Animals and Cancer Research
Sir Walter Bodmer F.R.S.
Text from the Animals in Medical Research Information Centre (AMRIC) Lecture 1989.
Originally printed in Science and Public Affairs, 1990
The first Director of the Imperial Cancer Research Fund, Ernest F. Bashford, produced a draft of a scheme for enquiring into the 'Nature, cause, prevention and treatment of cancer.' On the basis of this he was chosen to be the head of one of the first institutions specifically devoted to cancer research in the world, and at a time when very little was known about the disease. His scheme still makes sense today and included recommendations both for extensive statistical investigations of the distributions of cancer in human populations, and proposals for studying experimental cancers in animals. There was then no other experimental approach.
Bashford and his colleagues were among the pioneers in the investigation of transplantation of tumours from one animal to another. These investigations were complicated by a total lack of understanding of the problems of transplantation rejection due to individual differences. However, it was through attempting to solve this problem that we developed our modern understanding of the nature of graft rejection, the basis for distinguishing self from non-self, and of histocompatibility antigens and their role in the immune response, came into being. To these early investigations, we now owe the success of, for example, kidney transplantation as an effective therapy for kidney disease, and bone marrow grafts for certain types of leukaemias.
Bashford, who had worked with the great Paul Ehrlich, must already have been well aware of the great advances being made through animal experiments. For example, in the understanding of responses to microbes, the nature of the immune system, and the potential of this in developing a basic understanding of cancer processes.
It is this type of basic research that continues to underlie the fundamental advances that lead to major practical applications. Promoting a non-animal based, apparently useful, application might have led to the greater and greater refinement of the iron lung for the treatment of polio without (or with a very much delayed) discovery of the much more appropriate preventative approach of vaccination.
My aim now is to demonstrate the enormous contribution that animal experiments have made to our present understanding of cancer, and the continuing need for work on animals if we are to fully exploit advances in technology and understanding of the biology of molecules, cells and whole organisms, in order to achieve the prevention and cure of most cancers. Clearly, there is so much to cover that I must concentrate on areas with which I am most familiar, and where I believe there is the greatest promise.
The major advances in the treatment of leukaemia and testicular cancer, particularly through the use of chemotherapy, have depended critically on animal experiments to test drug efficacy and therapeutic regimes, as is generally the case in drug development.
Skippers' model experiments with mouse tumours paved the way for successful leukaemia chemotherapy. These are comparatively rare tumours, so much work is now being done on more common tumours, particularly the carcinomas. These have not responded to current therapeutic approaches, and that is why new information, ideas and approaches are needed. These will come from the fundamental advances in understanding that are now under way.
The combination of known carcinogens and skin painting with oil extracts led to the realisation that cancer development is at least a two-stage process: initiation (the production of the potential cancerous cells) and their promotion (the encouragement of their outgrowth in various ways). Now we know that this is an over-simplified view and that there are many steps involved, but the general notion of promotion remains a critical feature of cancer progression. It is important for prevention because, as the general evidence suggests, most of the major environmental factors that may be identified, for example in the diet, will be of a promotional nature. This idea of promotion also underpins the understanding of why when an individual stops smoking cigarettes, the cancer risk remains as it was when they stopped.
Cancer care in general, now and for the foreseeable future, will depend on many general surgical and medical techniques and procedures. Antibiotics are essential for dealing with immunosuppressed patients and imaging techniques, such as the computer-aided tomography (CAT) scan and magnetic resonance imaging, are a key feature of diagnostic procedures.
The fundamental nature of cancer and its causes
Cancer is not a single disease. It can affect any organ of the body to different extents and with different frequencies. This complexity means that cancer research encompasses virtually all aspects of fundamental cellular and molecular biology.
Patterns of cancer incidence, and other types of epidemiological studies, indicate that as much as 80% of cancer is in some sense caused by environmental factors. In principle, this much may therefore be preventable if the factors can be identified and controlled. However, the length of time it took to identify cigarette smoking as a major cause of lung cancer, and indeed of 30% of all cancers in this country, and the difficulties faced in the 30 years plus since this discovery (by Bradford-Hill and Doll) in implementing a truly effective anti-smoking policy, shows that preventability and actual prevention are far from equivalent.
Apart from cigarette smoking, the other major presumed environmental factors lie in the diet. But the relationship is complex, and hardly ever is a particular feature of the diet clearly linked to a particular difference in cancer incidence. Epidemiological studies have not resolved this issue and, undoubtedly, ideas based on our fundamental understanding of the cancer process are needed.
Viruses are increasingly recognized as significant causes of human cancer, especially the papilloma virus as a critical cause of cervical cancer, and the hepatitis B virus in the case of liver cancer. In the case of the papilloma viruses, this recognition has come entirely from modem techniques of molecular biology, which have also been the entire basis for the discovery of the HIV viruses as the cause of AIDS. New techniques also hold great potential for effective vaccination against these and other cancer causing viruses, as well as for the successful differential treatment and eventual cure of AIDS, and other diseases caused by related viruses.
Fundamental research, whether or not it makes use of animals, is as relevant to the problem of cancer prevention as it is for cure. In fact, the first priority in dealing with cancer is prevention, and then what cannot be prevented must be treated and cured. The key is prevention and cure, not prevention versus cure.
Basic theories on the mechanisms of cancer causation have been known for some time: carcinogens, de-differentiation (the loss of key specific characteristics in a cell), the immune system, viruses, changes in the chromosomes that carry the genes and (the bottom line) gene changes in somatic (non-germ) cells. These ideas can all be combined into the one main notion that cancer is a series of changes in gene expression, usually genetic mutations, that lead progressively from the normal to the malignant cancer cell.
The development of most of these ideas depended on work with animals. Peyton Rous showed that a fowl sarcoma could be propagated by means of a cell free preparation; a classic demonstration of a virus. The idea was slowly accepted, and it was work with the cancer-causing oncogenic viruses in animals that provided some of the most striking evidence - through the discovery of the oncogenes - for the genetic basis of cancer. This finding also advanced the discovery of HIV.
The discovery of antibodies by Behring, showing that acquired resistance to bacterial infection resided in blood serum, was the first of many animal experiments which formed the basis of our understanding of the immune system. This is the system which normally protects us from infections and is the basis for vaccination. Its role in transplantation and tumour response will be discussed later.
Clonality of tumours (all the daughter cells being derived by cell division directly from the first cell) has been demonstarated most elegantly by D. Williams' experiments on induction of mouse liver tumours in a particular strain of mice which is heterogenous for an x-linked marker. The tumours from this strain of mice contain a marker for a gene from the X-chromosome derived from the mother, which can be used to establish that they are derived from one, rather than several cells. This is fundamental evidence that tumours are due to genetic changes.
The development of techniques for growing mammalian (including human) normal and cancerous cells in the laboratory - tissue culture - has had a great impact on cancer research. It has enabled experiments aimed at understanding single cells or defined mixed cell populations to be done in the laboratory rather than in animals. Tissue culture is the ultimate pioneer of so-called 'alternatives' to animal experiments and is now a much more fundamental part of cancer research than work with animals. But the development of tissue culture techniques depended on the use of animal tissues; animal-derived tumours, on sources of animal serum, on the use of antibiotics and on a knowledge of vitamins and other growth constituents. All of these derive to a greater or lesser extent from work with animals.
Genetic changes in cancer cells and the control of growth
The discovery of specific genetic changes in cancers, especially human cancers, and their functional significance, is one of the most dramatic advances in cancer research in this century and has great promise for new approaches to prevention and treatment.
The very origins of molecular biology and molecular genetics lie in Griffiths' experiments with mice in the late 1920s, showing that heat-killed pneumococci could transfer virulence to living non-virulent bacteria. This led to the discovery of transformation and so, through the work of Avery and his colleagues, to the demonstration that DNA was the chemical substance of the gene. The rest is history: Watson and Crick's discovery of the structure of DNA, the elucidation of the genetic code and the basic mechanisms of protein synthesis and gene expression, and the development of recombinant DNA techniques allowing any stretch of human DNA to be isolated, multiplied in the test-tube and have its sequence read.
The direct molecular evidence for the genetic basis for cancer comes from the transformation of cells in culture to a cancerous-like state by Rous’ sarcoma and other oncogenic viruses. The transformed state, however, must first be established by the ability of the cells to cause tumours in suitable animal hosts.
Application of recombinant DNA techniques to the analysis of the genes and gene sequences of the oncogenic viruses showed which of their genes are responsible for their ability to cause a cancer. These are the oncogenes, which are altered or mutated forms of genes that occur in normal cells. Molecular genetics makes it easy to jump from one species to another, and so it could be shown that the viral oncogenes that cause cancers in mice and chickens were also oncogenes in human tumours. So now we have identified a class of genes whose particular role seems to be in the control of cell growth.
There are three ways in which this can happen. The genes may trigger the production of growth factors or their receptors (molecules on the cell surface where growth factors attach) or they may act indirectly, through the processes by which a growth signal is transmitted from the surface of a cell to the nucleus. Genes in this class are often found to be the genetic culprits in different human tumours. The first growth factor for cancer cells, epidermal growth factor (EGF), was found as a substance that led to the early opening of eye lids of new born mice. Oncogenes have been shown to be critical for the development of chronic myelogenous leukaemia (CML); raised epidermal growth factor receptors, or related receptors, are found in more aggressive breast and bladder tumours; and mutated ras oncogenes, involved somewhere in the signalling process, are found in many tumours including, for example, up to 70% of colorectal tumours: the commonest cancer in the Western World not due to cigarette smoking.
When a growth factor produced by a cancer cell stimulates it to divide itself, blocking the function of that growth factor which has been switched on inappropriately by the cancer cell, is one approach to specific therapy. Small cell lung carcinomas, which constitute 25% of all lung cancers, produce a growth factor called gastrin releasing protein (GRP), which is critical for its growth in up to 70% of cases. GRP was discovered through the action of a related protein (bom¬besin) in frogs, and assays using mouse cell lines established that it was a po¬tent growth factor, leading to the development of drugs to block its action. Now it is possible to grow human small cell lung carcinomas in mice to show that analogue peptides can block the action of GRP and inhibit tumour growth. This experi¬ment is a critical step, together with toxicity tests on the peptide, before it can be tested in clinical trials on human patients, which we hope will start soon. This is the work of Henry Rozengurt at the ICRF.
The sequence of testing first on animal and human cell cultures, and then on human tumours grown in mice before a phase I clinical trial, is com¬mon to nearly all new drug developments, whether conventional chemotherapy or using the newer approaches, such as the GRP blocking peptide.
Sex hormones, particularly oestrogens, are important for the growth of breast tumours in humans and in mice. One of the classic oncogenic viruses that causes breast cancers in mice was discovered by its transmission through the milk. The virus inserts DNA next to a gene which it switches on as a key step in the production of a cancer. Studying this mechanism led to the discovery of a class of growth factor like substances that play an important role in the control of differentiation at early stages in embryogenesis (development of the embryo). Already it seems clear that changes in the level of production of these factors are important in some human tumours.
The oncogenes described so far are 'dominant'. This means that cancer is caused by a change in just one of our two copies of a gene, even if a normal version of the gene remains. There is, however, another major class of genetic changes in cancer, which involve knocking out the function of a gene.
Two approaches led to their discovery. In one, pioneered by Henry Harris and George Klein, cancer cells are crossed with normal cells in the laboratory and the hybrid cells are shown not to cause tumours in animals. In other words, when the normal cell corrects the deficiency defect in the cancer cell, the resulting hybrid is no longer fully cancerous. These genes were identified by studying rare families who have a susceptibility to cancer. Knudson (1971) pointed out that some genes which were inherited through the family could also be genes that were altered in the somatic cells from which a cancer is derived step by step, without there being an underlying inherited susceptibility. The idea is that the inherited susceptibility is the first step in knocking out a gene’s function. Only one further step - knocking out the remaining normal gene's function - means there is no gene function in the cell, which can then lead to cancer. This will occur with much higher frequency than the simultaneous loss of function from both versions of a normal gene in an individual who has not inherited susceptibility. The conclusion is that, if we find the genes for inherited cancer risks, we may also find key genes for the development of a non-inherited form of a cancer.
Adenomatous Polyposis Coli (APC) is an inherited cancer susceptibility in which individuals have up to 1,000 small, pre-cancerous growths in their bowels - adenomatous polyps. These first develop in the early to mid teens, and each has a small chance of giving rise to a cancer, so that the overall probability that such an individual will at least one cancer of the bowel is very high if they are not treated. So far, the only effective treatment which greatly reduces the risk is surgically to remove most of the large bowel. The inheritance pattern is simple Mendelian dominance: i.e. on average 50% of the offspring of an affected individual will be affected. APC constitutes only about 0.5% of all colorectal cancer.
Using new molecular techniques which allow the study of DNA differences in families, together with the observation of a small deletion mutation in an isolated affected individual with many other abnormalities, the APC gene has been localized to a small region of chromosome 5. This knowledge has enabled researchers to show that the gene is involved in more than 40% of all colorectal cancers. The pattern of inheritance shows that when the gene’s normal function is completely knocked out, a cancer may be encouraged to develop. Thus, the normal function appears to hold back undisciplined growth of cells in the large bowel. Now we can pursue more or less standard, though very complex, techniques for trying to isolate the gene itself and determine its function.
The function of the gene may provide a variety of clues to prevention. For example, it may reveal substances produced at an early stage in human development that can be detected and used for screening. For treatment, it may reveal new pathways that are switched on to encourage the growth of the cells that could be blocked.
But the discovery of the gene will also open up another new and exciting possibility – creating mouse models of human cancer. This can now be done by growing cells derived from early mouse embryo cultures in the laboratory, and inserting the relevant mutation in the right position into these cells. These cells can then be incorporated into another mouse embryo, giving rise to animals which carry the relevant mutation in all their cells, just as if it had arisen spontaneously in a germ cell. Now we have a true model of the human disease that is entirely relevant, and using it we can examine conditions, for example, diet or other treatments, which may prevent the development of the polyps or which will inhibit their growth. Thus, in this way, we have the possibility of finding new approaches both to the prevention and the treatment of a significant fraction of bowel cancers, of which there are 25,000 new cases each year in this country and some 19,000 deaths. So far, only surgery is really effective in treatment and only then if the cancers are caught early enough.
The manipulation of mouse embryos, on which the production of transgenic mice depends is another major advance in biological and medical research which opens up totally new possibilities for the study of differentiation and development. These contribute not only to cancer research, but to research on many other diseases, as well as to novel approaches for the breeding and trans¬port of farm animals as frozen embryos.
There are a number of other examples of so-called 'recessive' genetic changes in cancer cells, each of which can now be identified and its functional basis studied and exploited using transgenic mouse models. One of these changes, on the short arm of chromosome 17, is intriguing since it was first found that there was often an apparent deletion of function in this part of the chromosome, for example in 50% or more of bowel cancers. Then it was realised by Vogelstein that the p53 oncogene defined by Lane and Crawford, and Levine and others had been shown to be in this part of the chromosome, and so they thought: 'Why not see whether that is responsible for chromosome 17's involvement in bowel cancer?' And that is exactly what was found, namely that in those cases where there is an apparent loss of function, the p53 gene is mutated. The explanation seems to be that the p53 mutation gives the tumour a selective advantage, although this is to some extent counteracted by the remaining normal version of the p53 gene. In this case, therefore, the apparent loss of function on chromosome 17 is actually due to another sort of positive change like those for the dominant oncogenes including ras.
Knowledge of where a gene is, is an essential step to finding a gene, and so its function, and this is the aim of studies on APC. It is also the basis on which the defects underlying Duchenne muscular dystrophy and, most recently, cystic fibrosis have been identified. Finding out where all the human genes are is the Human Genome Project, the results of which promise to provide the basis for major new approaches to the medical and health care of the next century. An important part of this project is the comparative analysis of the mouse and human genomes, since knowledge of one can greatly help knowledge of the other. Once we know where a gene is on the human genome, one can find the corresponding place on the mouse genome and create at will the relevant mouse mutation that is the true model for the human disease. There are already examples where knowing where the human gene is and looking at the corresponding mouse region and finding the gene of known function there, led to the identification of the human function, and vice versa.
The immune system and cancer
Another change commonly seen in colorectal carcinomas, and in certain other cancers, seems to be the loss of expression of an HLA tissue type; this is interesting because it signals that the cancer has had to escape an immune response directed at it. This is because the HLA class I determinants are an essential requirement for the function of immune T cells when they destroy a target cell. Thus if a mutation that does not express the HLA class I molecules arises in a cancer which is being attacked by the immune system, this cell will escape attack and so take over the cancer, making it resistant to the body's immune system. The relatively common observation of such changes suggests that immune attack on a cancer may be more common than was realised until recently.
Paul Ehrlich suggested that one role for the immune system was to recognize changes in tumour cells and so remove them, rather like it removes an invading infectious organism or a foreign tissue graft. For many years these ideas seemed attractive but then no real evidence for them could be found and they fell out of favour. Nevertheless people still tried to manipulate the immune system in order to treat cancers. Now, however, new understanding of the nature of the cellular immune response, together with observations of genetic changes in tumours that could be targets for immune response, and the changes in HLA which indicate that immune response has happened, have totally changed the picture.
As already mentioned, transplantation of tumours in animals paved the way for an understanding of transplantation immunity. When Hunter successfully transplanted a cock's spur onto its head, he did not realise that it was successful, not because of the better blood flow, but because there was no genetic difference between the transplant and its recipient which could lead to its rejection by the immune system. The normal immunological basis for transplantation rejection was worked out by Medawar in his classic experiments using skin grafting in mice, followed by the experiments of Billingham, Brent and Medawar which demonstrated the induction of tolerance in mice by injection of cells into neonates. The fundamental division of the immune cells, or lymphocytes, into B and T cells came from the observation that removal of a special organ, the bursa of fabricius, from chickens within two weeks of hatching makes them unable to produce antibodies later in life.
Many subsequent experiments with animals, involving transfers of immune cells from one animal to another and studying the immune response of animals to viruses and other substances, have led to our present level of understanding of this extraordinarily complex and important body system. This understanding, and its application using animal experiments, provides the basis for new approaches in exploiting the immune system for the treatment and prevention of cancers.
A particularly important recent discovery by Alain Townsend and his colleagues is that the immune T cells can recognize intracellular products, and not just molecules on the surface of a cell. This is because the intracellular proteins are broken down and the breakdown products themselves are expressed on the surface of a cell in a form that the immune T cell can recognise. This totally changes the picture of potential targets for the immune system to recognize in a cancer cell, and means that any genetic change leading to an altered protein is a potential target for immune attack. Thus, one can have a new view of the role of the immune system in cancer and the way it can be manipulated, even holding out the prospect in the longer term of some forms of vaccination. These developments will depend critically on the ability to create transgenic mice which carry some of the human genes that control human immune response. This will enable reagents to be made in the mouse that recognize relevant human differences.
Novel approaches to vaccination, such as are being developed for HIV and other viruses by enhancing the response of the lymphocytes which specifically recognize cellular changes, are also needed and will - depend entirely on suitable animal experiments, perhaps also including use of primates.
Bone marrow transplantation, which depends on the development of our knowledge of the immune system and on tissue typing, is now an effective form of therapy for some leukaemias.
Monoclonal antibodies (derived from a single antibody-forming cell), as first made by Kohler and Milstein, are remarkable specific reagents and their production depended on a knowledge of the immune system and its function, together with the development of tissue culture techniques and ways of obtaining suitable cell lines from tumours and crossing them in the laboratory. Monoclonal antibodies, mostly made in mice or rats, are used in almost all aspects of work in biomedical research, and their production itself now counts as an animal experiment. Once a hybrid cell making a particular monoclonal antibody has been isolated, the antibody itself can either be produced in tissue culture or by propagation in a mouse.
Monoclonal antibodies now form the basis for an extraordinary variety of investigations, from basic studies on the characterization of new cellular products to their use in diagnosis in the histopathology laboratory and their application in novel approaches to imaging tumours in people and to targeted therapy. Monoclonal antibody reaction to newly identified products may, for example, form the basis for more specific and sensitive screening tests, using stool samples, for the presence of early bowel cancer. This should pick up mostly early tumours that can be effectively cured by surgery. Failing primary prevention, this secondary approach is the next best step.
The use of monoclonal antibodies for in vivo diagnosis has depended on the development of suitable animal models. Human tumours can be grown in the nude strain of mice, which is a mutant strain whose immune system is deficient to such an extent that it will not recognize the foreignness of human tumours. Assessing the effectiveness of different forms of therapy against human tumours grown in nude mice is a major step forward in providing a more effective animal model for human tumours. In particular, antibodies that react with human tumours will react with them when they are grown in a mouse and it was in this way that the techniques for imaging tumours in people using antibodies tagged with radioactivity were developed. More work still needs to be done, but already these approaches are providing some effective additions and alternatives to other forms of imaging of human tumours.
The radioactivity attached to an antibody and delivered by it to a tumour is known as targeted radiotherapy. If enough of the antibody attaches to the tumour, and not too much to other critical normal tissues, then the tumour may be killed by the attached radiotherapy delivered by the antibody, with relatively minimal side effects. In exactly the same way, an antibody can be used for delivering a poisonous toxin, such as ricin, or indeed any other chemotherapeutic agent.
Although there are many possibilities for engineering the antibody molecule and devising new forms of attachment of radioactivity, toxins and other chemicals to improve stability and delivery, and while much work may still be needed to find antibodies to determinants that are sufficiently tumour - associated relative to normal tissue, to minimize side effects, these are technical problems that will undoubtedly be overcome.
The main limitation of the very promising monoclonal antibody approach to therapy is the difficulty of access to the tumour. The antibody is a big molecule and cannot always penetrate through the vascular system in sufficient amounts in the time available to deliver a sufficient dose of radioactivity or other drug to the tumour. Nevertheless, in several situations where antibodies can be inoculated into body cavities, such as the peritoneum or the cerebrospinal fluid surrounding brain and nervous system tissues, where it has a more direct access to the tumour than via the blood stream, promising results have been obtained. Early treatment when the tumour mass is small and access is much easier - namely adjuvant, monoclonal, antibody-based, targeted therapy - may also prove effective.
Other advances in cancer research
There are many other examples of advances in cancer research which depend on work with animals. Metastasis, the spread of a tumour, is a complex process involving many different tissues. This cannot easily, if at all, be recreated in the laboratory outside the whole animal. Another important feature of tumour progression is vascularization (development of new blood vessels) of the tumour and the influence that this has on its development. The ability of a tumour to differentiate seems to depend in many cases on its interactions with neighbouring tissues; some of this can be studied in tissue culture systems but eventually, especially if the observations can be exploited by the induction or blocking of cellular functions as a part of tumour treatment, then animal experiments become essential.
Many novel substances have been found through the application of recombinant DNA techniques. In many cases, these techniques enabled studies which could not otherwise have been done, such as with the interferons. Such substances, including those which stimulate the growth and development of cells of the bone marrow -collectively called cytokines - are beginning to play an important role in cancer therapy. A wide variety of different cytokines interact with each other and work in different ways on different cell types so that only studies on the whole animal can give a true picture of their effect.
Another new approach to treatment is to use lasers to activate drugs which are preferentially taken up by cancers and which then release 'singlet oxygen', which causes localised damage. In other words, only those tissues which have taken up the drugs, which include particular forms of phthalo-cyanines, will be damaged. Initial studies with animals have helped to define the basis for the preferential uptake of the drug into the tumours, and to provide the basis for the initiation of clinical trials in appropriate human situations.
Fundamental cellular and molecular studies, such as I have been describing, but including many other areas I could not cover, are defining new functions, new proteins and new enzymes which may play special roles in various stages of the cancer process. With every such defined function there is a potential for pursuing the classical pharmacological approach so successful in the development of cimetidine and the beta blockers, namely to identify low molecular mass compounds that specifically antagonize or block a particular function. As we learn about the transport processes across membranes of cells, it will become easier to block specific intracellular functions including, for example, those mediated by the ras and other oncogenes.
In each case the potential for such a blocked function to be the basis for a novel approach to therapy must be tried out in a suitable animal model, before patient studies and after tissue culture experiments. Only in this way is it possible to assess, at least to some extent, the overall effects of a treatment on body functions in the whole organism. The blocking of GRP (see above) function as a basis for treating small cell lung carcinoma is a typical example. This approach should not be limited by access of the blocking agent to the tumour and it should be possible to establish considerable specificity with respect to tumour versus normal tissue, and so minimise side effects. I believe this to be the cancer pharmacology of the future: an integration of classical approaches with the revolutionary new insights being provided by modern techniques of cellular and molecular biology.
Cancer research, I believe, is in a more exciting period than ever before. The advances I have described, and their potential, arise from the combination of modern techniques of cellular and molecular biology with experiments in animals. They hold out great promise that most cancers may be prevented or satisfactorily treated within the foreseeable future following, by now, reasonably defined overall lines of attack.
Animal experiments are an important part of the work of nearly every one of our laboratories at the ICRF (as they would be in most biomedical research laboratories), especially, of course, in the production of monoclonal antibodies. Nevertheless, the relative importance of animals numerically is gradually declining. The overall use of animals at the ICRF (95% rats and mice), which are all specially bred stocks, and often our own produce under specific pathogen-free conditions, has declined by a factor of more than two in the past ten years, whereas our scientific staff has increased nearly three fold. Directly attributable expenditure to this work is now only about 3-4% of our total expenditure on research. The trend is probably towards a continuing decline but an increasing complexity of the experiment, as in the production of transgenic animals. Much of our work now involves the use of the immuno-deficient nude strain of mice, particularly for growing human tumours, and will increasingly involve the very powerful technique of producing transgenic animals expressing various combinations of human genes.
At the same time, undoubtedly, there will be further developments of 'alternative' approaches that do not depend on the use of mammals. The more we learn about basic cellular functions and their evolution, the more often it becomes possible to study key human functions in another organism, such as yeast, or the geneticists' fruit fly Drosophila, and, of course, the famous E. coli bacteria which are the basis for most genetic engineering techniques. Thus, for example, the function of a protein which is often responsible for the development of resistance to drug therapy in cancers has been elucidated by studying a protein with a related function, concerned with the transport of molecules between the inside and the outside of the cell, in bacteria. Similar molecules in different guises may be responsible for different human diseases.
One of the best-known examples of bacterial tests in the cancer field is that for carcinogens based on the extent to which they cause genetic mutations in bacteria: the 'AMES test'. Extensions of this use include using tissue culture cells to look for the 'genotoxic' (gene-damaging) effects of different chemicals. This is an important aspect of assessing environmental factors in cancer but not, as I have already stated, perhaps the most critical one. For the effects of diet we must look to mechanisms by which components in the diet promote the growth of cancer cells rather than initiate them. And for this, basic functional studies, such as have been described above, are essential.
It is clearly impossible to be at the forefront of cancer research without doing some work with animals. The moral issue that must be faced and which may be answered differently by different people, is whether the achievements of medical research using animals, which are aimed at saving human lives and alleviating human suffering, justify the use of animals. For those who believe that the lives of people take precedence over the lives of animals, the answer is clear. For those who do not, they must tell us how many mice experimented on and killed, are worth a human life.
It is essential for the scientists involved to explain and justify what they are doing and to be prepared to discuss the issues openly with those who have a genuine concern as to what is done with animal experimentation. Those experts in support of the anti-vivisection campaign who say: 'Animal experiments have not and never can make any significant contribution in the fight against cancer', or; 'Most research funding continues to be misdirected into cruel and useless animal research', or; 'We now know how to prevent 80-90% of cancers', must be answered. These statements are patently not true. If those who make such statements know that what they are saying is untrue, then they are deliberately spreading misinformation. If they do not, then they are peddling their ignorance to the public, often behind a front of apparent scientific respectability.
Whatever one's beliefs, the scientific facts must speak for themselves and not be distorted. Our concern as scientists must be that the facts are presented in as dispassionate and objective a way as possible, while nevertheless accepting that there may be differing views as to whether the benefits that come from them justify the use of animals in research.
The case, to my mind and to most of those involved in medical research, is surely overwhelming. Now we can look forward to a period of time when there will be extraordinary advances in our ability to prevent or cure cancer.