February 15, 2008 — -- An ambitious program announced Thursday by a coalition of government agencies could lead to the end of animal testing to evaluate the safety for humans of new chemicals and drugs.
Three agencies -- the Environmental Protection Agency, the National Toxicology Program and the National Institutes of Health -- have signed a "Memorandum of Understanding" to develop and implement the new methods. The collaboration is described in today's edition of the journal Science.
The agreement is a "milestone," says Martin Stephens of the Humane Society of the United States. "We believe this is the beginning of the end for animal testing. We think the (conversion) process will take about 10 years."
The agencies acknowledge that full implementation of the shift in toxicity testing could take years because it will require scientific validation of the new approaches.
The Humane Society and other activist groups have long protested the use of animals to test the safety of chemicals, particularly those used in cosmetics and other personal products. The agencies noted that the public's "unease" with animal testing, in addition to a growing number of new chemicals and high testing costs, fueled the new collaboration.
Although there are no actual figures, Stephens says his "best guess" would be that about 10 million animals a year are used in toxicity testing, mostly mice, rats, rabbits, guinea pigs, and then lesser numbers of dogs, monkeys and other species.
Historically, toxicity has been identified by injecting chemicals into animals and seeing whether they were harmed.
"It was expensive, time-consuming, used animals in large numbers, and it didn't always work," says Francis Collins, director of the NIH's National Human Genome Research Institute.
The new systems the agencies hope to use rely on human cells grown in test tubes and computer-driven testing machines. They allow the scientists to examine potentially toxic compounds in the lab rather than injecting them into animals.
The EPA has begun evaluating 300 chemicals using the new methods.
The first phase should be finished this year, says Robert Kavlok, director of the National Center for Computational Toxicology.
Thousands of chemicals can be tested at one time under a method that uses a 3-by-5-inch glass tray with 1,536 tiny wells, each a fraction of a millimeter across, says Christopher Austin, director of NIH's Chemical Genomics Center.
A few hundred human cells grown in a test tube go into each well. Then, guided by a computer, the testing machine drips a different chemical into each well. After a while, the machine shines a laser through each well to see how many cells remain. A computer analyzes the toxicity of each compound based on how the cells react.
By comparison, it's taken the EPA 30 years to rigorously test 2,500 potentially toxic compounds, says Elias Zerhouni, director of the NIH.
All the data produced will be put into a public database.
"We think it is very important for the entire public worldwide to have access to these very precious experimental results," Kavlock says.
The agencies' agreement is the fruit of work begun in 2005 by the EPA and the National Toxicological Program to speed up toxicological testing. That resulted in a report by the National Research Council last year laying out how quicker testing might be done.
The federal agencies will start their efforts with compounds previously tested on animals to confirm that the cell-based tests are accurate, Collins says.
Animal testing won't disappear overnight, but the agencies' work signals the beginning of the end, Zerhouni says.
The testing shift began to take shape when scientists realized the same methods drug companies used to test compounds for therapeutic purposes could also be used to see whether they harmed cells.
It's a wonderful example of what scientists always hope for, Collins says.
"You develop a technology for one purpose, and you realize, 'Goodness! We can use it for something else!' "
He likened it to the evolution of the military's data transmission projects of the 1970s, asking, "Who would have thought that would result in the Internet and the Web?"