The cream formulation contains terbinafine hydrochloride, lidocaine, L-menthol, dipotassium glycyrrhizate, diethylene glycol monoethyl ether, isopropyl myristate, carbomer 940, sodium docusate, parabens, and other excipients.
We are attempting to analyze terbinafine-related genotoxic impurities, such as N‑nitroso terbinafine degradant and N‑nitroso terbinafine impurity A. However, due to matrix effects caused by the excipients, the recovery is very low (approximately 10%).
Are there any effective methods to reduce matrix effects in this case? Any advice would be greatly appreciated.
Best way would be to weigh every single excipient separately and spike with reference solution.
With this you could see what exactly is causing your bad recoveries and with the knowledge you could try to get rid of it. If polarity of excipient and analyte are quite different, then perfoming a liquid/Liquid partition would be an quick and easy solution.
If separation fails during sample preparation, you might still be able to resolve the issue using your HPLC.
But for this, you need to make the matrix influence visible.
If you have a syringe pump available, please proceed as follows:
Inject a blank.
After your column and before your MS, insert a T‑junction.
At this T‑junction, infuse your analyte solution via the syringe pump into the LC flow.
This will give you a high, stable baseline, and you will be able to see exactly at which retention times matrix effects occur in your run.
For example, if you observe a dip right before your analyte, and this dip is relatively narrow with a defined peak shape, then you can try to optimize your gradient or other chromatographic conditions.
Without the T‑infusion setup, these matrix effects are not visible, so keep that in mind.
You are right, but there are a few important points to consider:
First, the recoveries are around 10%. If an internal standard is used, the LOQ must still be detectable. In this case, sensitivity may already be maxed out, meaning that adding an internal standard will not solve the issue.
Second, internal standards, especially isotopically labeled ones, are quite expensive. If you invest some time in optimizing your extraction or chromatographic conditions, you may save a significant amount of money if your method ultimately does not require an internal standard.
Third, some structures are, for example, acidic at the alpha‑carbon. If you use deuterated standards, proton–deuteron exchange can occur, which will also change the mass and compromise quantification. In such cases, ¹⁵N-labeled internal standards are a better option, but they are also even more expensive.
So yes, using an isotopically labeled internal standard is often an easy but costly fix for matrix effects. But it’s worth considering these aspects before deciding on it.