How to read a certificate of analysis: what each test establishes, and what it leaves open

Last reviewed 9 August 2026

A certificate of analysis is a test report. It records what a named laboratory measured, on a named sample, on a named date, by a named method. A certificate that omits any one of those four leaves a corresponding question open.

Two numbers dominate the format: a purity percentage from high-performance liquid chromatography, and a mass from mass spectrometry. Both are real measurements. Neither establishes what it is commonly read as establishing. Three documents do the structural work below — ICH Q7, which contains the fullest published description of a certificate of analysis; ICH Q6A, which sets out what a specification consists of; and ISO/IEC 17025, which governs how a laboratory reports a result.

What the document is supposed to contain

ICH Q7 section 11.4 is the yardstick. Authentic certificates should be issued for each batch on request; the material name, grade, batch number and date of release should appear; certificates should be dated and signed by authorised personnel of the quality unit and show the name, address and telephone number of the original manufacturer. On the results themselves:

The Certificate should list each test performed in accordance with compendial or customer requirements, including the acceptance limits, and the numerical results obtained (if test results are numerical).

Three requirements are doing work there. Each test performed — the document is a list, and a short list is a statement about how little was measured. The acceptance limits — a result appears beside the criterion it was judged against. The numerical results obtained — a number, not a verdict. ICH Q3A says the same: Quantitative results should be presented numerically, and not in general terms such as “complies”, “meets limit” etc. A purity row reading Conforms withholds the measurement while keeping the appearance of having reported it.

Q7 is a manufacturing standard, and laboratory research material is not made under it. That is what makes it a useful yardstick rather than an unfair one: it is the clearest published account of what the document is for, so it renders omissions legible.

The header block

The identifying fields carry more of the document’s value than the results do, because a result with no sample attached to it is not evidence about anything in particular. ISO/IEC 17025:2017 governs this at clause 7.8.2.1, requiring a report to contain enough information to minimise any possibility of misunderstanding or misuse.

Chromatographic purity, and what area percent is a percentage of

The purity figure is almost always a chromatographic area percent: the integrated area of the main peak divided by the summed area of every peak the detector recorded. Detection is by ultraviolet absorbance, conventionally near 210–220 nm, where the peptide bond absorbs. The number is a ratio of detector responses, in one chromatogram, at one wavelength, under one separation.

The assumption inside the number

Area normalisation assumes that a given quantity of any species produces the same detector response. ICH Q3A names this openly, stating that procedures for estimating impurities can be based on analytical assumptions (e.g., equivalent detector response), and adding that where response factors are not close the practice remains acceptable only if a correction factor is applied or the impurities are being overestimated rather than under-reported.

For peptides the assumption is measurably false. Kuipers and Gruppen (2007) determined molar extinction coefficients at 214 nm for the twenty amino acids and for the peptide bond, reporting the peptide bond at 923 M⁻¹ cm⁻¹, tryptophan at approximately thirty times that value, phenylalanine, tyrosine and histidine at approximately six times, and methionine at approximately the same value as the peptide bond itself. Response therefore tracks the number of peptide bonds and the aromatic composition of whatever is eluting. A truncated sequence carries fewer peptide bonds and, if the residue lost was a tryptophan, far less absorbance — so it contributes a smaller peak than its mass warrants, and the main peak’s share rises accordingly. Two categories are absent from the arithmetic altogether: anything that does not absorb at the detection wavelength, and anything that does not elute under the gradient.

What it is not a percentage of

A lyophilised solid is peptide, plus counter-ions carried through from synthesis and purification, plus residual water, plus residual solvents. Chromatographic purity says nothing about any of them, because each is measured by a different method — counter-ion content by ion chromatography, water by Karl Fischer titration, residual solvents by gas chromatography.

The scale of the difference is documented in the reference-material literature. Melanson and colleagues (2018) assigned a purity value to a candidate certified reference material of angiotensin II by combining quantitative NMR, amino acid analysis by LC-MS/MS, and a mass balance in which the trifluoroacetate counter-ion was quantified by ¹⁹F-qNMR. They reported the counter-ion at approximately 25% of total mass and a final assigned purity of 691 ± 9 mg/g. McCarthy and colleagues (2023) set out that mass balance, in which chromatographic area is one factor and the summed weight-for-weight contributions of counter-ions, residual solvents and inorganic material are the other. A chromatographic purity in the high nineties and a mass fraction near 69% are not in conflict; they answer different questions, and only one is normally printed. The two are separated on the companion page on area percent and net peptide content.

Mass spectrometry, and what a conforming mass establishes

The second number is a measured mass reported against a calculated one. Two details decide whether the comparison means anything: whether the figures are monoisotopic or average masses, and whether the observed value is a deconvoluted neutral mass or a raw mass-to-charge ratio at a stated charge state. Monoisotopic mass is computed from the lightest isotope of each element and average mass from natural isotopic abundances; the two diverge as molecular size increases, so a certificate silent on which it reports has left a systematic difference unresolved.

A conforming mass establishes that the material has the elemental composition expected of the target sequence. Chrone and colleagues (2024) describe mass spectrometry as an optimal method for evaluating the authenticity and integrity of a synthetic peptide, precisely because the sequence is already known and the analysis is directed at confirming it.

What it cannot establish follows from arithmetic. Substituting a D-amino acid for its L-form changes no atoms, so formula and both masses are identical; leucine and isoleucine share the formula C₆H₁₃NO₂; transposing two residues leaves composition untouched. These are not hypothetical categories. D’Hondt and colleagues (2014) catalogue the impurities arising in solid-phase synthesis — deletion and insertion of residues, diastereomeric impurities from racemisation during deprotection, protection adducts, oxidation products, oligomeric species and unwanted counter-ions such as trifluoroacetate — alongside degradation routes including β-elimination and diketopiperazine, pyroglutamate and succinimide formation. Lian and colleagues (2021) regard structural isomers and peptide epimers as a distinct analytical problem requiring dedicated methods rather than a mass check.

ICH Q6A states the general principle: identification should discriminate between compounds of closely related structure likely to be present, and Identification solely by a single chromatographic retention time, for example, is not regarded as being specific. Two procedures separating on different principles, or a combined technique such as HPLC/UV diode array, HPLC/MS or GC/MS, is what the guideline regards as generally acceptable. The calculation is on the companion page on mass spectrometry confirmation.

The distance between a certificate and a specification

ICH Q6A defines four universal tests applicable to all new drug substances: description, a qualitative statement of state and colour; identification; assay, a specific stability-indicating procedure determining content; and impurities, covering organic and inorganic impurities and residual solvents. Water content by a water-specific method, inorganic impurities and microbial limits appear as further specific tests.

Peptides sit awkwardly in that framework, and the guidelines say so. ICH Q3A opens by excluding them: The following types of drug substances are not covered in this guideline: biological/biotechnological, peptide, oligonucleotide, radiopharmaceutical, fermentation product and semi-synthetic products derived therefrom, herbal products, and crude products of animal or plant origin. The reporting, identification and qualification thresholds widely assumed to sit behind a peptide purity figure do not, as a matter of scope, reach peptides.

USP wrote a general chapter into that gap. ⟨1503⟩ Quality Attributes of Synthetic Peptide Drug Substances addresses peptide content and assay, related-substance impurities, counter-ion content, water content, residual solvents, elemental impurities, microbiological contamination and bacterial endotoxins, with ⟨1504⟩ covering starting materials, and procedural chapters for acetic acid and trifluoroacetic acid in peptides, amino acid analysis and bacterial endotoxins.

Set that list beside a typical research certificate and the shape is clear: two attributes reported, the remainder unmentioned. Silence is not a negative result. An attribute absent from a certificate was not tested, which is a different statement from tested and found acceptable, and neither is the same as tested and found absent. Endotoxin and sterility are the sharpest instance, covered on the page on what research grade does not mean.

Lot-to-certificate traceability

A certificate is evidence about the sample a laboratory received. Connecting it to material in a container requires an unbroken chain: the lot marked on the container, the lot recorded on the certificate, the sample the laboratory logged in, and the party who selected and submitted it. ICH Q7 anticipates the awkward case, where the certificate is issued not by the original manufacturer but by someone downstream:

If new Certificates are issued by or on behalf of repackers/reprocessors, agents or brokers, these Certificates should show the name, address and telephone number of the laboratory that performed the analysis. They should also contain a reference to the name and address of the original manufacturer and to the original batch Certificate, a copy of which should be attached.

Three identities are named there and kept distinct: the laboratory, the original manufacturer, and the party issuing the new document. A certificate that does not name all three leaves the question of who stands behind the result open. The MHRA Inspectorate has made the wider point that credentials and authorisations are themselves forgeable, and that independent checks belong in a purchaser’s own procedures rather than in the documents supplied to them (Krasteva and Brown, 2020).

Questions that separate a lot-specific certificate from a generic one

1. Does a lot number appear, and does it match the container? One common to every certificate a supplier issues is a specification sheet formatted as a test report.
2. Is there a date of analysis, after the date of manufacture? A certificate predating the material describes a previous lot.
3. Is a laboratory named, with an address? A logo is not an identity. A named laboratory can be contacted.
4. Whose laboratory is it? The manufacturer’s quality unit, a contract laboratory engaged by the manufacturer, and one engaged by the seller are all legitimate and none is equivalent to another.
5. Are the chromatogram and mass spectrum reproduced, with axes, retention times and an integration table? A purity figure with no chromatogram cannot be checked against itself.
6. Is the method stated? Column, mobile phase, gradient, flow rate, detection wavelength, volume loaded. Without the wavelength, an area percent is not comparable with any other area percent.
7. Are the results numerical, with acceptance limits beside them? Per ICH Q7 and Q3A both: a number, against a criterion.
8. Does the document say what was not tested? The most informative certificates list every attribute in the specification, including those marked not performed.
9. Who selected the sample? Independence of the analysis is not independence of the sampling. A laboratory reports on the material it was sent.

What third-party tested asserts, and what it does not

The phrase has no defined meaning. At its strongest it denotes analysis by a laboratory with no commercial interest in the outcome, working to a documented method, whose competence has been independently assessed — in the United Kingdom, accreditation to ISO/IEC 17025 assessed by UKAS. Accreditation is granted against a schedule listing the specific tests and materials a laboratory has been assessed for, so a laboratory can hold accreditation and still perform a given analysis outside its accredited scope. The schedule, not the letterhead, settles that.

At its weakest, the same phrase denotes a report obtained by a seller, on a sample the seller chose, from a laboratory the seller paid, for a lot the reader cannot connect to the container in front of them. Both uses are in circulation and the wording alone does not distinguish them. The nine questions above do.

NovoVita’s position

NovoVita’s published statement is a floor rather than a point value, third-party tested at greater than 99% purity, because a floor holds for every lot where a figure quoted from one certificate holds for one. The full position is set out on the page on NovoVita’s certificate position. The two diluents in the catalogue carry no purity figure at all, since a peptide purity specification has no meaning applied to water.

Purity is a statement about composition. It is not a statement about effect.

References

1. International Council for Harmonisation. Good Manufacturing Practice Guide for Active Pharmaceutical Ingredients, Q7. Step 4 version dated 10 November 2000. Section 11.4, Certificates of Analysis.
2. International Council for Harmonisation. Specifications: Test Procedures and Acceptance Criteria for New Drug Substances and New Drug Products: Chemical Substances, Q6A. Step 4 version dated 6 October 1999. Sections 3.2.1 and 3.3.1.
3. International Council for Harmonisation. Impurities in New Drug Substances, Q3A(R2). Step 4 version dated 25 October 2006. Sections 1, 4 and 5.
4. ISO/IEC 17025:2017, General requirements for the competence of testing and calibration laboratories, clause 7.8, Reporting of results; clause 7.8.2.1, common requirements for reports.
5. United Kingdom Accreditation Service. Laboratory accreditation — ISO/IEC 17025, and published schedules of accreditation.
6. United States Pharmacopeia. General Chapters ⟨1503⟩ Quality Attributes of Synthetic Peptide Drug Substances; ⟨1504⟩ Quality Attributes of Starting Materials for the Chemical Synthesis of Therapeutic Peptides; ⟨503⟩ Acetic Acid in Peptides; ⟨503.1⟩ Trifluoroacetic Acid (TFA) in Peptides; ⟨1052⟩ Biotechnology-derived Articles — Amino Acid Analysis; ⟨85⟩ Bacterial Endotoxins Test.
7. Kuipers BJH, Gruppen H. Prediction of molar extinction coefficients of proteins and peptides using UV absorption of the constituent amino acids at 214 nm to enable quantitative reverse phase high-performance liquid chromatography-mass spectrometry analysis. J Agric Food Chem. 2007;55(14):5445–51. PMID 17539659. DOI 10.1021/jf070337l. Analytical measurement.
8. Melanson JE, Thibeault MP, Stocks BB, Leek DM, McRae G, Meija J. Purity assignment for peptide certified reference materials by combining qNMR and LC-MS/MS amino acid analysis results: application to angiotensin II. Anal Bioanal Chem. 2018;410(26):6719–31. PMID 30143839. DOI 10.1007/s00216-018-1272-7. Analytical method development, reference material.
9. McCarthy D, Han Y, Carrick K, Schmidt D, Workman W, Matejtschuk P, Duru C, Atouf F. Reference Standards to Support Quality of Synthetic Peptide Therapeutics. Pharm Res. 2023;40(6):1317–28. PMID 36949371. DOI 10.1007/s11095-023-03493-1. Collaborative study / methodology.
10. D’Hondt M, Bracke N, Taevernier L, Gevaert B, Verbeke F, Wynendaele E, De Spiegeleer B. Related impurities in peptide medicines. J Pharm Biomed Anal. 2014;101:2–30. PMID 25044089. DOI 10.1016/j.jpba.2014.06.012. Review.
11. Lian Z, Wang N, Tian Y, Huang L. Characterization of synthetic peptide therapeutics using liquid chromatography-mass spectrometry: challenges, solutions, pitfalls, and future perspectives. J Am Soc Mass Spectrom. 2021;32(8):1852–60. PMID 34110145. DOI 10.1021/jasms.0c00479. Review.
12. Chrone VG, Lorentzen A, Højrup P. Characterization of synthetic peptides by mass spectrometry. Methods Mol Biol. 2024;2821:83–9. PMID 38997482. DOI 10.1007/978-1-0716-3914-6_7. Methods chapter.
13. Krasteva E, Brown P. Qualifications of suppliers and customers online: reliable or fake news? MHRA Inspectorate blog, 21 December 2020. Regulatory publication.