Pfizer discontinued two obesity programs in the pipeline update it published alongside its second-quarter results on August 4, 2026. One was an oral GLP-1 receptor agonist peptide inherited from the roughly $10 billion Metsera acquisition. The other was a phase 2 small-molecule GIPR antagonist that had been the last survivor of the company's pre-Metsera obesity portfolio.
The cuts landed in the same announcement as an additional $2.5 billion in cost reductions, which lifts Pfizer's "Realigning Our Cost Base" program target to about $6.7 billion from $5.7 billion. For anyone tracking this class, though, the interesting part is not the accounting. It is that the two programs Pfizer killed represent the two most-debated bets in metabolic peptide design right now — oral delivery of a peptide, and blocking rather than activating the GIP receptor.
Research-context information only. Compounds discussed below are research peptides and supplements; some are investigational drugs not approved by the FDA. Protocols, doses, and reactions reported come from published research and self-reported community sources. Possession or use of investigational drugs outside an authorized clinical trial may be illegal in your jurisdiction. This article reports what has been documented, not what should be done. Consult a licensed physician for personal medical decisions. The two Pfizer/Metsera compounds discussed in this article were discontinued in-house and were never available outside Pfizer's own clinical trials.
What Pfizer Cut
PF-6796 — the oral GLP-1 peptide. This is the molecule Metsera carried as MET-224o, one of its two lead oral candidates, and it was in phase 1 when Pfizer stopped it. Pfizer's stated position in the update is that it remains committed to exploring both oral peptide and small-molecule approaches to GLP-1 receptor agonism, which reads as a program cut rather than an abandonment of the route.
PF-07976016 — the GIPR antagonist. This one is the bigger surprise. PF-07976016 is a small-molecule antagonist of the glucose-dependent insulinotropic polypeptide receptor, and it was the only asset to survive the obesity clearout that preceded Pfizer's rebuild through the YaoPharma license and the Metsera purchase. Reporting on the update indicates it was discontinued after review of phase 2a data generated on a liraglutide background — that is, the antagonist was being tested layered on top of an existing GLP-1 agonist rather than as a standalone.
That kill has a knock-on effect. In December 2025, Pfizer had announced plans to test PF-07976016 in combination with YP05002, the oral small-molecule GLP-1 it licensed from China's YaoPharma for $150 million upfront. With the antagonist gone, that combination goes with it, and the YaoPharma compound — now carrying the Pfizer code PF-08642534 — becomes the company's only remaining oral GLP-1 candidate.
The GIPR Paradox Is Still Unresolved
The PF-07976016 discontinuation is worth sitting with, because it lands in the middle of a genuine scientific disagreement rather than settling one.
The GIP receptor can be pushed in either direction, and both directions have produced human weight loss data:
| Approach | Representative compound | Reported outcome |
|---|---|---|
| GIPR agonism (+ GLP-1) | tirzepatide | ~21% mean weight reduction at 72 weeks (SURMOUNT-1) |
| GIPR agonism (+ GLP-1 + glucagon) | retatrutide | 30.3% at 104 weeks (TRIUMPH-1); 22.6% at 80 weeks (TRIUMPH-3) |
| GIPR antagonism (+ GLP-1) | maridebart cafraglutide (Amgen) | up to -19.9% at 52 weeks, phase 2 efficacy estimand |
| GIPR antagonism, small molecule | PF-07976016 (Pfizer) | discontinued at phase 2 |
Amgen's compound is an antibody-peptide conjugate — a GIPR-blocking monoclonal antibody linked to two GLP-1 agonist peptides — and its six-trial phase 3 MARITIME program has been running since 2025. So the antagonist thesis has not been falsified by Pfizer's decision. What failed was one small molecule, in one combination design, at one phase.
The practical read is narrower than the headlines suggest: a small-molecule antagonist stacked on liraglutide did not clear Pfizer's internal bar. Whether GIPR blockade beats GIPR activation remains an open question that phase 3 readouts in 2027 will address far more definitively than a quarterly pipeline table.
What Survived
Pfizer's obesity portfolio after the cut is narrower but still front-loaded on the injectable side:
- Berobenatide (previously PF-3944 / PF-08653944 / MET-097i) — the once-monthly injectable GLP-1 that anchored the Metsera deal. Pfizer has said it plans roughly ten pivotal studies in 2026, spanning chronic weight management and obesity-related comorbidities including knee osteoarthritis and obstructive sleep apnea. We covered the phase 2b VESPER data and the monthly-dosing thesis when it read out.
- PF-08653945 (MET-233i) — an ultra-long-acting amylin analog in the SOLIS-1 phase 2b study, tested both as monotherapy and in combination with berobenatide. Amylin is the same target class as cagrilintide.
- PF-08642534 (YP05002) — the licensed oral small-molecule GLP-1, now the only oral in the class Pfizer still holds.
- MET-034i — a GIPR agonist in phase 1, alone or with berobenatide. Note the direction: Pfizer retained an agonist while cutting the antagonist.
What This Actually Means for Buyers
Very little changes in the near term, and it is worth being precise about why.
Neither discontinued compound was ever sourceable. PF-6796 and PF-07976016 are proprietary Pfizer molecules under active patent, in phase 1 and phase 2 respectively. They never appeared in research vendor catalogs, they are not on any compounding bulk-substance list, and no version of either was available outside Pfizer's own trials. A discontinuation removes a future option, not a current one.
The route lesson is the durable one. Pfizer has now discontinued oral GLP-1 candidates from both sides of the chemistry — the small molecule danuglipron in April 2025 after a reported case of drug-induced liver injury, and now the oral peptide PF-6796 at phase 1 — while its injectable program keeps advancing into pivotal trials. That asymmetry mirrors what already governs the research-compound market: the peptides vendors actually stock are lyophilized powders reconstituted for subcutaneous injection, because oral bioavailability for this class remains a hard, largely unsolved formulation problem. Approved oral GLP-1s exist, but they rely on absorption-enhancer formulations that do not translate to a vial of powder and a syringe.
So the compounds available for research use today are the same ones that were available yesterday: semaglutide, tirzepatide, retatrutide, cagrilintide and the smaller set of related molecules with non-proprietary chemistry. Current per-milligram pricing across vendors is on the retatrutide, tirzepatide and semaglutide comparison pages, and active vendor discount codes are collected on the deals page.
