MariTide is Amgen's investigational name for maridebart cafraglutide, formerly AMG 133. It is a peptide-antibody conjugate: a fully human monoclonal antibody that blocks the GIP receptor, chemically attached to two GLP-1 analogue agonist peptides. The intended cadence is once-monthly (or less frequent) subcutaneous dosing. The long circulating time comes from the antibody backbone, not from a weekly peptide side chain.
That mechanism is the opposite of dual GLP-1/GIP agonists such as tirzepatide and Viking's investigational VK2735. Those drugs turn both receptors on. MariTide turns GLP-1 on and GIP off.
It is not FDA-approved. It is not a research-vendor "GLP-1/GIP" vial. Phase 2 now has a peer-reviewed New England Journal of Medicine paper, published online June 23, 2025. Phase 3 still has to prove 72-week efficacy, longer safety, and any cardiovascular, heart-failure, or sleep-apnea claim.
What the molecule is
Nature Metabolism describes AMG 133 as a bispecific molecule made by conjugating a fully human monoclonal anti-human GIP-receptor antagonist antibody to two GLP-1 analogue agonist peptides through amino acid linkers. The conjugation sites are cysteine substitutions at E384C. Average molecular weight is about 153,514 daltons.
In cell assays, AMG 133 fully antagonized native GIP at human GIP receptors (IC50 42.4 nM) and fully agonized human GLP-1 receptors (EC50 24.4 pM). Those are in vitro potencies, not human dose targets.
Amgen's June 23, 2025 ADA release repeats the same dual action and calls MariTide "the first monthly or less frequently dosed obesity treatment." That is positioning, not an FDA designation.
The antibody scaffold is why monthly dosing is even plausible. In the Phase 1 paper, mean terminal half-life was about 14 to 16 days for intact AMG 133 (antibody still carrying at least one GLP-1 peptide) and about 21 to 24 days for total AMG 133 (antibody with or without the peptides). Weekly GLP-1 or dual-agonist peptides do not have that backbone.
Why this is not tirzepatide, and not a research vial
Tirzepatide, sold as Mounjaro and Zepbound, is a peptide dual agonist of GLP-1 and GIP. VK2735 is another peptide dual agonist, still investigational. Retatrutide adds a glucagon-receptor agonist on top of GLP-1 and GIP. Eloralintide leaves the incretin pair and uses amylin.
MariTide is in the incretin conversation and still a different chemical class. It is a monoclonal antibody plus two attached peptides, not a single linear or acylated peptide. Sharing a receptor pair is not sharing a product, a schedule, or a label.
Research-vendor listings that use "GLP-1/GIP," "AMG 133," or a MariTide-like name are not Amgen's investigational conjugate. A COA check can tell you whether paperwork names a real lab. It cannot turn an unapproved research vial into the Phase 2 or Phase 3 product.
Nature Metabolism notes the live tension: genetics and antibody work support GIP-receptor inhibition, while dual agonists such as tirzepatide support GIP-receptor activation. Both combinations can produce weight loss. The paper says the precise reason is not known.
Phase 1: three monthly doses and a 150-day tail
The first-in-human study is NCT04478708, reported in Nature Metabolism on February 5, 2024. It was randomized, double-blind, and placebo-controlled in adults with obesity and without diabetes.
The published analysis covers 49 people in single-ascending-dose cohorts (21 mg to 840 mg) and 26 people in multiple-ascending-dose cohorts. Multiple-dose groups received AMG 133 or placebo subcutaneously every 4 weeks on days 1, 29, and 57. Weight was exploratory. Safety was primary.
Dose-dependent weight loss appeared after a single dose and after three monthly doses. In the 420 mg multiple-dose group, mean body-weight change reached -14.5% by day 85, versus +1.5% on placebo, and remained -11.2% as far as 150 days after the last dose. The abstract states that multiple-dose weight loss was maintained for up to 150 days after the last dose.
That tail is the monthly-dosing thesis. It is still a small Phase 1 signal. The 420 mg cell was eight people, and four completed all three doses. Gastrointestinal events were mostly mild nausea and vomiting after the first dose: 68% after dose one among people who received at least two doses, then 9% after later doses. Four people in the 420 mg group left after one dose. The authors flagged intra-patient dose escalation as the next design move.
Phase 2 design: two cohorts, two estimands
The Phase 2 record is NCT05669599. It is a completed, randomized, placebo-controlled, double-blind, dose-ranging study. Actual enrollment was 592. Primary completion was October 8, 2024. The primary endpoint was percent change in body weight from baseline to week 52.
ClinicalTrials.gov and the Amgen release describe two cohorts:
| Item | Cohort A (obesity, no T2D) | Cohort B (obesity with T2D) |
|---|---|---|
| Enrolled | 465 | 127 |
| Condition | Overweight or obesity without type 1 or type 2 diabetes | Overweight or obesity with type 2 diabetes |
| NEJM baseline | 63% female; mean age 47.9; mean BMI 37.9 | 42% female; mean age 55.1; mean BMI 36.5 |
| Part 1 arms (Amgen) | Monthly fixed 140 mg, 280 mg, or 420 mg; 420 mg every 8 weeks; two escalation arms from 70 mg to 420 mg over 4 or 12 weeks; placebo | Monthly fixed 140 mg, 280 mg, or 420 mg; placebo |
The NEJM abstract matches that Part 1 map, including the every-8-week 420 mg arm and the two escalation schedules.
Part 2 is a later re-randomization for people who lost at least 15% at week 52 and were still on investigational product. Amgen said they could be assigned to placebo, monthly 70 mg, 140 mg, or 420 mg, or 420 mg every 12 weeks. Those results are not the week-52 primary analysis.
The two week-52 numbers that circulate are not interchangeable.
| Estimand | What it asks | How missing data are handled (Amgen) |
|---|---|---|
| Treatment-policy (intention-to-treat) | What happened to randomized people, whether or not they stayed on drug | All endpoint data count. If a person stops and the endpoint is missing, the analysis treats later weight as similar to placebo. Amgen says this matches regulatory guidance. |
| Efficacy estimand | What would have happened if people had stayed on MariTide for 52 weeks | Data count while the person is on drug. After early stop, later weight is modeled from that person's response and a predicted off-drug path. |
Amgen says the gap between the two was driven by early discontinuations and by a conservatively defined treatment-policy rule. The treatment-policy range is the ITT number. The "~20%" and "~17%" headlines are the efficacy estimand.
Phase 2 week-52 results
The NEJM paper is the peer-reviewed treatment-policy account. The Amgen ADA release adds the efficacy-estimand range and the company's "no plateau" reading.
| Population | Treatment-policy (ITT) mean weight change at week 52 | Efficacy estimand (Amgen) |
|---|---|---|
| Obesity, no T2D (n=465) | -12.3% (95% CI, -15.0 to -9.7) to -16.2% (95% CI, -18.9 to -13.5) vs -2.5% (95% CI, -4.2 to -0.7) placebo | -16.3% to -19.9% vs -2.6% placebo |
| Obesity with T2D (n=127) | -8.4% (95% CI, -11.0 to -5.7) to -12.3% (95% CI, -15.3 to -9.2) vs -1.7% (95% CI, -2.9 to -0.6) placebo | -12.1% to -17.0% vs -1.4% placebo |
The NEJM abstract reports those ITT ranges across active arms, not one winning dose. The company's "~20%" and "~17%" lines are the top of the efficacy-estimand range (19.9% and 17.0% average weight loss).
On the treatment-policy estimand, HbA1c in the diabetes cohort changed by -1.2 to -1.6 percentage points with maridebart cafraglutide and +0.1 with placebo. Amgen cites a reduction of up to 2.2% on the efficacy estimand. The 2.2% figure is the company efficacy number, not the ITT range.
Amgen and investigator Ania Jastreboff said weight loss had not plateaued at 52 weeks. That is a company and presenter reading of the curves, not proof that loss continues through 72-week Phase 3. The release also lists Phase 2 improvements in waist, blood pressure, hs-CRP, and selected lipids. Those are not cardiovascular-outcome results.
Gastrointestinal events and dose escalation
Gastrointestinal adverse events were common. The NEJM abstract says they were less frequent with dose escalation and a lower starting dose. No unexpected safety signals emerged.
Amgen said most gastrointestinal events were mild to moderate, clustered around initial dosing, and less frequent when escalation was used, without a stated loss of efficacy. Discontinuation because of gastrointestinal events was up to 7.8% in the dose-escalation arms, lower than in the non-escalation arms. The company used a daily symptom tool (MINVR) in addition to standard adverse-event reporting.
A separate Phase 1 low-dose-initiation study (NCT06976372) tested still-lower starts. Amgen reported vomiting in 24.4% on a 21/70/350 mg sequence and 22.5% on a 35/70/350 mg sequence, and no gastrointestinal discontinuations in that primary analysis. That study is not the 592-person Phase 2 trial. Phase 3 chronic-weight-management protocols, per the same release, use an eight-week escalation (21 mg, then 35 mg, then 70 mg) before one of three target doses. Those milligram figures are trial-design numbers, not a consumer protocol.
What Phase 3 still must prove
Amgen's June 23, 2025 release said the MARITIME chronic-weight-management studies were actively enrolling. It also said Phase 3 outcome studies in atherosclerotic cardiovascular disease and heart failure, plus a Phase 3 obstructive-sleep-apnea study, would start in 2025.
The assigned example is MARITIME-2 (NCT06858878): Phase 3, randomized, double-blind, placebo-controlled, in adults with type 2 diabetes who have obesity or overweight. The primary objective is superiority to placebo for percent weight change at week 72.
The registry now lists MARITIME-2 as active, not recruiting. Actual enrollment is 1,105. Arms are high, medium, and low subcutaneous dose, plus placebo. Estimated primary completion is January 21, 2027. Enrollment complete is not a result. Week 72 is still not a cardiovascular-outcomes trial. HbA1c is secondary. Sleep apnea, heart failure, and atherosclerotic disease are other programs.
Nothing in the sources used here is an FDA approval, a completed Phase 3 efficacy readout, or a label for monthly obesity treatment.
What this does not mean
- It is not FDA approval. Maridebart cafraglutide remains investigational.
- It is not a weekly dual agonist. Tirzepatide and VK2735 activate GIP. MariTide blocks GIP. Cross-trial percentages are not a ranking.
- The ~20% and ~17% figures are the efficacy estimand. The NEJM intention-to-treat ranges are about 12.3% to 16.2% (obesity) and 8.4% to 12.3% (obesity with diabetes), versus 2.5% and 1.7% on placebo.
- "No plateau" is a 52-week company reading. It does not establish 72-week or multi-year continuation.
- Phase 3 is not done. MARITIME-2's primary endpoint is percent weight change at 72 weeks. Cardiovascular, heart-failure, and sleep-apnea claims are unproven.
- Amgen's "first monthly" line is positioning. It is not a regulatory status.
- This is not a research-vendor product. The conjugate used in NCT05669599 and NCT06858878 is not a catalog GLP-1/GIP peptide.
- Milligram figures above are trial-design and published-study numbers. They are not a dosing protocol.
What to watch next
- Whether week 72 remains the MARITIME-2 primary weight look. That is the registered primary endpoint today.
- The 2027 primary-completion window. January 21, 2027, is the current ClinicalTrials.gov estimate, not a locked readout date.
- Which estimand Amgen leads with in Phase 3. Treatment-policy is the regulatory-style ITT number. Efficacy estimand is the "if everyone stayed on drug" number.
- Part 2 of Phase 2, if published: continued treatment, washout, and every-12-week 420 mg maintenance in people who already lost at least 15%.
- Outcome trials. The June 2025 release planned ASCVD, heart-failure, and OSA starts in 2025. A start is not a result.
For the wider incretin map, see the ADA 2026 GLP-1 pipeline roundup. Live tirzepatide and semaglutide listing pages are price-and-vendor views of different products.
FAQs
What is MariTide?
Maridebart cafraglutide (formerly AMG 133): an investigational peptide-antibody conjugate that blocks GIP and activates GLP-1, designed for monthly or less frequent subcutaneous dosing. Not FDA-approved.
Is MariTide a dual GLP-1/GIP agonist like tirzepatide or VK2735?
No. Those drugs activate both receptors. MariTide activates GLP-1 and blocks GIP.
What did Phase 2 show?
In the NEJM ITT analysis of 592 adults, week-52 mean weight change was about -12.3% to -16.2% versus -2.5% placebo without diabetes, and about -8.4% to -12.3% versus -1.7% with type 2 diabetes. HbA1c fell 1.2 to 1.6 percentage points on that estimand, versus +0.1 on placebo.
Why do some articles say 20% weight loss?
That is Amgen's efficacy estimand (up to 19.9% and 17.0%), which models staying on drug. It is not the NEJM ITT range.
Is MariTide FDA-approved?
No. MARITIME-2 has a 72-week primary weight endpoint and estimated primary completion in January 2027.
Can someone buy MariTide from a research vendor?
No. Catalog "GLP-1/GIP" vials are not Amgen's investigational conjugate and were not the product in NCT05669599 or NCT06858878.
This article is for general informational purposes only and is not medical advice. Talk to a licensed clinician about personal medical decisions. Maridebart cafraglutide is an investigational peptide-antibody conjugate studied in controlled trials. Nothing here is a dosing protocol, an availability claim, or a recommendation to obtain an unapproved product.
Sources
- Jastreboff AM et al. Once-Monthly Maridebart Cafraglutide for the Treatment of Obesity - A Phase 2 Trial. N Engl J Med. Published online June 23, 2025. DOI 10.1056/NEJMoa2504214
- Véniant MM et al. A GIPR antagonist conjugated to GLP-1 analogues promotes weight loss. Nat Metab. 2024. DOI 10.1038/s42255-023-00966-w
- ClinicalTrials.gov NCT05669599. Phase 2 dose-ranging study of AMG 133 / maridebart cafraglutide
- Amgen. Results from Amgen's Phase 2 obesity study of monthly MariTide presented at the American Diabetes Association 85th Scientific Sessions. June 23, 2025
- ClinicalTrials.gov NCT06858878. MARITIME-2: Phase 3 maridebart cafraglutide in adults with type 2 diabetes who have obesity or overweight