Vendor comparison in this sector is difficult for a structural reason: almost every claim a supplier makes about itself is unverifiable from the outside, and the claims that can be checked are rarely the ones being advertised.
A workable evaluation therefore ignores adjectives entirely and asks only what the supplier can produce on request. The questions below are ordered so that a failure early on saves the effort of the rest.
1. Can they produce a lot-specific document?
Ask for the certificate of analysis matching the lot number printed on a vial you actually received. This single request separates suppliers with a real documentation chain from suppliers with a marketing asset.
A representative COA, a COA for a different lot, or a COA with the lot field blank are all the same answer: the documentation is not connected to the product.
2. Who ran the test?
There is nothing wrong with in-house analysis, and there is nothing automatically superior about an outside laboratory. What matters is that the document says which one it was, because the two support different conclusions.
A named third-party laboratory can be contacted independently. An unnamed one cannot, and 'third-party tested' with no laboratory named is not a verifiable statement.
- Is the testing laboratory named on the document?
- Does the laboratory publicly confirm that it performs the analysis in question?
- Is there a lot-lookup mechanism that returns the report, rather than an emailed image?
3. Is the evidence complete enough to interpret?
Purity, identity and net peptide content are three separate results. A supplier presenting one of them as though it covered the others is not necessarily being dishonest, but the reader is left drawing a conclusion the data does not support.
| Provided | Supports | Does not support |
|---|---|---|
| HPLC purity only | How much of the detected material is one component | What that component is; how much of the vial is peptide |
| Mass confirmation only | That a component of the expected mass is present | How much of it there is relative to impurities |
| Both, same lot | A reasonably complete composition picture for that lot | Sterility, endotoxin, or anything not tested |
| A purity number with no method | Nothing that can be checked | Any comparison against another supplier's number |
4. Claims that carry no information
Some of the most prominent signals in this market are unfalsifiable, self-assigned, or describe something other than what the reader assumes.
- Purity figures with no method, no lot and no chromatogram. There is nothing to check.
- 'Pharmaceutical grade' and 'lab grade'. Neither is a defined regulatory category for these materials.
- 'GMP' used without naming the facility, the registration, or who audited it.
- Accreditation logos on the supplier's own site with no registry entry. Accreditation belongs to a laboratory and is verifiable on the accreditation body's public register.
- Site design, checkout polish and review counts. These measure investment in the storefront.
- Aggregate testimonials. They are not analytical evidence and are trivially manufactured.
5. Operational facts worth establishing
- Whether the supplier synthesises, or resells material produced elsewhere. Both are legitimate; they imply different questions about upstream control.
- Where fulfilment actually happens, as distinct from where the company is registered.
- Whether storage and shipping conditions are stated, and whether they match the stability requirements of the material.
- Whether pricing is public. Quote-only pricing is not a red flag on its own, but it removes a comparison axis.
- How corrections and complaints are handled, and whether documentation errors have ever been publicly acknowledged.
6. Record what you could not determine
The most common failure in supplier evaluation is converting silence into a verdict. A supplier that has not published its testing relationships has not thereby failed a check — it has left a field blank, and the honest record says so.
Therapept's own directory follows this rule: every field is documented, self-reported, or explicitly not publicly stated. A comparison built on inferred values is a comparison of inferences.
Sources
- ISO/IEC 17025:2017 — General requirements for the competence of testing and calibration laboratories
- FTC — Health Products Compliance GuidanceSubstantiation expectations for claims made in commerce.
- FDA — Compounding and the FD&C Act sections 503A and 503B
This article is informational. It is not medical, legal or financial advice. Editorial policy · Report a correction
