The
Pharmaceutical Journal Vol 266 No 7133 p122
The drug development process |
|
5. Pharmacotoxicological studies (I) |
|
By Robin J. Harman, PhD, MRPharmS |
|
|
In this fifth article in the series on the drug development process, details are given of toxicological studies undertaken during drug development. These form the “safety” 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 and fourth articles dealt with the quality issues in pharmaceutical development.3,4 This article covers the safety aspects of the development process. Scientific assessment and achievement of quality, safety and efficacy form the three main criteria that any company wishing to place a medicinal product on the market must satisfy. The registration procedure utilises the considerable scientific knowledge that pharmacy undergraduates are uniquely taught. Safety issues cover all aspects of the preclinical development of a new product. They are covered in Part III of the marketing authorisation application (MAA).1 In essence, this covers the following:
Attention has already been drawn to the recognition of the importance of trying to limit the use of animals in drug development studies.1 Nevertheless, studies undertaken on animals are still a requirement of regulatory authorities in assessing the potential beneficial and harmful effects in humans. To assist applicant companies in the EC to provide the necessary information for each of the above topics, guidelines or Notes for Guidance have been produced by the EC Committee for Proprietary Medicinal Products (CPMP) of the European Agency for the Evaluation of Medicinal Products (EMEA). Although the guidelines have no legislative standing, applicants must justify their own processes where they differ from those given in the published guidelines. Adopted and draft EC safety guidelines are listed in Table 1 (below). Those that are currently under development within the International Conference on Harmonisation of the Technical Requirements for the Registration of Human Medicinal Products (ICH) and which are therefore also to be adopted within the EC are listed in Table 2. The guidelines on safety aspects of the MAA are published as a compendium
in Volume 3B of the “Rules governing medicinal products in the European
Union”.5 The following summary of safety requirements is based upon the
guidelines published in this volume. (The numbering of EC guidelines is
convoluted and inconsistent. Later documents, or earlier ones that have
been the subject of revision, are numbered as in Tables 1 and 2 above;
earlier ones still can only be referenced by their original EC designation.
The designation in Volume 3B of the “Rules” is given in the discussion
below.) |
|
|
Single-dose toxicity The guideline on single-dose toxicity covers the qualitative and quantitative study of toxic phenomena and their occurrence related to time after a single administration of a substance or a combination of substances. Its primary use is in determining what effects acute overdosage in humans might produce, and by what mechanism(s) death arises. It may also assist in designing repeated-dose toxicity studies in animals. The guideline specifically requires that the maximum amount of information be obtained from the minimum number of animals. Product specification The finished product and the active substance should ideally have similar impurity profiles; if this is so, the toxicity profiles will also be similar. If it is not so, comparative toxicity studies must be undertaken. The physical characteristics of the product’s ingredients will be the most likely cause of variations between the finished product and the active substance. The formulation to be marketed as the finished product should be used in large animal (eg, primates) studies. If the pharmaceutical formulation might lead to changes in the bioavailability of the active substance between species, this is particularly important. As indicated in a previous article,3 when a new excipient is used, toxicity studies must be carried out as if it was a new active substance. Combinations of substances in finished products (whose use is not normally recommended) require special studies. Each individual substance should be tested, as well as the combination, to exclude any unusual toxic effects. The active ingredient or finished product might degrade over time. The toxicity of any degradation products must also be determined.
Animals Equal numbers of both sexes of at least two mammalian species of known strain should be used for single-dose toxicity studies. Qualitative study of toxic signs should be recorded. Qualitative study of toxic signs and quantitative determination of the approximate lethal dose can be studied using rodents. To minimise use of animals, failure to find any difference in toxicity between sexes in the first rodent species permits use of only a single sex in the other acute toxicity studies. Each study should record for each animal used the species, strain, age, sex, weight, origin, whether vaccinated, whether free of specific pathogens, and the time spent in the laboratory prior to testing. Details of how the animals are kept, including the amount of water given and the diet, should also be recorded.
Administration The route and conditions of administration of the substance under test and the dose given must be detailed. Two routes of administration are essential, one of which should be the intended use in humans, and one to ensure circulatory access. If the intended use in humans is only by IV injection, only this route need be tested in animals. All details of the formulation of administration must be given. If it is not of neutral pH, possible formulation toxic effects must be taken into account. As wide a dose range should be used as possible to show all toxic effects. The dose-effect range and approximate lethal dose should be determined in rodents. Observations Regular observation is essential to track development of any toxicity. The period of observation should usually last 14 days, but may be extended as long as progressive toxic signs are seen.
Autopsy Some animals will die during the study; others that survive must be killed at the end. Bodies should be subjected to autopsy.
Presentation of data An assessment of morbidity should be given for each species used, at each dose, and for each route of administration. Calculations (eg, of lethal dose) must be fully described.
Repeated-dose toxicity - EC note for guidance CPMP/SWP/1042/99 For long-term treatment in humans with an active substance, repeated-dose toxicity studies are required. The length of the toxicity study depends upon the expected length of treatment in humans. The maximum length of repeated-dose toxicity studies is six months. (This has been agreed globally within the ICH process.) This is even if the treatment is to be continuous, or the body retains a single dose for a long (unspecified) period. For studies longer than three months, “sub-acute” toxicity studies may be carried out (see below) to determine the dose-range for the repeated-dose toxicity studies.
Specifications with regard to the substance and its administration The finished product and the active substance should ideally have similar impurity profiles; if this is so, the toxicity profiles will also be similar. If it is not so, comparative toxicity studies must be undertaken. The physical characteristics of the product’s ingredients will be the most likely cause of variations between the finished product and the active substance. The formulation to be marketed as the finished product should be used in large animal (eg, primates) studies. If the pharmaceutical formulation might lead to changes in the bioavailability of the active substance between species, this is particularly important. When a new excipient is used, toxicity studies must be carried out as if it was a new active substance. The active substance should be given by the route of administration intended for use in humans. Pharmacokinetic studies should be able to demonstrate how much is absorbed from the site of administration. When mixed with food or dissolved in drinking water, demonstration of reasonable absorption is necessary. Changes in consumption and growth of the test animal must also be taken into account. If the product is to be given other than orally, any local toxicity at the site of administration must be detected. The active substance should be given every day unless the rate of elimination is especially slow. If elimination rates are high, dosing more than once a day may be required. Three ranges of dose should be selected: high, intermediate and low. This will allow demonstration of a range of toxic effects, from organ toxicity at one extreme, to the desired therapeutic effect with minimal toxic effects and blood levels comparable with those expected in humans at the other. Control groups of animals not given the active substance should also be simultaneously tested.
Specifications pertaining to the experimental animal Pharmacokinetic and metabolic studies in different animal species will indicate which produces a response most similar to that expected in humans. The therapeutic range (the dose range between a therapeutic effect and toxic effects) should be determined. The choice of species and strain should be justified. Normally, equal numbers of both sexes should be used in the study. A recurrent problem is choosing the number of animals to be tested. International agreement has dictated that the minimum number of animals to show the required effects should be used. However, there must be sufficient to ensure that all possible toxic effects are seen and that animals can be subject to autopsy during the study without affecting the final statistical analysis. Additionally, there should be sufficient animals that some can be retained after the study to see if the toxic effects are reversible. At least two species must be tested, one of which should not be a rodent. The choice should include a species which produces a response close to that expected in humans.
Animal husbandry Strict control of the conditions (eg, the environment and diet) in which the animals are kept is essential. These must be detailed in the study report.
Observations As mentioned above, all data on test animals must be compared with controls. Accurate baseline data on all physiological, morphological and biochemical values for the control animals is therefore essential. General monitoring on both test and control animals is required throughout the study, including body weight, clinical chemistry, haematology and ophthalmology. Pharmacokinetic parameters and toxic effects will determine the frequency at which monitoring should take place. It is also possible that, if the drug is given in the food, intake may be affected. There must be compensation for this to ensure dosing levels are maintained. Comprehensive autopsy of all animals must take place. The guideline lists those tissues from control animals and those given the high dose that must be histologically examined during the autopsy. Less extensive histopathology is required in rodents: only those tissues that show macroscopic pathological changes need be examined. If only small numbers of other species are used, examinations must be conducted on all tissues from control animals and those given all doses.
Immunointerference Immunological changes produced by the drug may cause adverse reactions. Even if no immunological effects are expected, those organs most susceptible to such changes (the spleen, thymus and some lymph nodes) must be examined macroscopically and microscopically at the end of the study. Appendix A to the guideline lists those tissues that must be histologically examined during the autopsy. Appendix B describes the special requirements for drugs that are given by inhalation.
Repeated-dose tissue distribution studies - EC note for guidance CPMP/ICH/385/95 Pharmacology and toxicology studies can be interpreted more
effectively if the absorption, distribution, metabolism and excretion
of the active substance is well defined. Usually, single-dose tissue distribution
studies are sufficient. However, repeated-dose tissue distribution studies
may assist in the
Circumstances under which repeated-dose tissue distribution studies should be considered The following are instances when repeated-dose tissue distribution studies should be considered:
Design and conduct of repeated-dose tissue distribution studies All of the previous studies undertaken with the active substance should form the basis for the design and conduct of the repeated-dose tissue distribution studies. Using radiolabelled substances, dose levels, duration of dosage (usually using a minimum of one week) and organs to be examined all need to be considered.
Detection of toxicity to reproduction for medicinal products — EC note for guidance CPMP/ICH/385/95 All active substances must be tested for potential reproductive toxicity. Tests should be carried out using animals during defined stages of reproduction. For detection of potential long-term reproductive problems, studies over one or two generations of animals may be required. The type of study to be undertaken is determined by the expected use of the drug product, its formulation and route of administration, data already collected on toxicity, pharmacodynamics, pharmacokinetics and the known effects of related compounds. A range of studies should be undertaken covering all stages from conception to sexual maturity; observations should also be continued through one complete life-cycle (from conception in one generation to conception in the next generation). Any further studies undertaken to investigate detected effects on reproduction are chosen on a case-by-case basis.
Animal characteristics The characteristics of animals used in studies must be well defined (eg, by their health, age, prevalence of any abnormalities and fertility). If young mature adults and virgin females are used, the animals are likely to be of a similar age and weight. Mammals should be used, ideally of the same species and strain as used in other toxicological studies. The rodent order of mammals is the most commonly used; in embryotoxicity studies, a second non-rodent mammal (usually the rabbit) is required. Other test systems are continually being developed (eg, mammalian and non-mammalian cell systems, both in vitro and in vivo), but they tend to lack the integrated sophistication of animal studies. They are, however, a useful adjunct and may indirectly reduce the required number of animals.
General recommendation concerning treatment Dosage selection is critical. The highest dose is selected first, based on previous pharmacological, toxicological and pharmacokinetic studies. Lower doses are selected from the higher dose in intervals based on known toxicity and other factors. The same route of administration as is to be used in humans should be used. The dose should usually be given once daily but, again, pharmacokinetic data may decrease or increase this frequency. As with all toxicity studies, control groups must be used under identical conditions.
Proposed study designs — combination of studies Most commonly used is the three-study design to study a combination of effects on:
Fertility and early embryonic development tests will demonstrate toxicity prior to mating and implantation. In females, this includes effects on the oestrous cycle, tubal transport and implantation; in males, it will show effects on sperm motility and maturation, and libido. At least one species is used, usually the rat. Effects on the pregnant and lactating female, and on the development of the foetus and offspring during weaning, will be shown by studies on pre- and post-natal development, including maternal function. In case of delayed toxic effects, observations are continued through to sexual maturity. Adverse effects that might be expected include pre- and post-natal death of offspring, and altered growth and development. At least one species is used, usually the rat. Studies on embryo-foetal development will demonstrate altered foetal growth and structural changes prior to closure of the hard palate. Two species are normally used: the rat and the rabbit. Alternative study designs are accepted in certain cases, which must be individually justified. In a single-study design in rodents, the fertility and early embryonic development stage and pre- and post-natal development stage are combined. A two-study design can be similar, but with mandatory foetal examination. Alternatively, female treatment in the fertility study could be continued until closure of the hard palate (with examination of foetuses), combined with pre- and post-natal development studies. This option uses considerably fewer animals than the designs described above.
Statistics All results must be statistically interpreted using descriptive statistics (in which the relationship between variables and their distribution is determined) and inferential statistics (establishing the statistical significance of the results).
Data presentation Individual values for each animal in the study must be clearly tabulated, permitting the history of the animal to be followed from its conception to autopsy. Less frequent positive indications (eg, of clinical signs and autopsy findings) should be individually grouped in tables.
Testing of medicinal products for their mutagenic potential The guideline opens with the following definition of mutagenesis: “Mutagenesis refers to those changes in the genetic material in individuals or cells brought about spontaneously or by chemical or physical means whereby their successors differ in a permanent and heritable way from their predecessors.” Not only are future generations at risk from genetic changes, the individual taking the chemical is also susceptible to a potential cancer risk.
Objectives of a mutagenicity testing procedure There are different tests available for mutagenicity detection. The procedure chosen should be of maximum accuracy and of reasonable cost, and be capable of detecting the main types of genetic damage: mutations of either genes, chromosomes or genomes. The procedure must also take account of the different organisation of genetic material in prokaryotes and eukaryotes. It should also be recognised that the ability of different organisms and different test systems to metabolise xenobiotic compounds greatly varies. Both in vitro and in vivo tests must be used in such instances.
Proposed mutagenicity tests for medicinal products It is the characteristics of the compound under investigation that determines which combination of tests should be used. There are four categories of tests; one test from each category must usually be used. The most widely used tests are for gene mutations in bacteria, which are carried out using known and well-characterised bacterial strains. Such tests can detect frame shifts and base change mutations in DNA. Tests for chromosomal aberrations in mammalian cells in vitro utilise human lymphocytes or mammalian cell lines. Assessment of damage at mitotic metaphase during DNA replication indicates mutagenic potential. The test for gene mutations in eukaryotic systems can be used in both bacteria and organisms with complex eukaryotic chromosomal structures. Eukaryotes as complex as fungi and insects may even be used. Confirmation of the results from the above in vitro tests is required by the use of the fourth test, the in vivo test for genetic damage. Chromosomal damage is indicated by the bone marrow metaphase and micronucleus tests, and the dominant lethal test. The mouse spot test is widely used. Interpretation of the results Extrapolation of the above tests to assessing potential mutagenic damage in humans is especially difficult. Negative results in all four categories of test may suggest low mutagenic potential. The converse results might indicate a high mutagenic risk. Usually, however, the results produce some positive and some negative results, as a consequence of the tests’ different end-points. It may therefore be necessary to carry out supplementary tests, whose selection is based upon the intended therapeutic use and the drug’s other properties.
Risk/benefit considerations Consideration of many other factors is necessary in making a risk/benefit assessment of mutagenic potential. These include the pharmacokinetic, metabolic and toxicity profiles; and the intended therapeutic use, the age and reproductive status of the patient and the intended length of treatment with the medicinal product.
Genotoxicity: specific aspects of regulatory genotoxicity tests for pharmaceuticals - EC note for guidance CPMP/ICH/141/95 EC note for guidance CPMP/ICH/141/95 is an extension of, and in part replaces, some of the above guideline, “Testing of medicinal products for their mutagenic potential”. In its introduction, explanation is given of the different guidelines currently operational in the EC, in Japan by the Ministry of Health and Welfare, and by the US Food and Drug Administration (which in fact uses for pharmaceuticals the FDA Center for Food Safety and Applied Nutrition guidance). The guideline is highly technical in character, and details:
Explanatory notes, a glossary and (almost unique in EC guidelines) a detailed list of references used in the compilation of the text are also included.
Carcinogenic potential Detection of carcinogenicity is facilitated (but not guaranteed) by almost all known human carcinogens being carcinogenic in experimental animals. However, the converse is not always true, and extrapolating results from animal studies to humans is difficult and sometimes arbitrary. Factors which affect extrapolation include, in humans, the intended use of the compound, the dose, route of administration and, in animals, the species tested and the incidence of tumours in specific tissues.
Requirements for carcinogenicity studies Studies will need to be carried out if the agent is to be administered regularly, either continuously over at least six months, or frequently and intermittently such that the amount of drug administered is equivalent to the continuous administration. They are also required when the chemical is related to one of a similar structure of known carcinogenic potential, or where its biological action, long-term toxicity, or mutagenic study results may indicate a potential for carcinogenicity. Some agents used in the latter stages of terminal disease are themselves carcinogenic in the longer term. However, if the patient’s life expectancy is shorter than the period in which carcinogenic effects from the drug will develop, carcinogenicity studies are not required. Similarly, insoluble agents which are not systemically absorbed need not be tested in detail for carcinogenicity.
Species and strain selection Two species should be used in which the metabolism of the compound is known and should be preferably similar to the metabolism expected in humans. Species and strains known to be sensitive to known carcinogens must be used in the studies. Routine use of positive controls is not required. However, the detection of spontaneous tumours in the strains used (but not necessarily undergoing the testing) should be noted.
Dosage Anticipated dosage regimens in humans should ideally be used in the studies (with at least daily dosing), and absorption of the substance should be demonstrated. Three dose levels are to be used. The highest dose should generate a minimum toxic effect. This could be target organ toxicity, as shown by organ failure and subsequently identified pathological changes. The lowest dose should be two to three times the maximum anticipated human therapeutic dose or pharmacologically effective dose in animal studies. The mean of the high and low dose is chosen as the middle dose.
Practical features All animals used must be healthy, and studies started immediately after weaning. They must proceed for 24 months in rats and 18 months in mice or hamsters. A high survival rate may require extension of the study to 30 months in rats and 24 months in mice. Each group must comprise 50 animals, with 50 animals of each sex in the control groups. Provision of a uniform, clearly specified diet throughout the study is essential.
Additional monitoring Testing must be carried out to ensure maximum determination of carcinogenic data. Any additional data expected to be derived from the studies must not prejudice the primary objective.
Statistical design of the study All attempts must be used to minimise bias from the study and control groups. This can be as simple as ensuring that one part of the animal house is not environmentally different from another, and keeping a proportion of each group of animals similar to that in the main group if the study’s start has to be staggered because of the large number of animals undergoing testing.
Terminal investigations Animals may die during the study, or may have to be humanely killed. Similarly, all animals must be sacrificed at the end of the study. Autopsies are essential on all animals, irrespective of the reasons for their decease. Previously discovered toxic effects may suggest particular organs to examine, and determination of the reasons for toxic effects may be assisted by biochemical and haematological studies. All tissues and organs listed in an appendix to the guideline must be subject to microscopic examination from all animals given the high dose, and from all control animals. If visible signs of damage are noted, similar examinations are required in the affected tissues. The presence of tumours in any tissues requires the corresponding tissues in the middle- and low-dose animal groups to be similarly examined.
Principles of reporting on carcinogenicity studies All discovered tumours should be classified according to international definitions (eg, those produced by the World Health Organisation). Raw data presentation should include the number of animals examined and the results of macro- and microscopic examination. Results must also include the numbers of animals with tumours, the total number of malignant tumours found in an animal, and the length of survival of each animal.
Analysis of the data The test results must be assessed as follows:
Comparison of the results between each of the three dose groups and between the two control groups is required. It is also necessary to determine independently any dose-related effects of the substance. The guideline recognises that different studies may need different statistical approaches. Consideration of other factors (eg, death of study animals from other diseases, and early sacrifice of animals showing clinically detectable tumours) may also require statistical analysis. An increased incidence of tumours in animals receiving the substance compared with control groups is always significant. There may be an increased incidence or reduced latency of malignant tumours, an increased incidence of benign tumours, or local induction of tumours at an injection site.
Use of short-term carcinogenicity studies The guideline unequivocally states that short-term carcinogenicity studies are no substitute for formal carcinogenicity testing in animals. Positive results from short-term studies always require formal carcinogenicity testing for further development of the substance to a marketable product. |
Robin J. Harman is a freelance pharmaceutical and regulatory consultant based in Farnham, Surrey
