Cagrilintide
Cagrilintide is a long-acting amylin analogue[1]. Studies test it for weight loss[2]. In a Novo Nordisk-funded phase 3 trial, mean weight loss with cagrilintide alone was 11.5% at 68 weeks[3]. Gut events affected 54.0% of that group during treatment[3]. CagriSema combines cagrilintide with semaglutide[3]. Its results are not cagrilintide-only results. Trial findings do not establish the contents of a separately bought vial.
What is Cagrilintide?
Amylin is a pancreatic hormone involved in feeling full[2]. The drug was made as a stable form with a lipid group to give it a longer action[1]. It is not the same molecule as a GLP-1 drug. Semaglutide is the other drug in CagriSema[3].
The two names answer different questions. This drug can be studied on its own. CagriSema tests it alongside semaglutide. The published programme includes both, which makes the arm named beside each result important.
What has been published in humans?
What did cagrilintide alone do?
The earlier phase 2 trial enrolled 706 people in total[4]. The five groups receiving the drug held 506 people. Another 99 received liraglutide. Pooled placebo held 101. Novo Nordisk paid for the trial[2].
| Study and population | Single-agent finding | Interpretation and funding | Checked |
|---|---|---|---|
| NCT03856047; 706 adults without diabetes; cagrilintide groups 101, 100, 102, 102 and 101 | At week 26, trial-product mean weight loss ranged from 6.0% to 10.8% across 0.3-4.5 mg; placebo 3.0%. All tested doses differed from placebo, p<0.001[2]. | Novo Nordisk. Trial-product analysis assumes adherence; treatment-policy analysis also showed weight loss. | 2026-09-06 |
| Same trial: cagrilintide 4.5 mg versus liraglutide 3.0 mg | 10.8% versus 9.0% loss; estimated difference 1.8 percentage points, p=0.03[2] | Novo Nordisk. These are the named trial doses, not a general ranking of drugs. | 2026-09-06 |
| REDEFINE 1, NCT05567796; 302 assigned cagrilintide, 705 placebo, within 3,417 adults without diabetes | At week 68, treatment-policy mean weight change was -11.5% with cagrilintide 2.4 mg and -3.0% with placebo[3]. | Novo Nordisk. This analysis includes the effects of stopping treatment or using rescue interventions. | 2026-09-06 |
The phase 2 group counts come from the trial record[4]. They add to 506 for this drug, not 706. The larger number is the full trial. The abstract puts the group split in one sentence that is easy to misread.
A mean is a figure for a group, not a promise to each person. The confidence interval describes the uncertainty in the estimate. It is not a range within which every person's weight loss must fall. The trial group and the way the result was estimated both matter.
REDEFINE 1 also reports a trial-product analysis[3]. Its weight-loss estimate for the drug alone is 11.8%, compared with 2.3% for placebo. This asks what would happen with treatment taken as intended. It is not the same analysis as the 11.5% and 3.0% figures above.
What do studies show about Cagrilintide with other drugs?
What changes when semaglutide is added?
CagriSema results belong to the pair. In REDEFINE 1, 2,108 people were assigned the pair and 302 semaglutide alone[3]. All arms also received lifestyle intervention. Novo Nordisk paid for the work. The table keeps the two statistical questions apart.
| Study or outcome | Result | Source and boundary | Checked |
|---|---|---|---|
| REDEFINE 1, week 68 weight change, treatment-policy | CagriSema -20.4%; semaglutide -14.9%; cagrilintide -11.5%; placebo -3.0% | PMID 40544433; Novo Nordisk[3] | 2026-09-06 |
| Same analysis: combination minus comparator | Versus placebo -17.3 percentage points (95% CI -18.1 to -16.6); versus semaglutide -5.5 (-6.7 to -4.3); versus cagrilintide -8.9 (-10.1 to -7.7) | All p<0.001; these differences concern the combination[3] | 2026-09-06 |
| REDEFINE 1, trial-product weight change | CagriSema -22.7%; cagrilintide alone -11.8%; placebo -2.3% | Different estimand from the treatment-policy rows[3] | 2026-09-06 |
| REDEFINE 1, combination versus placebo | Waist change -17.5 versus -4.0 cm; systolic pressure -9.9 versus -3.2 mmHg | Table 2; Novo Nordisk. Not single-agent effects[3] | 2026-09-06 |
| REDEFINE 1, physical function | IWQOL-Lite-CT change 25.9 versus 14.0; SF-36 physical-function change 7.1 versus 3.6 | Combination versus placebo; higher scores mean better function[3] | 2026-09-06 |
| Phase 1b, NCT03600480; 96 randomised, 95 exposed | Week 20 loss: cagrilintide 1.2 mg plus semaglutide 15.7%; 2.4 mg plus semaglutide 17.1%; pooled placebo plus semaglutide 9.8% | PMID 33894838; Novo Nordisk. Every group received semaglutide 2.4 mg[5]. | 2026-09-06 |
| Same phase 1b comparison | Differences -6.0 percentage points (95% CI -9.9 to -2.0) and -7.4 (-11.2 to -3.5), respectively | A separate 4.5 mg cohort had 15.4% loss versus 8.0% matched control; difference -7.4 (-12.8 to -2.1)[5]. | 2026-09-06 |
In REDEFINE 1, the proportion losing at least 20% was 53.6% with CagriSema versus 1.9% with placebo[3]. For at least 25% loss, it was 34.7% versus 1.0%. For at least 30%, it was 19.3% versus 0.4%. These are treatment-policy estimates for the pair. They are not response rates for this drug alone.
What did the diabetes studies find?
The phase 2 diabetes trial compared three active groups[6]. The larger REDEFINE 2 trial compared CagriSema with placebo[7]. Neither design lets the pair's whole effect be assigned to this drug. Novo Nordisk paid for both trials.
| Trial | Findings | Source and limit | Checked |
|---|---|---|---|
| NCT04982575; 92 people; CagriSema 31, semaglutide 31, cagrilintide 30; 32 weeks | Weight change -15.6%, -5.1%, -8.1%; HbA1c change -2.2, -1.8, -0.9 percentage points, respectively | PMID 37364590; Novo Nordisk[6] | 2026-09-06 |
| Same trial, HbA1c comparison | Combination versus cagrilintide: -1.3 percentage points (95% CI -1.7 to -0.8), p<0.0001. Versus semaglutide: -0.4 (-0.8 to 0.0), p=0.075. | The HbA1c difference from semaglutide was not statistically significant[6]. | 2026-09-06 |
| Same trial, fasting glucose | Changes -3.3, -2.5 and -1.7 mmol/L for combination, semaglutide and cagrilintide | Combination differed from cagrilintide, p=0.0010, but not semaglutide, p=0.10[6]. | 2026-09-06 |
| Same trial, time in glucose range 3.9-10.0 mmol/L | Baseline 45.9%, 32.6%, 56.9%; week 32 88.9%, 76.2%, 71.7%, in the same group order | Continuous glucose monitoring; Novo Nordisk[6] | 2026-09-06 |
| REDEFINE 2, NCT05394519; 1,206 people; CagriSema 904, placebo 302 | Week 68 weight change -13.7% versus -3.4%; difference -10.4 percentage points (95% CI -11.2 to -9.5), p<0.001 | Novo Nordisk; treatment-policy estimand; no single-agent arm[7] | 2026-09-06 |
| REDEFINE 2, HbA1c | Change -1.8 versus -0.4 percentage points; difference -1.4 (95% CI -1.6 to -1.2). HbA1c at most 6.5%: 73.5% versus 15.9%. | Combination versus placebo[7] | 2026-09-06 |
What about tirzepatide or retatrutide?
REDEFINE 4 compared CagriSema against tirzepatide, not alongside it[13]. Novo Nordisk reported headline results from 809 people after 84 weeks. Under the efficacy estimand, weight loss was 23.0% with CagriSema and 25.5% with tirzepatide. Under the treatment-regimen estimand, it was 20.2% and 23.6%. The study did not meet its weight-loss non-inferiority endpoint.
That sponsor report describes an open-label comparison[13]. It does not test taking all three drugs together. The cited semaglutide trials also do not test adding it to retatrutide. Results from CagriSema cannot be transferred to a different pair.
What side effects and safety findings were recorded?
Cagrilintide alone in phase 2
Gut effects and reactions at the administration site were common in the phase 2 trial[2]. Gut events affected 41-63% across doses of the drug versus 32% with placebo. Nausea affected 20-47% versus 18%. Constipation and diarrhoea were also named. Novo Nordisk paid for the study.
Across the whole trial, 73 people stopped treatment for good[2]. Thirty stopped because of an adverse event. Twenty-nine withdrew from the trial. These totals are not drug-only counts.
| Registered safety count | 0.3 mg | 0.6 mg | 1.2 mg | 2.4 mg | 4.5 mg | Liraglutide | Placebo |
|---|---|---|---|---|---|---|---|
| Participants at risk | 101 | 100 | 102 | 102 | 101 | 99 | 101 |
| With a serious adverse event | 6 | 2 | 7 | 3 | 4 | 4 | 3 |
| Deaths | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
Source: NCT03856047, registered adverse-event groups; Novo Nordisk; checked 2026-09-06[4]. An event in a treated group is not by itself proof of drug causation.
REDEFINE 1: four groups kept separate
The full paper reports both single-agent and pair safety[3]. Its safety population includes 2,106 people on CagriSema, rather than the 2,108 assigned at random. The other group sizes are unchanged. Novo Nordisk paid for the work.
| Participants with an event (%) | CagriSema | Semaglutide | Cagrilintide | Placebo |
|---|---|---|---|---|
| Any adverse event | 92.3 | 89.7 | 84.1 | 82.3 |
| Serious adverse event | 9.8 | 5.0 | 8.9 | 6.1 |
| Event leading to permanent treatment stop | 5.9 | 3.6 | 2.6 | 3.5 |
| Gut event leading to permanent treatment stop | 3.6 | 1.3 | 1.3 | 0.6 |
| Fatal event | 0.1 | 0 | 0 | 0 |
| Gut event | 79.6 | 73.8 | 54.0 | 39.9 |
| Injection-site reaction | 12.2 | 2.6 | 16.9 | 3.0 |
| Allergic reaction | 5.2 | 5.6 | 7.6 | 5.5 |
| Neoplasm | 6.4 | 6.6 | 1.7 | 4.4 |
| Gallbladder disorder | 4.1 | 3.0 | 2.3 | 1.0 |
| Confirmed malignant neoplasm | 0.7 | 0.7 | 0.7 | 0.6 |
| Pancreatitis | 0.2 | 0.3 | 0 | 0 |
Source: PMID 40544433, Table 3; checked 2026-09-06[3]. Treatment stops, gut events, site reactions, allergic reactions, gallbladder disorders and pancreatitis use the on-treatment period. Other rows use the in-trial period.
The two deaths in the combination group were judged as suicide and cancer of unknown primary source[3]. No deaths were recorded in the group taking the drug alone. These are recorded outcomes, not claims that the drug caused them.
Other combination studies
In the phase 1b trial, 92 of 95 exposed people reported adverse events[5]. Of 566 events, 207 concerned the gut. That 37% figure describes events, not people. Most events were mild or moderate. All groups also received semaglutide, and Novo Nordisk paid for the trial.
In the 32-week diabetes trial, events were reported by 21/31 on CagriSema, 22/31 on semaglutide and 24/30 on this drug[6]. Mild or moderate gut events were most common. No level 2 or 3 hypoglycaemia was reported. There were no fatal events. Novo Nordisk paid for this trial too.
REDEFINE 2 studied the pair in people with diabetes[7]. The following rates therefore do not describe the drug alone. Novo Nordisk paid for the study.
| Participants with an event (%) | CagriSema, n=904 | Placebo, n=302 |
|---|---|---|
| Any adverse event | 90.2 | 85.4 |
| Serious adverse event | 10.4 | 12.9 |
| Event leading to permanent treatment stop | 8.4 | 3.0 |
| Gut event leading to permanent treatment stop | 4.8 | 0.7 |
| Fatal event | 0.4 | 0 |
| Level 1 hypoglycaemia | 11.9 | 7.9 |
| Level 2 hypoglycaemia | 6.0 | 3.3 |
| Level 3 hypoglycaemia | 0.2 | 0 |
| Gut disorder | 72.5 | 34.4 |
| Retinal disorder | 8.3 | 7.9 |
| Neoplasm | 7.0 | 6.6 |
| Allergic reaction | 5.1 | 6.0 |
| Injection-site reaction | 4.9 | 0 |
| Gallbladder disorder | 2.0 | 0.7 |
| Confirmed malignant neoplasm | 1.5 | 1.3 |
| Pancreatitis | 0.3 | 0 |
Source: REDEFINE 2, Table 3; checked 2026-09-06[7]. Hypoglycaemia, gut, allergic, site, gallbladder and pancreatitis rows are on-treatment. The other rows are in-trial.
Both people with level 3 hypoglycaemia also took sulfonylureas[7]. The four deaths in the combination group were attributed to cancer, sudden cardiac death, suicide and another noncardiovascular cause. No placebo participant died. Suicidal ideation or behaviour was recorded in 8/878 assessed people on the pair and 4/296 assessed people on placebo. That assessment has different denominators from the main safety table.
The REDEFINE 4 sponsor report names gut effects as the most common events[13]. It says most were mild to moderate and declined over time. That headline report does not supply a complete event table. Its general safety wording is the sponsor's account, not a quantified single-agent finding.
What dosage information do sources report for Cagrilintide?
The sources report study doses, not one dose for each reader. Trials of the drug and the pair used different designs. Not everyone reached the target dose[3]. The table reports amounts, not steps for raising a dose.
| Source | Reported amount and setting | Source type and funding | Checked |
|---|---|---|---|
| NCT03856047, PMID 34798060 | Once-weekly subcutaneous cagrilintide targets of 0.3, 0.6, 1.2, 2.4 or 4.5 mg over 26 weeks | Randomised phase 2; Novo Nordisk[2] | 2026-09-06 |
| NCT03600480, PMID 33894838 | Weekly cagrilintide targets of 0.16, 0.30, 0.60, 1.2, 2.4 or 4.5 mg alongside weekly semaglutide 2.4 mg | Phase 1b; Novo Nordisk[5] | 2026-09-06 |
| NCT04982575, PMID 37364590 | Weekly subcutaneous targets of 2.4 mg per drug, given alone or together, over 32 weeks | Phase 2 diabetes trial; Novo Nordisk[6] | 2026-09-06 |
| REDEFINE 1 and 2 | Weekly subcutaneous targets of 2.4 mg cagrilintide and 2.4 mg semaglutide; REDEFINE 1 also had single-agent arms | Phase 3; Novo Nordisk[3] | 2026-09-06 |
| PMID 42228334 | Single cagrilintide doses of 0.6 mg in the renal-function study and 0.9 mg in the liver-function study | Pharmacokinetic studies, not a weight-loss dosing guide[8] | 2026-09-06 |
In the phase 1b study of both drugs, the drug's half-life ranged from 159 to 195 hours[5]. Median time to peak concentration was 24 to 72 hours. Exposure rose with the dose. It did not alter semaglutide exposure or loss from the body in that study. These measurements do not define a schedule for an untested pair.
The 2026 kidney and liver studies included 33 and 32 people[8]. Exposure showed no clear trend as organ function declined. They used single doses and small groups. Novo Nordisk funded both studies and the writing support.
| Study, relative to normal function | Mild impairment | Moderate | Severe |
|---|---|---|---|
| Renal, AUC ratio (90% CI) | 1.23 (0.91-1.66) | 1.18 (0.87-1.59) | 1.21 (0.87-1.68) |
| Hepatic, AUC ratio (90% CI) | 0.99 (0.89-1.11) | 1.01 (0.91-1.12) | 1.11 (0.96-1.30) |
Source: PMID 42228334; checked 2026-09-06[8]. The renal study recorded 21 events in 11 people, and the hepatic study 16 in nine. Neither recorded a serious event, event-related withdrawal or death. This is a description of those studies, not advice to adjust or retain a dose.
What do animal and laboratory studies add?
Kruse and colleagues reported the development of a stable, lipidated amylin analogue[1]. Their paper describes structure-activity work that led to this drug. That work concerns how the molecule was developed. It does not test the contents of a consumer vial.
The design abstract gives no separate animal weight-loss rate to report here[1]. Human weight-loss figures above come from human trials, not animal models. The abstract does not state who paid for this work.
What has direct vial or product testing found?
Mendias and Awan included this drug in a preprint analysing submitted peptide tests[10]. They report no external funding. The samples were self-selected, not a random market sample. The following figures pool compounds; none is a drug-only rate.
| Pooled finding | Result |
|---|---|
| Amount and purity analysis | 6,285 reports |
| Median amount relative to label | 101.80%; IQR 95.00-109.00% |
| Median purity | 99.80%; IQR 99.50-99.90% |
| Failed either model's criteria | 41.6% with the looser model; 71.1% with the stricter model |
| Measurable endotoxin | 36 of 243 tested reports; 0.5-40 EU/mL |
Source: DOI 10.20944/preprints202604.1748.v1; checked 2026-09-06[10]. The models require amount within 90-110% and purity at least 98.0%, or amount within 95-105% and purity at least 99.5%. These are study benchmarks, not formal release decisions. The dataset does not establish sterility, stability or every vial's contents.
What do people report about Cagrilintide?
No firsthand treatment account is quoted on this page. Trial questionnaires and trial adverse-event reports are not the same as freely shared forum accounts. No claim about what most users experience is made.
What is the current evidence status, and what is the regulatory status?
There are published human results for this drug alone and for CagriSema. The two remain separate. A late trial phase or published result does not establish approval. The EU central index has no entry for cagrilintide[11].
| Programme | Registry state | Sponsor | Checked |
|---|---|---|---|
| NCT03856047, single-agent phase 2 | Completed; 706 actual participants; results posted[4] | Novo Nordisk | 2026-09-06 |
| NCT05567796, REDEFINE 1 | Phase 3; active, not recruiting; registry estimate 3,400[12]. The paper reports 3,417 randomised[3]. | Novo Nordisk | 2026-09-06 |
| NCT07220642, cagrilintide weight-management study | Phase 3; active, not recruiting; estimated enrollment 300[12] | Novo Nordisk | 2026-09-06 |
| NCT07220759, cagrilintide with type 2 diabetes | Phase 3; active, not recruiting; estimated enrollment 330[12] | Novo Nordisk | 2026-09-06 |
The EU index covers central EU authorisations, not every national register. It does not settle UK approval. No UK product-authorisation conclusion is made from that search. A trial registration date is also not the date a product first appeared in the market.
See also
References
- ↑ a b c d Kruse T et al. Design of a long-acting amylin analogue. J Med Chem, 2021. PMID 34288673; DOI 10.1021/acs.jmedchem.1c00565. Used for the drug's design and identity; abstract checked 2026-09-06.
- ↑ a b c d e f g h i Lau DCW et al. Single-agent weight-loss trial. Lancet, 2021;398:2160-2172. PMID 34798060; DOI 10.1016/S0140-6736(21)01751-7. Results, doses, harms and Novo Nordisk funding; checked 2026-09-06.
- ↑ a b c d e f g h i j k l m n o p q r s t Garvey WT et al. REDEFINE 1. NEJM, 2025. PMID 40544433; DOI 10.1056/NEJMoa2502081; NCT05567796. Full paper: each drug and the pair, outcomes, harms, doses and funding; checked 2026-09-06.
- ↑ a b c d e Novo Nordisk. Phase 2 trial and posted results. ClinicalTrials.gov, NCT03856047. Group sizes, serious events and trial state; checked 2026-09-06.
- ↑ a b c d e Enebo LB et al. Phase 1b study of the two drugs. Lancet, 2021;397:1736-1748. PMID 33894838; DOI 10.1016/S0140-6736(21)00845-X; NCT03600480. Results, doses, blood levels and Novo Nordisk funding; abstract checked 2026-09-06.
- ↑ a b c d e f g Frias JP et al. Three-arm trial in type 2 diabetes. Lancet, 2023;402:720-730. PMID 37364590; NCT04982575. Glucose, weight, safety and funding; abstract checked 2026-09-06.
- ↑ a b c d e f Davies MJ et al. REDEFINE 2. NEJM, 2025. DOI 10.1056/NEJMoa2502082; NCT05394519. Full paper: the pair's effects, harms, dose and funding; checked 2026-09-06.
- ↑ a b c Lauenborg BW et al. Kidney and liver function studies. Clin Pharmacokinet, 2026. PMID 42228334; NCT04209049 and NCT05564104. Single-dose design, exposure, harms and Novo Nordisk funding; full-text funding statement checked 2026-09-06.
- ↑ a b c FDA/NCATS. Cagrilintide substance record. GSRS, UNII AO43BIF1U8. Names, codes and peptide scaffold checked 2026-09-06.
- ↑ a b Mendias CL, Awan TM. Peptide identity, amount and purity tests. Preprint, 2026; DOI 10.20944/preprints202604.1748.v1. Pooled test findings and limits, not rates for this drug alone; checked 2026-09-06.
- ↑ a b European Commission. Active-substance index. Union Register. Central EU search for this drug; checked 2026-09-06.
- ↑ a b c d ClinicalTrials.gov. Cagrilintide trial records. Exact-name query. Includes NCT05567796, NCT07220642 and NCT07220759. Current trial state and planned group size; checked 2026-09-06.
- ↑ a b c Novo Nordisk. REDEFINE 4 headline results. Sponsor report, 23 February 2026; NCT06131437. Design, weight-loss estimates and reported harms; checked 2026-09-06.