Stability-Indicating Method: Why a Passing Assay Can Mislead

An assay result of 99.6 per cent tells you the material is strong. It does not tell you whether the method that produced the number was capable of noticing if the material were decaying inside the drum. Those are two different questions, and only one of them is answered by the figure on the certificate. The second question has a name, and it is worth knowing because it is the one that decides whether a re-test date means anything.

The term is stability-indicating, and the industry's own GMP text for this material family states the requirement plainly: test procedures used in stability testing should be validated and should be stability indicating.

What "stability-indicating" actually requires

The idea is not complicated. A stability-indicating method is one that can distinguish the intact molecule from its degradation products, so that a fall in the measured value means degradation rather than assay noise. Everything else about the method follows from that single capability.

Three properties are usually implied, and they are worth separating because they fail independently.

Property	What it means	How it fails in practice
Separation	Degradants elute clear of the main peak	Two peaks co-elute and the assay reads high by exactly the amount lost
Specificity	The detector responds to the parent and not equally to the degradants	A non-specific detector responds to whatever is in the cell, including degradants
Mass balance	Degradation measured by the assay is accounted for by the impurity results	Impurities rise, the assay does not fall, and nobody can say which method is right

Mass balance is the one that gets skipped. It is also the one that catches a method problem the other two would let through, because it forces the two results to agree with each other rather than each to look acceptable alone.

Why this is not an abstract concern on a bile acid

Bile acids are chemically robust, which is exactly why the question gets dropped in this category. But robust is not inert, and the forces that matter are well known.

• Hydrolysis at the side chain and epimerisation at the hydroxyl centres, both of which move the molecule without necessarily changing how much acid is present.
• Oxidation and photodegradation, both of which are documented stability-relevant routes for this class and both of which change the chromophore behaviour rather than the titratable acidity.
• Keto-enol driven interconversion in derivatives that carry a ketone, which matters for the intermediates in this family rather than for the finished acid.

Every one of those routes can produce a degradation product that still carries the carboxyl group. And a method that measures the carboxyl group will not see any of it.

The assay-method question nobody asks

Here is the part that surprises buyers in this category, and it is worth stating with both texts open rather than from memory. The two pharmacopoeial texts release this material on two different kinds of measurement. The USP ursodiol monograph specifies an HPLC assay: mobile phase acetonitrile and water (55:45) adjusted to pH 3.0, an epiandrosterone internal standard, and a measurement taken from a separated parent peak. The European monograph, working from the current European Pharmacopoeia text, releases the same substance on an alkali titration: 0.350 g dissolved in 50 mL of ethanol (96 per cent) that has been neutralised to phenolphthalein, 50 mL of water added, and the solution titrated with 0.1 M sodium hydroxide to a pink endpoint, where 1 mL of the titrant is equivalent to 39.26 mg of the substance.

Read the two descriptions side by side and the difference is not a matter of style. One method separates the molecule from whatever sits next to it and measures the parent. The other adds up everything that neutralises alkali and reports the total, with the endpoint judged by eye against a colour change.

This is not an argument that one pharmacopoeia is careless. It is an argument that the monograph defines the release test and the stability programme defines the assurance, and the two are separate documents owned by separate people. A supplier can be fully compliant with a release test that has no ability to see the thing you are worried about.

What a forced-degradation package should contain

The evidence that a method is stability-indicating is produced once, during development, and then archived. Getting it out of the archive is the practical task, and it goes faster when you know what the package is meant to hold.

• The stress conditions, named and quantified. Acid, base, oxidative, thermal and photolytic stress, each with the concentration, temperature and duration applied. For a bile acid the informative routes are hydrolysis at the side chain and oxidation on the ring system, so a package that lists only heat has not tested the chemistry that matters here. Photostability follows the ICH Q1B design, with a dark control and a measured exposure rather than a claim of exposure.
• The chromatograms, not only the numbers. The demonstration is that the degradant peaks resolve from the parent. A table of percentages without a resolution figure has not shown that the separation works.
• Resolution and peak-purity data from the stressed sample. This is the line that answers the specific question, which is whether the parent peak is clean underneath.
• The mass-balance reconciliation. Assay loss against impurity gain, on the same stressed sample, with the degradants identified well enough to be counted.

A worked example makes the logic concrete. Take a sample stressed under oxidative conditions, and suppose the assay falls from 99.6 per cent to 97.9 per cent, a loss of 1.7 percentage points. If the impurity profile shows one oxidative degradant rising from 0.1 to 1.8 per cent, the two figures nearly agree and the method is behaving as it should. If instead the assay falls by 1.7 points while the impurities rise by 0.1, one of two things is true: the degradant carries no chromophore and the detector cannot see it, or the assay is responding to something the impurity method does not measure. Either way the reconciliation is what exposed it, and neither figure on its own would have.

How to test the claim without access to the laboratory

You do not need to run the method to evaluate whether it is stability-indicating. You need four documents, and all four are the supplier's own.

1. Ask for the forced-degradation summary. A stability-indicating method is demonstrated by stressing the material and showing the degradants resolve. Ask for the stress conditions used and the resolution obtained. A summary that lists conditions but no resolution data has not demonstrated anything.
2. Ask for the peak-purity result from one stressed sample. This is the single cheapest piece of evidence that the main peak is not hiding something underneath it.
3. Ask how mass balance was checked. The good answer names the degradants and shows the totals adding up. The weak answer says the assay was unchanged.
4. Ask what happens to the re-test date if the assay method changes. This is the forward-looking question. A method swap invalidates the stability data collected on the old method unless a bridging study is done, which is why the answer should reference a comparability exercise rather than an assurance. Our note on what a re-test date actually requires sets out the logic; this is the practical face of it.

None of these requests is unusual, and none requires disclosing anything commercially sensitive. If a supplier cannot produce the first two, the stability programme exists on paper and the re-test date is a convention rather than a conclusion.

What this means for the numbers you already hold

An assay figure and an impurity profile are two views of the same batch, taken through two different instruments. When they agree, the batch is behaving. When they disagree — impurities rising while the assay holds flat — you are looking at the signature of a non-specific assay, and no amount of additional batch testing will resolve it. The fix is a method, not a retest.

That is why the related-substances arithmetic deserves to be read as the sensitive half of the pair rather than as the secondary test. On a molecule this stable, the impurity numbers move first and the assay follows or does not follow. Which of the two happens is the evidence about the method.

Frequently asked questions

What does stability-indicating mean?

It means the method can distinguish the intact molecule from its degradation products, so a change in the measured value reflects real degradation rather than assay variation. That requires resolution of the degradants, a detector that does not respond equally to them, and a mass balance that reconciles the assay movement with the impurity results.

Is a titration assay stability-indicating?

Not in the sense the term is used. A titration measures total acid equivalents, and degradation products of a bile acid can retain the carboxyl group, so they can contribute to the titrated value while no longer being the parent molecule. It is a legitimate release test where the monograph specifies it; it is not a demonstration that degradation would be detected.

How do I check the claim without access to the laboratory?

Ask for four things: the forced-degradation summary with resolution data, a peak-purity result from a stressed sample, the mass-balance reconciliation, and the position on what happens to existing stability data if the assay method changes. All four are documents the supplier already holds.

Why is mass balance harder than peak purity?

Peak purity is a property of one injection. Mass balance requires the assay loss and the impurity gain to reconcile across the same sample set, which means the two methods must be calibrated to each other and the degradants must be identified well enough to be counted. It is the check that catches a specificity problem the other two would pass.

Does the pharmacopoeia require a stability-indicating method?

The monograph defines the release test. The stability requirement sits in the GMP framework for active substances, which states that stability test procedures should be validated and stability indicating. The two obligations are separate, and compliance with the first does not discharge the second.

What changes if my supplier switches assay method?

The stability data collected under the previous method does not automatically carry over, because a different method may have different specificity. The defensible route is a comparability exercise on retained samples. If the answer is that the method changed and the re-test date did not, ask what the comparison was.

Where this leaves you

An assay figure is a statement about a measurement, not about the material's future. The question that connects the two is whether the method could see the future arriving. Ask for the forced-degradation evidence before you rely on the re-test date, and read the impurity profile as the test that moves first.

What we release against, and the method behind each certificate line, is set out on our ursodeoxycholic acid page. If you want the method references and forced-degradation summary discussed against your own stability protocol, you can raise it through a quotation request. For what each certificate line proves on its own, see our breakdown of the certificate of analysis.

This article describes analytical methods and stability testing frameworks for quality, regulatory and procurement professionals. It states no therapeutic use, dose or indication for any substance, and no order quantity or lead time. Confirm all requirements against the current versions of the referenced guidelines and pharmacopoeial texts.