In December 2016, I was invited by the European Commission to speak on a scientific panel at the conference Non-Animal Approaches – The Way Forward. The event was organised as part of the EU’s response to the citizen’s initiative ‘Stop Vivisection,’ which presented more than one million supporting signatures from across the EU to the European Commission in May 2015, calling for an end to animal experiments.
The final report of the European Commission conference has just been published. The key message of harmonisation – working together to replace animals in science – rings true, but spending even more time, money and effort in trying to fix the broken well of animal experimentation does not. Animals do not model human disease and it seems that the more we invest scarce public health resources in these failing animal models, the further we get from understanding human disease mechanisms. Transitioning away from failing animal models to human-based approaches to develop medicines for human disease must become the central thread of 21st century research.
My scientific career has been dedicated to developing practical tools for studying and developing treatments for human disease without harming animals. I am not alone in this. Recent decades have seen enormous strides in the development of non-animal technologies. These tools are known to be more relevant and predictive than ineffective, harmful animal research, but are not universally accepted by regulatory agencies. However, an exciting advance this month has seen the FDA announce a year long partnership with Emulate Inc., to evaluate the potential for advanced, non-animal methods in food safety testing. Initially, scientists will compare the human-based technology with the results of animal tests- but if successful, this could truly signal the beginning of the end for animal research.
Animal research certainly fails animals, in terms of the distress and suffering caused, and just as importantly, animal research often fails people, too, in terms of the slow, unproductive route to useful treatments. More than 90 percent of drugs that have passed animal trials for safety and efficacy are not successful in treating the human disease for which they are intended. It takes from three to five years and millions of dollars to put one chemical substance through a battery of animal safety tests (producing results that predict human outcomes around 50 to 60 percent of the time). Using new non-animal test systems, it would take an average of 20 minutes and $5,000 to test a chemical: producing results that are currently as predictive as animal tests but, crucially, that are improving rapidly as we extract more meaning from the huge amount of new data produced in these human-relevant, non-animal systems.
There are clearly issues on the “translation” of results from other species to humans, but there are also issues with the conduct and reporting of much of the animal research. More than 80 percent of studies using animals demonstrate experimental bias, such as the choice of animals for a particular study. For example, animal research has a huge gender bias; eight out of ten experiments only use male animals. Modelling diseases that specifically affect humans using non-human animals is hard to make sense of, but modelling diseases that affect women using male, non-human animals? This makes even less sense and is unlikely to ever provide relevant results.
The rampant selection bias that pervades so much research using animals means that millions of animals are experimentally manipulated, operated on, treated with test compounds and otherwise harmed, stressed and distressed to no appreciable result. The issues with how animal experiments are carried out also mean that less than 20 percent of published animal research is discriminating, rigorous and trustworthy.
There are guidelines on the reporting of animal experiments that aim to improve the quality of science, and which could in theory lead to a reduction in animal use. However, whilst the guidelines are endorsed by more than 300 scientific publications, uptake has been slow. Some funding bodies require their use, yet animal researchers are under no obligation to adopt the guidelines before being granted permission to use animals. Issues of biased reporting, skewed results, and ultimately the waste of millions of animals’ lives, remain.
Surely we can all agree that replacement of animals in testing and research is morally, ethically and scientifically the only way forward? A recent market research poll, commissioned by the UK’s Department of Business, Energy and Industrial Strategy, indicates that more than a quarter of UK citizens would support an outright ban of the use of animals in research. In November 2016, a publication in the scientific journal Drug Discovery Today used expert analysis of five human disease areas to show that animal models have failed to provide answers or treatments for any of them. This paper calls for a transition away from failing animal models towards human-relevant methods.
The inescapable truth is that attempting to model human diseases in non-human animals will always be problematic, and can delay progress in finding much-needed cures. Animal replacement remains the most promising way forward. The UK published its delivery plan to reduce animals in research in 2014, whilst earlier this year the Netherlands opened discussions about phasing-out animal testing by 2025. Enormous investments have been made in developing more promising solutions, including EUToxRisk to find non-animal solutions for chemical safety assessment. In the US, a joint investment by the National Institutes of Health and Defense Advanced Research Projects Agency is building mini-organs and linking them together in a system they are calling “human-on-a-chip.” These are the types of investments that will bring light to the future of medicine and environmental safety. We look forward to seeing the results from the ground-breaking partnership of the FDA and Emulate Inc and to an end for animals in research.