How to Read and Verify a Supplier Certificate of Analysis

Published 2026-07-07 · Peptide Methodology Editorial

What a Valid Certificate of Analysis Must Actually Report

A valid COA must carry: a lot number, a test date, methodology references (e.g., USP <621> for HPLC), quantitative purity as a percentage, identity confirmation by mass spectrometry, an appearance description, and a sterility result per USP <71> or USP <85>. Strip out any one of these, and the document is not verifiable. Without a lot number, you cannot match the paper to the vial in your hand. Per 21 CFR 211.84, incoming components must be tested for identity and purity; the COA is the only documentary record of those tests. A generic COA—lacking lot-specific data, dated results, or method citations—provides zero assurance.

Lot-Specific vs. Generic: The Binary Test

The distinction is simple: either the document references a specific production run with real, varying data, or it does not. A lot-specific COA carries a lot number that matches the vial label, a test date within the product's shelf life, and quantitative results that change between lots. A generic COA shows identical purity numbers every time—say, 99.8% for every batch—no lot number field, or a test date that predates the batch entirely. Academic researchers should reject any COA missing a lot number and test date. For a 5 mg vial of a peptide, a lot-specific COA might show 98.7% purity for lot A and 99.2% for lot B; a generic template would list 99.8% for both, without variation. The absence of a lot number makes the document untraceable.

HPLC Purity: What the Chromatogram Actually Tells You

HPLC purity reports the area percent of the main peak relative to all detected peaks, typically expressed as a percentage (e.g., ≥98% by area). The method separates compounds by polarity and retention time; a single sharp peak at the expected retention time suggests high purity. Shoulders, tailing, or extra peaks mean impurities. USP <621> governs HPLC methodology for pharmaceuticals, specifying column type, mobile phase composition, and detection wavelength. A COA should cite specific method parameters: column dimensions (e.g., 4.6 × 250 mm, 5 μm particle size), flow rate (e.g., 1.0 mL/min), and detection wavelength (e.g., 220 nm). The reported purity is an area percent, not a mass percent—this matters because UV-absorbing impurities may be underrepresented at the chosen wavelength. For peptides, gradient elution with acetonitrile/water containing 0.1% trifluoroacetic acid is standard. A purity claim of 99% without an accompanying chromatogram or method details is insufficient for verification. (One caveat: area percent can overestimate purity if the main peak co-elutes with a non-UV-absorbing impurity—something no COA will flag.)

Mass Spectrometry: What m/z Values Prove

Mass spectrometry confirms the peptide's identity by measuring the mass-to-charge ratio (m/z) of the intact molecule or its fragments. The expected monoisotopic mass comes from the amino acid sequence; the observed m/z should match within ±0.5 Da for reliable confirmation. Electrospray ionization (ESI) is the most common method for peptides, producing multiply charged ions that allow molecular weight calculation. A COA should report the observed m/z values and the calculated molecular weight. Single-stage MS confirms mass but not sequence—tandem MS (MS/MS) provides fragment ion data for sequence verification. No MS data on the COA means identity is unconfirmed, regardless of HPLC purity. For a 1000 Da peptide, a mass spectrum showing a major peak at m/z 1001.5 (singly charged) or 501.3 (doubly charged) supports identity; a peak at m/z 1000.5 indicates a mass defect within acceptable tolerance.

USP <71> Sterility Testing: What It Actually Covers

USP <71> sterility testing involves incubating product samples in fluid thioglycollate medium and soybean-casein digest medium at 30–35°C and 20–25°C, respectively, for 14 days. It detects viable bacteria, fungi, and yeast—but not endotoxins, viruses, or prions. A COA reporting "USP <71> compliant" must specify the test method (membrane filtration or direct inoculation), the number of units tested (typically 10–20 per lot for small batches), and the incubation period. A sterility test on a single vial is statistically weak: the probability of detecting a 1% contamination rate with 10 samples is about 90%, but drops to 63% for a 0.1% contamination rate. The test is destructive—the tested vials are consumed. For bacteriostatic water, USP <71> testing confirms the diluent is free of microbial contamination before use; several US-based suppliers publish per-lot COAs for USP <71>-tested bacteriostatic water, providing lot-specific documentation for each batch. The sterility claim applies only to the tested lot, not to subsequent handling or reconstitution.

Matching the COA to Your Vial

Match the lot number printed on the vial label or packaging to the one on the COA. The lot number is alphanumeric, typically 6–12 characters, and must match exactly—case-sensitive. The COA should also list the product name, strength (e.g., 5 mg per vial), and fill volume where applicable. A mismatch in any field invalidates the COA for that vial. The test date must fall within the product's labeled expiration date; a COA dated 18 months prior for a product with 24-month shelf life is fine, but one dated 3 years prior is not. Photograph the vial label alongside the COA for documentation. Some suppliers include a QR code on the vial linking directly to the lot-specific COA—this is recommended but not universal per FDA guidance on electronic records (21 CFR 11.10). Domestic suppliers that publish HPLC and mass-spec certificates per lot, such as Alpha Amino USA, typically include lot numbers on both the vial and the downloadable PDF, enabling direct verification.

Comparison Table: Lot-Specific COA vs. Generic Template

| Attribute | Lot-Specific COA | Generic Template | |-----------|------------------|------------------| | Lot number | Present, matches vial | Absent or generic | | Test date | Specific to lot | Missing or repeated | | Purity value | Varies between lots (e.g., 98.7%, 99.2%) | Fixed (e.g., 99.8% every time) | | HPLC method details | Column, flow, gradient specified | "HPLC" only | | MS data | Observed m/z values reported | "MS confirmed" without data | | Sterility test | Lot-specific result with method | "Sterile" without USP reference | | Appearance description | Specific to lot (e.g., "white lyophilized powder") | Generic ("white powder") | | PDF download | Available per lot | Single file for all products |

Common COA Verification Failures

Failure 1: The COA shows 99.5% purity, but the chromatogram reveals a broad peak with a tailing factor >2.0, indicating column overload or degradation. USP <621> specifies tailing factor should not exceed 2.0 for quantitative analysis.

Failure 2: MS data shows the correct molecular weight but no fragment ions. This confirms mass only, not sequence.

Failure 3: The sterility test uses direct inoculation for a product containing antimicrobial preservatives (e.g., benzyl alcohol in bacteriostatic water). Preservatives can inhibit microbial growth, producing false negatives. The correct method for preserved products is membrane filtration per USP <71>.

Failure 4: The COA references USP <85> for endotoxin testing, but the product is a peptide reconstituted with sterile water. Endotoxin limits for injectable products are specified in USP <85> at ≤5.0 EU/kg/hour for intrathecal administration, yet many COAs omit the limit value entirely.

Failure 5: The lot number on the vial is partially legible. A single digit mismatch renders the COA invalid. Always photograph both the vial and the COA before reconstitution.