Skip to main content

Ld50 Toxicity Test

Volume 13: debated on Tuesday 24 November 1981

The text on this page has been created from Hansard archive content, it may contain typographical errors.

Motion made, and Question proposed, That this House do now adjourn.— [Mr. Brooke.]

11.45 pm

The LD50 toxicity test is a method of trying to determine how poisonous a chemical or pharmaceutical substance is by forcibly feeding it to live animals and then determining what single dose is required to kill 50 per cent. of them within a maximum 14-day period, hence the expression "lethal dose 50", or LD50. The technique generally used is to push a tube down the throat of the animal and then feed it with the substance, or the chemical may be applied to its skin or, in some cases, a vapour may be used which the animal is forced to breathe.

An official report to the Home Office in 1979 by the advisory committee on the administration of the Cruelty to Animals Act 1876 said rather baldly in paragraph 13:
"LD50s must cause appreciable pain to the animals subjected to them."
It would be a good deal more accurate to say that they cause agonising pain, convulsions, bleeding, diarrhoea, and eventually lingering death over a number of days.

The number of animals used is not insignificant. In 1980, it was 484,849, according to official statistics. The test was devised by a Dr. J. W. Trevan in 1927, more than 50 years ago, to measure the strength of highly poisonous but potentially beneficial drugs such as diphtheria antitoxin, digitalis and insulin. Unfortunately, its use has been extended to determine the acute toxicity—in normal speech, poisonous potential—of industrial chemicals, natural compounds, food additives, pesticides, detergents and even substances used in cosmetics. The LD50 test was developed originally for drugs. It has now been taken over for the toxicological evaluation of all other classes of substances, hence the use of nearly 500,000 animals for this purpose in 1980. Mice, rats, birds, fish, rabbits and guinea pigs are used mostly, but sometimes dogs and, very rarely, monkeys.

Unfortunately, its use has been built into the law. In the United Kingdom regulations under the Medicines Act 1968 and the Health and Safety at Work etc. Act 1974 refer to it specifically. An EEC directive is being considered to harmonise practice in the Common Market, and many other countries have similar legislation. The consequence is that the test is widely employed not because it gives useful results scientifically but to satisfy legal requirements in marketing products within the United Kingdom and in export markets.

If a manufacturer produces a substance which subsequently causes damage or harm to a person and that person sues the manufacturer, he can argue in his defence that the LD50 test was carried out according to the rules and thereby seek to satisfy the court that he did nothing wrong in marketing the product.

However, the test has serious defects. The results can vary drastically according to the species of animal used. One compound administered according to the LD50 procedure to male mice produced a slightly toxic effect. Given to male rats, for the same purpose, it qualified as highly poisonous and therefore subject to strict controls. This example is quoted in an article published this year—I stress this year, 1981—by G. Zbinden of the Institute of Toxicology in the Swiss Federal Institute of Technology and University of Zurich, an institution of international reputation in scientific matters.

The same article draws attention to variations in results from the tests according to the sex, age, diet, genetic strain, health, degree of starvation, and method of dosing of the animals used. The temperature and humidity of the laboratory, the type of cage, and even the bedding material of the animal could affect the results. There are also seasonal variations.

A special international study was undertaken in the late 1970s of the variability of the LD50 test, covering 100 laboratories in 13 countries. By the end of 1979 information from 80 laboratories became available. Despite careful control of experimental conditions—caging, feeding, weight and age of the animals used—the results for the five substances in this control test showed marked discrepancies as between different laboratories.

More fundamental is the objection that species difference—different types of animals reacting differently to the same chemical—is a major obstacle to the extrapolation of results from animals to man. I shall quote what Professor Zbinden said in his article:
"Unfortunately, even with carefully designed animal experiments it is sometimes not possible to predict all relevant symptoms of an acute intoxication in man. Human response to chemical substances is often quite peculiar, and valid experimental model systems to assess these reactions are lacking. For this reason, it is very important to intensify research which will permit us to expand the range of toxicological knowledge and to improve the predictability of the test procedures. Governments and industry would be well advised if part of the funds now spent for routine investigations would be allocated to the study of better testing methods."
We should recognise that in the half-century which has elapsed since this test was devised, a revolution has occurred in the biological sciences, and alternative techniques could and should be developed. A single-cell culture system can give a rough guide to the human lethal dose of a substance, if that substance acts by damaging the cell function. This would filter out the more poisonous chemicals and drugs, which could then be discarded without further testing and without poisoning unnecessarily thousands of animals.

Some work carried out in Sweden has shown that a simple cellular system derived from human cancer of the cervix can be very useful. The system is based on a metabolic inhibition test supplemented by microscopy. For 39 out of 52 drugs, dosages toxic to the cell were grossly similar to the estimated human lethal doses. The advantage of cellular systems such as those used in Sweden is that they are controlled and, of course, are much cheaper. That is one reason why the Swedish work was initiated.

The greater complexity of living animals—there is a difference between testing on cells and testing on live animals—is of no value when based on fundamental and unalterable differences between the species. In other words, one cannot simply extrapolate from animal to man. Increased demand for testing by regulatory bodies has spurred the investigation of these cheaper non-animal tests, but continued reliance on the animal method has delayed the development of new and more effective procedures.

It is possible to get a more sophisticated assessment by using tissues from various organs such as the liver, nervous tissue, muscle cells or the thyroid gland. One of the advantages of tissue culture is that human cells can be employed to reduce the problem of species difference between animals and humans. I shall quote briefly from the article by Professor Zbinden:
"For the recognition of the symptomatology of acute poisoning in man and for the determination of the human lethal dose the LD50 test in animals is of very little value. Much more relevant information could be gained from a well conducted comprehensive short term test which includes clinical and biochemical monitoring and gross and microscopic pathology. Such a test could be done with small numbers of animals … The use of the LD50 test as basis for selection of doses for sub-acute and chronic toxicity tests and other procedures … is obsolete".
These alternative techniques can be supplemented by theoretical studies of new chemicals which belong to classes of substances whose mechanism of poisoning is already well known. However, if more humane and more satisfactory techniques are to be developed, both Governments and industry must allocate more funds to investigate better testing methods. Of course, there would have to be some change in the regulations under the Medicines Act 1968 and the Health and Safety at Work etc. Act 1974. Special responsibility must rest on the Government, and particularly on the Home Office, to change that legislation.

First, the LD50 test is obsolete. I have already cited the evidence from the special study made in Zurich, which categorically reaches that conclusion. Secondly, it is now widely used for purposes for which it was not originally designed. It was developed in 1927—more than 50 years ago—for testing the poisonous nature of special substances that were, and no doubt still are, of enormous importance as potential medicines. Substances such as digitalis and insulin are extremely important. It was not designed as an all-purpose test for industrial chemicals or for the compounds that are found in detergents, cosmetics, and pesticides, although over the years it has been steadily extended to them. Thirdly, it produces results that are scientifically unreliable and cannot be extrapolated from animals to humans. Again, I cite the findings of the Zurich paper, which are specific and important. Fourthly, it causes agonising pain and death to nearly ½ million experimental animals each year in this country alone. All hon. Members will agree that we should avoid that if possible and if there are other options.

Fifthly, the main use of the test now is to satisfy legal—not scientific—requirements. It does not have a sound scientific base. The regulations under the various Acts have incorporated the test as a requirement and, as a result, it is being widely used for legalistic rather than truly scientific purposes. Sixthly, there are other, supplementary techniques that could drastically reduce both the number and type of animals used and might ultimately eliminate their use altogether.

I am aware that cruelty to animals is an emotive subject. All civilised people are anxious to make arrangements that prevent or limit it. I am equally aware that there are powerful schools of thought that claim that the experiments made on animals are necessary either for the advancement of science or to ensure the safety of substances that are widely used both in medicine and for various industrial purposes.

I emphasise—I am sure that no one can dispute it—that in the past 10 or 15 years we have had a biological revolution. It is now possible for scientists and biologists to manipulate the fundamental structure of life. We can disentangle the genetic code that determines the nature of living things. The advances in that direction have been enormous. I do not believe that it would be impossible to develop and use similar techniques so as to avoid inflicting violent pain on hundreds of thousands of animals every year. Perhaps we cannot eliminate the LD50 toxicity test altogether, but I am convinced that we could go a long way towards it, and that Government action is now required to set out on that road.

12.1 am

The hon. Member for Sheffield, Heeley (Mr. Hooley) has selected for debate tonight a subject about which there is a good deal of continuing discussion among those concerned with the use of live animals in experiments. It is right that the House should join in the discussion.

Let me make it clear that the Government fully understand the strength of feeling on the matter. I want to leave no doubt in anyone's mind that we welcome the constructive discussion on the way forward in what is inevitably an emotive area.

I am fully aware that there are those who take the view that no animal experiments are justifiable and that it is morally wrong to use animals in that way. I do not think that the hon. Gentleman would hold that view, but we respect the sincerity of those who do. However, all the recent discussions, both in Parliament and outside, have made clear the overwhelming view that experiments on animals should be permitted to continue but that they should be regulated and controlled to preserve a proper balance between the welfare of animals and the legitimate needs of scientific advance. That should be the guiding principle.

The hon. Gentleman has expressed particular concern about the LD50 test and the possibility of using other methods that do not require the use of animals. I understand his concern. As a general principle, the Government support the view that alternatives to living animals should be used wherever practicable. There is no difference between us on that score. At the same time, as I shall explain, we should be quite clear about the extent to which it is practicable to use alternatives to animals, either now or in the foreseeable future. It is regrettable, but unavoidable, that many sorts of biomedical research will require the use of animals for some time to come. It is necessary to dispel any misconception that there may be on this point.

The LD50 test has been the subject of a good deal of controversy in the past few years. That led the Home Secretary at the time to ask his advisory committee on the administration of the Cruelty to Animals Act 1876 to inquire into the test; the extent of its use in accordance with statutory requirements and otherwise; the scientific necessity and justification for the test in its various applications; and to make recommendations in the context of his powers under the Act. The advisory committee made a thorough study of those issues and its views were set down clearly in its report to my right hon. Friend the Home Secretary in 1979.

As the hon. Gentleman broadly indicated, the LD50, the lethal dose 50 per cent., is a numerical index that gives some information about the acute toxicity of a chemical substance in experimental animals. The LD50 test is only one of a number of types of acute toxicity experiments, all of which are recorded together in the Home Office statistics published annually without distinction, so a figure for it can only be roughly estimated. The estimated number of animals used for LD50 tests in 1977, for the purposes of the advisory committee's study, was 229,500 out of a total of about 560,000 acute toxicity animal experiments, which represents about 10 per cent. of the total number of experiments. Of the 229,500 some 60 per cent. were mice; some 27 per cent. rats, some 11 per cent. fish and about 2 per cent. birds. In 1980 the total of acute toxicity animal tests was about 485,000, which represents a significant fall. No estimate has been made of the number of LD50 tests, but if the same comparison is applied about 199,000 animals were used. The LD50 test was originally devised in 1927 to overcome difficulties arising in measuring the potency of highly toxic yet potentially beneficial medicines of natural origin, such as diphtheria anti-toxin, digitalis and insulin. Though there are still occasions when it may be used for its original purpose—the standardisation of therapeutic substances known to be toxic—it has in recent years come to be used increasingly in many countries for a different purpose, that is as a step in the estimation of the relative safety of new substances such as medicines, industrial chemicals, pesticides, weed-killers, detergents and food additives.

A preliminary step in the LD50 test is to determine the approximate dose of a substance which is the minimum lethal dose. This usually involves its administration in geometrically ascending doses. At this stage at least three dose levels are required with a minimum of two animals in each group. From the results an estimate is calculated from a graphical plot of the relationship between the number of deaths and the dose of the substance given. The LD50 figure is the quantity estimated to kill half the number of animals. For the LD50 test, five dose levels are commonly employed with 10 animals in each group. Some 14 days after dosing any surviving animals are painlessly killed.

In carrying out its expert review of the test, the advisory committee paid particular attention to the question of pain and alternatives to the test. It did not find it possible to make any judgment on the extent of the pain suffered which would be applicable to all animals subjected to LD50s, because the substances administered varied enormously in their toxicity and the sizes of the doses vary with different groups. In this connection, the requirements of the pain condition attached to all licences granted under the 1876 Act shorten the period over which severe pain is felt. Nevertheless, the committee was concerned that LD50s must cause appreciable pain to a proportion of the animals subjected to them.

The possibilities of determining LD50 values without experiment through the development of computer-based information retrieval systems was considered by the advisory committee. In particular, it considered a claim that the LD50 of any compound administered to a rat orally could be determined by computer from the compound's molecular formula. It reached the view that the limitations of computer-based figures in this field were such that it doubted whether the technique involved would result in an appreciable reduction in animal usage in the foreseeable future.

After spending two years receiving and examining evidence from animal welfare bodies, industrial associations, advisory and regulatory bodies and medical and veterinary experts, the advisory committee concluded that for the proper testing of new substances some acute and chronic toxicity tests must be carried out on whole live animals; and that where it is necessary to estimate the lethal properties of a substance there was no evidence that there was any better test or one more economical in the use of animals.

Accordingly, the committee recommended that LD50 tests should be allowed to continue but that those who prescribe them or carry them out should always bear in mind that for safety evaluation purposes an LD50 test giving a reasonable indication of the order of toxicity is all that is required; and that wherever practicable a "limit test" —that is, the largest dose-equivalent that a human being might take—should be used in preference to an LD50. If this does no harm to the animal, there is no need to use more extreme tests.

The committee also recommended closer liaison between the Home Office and the bodies concerned with the framing of relevant regulations, both here and in the European Community, relating to the transport storage, marketing and use of toxic substances; the possibility of a uniform code of laboratory practice; that LD50 tests should not be begun at a time which will not ensure adequate supervision of the animals during the expected period of maximum effect; and special controls over the use of primates in any kind of experiment.

My right hon. Friend accepted all these recommendations in principle, and, within the limits of the resources available, they have been carried forward. Through the authorities responsible for registered places where experiments are carried out, licensees under the 1876 Act were urged to take careful account of these conclusions and to act accordingly.

I am aware that there continues to be concern about the use and value of the test. I am equally aware of the constructive consideration which is being given to the question of alternatives, which was discussed by the hon. Gentleman. However, the present state of knowledge does not enable us to avoid the dilemma which arises from the fact that, on the one hand, many new drugs and other substances confer great benefit but carry some risk of toxicity against which the public have a right to be protected; while, on the other hand, there is concern for the welfare of animals. In this situation, we have to strike a balance.

In talking about the defects the hon. Gentleman perhaps overstated his picture. We accept that there may be defects, but the problem is that there is nothing better. It is easy to quote variability, for example, but that would lead to more extensive and larger tests. A species' difference is important, but who can use man as a test species? The variation is minimised by the use of several different species.

Also with regard to alternatives, perhaps cell cultures have a future, but the validation of such tests has far to go. We cannot call the LD50 test obsolete until it is satisfactorily replaced, which has not yet happened.

As I have already explained, we support the use of alternatives where they are practicable. It is my understanding that, where proven alternatives exist, researchers adopt them, not only on humanitarian grounds but because such methods are generally quicker, cheaper and more reliable. Indeed, the use of alternatives is believed to be one of the factors that has led to the levelling off of the number of experiments performed on animals over the last 10 years or so, and indeed to the fall in numbers over the last three years. The Government regularly bring to the attention of all those carrying out experiments under the Cruelty to Animals Act 1876 the importance of taking every reasonable step to confirm before using live animals that their investigations cannot be effectively carried out by any alternative means. Licensees are also urged to give thought to the possibilities of developing new alternatives.

However, calls for the Government to promote research into the development of alternatives tend to ignore a number of important factors. There is likely to be disagreement among scientists about new alternatives in any particular area, and about the relative costs of them in terms of manpower, materials and time.

I understand that the Medical Research Council takes the view that it is not practicable for scientists to be engaged specifically in devising alternatives as their sole or main activity separately from research experiments; and that the person conducting an experiment is in the best position to decide, from discussions with his colleagues, from his own experience and from the literature, whether techniques of his own or others will serve his purpose without using animals. It is also through knowledge of his own techniques and those of others that he is best able to judge during the course of his experiments when an alternative could usefully be developed.

Those well-established approaches to the conduct of research are far more likely, in the council's view, to be fruitful and economical in the use of animals than for scientists to be engaged in the search for alternatives to the use of animals in other scientists' experiments.

This is not a complacent attitude. It is one of realism. We understand the concern that the hon. Gentleman expressed. There is no difference between us in wishing to see a continuing reduction in the use and number of animals subjected to experiments which may involve pain; but this must be consistent with our overall obligation towards human and animal health.

The use of the LD50 test and the question of alternatives will, without doubt, continue to be the subject of debate both in Parliament and in the country. We hope that the debate will be both realistic and constructive, and that it will take proper account of the factors which must be weighed. I am grateful to the hon. Gentleman for raising this issue and hope that I have been able to show to the House that, while we appreciate and understand his concern, there are very real difficulties in advancing at the pace he advocates to the goal which both he, and I, would wish to see.

Question put and agreed to.

Adjourned accordingly at Fourteen minutes past Twelve o'clock.