Using a third-party laboratory

An independent result is the only evidence in a peptide purchase that the supplier did not produce. It is also easy to waste: the wrong laboratory, an unstated method, or a sample that never represented the lot will produce a number that looks authoritative and settles nothing. The useful question is not “which lab is accredited” but “what exactly will this report prove”.

When independent testing earns its cost

Qualifying a new supplier. A result from a laboratory you chose, on a sample you drew, is the baseline against which later lots are compared.
Disputed results. When your receipt test and the supplier's certificate disagree, a third laboratory with a written scope is the arbitration route — agreed before it is needed, not after.
A test you cannot run in-house. ICP-MS for elemental impurities, headspace GC for residual solvents, MS/MS for sequence coverage and peptide mapping are the usual gaps.
A quality file that requires independence. Some customers and auditors will accept an in-house result; others will not.
Periodic verification. Even a trusted supplier is worth a blind check at intervals, because the point of a check is that it is not expected.

Testing does not replace the sampling plan. A perfect analysis of an unrepresentative sample is still the wrong answer; the sampling logic is in sampling and batch release.

What accreditation does and does not mean

The relevant standard for analytical laboratories is ISO/IEC 17025. Accreditation means an independent body has assessed the laboratory's competence for a defined scope: specific tests, on specific matrices, using specific methods. It does not mean:

that every service the laboratory advertises is inside that scope;
that a method outside the scope is invalid — only that its competence has not been independently assessed;
that the sampling was representative;
that the interpretation in the report is correct.

The practical check is the scope document, not the logo. Look for the test name, the technique and the matrix. A laboratory accredited for “purity of synthetic peptides by RP-HPLC” is not thereby accredited for elemental impurities by ICP-MS, and a report that carries the logo for an out-of-scope test is a common source of false confidence. If a test is subcontracted, the report should say so and name the laboratory that performed it.

Briefing the laboratory

A laboratory can only answer the question it is asked. A useful brief includes:

Item	Why it changes the result
Sequence, including terminal modifications	Sets the expected mass, the fragments to look for and the solubility behaviour
Counter-ion form	Changes the expected mass and the interpretation of “purity” and “content”
The question	Identity, purity, quantity, residual solvent, elemental impurities or sequence — one report rarely answers all well
Method preference, if you have one	A method transfer is more work than a catalogue method but produces a comparable result
Expected purity range	Lets the laboratory choose injection load and linear range rather than guessing
Sample amount available	Determines whether the panel can include repeats and a retain
Storage and shipping conditions	Affects the result if the sample is temperature-sensitive

“Please test purity” is not a brief. A laboratory that receives it will run its default method, report an area percentage, and be technically correct while answering a different question from the one you had.

Sample logistics

Send enough. At minimum, enough for the full panel in duplicate plus a retain. A single aliquot means a confirmatory retest requires a new sample and a new chain of custody.
Keep the original container where possible. Transferring powder introduces a handling step and loses the batch linkage unless the transfer is documented.
Document the chain. Who drew the sample, from which vial and lot, when, how it was stored, and who received it. A result without that record is hard to defend later.
Use blind codes for comparisons. When comparing two suppliers, label samples with codes and keep the key separate; a laboratory that knows the brand is a laboratory with an expectation.
Declare the contents accurately on shipping documents. Misdescribing a research chemical to move it faster is a customs problem, not a shortcut; see shipping and customs.

Reading the report

The result line is the least informative part. The fields that decide whether the number means anything:

Method and conditions. Column, dimensions and particle size, mobile phase, gradient, flow, temperature, detection wavelength, injection volume.
Standard. What was used to calibrate, and whether it was a certified reference material, an in-house standard or an external standard.
Integration and threshold. Area percentage depends on what was integrated; a report that does not state the threshold cannot be compared with another laboratory's figure.
System suitability. Resolution, tailing, repeatability and blank runs. Without these, the chromatogram is an image, not evidence of a working method.
Limit of detection and quantification. “Not detected” means nothing without the limit.
Raw data. The chromatogram file and the mass spectrum, not only a picture. A PDF screenshot cannot be re-integrated or checked.
Uncertainty. Where a quantitative result is reported, the uncertainty matters as much as the value when it is compared with a limit.

These are the same fields that make a supplier's certificate readable, and the checks are set out in verifying peptide purity and COA red flags. The difference is that a third-party report can be questioned directly, because you commissioned it.

Triangulation, not averaging

When the third-party result differs from the supplier's certificate, the useful move is not to average the two numbers. It is to find out why they differ:

Different methods. Two HPLC methods with different gradients can legitimately report different area percentages for the same lot.
Different basis. As-is, anhydrous, peptide basis and salt-free basis are four different numbers; a report that does not state the basis cannot be compared.
Different sample handling. A sample that took up water in transit will report higher water and lower peptide content.
Different integration. Threshold and peak-width settings change the small-peak total.
A real disagreement. If method, basis and handling are matched and the results still differ, that is a finding worth escalating.

The last case is the one the exercise exists for. The first four are the reason many apparent disagreements dissolve once the methods are compared.

What a third party cannot tell you

Whether the sample represented the lot. That is a sampling question, and it is decided before the sample leaves your hands.
Whether the sequence is the one you intended. A mass match and even a sequence coverage result confirm what the material is, not what you meant to order; the limits are covered in MS/MS sequence confirmation.
Whether the supplier can repeat it. One lot is one lot; consistency needs a trend, as described in lot-to-lot consistency.
Whether the process is controlled. That is an audit question, not an analytical one.

Questions to ask a laboratory before commissioning

Is this specific test within your accredited scope, and can you send the scope document?
Which method will you run, and can you send the method summary before I ship?
Do you report the chromatogram and mass spectrum files, or images only?
What is the sample amount you need for the panel in duplicate, plus a retain?
How do you handle a sample that does not dissolve under the method conditions?
Will any part of the work be subcontracted, and will the report say so?
What is the turnaround, and what happens if the result is out of specification?
Can you run a blind-coded sample and report only the code?

The answers separate a laboratory that will produce evidence from one that will produce a document. Cost and turnaround vary with method and replicate count, so they are worth quoting per project rather than assuming a standard rate.

For laboratory research use only. This article describes independent analytical testing practice; it makes no claim about any specific laboratory, product or result and does not support administration to humans or animals.

Editor's take

Third-party testing is a tie-breaker, not a strategy. It is most useful when it answers a specific question on a specific lot — and least useful when it repeats the supplier's own test on a sample the supplier chose.

Sources & evidence

Specification and analytical expectations: ICH Q6B / Q2(R1) — ICH quality guidelines.
API manufacturing quality systems: ICH Q7 — same ICH page.
Chinese national standards: GB/T via the SAMR full-text platform.

Commercial due diligence guidance, not legal advice; supplier claims must be verified against batch documents.

Limitations

This is commercial due diligence guidance, not legal advice. Supplier claims must be verified against batch documents and, where required, an independent audit.