Elemental impurities: from a PDE to a limit you can measure

Two numbers that are constantly confused

An elemental impurity limit is discussed in two units, and they are not interchangeable. The permitted daily exposure, or PDE, is an amount per day: the mass of the element a patient may receive every day for a lifetime. A concentration limit is an amount per gram: the mass of the element in a gram of the material. The first is a toxicological judgement. The second is a line you can test against, and it exists only once the daily dose of the product has been decided.

Most disagreements about elemental impurities - between a specification writer and a contract laboratory, or between a buyer and a supplier - come from sliding between the two. A limit of one microgram per gram means nothing until the dose it was derived from is named.

Where a PDE comes from

The PDEs in use in the major markets come from the ICH Q3D guideline, which classifies the elements by how likely they are to be present and by their toxicity, and then assigns each one a permitted daily exposure for each route of administration: oral, parenteral and inhalation. The three routes produce different numbers for the same element, because the fraction absorbed and the sensitivity of the route differ. For the four elements the guideline groups together - lead, cadmium, arsenic and mercury, the ones with no useful therapeutic role and no realistic prospect of control at source - the oral PDEs are 5, 2, 15 and 3 micrograms per day respectively.

Three details in that paragraph matter more than the numbers. The PDE is per day, not per kilogram of body weight, so it does not scale with the patient. It is route-specific, so a figure taken from an oral table cannot be applied to an injectable. And it is set for chronic exposure, which is why it is a control level rather than a line separating safe from unsafe.

Turning a PDE into a concentration limit

The conversion is a division with one input that is frequently left unstated:

concentration limit (micrograms per gram) = PDE (micrograms per day) divided by maximum daily dose (grams per day)

A worked example. A solid oral product with a maximum daily dose of 0.5 g. For lead, with an oral PDE of 5 micrograms per day, the limit is 5 divided by 0.5, giving 10 micrograms per gram, or 10 ppm. For cadmium, 2 divided by 0.5 gives 4 micrograms per gram. Halve the dose to 0.25 g and both limits double; a product dosed at 4 g per day gets limits a fifth of the figures in the guideline table.

That table is where specifications go wrong most often. A limit copied from a published table is correct only for the dose that table assumes, and the dose is rarely written down next to the number.

The 10 gram assumption, and what happens above it

Q3D builds its option 1 concentration limits on the assumption that the maximum daily dose of the product is 10 g per day or less. That is what allows the table to be used as a default: below 10 g the computed limits are within reach of modern instrumentation. Above it the arithmetic keeps going and the limits keep falling - at 20 g per day the lead limit is 0.25 micrograms per gram, close to the bottom of what an inductively coupled plasma mass spectrometer quantifies comfortably in a straightforward digest. A product dosed that high needs a documented rationale, not a table lookup.

Two other parts of the option 1 arithmetic are worth stating. Compliance is judged on the sum of contributions across all the specified elements, each contribution expressed as a fraction of its own limit, and the total must not exceed 1.0 - so an element at half its limit uses half of the allowance. And where an element is not otherwise specified, a control threshold of 2.5 micrograms per gram is the default used for the assessment.

What that asks of the analytical method

A limit is only a limit if the method can see below it. The practical criterion is that the limit of quantitation sits at or below roughly 30 percent of the specification limit, so the method can report a value well inside the limit instead of "less than something". At a lead limit of 10 micrograms per gram that is comfortable. At 0.5 micrograms per gram it takes an appropriate digest, a clean instrument and a calibration that brackets the low end.

Three consequences follow. The result is reported on the material as supplied, so the dilution arithmetic has to trace back to a weighed mass of the drug substance, not of a working solution. The digestion has to be complete for the elements of interest, because a partial digest can leave refractory species undissolved and produce a false negative that no review will catch. And the response has to be validated in the matrix, because a matrix that suppresses signal by 20 percent also reports 20 percent low, and a limit test has no way to notice.

Validation expectations for the procedure itself sit in the ICH Q2 family: specificity, linearity, accuracy, precision and the range over which each holds. Where the procedure is a pharmacopoeial general chapter, that chapter carries its own system suitability requirements and they take precedence over a generic validation plan.

Where a reference standard does and does not fit

Elemental impurity testing uses calibration standards, not pharmacopoeial reference standards, and the two are different objects. An ICP calibration standard is a solution of known concentration, and its quality is judged from the certificate of that solution: element, concentration, matrix, uncertainty and expiry. A pharmacopoeial reference standard is a material whose suitability is defined against the assays in a monograph, and its information leaflet states that the quantity in the vial is approximate and not to be treated as an analytical quantity.

What a buyer should look for in a calibration standard is a certified concentration traceable to a national metrology institute, an uncertainty stated with its coverage, and a matrix that matches the digest so no mismatch is introduced. For anything used as a certified reference material, the producer is assessed against ISO 17034, which is the frame for what an assigned value and its uncertainty are supposed to mean. A certificate quoting a concentration without traceability, or a purity without an uncertainty, cannot support a limit test.

What belongs in the specification

An elemental impurity line is complete when it names the element and its limit, the basis on which the limit applies, the maximum daily dose the limit came from, the procedure, and the testing frequency. Anything less leaves the line ambiguous at the bench.

1. The element and its limit, in micrograms per gram, with the basis stated - as supplied or on a dried basis.
2. The maximum daily dose used in the conversion, so the arithmetic can be rechecked.
3. Which option the limit comes from: computed from the PDE, a control threshold, or a limit taken from a monograph.
4. The procedure, its system suitability requirements, and the limit of quantitation relative to the limit.
5. How the sum across several elements is handled, when more than one is specified.
6. Whether testing is per batch or at a reduced frequency, and what that reduction rests on.

Traps worth naming

• Copying a limit from another product. The limit depends on the daily dose of your product, not on the element alone. The same division by dose appears in how a mutagenic impurity limit is derived, with different toxicology behind it.
• Testing a standard instead of the material. The limit applies to the drug substance or the product. A result obtained on a bought-in standard is a statement about that standard.
• Reporting "less than" from an insensitive method. If the limit of quantitation sits above the limit, a compliant report was never possible, however good the material was.
• Using an oral PDE for an injectable. The parenteral figures are lower, and the difference is not a rounding error.
• Forgetting water and solvent in the basis. A result corrected to a dried basis describes a different material from the one on the balance - the same basis arithmetic that purity assignment runs into.
• Adding up elements as if each were allowed at its limit. Each contribution is a fraction of its own limit, and the fractions add to a total that must stay at or below 1.0.

Where to start

If the maximum daily dose and the route are known, the limit is arithmetic once the PDE is agreed. The two things worth settling before any material is bought are the dose basis and the basis of the material. Send the maximum daily dose, the route and the elements of interest to technical@rssechem.com. The impurity standards index is organised by parent drug substance, and the certificate of analysis page shows the fields we report.