What a peptide test can and cannot tell you

Every later lesson relies on one habit: before you read a number, ask which question it answers. This lesson gives you a method for taking any report apart, line by line, so you can see for yourself what has and has not been shown, and it sets out the questions no single test can answer.

What is a test actually made of?

The ICH guideline Q6B defines a specification as "a list of tests, references to analytical procedures, and appropriate acceptance criteria which are numerical limits, ranges, or other criteria for the tests described" (ICH Q6B). Break that sentence down and every result has four parts:

1. The question. What property is being checked: identity, purity, amount, endotoxin, water and so on.
2. The method. The analytical procedure used to answer it, such as HPLC with a UV detector or mass spectrometry.
3. The output. What the method produces: a trace, a spectrum, a number with units, or a pass or fail at a stated sensitivity.
4. The acceptance criterion. The limit the output is judged against. Without it, "pass" has no meaning.

Most confusion about peptide reports comes from losing one of those parts. A purity figure read as an identity check has lost its question. A "pass" with no limit has lost its acceptance criterion. A milligram figure with no method has lost the means to judge it.

Which questions can a peptide test answer?

The European Medicines Agency's guideline on synthetic peptides gives a non-exhaustive list of tests a peptide substance specification may include: appearance, identification, purity, high molecular weight impurities where relevant, assay or content, counter-ion identity and content, residual TFA, water content, residual solvents, elemental impurities, bacterial endotoxins and microbiological purity (EMA, synthetic peptides guideline, 2025). A buyer's test usually orders only the first few.

Each row is independent. HPLC purity does not name the molecule, a mass spectrum does not count milligrams, and an endotoxin assay ignores the peptide entirely. The course takes the rows in this order because each depends on the one before: identity first, since every other number assumes the right molecule. The methods section covers each technique, starting with HPLC-UV and LC-MS.

Why does the method matter as much as the number?

A method can only see what it was built to see, and only down to a certain level. ICH Q2(R2) asks laboratories to show that a procedure is specific, accurate and precise across a stated range before its results are relied on, and notes: "Where one analytical procedure does not provide sufficient discrimination, a combination of two or more procedures is recommended" (ICH Q2(R2)).

In practice, two reports can print the same word, "purity", and mean different things if one used a validated HPLC method with a stated reporting threshold and the other did not say how the figure was produced. A result without a named method cannot be compared with anything, including a later result on the same vial.

What does a test never answer, however good the laboratory?

A result attaches to the sample analysed, on the analysis date, for the properties tested. It says nothing about vials that were not analysed, what happened to the vial afterwards, or properties nobody ordered: if endotoxin was not tested, the report is silent on it, not reassuring about it. ICH Q6B describes specifications as "one part of a total control strategy" (ICH Q6B), alongside manufacturing controls and stability testing. A buyer's single-vial test is a much smaller slice than that, and should be read as one.

Batches first. The manufacturing guideline ICH Q7 defines a batch as a quantity of material "expected to be homogeneous within specified limits", identified by its own batch number, and expects samples to be "representative of the batch of material from which they are taken" (ICH Q7). So before you rely on any report, match its batch code to the code on your vial. A report with a different code, or no code, is not evidence about your vial.

Time next. The stability guideline ICH Q1A(R2) exists because the quality of a substance "varies with time under the influence of a variety of environmental factors such as temperature, humidity, and light" (ICH Q1A(R2)). A vial left in a hot parcel after analysis is no longer the vial the report describes. The reverse also holds: a low result on a vial that sat in heat for days does not, on its own, show the seller sold poor material.

Sterility last. Even a dedicated sterility test has a built-in limit: the US FDA notes that sterility tests "are limited in their ability to detect contamination because of the small sample size typically used" (FDA, aseptic processing guidance). One clean vial cannot show a batch is sterile, and a report that never ran the test says nothing either way.

What does build a picture of a seller is many results over time, from vials the seller did not choose. Blind testing explains why who picked the vial matters, and how the Index scores vendors shows how repeated anonymous purchases are combined. Neither turns a record into a promise about the next order.

How do you take a report apart? A worked example

The report summary below is invented for teaching and describes no real product or laboratory. Sort each line into question, method, output and limit.

Only two of the five lines carry enough information to interpret, and even those lack acceptance criteria. That is common and proves nothing wrong, but it tells you exactly which questions remain open and what to ask the issuer for. The COA checker runs the same sorting on a real certificate, and the which test tool works the other way: start from your question and it names the test. Next: Identity: why it comes first.

Check your understanding

1. A report says "Purity 99.2%" and nothing else about the molecule. Has identity been shown? No. Purity answers how clean the sample is, not which molecule it is. Identity needs its own method, usually mass spectrometry.
2. A certificate says "Endotoxin: pass". What is missing? The method, the units of the result and the acceptance limit. Without them you cannot tell what was measured or what "pass" was judged against.
3. Your vial's report does not mention sterility. Can you treat that as a clean result? No. A property that was not tested is unknown. Silence on a report is never a pass.