I read on a regular basis that some new technology is going to replace the need for animal tests in drug development. This is true if your time frame is measured in decades but if the reader thinks these new technologies are going to eliminate animal testing in her lifetime, she is probably wrong. Two recent examples of this are the organism on a chip and a computer model of supposedly an entire organism.
Alicia Graef writing in her article, ‘Organ Chips’ Could Replace Animal Experiments states:
Popular VideoIt turns out President Trump's budget has $2 trillion error in it:
A $70 million research project that will develop transparent silicon microchips with hollow channels that contain actual living human tissue and pumps to replicate organ function is underway and is predicted to provide faster, cost-effective and more accurate results for testing diseases, toxins and pharmaceuticals – all on something about the size of a flash drive.
Someday, we will have the knowledge and technology to do this and if I were in charge of funding research, this type of research would be high on my list. However, science is a very long way from putting these chips into drug development as truly predictive modalities. Graef alludes to the real problem drug development has at present:
Popular VideoIt turns out President Trump's budget has $2 trillion error in it:
“Given the differences in cellular biology in the brains of rodents and humans, development of a brain model that contains neurons and all three barriers between blood, brain and cerebral spinal fluid, using entirely human cells, will represent a fundamental advance in and of itself,” said John Wikswo, the Gordon A. Cain University Professor and director of the Vanderbilt Institute for Integrative Biosystems Research and Education (VIIBRE), who is orchestrating the multidisciplinary effort.
Note I said graef alludes to the problem: rats and humans are different organisms and as each is a complex system these differences are going to make extrapolation, at the level or organization where drugs act, impossible. If rat brains are used on the chip, then the chip will be predictive, in all likelihood, for rats but not for humans. There are many difficulties with developing chip technology and ultimately human tissues must be used.
The reason I say Graef alludes to the problem is that nowhere does she then follow-up on the above and say explicitly that animal models as used today are not predictive for human response. The reader is led to believe that these chips will be an improvement on animal models but that animal models as used today are still very useful. This notion is reinforced by no doubt well meaning, but nonetheless scientifically illiterate, animal activists:
“This is an exciting example of how modern-day innovation can produce a humane and more reliable approach to understanding the inner workings of human disease without the need for animal suffering. The USA appears to be leading the way in funding alternatives, it is now time for the UK to catch up,” said Dr. Katy Taylor, scientific adviser for the British Union for the Abolition of Vivisection.
The second advance. According to ScienceDaily: “In a breakthrough effort for computational biology, the world's first complete computer model of an organism has been completed, Stanford researchers reported last week in the journal Cell.” The scientists reviewed the literature and input data from 900 papers regarding chemical interactions in the bacterium Mycoplasma genitalium.(Karr et al. 2012) This resulted in inputting 1,900 experimentally determined parameters. Make no mistake; this is a huge accomplishment. But it by no means lives up to the hype. The article continues:
Most biological experiments, however, still take a reductionist approach to this vast array of data: knocking out a single gene and seeing what happens. "Many of the issues we're interested in aren't single-gene problems," said Covert. "They're the complex result of hundreds or thousands of genes interacting." This situation has resulted in a yawning gap between information and understanding that can only be addressed by "bringing all of that data into one place and seeing how it fits together," according to Stanford bioengineering graduate student and co-first author Jayodita Sanghvi. Integrative computational models clarify data sets whose sheer size would otherwise place them outside human ken.
All of this is true and is also the reason why, based on the knowledge of life we currently have, it is still impossible to reproduce even a relatively simple life form on a computer. (See what the characteristics of a complex system are and what this means for animal modeling in our article Animal models in an age of personalized medicine.) The whole organism on a computer claim will not end animal experiments or even begin to end animal experiments, as the whole organism is not there yet.
Animal protection organizations jump on announcements like these for two reasons. 1. They claim credit for these advancements based on the pressure they are putting on society to end animal experiments. AND if you just keep sending them money these types of advancements will continue. In reality, these advancements were inspired by about everything except animal protection organizations. 2. Most animal protection organizations are scientifically illiterate and do not understand what the above actually means or, for that matter, what the scientific problem with animal testing really is.
In light of the fact that we are a long way from totally replacing animal-based research, and or testing, with computers, the question society should be asking is still the same as it was last week and last decade: Is animal-based research a predictive modality for human response to drugs and disease? If the answer is no, then animal models should not be used for experiments where the results are claimed to be predictive. Period. Regardless of what else is or is not available. Developing computer-based technology to predict human responses is a great idea and should be pursued. Developing a cure for cancer is also a great idea and should be pursued. But in the meantime, animal models should not be used as predictive models and homeopathy should not be used to treat cancer. Neither work! This is where animal protection groups should be focusing. As long as animal protectionists blindly go along with the myth that animal models are predictive, society will blindly demand predictive nonanimal technologies before allowing animal tests to cease. As I might have said once or twice, loving animals does not make you a scientist. If you, as an animal protectionist, do not have a doctorate in science yet think you are competent to speak to the science of vivisection, I suggest you look familiarize yourself with the Dunning-Kruger effect. Ditto if you have a doctorate but, when you read my writings, you don’t really understand what I am saying.
The notion that animal-based research is predictive for human is unquestioned by society in general and this is illustrated by the following from the UK. Mike Hancock, an MP from Portsmouth South, formally asked the Secretary of State for the Home Department “what recent research her Department has (a) commissioned and (b) evaluated on the efficacy of animal experiments.” Lynne Featherstone replied that
The Government has not commissioned or evaluated any formal research on the efficacy of animal experiments.
This is because they accept as fact the notion that it is predictive and see no reason to waste money on reinventing the wheel. The lobbyists in the UK have done their job well.
This is for a number of reasons. Put simply, animal research is already routinely subject to a number of stringent tests.
By tests Featherstone means regulatory tests not scientific tests. Featherstone continues:
First and foremost, unless an experiment is judged to be potentially efficacious it cannot be licensed under the Animals (Scientific Procedures) Act 1986.
To the first approximation what Featherstone means is that animal models are indeed predictive. But that is not all that she is saying. This verbiage is the usual justification for animal-based research in general including basic research with animals. The vivisection activist will say: “Well, basic research that uses animals might lead to something important. You can’t prove that it is impossible that basic animal-based research will lead to anything useful.” And the vivisection activist is right. Essentially anything might lead to something useful. Biomedical advances have come from very unusual places. Dr. Swan observed sailboats in the Pacific Ocean off the coast of southern California and had the idea of attaching a sail onto a catheter thus allowing it to float into the heart with the blood. This was used to develop a catheter for determining the pressures inside the heart, the output of the heart, and imaging the heart among other things. Novel ideas have even come from dreams. Yet, no one funds scientists to go the beach or take naps. The reason being that while almost anything is possible in terms of leading to cures for human disease, not all things are equally probable. The smart thing to do would be to fund types of research that have a high success rate and or a basis in science for expecting success. Animal-based research has neither. Human-based research does (as does basic research in the physical sciences). But lobbyists, and hence Congress, do not care about doing the smart thing. Congress just cares about raising the money to get re-elected, which comes from lobbyists, be it directly or indirectly.
Where an application relates to ongoing work, inspectors will assess the evidence of benefits from previous work in advising on authorisation.
For a thorough debunking of this concept see our article: An analysis of the Bateson Review of research using nonhuman primates. The people assessing the value of the research are, for the most part, other animal modelers or people with an indirect vested interest in animal modeling. Guess the outcome. Despite minimal returns on the investment, the people with a vested interest say every time that more money is needed for such projects and that the animal-based research in question should be allowed to proceed. You don’t need a doctorate in science to see the conflict of interest here.
Also, local ethical review processes in licensed establishments review the conduct of the work undertaken under licence at their establishments.
Local review boards are composed primarily of the people with a vested interest in the process. Note what one of the inventors of local ethics boards, Carbone, stated:
Few people realize that virtually nothing is prohibited by the Animal Welfare Act, so long as it can be justified to the animal care and use committees. Nor do IACUCs, by and large, function by rejecting animal protocols when the ethical costs are too high. Unlike granting and funding agencies where money is a limited resource, IACUCs have no limit on the number of protocols they can approve. They can approve all or none, but as Russow points out, their general operating philosophy is roughly: Given that this project is going to be done, is it being carried out as humanely as possible (Russow 1998)? This is especially true if a project has been favorably peer-reviewed by a competitive granting agency such as the NIH. Given this as their starting point, how does one assess the effectiveness of IACUCs? . . . I’ve worked in IACUC protocol review since its mandated inception in 1986 and have rarely seen a protocol rejected. (Carbone 2004) p183-4
In addition, research councils and charities evaluate the research projects carried out under the 1986 Act for which they provide funding. Work funded by pharmaceutical companies is subject to internal scrutiny within those companies, and the safety and efficacy testing needed before people are exposed to new drugs is evaluated by the relevant regulators.
All of the above, that are supposedly scrutinizing the process, are composed of the people doing animal-based research along with their lackeys and have the same philosophy referred to by Carbone.
Featherstone’s comments are just one example. In the past, I have referenced numerous scientists that claim animal models are predictive and or that animal-based research is the best way to advance medical care and so on. The reason animal experimentation persist is not because society does not have adequate replacements. Society, and most animal protection groups, do not understand exactly how animal models are used and how completely useless they are for predicting human response. If you are considering donating money to an organization that supposedly opposes vivisection, check and see what their position is regarding the ability of animal models to predict human response to drugs and disease and how they are going about ending the practice. I think the presence of an informed, educated donor base would eliminate >90% of anti-vivisection donations in the US and Europe which would then force so-called animal protection groups to become competent. (For more on this topic see the three essays on animal protection on the AFMA website and recent posts regarding HSUS (here and here).)
Reality is not always pleasant, but that does not make it any less true.
Carbone, Larry. 2004. What Animals Want: Expertise and Advocacy in Laboratory Animal Welfare Policy. Oxford: Oxford University Press.
Karr, Jonathan R, Jayodita C Sanghvi, Derek N Macklin, Miriam V Gutschow, Jared M Jacobs, Benjamin Bolival, Nacyra Assad-Garcia, John† Glass, and Markus W Covert. 2012. A Whole-Cell Computational Model Predicts Phenotype from Genotype. Cell 150 (2):389-401.