Peptide API impurities and specifications: thresholds, tests and what regulators expect in 2026

Peptide APIs sit between small molecules and proteins, and the impurity rules reflect that. The ICH guidelines most quality teams use for small molecules do not apply, or apply only partly, so peptide specifications rest on pharmacopoeial thresholds, an EMA guideline that came into effect in June 2026, and FDA product-specific guidance that was rewritten in July 2026. This article sets out what counts as a peptide impurity, which thresholds apply in the EU and US, what a peptide API specification should contain, and what to check before accepting one.

Why small-molecule impurity rules do not apply

ICH Q3A(R2), the standard guideline on impurities in new drug substances, is explicit: “The following types of drug substances are not covered in this guideline: biological/biotechnological, peptide, oligonucleotide…” (ICH Q3A(R2)). ICH Q6A on specifications says it “may be applicable to… synthetic peptides of low molecular weight” but is “not sufficient to adequately describe specifications of higher molecular weight peptides and polypeptides” (ICH Q6A).

EMA’s guideline on the development and manufacture of synthetic peptides (EMA/CHMP/CVMP/QWP/367182/2025), adopted in December 2025 and in effect from 1 June 2026, confirms that synthetic peptides “are fully or partially excluded from the scope of ICH Q3A/B… ICH Q6A/B… and ICH M7”, and fills the gap with peptide-specific expectations.

What counts as a peptide impurity

EMA divides impurities into peptide-related impurities, which contain structural elements of the peptide sequence, and non-peptide impurities. The guideline lists the main sources:

Category	Source	Examples
Peptide-related, from starting materials	Impurities in amino acid derivatives carried into the sequence	Wrong enantiomers or diastereomers, wrong amino acids (Ile for Leu), β-alanine residues, dipeptides, incorrectly protected amino acids
Peptide-related, from the process	Incomplete or unwanted reactions during SPPS, LPPS, fragment condensation and cleavage	Deletion, truncated and insertion sequences; diastereomers from epimerisation; adducts on Trp, Tyr or Met from cleavage; incomplete side-chain deprotection
Peptide-related, from degradation	During manufacture or storage	Oxidation, hydrolysis, isomerisation, deamidation, aspartimide formation, diketopiperazine and pyroglutamic acid formation, β-elimination, dehydropeptides, disulfide exchange, acetylation
High molecular weight	Association of peptide chains	Dimers, trimers, oligomers, aggregates
Non-peptide	Process reagents and materials	Reagents and their by-products, residual solvents, elemental impurities, potential mutagenic impurities, residual TFA

The route of synthesis determines which of these can appear; our comparison of SPPS, LPPS and hybrid manufacturing explains why. Aggregation and chemical degradation are also influenced by formulation and processing, including lyophilisation (Zapadka et al., 2017).

Thresholds: EU and US compared

Framework	Reporting	Identification	Qualification / limit	Applies to
ICH Q3A(R2), for comparison only	0.05%	0.10% (or 1.0 mg/day)	0.15% (or 1.0 mg/day)	Small-molecule drug substances, maximum daily dose up to 2 g; not peptides
Ph. Eur. general monograph, cited by EMA	0.1%	0.5%	1.0% (qualification)	Peptide-related impurities in synthetic peptides, EU
EMA, synthetic version of a recombinant reference	Quantification limit 0.1%	Full profile comparison at 0.1–0.5%	Compared impurity by impurity with the EU reference product	Generics of recombinant peptides, EU
FDA draft PSG, semaglutide injection (July 2026)	0.1%	0.5%	Common impurities ≤ reference level or 1.0%, whichever is greater; new impurities reduced below 1.0%; total ≤ reference	Recombinant, synthetic or semi-synthetic generic semaglutide, US

Sources: ICH Q3A(R2); EMA, 2025; FDA draft PSG.

FDA’s change of course in 2026

In May 2021, FDA issued guidance on ANDAs for certain highly purified synthetic peptides that refer to listed drugs of rDNA origin. It covered glucagon, liraglutide, nesiritide, teriparatide and teduglutide, treated new impurities above 0.5% as not acceptable, and required impurities between 0.10% and 0.5% to be identified, characterised and justified with comparative immunogenicity risk assessment (FDA presentation).

On 28 July 2026, FDA withdrew that guidance because it “no longer reflects FDA’s current scientific thinking”, published 17 revised draft product-specific guidances, and said it plans to revise the general guidance (FDA). The revised semaglutide guidance is more permissive on numbers but asks for more comparison. Higher impurity levels “do not preclude an application from being submitted as an ANDA” but are subject to scientific review, and the guidance recommends assessing aggregates and non-peptide impurities as part of the immunogenicity evaluation.

What a peptide API specification contains

EMA’s guideline gives a non-exhaustive list of specification tests for synthetic peptide active substances. The table adds what each test tells a buyer.

Test	Method (as listed by EMA where it names one)	Why it matters
Appearance	Visual	Gross quality and consistency
Identification	Validated method chosen by the manufacturer	Confirms the right sequence and modifications
Purity: total, specified and unspecified impurities	LC, with a method able to support the 0.1% reporting threshold	Core measure of peptide-related impurities
High molecular weight impurities	SEC-LC or other techniques	Aggregates and oligomers; EMA requires them to be named, not just “HMW impurities”
Assay / content	LC, elemental analysis, amino acid analysis, Kjeldahl nitrogen or qNMR	How much of the powder is actually peptide
Counter-ion identity and content	For example, acetic acid	Defines the salt form; EMA requires a justified upper limit
Residual TFA	Validated method chosen by the manufacturer	TFA often remains from purification
Water content	Validated method, commonly Karl Fischer titration	Peptides are often hygroscopic
Residual solvents	GC, per ICH Q3C	Process solvents such as DMF and acetonitrile
Elemental impurities	Per ICH Q3D	Relevant where metal catalysts are used
Bacterial endotoxins	Per pharmacopoeia	Relevant for APIs used in parenteral products
Microbiological purity	Per pharmacopoeia	Bioburden control

EMA also mentions process-specific tests where relevant, such as residual fluoride when HF is used for deprotection.

Purity is not content

Chromatographic purity is the share of the peptide signal that belongs to the main peak. It says nothing about water, counter-ion or other non-peptide mass in the powder. A peptide can be highly pure by HPLC and still contain substantially less than 100% peptide by weight. That is why the specification needs both purity and assay, and why counter-ion and water are controlled. Most approved peptide APIs are acetate salts, while reversed-phase purification often leaves TFA, which is exchanged or reduced during processing (Sikora et al., 2020).

EMA adds two practical points: peptide reference standards need a defined qualification approach that prevents drift in their assigned peptide content, and because peptides are often very hygroscopic, moisture uptake by the reference standard should be controlled and, where appropriate, its moisture content confirmed before use.

Pharmacopoeial monographs

Where a Ph. Eur. monograph exists, it sets the specification and enables a Certificate of Suitability. Older synthetic peptides such as octreotide have monographs and CEPs listed in the EDQM database. For semaglutide, the same database returned no record on 11 October 2026, and we could not confirm a Ph. Eur. or USP monograph when writing this article. Until one exists, the specification is justified in the dossier on the basis of the guidelines above and, for generics, comparison with the reference product.

Semaglutide: what the 2026 guidance adds

The reference semaglutide product’s peptide backbone is made by yeast fermentation and then chemically modified (DailyMed), while many generic APIs are synthetic. Both regulators now ask generic applicants to show sameness with the reference product before anything else:

• FDA’s draft PSG recommends comparative characterisation to support active-ingredient sameness and the impurity comparison in the table above.
• EMA’s draft product-specific bioequivalence guidance for semaglutide (EMA/106257/2026, consultation open until 31 October 2026) states that semaglutide substance sameness should be demonstrated before a bioequivalence study (EMA).
• EMA’s synthetic peptides guideline expects comparative forced degradation studies, including shear and mechanical stress, when a synthetic product is compared with a recombinant reference.

A buyer’s checklist for an impurity package

1. An impurity table for at least three batches, listing each impurity above 0.1% by relative retention time, identity (where known) and level.
2. Identification of every impurity above 0.5%, with the evidence (typically LC-MS).
3. Method validation summaries showing the purity method separates closely related sequences and diastereomers, with a quantification limit that supports 0.1%; additional orthogonal methods where one method is not enough.
4. Aggregate data, with the aggregate species named.
5. Assay, counter-ion, residual TFA and water results on the same batches, so the mass balance can be checked.
6. Stability data showing how degradation impurities develop under the proposed storage condition.
7. For generics, a comparison with the reference product, aged as the guidance describes.

See how to qualify a peptide API supplier for the wider document and audit checklist, and our buyer guides for Europe and the USA.

Winhycare Pharma

Winhycare Pharma supplies Semaglutide API on a B2B basis from Hyderabad, India; the product page sets out its specification targets, and each batch is supplied with a Certificate of Analysis. Our CRDMO services include analytical method development and impurity profiling for clients’ own peptide programmes, and our lyophilisation CMO service takes client-supplied API into vials for research and development. Our research peptide range and custom peptide synthesis are for research use only, not for human or veterinary use. More on the division: Winhycare Pharma.