In his classic Microbe Hunters, American microbiologist Paul de Kruif wrote of nineteenth century researchers responding to diphtheria epidemics that filled Paris hospitals with dying children. “Their diphtheria soup paralyzed rabbits!” de Kruif wrote. “... The delighted experimenters watched... [as] they died in a clammy, dreadful paralysis.” Cruel as it was, animal testing helped produce a successful antitoxin in the 1890s, eventually saving thousands of children from terrible, gasping deaths.

 

Since those early days of experimentation, millions of rabbits, dogs, guinea pigs, and sheep, as well as laboratory rats and mice, have been exposed to pathogens by scientists hoping to elucidate disease mechanisms. But what is the future of animal testing today, when alternatives are available that are often faster, cheaper, and more accurate—as well as, most would say, more humane?

 

According to a “Guidance to Industry and Reviewers” issued by the U.S. Food and Drug administration in 2006, animal studies of pharmaceuticals “are designed to permit the selection of a safe starting dose for humans, to gain an understanding of which organs may be the targets of toxicity, to estimate the margin of safety between a clinical and a toxic dose, and to predict pharmacokinetic and pharmacodynamic parameters.”

 

Animals are still widely used in basic and applied research, as well as in safety testing, but corporations, universities, and government agencies are under pressure to find new approaches. Beyond concerns about the ethical treatment of animals, much animal testing is a blunt instrument, and has never been subjected to the kind of validation studies that new tests undergo. Thomas Hartung, director of the European Centre for the Validation of Alternative Methods (ECVAM), calls the old toxicity tests “simply bad science.” They may over- or underestimate toxicity or carcinogenicity, principally because of species differences. A child, for example, can happily eat an amount of chocolate that might poison a dog of the same size, because dogs metabolize theobromine much more slowly.

 

Until the year 2000, the number of animals used in all forms of testing had been dropping—to about half the levels seen in the 1970s. The reasons for the drop included growing concerns about animal welfare and a major reduction in animal testing of cosmetics and personal care products. But in recent years, the number has slowly risen, owing to the use of transgenic animals, including “knock-out” and “knock-in” mice (mice with particular genes either inactivated or inserted).

 

These changes can be seen in European and Canadian statistics. Figures on use in the U.S. are harder to come by as the Animal Welfare Act says the use of mice and rats—the most common lab animals—doesn’t have to be recorded. The United Kingdom’s Research Defence Society (RDS) estimates that the U.S. uses 15 million animals a year. Animal rights groups think it’s more. Various estimates of worldwide use range from 50 million to 100 million.