Over 100,000 chemicals are found in consumer products, and for most, little is known about their toxicity. Traditional toxicology tests rely heavily on animal models, which in addition to their ethical concerns, take too long and are too expensive to meet this need.
Fortunately, important steps are taking place to reduce, and maybe one day replace, animal use for this purpose.
“There is an urgent, worldwide need for an accurate, cost-effective and rapid way to test the toxicity of chemicals, in order to ensure the safety of the people who work with them and of the environments in which they are used,” noted Daniel Russo, lead author of a new study which reported the development of a computational method to predict the toxicity of new test compounds. “Animal testing alone cannot meet this need,” he said.
The researchers focused their efforts on data collected in acute oral toxicity tests, those conducted to assess the effects of test compounds that are administered by mouth. Animal tests used for this purpose determine the dose of a test compound that kills half of the population of animals tested, and there are currently no in vitro tests that regulatory agencies accept as stand-alone replacements for these animal tests.
To address this problem, Daniel, and other researchers at Rutgers University developed a novel, high-speed algorithm to predict the toxicity of chemicals. This algorithm is automatically able to extract data from PubChem, a public database with information on millions of chemicals. It then compares fragments of tested chemicals to those that are untested, and uses computational models to make a prediction about the toxicity of the untested chemical.
The algorithm demonstrated a promising success rate in predicting the oral toxicity of several groups of chemicals and highlighted new factors that could help determine the toxicity of a chemical.
The researchers see the value of this approach in helping to prioritize potentially hazardous chemicals for testing, which would significantly reduce the number of animals used in such studies. These kinds of algorithms could also be developed to predict different types of toxicity as well, having even farther-reaching effects on reducing animal use.
NAVS is excited to see innovative thinking like Daniel’s being used to help reduce and replace animal use. In the coming weeks, we will be reviewing the next round of research proposals using animal alternatives that we may fund through the International Foundation for Ethical Research. We will also once again be taking part in our annual visit to the Intel International Science and Engineering Fair. There, we will meet and award prizes to high school students from all around the world who are leading the way in developing smarter, more humane research methods–any one of them may make the next breakthrough in using alternatives to animals in science
Want to do more to support smarter, more humane science? Donate to NAVS today!
Source: “New algorithm allows for faster, animal-free chemical toxicity testing,” ScienceDaily, April 16, 2019.
Russo, D. et al. “Nonanimal models for acute toxicity evaluations: Applying data-driven profiling and read-across.” Environmental Health Perspectives, April 2019.