25
control of specific impurities that cannot conveniently be controlled by LC or GC. Existing TLC tests that do not follow this recommendation will be replaced gradually as soon as information on suitable LC or GC tests becomes available.
Where the counter-ion of an active substance is formed from a lower organic acid, a test for related substances of the organic moiety is usually not considered necessary (for example, magnesium lactate used as a source of magnesium).
Monographs frequently have to be designed to cover different impurity profiles because of the use of different synthetic routes and purification procedures by producers. The usual practice is to include a general LC test, supplemented where necessary by other tests (LC, GC, CE, TLC, or other techniques) for specific impurities. It is, however, becoming increasingly impractical in some cases to design a single general test and in such cases more than one general test is included and the scope of the different tests is defined in the tests themselves with a cross-reference in the IMPURITIES section.
Monographs cover a number of specified impurities designated in the IMPURITIES section. Specified impurities are those that occur in current batches of the substances used in approved products and for which an individual acceptance criterion is provided. Wherever feasible, monographs also have an acceptance criterion for other impurities (at the identification threshold for the substance) and a limit for the total of impurities (or a limit for the total of impurities other than a number of identified specified impurities) above the reporting threshold. The acceptance criterion for specified impurities may be set at the identification threshold for the substance. The acceptance criteria for specified impurities take account of both:
?
qualification data, where applicable, the limit being set at a level not greater than that at which the impurity is qualified; the information on qualification is provided by the producer and the compatibility of the limit with the qualification data and approved specifications is checked by the competent authorities during elaboration of the monograph and/or during the Pharmeuropa comment phase; and
batch analysis data, the acceptance criteria being set to take account of normal production; data is provided by the producer for typical batches and checked during elaboration of the monograph on not fewer than 3 batches.
?
All decisions on impurity acceptance criteria should be based on the real impurity content (meaning after application of correction factors (CF)) in representative batches examined. Impurities need to be specified and located appropriately in the chromatogram if the reported batch values for an impurity are:
? ?
above the applicable limit for unspecified impurities before correction and cross the limit downwards when corrected (overestimation, CF<1), or
below the limit for unspecified impurities before correction and cross this limit upwards when corrected (underestimation, CF>1).
Usually, no correction factor will be given if the reported batch values for an impurity are below the applicable limit for unspecified impurities before correction and below the reporting threshold (disregard limit) after correction.
In any case, correction factors between 0.8 and 1.25 (corresponding to response factors of 1.2 to
26
0.8) are not given in monographs. Further information on the indication of correction factors is given in chapter II.5.8.2.b.
Response and correction factors. According to general chapter 2.2.46. Chromatographic separation techniques, the relative detector response factor, commonly referred to as response factor, expresses the sensitivity of a detector for a given substance relative to a standard substance. The correction factor given in the monograph is the reciprocal value of the response factor. The response factor can be calculated from the following formula:
??????=
????????× ????????
RRF = response factor
Ai = area of the peak due to the impurity
As = area of the peak due to the test substance
Cs = concentration of the test substance in milligrams per millilitre Ci = concentration of the impurity in milligrams per millilitre.
For the calculation, the mean of the area ratios over the whole range of linearity or the ratio of the slopes of the respective linearity regression equations may be used.
The response factor can be determined from the formula above, by preparing solutions of defined concentrations of the impurity and the test substance and measuring them by LC/UV at a given wavelength and flow rate. The concentration of the impurity and that of the test substance should be in the same order of magnitude and the measurement should be carried out using a calibration curve determined at several points around the concentration which corresponds to the acceptance criterion of the impurity. The weighings of impurity and test substance should both be corrected for the purity. Ideally, the chromatographic purity and water/solvent content of the impurity and the test substance should be determined beforehand. A provisional value might be assigned on the basis of the formula:
??????????????(%)=[100?(??????????+????????????????)]×
???????????????????????????? ???????????? (%)
100Suitable methods should be chosen when only a small amount of the impurity is available, e. g. TGA, coulometry for water/solvents, and LC to estimate its purity by injecting a concentrated solution of the impurity. If the available amount of impurity is still too small, values given in the certificate of analysis may be used.
It is also important to consider the form (base/acid or salt) of both the impurity and the test substance and to introduce an additional correction factor for the molecular mass ratio when impurity and test substance used for the determination are present in different forms. Preferably the response factors should be determined in two laboratories using the same protocol. If different detector types (diode-array detector (DAD) and variable wavelength detector (VWD)) are available, these may also be considered for the measurement of response factors.
Separation methods. For pharmacopoeial purposes the objective of a purity test using a separation method will usually be the control of impurities derived from one or more known manufacturing processes and decomposition routes. However, the experimental conditions are chosen for the test, especially the detection system, so as not to make it unnecessarily narrow in
27
scope. Chromatographic purity tests may often be the best means of providing a general screening of organic impurities derived from new methods of manufacture or accidental contamination. It may be advantageous to supplement a chromatographic test with other chromatographic or non-chromatographic tests.
As mentioned above (part II.4.8), a chromatographic system applied to purity testing may, when suitable, be applied also for identification.
When a related substances test based on a chromatographic technique is carried out, a representative chromatogram is published with the monograph in Pharmeuropa. Ultimately, the chromatogram will not be published in the Ph. Eur. but will be transferred to the EDQM Knowledge Database.
When no individual impurity is available as a reference substance or when a large number of impurities may be detected in the substance, a representative chromatogram will be supplied with the available reference substance (e.g. substance to be examined spiked with the impurities). Monographs should provide a reliable means of locating all specified impurities on the chromatogram. Identification of unspecified impurities is necessary if a correction factor is to be applied. Peaks may be located using:
? ?
a reference standard for each impurity;
a reference standard of the substance to be examined containing some or all of the specified impurities, provided with a chromatogram.
Location by relative retention is not generally considered sufficient for pharmacopoeial purposes, notably for gradient elution. Where a reference standard containing a mixture of impurities is to be used, a sample of each specified impurity should be provided to the EDQM to enable the establishment of the reference standard.
In general, relative retention is given to 1 decimal place. However, it is given to 2 decimal places where necessary to indicate the elution order of closely eluting peaks. General considerations applying to separation techniques:
?
high concentrations/loadings are normally used since the symmetry of the principal peak or shape of the spot is not critical in impurity testing so long as there is no interference. When using an external standard in quantitative determinations the response of the principal peak need not be in the linear range of the detector;
in general tests for related substances, the substance to be examined should not to be chemically modified (e.g. derivatisation) before purity testing since the impurity pattern may be modified;
similarly, extraction of the free base or acid prior to impurity testing is to be avoided.
?
?
II.5.8.1. Thin-layer chromatography (TLC) (2.2.27.)
TLC methods should only be used for control of a specified impurity and where LC, GC or CE methods are inappropriate (usually due to a lack of a suitable detection system).
Commercially available pre-coated plates, described in general chapter 4.1.1. Reagents, are to be used; the trade name of the plate found suitable is indicated in a footnote to the draft monograph,
28
and posted to the EDQM Knowledge Database after adoption of the monograph. In general chapter 4.1.1. Reagents, besides information on the coating material used (type of coating material, type of binder), a suitability test procedure is described under TLC silica gel plate R. The monograph must describe the type of plate and include a system suitability requirement. Often the substances that would be best suited for a system suitability test will not be readily available individually; a sample of the substance to be examined containing them as contaminants or even a deliberately spiked sample may then be prescribed. Permissible variations to the different parameters are indicated in general chapter 2.2.46. Chromatographic separation techniques.
If any pre-treatment is required or if the chromatography is carried out in unsaturated conditions for the satisfactory conduct of the test, then this information is included in the text of the monograph (especially applicable to the use of reverse-phase plates).
One or more dilutions of the substance to be examined will often prove adequate for reference purposes, provided the impurities to be compared exhibit a similar behaviour under the chosen chromatographic conditions. This implies that the spots to be compared must be sufficiently close in RF value to minimise errors introduced by different diffusion of the substances during their migration. Otherwise, reference solutions containing the specified impurities are to be employed. It may be necessary to instruct the analyst to disregard a spot – often due to the non-migrating counter-ion of a salt – remaining on the starting line.
Summation of the responses exhibited by each individual spot is only acceptable when appropriate equipment is prescribed. It is not recommended to set a limit or limits for the concentration of impurities without a limit on their number, otherwise the total theoretical impurity level would be unacceptably high. This situation may be counteracted by limiting the impurities on 2 or more levels, allowing only a defined number to be at the higher level and the rest below the lower level. As examples, the test may specify that no contaminant may exceed a relative concentration of 1 % and that only 1 may exceed 0.25 % or that no contaminant may exceed a relative concentration of 1 %, only 1 contaminant above 0.5 % and no more than 4 contaminants above 0.25 %.
II.5.8.2. Liquid chromatography (LC) (2.2.29.)
Defining the appropriate chromatographic system will often be one of the major problems to be dealt with in elaborating a pharmacopoeial purity test based on chromatography. In LC the matter is further complicated by the existence of numerous variants of stationary phases, especially amongst the chemically bonded reverse-phase materials for which not only brand-to-brand but occasionally also batch-to-batch variations occur that can influence a given separation. Once the type of stationary phase tested has been found to show a satisfactory separation it must be defined by, for example including the following information: type of particles (irregular or spherical), particle size, specific surface area (m2/g), pore size (nm), and when using reverse-phase columns the extent of carbon loading (%) and whether the stationary phase is end-capped or otherwise treated to inactivate the residual silanol groups (this is particularly important when basic substances are to be examined and there is a risk of peak tailing).The trade name of the stationary phase(s)/column(s) found suitable during elaboration of the monograph is indicated in a footnote to the draft monograph and transferred to the EDQM Knowledge Database after adoption of the monograph.
29
In describing the chromatographic system, mention must be made of the column dimensions (length and internal diameter), nature of the stationary phase (as detailed previously) including any steps to prepare or pre-treat it, composition and flow rate of the mobile phase including elution programme (if any), column and autosampler temperature (if differing from room temperature or especially if thermostated), method of injection (if important), injection volume and method of detection.
Depending on the detection wavelength selected, the analyst should choose a suitable grade of solvent when preparing the mobile phase. The following guidance applies for the most frequently used solvents methanol and acetonitrile from the 7th edition of the Technical Guide on. If water is used as a component of the mobile phase, water for chromatography R should be used.
Wavelength intervals λ ≥ 250 nm 220 nm ≥ λ < 250 nm λ < 220 nm Acetonitrile grade Acetonitrile R Acetonitrile for chromatography R Acetonitrile R1 Methanol grade Methanol R Methanol R1 Methanol R2
Permissible variations to the different parameters are indicated in general chapter 2.2.46. Chromatographic separation techniques.
Test and reference solutions are wherever possible prepared using the mobile phase as the solvent in order to minimise peak anomalies.
Since many active pharmaceutical substances are synthesised by a number of synthetic routes, the list of potential impurities to be limited may be large and the analytical challenge to separate them is great. For the sake of robustness and reproducibility, isocratic elution is to be preferred while setting up a compendial method. However, isocratic liquid chromatographic methods may not be sufficiently selective so there is an increasing need to employ gradient methods.
When a gradient system is described, all necessary parameters must be clearly given, e.g. composition of mobile phases, equilibrium conditions, gradient conditions (linear or step), etc. In general the return to the initial conditions is not prescribed in monographs as this is considered to be instrument specific. Should this information be considered important (e.g. ion exchange chromatography), it may be added as a note to the draft monograph and transferred later to the EDQM Knowledge Database.
For gradient elution in LC an important parameter to be considered is the volume between the solvent mixing chamber and the head of the column. This volume is referred to as the dwell volume, D (other terms employed include: effective system delay volume, dead volume and delay volume). The dwell volume is dependent on the configurations of the pumping system including the dimensions of the capillary tubing, the solvent mixing chamber and the injection loop;. Large differences in dwell volume from one pumping system to another will result in differences in elution of peaks. The greatest effect of differing dwell volumes on retention times is for those substances that are not strongly retained. Thus, gradient systems should be conceived with an initial isocratic phase so that analytes are not eluting too close to the injection peak allowing the correction for marked differences in dwell volume from one gradient pumping system to another. The dwell volume of the pumping system employed to develop the method should be equal to or less than 1.0 mL. If the method is developed using a system with a dwell volume greater than
