Drug safety testing and public health
Despite the fact that regulatory requirements relating to testing in a rodent and a non-rodent species are being met, there is nevertheless a high incidence of adverse drug reactions (ADRs) in the human population. According to the British Medical Journal, five per cent of all hospital admissions are the result of ADRs (10) - and the annual number of people ultimately dying from ADRs translates into a total figure of 18, 000 deaths - more than five times the number of people killed every year in road traffic accidents.
Significantly, human liver damage is the most frequent reason cited for withdrawal of an approved drug and accounts for more than 50% of cases of acute liver failure (11). This important phenomenon is, in no small measure, explained by scientific evidence strongly suggesting that human liver damage has the poorest correlation with regulatory animal toxicity tests (12).
Animals frequently receive very high doses of a test compound (often thousands of times more than any conceivable human exposure) in an attempt to determine whether the substance poses a health risk (to humans). A sad example is the lifetime rodent bioassay (LRB) for predicting human carcinogenicity. In this experiment, rats are exposed to potentially cancer-causing chemicals for as long as two years.
Defenders of the rodent bioassay claim that all known human carcinogens cause cancer in rodents. However, little attempt has been made to validate this assertion against human cancer data (13). Many toxicologists have expressed serious doubts as to the statistical relevance of the LRB (14). It is now acknowledged that this test yields many false positives(chemicals that cause cancer in rodents but not in humans). More significantly, it also yields false negatives. These are chemicals that, while harmless in rodents, are dangerous to people. To add to the confusion, using different strainsof rats produces conflicting results (15).
Disturbingly, the LRB is still considered by regulators to be 'the standard' for predicting human carcinogenicity.
The food industry has not been slow to exploit the muddled results produced by the LRB. For example, the widely consumed artificial sweeteners saccharineand aspartamecause cancer in rodents (saccharineleads to bladder cancer, and aspartameto leukaemias). The government, nevertheless, permits these products to be mass marketed because it has allowed itself to be convinced by the food industry that the animal tests are irrelevant.
We are repeatedly told that such tests are 'necessary' - only for them to be discounted when their results are commercially inconvenient. What is the general public to make of this? Should animal tests be trusted? Of equal significance is whether the health authorities can be trusted.
The potentially negative implications of such data for public health do not stop there. Regulators not only allow industry to back their health claims with questionable animal data, but they also allow industry's 'experts' to determine how much of a chemical can be 'safely' consumed. This accepted volume is referred to as the ADI - the acceptable daily intake. It is typically expressed in mg/ kg/day. The ADI does not take into account differences between individuals with respect to age, gender, weight, allergies, etc.
The ADI dosage is derived from data emerging from animal tests, to which adjustments are made in order to incorporate a safety factor. Usually, this means the original 'safe dose' being stepped up by a multiple of ten to account for intraspecies sensitivity (individual variation within the same species), interspecies difference (between species), and other factors of concern (completeness of data). The appropriateness of this arbitrary 'multiple of 10' has, understandably, been seriously questioned. Professor Frederick vom Saal at the University of Missouri-Columbia was quoted by Natureas saying:'The evidence is that there can be as much as a 1, 000- fold, or greater, range of responses to these chemicals in different strains of mice. The regulatory default assumption of a ten-fold correction or safety factor for genetic variability is completely out of touch with the data' (16).
Not surprisingly,when pressed to make a scientific case for animal tests, not even the Department of Health (DoH) can muster a plausible body of evidence. As part of its 2002 inquiry into Animals in Scientific Procedures, the House of Lords Select Committee asked the DoH to provide published scientific papers that support the validity of toxicity tests on animals. One of the papers (17) submitted by the DoH actually lists several reasons why data obtained from animals are usually not predictive of human effects. One difficulty is that 'the life span of humans is from 4.4 to 66 times that of common test species. Thus, there is generally a much longer time available for many toxicities to be expressed or developed in people than in test animals'.
Another (18) contains this damaging admission: 'Two reviews addressed those drug cases where the clinical (human) toxicity was so severe as to lead to withdrawal from marketing in the approximate period 1960 - 1990... In one report,only 4 of 24 cases werepredictable from animal data; in the other report, only 6 of 114 clinical toxicities had animal correlates.' Given the lack of predictive value of animal tests, it is no wonder that adverse drug reactions are now the fourth leading cause of death in theUK, after heart disease,cancer and stroke! (19)
Equally,it should come as no surprise that, in 2004, Home Office Minister, Caroline Flint MP, during an answer to a parliamentary question,stated:'The Home Office has not commissioned or evaluated any formal research on the efficacy of animal experiments.' When pressed by Michael Hancock MP about the future,instead of trying to undo the damage, the government dug itself even deeper into its hole by admitting that it had no plans to commission such a study (20).
REACH (Registration,Evaluation and Authorisation of Chemicals)
Media reports throughout 2004 and 2005 have suggested that the chemical industry is being brought to heel by a timely proposal put forward by the EU,entitled REACH. In principle,EU authorities are right in wanting to assess 30,000 (tailored down from an original figure of 100,000) man-made chemicals for their adverse effects on human health and the environment. However,the prospect that this assessment will not be completed for several decades at least,and will involve a very large number (many millions) of animals, is cause for serious concern.
In January 2005,the European Parliament (EP) began its formal consideration of REACH at a public hearing. Three main committees of the EP (environment, industry and internal markets) reviewed REACH in detail, including 1000 tabled amendments. The EP first reading was expected to be completed with the Plenary (whole Parliament) vote during the week 14 - 17 November 2005, to be followed by a second reading by as early as April/May 2006. Barring any major delays, these EU regulations could become law in late 2006. However, there is likely to be a time lag between entry into force of these regulations and their implementation, ranging from three to 11 years, based on considerations such as production volume and public health risk.
Caroline Lucas MEP has stated:
"As a Member of the European Parliament, I have become increasingly aware of the threat posed by chemicals in our food, our homes and the environment. Throughout discussions of the European Commission's proposal for a new regulatory regime for chemicals (REACH), I have argued that the protection of human health and the environment is of paramount importance. At the same time, however, I have made the case that this does not have to lead to yet more cruel, outdated and inefficient animal tests. To the contrary, the new drive for safer chemicals can be used to generate new momentum for the use and development of non-animal testing strategies" (21).