What a pass leaves out

There is a moment on almost every certificate in the laboratory compound sector where the numbers stop. Identity is matched against a reference spectrum. Purity is printed to a decimal place. Content is reported in milligrams against the label. Then the line that concerns bacterial contamination, the line most likely to ruin a cell culture or skew an assay, ends in one word. Pass.

That word has been accepted for so long that few people stop to ask what it means. We did. The answer is the reason this journal exists, and the reason every certificate Crescent Bioresearch publishes carries the measured endotoxin value beside the limit it was judged against.

01What endotoxin is, and why a laboratory cares

Endotoxin is lipopolysaccharide, a structural molecule from the outer membrane of Gram-negative bacteria. It is released when those bacteria die or divide, and it does not leave with them. Filtration that removes living cells does not remove it, and it survives the ordinary heat that kills them. A preparation can be sterile and still carry endotoxin.

It matters because immune cells are built to notice it. In culture, endotoxin can switch on inflammatory signalling at concentrations far below anything visible, and a model that is reacting to contamination is no longer answering the question it was set up to answer. One published case from a regenerative medicine laboratory traced an endotoxin problem in cell culture to a medium supplement whose certificate carried no endotoxin value at all. The laboratory set its own criterion, below 1.0 EU per mL in the culture supernatant, and resolved the problem by moving to a supplier that reported the value for every lot [7].

Endotoxin is measured in Endotoxin Units, written EU. The standard test uses Limulus amebocyte lysate, an extract of horseshoe crab blood cells that clots in the presence of endotoxin, a reaction characterised by Levin and Bang [8]. The method is set out in United States Pharmacopeia General Chapter <85>, the Bacterial Endotoxins Test [1].

02What a pass actually says

Under USP <85>, a pass has one meaning. The article meets the requirement when its endotoxin concentration is less than the limit specified for it [1]. A pass is therefore a statement about two numbers, the measured value and the limit, and a certificate that prints only the word has removed both of them. Regulated manufacturing works the same way: a result is judged against the limit at a defined dilution, and a lot that fails at that dilution is rejected [2].

Consider what that hides. Against a limit of 40 EU per vial, a lot measured at 13.4 EU passes. So does a lot measured at 39.5 EU. Both of those figures come from our own published certificates, two lots judged against the same 40 EU limit. One carries almost three times the endotoxin of the other. On a certificate that prints only the word, they are identical.

03A limit chosen for someone else

Even when a certificate prints its limit, the limit belongs to a context. In pharmaceutical practice, USP <85> derives the endotoxin limit from a formula tied to how a finished medicine is used [1, 3]. A laboratory compound has no such context.

The limit printed on its certificate was chosen by the supplier or the testing laboratory, and it may have nothing to do with the work the material is bought for. A research entity running a sensitive cell line will hold its own acceptance criterion, often far below the figure on the page. Our own catalog shows how wide the gap can be: eight lots are judged against 40 EU per vial and five against 350 EU per vial, by two different forms of the test. Both kinds pass. They do not pass the same test.

This is the point that matters most. Only the measured value lets a research entity declare pass or fail for its own work. A supplier’s pass answers the supplier’s question. The number answers everyone’s.

04Every form of the test produces a figure

It is sometimes suggested that endotoxin testing only yields pass or fail. The standard says otherwise. The gel-clot assay in USP <85> reports an endotoxin concentration from the endpoint of a dilution series, and when no dilution clots it instructs the analyst to report the result as less than the lysate sensitivity multiplied by the dilution [1, 3, 5]. The kinetic turbidimetric and chromogenic methods read the value directly from a standard curve. Every form of the test can put a number on the page.

That includes the results that read below detection. When four of our lots report < 5 EU per vial, that is the laboratory’s measured result: the vial came in below the assay’s limit of detection, 5 EU per vial, at the dilution tested. It is a figure with a unit and a boundary, and it can be compared with any criterion a laboratory chooses. A missing number is a reporting choice. It is not a limitation of the science.

05A word cannot be checked

A measured value does not sit on a certificate alone. It arrives with the volume the vial was reconstituted in, the sensitivity of the lysate, the dilution, the limit and the report number of the laboratory that ran it. Those figures have to agree with each other. A value that cannot be reconciled with its own sensitivity and dilution stands out. A report number can be taken back to the issuing laboratory and matched against its record.

A single word carries none of that structure. A certificate copied from another lot, edited after the fact, or issued without a test at all can print pass as easily as a genuine one, and nothing on the page would contradict it. That is not an accusation against any supplier. It is a description of what the format allows. A pass-only certificate is not shown to be false. It simply cannot be shown to be true.

We have seen the difference on our own record. One of our lots was tested twice for endotoxin by two different laboratories. The first certificate returned the word pass, twice, with the threshold field left empty. The second, on the same lot, returned 92.7 EU per vial against a limit of 350. Both describe the same vial. Only one of them tells a laboratory anything it can use.

The numerical endotoxin value closes that gap. It turns the endotoxin line from a claim into a record, one that can be read, compared and verified by the people who rely on it.

06What the sector prints today

Before we wrote this, we read the endotoxin line on twenty-five certificates of analysis published by suppliers in the laboratory compound sector. The result is below. Most did not carry a number. Close to half did not carry an endotoxin line at all.

07What a complete endotoxin line carries

A certificate that a research entity can actually use answers six questions on the endotoxin line. If any of them is blank, the line is incomplete.

08What we publish, and why

Every lot Crescent Bioresearch supplies is tested by an independent third-party laboratory for purity and for bacterial endotoxin before it is listed. The full signed certificate is published under the lot number, and the endotoxin line on every signed certificate carries the measured value, the limit, the method, the lysate sensitivity and the volume. Today those values run from below 5 EU per vial to 92.7 EU per vial, and every one of them sits beside the limit it was judged against. When a new lot arrives, its certificate replaces nothing; it is added to the record.

We do this because a certificate of analysis was always meant to be a set of measurements that the reader can check. Somewhere along the way, the endotoxin line became the exception. We think it should be the rule, and we think the laboratories that rely on these materials deserve to see what was measured, not only what was decided.

This is not a premium. It is the floor. A number on the endotoxin line costs a supplier nothing but the willingness to show it.