Analysing the monographs

The European Pharmacopoeia (Ph. Eur.)1 monographs are subject to regular adaptation in order to cope with progressing quality requirements. This contribution outlines the evolution in analytical requirements and techniques in monographs for organic substances. These monographs generally carry the subheadings definition, characters, identification, tests, assay and impurities.

The general style is being made shorter and more direct. Since the current 5th edition does not only contain newly developed monographs, a mix of different styles can be observed.

Definition

The definition mainly reports the chemical name and limits for the content, which is calculated on the ’dry‘ material, in which case the loss on drying is taken into account, or on the ’anhydrous‘ material, in which case the water content is taken into account. These techniques are further discussed under Tests.

The content limits are of course dependent on the purity of the substance: generally, the limits are more stringent for a synthetic substance than for a natural substance, such as an antibiotic, obtained by fermentation. Indeed, the first can be purified more easily during the different stages of the synthesis, whereas the second contains fermentation impurities that are often difficult to remove.

But the analytical technique also has an influence on the content limits. So it is observed that for classical titration techniques 99-101% limits are often prescribed, whereas for UV spectroscopy (UV) assay methods 97-103% limits are usual. This is related to the better reproducibility that can be obtained with titrations and the more limited precision of the spectroscopic method, certainly if specific absorbance values are used and the calibration of the spectrophotometer therefore becomes an important factor. For liquid chromatographic (LC) methods also, the limits are broader; classically 98-102% for substances that can be obtained in a high degree of purity and this is related to the reproducibility of the method but also to the specificity. Indeed, contrary to titration methods, impurities do not contribute to the final assay value in LC. In view of this, it would be more logical to have asymmetrical limits at around 100%. For a long time, it has been Ph. Eur. policy to combine a less specific assay method with a more specific chromatographic test for related substances. This has the advantage that good precision is obtained in the assay, but the accuracy is less due to interference of impurities. However, the amount of impurities is normally limited by the related substances test. This has been discussed extensively in a recent paper2.

An advantage of titrations is that one does not need a reference substance with an officially stated content. Reference substances make assays more expensive. Indeed, though not generally specially treated to improve purity (production batches of good quality are normally used), they have to be examined extensively in a collaborative study with five or more laboratories to determine the content of the active. This mainly implies the performance of titrations, liquid chromatographic and/or other techniques to determine the impurities, loss on drying and/or water content and gas chromatography (GC) to determine the volatile components.

Characters

The characters give general information about properties such as appearance (colour, crystallinity and polymorphism), solubility and hygroscopicity. The information in this paragraph is not to be considered as strict analytical requirements, although clear deviation from the described properties indicates noncompliance. Also, analytical methods to make accurate measurements are not available. Indeed, the Ph. Eur. does not describe a method for colour determination of solids and the methods for determination of solubility and hygroscopicity are approximate. To determine whether a substance is crystalline, reference is made to X-ray powder diffraction. If, for a given substance, a well-defined crystalline form has to be used for reasons of activity, then a test (melting point, infrared (IR) X-ray) will be provided under identity. Melting points are reported merely for information and will probably disappear from the characters in future. Characters such as odour or taste are no longer reported in recent monographs.

Identity

The identity requires selective techniques with high specificity, but they do not have to be sensitive.

Identification tests are expected to confirm the identity mentioned on the label and not to identify totally unknown structures. When the substance is also to be used in (hospital) pharmacies, where less sophisticated equipment is available, a second series of identifications is provided, which may be used only if the first has already been used upstream in the distribution chain. Infrared spectroscopy is taking over more and more from techniques such as colour reactions, melting point and UV spectroscopy. IR is often used alone, or in the case of some salts, in combination with a test for the counter-ion. For a period, IR has been used with a reference spectrum, but in recent monographs, reference substances are introduced systematically. These reference substances are easier to produce than those for quantitative work because the exact content is not determined. It is true that much better specificity can be obtained when the spectra to be compared have been obtained on the same equipment. When needed for reasons of specificity, an additional test is added to the IR. UV is less specific than IR because the spectrum is simpler. On the other hand, it is easier to interpret UV results (wavelength of maxima and specific absorbance values), whereas in IR at least some experience is needed to decide whether or not the spectra are identical.

Chromatography is also a selective technique. Thin-layer chromatography (TLC) has been extensively used. However, it is observed that IR replaces TLC in recent monographs. Both techniques need a reference substance, but IR is faster and gives better specificity. Also, many TLC methods use several reference substances in order to check the resolution of the method, which can be influenced by chromatographic conditions such as the quality of the stationary phase. The use of TLC is therefore more restricted to the second identification series.

Liquid chromatography (LC) and gas chromatography are less frequently used – the latter also because few substances are volatile. LC has the advantage that in the same run, identity, purity and assay may be performed.

The specific optical rotation is used for identification of an enantiomer. This can also be achieved by enantioselective LC, which is much more difficult to perform and needs reference substances. The angle of rotation is used to distinguish a racemate from an enantiomer.

Tests

Tests need to be sensitive in the first place because small amounts of impurities must be detected.

General tests for control of purity will be discussed at first, where-after tests for related substances will be treated.

In recent monographs, the appearance of solution is prescribed only for substances for parenteral use. In this case the clarity of a solution is of primordial importance but most often the colour is examined too. In recent monographs, when relatively higher amounts of colour are allowed, the colour is often checked by an absorbance test in the 400-450 nm zone. Very weak colours are difficult to measure in a spectrophotometer with short cells.

Impurities with acidic or alkaline character are detected by the pH test or the acidity/alkalinity test. This latter test, which involves a titration, is prescribed only when the buffer capacity of the tested solution is sufficiently weak.

Optical rotation is used in recent monographs for control of the enantiomeric purity or, in the case of racemates, to control the absence of optical activity. In the latter case the test is prescribed only if it has been shown that the method is sufficiently sensitive.

The heavy metal test and the sulphated ash test, for control of mineral purity, are prescribed as general quality control tests. Both consume a lot of substance (1 g) and are laborious. The heavy metals test is prescribed in function of the daily dose, the usual period of treatment and the route of administration. Six methods are available and in the 5th edition the sensitivity in all the methods may be improved by filtration of the very weakly coloured solutions on a filter, which facilitates the interpretation. In each method a supplementary test has been added that allows the control for interference of the substance examined with the test. It is true that more performant techniques (AAS, ICP-MS) exist nowadays for control of heavy metals, but their introduction in daily routine analysis is not considered yet due to insufficient availability of equipment. The sulphated ash test may be omitted for very expensive substances.

Simple tests for extraneous cations or anions are still in use, but colour reactions and UV tests for related substances are largely replaced by chromatographic tests.

The determination of water by the Karl Fischer titration (KF) is more specific than the loss on drying (LoD) test, which allows a more general measurement of volatile substances. The KF test is normally performed in solution, whereas the LoD does not always eliminate solvents, which are bound inside the solid structures. More specific information on the residual solvents is obtained by GC. In general, the ICH rules for residual solvents3 are of application and monographs do not repeat this. Only when levels higher than ICH levels are allowed, the test will remain prescribed and where needed, a method is described.

TLC tests for related impurities are replaced systematically by LC. This is due to the better specificity of LC but also to the fact that non-instrumental TLC does not allow for making totals of impurities. TLC scanners that allow the production of such quantitative TLC results are not used because it was considered that LC gives at least equally good results and LC equipment is more generally available. However, TLC remains important for analysis of plant materials, for example, where the complex composition of the sample is less compatible with the use of expensive columns that are easily contaminated.

The intention of the 5th edition was to have LC methods that are fully compliant with ICH prescriptions for impurities, although these are described for new drug substances only. This implies that for a maximal daily dose of not more than 2 g/day, the reporting threshold is 0.05% and the identification threshold is 0.10%. The main problem is that specified impurities must be located in the chromatogram whereas the monograph does not always provide the possibility to do so. The best method to locate specified impurities is with the help of reference substances of individual impurities or of a reference substance for identification of impurities, which contains a mixture of impurities and of which an example chromatogram is available. However, some monographs do not provide this and only give relative retention times (RRT) for peak identification. In some cases even the RRT are lacking. It has been shown that RRT are not sufficient for peak identification in a chromatogram4. When the specified impurities are limited at a level of 0.1%, there is no real problem of decision-making because all the impurities, specified and unspecified, are limited at the same level. In some cases, however, the specified impurities are limited at levels above 0.1% and it is then important that the specified can be distinguished from the unspecified ones. According to the 5th edition, impurities that are not listed are automatically limited at 0.1%. This brings a sudden change in monographs with, for example, a statement such as “any other impurity is not greater than 0.5%”, where previously all other impurities were allowed to be present up to 0.5%, whereas now the limit is valid for the listed impurities only. Unfortunately, the impurity lists have not been adapted accordingly with the consequence that the limits for a number of impurities have suddenly and unwillingly been changed in the new edition. Much of the harm is due to the fact that one has tried to adapt in one movement all the monographs, for new and for old drug substances. It can be expected that major efforts will be made to normalise this situation as soon as possible.

Capillary electrophoresis (CE) is a relatively new separation technique based on differential migration velocity of compounds in an electric field. Because of the different separation mechanism it is considered to be complementary to LC. It has been introduced to a rather limited extent in monographs, for the test on related substances, mainly in cases where LC could not yield satisfactory results. In general it can be expected that CE will not replace LC for assay purposes, because the precision obtained in CE is not superior to the precision in LC.

Assay

The assay has already been partly discussed under ‘Definition’ (see above). Assay methods do not need to be sensitive but should be accurate and reproducible. Specificity contributes to better accuracy.

UV assays are almost absent in new monographs. The UV methods either use specific absorbance values or a reference substance. The latter leads to better reproducibility but obviously is more expensive. UV methods are not selective. Titration methods are reproducible but not very accurate since impurities may react too. LC assays are now introduced more frequently, in the first place because they are also used for simultaneous testing for related substances, but also because they are more selective and accurate. Another argument in favour of LC is that the active part of the molecule is determined, whereas in many titrations the counter-ion is titrated. Microbiological assays for antibiotics have been replaced by LC in many cases. At first, LC was introduced for the semi-synthetic and the less complex natural antibiotics, but more recently LC is also used for the more complex ones and when necessary, several active components are summed to give the final content value.

For salts of basic antibiotics, the content was originally expressed in terms of the base in order to obtain values that were more easily comparable to the original potency figures, expressed in International Units/mg. This sometimes gave very low content values for salts and consequently gave an erroneous impression of poor purity. This situation is now being adapted so that contents are expressed in terms of the salt and therefore yield values that are closer to 100%. It should be emphasised that there is no official relationship between potency and content expressions, which means that they cannot be translated into each other.

Impurities

The list of impurities is undergoing major changes. At this moment the situation is rather complex; some monographs have no impurities list and in those having one, several different subheadings are used. There is consensus that this should be adapted as soon as possible in order to be fully compliant with the ICH prescriptions.

Concluding remarks

The Ph. Eur. is very much alive because it is undergoing interesting changes that will certainly help to confirm the eminent place it has conquered worldwide in past decades.

References

1. European Pharmacopoeia, 5th Edition, 2004, EDQM, Council of Europe, Strasbourg, France.
2. S. Görög, J. Pharm. Biomed. Anal., 36 (2005) 931- 937.
3. International Conference on Harmonisation, http://www.eudra.org/emea.html
4. P. Dehouck, D. Visky, Z. Kovács, B. Noszál, E. Adams and J. Hoogmartens, LC•GC Europe November 2003, 764 – 771.