In this fourth article in the series on the drug development process, more details are given of aspects of pharmaceutical development. These form the "quality" aspects of the quality, safety, and efficacy criteria that have to be met in the development of a new pharmaceutical product
The first article in this series gave an overview of the drug development process.1 The second described the options available in the United Kingdom and the rest of Europe for successful registration of pharmaceutical products to allow marketing in member states of the European Union (EU).2 The third article covered in detail the "quality" stages of the development process.3 This fourth article continues that coverage.
Guideline CPMP/ICH/281/96 is a derivative of the "Note for guidance on validation of analytical procedures: definitions and terminology" (see below) and provides guidance and recommendations on how to consider the various characteristics for each analytical procedure. It also indicates the data that should be supplied to support a marketing authorisation application (MAA). The guideline recognises that the analytical procedures used for biological and biotechnological products may be dealt with in ways that differ from those described. The arrangement of the sections in the guideline is itself instructive as to the ways in which the analysis should be carried out.
Specificity The objective of the analytical procedure will in part determine the way in which specificity is determined during the validation of identification tests, the determination of impurities, and the assay. If it is not possible for an analytical procedure to be specific for a particular chemical, two or more analytical procedures are recommended.
Closely related chemical structures must be identifiable in the analytical validation. It is acceptable for a positive result to be obtained by comparing an analysis of a sample containing the analyte with an analysis of a sample without the analyte.
Assay and impurity tests, usually using chromatographic techniques, must be carried out to determine specificity. If a non-specific assay is used, additional analysis must be carried out to ensure specificity is demonstrated.
If impurities and/or excipients are present in the material, it must be shown that the analyte and the other components can be separately identified. Reference samples can be "spiked" with known impurities and/or excipients and the result compared with actual samples. If impurities are not available, the test results should be compared with another established method (eg, that cited in a pharmacopoeia). Tests should also be carried out on samples that have been stored or subjected to known stresses (eg, humidity, light or oxidation conditions).
Linearity The "range" of the analytical procedure is the limits within which the results from the test can be considered confident (see below). There should be a linear relationship across the entire range of the analytical tests. Testing should be carried out on the stock solution of the test substance using a range of concentrations, or on separate weighings of synthetic mixture of the product ingredients. Statistical analysis of a linear relationship (eg, calculation of a regression line by the method of least squares) should be carried out. Data from this analysis (eg, the correlation coefficient, the y-intercept, the slope of the regression line, and the residual sum of squares) and a plot of the analysis should be included in the MAA. A minimum of five concentrations are recommended for determination of linearity.
Range Closely related to the linearity is the range of results from the anlaytical procedure. The guideline suggests a number of minimum specified ranges:
Accuracy The guideline defines the methods of defining accuracy of the assay for active substances and medicinal products and for impurities, and also indicates recommended data for submission.
One method of defining accuracy of the assay of the active substance is by comparing the results with those obtained using a well-established procedure. For a medicinal product, it is possible to determine the accuracy of the assay by applying the same procedure to a predetermined mixture of known concentrations of the ingredients.
The method suggested for determining the accuracy of the assessment of the impurities is to spike samples of the substance or product with known quantities of the impurity. Alternatively, the results can be compared with those obtained from a well-established procedure.
A minimum of nine assays over a minimum of three concentrations that cover the specified range should be used to assess accuracy.
Precision The precision of the analytical validation is determined by measurement of the repeatability of the tests and their reproducibility. The guideline also specifies the data recommended to be included in the MAA.
Detection limit The limit of detection can be based upon visual evaluation, upon signal-to-noise (for those analytical procedures that exhibit baseline noise), or upon the standard deviation of the response and the slope. The guideline also states that the method used for determining the detection limit and the actual detection limit itself should be included in the data.
Quantitation limit Similar approaches to those used for the detection limit should be used for the determination of the quantitation limit.
Robustness The robustness is a measure of the reliability of the analysis in the event of changes in method parameters. This is expressed as a series of system suitability parameters (eg, a resolution test) that ensures that the analytical procedure used remains valid throughout any variations. Such variations may include the stability of analytical solutions or extraction time; others may arise in the use of liquid or gas chromatography.
Guideline CPMP/ICH/381/95 provides a collection of terms and their definitions, but states clearly that it is not intended to indicate how analytical validation is to be achieved. It provides a common terminology to replace those previously in use in the three regions involved in the ICH process (the EU, the United States and Japan).
Types of analytical procedures to be validated The guideline describes the four most commonly used analytical procedures:
As has been described in the guideline on methodology (CPMP/ICH281/96 above), the parameters used in the majority of analytical testing procedures are accuracy, precision, repeatability, intermediate precision, specificity, detection limit, quantitation limit, linearity, range, and robustness. Each of these terms is described in the glossary of this guideline. The circumstances when revalidation may be required are also explained (eg, when there are changes in the synthesis of the active substance, the medicinal product's formulation, and in the analytical procedure).
Guideline CPMP/ICH/380/95 describes the core stability data required to accompany an MAA for new active substances and medicinal products. (It does not cover the data that must be submitted in abridged MAAs, clinical trial applications or variations to approved marketing authorisations; equally, the data required in an MAA for existing active substances and medicinal products are given in the following guideline.) This guideline states that data generated in one of the three areas that are party to the ICH process are valid in any of the other two areas.
The primary purposes of the guideline are to explain what data are required to demonstrate that the quality of an active substance or a medicinal product is maintained under a variety of environmental stresses (eg, temperature, light and humidity), and to determine the recommended storage conditions, retest periods, and shelf-lives.
The data requirements described differ slightly for active substances and for complete medicinal products. All the following areas of stability testing are required to be carried out for active substances. All except "stress testing" and "formal studies" are required for medicinal products.
The areas of stability testing required are as follows:
General For an active substance, information on its stability forms a core part of the stability evaluation of the product. The information gained on stability of the active substance is used in determining the stability of the formulated final product.
Stress testing All chemicals are inherently unstable and degrade along predetermined pathways. Stress testing assesses the degree of stability residing in the active ingredient.
Formal studies The active substance must remain within the predetermined specification limits during its expected shelf-life and the retest period.
Selection of batches For active substances, at least three batches must be subjected to accelerated and long-term stability tests of at least 12 months' duration. Equally importantly, the quality of the ingredients used in the manufacture of the active substance must remain the same during initial studies, manufacturing scale-up, and the final manufacturing process. This is to permit extrapolation of the stability test results. After the marketing authorisation approval has been granted, the first three production batches of active substance must be placed under long-term stability testing if the results were not available for submission with the MAA.
For the finished medicinal product, the containers and closures to be used in distribution of the product must be the same as those used in the stability tests. The guideline states that "data on laboratory scale batches is not acceptable as primary stability information".
Test procedures and criteria Initial tests on the active substance will have identified the ways in which degradation of the active substance will be most likely (eg, physical, chemical and/or microbiological). Tests must, therefore, be selected that most closely monitor those pathways for degradation.
The range of testing on the formulated product is similar, but with particular attention on the preservative efficacy testing and assays on stored samples.
Specification The active substance used in preclinical and clinical studies will indicate the limits of acceptability of degradation products. For the finished formulated product, the shelf-life specification must be justified based on changes observed on storage. Upper limits for degradation products must be stated and will be influenced by the changes seen in finished product samples prepared for preclinical and clinical trials.
Storage conditions For an active substance, confidence is required to ensure that the material remains stable during storage and subsequent use. Similar storage conditions should be used in storing the active substance and the finished medicinal product. Long-term testing should be carried out at 25C +/- 2C and 60 per cent relative humidity (RH) +/- 5 per cent for at least 12 months at the time of submission of the data; accelerated testing should be carried out at 40C +/- 2C and 75 per cent RH +/1 5 per cent over at least six months. Data from accelerated testing at an intermediate condition may be used to indicate storage stability when the materials are stored outside the label conditions (eg, on shipment).
Whereas the failure to meet the required specification in accelerated testing is defined as a "significant change" for the active substance, for the finished medicinal product there are slightly less stringent requirements. A "significant change" for the finished medicinal product includes a 5 per cent loss of potency from the initial assay value of a batch, any specified degradation product exceeding its specification limit, the product exceeding its pH limits, or dissolution exceeding the specification limits for 12 tablets or capsules.
Testing frequency The frequency of testing should be adequate to define clearly the stability characteristics of the active substance or medicinal product. Testing under the defined long-term conditions should take place every three months for the first year, every six months for the second year, and annually thereafter. For the finished medicinal product, matrixing or bracketing can take place if justified.
Packaging/containers The packaging used in the storage and distribution of the active substance or the finished product should be a mock-up of or the same as that used in the testing process.
Evaluation The stability study is intended to ascertain that the active substance and the finished product will remain within defined physical, chemical, microbiological and quality limits when retested at predetermined times. One way in which this can be assessed is to "determine the time at which the one-sided 95 per cent confidence limit for the mean degradation curve intersects the acceptable lower specification limit". Applying appropriate statistical limits can be beneficial for batches of material where the variability is low.
It may be possible to extrapolate real-time data beyond the observed range to define the expiry date at submission of the data, especially if this is supported by the accelerated testing data. However, this must be justified in the submission. Any actions undertaken on the stored finished medicinal product (eg, reconstitution or dilution) must also be evaluated in the stability testing.
Statements/labelling It may be possible to state a storage temperature range which should be based upon the stability evaluation. Specific storage requirements (eg, "do not store at freezing or below") may be applicable. Vague terms (eg, "ambient temperature" or "room temperature") are not to be used.
This stability testing note for guidance is an annex to the previous guideline, which should be consulted for basic principles. It is short and merely defines new dosage forms.
A "new dosage form" is defined as a medicinal product that comprises a different pharmaceutical product, but contains an active ingredient that is already approved by the regulatory authority. It includes products given by a different route of administration, a new delivery system, or a different dosage form of the same route of administration. The testing carried out should be based on that used for the parent active substance, although a reduced stability database (eg, six months' accelerated and six months' long-term data) may be acceptable when justified.
This photostability note for guidance is an annex to the guideline CPMP/ICH/380/95 (see above), which should be consulted for basic principles. It does not include photostability testing of the product during or after administration.
Light testing is an integral part of stability testing. All new active substances and new medicinal products should be subjected to light testing to ensure unacceptable degradation does not occur. Any changes made to the product require light testing to be repeated.
Testing should be carried out on the active substance, on the exposed product outside the immediate pack, on the product in the immediate pack and, if necessary, on the product in the marketing pack. The extent of acceptable limits for change of the product is to be defined by the company making the marketing authorisation application.
The sources of light to be used for testing are recommended in the guideline and include:
Samples of the material should be exposed to light that gives not less than 1.2 million lux hours and an integrated near UV energy of not less than 200 watts per hour per square metre.
Active substance There are two elements to the photosensitivity challenge: forced degradation testing and confirmatory testing. For forced degradation studies, simple solutions or suspensions of the active ingredient can be tested in chemically inert and transparent containers. It is possible that the degradation products identified during forced testing would not occur under normal testing conditions. In such instances, if the degradation products are absent from the confirmatory studies, then they do not need to be investigated further. The purpose of confirmatory studies is to ensure that the ideal conditions for handling, packaging, and labelling are used.
It is possible to carry out photostability testing of the active substance using only a single batch of the material. If it is clear that the active material is photostable or photolabile, no further tests need be carried out. If the results are equivocal, additional batches should be tested.
Samples are examined at the end of the testing period. Any obvious signs of physical deterioration (eg, a change in colour, or the development of an opaque solution) should be recorded.
Medicinal product Tests are carried out in the following order: the fully exposed medicinal product, the product in any intermediate pack and, finally, on the packaging to be used for the distribution of the medicinal product. Testing is continued until it is clear that the product is adequately protected from exposure to light. If the immediate pack is clearly impermeable to light (eg, it is an aluminium tube), the tests should be carried out only on the directly exposed product.
If a product is to be used in such a way that it will be exposed to light on use (eg, a skin cream), tests should be carried out to ensure that it is photostable on use. Equally, a product should be tested in such a way that it receives the maximum exposure to the light source (eg, tablets should be exposed in a single layer).
Samples should be examined to ensure that there have been no gross physical changes. Assay of possible degradation products should also be carried out. For solid oral dosage forms, at least 20 tablets or capsules should be tested. Care must also be taken in sampling from other preparations that may have broken down into non-homogenous components.
If there have been significant changes in the product, it may be necessary to have special labelling or packaging such that the product remains of an appropriate quality throughout its life.
If the therapeutic action is to be prolonged, or attempts are made to reduce possible toxic effects, a formulation may be modified such that the release of active substance is the rate-limiting step in its absorption. Such formulations are described as "prolonged release", and the guideline uses this term in preference to the terms "controlled release" or "sustained release".
Development pharmaceutics The reasons for the development of the prolonged release formulation should be given, along with the pharmacokinetic and physico-chemical parameters of the active substance. It should be explained how the prolonged release is to be achieved, whether the formulation is a single disintegrating unit or a multiunit pelletised system, and the release mechanism and kinetics (if known).
In vitro testing using a pharmacopoeial dissolution test should be carried out to determine the release rate of the active substance. Each strength of the prolonged release product should be tested, and any actions to be undertaken prior to administration by the patient (eg, if the tablet is to be halved) should also be reflected in the testing procedure. In vivo testing will permit determination of the bioavailability of the active ingredient. Correlation of the in vitro and in vivo testing results should be attempted.
Manufacturing process validation The details of the manufacturing process should be given that include the critical process parameters. If the validation has been carried out using small-scale batches, the dissolution characteristics of the scaled-up manufacture of the product should be given.
Control tests A dissolution test is normally included in the finished product specification. This ensures consistent substance release from batch to batch and allows acceptance limits to be set for the dissolution profile during the product's shelf-life. The number of dosage units tested should be stated. At least three points should be determined: an early one to ensure that all the active ingredient has not been eliminated immediately, an intermediate point, and a later point to demonstrate that the majority of the substance has been released. Like all test methods, the dissolution assay should undergo analytical validation, which includes determining the stability of the active substance in solution in the dissolution medium. At least three batches of production scale should be validated, and individual dosage units dissolution test results included.
Stability The dissolution profile of the active substance must be maintained throughout its proposed shelf-life.
Changes to products If in vivo/in vitro testing correlation has been achieved, minor modifications to the composition of the product or its manufacture may be acceptable without further correlation being obtained. More significant changes usually require the correlation to be reassessed.
Robin J. Harman is a freelance pharmaceutical and regulatory consultant based in Farnham, Surrey
| 1. Harman, RJ. The drug development process: 1. Introduction and overview. Pharm J 1999;262:334-7. |
| 2. Harman, RJ. The drug development process: 2. Administrative processes and options for registration of medicines. Pharm J 1999; 262:928-31. |
| 3. Harman, RJ. The drug development process: 3. Quality issues (I). Pharm J 1999;263:394-8. |
| 4. Rules governing medicinal products in the European Union (vol 3A). Guidelines: Medicinal products for human use - quality and biotechnology. Luxembourg: European Communities, 1998. |