The FDA published 17 revised draft product-specific guidances for peptide drug products on July 28, 2026, with the Federal Register notice following on July 29. Product-specific guidances are the agency's technical roadmap for generic developers: they spell out what a company has to demonstrate to file an abbreviated new drug application against an approved reference product. The batch covers semaglutide, tirzepatide, liraglutide, teriparatide, pegcetacoplan, glucagon, dasiglucagon, calcitonin salmon, and vosoritide.
The more consequential line in the announcement is the withdrawal. The FDA is pulling its May 2021 guidance on ANDAs for certain highly purified synthetic peptide drug products that reference listed drugs of rDNA origin, on the grounds that it no longer reflects the agency's current scientific thinking. That single document has been the governing text for whether a peptide made by chemical synthesis can be treated as the same active ingredient as one made recombinantly — which is the threshold question for nearly every GLP-1 generic program in development.
Research-context information only. This article reports on published regulatory documents. Nothing here is medical or legal advice. Semaglutide, tirzepatide, and liraglutide are FDA-approved as prescription medicines; the research-use-only material sold by peptide vendors is not FDA-approved for human use and has not been evaluated for safety, purity, or potency in that channel. Consult a licensed physician for personal medical decisions.
What the FDA actually published
Seventeen revised draft PSGs, spanning nine active ingredients across multiple reference applications:
| Active ingredient | Reference applications covered | Therapeutic area |
|---|---|---|
| Semaglutide | 2 (NDA 209637, NDA 215256) | Type 2 diabetes, obesity |
| Tirzepatide | 2 (NDA 215866, NDA 217806) | Type 2 diabetes, obesity |
| Liraglutide | 2 (NDA 022341, NDA 206321) | Type 2 diabetes, obesity |
| Teriparatide | 2 (NDA 021318, NDA 218771) | Osteoporosis |
| Pegcetacoplan | 2 (NDA 217171, NDA 215014) | Geographic atrophy, PNH |
| Glucagon | 3 (NDA 210134, NDA 020928, NDA 212097) | Severe hypoglycemia |
| Calcitonin salmon | 2 (NDA 017769, NDA 017808) | Osteoporosis, hypercalcemia |
| Dasiglucagon | 1 (NDA 214231) | Severe hypoglycemia |
| Vosoritide | 1 (NDA 214938) | Achondroplasia |
The revisions concentrate on five areas: submission of recombinantly, synthetically, or semi-synthetically produced peptides as ANDAs; innate immune response testing; impurity thresholds; higher order structure assessment; and biological activity assessment.
Comments are due September 28, 2026. Sponsors can propose alternative bioequivalence approaches with justification, though in practice deviating from a PSG invites more review scrutiny, not less.
Why pulling the 2021 guidance is the real signal
Semaglutide and tirzepatide are approved under new drug applications, not biologics license applications. That is a legal distinction with enormous commercial consequences: it means a copy can be filed as an ANDA — a generic, with no comparative clinical efficacy trial required — rather than as a biosimilar, which needs a comparative program that costs an order of magnitude more and takes years longer.
The catch is that the reference products are made recombinantly, in engineered cells, while most would-be generic manufacturers plan to make the same molecule by solid-phase chemical synthesis. Two routes to the same amino acid sequence produce different impurity profiles: recombinant production yields host-cell-related impurities, synthesis yields sequence-related ones such as deletion, insertion, and diastereomeric variants. The 2021 guidance set out how the agency would decide whether a synthetic version could still be called the same active ingredient.
Withdrawing it without a finished replacement leaves developers between frameworks. The PSGs published this week carry the interim expectations, and a broader updated framework has been signaled for later in 2026. For programs already in flight, revised impurity thresholds and higher-order-structure expectations can mean re-running characterization work rather than simply refiling.
What this changes for price and access
In the near term for US buyers: nothing. A product-specific guidance is not an approval and does not touch patent exclusivity. The gating factor on US generic entry remains the patent estate — core semaglutide protection runs into December 2031, tirzepatide later still, and both companies have filed extensive follow-on applications around formulation, device, and dosing regimen.
The one GLP-1 already past that gate is liraglutide. The FDA cleared a generic against the diabetes reference product in December 2024 and granted full approval of a generic against the weight-management reference product in August 2025 — the first time a GLP-1 receptor agonist faced true generic competition in the US.
Outside the US the clock is running faster. Semaglutide data exclusivity lapsed in Canada on January 4, 2026 after a maintenance-fee issue caused key patent protection to expire earlier than expected, and multiple manufacturers have filed generic applications with Health Canada, with pricing signaled well below the branded equivalent. Canada is the live experiment for what happens to a GLP-1 market when the exclusivity wall comes down.
None of this alters the research-use-only channel today. Per-mg pricing and COA documentation by vendor are on the semaglutide and tirzepatide pages, and active vendor codes are on the deals page.
