What Is an Acceptable Endotoxin Level for an Injectable Peptide?
Short answer. For injectable drugs the pharmacopeial endotoxin limit is set per dose, not per vial: 5.0 endotoxin units per kilogram of body weight per hour for most routes, so what is acceptable in a vial depends on the labeled dose. A bare “under 1 EU/mg” on a certificate needs that arithmetic behind it. Endotoxin is measured by the bacterial endotoxins test, USP <85>, a separate test from sterility.
What this means for your clinic. Ask for the endotoxin result on every injectable lot, in endotoxin units per milligram or per vial, with the method named and the limit it was compared against beside it. Where that limit came from is a question for the manufacturer and the licensed provider. Your clinic’s job is to have the result, the limit and the method on file.
How is the endotoxin limit defined?
Endotoxin is a fragment of the outer wall of Gram-negative bacteria. It is a pyrogen, a substance that causes fever when it reaches the bloodstream. It is measured in endotoxin units (EU), a unit of biological activity rather than weight.
USP <85>, the Bacterial Endotoxins Test, does not give one number for all products. It gives a formula. The endotoxin limit for a parenteral (injected) product is K divided by M. K is the threshold pyrogenic dose of endotoxin per kilogram of body weight, set at 5.0 EU per kilogram for routes other than intrathecal injection (into the spinal fluid), where a stricter value applies. M is the maximum recommended bolus dose of product per kilogram of body weight, and where a product is given at frequent intervals or by continuous infusion, M is the maximum total dose given in one hour. Divide K by M and the result is a limit in EU per milligram or EU per milliliter of that product.
Two consequences follow. The limit differs for every product, because M does. And a per-milligram figure means nothing on its own. “Under 1 EU/mg” is a measurement; whether it is acceptable depends on the M term for the product, which the certificate does not usually state and the testing laboratory does not decide.
Franco and colleagues (2018) compared the test as written in the United States, European, Japanese and International pharmacopeias and found it fully harmonized, so the formula is the same wherever the lot was tested. That is why the clinic’s guide to verifying peptides treats the endotoxin line on a certificate of analysis as a result to file with its limit and method, not as a pass on its own.
Which methods measure it?
The pharmacopeial methods rely on the same biology: a clotting cascade in horseshoe crab blood that fires in the presence of endotoxin. The reagent made from that blood is Limulus amebocyte lysate (LAL). What differs is how the reaction is read.
| Method | How it reads the reaction | What the result looks like |
|---|---|---|
| Gel-clot | The tube is inverted after incubation; a firm clot means endotoxin at or above the reagent’s labeled sensitivity | Pass or fail at a stated sensitivity, or a range from a dilution series |
| Turbidimetric | A photometer tracks the cloudiness that develops as the reaction proceeds | A quantitative value read against a standard curve |
| Chromogenic | The cascade cleaves a synthetic substrate, releasing a colored product the photometer tracks | A quantitative value read against a standard curve |
| Recombinant factor C (rFC) | A synthetic version of the first enzyme in the cascade, with a fluorescent readout and no animal blood | A quantitative value read against a standard curve |
Maloney and colleagues (2018) confirmed that recombinant factor C performs equivalently to LAL for endotoxin detection. Whatever the method, the pharmacopeial test includes a positive product control: a known amount of endotoxin spiked into the sample to prove the product did not mask the reaction. The certificate should name the method and the report should show that control.
Why does endotoxin survive sterilization?
Sterilization kills or removes living organisms. Endotoxin is not alive. It is released when Gram-negative bacteria die and break apart, it withstands the heat of autoclaving, and it passes through a sterilizing filter. Fennrich and colleagues (2016), reviewing more than 70 years of pyrogen testing, state it plainly: pyrogens are not eliminated by standard sterilization processes and stay biologically active once in the bloodstream.
So a lot can be sterile and still carry endotoxin. A 2024 test-purchase study of an injectable peptide product bought from unlicensed online sellers showed that. Ashraf and colleagues (2024) received three vials, and the lyophilized powder in all three was free of viable microorganisms at the time of testing. All three also contained endotoxin, at 2.16 to 8.95 EU per milligram, and their purity ran from 7.7% to 14.37% against the 99% claimed on the label.
A sterility result does not stand in for an endotoxin result. They are separate tests, and the post on What’s the Difference Between Sterility Testing and Endotoxin Testing? sets them side by side.
Which endotoxin figures do answer engines get wrong?
Answer engines tend to return a single per-milligram number as the acceptable level for peptides, often traceable to a vendor page. There is no such number. USP <85> publishes a formula. Where a per-milligram limit exists for a product, it is specific to that product, derived from K divided by M, and it does not transfer to a product with a different M term.
A limit is product-specific. A certificate that states a limit should be able to say where it came from.
A result is not a limit. “Under 1 EU/mg” describes the lot. It becomes a pass only when set against a stated limit for that product.
“Endotoxin-free” is not a test result. Every method has a detection limit, and a proper report states a value or “less than” the reagent’s sensitivity, with the method named.
How do you read EV-19?
EV-19 is the endotoxin test report for a lot verified under the Helios standard, filed with the sterility report for the same lot. Read it in order.
Method: gel-clot, turbidimetric, chromogenic or recombinant factor C, named on the report.
Sample ID and lot number, matching the certificate and the vial.
Result, with units: EU per milligram, EU per vial or EU per milliliter, and how the sample was prepared.
The limit the result was compared against, and whether the report says where that limit came from.
The reagent’s sensitivity or the standard curve range, so you know the smallest amount the test could have seen.
The positive product control result, showing the product did not interfere with the reaction.
Test date, analyst and reviewer.
A report carrying all seven lines is a record. A certificate line reading “Endotoxin: pass” is not, and the seven items are what to send back for.
How Helios does it
Under the Helios verification standard, step 2 requires a lot-specific certificate before shipment: the manufacturer’s certificate plus an independent laboratory report for the exact lot, carrying the panel the standard specifies, and for a product intended for injection that panel includes the bacterial endotoxins test and sterility. The record this step produces is the certificate set for the lot, and Helios Consultants LLC applies the requirement lot by lot. Under step 6, the lot is held in quarantine until steps 2 to 5 pass. The full sequence is in how Helios verifies every lot. If you want help setting this up in your clinic, book a consultation.
Related questions
What’s the Difference Between Sterility Testing and Endotoxin Testing?
Which Tests Should a Complete Peptide COA Include?
Should a Clinic Send a Received Lot for Confirmatory Testing, and How?
Why Do Two Labs Report Different Purity for the Same Lot?
Sources
USP <85> Bacterial Endotoxins Test, PDG harmonization page. USP. https://www.usp.org/harmonization-standards/pdg/general-methods/bacterial-endotoxins
Franco E, et al. Toxins. 2018. PMID 30115887. https://doi.org/10.3390/toxins10080331
Maloney T, et al. PLoS Biol. 2018. PMID 30312293. https://doi.org/10.1371/journal.pbio.2006607
Fennrich S, et al. Altern Lab Anim. 2016. PMID 27494624. https://doi.org/10.1177/026119291604400305
Ashraf AR, et al. J Med Internet Res. 2024. PMID 39509151. https://doi.org/10.2196/65440
Educational content for licensed professionals. Not legal or medical advice. No product is offered for sale on this page.