Monthly GLP-1: Why Pfizer, Amgen, and Ascletis Are Stretching the Dose Interval

The big obesity drugs are weekly because that is what their chemistry allows. Native GLP-1 lasts about two minutes in the bloodstream. Semaglutide stretches that to about seven days through a fatty-acid tail that binds to albumin; tirzepatide pushes it to about five. Both are engineering achievements, and both top out around once-a-week dosing. The next race is to make a monthly injection work, and three different chemical approaches are converging on that target.

Amgen's MariTide is a peptide bolted to a monoclonal antibody, borrowing the antibody's natural multi-week persistence. Pfizer's berobenatide, the lead asset from its $4.9 billion Metsera acquisition, is a peptide with a more aggressive lipid modification that pushes its half-life to about 15 days. Ascletis's ASC30 is a small-molecule GLP-1 packaged in a subcutaneous depot with a 46-day half-life that could go quarterly. None of these match weekly tirzepatide on raw weight loss. The bet is that the patient who skips fewer doses loses more weight in real life than the patient on a stronger drug who quits.

Trial discontinuation rates make GLP-1 drugs look reasonable. The Wegovy label cites discontinuation for nausea at 1.8% and for vomiting at 1.2%, against placebo rates near zero. Real-world data tell a different story. A 2025 Truveta analysis of 125,474 GLP-1 starters found that 64.8% of patients without diabetes had stopped within a year, and 46.5% of those with diabetes had done the same. A Prime Therapeutics study put obesity-without-diabetes three-year persistence at 8.1%; semaglutide led at 14.3%, daily liraglutide trailed at 2.5%. Amgen's own analysis reports that 58% of patients regain weight within twelve months of stopping.

A share of that dropout is side effects. Some of it is cost, some is insurance churn, some is the simple grind of remembering a weekly injection. The dosing-interval thesis is that compressing the injection schedule from fifty-two times a year to twelve cuts the friction enough to keep patients on therapy long enough to benefit. Patient-preference surveys say the marginal disutility of weekly versus daily is small; the marginal value of monthly over weekly is smaller still. But disutility studies measure what patients will pay extra for, not what they will quit over. The real-world persistence gap, weekly versus weekly-with-grocery-bag-of-pens-in-the-fridge, may be larger than the survey data suggest.

Plasma residence time is the engineering problem. To dose a drug once a month while keeping its concentration above the efficacy threshold the whole time, you need a half-life of roughly two to three weeks. Native GLP-1 has a half-life measured in minutes because the enzyme DPP-4 cleaves it almost immediately. Semaglutide and tirzepatide both attach long fatty-acid chains to specific lysine residues so the molecule binds to circulating albumin, which shields it from proteolysis and slows its clearance through the kidneys. That chemistry gets you to weekly. It does not easily get you to monthly.

Three approaches have stretched it further. The first is to attach the peptide to a much larger antibody, so the conjugate inherits the antibody's natural plasma persistence. Antibodies live for weeks in circulation through a recycling pathway involving the neonatal Fc receptor. Amgen's MariTide does this: it is a monoclonal antibody that antagonizes the GIP receptor, with two GLP-1 agonist peptides conjugated to it. The second is to push the albumin-binding chemistry harder. Pfizer's berobenatide uses the HALO lipidation platform from Metsera, which produces a peptide half-life of about 15.5 days, roughly twice tirzepatide's. The third is to give up on persistent plasma levels entirely and use a subcutaneous depot. Ascletis's ASC30 is a small-molecule GLP-1 agonist in a long-acting depot formulation, with a measured half-life of 46 days for the standard depot and 75 days for a maintenance formulation now in development.

MariTide (maridebart cafraglutide) is the most clinically advanced monthly candidate. Its structure is unusual: a full monoclonal antibody that blocks the GIP receptor, with two GLP-1 receptor agonist peptides linked to it. The combination is therefore a dual mechanism, but the opposite of tirzepatide. Tirzepatide agonizes both GLP-1 and GIP receptors. MariTide agonizes GLP-1 and blocks GIP. There is a biological argument for either approach, and the comparison will eventually run head-to-head.

The Phase 2 data, published in the New England Journal of Medicine in September 2025, enrolled 592 adults across two cohorts. In obesity without type 2 diabetes (n=465), MariTide produced up to 20% mean weight loss on the efficacy estimand and 12.3 to 16.2% on the treatment-policy estimand. In type 2 diabetes (n=127), weight loss reached 17% with up to 2.2 percentage points of HbA1c reduction. The catch was tolerability. Without dose escalation, 12 to 27% of participants discontinued for GI events; vomiting in some arms ran 43 to 92%. Adding a slow escalation cut discontinuation to as low as 7.8%.

The Phase 3 MARITIME program now uses a sharper escalation: 21 mg, then 35 mg, then 70 mg, stepped over eight weeks, then continued monthly. MARITIME-1 (non-diabetic obesity) and MARITIME-2 (T2D) are enrolling, with primary completion in 2027. Additional Phase 3 trials cover cardiovascular outcomes, heart failure, and obstructive sleep apnea. The structure means MariTide arrives later than the weekly incumbents, but with a defensible niche if the GIP-antagonism mechanism produces something the dual-agonist approach cannot.

Berobenatide is the lead asset from Pfizer's $4.9 billion acquisition of Metsera in late 2025, which carried up to $22.50 per share in additional contingent value rights tied to monthly approvals and combination milestones. The molecule (PF-07976094, formerly MET-097i) is a 41-amino-acid peptide with a HALO lipidation that produces a 15.5-day half-life and lets the company dose monthly through a 0.5 mL subcutaneous injection. Pfizer is developing the asset on two tracks: a weekly version with a Phase 3 launch around 2028 and a monthly maintenance version targeted for 2029.

ADA 2026 was the validation point. VESPER-1 extended to 32 weeks showed 15.9% non-placebo-adjusted weight loss with no plateau. VESPER-2 in type 2 diabetes produced 10.2% placebo-adjusted weight loss and a 2.2% HbA1c drop at 28 weeks. VESPER-3, the monthly-maintenance trial, produced up to 12.3% placebo-adjusted weight loss at 28 weeks after a weekly lead-in, with 9.3% overall adverse-event discontinuation but rates as high as 20.8% at the top monthly dose. BioPharma Dive's ADA wrap called the data "foundational," Leerink called Pfizer back into the obesity conversation, RBC's Trung Huynh called it "solid but relatively undifferentiated" against Wegovy HD and tirzepatide.

Pfizer's plan is ten Phase 3 trials in 2026, plus more than twenty obesity studies overall. VESPER-6 is the registrational monthly-maintenance trial. The strategic logic of the acquisition rests less on this asset's headline weight number, which trails tirzepatide, and more on the combination case: pairing the monthly GLP-1 berobenatide with Metsera's monthly amylin analog MET-233i, the same combination structure that the CVR payouts in the Metsera deal explicitly reward.

Ascletis's ASC30 is a different kind of monthly. The molecule is a small-molecule GLP-1 receptor agonist, not a peptide, formulated as a long-acting subcutaneous depot. The pharmacology runs through depot biology rather than peptide engineering: the drug is released slowly from the injection site over weeks, and the formulation rather than the molecule controls the dosing interval. The standard depot has a 46-day half-life. A maintenance formulation reported a 75-day half-life in early human PK, which would support quarterly dosing.

The US Phase 2 trial enrolled 65 adults with obesity at three doses of 400 mg given at day 1, week 4, and week 8. Placebo-adjusted weight loss was 6.3% at 12 weeks, 7.5% at 16 weeks, and held at 6.4% at 20 weeks and 5.8% at 24 weeks, four months after the last injection. No patient discontinued for adverse events. The numbers are well below the weekly and monthly incumbents on peak efficacy, but the maintenance pattern is striking: the drug holds its weight effect long after dosing stops. Ascletis is also developing ASC30 as an oral daily tablet in parallel, which posted 7.7% placebo-adjusted weight loss at 13 weeks.

The depot approach is the longest-shot of the three. It introduces depot-specific complications (injection-site reactions, irregular release kinetics) and trails the antibody-conjugate and lipidated-peptide approaches on absolute weight loss. But if the maintenance pharmacology holds in Phase 3, it points at something the other approaches do not yet promise: a true once-quarterly obesity drug for stable maintenance after weight loss is established.

Monthly dosing inherits a problem that weekly dosing solved. When a GLP-1 produces nausea or vomiting on a weekly schedule, the patient can skip a dose, drop a step in the titration, or wait out the worst of the side effect with the knowledge that drug levels will fall within a few days. On a monthly schedule, the drug stays in the system for weeks. There is no easy way to back out a dose that turned out to be intolerable.

The Phase 2 data already hint at this. MariTide without escalation produced GI discontinuation rates between 12 and 27%; the Phase 3 dose schedule had to add a slower titration to control it. Berobenatide's VESPER-3 saw 11.1% to 20.8% discontinuation across monthly doses, higher than what Zepbound posts on a weekly schedule. The drugs are engineered for long persistence, but the patients still have stomachs. Slow titration is the working answer, but it adds a step that weekly dosing did not need, and it raises the question of whether the convenience argument survives once patients understand the trade-off.

The obesity market has segmented along the dosing axis in the past year. Weekly injectables (Wegovy, Zepbound, Mounjaro) remain the franchise and will stay so through the late 2020s. Oral daily (Foundayo, Wegovy pill, aleniglipron, elecoglipron) is the fastest-growing segment and accounted for roughly a third of new-to-brand prescriptions within eight weeks of the January 2026 Wegovy pill launch. Monthly injectables (MariTide, berobenatide, ASC30) are still pre-approval but advancing on three different chemistries. Combinations (CagriSema, future Roche petrelintide-plus-enicepatide) cut across the axis as the higher-efficacy play.

Monthly is not positioned to displace weekly; the strongest weekly drugs still produce more weight loss in the same trial. The case for monthly is durability and adherence. If a real-world patient on a monthly drug loses 12% over a year because they did not quit, while a patient on a weekly drug quits at six months after losing 8%, the monthly drug wins the actual cohort outcome even though it loses every short-term trial. That is the bet. Whether it pays off depends on Phase 3 readouts that begin in 2027 and on whether the tolerability trap can be managed at scale.

Monthly GLP-1 is the maintenance-phase play for a market that has spent five years optimizing for peak weight loss. Three drugs with three chemistries (antibody conjugate, lipidated peptide, subcutaneous depot) are now in or near registrational trials, all of them backing the same hypothesis: that the patient who actually stays on therapy loses more weight than the patient on a stronger drug who quits. The data behind that hypothesis are real (Truveta 64.8% one-year discontinuation, Prime 8% three-year persistence) but the assumption that monthly dosing fixes them is still untested at scale.

The near-term competitive map is set. MariTide arrives first with the strongest weight number among monthly candidates (20% Phase 2) and a defensible GIP-antagonist mechanism. Berobenatide arrives second with the cleanest lipidation chemistry, lower peak efficacy, and a combination strategy that rests on pairing with the monthly amylin analog Pfizer also acquired. Ascletis ASC30 arrives later with a quarterly target that the other two cannot match. Whether any of them dislodge the established weekly franchises will turn on Phase 3 cardiovascular outcomes, real-world adherence data that does not yet exist, and whether monthly's tolerability trap can be solved by titration. The bet is reasonable. The data to settle it are years away.