CagriSema REDEFINE 1 results: the two headline numbers explained

What CagriSema is

CagriSema is an investigational once-weekly combination of two drugs. Cagrilintide is a long-acting amylin analog. Semaglutide is the GLP-1 medicine already sold as Wegovy and Ozempic. The combination tests whether two receptor systems involved in appetite and glucose regulation can produce more weight loss than either component alone.

REDEFINE 1 studied adults with obesity or overweight without diabetes, while REDEFINE 2 enrolled adults with type 2 diabetes. Both phase 3 trials were published in the New England Journal of Medicine in 2025. The results are easy to misread because Novo Nordisk's first release and the peer-reviewed paper emphasized different estimands, or different ways of handling treatment discontinuation and missing data.

What REDEFINE 1 showed

REDEFINE 1 randomized 3,417 adults for 68 weeks to CagriSema, cagrilintide alone, semaglutide alone, or placebo. Everyone also received lifestyle counseling. The combination target was 2.4 mg of cagrilintide plus 2.4 mg of semaglutide, reached through gradual dose escalation.

Under the treatment-policy estimand, which followed the randomized groups regardless of whether participants stopped treatment or used another weight-management intervention, mean weight change was 20.4% with CagriSema and 3.0% with placebo. The same paper also showed that the combination produced more responders at the 5%, 20%, 25%, and 30% thresholds than placebo.

Dose exposure matters when reading the result. Novo Nordisk's trial presentation reported that 57.4% of participants assigned to CagriSema were receiving the highest dose at week 68. That doesn't invalidate the average. It shows why a flexible, tolerability-driven escalation plan shouldn't be described as though every participant took 2.4 mg plus 2.4 mg for the full trial.

Gastrointestinal events were common: 79.6% of the CagriSema group reported events such as nausea, vomiting, diarrhea, constipation, or abdominal pain, compared with 39.9% on placebo. Most were described as transient and mild to moderate, but the combination didn't remove the tolerability burden associated with these pathways.

Why the press release moved the stock

Novo Nordisk announced a 22.7% mean reduction in December 2024 using the trial-product estimand, which estimates the effect if participants stayed on treatment. The peer-reviewed REDEFINE 1 paper reported 20.4% under the treatment-policy estimand, which keeps participants in their randomized groups after discontinuation or rescue intervention. Both numbers come from the same trial, but they answer different questions.

The market had expected a result closer to 25%, so the 22.7% release triggered a sharp selloff. The useful lesson isn't that one number was false. Every obesity-trial headline needs four labels beside it: population, duration, dose exposure, and estimand. Without those, a cross-trial ranking can compare adherence under one method with real-world discontinuation under another.

How the amylin arm works

Cagrilintide activates amylin receptors, while semaglutide activates GLP-1 receptors. Amylin receptors are complexes built from the calcitonin receptor and receptor-activity-modifying proteins. Relevant signaling occurs in brain regions involved in meal termination, including the area postrema and related hindbrain circuits.

The two pathways are different at the receptor level, but their downstream effects overlap. Both can reduce food intake, slow gastric emptying, and influence post-meal glucagon. The rationale for combining them is therefore not that each effect is separate. It is that two receptor systems may reinforce appetite control strongly enough to add efficacy, while also creating an overlapping gastrointestinal side-effect burden.

The press release number versus the paper number, decoded

The treatment-policy estimand asks what happened across everyone randomized, whether or not they stayed on treatment. In REDEFINE 1, that estimate was 20.4%. The trial-product estimand asks what the outcome would look like if participants stayed on treatment, and that estimate was 22.7%.

Neither number should be silently swapped for the other. When comparing CagriSema with semaglutide, tirzepatide, or another obesity drug, use results calculated under similar assumptions and treat cross-trial rankings as approximate.

How to compare the result without overreading it

A trial average is not a personal forecast. REDEFINE 1 included a broad group, but eligibility criteria still selected participants who could enter a long obesity trial and follow its schedule. Someone's response can land well above or below the mean because weight history, side effects, treatment persistence, other medicines, and biology all shape the result.

Responder thresholds add useful context. The paper reported more participants reaching losses of at least 5%, 20%, 25%, and 30% with CagriSema than with placebo. Those thresholds show that the mean wasn't created by a tiny group of extreme responders. They still don't tell a new patient which threshold they would reach, and they don't replace the full distribution of outcomes.

Safety numbers need the same discipline. Saying that most gastrointestinal events were mild or moderate describes how investigators graded them. It doesn't mean they were trivial for every participant, and it doesn't erase discontinuation or the fact that fewer people reached the highest combination dose than a headline might imply.

Finally, trial duration matters. REDEFINE 1 ran for 68 weeks, while the direct REDEFINE 4 comparison ran for 84 weeks. A percentage from one time point shouldn't be ranked against another without the date attached. The strongest comparison is the randomized head-to-head result, not a leaderboard assembled from separate trials.

The control arms also matter. CagriSema was tested beside each component alone, not only against placebo, so the trial could ask whether the combination added value beyond semaglutide or cagrilintide by itself. That is stronger evidence for the combination than a single-arm weight-loss report. It still doesn't settle how CagriSema compares with every approved or investigational obesity medicine, which is why REDEFINE 4 carries so much weight.

How CagriSema stacks up against the rest of the field

Every next-gen obesity drug right now follows the same basic formula: start with a GLP-1 backbone, then bolt on a second hormonal arm that pulls a different lever. Tirzepatide adds GIP, an incretin hormone that improves how fat tissue stores fat and sharpens insulin sensitivity. Survodutide adds glucagon receptor agonism, which raises energy expenditure and burns liver fat, the mechanism behind its strong showing in liver disease trials. Retatrutide stacks both GIP and glucagon on top of GLP-1, making it a triple agonist. CagriSema adds amylin, working through central satiety, slower gastric emptying, and glucagon suppression.

Published and announced results span different populations, durations, and statistical methods. STEP 1 reported about 14.9% for semaglutide at 68 weeks. SURMOUNT 1 reported up to about 20.9% under its treatment-regimen estimand and 22.5% under an efficacy estimand for tirzepatide at 72 weeks. REDEFINE 1 reported 20.4% and 22.7% for CagriSema under treatment-policy and trial-product estimands, respectively. Those pairs show why a single headline number can mislead.

REDEFINE 4 supplied the direct comparison that REDEFINE 1 could not. In February 2026, Novo Nordisk reported 23.0% mean weight loss with CagriSema and 25.5% with tirzepatide under an efficacy estimand. Under the treatment-regimen estimand, the figures were 20.2% and 23.6%. The trial did not establish CagriSema as noninferior to tirzepatide. These are company-reported topline results, so the complete paper still matters.

What CagriSema still doesn't have

CagriSema has phase 3 obesity data and a direct active-comparator trial, but several long-term questions remain open. REDEFINE 3 is designed to evaluate cardiovascular outcomes and is ongoing, so there is no completed CagriSema cardiovascular outcomes result yet. There is also no completed kidney outcomes program or long-duration evidence showing what happens over several years.

Compare that to what semaglutide already has locked down. The SELECT trial showed a 20% reduction in major adverse cardiovascular events in adults with obesity and established cardiovascular disease, without diabetes. Semaglutide also carries a MASH approval. Tirzepatide has approvals spanning obesity, type 2 diabetes, obstructive sleep apnea, and MASH. CagriSema has none of that yet. It's a reasonable bet that adding cagrilintide won't cancel out semaglutide's cardiovascular benefit, but that stays an assumption, not a fact, until a dedicated CV outcomes trial reads out.

The realistic read

CagriSema produced clinically substantial weight loss in a large phase 3 trial, and the 20.4% treatment-policy result is the number that best reflects the randomized comparison published in the paper. The 22.7% trial-product estimate is useful too, as long as it is labeled as an on-treatment scenario rather than presented as the only result.

The direct REDEFINE 4 topline comparison was less favorable: CagriSema did not establish noninferiority to tirzepatide and had lower mean weight loss under both reported estimands. CagriSema belongs in the leading group of investigational obesity combinations, but current evidence doesn't support calling it superior or equivalent to tirzepatide. Approval, long-term outcomes, access, and performance outside trials remain open questions.

Educational content only. Not medical advice.

Continue with source-linked research guides