Impurity thresholds: where the common numbers come from

Impurity thresholds: where the common numbers come from — setting out what I have, and where I think it stops being reliable.

I have spent a fortnight trying to pin impurity thresholds down and I want to set out where I have got to, because I suspect the honest answer is duller than the thread this will produce.

What I have: a consistent observation across a small number of cases, collected the same way each time. What I do not have: any controlled comparison, or any reason to think my cases are representative.

The specific question is whether the pattern survives once the obvious confounder is removed. I cannot remove it with what I have.

Oxidation at methionine and tryptophan: adds 16 per oxygen. Usually elutes earlier. Oxidation is common in storage, especially if the solution is exposed to light or if antioxidants are not present.

Written in the hope of being told what I have missed.

Deamidation at asparagine or glutamine adds approximately one dalton and frequently produces a close-eluting pair. It is the impurity most likely to be integrated into the main peak by accident.

The confident answers on impurity thresholds and the well-sourced answers are not the same answers, which is the most useful thing I have learned reading this category.

Coming back to post #3, because the follow-up matters more than the original answer.

Answering the impurity thresholds question as asked, then the question I think is meant. As asked: yes, with the qualification below. As meant: it depends on how the first measurement was taken.

Adding a note of thanks rather than an opinion. I did not know most of that.

Reporting thresholds matter: below a stated threshold, peaks are usually not reported at all. A clean-looking table may reflect a high threshold rather than a clean synthesis.

Take it as a starting point and not as a specification.

An impurity profile that changes between lots is more informative than the total. A stable profile suggests a controlled process; a shifting one suggests something is varying.

Storage-related degradation and synthesis-related impurity look different on a chromatogram. A growing oxidation peak over time is not a manufacturing finding.

What I would want before treating impurity thresholds as settled: the method, the sample, and whether anyone tried to find the opposite result. Two of the three are usually missing.

Trifluoroacetate is a counter-ion rather than an impurity, and it appears in the mass balance rather than in the chromatogram. Conflating the two accounts for several confused threads here.

Take the reasoning and check the arithmetic; I do not always get it right.

Where an impurity is identified rather than merely counted, the certificate is telling you the manufacturer has characterised its own process. That is a meaningful difference in documentation quality.

The honest answer is that it depends, and here is what it depends on.

This follows post #12 rather than contradicting it.

Two questions I would want answered before drawing anything from the impurity thresholds data above: how were the cases selected, and what happened to the ones that dropped out.

Worth separating two things that post #10 runs together.

Comparing impurity profiles across suppliers is much more informative than comparing purity figures, and almost nobody has the documents to do it.