Ethics of Research involving animals
Use of the concepts ‘Alternatives’ and ‘Replacements’
11.5 Before we consider these different areas in more detail, we need to be clear what is meant by the term alternative in the context of animal experiments. To the general public, an alternative is likely to mean an alternative method that does not involve using an animal. This is the principle encompassed by UK and EU laws, which require that animal experiments can only be carried out if the purpose of the programme of work ‘…cannot be achieved satisfactorily by any other reasonably practical method not entailing the use of protected animals.’2 However, in recent years, the term ‘Alternative’ has been applied to all of the Three Rs as an overarching term referring to any procedure that reduces the harms caused to animals in experiments, not only by replacing them (Replacement), but also by reducing the numbers used (Reduction) or by causing less animal suffering (Refinement). Such a conceptual muddle is unhelpful and in this chapter we focus exclusively on Replacements since this is the area in which there is most debate about the potential to improve on current practice.
Definition and scope of Replacements
11.6 Animal experiments are carried out to try to answer scientific questions. The term ‘Replacement’ is used to encompass methods that permit a given scientific purpose to be achieved without conducting experiments or other scientific procedures on living animals.3
For complete replacement of animals, an alternative method should not require any animal-derived biological material. Examples of such methods or approaches include the use of predictions based on the physical and chemical properties of molecules, mathematical and computer studies of biological processes, analysis of epidemiological data, research involving human participants or research on isolated human cells and tissues in culture (see Box 11.1). However, many methods considered as Replacements also use some biological material obtained from living or humanely killed animals. These include research on cells and tissues derived from living or humanely killed animals for culture in vitro and animal-derived growth supplements such as serum derived from fetal or newborn calves. These methods can be called incomplete Replacements.4
| Box 11.1: Complete and incomplete replacements
Computer studies and in vitro methods Human studies In many types of biomedical and toxicological research, animals are used because ethical considerations preclude conducting the experiments on humans. However, a number of approaches have been suggested which, in some cases, might replace the use of animals with studies on humans. These include non-invasive brain scanning to replace some experiments on primates,‡ and studies on ultra-lowdose ADME metabolism in human volunteers in the early stages of selection for potential medicines.∫ Human tissue samples can be used both for direct examination (e.g. histopathology) and in cell culture and other in vitro techniques.** * Using, for example, molecular modelling and the development of quantitative structure-activity relationships; see Combes RD and Judson P (1995) The use of artificial intelligence systems for predicting toxicity Pest Sci 45:179–94; Combes RD and Rodford R (2003) The use of expert systems for toxicity prediction – illustrated with reference to the DEREK program, in Modelling Environmental Fate and Toxicity Cronin M and Livingstone D (Editors) (London: Taylor & Francis); Assessing the cumulative effect of mutations: Kirkwood TB & Proctor CJ (2003) Somatic mutations and ageing in silico Mech Ageing Dev 124:85–92. † The European Collection of Cell Cultures (ECACC) operates a cell bank of peripheral lymphocytes from approximately 40,000 donors. Forty percent are in the form of lymphoblastoid cell lines representing around 450 genetic disorders. These lines are useful for the analysis of the role of genes in disorders that have a genetic component, for example cardiovascular diseases, Alzheimer’s disease or depression. |
11.7 Tests using invertebrates, or early developmental stages of vertebrates (i.e. before they reach the point at which their use in experiments and other scientific procedures is regulated), are also sometimes described as Replacements, even though they do not replace animals per se. For example, the horseshoe crab (Limulus) can be used to replace the pyrogen test for microbial contamination of biological fluids, which was previously carried out in rabbits.5
11.8 The term Replacement can be misleading in that it implies that an animal technique is already in place, and that a non-animal technique can directly and completely replace it. Sometimes, non animal methods may directly replace an established animal test, but they are often simply the best or only method of addressing certain scientific problems, and are used within multidisciplinary research programmes to reduce overall reliance on animal experiments. In other words they may displace or avoid, rather than replace animal experiments. We take the view that the concept of Replacement is best understood in a broad sense.
Complete Replacement
11.9 The most obvious targets for Replacement are the established animal methods used to comply with testing regulations or standard operating procedures for the toxicity testing of chemicals and biological medicines. Considerable effort has been directed to replacing these tests, such as the Draize eye-irritancy test in rabbits (see Box 11.2). Complete Replacement of these procedures has not yet been achieved, although in vitro tests are being increasingly used to identify strongly irritant and corrosive chemicals, so that animal tests are not required to screen out these compounds.6
| Box 11.2: The Draize test Developed in 1944, the Draize test, along with the LD50 (paragraph 9.14) is an animal test for toxicity. It involves placing the tested substance directly into the eye of a live, conscious animal and observing the results. The test is usually performed using albino rabbits. In 1999, 3500 Draize tests were undertaken. The test has been recently replaced by alternative approaches and in 2003 a total of 33 eye tests, including Draize and other tests, were undertaken.* Many people are concerned that the Draize test causes suffering and it has received much attention from animal protection groups. Some scientists also claim that the test is invalid because of differences between the human and rabbit eye. Rabbits have a third eyelid, a thinner cornea, a more alkaline eye than the human eye, and produce less tear fluid to wash away irritants.† It is claimed that the Draize test overestimates how irritating a product is to the human because rabbits’ eyes are more sensitive. The test is also thought by some to be imprecise because it is purely observational. The toxicity is evaluated by an investigator rather than quantitatively measured.‡ The Draize test is still widely used in the USA. In the UK it is no longer used for the testing of cosmetic products and ingredients, following the ending of animal testing for cosmetics. However, it is still used as a safety test for noncosmetic products and chemicals, and is recommended for regulatory risk assessments of chemicals and a range of manufactured products that may be deliberately or accidentally brought into contact with the eyes.∫ The Home Office has published guidance for the test. These include the following stipulations: testing should only take place when in vitro screening tests have been used to identify, classify and eliminate materials with obvious irritant potential; it should not be carried out with strongly acidic or alkaline substances, nor with substances which are already known to produce severe adverse effects on the skin.** In response to a study which claimed that a variety of valid alternatives existed,†† the Home Office concluded in 2001 that the currently available alternatives to the Draize test had significant limitations and were not suited to replace live animal use.‡‡ Research aiming to develop alternatives to the Draize test continues. This includes, for example, the use of human eye tissue obtained from tissue and organ donors, and protein solutions that can be manufactured to be sensitive to potential irritants.∫∫ * Home Office (2004) Statistics of Scientific Procedures on Living Animals Great Britain 2003 (London: HMSO). † Kaufman SR (1989) Problems with the Draize Test. Perspectives On Animal Research, Vol. 1, available at: http://www.curedisease.com/Perspectives/vol_1_1989/Problem s%20with%20the%20Draize.html. Accessed on: 16 Jun 2004. ‡ The Group for the Education of Animal Related Issues Continued |
| (GEARI), available: http://www.geari.org/faqdraize.html. Accessed on: 16 Jun 2004. ∫ See European Commission Directive on dangerous substances 67/548/EEC, Directive on plant protection products 91/414/EC and Directive on medicinal products for human use 2001/83/EC and their UK counterparts. ** Statement by Secretary of State for the Home Department, House of Commons. Hansard Written Answers for 16 Jan 2001 (pt 21), available at: http://www.parliament.the-stationeryoffice. co.uk/pa/cm200001/cmhansrd/vo010116/text/10116w21 .htm. Accessed on: 16 Jun 2004. †† Wilhelmus KR (2001) The Draize eye test Surv Ophthalmol 45:493-515. ‡‡ Statement by Secretary of State for the Home Department, House of Commons. House of Commons (2001) Hansard Written Answers for 16 Jan 2001 (pt 21), available at: http://www.publications.parliament.uk/pa/cm200001/ cmhansrd/vo010116/text/10116w21.htm. Accessed on: 6 May 2005. ∫∫ For example, EpiOcular developed by the Mattek Corporation, available at: http://www.mattek.com/pages/products/epiocular Accessed on: 16 Jun 2004; the Irritection Assay system developed by Invitro International, available at: http://www.invitrointl.com/products/irritect.htm. Accessed on: 16 Jun 2004; the Agarose Diffusion method, Cottine M et al. (1993) Critical evaluation of an Agarose Diffusion method for the assessment of eye irritancy ATLA 21: 427–40, available at: http://altweb.jhsph.edu/publications/journals/atla/ atla21_4/atla21_4b.htm. Accessed on: 16 Jun 2004; see also Draize FAQ The Group for the Education of Animal Related Issues (GEARI), available at: http://www.geari.org/faqdraize.html. Accessed on: 16 Jun 2004. |
11.10 A major success in the use of Replacements in toxicity testing was achieved in 2000, when, following a successful validation led by ECVAM (Box 2.5 and paragraph 11.32), an in vitro test for phototoxicity7 was adopted as a standard test guideline by the EU, and two years later by the OECD.8 In basic biomedical research there are also examples of where Replacement methods have successfully been applied to established methods or techniques in a particular research field. For example, monoclonal antibodies were usually produced in mice (by the ascites method (see paragraph 5.25) before in vitro methods were developed. Here, the deployment of a non-animal alternative method can be seen as complete Replacement.
Non-animal techniques as 'advanced' methods
11.11 Animal experiments are often only one part of a scientific study or programme of research. For example, developing an effective vaccine against West Nile virus, a fatal infection of horses transmitted by mosquitoes, includes the following: molecular studies of the virus, studies of virus growth and development in insect and mammalian cell lines, epidemiological studies of vector populations and disease incidence in the field, mathematical modelling of the transmission and spread of disease, and clinical studies. This work usually involves very little experimental live animal use. Some laboratory infection of horses (or small-animal models) is undertaken to examine the progression of the disease in a controlled manner, to discover the exact means of insect transmission and to develop and test candidate vaccines. In this case, the molecular and epidemiological studies are not Replacements for the animal work; they are addressing different scientific questions within the research programme.
11.12 The terms ‘advanced’ or ‘complementary’ have been applied to many non-animal methods (for example, the molecular biology and mathematical modelling techniques mentioned above) to indicate that these methods have been developed to answer specific scientific questions that animal tests cannot address. They were not developed exclusively to replace animals for ethical reasons, and it is therefore unhelpful to refer to them in claims that all animal research could easily be replaced, if there was only a will to do so.
Non-animal methods as adjuncts
11.13 Non-animal methods may act as an adjunct to animal experiments rather than replace them, but in so doing, they may serve to reduce the total number of animals used in a programme of work. A classic example is the screening of anti-cancer drugs in nude9 mice with human tumours. An initial screen using cell cultures can be used to demonstrate basic tumour cytotoxicity, and only the active toxins are tested in vivo. The same principle is used in high-throughput screening of potential medicines (see paragraph 8.6). This approach involves testing a large range of potentially useful candidate chemicals for a particular purpose in non-animal systems (especially computer prediction studies and in vitro tissue culture) using techniques that can be carried out very rapidly. Those chemicals with desirable biological activity (efficacy) and devoid of undesirable activity (toxicity) can then be selected for further study. In this way, it is possible to reduce the numbers of animal tests required to assess a given number of chemicals. The severity of animal tests can be minimised by screening out substances that are likely to be toxic at an early stage. In recent years, high-throughput screening has become widely adopted by the pharmaceutical industry (see paragraphs 8.4–8.6).
Alternative approaches
11.14 Another equally important concept is the use of an alternative approach to an experimental goal enabling the avoidance of animal use. Even where there are no obvious alternatives, any proposed scientific study should consider at an early stage not only whether the animal experiment is the most appropriate and only method of addressing each research question, but also whether the question is worth asking, and whether it justifies causing pain and suffering to a sentient animal. In other words, the first alternative to consider is the option not to carry out the experiment at all. For example, within the REACH testing programme (see Box 9.2) the first consideration might be whether a particular test is actually necessary, regardless of whether there are, for example, adjunct Replacement methods that could be used in research.
Footnotes2 See Home Office (2000) Guidance on the Operation of the A(SP)A 1986 (London: TSO), Chapter 5. Note that animals here
means the animals covered by the Act and therefore only vertebrates, and one species of octopus; see Section 5.5 (a) of the A(SP)A, Article 7.2 of Directive 86/609.
3 Balls M (1994) Replacement of animal procedures: alternatives in research, education and testing Lab Anim 28: 193–211; Balls
M (2002) Future improvements: replacement in vitro methods ILAR J 43, Supplement: 569–73; Gad SC (2000) Alternatives to in vivo studies in toxicology, in General and Applied Toxicology, Vol. 1, 2nd ed, Ballantyne B, Marrs TC and Syversen T (Editors)
(London: Nature), pp401–24.
4 Although this practice does not replace the use of animals per se, it replaces the carrying out of procedures on living animals.
Sacrificing the life of one animal can save the lives of many other animals, as its organs and tissues can be used in many different experiments. The humane killing of an animal carried out according to methods prescribed in Schedule 1 of the A(SP)A is not counted as a procedure.
5 The pyrogen test is used to determine whether a substance is fever inducing. The test involves injecting a sample of the
substance being tested, usually into rabbits. The rabbits must be individually held in a fixed position for a number of hours in
a cage. Through temperature probes placed in the rectum of the animal, increased temperature is measured and, if recorded,
gives an indication of pyrogen contamination. 3R Research Foundation Switzerland, 3R Training: Rabbit in vivo pyrogen test,
available at: http://3r-training.tierversuch.ch/content.php?ctool_page_id=134〈=en. Accessed on: 6 May 2005; Liebsch M
(1995) History of the LAL-test: validation and regulatory acceptance ALTEX 12: 76–80.
6 See OECD (2001) Series On Testing And Assessment, Number 33: Harmonised Integrated Classification System For Human
Health And Environmental Hazards Of Chemical Substances And Mixtures: ENV/JM/MONO(2001)6; Chapters 2.2 (Skin Irritation/Corrosion) and 2.3 (Eye Irritation/Corrosion), available at:
http://www.oecd.org/LongAbstract/0,2546,en_2649_34365_2671862_1_1_1_1,00.html. Accessed on: 6 May 2005.
7 While a medicine by itself may have no toxic effects, this may change in combination with light. Phototoxicity studies test
whether the toxic properties of a compound change when exposed to light. This is important if a compound is applied to a
specific area of the body that may be exposed to light, in the form of a skin cream for example. A phototoxic compound may
enhance the possibility of ultraviolet (UV) light inducing skin cancer. TNO Nutrition and Food Organisation Phototoxicity: the
combined effect of sunlight and pharmaceuticals on skin, available at:
http://www.voeding.tno.nl/ProductSheet.cfm?PNR=ZE_226A. Accessed on: 29 Apr 2005.
8 Animal tests for phototoxicity carried out before this date could in principle have been replaced by the alternative method.
The new test did not in fact replace an existing EU or OECD test guideline for an animal test until 2000/2002.