Quantitative LC-MS/MS is the standard technique for measuring drugs and metabolites in biological fluids. Its sensitivity and selectivity are exceptional, but its response is not always stable: ionisation efficiency can change from sample to sample because of co-eluting matrix components. An internal standard corrects for this and for losses during sample preparation. Stable-isotope-labelled (SIL) internal standards, most often deuterated analogues of the analyte, are widely regarded as the best choice. This article explains why, and how to use them well.
What an internal standard does
An internal standard (IS) is a compound added at a fixed concentration to every calibration standard, quality control sample and study sample, usually at the very start of sample preparation. Quantitation is based on the ratio of analyte response to IS response. If extraction recovery drops, an injection is slightly smaller, or the matrix suppresses ionisation, the IS should be affected to the same extent as the analyte, so the ratio stays constant.
That only works if the IS behaves like the analyte in every step: extraction, chromatography and ionisation. A structural analogue behaves similarly. An isotopically labelled version of the analyte itself behaves almost identically.
Why stable-isotope labels work so well
Replacing hydrogen with deuterium, or carbon-12 with carbon-13, changes the mass but barely changes the chemistry. The SIL-IS therefore:
- extracts with essentially the same recovery as the analyte
- co-elutes, or nearly co-elutes, so it experiences the same matrix effects at the same moment
- ionises with very similar efficiency
- is distinguished from the analyte by the mass spectrometer through its different precursor and, usually, product ion masses
The ICH M10 guideline on bioanalytical method validation, adopted by the FDA, EMA and other regulators, states that a stable-isotope-labelled analyte is recommended as the internal standard for mass spectrometric methods, where available.
Choosing a labelled internal standard
Mass difference
Natural isotopes, mainly carbon-13, give every molecule an M+1, M+2 and sometimes M+3 signal. If the IS is only 1 or 2 Da heavier, the analyte's natural isotope peaks overlap with the IS channel, especially at high analyte concentrations. A mass difference of at least 3 Da is a common rule of thumb, and more is needed for molecules containing chlorine, bromine or sulfur, which have significant M+2 isotopes.
Label position
Deuterium on heteroatoms, such as OH, NH or SH, exchanges rapidly with protons in water or methanol and is lost. Deuterium on carbon adjacent to carbonyl groups or other acidic positions can also exchange under some conditions. Good labels sit on stable aliphatic or aromatic carbon positions. Check that the labelled atoms remain in the product ion used for MRM. If the fragment that carries the label is lost, the IS product ion may coincide with the analyte's.
Isotopic purity
No labelled compound is 100% labelled. A standard with 99 atom % D per position may still contain small amounts of the unlabelled analyte, sometimes called the d0 contribution. This unlabelled material appears in the analyte channel and can bias the lower limit of quantitation. The COA should state isotopic enrichment and, ideally, the proportion of unlabelled species.
Deuterium versus carbon-13 and nitrogen-15
Deuterium labels are cheaper and easier to introduce, which is why deuterated standards dominate. However, extensive deuteration can shift retention time slightly in reversed-phase LC, with the labelled compound usually eluting a little earlier. If matrix effects change sharply across the peak, even a small shift can reduce how well the IS compensates. Carbon-13 and nitrogen-15 labels do not cause measurable retention shifts and do not exchange, so they are preferred for the most demanding assays, at higher cost.
Cross-talk and interference limits
Two kinds of interference must be checked during validation:
- IS contribution to the analyte channel, from unlabelled impurity in the IS. ICH M10 expects the response in a blank sample spiked with IS only to be no more than 20% of the analyte response at the lower limit of quantitation.
- Analyte contribution to the IS channel, from natural isotopes or impurities. The response in the IS channel from a sample at the upper limit of quantitation without IS should be no more than 5% of the average IS response.
If these criteria fail, options include using a higher-mass-shift standard, reducing the IS concentration, or selecting different transitions.
Setting the IS concentration
The IS concentration should give a robust, precise signal without overwhelming the detector or contributing significant unlabelled material to the analyte channel. Many laboratories choose a concentration that gives a response similar to the analyte in the lower to middle part of the calibration range. Consistency matters more than any particular value: the same amount must be added to every sample.
Monitoring IS response
Regulators expect IS response to be monitored across each analytical run. A sample whose IS response is markedly different from the others may indicate a pipetting error, extraction failure or unusual matrix effect. Laboratories define acceptance criteria and investigation procedures in their SOPs. A consistent drift in IS response across a run can also reveal instrument problems.
Sourcing labelled standards
Labelled analogues of common drugs are often available from catalogues, but for new drugs, metabolites and nitrosamine drug substance-related impurities, custom synthesis is frequently required. When ordering, specify the label position and number, minimum isotopic enrichment, maximum unlabelled content, chemical purity and the characterisation data you need, typically ¹H NMR to confirm label position and mass spectrometry to confirm isotope distribution.
Summary
A well-chosen SIL internal standard is the single most effective tool for robust LC-MS quantitation. Look for an adequate mass shift, stable label positions, high isotopic purity and low unlabelled content, verify cross-talk during validation, and monitor IS response in every run.
FAQs
Why is a deuterated internal standard better than a structural analogue?
It behaves almost identically to the analyte in extraction, chromatography and ionisation, so it compensates more accurately for losses and matrix effects.
Can deuterium labels be lost?
Yes. Deuterium on OH, NH or SH groups exchanges rapidly with solvent, and labels next to carbonyl groups can exchange under some conditions. Choose stable carbon positions.
What cross-talk limits does ICH M10 set?
IS interference in the analyte channel should be no more than 20% of the LLOQ response, and analyte interference in the IS channel no more than 5% of the IS response.