Question of the month - November 2018
What level of characterisation is sufficient for impurity reference standards?
None of the regulatory sources really give a clear answer. The International Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use (ICH) states in their Q3A and Q3B guidelines that impurity reference standards (IRSs) “should be evaluated and characterised according to their intended uses.”
So the question really is: what is my intended analytical use? For the most part, the use will be qualitative or quantitative.
For qualitative purposes, a full identification is necessary. At LGC, we think that 1H-NMR, 13C-NMR, MS, IR, CHN and Melting Point (MP) are the necessary techniques to confirm identity. MS, IR, CHN and MP give a first indication of structure and particularly allow comparison against literature data. On the other hand, H-NMR and C-NMR are great techniques to elucidate structures from scratch, especially where there is no literature available. In case the identity is not secured by the aforementioned techniques, you can perform additional NMR experiments including DEPT, HSQC, HMBC, NOESY and COSY to prove identity at the highest scientific level. We have further techniques available at LGC, should identity still be an issue for a certain substance, but let’s not get into that here: contact us if you any further questions or specific requirements.
Additionally, for qualitative applications, you should check your IRS candidate for purity. This should be over 85%, otherwise you might face difficulties interpreting the IR and NMR data correctly.
For quantitative applications, you should also determine the assay correctly, in addition to the techniques employed for the identity checks above. This determination is often performed by the 100% method (aka mass balance): you analyse and subtract from 100% the volatile contents (water and residual solvents) present in the candidate material, and multiply by the chromatographic purity, to account for organic traces present besides your target molecule. You can also use other approaches to check for the assay, like qNMR.
We do not recommend using qualitative IRSs (aka research materials) quantitatively, for many reasons. We discuss this in more detail in a recent Analyte Guru blog with Thermo Fisher, or in our new white paper, Characterisation of non-compendial impurity reference standards: how good is good enough?.
If you want to learn more, we also deal with this subject in great detail in our on-demand webinar.