Скачать 70.34 Kb.
THE THREE Rs CONCEPT & PROGRESS WITH IMPLEMENTING NON-ANIMAL ALTERNATIVES IN BIOMEDICAL RESEARCH & TESTING
Robert D Combes (FRAME)
The Three Rs concept of reduction, refinement and replacement as applied to laboratory animal experimentation was first proposed by Russell & Burch in their treatise entitled The Principles of Humane Experimental Technique, published in 1959 (Russell & Burch, 1959). The gradual adoption of the Three Rs approach to laboratory animal experimentation has resulted in best scientific practice, whilst achieving, as far as possible, a reduction in the numbers of animals used and refinements in experimental techniques, such that any pain, suffering or distress caused to laboratory animals is minimised (Balls et al., 1995a). Also, and most importantly, there has been an increase in the use of non-animal methods, mainly either as screens before animal experimentation is undertaken, or to be used as adjunct methods for investigating biological mechanisms or as complete replacements to obviate the need for some existing animal methods. There are several comprehensive texts, conference proceedings and handbooks dealing with issues concerned with the ethics of animal experimentation and the application of the Three Rs concept, including Tuffery (1995), Smith and Boyd (1991), O’Donoghue (1998), van Zutphen and Balls (1997), Wolfensohn and Lloyd (1998), Poole (1999) and Grayson (2000).
Legislation on animal experimentation
In many countries, including Member States of the European Union, investigators are compelled by legislation to declare that they have given full consideration to the possible use of replacement alternatives, when completing a project licence application form (Dolan, 2000). This requirement is endorsed by EC Directive 86/609/EEC, and the Council of Europe Convention for the Protection of Vertebrate Animals Used for Experimental and Other Scientific Purposes (1986) which require that:
“An experiment shall not be performed if another scientifically satisfactory method of obtaining the result sought, not entailing the use of an animal, is reasonably and practicably available.”
In the UK, the above is applied in the form of the Animals (Scientific Procedures) Act 1986 to protected species (i.e. all living vertebrates, except humans, Octopus vulgaris, and foetal and larval forms of vertebrates, at specific stages of development). The legislation is enforced by a Home Office Inspectorate and a licensing procedure, with individual inspectors visiting establishments approved for carrying out animal experimentation.
The above legislation is based on the so-called extrinsic argument that animals can be used in experiments under certain circumstances, after assessing the costs in terms of the numbers of animals to be used and the severity of the proposed procedures and the benefits to society in general (e.g ranging from increased fundamental knowledge and developing a new cure for a disease, to the marketing of safe products). Thus, in the UK, every scientific project involving protected species and regulated procedures (e.g. withdrawing blood or dosing with a potentially toxic chemical) has to be approved according to the cost-benefit analysis. In general, inspectors seek to approve research that causes the least cost to the animals with the greatest likely benefit and involving the highest quality research. In addition to regulation by the HOI, prospective licence holders now have to satisfy the requirements of a local ethical review process (ERP) in the establishment where they work before their proposals for animal work are submitted to the Home Office (Combes, 1999a). Currently, certain work in the UK is officially banned, and this includes the testing of cosmetics products and new ingredients intended mainly for cosmetics, and also tobacco and alcohol products. In addition, no experiments on great apes or the production of monoclonal antibodies by using the mouse are allowed.
Each establishment authorised to undertake animal experimentation in the UK is issued a certificate of designation, which is held by a Certificate Holder who is responsible for ensuring that all work is conducted according to legislative requirments. Additionally, several individuals at the establishment are responsible for adhering to legislation and for ensuring that work is only conducted according to the specifications laid down in each particular project licence. These individuals are the Personal Licence Holder, the Day to Day Care Person (for the animals concerned), the Personal Licence Holder (who conducts permissible scientific procedures on the animals), and the Named Veterinary Surgeon (who is directly answerable and responsible for the overall welfare of the animals, and who should be unconnected with the research work).
Scientists applying for licences in the UK have to undergo rigorous training courses covering aspects of legislation, veterinary science, husbandry, health monitoring, bioethics, licence management and the Three Rs. Project licence applicants have to sign a declaration that they have failed to find non-animal methods to achieve their stated scientific objectives. They are being increasingly encouraged to adopt a decision-tree approach to animal experimentation in which they should establish first if any published information exists to allow the use of a replacement method to achieve all or some of their objectives. Failing this, they need to establish initially if any existing replacement method can be adapted or a new technique developed to meet their needs. If not, then they should consider if the research strategy can be modified to permit the use of a replacement method. If none of these is possible, then applicants need to apply reduction and refinement techniques as far as possible without compromising the scientific quality of the work.
The Three Rs concept and alternatives
Many scientists relate the term ‘alternatives’ only to the use of techniques which completely replace the use of animals in research. Accordingly, they dismiss the idea that alternatives could be implemented in their research, as their work involves the use of in vivo methods. However, alternatives can be considered to include not only replacement methods but also reduction and refinement.
The Three Rs
Reduction is defined as a means of lowering the number of animals used to obtain information of a given amount and precision, refinement as any development leading to a decrease in the incidence or severity of inhumane procedures applied to those animals which have to be used, and replacement as any scientific method employing non-sentient material which could replace methods that use conscious living vertebrates. The importance of this approach lies in its combination of animal welfare considerations with good science and best practice (Combes, 1999a).
It is important to understand that replacement, reduction and refinement alternatives can affect each other. For example, the use of non-animal methods does not only offer the possibility of reducing the numbers of animals that might have to be used subsequently, for example, when screening out candidate chemicals during the early stages of product development, but can also lead to refinement of such animal experiments, when they are necessarily used. Also, reducing group sizes in an experiment might increase the suffering experienced by each individual animal, an example of greater reduction compromising welfare.
Russell and Burch suggested three main ways for reducing animal use: better research strategy, better control of variation, and better statistical analysis. A useful working definition of reduction is: the use of fewer animals in each experiment without compromising scientific output and the quality of biomedical research and testing, and without compromising animal welfare (Festing, 1994; Festing et al., 1998).
Although progress has been made, particularly as a result of scientific developments in in vitro alternatives in drug development and in the development of specific pathogen-free animals, there still appears to be scope for further reduction through better experimental design and statistical analysis, and the more widespread use of inbred strains, particularly in fundamental research and in toxicology. For example, a survey of 48 experiments, described in papers published in two mainstream toxicology journals, showed that 67% had obvious statistical errors. About a third of the experiments were unnecessarily large to achieve the stated aims of the research (Festing, 1996; Festing and Lovell, 1996). In some cases, the more widespread use of genetically-defined animals could also lead to an improvement in the scientific quality of the work, which in the long run should reduce animal use (Festing, 1990).
There are at least four main ways for achieving reduction: (a)- implementing better research strategies; (b)- having a high level of control of experimental variation; (c)- using powerful and appropriate statistical tests, and (d)- harmonising international regulatory testing guidelines. The implementation of better research strategies involves taking an initial decision as to whether or not an experiment is worth doing at all, carefully considering the use of non-animal methods to achieve the same research objectives, having a clear hypothesis for testing, and lastly ensuring that suitably qualified and trained staff, as well as appropriate facilities, are available, to avoid mistakes and the necessity to repeat experiments.
More control of experimental variation can often be achieved by selecting an appropriate experimental design, using the correct sex, strain and species of animal, using inbred strains, applying blocked and factorial arrangement of treatments and minimising operator error by employing suitably trained personnel in the use of reliable equipment. In order to use optimum statistical tests it is important to select them before experimentation, in relation to the design being used. It is also necessary for staff to have better training in statistics, and to encourage consultation with statisticians, to use sufficient numbers of appropriate control animals, and to clearly identify the fundamental experimental unit for comparison purposes. Lastly, harmonisation of international guidelines, especially through the efforts of the International Conference on Harmonisation (ICH), over the last few years has resulted in a substantial decrease in the numbers of animals used in the preclinical safety assessment of new pharmaceuticals (Lumley and van Cauteren, 1997). Thus, harmonisation can reduce the need for repeat testing, eliminate the requirement for redundancy in testing (where more than one test provides the same information), minimise group sizes (e.g. by obviating the need to use both sexes), and result in the adoption of shortened protocols, involving fewer animals with less severe treatments and procedures, than in previous tests.
The objective of applying refinement is to minimise the suffering and distress experienced by laboratory animals, and to maximise their well-being (Flecknall, 1994; Smaje et al., 1998). Refinements should be applied to animals from the moment they are born until they die. Methods for applying refinement include environmental enrichment, choosing species that suffer the minimum amount, handling them competently in a compassionate and caring manner, applying anaesthesia and analgesia (where appropriate), using the least invasive procedures and applying humane endpoints where possible, and humanely killing animals.
There are numerous ways to enrich the environment of laboratory animals, but these must be applied with respect to the particular type of animal being used and also in relation to the type of work being undertaken. Thus, inappropriate methods might increase the level of experimental variation and therefore necessitate using larger numbers of animals. Examples of environmental enrichment include using nesting boxes for rodents in cages, solid, as opposed to grid, flooring, and group housing. There are also important issues concerning the housing of animals during their transport which should be considered.
There are several opportunities for using more humane endpoints in animal experimentation (Hendriksen and Morton, 1999). Thus, experiments should be terminated when the objective of the experiment has been achieved. A good example of this is the suggestion to replace the LD50 acute toxicity test with the Fixed Dose Procedure, or FDP (van den Heuvel et al., 1990). The former test relies on death of half of the animals as an endpoint, whereas the FDP does not. Also, pilot studies are useful to ascertain the likely consequences of an experimental approach, and also non-invasive techniques (e.g. blood and urine sampling) can be used.
The large recent interest in developing and using genetically-modified (transgenic) animals has generated great concerns about the welfare of the animals concerned (Mepham et al., 1998). Thus, it is known that the introduction of foreign genes into the genomes of recipient animals can cause unexpected deleterious effects. Such effects might be cryptic, subtle and therefore difficult to detect. Also, genes can be introduced specifically to cause disease or an abnormality, and in some cases enormous numbers of animals are wasted in the process of generating new strains.
There has been much progress in the refinement of animal experiments over the last decade. However, we need to gather much more information about the levels of pain and suffering that different species experience when subjected to a variety of procedures, and as a result of their handling and housing. There is also an ongoing need to educate laboratory animal scientists about refinement issues, and more research is required on the scientific benefits to be gained from applying refinement techniques in the laboratory. Lastly, it is important that establishments where animal experimentation occurs foster a culture of animal welfare and care among their staff.
Replacement alternatives encompass methods that permit a given purpose to be achieved without conducting experiments or other scientific procedures on animals (Balls, 1994; Gadd, 2000). Replacement alternatives have been categorised as: (a)- relative or absolute; (b)- direct or indirect, and (c)- total or partial. Relative replacement involves the humane killing of a vertebrate animal to provide cells, tissues and/or organs for in vitro studies. Absolute replacement, however, is where animals are not need at all, for example through the use of cultures of human and invertebrate cells and tissues. An example of direct replacement involves using guinea-pig skin in vitro to provide information that would have been obtained from tests on the skin of living animals. Indirect replacement is, for example, when the pyrogen test in rabbits for microbial contamination of biological fluids is replaced by the Limulus amoebocyte lysate (LAL) test or a test based on whole human blood (Hartung and Wendel, 1997). Total replacement can be achieved by deciding not to conduct an animal procedure because of lack of justification or reliability of the method. In contrast, partial replacement involves using non-animal methods as pre-screens in toxicity testing strategies.
Types of replacement methods
The range of replacement alternative methods and approaches includes: (a)- the improved storage, exchange and use of information about animal experiments already carried out, so that unnecessary repetition of animal procedures can be avoided; (b)- predictions based on the physical and chemical properties of molecules; (c)- mathematical and computer models; (d)- the use of in vitro tissue culture methods; (e)- ‘lower’ organisms with limited sentience and/or not protected by legislation controlling animal experiments; (f)- early developmental stages of vertebrates before they reach the point at which their use in experiments and other scientific procedures is regulated, and (g)- human studies, including human volunteers, post-marketing surveillance and epidemiology.
Mathematical and computer modelling comprises a variety of approaches, including (a)- modelling of quantitative structure-activity relationships (Combes and Judson, 1995); (b)- molecular modelling by using computer graphics; and (c)- modelling of biochemical, physiological, pharmacological, toxicological and behavioural systems and processes. Tissue culture techniques are also varied, increasing in complexity from subcellular (cell-free) fractions, through cell-lines and primary cells grown submerged in liquid suspension as monolayers or as co-cultures (mixtures of different cell types) and three-dimensional organotypic cultures, to tissue slices or fragments and perfused organs, all consisting of animal or human cells.
The use of mammalian cells in tissue culture
A variety of differentiated mammalian cells can be cultured including embryonic stem cells and micromass cultures, bone marrow stromal cells, brain re-aggregate cultures, hippocampal cultures, chondrocytes, cardiomyocytes, isolated proximal tubules, umbilical vein endothelial cells, immunocytes, mast cells, as well as skin and eye cells (e.g. keratinocytes and corneal cells). It is becoming increasingly possible to obtain animal and human cells and tissues, for example from sources such as the European Collection of Animal Cell Cultures (ECACC) and the Human Cell Culture Centre (HCCC, in the USA). Common sources of human cells include material from cadavers, diseased tissue, skin strips, peripheral blood, buccal cavity smears, hair follicles and surgical waste from biopsy material, which is unsuitable for transplantation purposes. The safe acquisition of surgical waste and non-transplantable human tissues, as sources of a wide variety of viable human cells, is being facilitated by the establishment of a number of human tissue banks (e.g. the International Institute for the Advancement of Medicine in the USA and the UK Human Tissue Bank (Anderson et al., 1998)).
A major problem with culturing such cells from both animals and humans is their relatively short longevity and their tendency to de-differentiate and lose their specialized status in culture. In the past, it has been necessary to rely on the use of cell-lines, usually obtained from tumours, although this has restricted the variety of types of cells available, and also they still tend to de-differentiate.
A potential solution to this problem is to generate and use immortalized cells, and to produce cell lines that have been genetically-engineered to express certain desired specialized functions, such as cytochrome P450 (CYP) activity to act as indicators of metabolism-mediated toxicity. Immortalization involves introducing oncogenes (e.g. SV40) into cells via electroporation so that the oncogene is expressed. The idea is to generate a new cell type, that possesses the properties of both primary cells and cell lines, namely retention of differentiated status and longevity in culture. This technique has been used to immortalize a variety of different human and animal cell types, but unfortunately immortalization is often incomplete and the methodology needs considerable improvement (Pfeifer et al., 1993). In the meantime, attempts are being made to achieve the same objective by other methods, including the inhibition of intracellular activities resulting in oxidative stress (Coecke et al., 1999).
There also numerous ways of investigating the viability and function of cells in tissue culture. These techniques include ATP content, protein synthesis, total protein contact, intracellular levels of calcium and reduced glutathione (to indicate cellular stress), enzyme leakage, membrane damage, oxidative metabolism and other changes in metabolism, cell shape, and the integrity of intercellular gap junctions. Several dyes, such as neutral red, fluorescein and kenacid blue have been variously used for measuring these cellular properties, as well as the non-toxic dye, Alamar Blue (an indicator of metabolic activity).
Under appropriate culture conditions, for example the correct relative humidity and in the presence of certain growth factors, it is possible to grow cells as a multilayer, each layer exhibiting morphological and functional specialization (spatial differentiation). The cells are able to communicate with one another in three dimensions via gap-junctions. Examples of such organotypic cultures include the human epidermal equivalent skin model developed in the FRAME Alternatives Laboratory in the University of Nottingham Medical School, based in the UK (Clothier et al., 1995). This model arises from the multilayer formation of individual keratinocytes obtained from separate donor patients undergoing cosmetic surgery. Cross-sections of this system show three morphologically distinct layers corresponding to the outer, heavily cornified stratum corneum and comprising keratinised dead cells, underneath which is a layer of flattened squamous cells, and then a third layer of more rounded cells (Ward et al., 1997). This multilayer system can be cultured on an inert filter in inserts placed within multi-well plates, and maintained at the air-liquid interface, such that the outer layer is in contact with air and the layers beneath are exposed to nutrients and growth factors in liquid medium from below.
This two-compartment model reflects to some extent the situation found in situ in the skin, and enhances the use of the system as a model for investigating the toxicity of topically-applied substances, such as cosmetics and toiletries, and also facilitates the testing of insoluble test materials, such emulsions and creams.
The cells in the model retain physiological and biochemical activities, such as intercellular tight junctions and mixed function oxidase and esterase enzymatic activities, and the ability to produce cytokines following appropriate stimulation. Thus, the epidermal equivalent model can be used to detect skin irritants, barrier function (skin permeability), and the cutaneous metabolism of chemicals (van der Sandt et al., 1999). A similar multilayered system, consisting of immortalized human corneal cells has been developed elsewhere as a model for detecting chemicals causing eye irritation.
The division of the model into two compartments facilitates the assessment of various skin properties, such as barrier function and metabolism. The former property can be investigated, with and without prior exposure to a potentially damaging material, by following the passage of a topically-applied marker substance (e.g. a fluorescent dye) across the multilayer and into the bottom compartment. Metabolism of topically-applied substances can be studied by analysing the bottom compartment for the presence of any metabolites.
Until recently, a common criticism of in vitro tissue culture systems for toxicity testing was the difficulty of using them to study chronic long-term and reversible effects. However, this is no longer impossible, and several experiments, where such effects have been investigated, have shown that the information obtained is useful for distinguishing between the relative toxicities of chemicals. For example, it is possible to distinguish between mild and moderate skin irritants based on the rates of recovery from long-term dosing, rather than merely on initial levels of toxicity. The two-compartment model described above, together with the use of non-toxic indicators such as fluorescein and alamar blue, has also facilitated such investigations. This is because skin cells, cultured on an inert filter, can be removed easily from an initial exposure, washed and then analysed for their response using non-toxic indicator before re-challenging the cells and exposing them to fresh indicator. Also, hollow fibre technology is being used increasingly to generate long-term cultures of very high densities of cells (Combes, 1999b). This same technology has been applied to the development of commercially-available methods, such as Technomouse and Harvestmouse, for the in vitro production of monoclonal antibodies, as a complete replacement for the ascites mouse (Marx et al., 1997). Such developments have prompted many countries in Europe, and recently the USA, to ban the use of mice for this purpose altogether, or to encourage researchers to avoid using animals, wherever possible.
The epidermal organotypic model is being improved by developing a co-culture collagen gel system comprising keratinocytes, dermal fibroblasts, HMEC-1 endothelial cells and neural cells, to include other tissue components involved in mediating and effecting the inflammatory response. The original model is an example of a reconstituted epidermal system, and the proposed new co-culture model is an example of reconstituted skin by reconstruction, where collagen and fibroblasts are being used as a cellular dermal substitute. As a consequence of the widespread interest in using in vitro organotypic skin culture models for toxicity testing, and also for developing treatments to control the ageing process, several other commercially-available types of systems are available. The main ones are reconstituted skin by fabrication, where an acellular dermal substitute (e.g. collagen matrix) only is used, and reconstituted skin by recombination, where epidermal keratinocytes are grown on a layer of dead cells derived from a de-epidermalised dermis. These complicated models can exhibit levels of barrier function, biochemical properties and responses to toxic insult that more closely resemble those of intact skin than exhibited by more simple multilayer models or monolayers of one cell type (Roguet and Schaeffer, 1997).
The application of replacement methods in biomedical research and testing
There has been considerable effort to find replacement methods for toxicity testing, despite the fact that the numbers of animals used for this purpose are considerably lower than those used for other reasons, especially fundamental research. There are several motives for this, including the fact that regulatory guidelines require the use of animals, and test protocols often necessitate that toxic doses of chemicals are administered to animals, causing suffering (see Fielder et al., 1997; Salem and Katz, 1999; Castell and Gуmez-Lechуn, 1997). Also, as mentioned elsewhere, new methods have to be validated according to strict criteria, and the eventual regulatory acceptance of alternatives has important implications for their eventual application to other areas of animal experimentation.
Advantages and limitations of using replacement methods
Replacement methods have several advantages over using animals. First, in vitro systems can provide information in a cost-effective and time-saving manner. Secondly, information generated from replacement methods can sometimes be adequate to produce more reliable and reproducible data than can be obtained in animals. In other cases it can be used to increase the efficiency of whole-animal studies and decrease the number of animals used. In vitro systems are very useful for mechanistic investigations at the molecular and cellular level, as well as for target organ and target species toxicity studies. The ability to use human tissues in vitro obviates the need for cross-species extrapolation. As a consequence, data that are more relevant to the human situation can be generated. A good example of the potential benefits of this relate to the possibility of using human cell-based transformation systems as in vitro models of carcinogenesis and to detect human carcinogens (Combes et al., 1999). Lastly, an absence of complex body systems, which might act as confounding factors, can permit research which cannot be investigated in animals.
However, the use of in vitro systems has several limitations that include: (a)- the lack of integrated systemic mechanisms of absorption, distribution, metabolism and excretion; (b)- an absence of the complex, interactive effects of the immune, blood, endocrine and nervous systems; (c)- appropriate non-animal models are not yet available for all tissues and organs, or for all diseases and toxicity endpoints, and (d)- the nature of the test compound can complicate the interpretation of in vitro toxicity studies, due to the generation of artefactual conditions in tissue culture media.
It is important to consider these limitations and advantages in the development and utilisation of replacement alternatives, as well as during the interpretation of data obtained during their usage.
A major application of replacement methods has been for screening purposes. The principle of screening involves testing a large range of candidate chemicals for a particular purpose in rapid non-animal systems (especially computer prediction models and in vitro tissue culture). Those chemicals with desirable biological activity (efficacy) and devoid of undesirable activity (toxicity) are then tested in animals (partial replacement), having eliminated those chemicals that do not show the required activity profiles, on the basis of likely toxicity and lack of efficacy. In this way, it is possible to reduce the numbers of animal tests required to assess a given number of chemicals, and also the severity of testing on animals should be minimised by screening out ones that are likely to be toxic at an early stage. In recent years, the above approach has assumed great importance in the pharmaceutical industry in particular due to the need for high throughput screening of the very large numbers of novel molecules that can now be synthesised via combinatorial chemistry. This has prompted the development of toxicogenomics, in which changes in patterns of the expression of a series of genes, known to code for proteins involved in specific relevant processes such as stress and recovery, are monitored in response to toxic challenges (Atterwill et al., 1999).
Many new screening systems are being developed in a variety of cell types and organisms, including yeast, and in mammalian and human cell lines, using genetically-engineered cells with specific reporter genes. In this way, DNA sequences coding for a receptor are introduced on a vector molecule into the recipient cells, and the sequences can be linked to the corresponding DNA response element sequence. The latter stimulates specific gene expression in response to the binding of the receptor to a chemical ligand. In order to facilitate detection of transcription, the promoter region of the response element is ligated to the promoter region of a reporter gene, whose product can be detected colorimetrically. Receptor activation by a test chemical results in stimulation of expression of the reporter gene product, as this comes under the control of the response element (Combes, 2000).
Non-animal tests are also often more useful than animal tests for investigating the mechanisms of biological activity of chemicals. Thus, it is possible to study possible synergistic and antagonistic effects between chemicals in mixtures, the contribution of sample impurities and metabolites to the overall toxicity of a test sample, the activating and detoxifying effects of specific enzymes, structure-activity relationships (SAR), species- and organ-specific toxicity and receptor-binding affinity. Also, most importantly, it is possible to obtain accurate information on dose-response relationships, potency thresholds and relative toxicities in terms of inhibitory concentration values (e.g. IC50s), more easily than via animal experiments.
The use of test batteries and hierarchical testing schemes
The application of non-animal methods to toxicity testing has evolved from non-regulatory, in-house investigative and screening studies, through the use of other partial replacement approaches comprising the use of test-batteries to hierarchical testing schemes, and eventually to the use of complete replacement methods for regulatory purposes.
Batteries of in vitro tests have been developed and used to overcome the problem that a process in vivo might occur via several different mechanisms. Each test in the battery is based on detecting an endpoint which involves one of the mechanisms known or suspected to be relevant to the overall biological effect. These tests can all be non-animal tests, as well as animal assays, which are applied in a sequential or hierarchical approach, beginning with the non-animal methods. A good example is the use of several in vitro tests to establish the potential of a chemical to elicit genotoxic effects (DNA damage), as a screen for both inherited mutation and also carcinogenesis. Because DNA damage can arise in different ways, several tests (mainly gene mutation in bacteria and mammalian cells and chromosome damage in mammalian cells) are used. Chemicals, especially those exhibiting genotoxicity in vitro, are then further tested in animals to see if the potential effects detected in vitro are realised in the whole animal (Mitchell and Combes, 1997). A further example of a hierarchical approach to testing involves the assessment of skin sensitising potential. Here, chemicals can first be investigated for their ability to penetrate the skin, using isolated animal or human skin samples, and also for their possession of structural alerts for skin sensitisation (an SAR approach by using the DEREK expert system, see Combes and Judson, 1995). Those chemicals exhibiting both attributes can be discarded at this stage without the use of animals, whereas any further testing is undertaken using the local lymph node assay (a refined and more accurate method than the standard animal test, see Basketter et al., 1999).
Complete replacement methods and the need for validation
Despite much research, few complete replacement methods have been developed for regaulatory toxicity testing and related studies. There are two major reasons for this regrettable situation. The first relates to the fact that the development of non-animal models is very complicated and relies on having an in-depth understanding of mechanisms of toxicity. The second reason is due to the need for new methods to be rigorously validated. Validation refers to the process whereby the reliability and relevance of an assay are assessed for a particular purpose. Thus, validation is intended to assess whether a new method is going to provide useful data for predicting toxicity, and whether the same information can be obtained whenever and wherever the test is conducted using appropriate equipment, materials and by personnel with relevant training and expertise. It is also necessary for there to be a prediction model available, to allow the information from the non-animal test to be interpreted in relation to the data obtained from the animal test that is being replaced (Balls et al., 1995b).
The scale and complexity of validation is dictated by the purpose of the assay. For example, where a method is to be used for in-house purposes only, then validation can be conducted on a limited scale. If, however, an assay is designed to replace an animal method and be used widely for regulatory purposes (e.g. safety testing), then validation trials are conducted at several different laboratories in international collaborative trials. The European Centre for the Validation of Alternative Methods (ECVAM) was established by the European Commission in 1993 in Northern Italy, with the express purpose of undertaking research into alternative methods and facilitating and organising their validation. More recently, ECVAM, together with a another organisation in the USA, the Interagency Co-ordinating Committee on the Validation of Alternative Methods (ICCVAM), and in conjunction with the Organisation for Economic Co-operation and Development (OECD), agreed some internationally-harmonised criteria for validation, which are intended to facilitate the regulatory acceptance of alternative methods (OECD, 1996).
Some validation studies, such as the one designed to replace the Draize eye irritation test, were largely unsuccessful in supporting the use of one or two tests as complete replacement methods. This reflects the numerous problems that have been encountered during validation studies, not the least of which have been their complexity and high cost, the lack of relevant non-animal models, lack of co-operation between scientists involved, the use of unsuitable protocols, and problems finding sufficient high quality and reliable reference animal and human data.
However, despite the above problems and constraints, several new replacement methods have been successfully validated recently for percutaneous absorption, phototoxicity testing and for corrosivity testing (Fentem et al., 1998; Spielmann et al., 1998) However, regulatory acceptance does not automatically follow from scientific validation, and the decision of the OECD to recommend the above methods instead of animal tests in its testing guidelines is still awaited following months of deliberation. In February 2000, the Competent Authorities of the 15 Member States of the EU, however, agreed that three specific in vitro test methods should be formally adopted by the European Commission and published as part of the Annex V list of recommended testing methods in the EU. These tests are: (a)- for corrosivity testing (the Transcutaneous Electrical Resistance (TER) procedure and the EPISKINTM reconstructed human skin test), and (b)- the 3T3 NRU phototoxicity test (using mouse fibroblasts and neutral red release).
The results of another international validation, the Multicentre Evaluation of In vitro Cytotoxicity (MEIC) study, have been published recently (Eckwall, 1999). In this study, 61 in vitro assays were compared for their ability to predict the acute toxicities (human lethal blood concentrations) of 50 reference chemicals, for which human data are available. Overall, the data obtained, especially from those assays based on human cells, suggest that single doses of chemicals can often be lethal to humans via interference with basal cell functions, and certain in vitro assays can be used to predict these lethal concentrations, by providing models of the basal cell functions affected.
The prospects for the scientific validation of replacement methods for other toxicity endpoints is variable. The results of an ongoing validation study on potential screening methods for embyrotoxicity and teratogenicity, for example using embryonic stem cells (Bigot et al., 1999), look promising. In the case of skin irritation, several new tests and protocols are being assessed in an ongoing validation study. However, more research is required before skin sensitisation can be modelled in vitro, although several new approaches, based on co-culture systems, are being developed. There are also intense efforts to establish new and improved models for eye irritation, based on the mechanisms involved in chemical insult of the human cornea. Based on previous experience, it takes some 10 years between test inception and regulatory acceptance, and thus the complete replacement of animal toxicity tests is a long way off (Purchase, 1997). Great advances are being made, nevertheless, in the use of genetically-engineered cell lines, physiogically-based pharmacokinetic modelling and the application of integrated testing approaches for investigating acute, systemic and target-organ toxicity, for example (DeJongh et al., 1999). Also, new models for biomedical research are being developed on a case by case basis.
|Driver Testing, Training, Examining and Licensing: Implementing New European Union Requirements (Directive 2006/126ec – the third Directive on driving licences)||Related or contrasting ideas may be found in the sections on Animal Rights, Knowledge, Life, Medical Ethics, Nature, Progress, and Science|
|Professor and Endowed Chair, Pennington Biomedical Research Center||Testing Techniques Newsletter (Software Research)|
|Tänkande ur flervetenskapliga perspektiv|
«rational and social animal». The progress of genomics reveals a remarkable non linear relationship between the fast increases of...
|Concerning the Specific Programme “Cooperation” implementing the Seventh Framework Programme (2007-2013) of the European Community for research, technological development and demonstration activities|
|Biomedical Diagnostics Institute, National Centre for Sensor Research, Dublin City University, Dublin 9||1. ir microspectroscopy as a tool to discriminate nano-molecular alterations at cell or near-cell level in biomedical research|
|Joint wmo technical progress report on the global data processing and forecasting system and numerical weather prediction research activities for 2008||Joint wmo technical progress report on the global data processing and forecasting system and numerical weather prediction research activities for 2009|