CagriSema Lost to Tirzepatide by 2.5 Percentage Points at 84 Weeks. The Amylin Bet Didn't Pay Off.

In early August 2026, Novo Nordisk reported detailed results from REDEFINE 4, an 84-week head-to-head Phase 3 trial that compared CagriSema against Eli Lilly's tirzepatide 15 mg. Both drugs are once-weekly subcutaneous injections. Both target the two most common comorbidities associated with obesity. Both had shown strong weight-loss results in earlier trials.

The trial enrolled 809 adults with obesity and one or more comorbidities. Mean baseline body weight was 114.2 kg. The primary endpoint was non-inferior weight loss for CagriSema versus tirzepatide at 84 weeks.

The results: CagriSema (cagrilintide 2.4 mg plus semaglutide 2.4 mg fixed-dose combination) reached 23.0% mean weight loss. Tirzepatide 15 mg reached 25.5%. Under a stricter analysis (the treatment-regimen estimand, which accounts for people who stopped therapy or added other interventions), the numbers were 20.2% for CagriSema and 23.6% for tirzepatide. Either way, CagriSema missed non-inferiority.

This is a substantive moment in obesity pharmacology. CagriSema was Novo Nordisk's answer to tirzepatide, the drug that had opened up a 5-plus-percentage-point efficacy gap over semaglutide alone. Novo bet that adding cagrilintide, an amylin analog, to semaglutide would close or overtake that gap. It didn't. This piece walks through why, and what the miss teaches about how obesity drugs get built.

REDEFINE 4 was a randomized, open-label, active-comparator Phase 3 trial. Open-label means both patients and investigators knew which drug each participant was taking; the comparator was an approved drug (tirzepatide), and blinding an injectable comparison of two different pen devices would have been operationally difficult. Active-comparator means the trial did not include a placebo arm; the question was not whether CagriSema works, but whether it works as well as tirzepatide.

Enrollment. 809 adults with obesity and at least one obesity-related comorbidity (type 2 diabetes, cardiovascular disease, sleep apnea, or others). Mean baseline body weight was 114.2 kg.

Duration. 84 weeks, a longer window than most obesity trials. The extra weeks matter: weight-loss curves typically plateau in the second half of a trial as physiologic counter-regulation and diminishing dose response take hold. 84 weeks captures more of that plateau than the 68-week SURMOUNT-1 tirzepatide trial or the 68-week STEP 1 semaglutide trial.

Dosing. CagriSema at the 2.4 mg / 2.4 mg fixed-dose combination (2.4 mg of cagrilintide plus 2.4 mg of semaglutide, both administered together in one weekly injection). Tirzepatide at 15 mg weekly, the highest approved dose.

Primary endpoint. Non-inferior weight loss for CagriSema versus tirzepatide at Week 84. Non-inferiority trials use a pre-specified margin: the trial declares success if the CagriSema result is within a certain distance of the tirzepatide result. The exact non-inferiority margin for REDEFINE 4 has not been fully disclosed, but the observed 2.5 percentage point gap (23.0% versus 25.5%) exceeded it.

Two estimands. Regulatory obesity trials now typically report two analyses. The treatment-policy estimand assumes all randomized patients contribute to the analysis regardless of what happens after randomization (people who stop early are still counted). The treatment-regimen estimand focuses on people who stayed on assigned therapy without adding weight-loss interventions. The two views produce different numbers because obesity trials have real-world dropout, dose modification, and rescue-therapy use. REDEFINE 4 reported both.

Cagrilintide is a long-acting analog of amylin, a hormone that beta cells in the pancreas release alongside insulin after a meal. Native amylin has a very short half-life (roughly 10 minutes) and cannot be given as a once-weekly drug. Cagrilintide is engineered to survive in circulation for the six-plus days needed for weekly dosing.

Amylin's biology is complementary to GLP-1's biology, which is why the combination made sense on paper.

Where amylin acts. Amylin binds a heterodimeric receptor complex made of the calcitonin receptor plus one of three receptor activity-modifying proteins (RAMPs). The main site of action for appetite is in the brainstem, specifically the area postrema and nucleus tractus solitarius (NTS). These brainstem regions process peripheral satiety signals (including gastrointestinal stretch and glucose) and coordinate meal termination. Activating amylin receptors there tells the brain to stop eating.

How that differs from GLP-1. GLP-1 receptor agonists like semaglutide act primarily in the hypothalamus, a different appetite-regulating brain region, and also in the gastrointestinal tract to slow gastric emptying. The hypothalamic and brainstem circuits are separate pathways that both converge on food intake. Activating both at once, the amylin plus GLP-1 hypothesis went, should produce more appetite suppression than activating either alone.

Gastric emptying. Amylin also slows gastric emptying, but through brainstem-mediated pathways rather than through the direct smooth-muscle mechanism that GLP-1 uses. In principle, adding amylin's gastric-emptying effect to GLP-1's should produce more pronounced fullness after meals.

Preserving lean mass. Some preclinical work has suggested amylin agonists may reduce fat mass proportionally more than lean mass compared to GLP-1 monotherapy. If confirmed in humans, that would matter for functional outcomes: patients on obesity drugs are generally trying to lose fat mass, not muscle mass.

Each of these was a real mechanistic argument. What REDEFINE 4 shows is that adding cagrilintide to semaglutide does produce more weight loss than semaglutide alone (REDEFINE 1 showed 20.4% versus 3.0% for placebo at 68 weeks, better than semaglutide monotherapy's roughly 15% at that duration) but does not close the gap with tirzepatide.

Tirzepatide takes a different approach. It is one peptide that activates both the GIP receptor and the GLP-1 receptor. Glucose-dependent insulinotropic polypeptide (GIP) is another incretin hormone, released by K cells in the upper small intestine in response to nutrient ingestion. Its main historical role in metabolic physiology was thought to be stimulating insulin release after meals.

What GIP adds to a GLP-1 backbone, based on what tirzepatide has revealed:

Central appetite effects. GIP receptors are expressed in the hypothalamus and hindbrain and, when activated, contribute to appetite suppression through pathways that partly overlap with GLP-1 and partly do not. The overlap plus the divergence appears to produce more overall appetite suppression than GLP-1 alone.

Adipose tissue effects. GIP receptors are expressed on fat cells (adipocytes). GIP signaling in adipose tissue influences lipid metabolism, thermogenesis, and possibly adipose tissue expandability. This is a substantially different effect from anything amylin does, and it may explain part of tirzepatide's fat-mass reduction efficacy.

Nausea attenuation. A counterintuitive tirzepatide finding: GIP receptor activation may partially blunt GLP-1-induced nausea, even while contributing to appetite suppression. If confirmed, this would mean tirzepatide can push a higher net dose of appetite suppression than semaglutide alone for the same tolerability.

Bone and other tissue effects. GIP receptors also appear on osteocytes and bone marrow. Whether GIP agonism benefits bone metabolism during rapid weight loss is an open question. Semaglutide and tirzepatide have both been associated with mild bone density concerns during large weight losses, and the GIP arm may or may not partially offset that.

What matters for the head-to-head comparison: GIP plus GLP-1 in one molecule produced 25.5% weight loss at 84 weeks in REDEFINE 4. Amylin plus GLP-1 in a fixed-dose combination produced 23.0%. The 2.5 percentage point gap suggests, but does not prove, that GIP's adipose tissue and central effects offer something the amylin brainstem pathway does not fully replicate.

The REDEFINE 4 result is one trial. Reading too much into any one head-to-head can mislead. But several structural features of the comparison make the outcome interesting.

Fixed-dose combination versus unimolecular dual agonist. CagriSema is two separate active ingredients formulated together. Tirzepatide is one molecule that hits two receptors. The unimolecular design ensures both signals arrive at the same time to the same target tissues at the same relative concentrations, whereas the fixed-dose combination approach can produce timing and distribution differences between the two components. Whether that engineering difference contributes to the observed weight-loss gap is unclear, but it is a real structural difference between the two approaches.

Adipose tissue reach. Amylin's main site of appetite action is the brainstem. GIP hits adipose tissue directly, plus the hypothalamus. If some of the additional weight loss on tirzepatide reflects direct adipose effects (lipolysis, fat mass reduction, thermogenesis), amylin does not have a mechanistic counterpart to that effect. The mechanistic circle amylin plus GLP-1 covers is real but may be narrower than GIP plus GLP-1.

Dose ceiling and tolerability. Semaglutide 2.4 mg is at or near its tolerability ceiling for the obesity indication. Cagrilintide 2.4 mg was chosen based on Phase 2 dose-ranging. If either component has more efficacy at higher doses but did not get pushed to those doses in REDEFINE 4 because of nausea, that would explain part of the gap. Novo Nordisk has disclosed plans to run a higher-dose CagriSema Phase 3 in the second half of 2026, which suggests the company sees room to push doses higher.

Duration effects. At 84 weeks, both drugs' weight-loss curves are into the plateau zone. If cagrilintide's contribution manifests more strongly early in therapy (rapid meal-termination effects) and tirzepatide's builds over time (metabolic effects on adipose tissue), the 84-week readout might be less flattering to CagriSema than a shorter-duration comparison would have been. This is speculative but plausible from the biology.

The practical read: amylin biology is real and cagrilintide does contribute to appetite suppression and weight loss on top of semaglutide. But GIP biology is broader, and tirzepatide's unimolecular design lets that broader biology reach more tissues. In this trial, at these doses, over this duration, the broader mechanism won.

The REDEFINE 4 miss reshapes how to read the broader combination-obesity-drug landscape.

Retatrutide (Eli Lilly). Retatrutide is a triple agonist: GLP-1, GIP, and glucagon. It has produced roughly 28.7% weight loss at 68 weeks in the Phase 3 TRIUMPH-4 trial at the 12 mg dose, the highest weight-loss magnitude published in any obesity Phase 3 to date. If GIP plus GLP-1 (tirzepatide) beat amylin plus GLP-1 (CagriSema), and glucagon plus GIP plus GLP-1 (retatrutide) beat both, the pattern suggests each additional receptor system reaches tissues the previous combinations did not. Glucagon receptor agonism adds hepatic effects and thermogenesis; combined with GIP's adipose effects, the tissue coverage of retatrutide's three-receptor design is broader than either CagriSema or tirzepatide.

Survodutide (Boehringer/Zealand). Survodutide is a GLP-1 plus glucagon dual agonist that has produced roughly 19% weight loss at 46 weeks in Phase 2. The Phase 3 program is ongoing. Survodutide is a data point on the glucagon-add-on approach without GIP. Its results will help separate what glucagon adds from what GIP adds.

Mazdutide (Innovent). A GLP-1 plus glucagon dual agonist tested primarily in Chinese patients. Recent trials have shown roughly 15-17% weight loss at Week 48, comparable to semaglutide-monotherapy magnitudes but at substantial cost advantages. Mazdutide's data reads as competitive with semaglutide, though not with tirzepatide.

Next-generation amylin. The REDEFINE 4 miss does not close the amylin story. Novo Nordisk has an amylin monotherapy program (amycretin, an oral formulation) that has shown roughly 22% weight loss in Phase 1b at Week 36. If amylin monotherapy can deliver in that range, an oral formulation might carve out a segment where injection burden matters. Cagrilintide plus higher-dose semaglutide (planned Phase 3 in H2 2026) may still produce competitive results.

The overall trajectory. Each successive class combination (semaglutide alone at 15%, semaglutide plus cagrilintide at 20-23%, tirzepatide at 25.5%, retatrutide at 28.7%) has extended the ceiling by 2 to 5 percentage points. The ceiling may not extend indefinitely: at some point weight loss is limited by lean-mass preservation, cardiovascular tolerability of rapid weight loss, and patient adherence rather than by receptor coverage. Retatrutide may be at or near that ceiling. That would leave differentiation to move from raw efficacy toward tolerability, delivery format (oral, injectable, implant), duration of action, and price.

CagriSema was submitted to the FDA in December 2025 based on the REDEFINE 1 and REDEFINE 2 registrational trials. An FDA decision is expected late 2026. The REDEFINE 4 head-to-head miss does not directly affect the FDA decision, since REDEFINE 4 is a head-to-head comparison and the label submission is based on placebo-controlled data.

What the miss does affect:

Commercial positioning. Novo Nordisk had positioned CagriSema as its answer to tirzepatide. The 25.5% versus 23.0% comparison at 84 weeks makes that positioning harder to defend. Sales teams will have to lead with different messaging (perhaps tolerability profile, dosing convenience, or price) rather than efficacy superiority.

Payer conversations. In payer contracting discussions, comparative efficacy data feeds into rebate and formulary tier decisions. A CagriSema that clearly outperformed tirzepatide would command a stronger commercial position. A CagriSema that lags tirzepatide will need to compete on price or on patient-selection criteria.

Pipeline reprioritization. Novo Nordisk is running a higher-dose CagriSema Phase 3 in H2 2026, plus continuing work on amycretin (oral amylin), amylin plus retatrutide-class candidates, and the Wegovy 7.2 mg higher-dose semaglutide submission. The head-to-head miss may accelerate reprioritization toward next-generation candidates rather than incremental CagriSema iterations, though Novo has not signaled that publicly.

Novo Nordisk share price. Shares fell roughly 15% on the initial CagriSema head-to-head report in February 2026. The August 2026 detailed data confirmed rather than changed the picture, and shares have absorbed the news across the year. The bigger stock-market question is whether Novo can defend its GLP-1 franchise economics against tirzepatide's momentum and Lilly's expanding oral obesity portfolio (Foundayo, orforglipron).

For patients evaluating obesity drug choices, the REDEFINE 4 comparison offers useful information but should not be the sole decision factor.

Head-to-head data is the strongest form of comparison. A trial that randomizes patients between two drugs and follows them for the same duration is more informative than comparing two separate placebo-controlled trials done at different times with different patient populations. REDEFINE 4 is that direct comparison. On raw weight-loss efficacy at 84 weeks, tirzepatide has the edge.

Individual response varies substantially. Group averages of 23% versus 25.5% mask a wide distribution. Some patients on CagriSema will lose more than 30%; some patients on tirzepatide will lose less than 15%. Individual response reflects adherence, tolerability, dose titration success, and biological factors that are not fully understood.

Tolerability may matter more than headline efficacy for some patients. CagriSema's safety profile in REDEFINE 4 was described as generally well-tolerated. Whether specific nausea, vomiting, or other adverse event rates differ meaningfully between CagriSema and tirzepatide at these doses would be worth watching in the full trial publication. For patients who cannot tolerate tirzepatide's higher doses, CagriSema at these approved doses might deliver more usable weight loss.

Access and cost differ. Tirzepatide has been on the market for obesity since 2023 (Zepbound) and diabetes since 2022 (Mounjaro), with prescription supply and insurance coverage patterns established. CagriSema is not yet approved. When it launches (expected late 2026 to early 2027), initial insurance coverage will be limited and price will be at branded GLP-1 rates.

Prescribing decisions belong with a clinician. For patients considering obesity drug therapy, the comparison of drugs is one input among many. Comorbidities, medication interactions, insurance coverage, patient preference on injection frequency and device design, and clinical monitoring capacity all matter. A prescriber can weigh those together in a way that a summary of head-to-head efficacy cannot.

REDEFINE 4 showed that CagriSema, Novo Nordisk's fixed-dose combination of the amylin analog cagrilintide and semaglutide, produced 23.0% mean weight loss at 84 weeks against tirzepatide's 25.5%. CagriSema missed the non-inferiority primary endpoint. Under the stricter treatment-regimen analysis, the gap held at 20.2% versus 23.6%.

The result validates that amylin biology adds real appetite suppression on top of GLP-1: CagriSema clearly outperforms semaglutide monotherapy. But GIP biology (via tirzepatide's unimolecular dual-agonist design) reaches tissues that amylin does not, and in this trial that broader tissue coverage translated into a modest but consistent efficacy edge.

The bigger takeaway is about how obesity drug development is now organized. Each successive combination or unimolecular multi-agonist has extended the weight-loss ceiling by 2 to 5 percentage points, with retatrutide's three-receptor design at roughly 28.7% currently sitting near the top. Whether the ceiling can extend much further before other constraints (lean mass preservation, cardiovascular tolerability of rapid weight loss, patient adherence) become binding is an open question.

For Novo Nordisk, the near-term commercial task is to defend CagriSema's launch expectations against a drug that beat it head to head, then work through the higher-dose CagriSema Phase 3 and the amycretin oral-amylin program while Lilly's tirzepatide franchise and retatrutide pipeline continue to widen the competitive gap. For patients, the choice among these drugs remains individual: head-to-head averages inform expectations but do not determine outcomes.