Peptide Impurity Thresholds: Reporting, Identification and Qualification
A research-peptide certificate of analysis typically discloses one number about impurities, and it does so by implication: purity 98.6%, therefore impurities 1.4%. That single figure is doing the work of what regulatory chemistry treats as three separate obligations, each triggered at a different level and each demanding a different kind of evidence. Collapsing them into one percentage is not a simplification for a lay audience. It removes the information that determines whether a batch is characterized at all.
This article is about the mechanics of that three-tier structure, why the arithmetic behaves counter-intuitively for peptides specifically, and what a research buyer can and cannot conclude from a purity figure. A companion piece in this section deals with the separate question of where the familiar 98% threshold came from; here the concern is the ladder underneath it.
Three thresholds, three obligations
The framework most often cited is ICH Q3A(R2), Impurities in New Drug Substances. Its Attachment 1 sets three limits for a substance at a maximum daily dose of 2 g or less:
- Reporting threshold, 0.05%. Above this level an impurity must appear in the analytical results at all. Below it, the peak may be disregarded.
- Identification threshold, 0.10% (or 1.0 mg per day, whichever is lower). Above this level the impurity must be structurally identified — you must say what the molecule is, not merely that a peak exists.
- Qualification threshold, 0.15% (or 1.0 mg per day, whichever is lower). Above this level the impurity must be qualified: there must be data establishing the biological safety of that specific species at that level.
Above 2 g per day the thresholds tighten to 0.03%, 0.05% and 0.05% respectively.
The guideline also draws a distinction that a purity percentage erases entirely. A specified impurity is “an impurity that is individually listed and limited with a specific acceptance criterion” in the substance specification. An unspecified impurity is “limited by a general acceptance criterion, but not individually listed with its own specific acceptance criterion.” A specification built this way is a table with named rows. A purity claim is a single cell.
The scope sentence almost nobody quotes
ICH Q3A(R2) opens by stating what it covers, and the exclusion list is explicit. The guideline does not apply to “biological/biotechnological, peptide, oligonucleotide, radiopharmaceutical, fermentation product and semi-synthetic products derived therefrom, herbal products, and crude products of animal or plant origin.”
Peptides are named. The most-cited impurity framework in the industry excludes, in its own scope section, the class of molecule this catalog consists of. That does not make its three-tier logic irrelevant — the logic is sound and it is the vocabulary everyone uses — but it does mean that anyone invoking “the ICH limits” for a peptide is borrowing a framework rather than applying one.
The framework that does apply, and its inverted shape
In May 2021 FDA’s Center for Drug Evaluation and Research published ANDAs for Certain Highly Purified Synthetic Peptide Drug Products That Refer to Listed Drugs of rDNA Origin, covering five peptides: glucagon, liraglutide, nesiritide, teriparatide and teduglutide. It is the clearest United States statement of peptide-specific impurity expectations, and its construction is the opposite of a purity floor.
The guidance states that applicants “should identify each peptide-related impurity that is 0.10 percent of the drug substance or greater,” and that a proposed generic “does not contain any new specified peptide-related impurity that is more than 0.5 percent of the drug substance.” It adds a comparative criterion: for each impurity present in both the proposed product and the reference product, the level “is the same as or lower than that found in the RLD.”
Three features are worth extracting. The identification threshold is 0.10% — the same figure as ICH Q3A, arrived at for a class ICH Q3A excludes. The controlling limit is a ceiling on the worst individual new impurity, not a floor on total purity. And the decisive comparison is impurity-by-impurity against a known reference, not percentage-against-percentage.
A specification of that shape cannot be satisfied by any single number, however high. Two lots can both report 99.0% and differ completely in whether either would pass.
Why the arithmetic inverts for peptides
Here is the part that makes peptide impurity thresholds behave differently from small-molecule ones, and it is arithmetic rather than opinion.
Solid-phase synthesis generates impurities that are structurally close relatives of the target — deletion sequences missing one residue, insertion sequences carrying an extra one, incompletely deprotected species, oxidation and hydrolysis products, isomers, dimers. They are numerous, individually small, and chemically similar to the peptide itself. Three recent characterization studies show the scale:
- Li and colleagues, using liquid chromatography with high-resolution mass spectrometry, identified and quantified 18 structurally related impurities in a synthetic oxytocin study material, totaling 31.1 mg/g with an expanded uncertainty of 1.7 mg/g (Analytical and Bioanalytical Chemistry, 2021; 413:1861–1870). That is a little over 3% by mass, distributed across eighteen species.
- Cheng and colleagues characterized 23 structurally related impurities above 0.1 mg/g in a China Pharmacopoeia thymalfasin standard, spanning deamination, amination, succinimide formation, amino-acid insertion and deletion, dimers and isomers (Analytical and Bioanalytical Chemistry, 2022; 414:8035–8045).
- Huo and colleagues characterized 33 impurities in Cbf-14, a designed fourteen-residue antimicrobial peptide — one process-related and thirty-two degradation products, comprising 15 hydrolysates, 5 isomers, 4 acetylated isomers, 2 aldimine derivatives and 6 oxidized species, with 7 independently synthesized to confirm the assignments (Analytical and Bioanalytical Chemistry, 2022; 414(22):6485–6495).
Now apply the thresholds. Suppose a peptide lot reports 98.6% chromatographic purity, leaving 1.4% of detector area attributable to something other than the target. If that 1.4% is distributed across roughly twenty-five species in the pattern these studies describe, the average individual impurity sits near 0.056% — above the 0.05% reporting threshold and below the 0.10% identification threshold.
The consequence is uncomfortable and it is structural, not a criticism of any particular supplier. A peptide can be simultaneously 98.6% pure and carry two dozen distinct species none of which the framework obliges anyone to identify. Purity and characterization come apart. Raising the purity number does not close the gap — it narrows each slice further and pushes more of them below the identification threshold, not fewer.
This is also why the international metrology work on peptides corrects explicitly for impurities rather than inferring them. The CCQM-K115.b key comparison on synthetic oxytocin assigned peptide-related impurities a value of 31.6 mg/g with an expanded uncertainty of 1.4 mg/g as a distinct measurand, separate from the peptide content itself.
What “impurity” is measured against
One more distinction collapses inside a single percentage. Chromatographic area-percent purity compares peak areas within one chromatogram. It says nothing about material in the vial that the detector cannot see — water, counterion, residual solvent, excipients — because those never become peaks.
That is why the 1.4% above is described as “detector area attributable to something other than the target,” not “1.4% of the vial contents.” The two are different quantities, and only the first is what an HPLC-UV purity figure reports. A batch can be 99% pure by area and considerably less than 99% peptide by mass.
What this means for a research certificate
Nothing in the regulatory frameworks above binds a research-grade supplier. Research chemicals are not drug substances and no ANDA is being filed. The frameworks are useful here as a vocabulary for what a certificate discloses, and by that measure the honest statement of what a standard research COA supports is narrow:
- It supports a reporting-level claim: peaks above the method’s threshold were present in these proportions.
- It does not support an identification-level claim about the minor species, unless the certificate carries a mass-spectrometry section that names them.
- It does not support a qualification-level claim about any of them, and no research certificate in this market does.
Our own TB-500 10MG batch certificate is a fair example to hold to that standard. Report DBAV-TB500-10-051226, issued by Testides Analytical, reports 98.62% by HPLC-UV against a 10 mg stated fill, with a measured content mass and fill accuracy on the same document. That is a named laboratory, a report identifier, a date, a method and a content figure — which is what makes the number checkable rather than asserted.
It is also HPLC-UV only. There is no mass-spectrometry section on it, so it reports chromatographic purity and content mass and does not identify the minor species behind the remaining 1.38%. We would rather write that sentence than publish a purity figure and let it imply an impurity table that does not exist.
The practical position
Ask what the certificate reports rather than what percentage it reports. A number above a reporting threshold is a measurement; a named impurity above an identification threshold is characterization; a qualified impurity is a safety judgment. Most research certificates in this category, ours included, deliver the first and are silent on the second and third — and a supplier that says so is giving you more usable information than one advertising a higher number with the same silence underneath it.
All materials described here are supplied for research use only. They are laboratory chemicals intended for in vitro and preclinical research conducted by qualified personnel. They are not for human use, not for veterinary use, and not for diagnostic or therapeutic application. Nothing on this page is guidance for administration to people.
References
- ICH Q3A(R2), Impurities in New Drug Substances, International Council for Harmonisation. Scope exclusions and Attachment 1 threshold table.
- U.S. Food and Drug Administration, Center for Drug Evaluation and Research, ANDAs for Certain Highly Purified Synthetic Peptide Drug Products That Refer to Listed Drugs of rDNA Origin: Guidance for Industry, May 2021.
- Li M, Josephs RD, Daireaux A, et al. Identification and accurate quantification of structurally related peptide impurities in synthetic oxytocin by liquid chromatography–high-resolution mass spectrometry. Analytical and Bioanalytical Chemistry, 2021; 413:1861–1870. DOI 10.1007/s00216-021-03154-5.
- Cheng Y, Wu P, Kan Y, Li M, Li H. Identification and determination of structurally related peptide impurities in thymalfasin by liquid chromatography–high-resolution mass spectrometry. Analytical and Bioanalytical Chemistry, 2022; 414:8035–8045. DOI 10.1007/s00216-022-04336-5.
- Huo Y, Xu K, Lu Y, Ma L, Zhou C, Hang T, Song M. Characterization of structurally related peptide impurities using HPLC-QTOF-MS/MS: application to Cbf-14, a novel antimicrobial peptide. Analytical and Bioanalytical Chemistry, 2022; 414(22):6485–6495. DOI 10.1007/s00216-022-04205-1.
- CCQM-K115.b, Key Comparison Study on Peptide Purity — Synthetic Oxytocin, final report, Bureau International des Poids et Mesures and National Institute of Metrology (Beijing), 2020.