Study shows thousands of animals could be saved from suffering and dying in toxicity tests

A paper published last year (2019) ^[1] examined the results of developmental toxicity studies in animals. The aim of these tests is to determine if chemicals and drugs may pose any hazard to the development of unborn children, when their pregnant mothers are exposed to them. An infamous example of a drug causing developmental toxicity is thalidomide, which caused babies to be born with malformed limbs. It estimated that up to 7% of new-borns each year (~8 million children) have birth defects, correlated with increasing use of pharmaceuticals and other environmental factors ^[2].

The study compared data from developmental toxicity studies in rats and rabbits. It concluded, based on an analysis of more than 800 substances, that “it appears far-fetched to demand a developmental study in both species”.

This was based on the fact that these tests appear to be, intrinsically, very poorly reproducible. Results between the two species were found to differ up to 25-fold, solely due to poor reproducibility. This type of error was so great that it overwhelmed differences in test results due to the species used (i.e. because tests show different results in rats and rabbits anyway). Crucially, both factors play a role in generating data that differ across species, making results difficult, if not impossible, to inform any risk (or safety) for humans.

We naturally welcome this finding, and will urge those who commission these tests, and who accept their results, to explain how they intend to transition away from requiring data from two species as a matter of urgency. At the same time, we will remind them of other scientific evidence, showing that no animal-based developmental toxicity test is sufficiently predictive for humans, and that the need for a move towards more humane, human-relevant and predictive tests is irrefutable.

Humane tests using human stem-cells, organoids (tiny 3D ‘organs’ grown in the lab), body-on-a-chip (different types of human cells grown together and interconnected on ‘chips’) and ‘omics’ techniques (looking at changes in gene expression and protein production) are far more relevant to human safety, particularly when used together.

These non-animal methods allow researchers to incorporate human genetic diversity into the tests. Some of them have been formally validated (unlike the animal tests)—some for almost twenty years—yet are still waiting to be accepted by regulators of the chemical and pharmaceutical industries, despite accuracies of 80-100% ^[3–5] amid widespread acceptance and use by the research community. They are also quicker and cheaper—which is important, as many tens of thousands of chemicals are in regular use worldwide, with little or no safety information relevant to humans.

It is extremely disappointing that the pace of change is so slow. Animal tests in this field have been known to be very poor predictors of human risk and safety for many years, with poor concordance across species, and high rates of false positive and false negative results with regard to humans ^{[6, 7]}. They are, simply, not suitable for their stated purpose, yet continue to be used in in preference to scientifically superior—as well as humane—methods. Testing for just one substance is estimated to involve the use of more than 1,300 animals. Overall, this type of testing accounts for the use of around twelve percent of all animals used in chemical risk-assessment ^[8].

At Cruelty Free International we will continue to push for developmental toxicity testing to be animal-free, as scientific evidence demands. However, it is clear that an important, immediate first step towards this overdue change should be an end to tests that use two different species. This would save many thousands of animals from a great deal of suffering and death, as well as expediting a shift to testing methods that will not harm animals at all. This will also benefit people, by providing data directly relevant to them.

References

1. Braakhuis HM, Theunissen PT, Slob W et al. Testing developmental toxicity in a second species: are the differences due to species or replication error. Regul Toxicol Pharmacol 2019;107:104410.
2. Garg RC, Bracken WM, Hoberman AM et al. Reproductive and developmental safety evaluation of new pharmaceutical compounds. In: Gupta RC, ed. Reproductive and Developmental Toxicology. London: Academic Press, Elsevier 2011:101-28.
3. Worley KE, Rico-Varela J, Ho D et al. Teratogen screening with human pluripotent stem cells. Integr Biol (Camb) 2018;10:491-501.
4. Marikawa Y, Chen HR, Menor M et al. Exposure-based assessment of chemical teratogenicity using morphogenetic aggregates of human embryonic stem cells. Reprod Toxicol 2020;91:74-91.
5. Luz AL, Tokar EJ. Pluripotent Stem Cells in Developmental Toxicity Testing: A Review of Methodological Advances. Toxicol Sci 2018;165:31-39.
6. Bailey J, Knight A, Balcombe J. The future of teratology research is in vitro. Biogenic Amines - Stress and Neuroprotection 2005;19:97-145.
7. Bailey J. Developmental toxicity testing: protecting future generations? Alternatives to Laboratory Animals 2008;36:718-21.
8. Rovida C, Hartung T. Re-evaluation of animal numbers and costs for in vivo tests to accomplish REACH legislation requirements for chemicals - a report by the transatlantic think tank for toxicology (t(4)). ALTEX 2009;26:187-208.