lilly's selective amylin receptor agonist, which reported Phase 2 results in 263 participants over 48 weeks, with fixed doses producing 9.5% to 20.1% against 0.4% for placebo. Neither is approved. Both sit at earlier development stages than CagriSema, in trials with different populations, durations and endpoints, so their percentages cannot be ranked against the REDEFINE figures.