Certificates of Analysis, Peptide Purity, and the Role of HPLC in the Published Literature

A certificate of analysis (COA) for a research peptide is, in practice, a summary of analytical measurements. The two most commonly reported measurements are chromatographic purity and identity/content, and the published literature on both is largely built on high-performance liquid chromatography (HPLC) and HPLC coupled to mass spectrometry. The studies summarised below are analytical and reference-material studies; neither was designed to evaluate outcomes in people, and this guide reports only what the authors measured and in what system. On the content and traceability side, Feng and colleagues described the development of amyloid-beta solution certified reference materials (CRMs), Aβ40 (GBW09874) and Aβ42 (GBW09875), prepared from high-purity amyloid-beta raw material and assigned certified values with uncertainties of 7.58 ± 0.30 and 7.62 ± 0.30 μg g⁻¹ [1]. That work was performed entirely in vitro on solutions and materials, not in animals or people [1]. The authors reported that they used two independent certification strategies — amino acid-based isotope dilution HPLC mass spectrometry (ID-LC-MS) and sulfur-based HPLC isotope dilution inductively coupled plasma mass spectrometry (HPLC-ID-ICP-MS) — and stated this was the first application of these strategies to certify candidate amyloid-beta solution CRMs [1]. The stated motivation was that results from different amyloid-beta assays vary substantially, and that such variation can be addressed by standardising assays through CRMs and an established traceability chain [1]. Two aspects of that study map directly onto what a COA is supposed to communicate: the candidate CRMs were assessed for homogeneity, which the authors reported as good, and for stability, which was demonstrated for at least 5 days at −20 °C and 14 months at −70 °C [1]. The authors described the intended use of these materials as value assignment to secondary calibrators or CRMs with a clinical matrix [1]. On the purity and related-substances side, Smith and colleagues reported a reversed-phase HPLC method for separating and quantifying bovine, porcine, and human insulins and related substances in bulk insulin crystals and in injectables, using a sulfate buffer/acetonitrile mobile phase at 40 °C, developed after studying how temperature affected the separation of components in insulin injectables [2]. The chromatographic measurements themselves were laboratory analyses of drug substance and product, while the comparator method in that regulatory context was an animal assay: the United States Pharmacopeia official potency method described by the authors was a rabbit bioassay measuring depression of blood sugar concentrations [2]. The authors reported that HPLC reduced analysis time to 1/60 of that required for the bioassay and yielded more information about purity than the percent nitrogen determination that was one of the USP official procedures [2]. Comparing results across 40 lots of bulk crystalline insulin using a species-by-species potency/area conversion factor, the authors reported that HPLC estimates generally fell within the 95% confidence interval for combined independent bioassays [2]. Read together, these papers illustrate the two distinct questions a COA touches. One is chromatographic separation and quantitation of a main peak against related substances in a manufactured lot [2]. The other is metrological traceability — whether the number on the certificate is anchored to a characterised reference material with stated uncertainty, homogeneity data, and stability data [1]. Neither study reports on physiological effects in humans, and neither supports conclusions beyond the analytical systems described [1][2].