BBC Future

Will We Ever?

Will we ever… eliminate animal experimentation?

About the author

Alla Katsnelson is a freelance science writer specialising in biology, health and medicine. She writes for publications including Nature, Scientific American, and The Scientist. In a past life, she did a doctorate on mammalian brain development at the University of Oxford. You can find her online at


Will we ever… eliminate animal experimentation?

(Copyright: Science Photo Library)

Arguably one of the most heated debates in science, efforts to reduce the number of animals used in studies face many barriers, says Alla Katsnelson.

One of the most, if not the most, contentious issues in science is the use of animals in research. Scientists experiment on animals for a host of different reasons, including basic research to explore how organisms function, investigating potential treatments for human disease, and safety and quality control testing of drugs, devices and other products. Its proponents point to the long list of medical advances made possible with the help of animal research. Opponents believe it is cruel and meaningless, as observations in animals often do not translate directly to humans.

In 1959, William Russell and Rex Burch proposed their “3Rs” guidelines for making the use of animals in scientific research more humane: restrict the use of animals; refine experiments to minimise distress; and replace tests with alternative techniques. Over the course of five decades their guidelines have become widely accepted worldwide, and while the reliability of published reports on the numbers used varies, they do at least provide a snapshot of historical trends. Around 29 million animals per year are currently used in experiments in the US and European Union countries. (Rats and mice make up around 80% of the total.) This is less than half the total in the mid-1970s – a significant drop, but one that has plateaued in the last decade.

“In the late 1980s, people thought animal research was singing its swan song,” says Larry Carbone, a senior veterinarian at the University of California in San Francisco. Fresh out of veterinary school in 1987, Carbone landed a job as an animal vet at Cornell University, in New York State. At that time the numbers of animals being used in experiments and testing was on the decline: the campus was building a new multi-storey biotechnology facility, with just three rooms containing animal breeding and living facilities.

But then came the development of tools that could selectively modify individual genes in mice. This proved to be such a powerful and popular technique that the decreasing trend in animal use ground to a halt.

Now, a raft of novel experimental techniques may help to push numbers down again. Improvements in imaging methods that offer a peek inside the bodies of animals allow scientists to get more and better data from each experiment than before. For example, researchers previously had to cull multiple mice at different stages of tumour development, but now they can non-invasively watch the disease unfold in a single living animal using a fluorescent dye. Similarly, as brain-imaging techniques become more advanced, some questions that are now addressed with experiments in monkeys might be better answered by peering into the human brain. “My prediction is that human volunteers will be able to replace monkeys more and more in the next 10-20 years,” says Carbone.

Meanwhile in vitro advances are also pointing towards reliable alternative methods. One such advance is the ability to re-program human skin cells into a primordial, stem cell-like state. These “induced pluripotent cells” could be converted into any specialised cell in the body, like liver or kidney cells, and these could be generated from people with a particular illness, giving researchers a potent and patient-specific model of that disease in a dish. Lab-on-a-chip technologies – and perhaps one day, lab-grown organs – could also provide increasingly sophisticated ways to identify disease mechanisms or test prospective medicines.

Finding alternatives

Trends also show that some sectors are doing more than others to reduce animal use. Some believe technological advances will one day make animal studies unnecessary, while others argue that “non-living” models will never be capable of reliably replicating all of the uses of laboratory mice and other creatures.

When many people think about animal testing, they imagine rows of rodent cages in a pharmaceutical company lab. But according to data from European Union countries, the pharmaceutical sector uses almost half the number of animals that academic labs do, and animal use in drug development dropped significantly between 2005 and 2008 – the most recent statistics available. There are two reasons for this, says Thomas Hartung, Director of the Center for Alternatives to Animal Testing at Johns Hopkins University, in Baltimore, Maryland. First, drugs are increasingly designed to target specific molecular mechanisms, and these are best identified in culture dishes rather than live animals. Second, conducting experiments in 1,536-well cell culture dishes is vastly less expensive than in animals, so companies are motivated to use alternatives whenever they are available.

In the US and the EU, a drug’s efficacy and safety must be tested in animals before it enters human testing, though a 2010 directive from the EU calls for alternatives to be used when possible. Jan Ottesen, vice president of lab animal science at Danish company Novo Nordisk, which makes insulin and other drugs for diabetes and haemophilia, says his company actively seeks out tests that can replace animal use without compromising patient safety. Novo Nordisk decided 15 years ago to replace animal tests with cell cultures to verify the quality of each batch of drugs before it goes to market. The company had to provide the authorities with data proving that other tests worked just as well. It took until 2011 for the company to complete the switch. 

However, for some types of experiments there are no equivalent non-animal options, says Ottesen. For example, in searching for new drugs that decrease joint pain due to arthritis, you need a model that mimics the human condition. The important thing, he stressed, is to set up the experiment so as to avoid unnecessary pain. For safety and toxicological testing of drugs, he adds, “I cannot see for the foreseeable future how we can completely avoid it. Having said that, all the replacements that can be implemented should be implemented."

Under pressure

Safety testing of substances other than human and veterinary drugs, such as cosmetics, toiletries, household cleaning products and industrial chemicals might be a different story. Currently, says Hartung, such tests are outdated and inaccurate, with toxicity in rodents predicting problems in humans just 43% of the time. Meanwhile, tens of thousands of these substances have undergone no toxicity testing at all.

Addressing this gap with animal studies alone would be expensive and impractical. An overhaul of chemical safety regulations in the EU called REACH and a toxicology modernisation initiative led by the US National Institutes of Health, are driving the search for alternatives.

Hartung believes that with enough investment and coordination, animal tests on products in this category can be replaced completely. He is leading the Human Toxome Project, an initiative that aims to map the ways substances disrupt hormones and endanger health, as well as to develop advanced, non-animal lab tests for toxicity testing. It’s slow going, Hartung concedes. “We don’t have human data to compare with, or really high-quality animal data,” he says, adding that this makes it tough to evaluate the quality of the tests.

Meanwhile, almost four in ten animals are used in basic, as opposed to applied, biological research – and this proportion is growing. Sarah Wolfensohn, a veterinary surgeon who heads Seventeen Eighty Nine, a consultancy advising researchers on animal welfare, based in Swindon, UK, says this is in part because a lot of this type of work is carried out in academia where the financial and performance pressures that motivate interest in non-animal-based techniques are weaker than in the commercial sector.

Other factors play a role too, she says. "For example, if a senior professor in academia has spent his entire career developing experimental techniques on monkeys' brains and young researchers now tell him 'actually we don't need to do this, we can do it on a computer', it undermines his approach."

But just as important as reducing the numbers of animals used, adds Wolfensohn, is “to make sure they are being used in the best way and that their welfare is maximised, so as to get the best quality results, to make sure they are not wasted.”

Overall, pressure to limit the use of animals in research – either for financial, scientific or moral reasons – is rising. Meanwhile, the use of animals in many areas of life-science research is on the decline, experts note, even if genetic work in mice is still keeping numbers up. “I think this is temporary,” says Andrew Rowan, President and Chief Executive Officer of animal protection group Humane Society International. “I think it is going to start going down again as we improve our technologies.” How soon this might happen is too difficult to tell.

If you would like to comment on this story or anything else you have seen on Future, head over to our Facebook page or message us on Twitter.