If your head has spent twenty years talking about food, a brain scan showing that noise easing is not a small thing. It is a picture of something you were told to fix with discipline.
Here is the unpopular part, early. The scan is the least consequential number Novo Nordisk released about CagriSema this year, and the company’s own head-to-head trial is the one patients should read twice.
CagriSema is a once-weekly fixed-dose combination of cagrilintide, a long-acting amylin analogue, and semaglutide, the drug in Ozempic and Wegovy. It is not approved for use in the United States. Novo Nordisk submitted a New Drug Application to the FDA on December 18, 2025, and in its February 23, 2026 company announcement the firm said an FDA decision is anticipated by late 2026. The clearest evidence on CagriSema food noise comes from a 52-week functional MRI study of 113 adults, presented at a conference in September 2026 and not yet peer reviewed.
The CagriSema Food Noise Study: 113 Adults, 52 Weeks, One Scanner
The study is registered as NCT06267092. Novo enrolled 130 adults with a BMI of 27 or higher, no type 2 diabetes, and low craving control, defined as a disinhibition score of 8 or more out of 16 on the Three-Factor Eating Questionnaire. Participants were randomized to weekly CagriSema or placebo for 52 weeks after a two-week baseline period. Novo presented the design and results at the European Association for the Study of Diabetes annual meeting in Milan on September 30, 2026, and the abstract won the meeting’s Best Abstract Award.
The measurements were appetite after a standardized meal on a 100 mm visual analogue scale, food eaten at an unlimited lunch, dinner and snack box, three eating-behavior questionnaires, body weight, and blood-oxygen-level-dependent fMRI one hour after a standardized breakfast or lunch.
The appetite result, as reported in the EASD abstract, is an estimated treatment difference of 16.9 mm lower on that appetite scale for CagriSema than placebo at week 22 (95% CI -22.9 to -10.9), and 19.6 mm lower at week 52 (95% CI -25.6 to -13.5). Food eaten during the test meals was 4,212 kJ lower at week 22 and 3,392 kJ lower at week 52 than placebo. Those figures describe a difference between groups, not a promise that any individual eats 4,212 kJ less.
Of the 130 enrolled, 113 completed 22 weeks of treatment, 54 on CagriSema and 59 on placebo, and 105 completed 52 weeks, 49 on CagriSema and 56 on placebo. The body weight result that traveled furthest in the press coverage, 22.4% at week 52, is also an estimated treatment difference between groups, not the average weight lost by people taking the drug.
What the Scans Showed: Thalamus, Putamen, and a Word Novo Chose Carefully
At 22 weeks, compared with placebo, the CagriSema group showed a lower blood-oxygen response to images of high-calorie foods in the postcentral gyrus, and a higher response in several thalamic nuclei and the putamen. At 52 weeks the lower response appeared in the supplementary motor area, with higher responses in right frontal and parietal regions, the thalamus and cerebellar regions. The whole-brain analyses used 65 people on CagriSema and 62 on placebo.
That is a map of where a drug changes blood flow in response to pictures of food. It is not a recording of what anyone does at eleven at night. fMRI measures oxygenation as a proxy for activity, and the abstract describes regions it labels as related to satiety, hunger and reward. Novo’s own materials say the pattern suggests treatment related modulation of food cue reactivity. Suggest is the word the sponsor used, and it is the right word.
Two things are worth saying plainly. The study was funded and run by the company that owns the drug, and a Best Abstract Award is a conference honor, not peer review. The same data were presented at the American Diabetes Association meeting in June 2026 as abstract 1035-OR, with 22-week results only, and published in the journal Diabetes as a supplement abstract on June 5, 2026.
A scan can show where a drug lands. It cannot show what that is worth.
REDEFINE 4: CagriSema Lost the Head-to-Head and Novo Printed It
On February 23, 2026, Novo Nordisk published company announcement 13/2026 with the headline results of REDEFINE 4, an open-label trial of 809 randomized people with obesity and at least one comorbidity, mean starting weight 114.2 kg, comparing CagriSema with tirzepatide 15 mg over 84 weeks.
Under the analysis that counts the effect if all participants stayed on treatment, CagriSema produced 23.0% weight loss against 25.5% for tirzepatide. Counting the effect regardless of adherence, the figures were 20.2% and 23.6%. The primary endpoint, showing that CagriSema was not inferior to tirzepatide on weight loss, was not achieved.
Novo’s chief scientific officer, Martin Holst Lange, called the company pleased with the 23% result and described cagrilintide as adding to semaglutide. Both things are in the same document: a drug that helps people substantially, and a company statement that its combination did not match tirzepatide in a trial it ran itself.
Publishing the trial you lost is worth more than any conference award.
The Pivotal Numbers, Because the Poster Is Not the Product
The FDA application rests on REDEFINE 1 and REDEFINE 2, both published in the New England Journal of Medicine on June 22, 2025. They are the numbers a patient is most likely to be handed in a clinic brochure, so they are worth reading in the form the journal printed them.
REDEFINE 1, by Garvey and colleagues, randomized 3,417 adults with obesity or overweight and no diabetes across four arms for 68 weeks. Weight loss was 20.4% on CagriSema against 3.0% on placebo when the analysis counted all participants who were randomized, and 22.7% against 2.3% when it counted the effect if all participants stayed on treatment. Gastrointestinal adverse events were reported by 79.6% of the CagriSema group and 39.9% of the placebo group. Permanent discontinuation because of adverse events was 5.9% against 3.5%.
REDEFINE 2, by Davies and colleagues, randomized 1,206 adults with type 2 diabetes for 68 weeks. Weight loss was 13.7% against 3.4% under the analysis that counts all participants randomized. Gastrointestinal adverse events were reported by 72.5% of the CagriSema group and 34.4% of the placebo group.
Those two ways of counting are why the same drug appears as 20.4% in one sentence and 22.7% in the next. Neither is dishonest. Anyone quoting only the larger one is selling, and you can compare that pattern against the way tirzepatide side effect numbers are usually presented.
The trial dose was 2.4 mg of each drug once weekly, reached by gradual increase, and dose adjustment was allowed in the trials. The dose that suits any individual is a prescriber’s decision, and nothing here is a dosing instruction.
CagriSema Food Noise at the Clinic: What You Cannot Buy Yet
Novo’s release of October 20, 2026 carries a single line at the bottom: CagriSema is not approved for use in the United States.
The compounding question is answered on FDA’s own page on unapproved GLP-1 drugs. The agency writes that “Retatrutide and cagrilintide cannot be used in compounding under federal law,” adding that “these are not components of FDA-approved drugs and have not been found safe and effective for any condition.” FDA lists the page as current as of September 25, 2025.
That matters if you have seen a vial labeled cagrilintide or a product called CagriSema for sale. No trial has scanned the brains of people using any such product, and no human trial has tested those vials at all. The 113 adults in the food noise study received a Novo-manufactured fixed-dose pen, not a compounded vial, and the 3,417 adults in REDEFINE 1 did the same.
The semaglutide half of the combination is a different conversation, with its own approved products and its own long list of effects, which we cover in semaglutide side effects.
If you are weighing a clinic for any GLP-1 or peptide treatment, the red flags we check before paying are a faster read than any brochure, and the clinic directory lists practices by city. We do not verify or endorse any clinic’s license, and a reader should confirm each license with the state board. The compounds library explains what each peptide is and which ones have human data.
The useful picture is narrower than the headlines. The brain data is early and company-run, the head-to-head result is published and unfavorable, and the product is not for sale in the United States.
Frequently asked questions
Is CagriSema FDA approved?
No. Novo Nordisk submitted a New Drug Application to the FDA on December 18, 2025, and said in its February 23, 2026 announcement that a decision is anticipated by late 2026. The company’s October 20, 2026 release states that CagriSema is not approved for use in the United States.
Does CagriSema stop food noise?
It reduced appetite and food-related thoughts compared with placebo in a 52-week study of 113 adults who completed 22 weeks of treatment. Appetite scores were 16.9 mm lower than placebo at week 22 and 19.6 mm lower at week 52 on a 100 mm scale. The study was not peer reviewed at the time of writing, and reduced appetite is not the same as silence.
How much weight do people lose on CagriSema?
In REDEFINE 1, adults without diabetes lost 20.4% of body weight against 3.0% on placebo when all participants randomized were counted, and 22.7% against 2.3% when the analysis counted only the effect if all participants stayed on treatment, over 68 weeks. In REDEFINE 2, adults with type 2 diabetes lost 13.7% against 3.4%. Individual results vary.
Is CagriSema better than Zepbound or Mounjaro?
Not on weight loss in the trial that tested it. In REDEFINE 4, an open-label 84-week trial of 809 people, CagriSema produced 23.0% weight loss against 25.5% for tirzepatide 15 mg when counting the effect if all participants stayed on treatment, and 20.2% against 23.6% when counting all participants randomized. The trial did not meet its non-inferiority endpoint.
Can I get compounded CagriSema or cagrilintide?
Not lawfully. FDA states that cagrilintide cannot be used in compounding under federal law and is not a component of an FDA-approved drug. Products sold under those names sit outside both the trials described here and federal compounding rules, and no human trial has tested them.
When will CagriSema be approved in the US?
Novo says a decision is expected in the fourth quarter of 2026, and that guidance is the company’s, not the FDA’s. The FDA does not publish a decision date for a pending application, so no date can be confirmed until the agency acts. Novo has also said the timing may change.
This article is for information only and is not medical advice. It reports what trial publications, company announcements and regulators state, and it does not recommend any product, dose or clinic. Decisions about treatment belong with a licensed prescriber who knows your history.
Sources: Novo Nordisk, September 30, 2026; Uhre and colleagues, EASD 2026 abstract; Diabetes (ADA), abstract 1035-OR, June 5, 2026; Garvey and colleagues, New England Journal of Medicine, June 22, 2025; Davies and colleagues, New England Journal of Medicine, June 22, 2025; Novo Nordisk company announcement 13/2026, February 23, 2026; Novo Nordisk US, December 18, 2025; Novo Nordisk, October 20, 2026; FDA, page current as of September 25, 2025; ClinicalTrials.gov NCT06267092.