On 28 July 2026 the FDA released 17 revised draft product-specific guidances for generic peptide drugs — and withdrew the 2021 synthetic peptide guidance that underpinned the field. Here is what changed on impurities, higher order structure and immunogenicity testing.
On 28 July 2026, the US Food and Drug Administration published a batch of 17 revised draft product-specific guidances (PSGs) for peptide drug products, followed a day later by a Federal Register notice under the standing PSG docket FDA-2007-D-0369. Comments close on 28 September 2026.
The product list reads like a map of modern metabolic and specialty medicine: semaglutide, tirzepatide and liraglutide sit alongside glucagon, dasiglucagon, calcitonin salmon, teriparatide, pegcetacoplan and vosoritide. All are injectable peptides, and all are second or third revisions rather than new documents — a pattern the consultancy Lachman flagged as unusual, since FDA normally posts a quarterly mix of new and revised PSGs across many drug classes rather than a single-class batch.
Alongside the release, FDA said it is withdrawing its May 2021 guidance "ANDAs for Certain Highly Purified Synthetic Peptide Drug Products That Refer to Listed Drugs of rDNA Origin", because it no longer reflects the agency's current scientific thinking; a replacement is expected later in 2026. That withdrawal is arguably the bigger story, since the 2021 document was the intellectual backbone of synthetic generic peptide development for five years.
A PSG is FDA's published recommendation for how to demonstrate that a proposed generic is equivalent to one specific brand-name reference listed drug (RLD). It tells a developer which bioequivalence study design to run — or when studies can be waived — and what analytical evidence the Office of Generic Drugs expects to see in an abbreviated new drug application (ANDA).
Two things are worth stating precisely. PSGs are not regulations: draft guidances represent FDA's current thinking, are non-binding on both agency and industry, and permit alternative approaches that satisfy the applicable statutes. They are nonetheless the clearest available signal of how reviewers will actually assess a submission, so developers treat them as the default specification and negotiate departures through controlled correspondence or a pre-ANDA meeting — a route these revisions explicitly expand, including for non-clinical assays used to assess risk in recombinant generic peptides.
A conventional small-molecule generic is comparatively tractable. Aspirin is aspirin; a handful of analytical methods establish that the active ingredient is the same molecule as in the brand, and the regulatory question shifts to how much of it reaches the bloodstream.
Peptides occupy an awkward middle ground. Under US law, a polymer of 40 or fewer amino acids is regulated as a drug, and is therefore eligible for the ANDA pathway, while anything larger is a biologic requiring the biosimilar route. So a 31-amino-acid peptide such as semaglutide is legally a small molecule and scientifically something much closer to a protein. It has secondary structure, it can aggregate, and it can provoke an immune response. The ways it can go subtly wrong are numerous: a deleted or inserted residue, a racemised amino acid, an oxidised methionine, a deamidated asparagine, a truncated chain, or a lipid side chain attached in the wrong place.
Critically, few of those defects change molecular weight enough to be obvious, and many co-elute with the main peak on any single chromatographic method. This is why the literature on generic peptide characterisation converges on one word: orthogonal. As Kuril and colleagues set out in Analytical Biochemistry (2024), establishing sequence identity, purity, structural integrity and stability requires multiple independent techniques whose failure modes do not overlap — typically reversed-phase and ion-exchange chromatography, high-resolution and tandem mass spectrometry, peptide mapping, amino acid analysis, circular dichroism and NMR.
The revised PSGs update FDA's recommendations across 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. New or expanded sections cover higher order structure comparison against the RLD, a more comprehensive analysis of the RLD's active ingredient, pre-ANDA meeting procedures, and — for pen and autoinjector presentations — device and user interface assessment.
The logic connecting these is manufacturing route. A brand peptide made recombinantly in yeast or E. coli and a generic made by solid-phase synthesis converge on the same primary sequence but diverge completely in impurity profile: host cell proteins and DNA on one side; deletion sequences, protecting-group adducts and residual synthesis reagents on the other. Sameness of the active ingredient does not imply sameness of everything else in the vial.
Impurities are where peptide risk concentrates. Under the framework carried forward from the now-withdrawn 2021 guidance, FDA set an identification threshold of 0.10% for specified peptide-related impurities and treated any new peptide-related impurity above 0.5% as generally incompatible with the ANDA pathway. The scientific case for that stringency is well documented. Mattei and colleagues, in Frontiers in Immunology (2025), screened teriparatide and 34 of its impurities using computational T-cell epitope prediction, HLA binding assays and ex vivo human PBMC assays. Eight of nine tested impurities elicited T-cell responses in 24–48% of donor samples, against 19% for the reference product — and the authors identified a tolerogenic epitope in the parent molecule whose disruption by an impurity plausibly explains the difference.
Analysis: the practical significance is that a manufacturing change producing a chemically trivial variant can produce an immunologically non-trivial one. FDA's addition of innate immune response testing and biological activity assessment is consistent with a regulator asking developers to characterise that risk analytically rather than clinically. De Groot and colleagues (Pharmaceutical Research, 2025) argue the same from the industry side: orthogonal immunogenicity risk assessment can substitute for clinical immunogenicity studies in the generic context.
Peptides were a specialty niche for most of their history. They are not one now. Market analysts put the global peptide therapeutics market in the region of $58 billion in 2026 and the peptide CDMO segment at roughly $5.5 billion, with forecast growth well above the pharmaceutical average. These are commercial market-research estimates, not regulatory data, and should be read as directional.
GLP-1 receptor agonists are the reason. Semaglutide and tirzepatide converted peptide manufacturing from a boutique capability into a capacity constraint, and the capital response has been extraordinary: CordenPharma has committed around €900 million to GLP-1 capacity, and Eli Lilly has invested over $6 billion in a Huntsville, Alabama API site intended to produce peptide medicines including the oral GLP-1 orforglipron.
Analysis: that build-out was sized for branded demand. As key GLP-1 patents approach expiry across various markets, the same synthesis suites, purification trains and analytical laboratories become the infrastructure for generic entry — which is why a PSG telling developers what evidence FDA wants is commercially consequential rather than merely procedural.
For an ANDA sponsor, the direction of travel is heavier analytical characterisation in exchange for a lighter clinical burden. The revisions retain the possibility of waiving in vivo bioequivalence studies under 21 CFR 320.22(b)(1) and clarify that non-clinical methods can support equivalence conclusions — but the price is a comparative data package covering impurities, higher order structure, biological activity and immunogenicity risk, plus device comparability for pen presentations.
For CDMOs, analytical capability becomes a differentiator on par with synthesis capacity. Mass spectrometry, higher order structure methods and immunogenicity risk assessment are no longer late-stage checkboxes; they shape process development, because an impurity that cannot be controlled below threshold at commercial scale can disqualify the ANDA route entirely.
The caveat: these are drafts, the 2021 foundational guidance has been withdrawn without its replacement yet published, and FDA will revise in response to comments due 28 September 2026. Anyone building a development plan on the July text should expect the details — impurity thresholds above all — to move.
| Term | Meaning |
|---|---|
| ANDA | Abbreviated New Drug Application — the US filing route for a generic drug, relying on the brand's safety and efficacy data rather than repeating trials. |
| PSG | Product-Specific Guidance — FDA's published, non-binding recommendations for demonstrating equivalence to one named reference product. |
| RLD | Reference Listed Drug — the approved brand product a generic is compared against. |
| Bioequivalence (BE) | Demonstration that the generic delivers the active ingredient at a comparable rate and extent to the RLD. |
| Higher order structure | A peptide's folded three-dimensional shape (secondary and above), distinct from its amino acid sequence. |
| Peptide-related impurity | A variant arising from the peptide itself — deletion, insertion, truncation, oxidation, deamidation, racemisation — as opposed to a process reagent. |
| Immunogenicity | The capacity of a product or impurity to provoke an unwanted immune response. |
| Innate immune response testing | Assays for non-antibody-mediated immune activation, for example by residual process material or aggregates. |
| Orthogonal methods | Independent analytical techniques with non-overlapping blind spots, used together so one catches what another misses. |
| CDMO | Contract Development and Manufacturing Organisation — a company that develops and manufactures drug products on behalf of others. |
| Q1/Q2 | Qualitative and quantitative sameness of a formulation's inactive ingredients relative to the RLD. |
| SPPS | Solid-phase peptide synthesis — the dominant chemical route to synthetic peptides, building the chain one residue at a time on a resin support. |
No. A PSG describes what evidence FDA expects in an application. It says nothing about patent or exclusivity status, and it approves nothing. It makes the technical path clearer, not shorter.
No. Draft guidances state FDA's current thinking and are explicitly non-binding on FDA and industry. Developers may use alternative approaches that satisfy the applicable statutes and regulations.
US law draws the line at 40 amino acids. At or below that count a polymer is regulated as a drug and can use the ANDA pathway; above it, the product is a biologic requiring a biosimilar application.
FDA stated it no longer reflects the agency's current scientific thinking. Analysis: the intervening five years produced considerably more data on peptide impurity immunogenicity and higher order structure methods than existed when it was written; a replacement is expected later in 2026.
Analysis: the addition of higher order structure comparison and innate immune response testing as explicit expectations. Both push generic peptide submissions toward a characterisation package resembling biosimilar comparability more than traditional small-molecule sameness.
Potentially. Waivers under 21 CFR 320.22(b)(1) remain available, and the revisions clarify that non-clinical methods can support equivalence conclusions — but the supporting analytical package is correspondingly larger.
28 September 2026, under docket FDA-2007-D-0369.
Regulatory:
Peer-reviewed:
Trade press and analysis:
Disclaimer: This article covers FDA-regulated pharmaceutical products and the US generic approval pathway. It is regulatory and industry commentary, not medical advice, and it is not guidance on research-use materials. Statements labelled "Analysis" are our interpretation rather than FDA policy. All products sold on this site are supplied strictly for in-vitro laboratory and research use, not for human consumption.
Learn the correct technique for reconstituting lyophilised research peptides using bacteriostatic water for accurate, contamination-free preparations.
Research GuidesProper storage is critical to maintaining peptide integrity. Learn the correct temperature requirements and handling practices for lyophilised and reconstituted peptides.
99%+ HPLC purity · COA verified every batch · Next-day dispatch to Melbourne
Shop All Peptides