#289      37 min 07 sec
Fur and against: Scrutinizing the efficacy of animal testing and its alternatives

Toxicologist and pharmacologist Prof Thomas Hartung explains why animal testing is often unnecessary or of questionable efficacy. He discusses the emerging protocols and technologies that enable development of safe products without the need to conduct animal testing. Presented by Dr Dyani Lewis.

"This antibody was tested in rats and monkeys at 500 times higher doses than finally going into humans. And the first six volunteers which received this drug were in intensive care six hours after. So it is not necessarily protective just because monkeys have not had a problem with a substance." -- Prof Thomas Hartung




Prof Thomas Hartung
Prof Thomas Hartung

Thomas Hartung is Professor of toxicology, pharmacology, molecular microbiology and immunology at the Bloomberg School of Public Health at Johns Hopkins University, and the University of Konstanz, Germany. He is also Director of the Center for Alternatives to Animal Testing at Johns Hopkins University and has devoted his career to promoting a paradigm shift in toxicity testing to improve public health, via a strong evidence base. As the previous Director of the European Centre for the Validation of Alternative Methods of the European Commission (2002-2008), Thomas was involved in the implementation of the European REACH legislation and the ban on animal testing for cosmetics. Through this EU leadership, and as the current Director of the Center for Alternatives to Animal Testing, he assumed a key role in promoting the 2007 NRC vision document “Toxicity Testing in the 21st Century – a vision and a strategy” and the success in the translation of concepts for evidence-based medicine to toxicology. This aims for systematic assessment of the quality of all tools for regulatory toxicology, and the development of new approaches based on annotated pathways of toxicity. To the field of animal alternatives for experimentation and testing, Thomas brings a broad background in clinical and experimental pharmacology and toxicology, documented in more than 350 publications. Thomas has been the recipient of many awards including the 2009 Russell & Burch award of the Humane Society of the US, the 2010 Agilent Thought Leader award, and the 2006 US Society of Toxicology Enhancement of Animal Welfare Award.

Credits

Host: Dr Dyani Lewis
Producers: Eric van Bemmel, Kelvin Param, Dyani Lewis
Audio Engineer: Gavin Nebauer
Voiceover: Nerissa Hannink
Series Creators: Kelvin Param & Eric van Bemmel

View Tags  click a tag to find other episodes associated with it.

 Download mp3 (33.9 MB)

VOICEOVER
This is Up Close, the research talk show from the University of Melbourne, Australia.  

DYANI LEWIS 
Hi, I'm Dyani Lewis, thanks for joining us.  When we take a pill, put on makeup or purchase a new sofa, we usually take it for granted that these products are safe for us.  The confronting reality is that for decades our safety has largely relied on drugs and cosmetics and household chemicals being tested on animals.  But do we need to test on animals as much as we still do?  Few of us would now argue that our vanity should be at the expense of animals, but what about testing for drug safety?  What are the alternatives to animal testing, and how should product researchers and government regulators change their processes if the alternatives to animal testing are as good as, or perhaps even better than, the tests we conduct on animals?  I'm joined on Up Close today by Professor Thomas Hartung, a pharmacologist, toxicologist and leader in the development of alternatives to animal testing.  Professor Hartung is director of the Centre for Alternatives to Animal Testing at the Johns Hopkins Bloomberg School of Public Health in Maryland, and he is here as a guest of RSPCA Australia and the University's Animal Welfare Science Centre and Office for Research Ethics and Integrity.  Welcome to Up Close, Thomas.

THOMAS HARTUNG 
Thanks for having me.

DYANI LEWIS 
Thomas, why do we use animals for safety testing?  Where did this practice come from and why has it become such a standard practice?

THOMAS HARTUNG 
I think this is simply to do with the fact that small animals entered the laboratory in the '20s of the last century, exactly when we started to realise that products have problems.  So we solved some of the problems of drugs and cosmetics at the time by using the latest technology, which was animals.

DYANI LEWIS 
Do we know how many animals are used each year for these kind of scientific purposes?

THOMAS HARTUNG 
The statistics are very different in different parts of the world, but people have tried to extrapolate, roughly 50 million to 60 million are put into animal experiments worldwide.  Of these, roughly 10 per cent go into safety testing of chemicals, drugs, another 10 to 15 per cent into the testing of vaccines.

DYANI LEWIS 
So what kinds of tests are animals actually subjected to for safety testing?

THOMAS HARTUNG 
Everything you can possibly imagine.  They are exposed to chemical substances in safety testing. The possible ways, which means through the skin, through the air, through the gastrointestinal tract, injected.  Whatever we feel is best reflective of the process.  Then we are observing effects like development of cancer, the development of local necrosis, as we call it, a local destruction of tissue, to organ toxicities, failure of the kidneys and others, some of them being terribly painful and leading in the end to the death of the animal.  Others are pretty harmless and are stood very easily as, for example, many substances we are using are non-toxic and do not really harm the animals.  

DYANI LEWIS 
But ultimately those animals will be at the expense of these tests.

THOMAS HARTUNG 
The animals will typically be sacrificed.  So we need to ask what animals are used, and this changes also public perception very much if you give this piece of information.  About 80 per cent of the animals we are using are rats, mice and rabbits and about 10 per cent are fish and frogs.  The very precious organisms we talk about, monkeys, non-human primates, cats, dogs, they are altogether much less than five per cent and this changes perception, though all these animals are sentient beings.  A single rabbit which we can save is important and they're not feeling less pain than a dog does, though we are more attached most probably to dogs, most of us at least.

DYANI LEWIS 
There are obviously many different products that are developed for human use, from drugs to foods to chemicals in carpets and things like that.  How do we decide what products require testing on animals and which aren't tested on animals?

THOMAS HARTUNG 
That's actually pretty odd.  Most of it is historical.  You cannot explain why rationally a pesticide is tested in some 30 animal experiments and most chemicals, industrial chemicals, are never tested.  Most of our food additives which we have in many, many thousandfold higher concentrations than pesticides in the same products are not tested at all.

DYANI LEWIS 
You mentioned it's historical, but does that mean that the end product that we are getting has different levels of safety for us?

THOMAS HARTUNG 
Yes, it has.  It makes no difference what the purpose of a substance is, it does not make a difference whether it's natural or synthetic.  Most of the dangers come as big surprises.  There's, as I said, a historical reason for this, DDT and other pesticides led to big debates in the '60s after Rachel Carson's book, Silent Spring and we have developed [a] high level of consciousness of these problems.  At the same time, most people are not aware that we are having some 4000 food additives and we're missing 80 per cent of the data on the safety of these. Or of the new chemicals which entered the market in the US over the last 30 years.  Only one in eight has any toxicological data.

DYANI LEWIS 
So what about the safety for the environment?  Just because a substance might be safe for humans obviously doesn't mean that it is safe for other animals; birds or fish or even more distantly related animals like invertebrates.  So do we test for safety for other animals as well?

THOMAS HARTUNG 
Even less than for humans obviously.  There's reasonable testing on fishes, but we take one fish for all the hundred thousands of different types of fish and arguably they differ more even than a mouse or rat we test on to a human than different fish species, which are really quite different.

DYANI LEWIS 
Coming back to human products, animal testing rests on the assumption that we can work out from what might happen when humans are exposed to the substance.  We can work out what the risk is based on what's going to happen in an animal test species.  How well do animals actually represent humans?

THOMAS HARTUNG 
Nobody knows, that's the $10,000 question, or actually it's the $10 million or $10 billion question, because if we would know, we would know whether our investment of some billions every year we spent on safety testing are well invested.  If you look into how well do animals predict each other, then we get an estimate and we have pretty good data on mice to rats, which are obviously pretty similar, and for all of the more complex questions we can ask as toxicologists, it's only about 60 per cent. 

DYANI LEWIS 
So if something's dangerous for a mouse, there's a 60 per cent chance that it will be also dangerous for a rat?

THOMAS HARTUNG 
Exactly, so I'm referring, for example, to carcinogenicity, so does a substance produce cancer.  The correlation that we have the best possible data is only 57 per cent.  Similar numbers we have for reproductive toxicity, does it produce malformations in embryos, 61 per cent.  This tells us that the prediction of human health effects is likely much lower by using these animals.

DYANI LEWIS 
Does that mean that we should be testing on animals that are more closely related to us?

THOMAS HARTUNG 
This would be one approach.  It is actually something we do; we use non-human primates, but this is first of all a costly exercise, these are very precious creatures.  We also have much more resistance by the general public in using these animals and we are not necessarily getting much better data.  We had a scandal a couple of years ago, some might recall it was an antibody, TeGenero, and this antibody was tested in rats and monkeys at 500 times higher doses than finally going into humans.  And the first six volunteers which received this drug were in intensive care six hours after.  So it is not necessarily protective just because monkeys have not had a problem with a substance.

DYANI LEWIS 
Presumably then the success rate for taking a drug to market, using these tried and true processes going through clinical trials, first starting in animals, must be quite low then?

THOMAS HARTUNG 
It is and it's getting actually lower and lower.  The latest data suggests that only five per cent of drugs which go into human trials make it.  So all the promise of being safe and being efficacious was not held.  This is a dramatic problem for the pharmaceutical industry because it is now about $1.4 billion to develop a new drug, get it to the market, and a lot of this is simply due to the fact that we put our money on the wrong horses.

DYANI LEWIS 
And I guess the opposite could be true as well; we might be discounting drugs that could be very useful in humans, but because they haven't shown their efficacy in animals, we discount them before they even have the chance to get anywhere.

THOMAS HARTUNG 
This is a very often overlooked problem.  Let's take an example; we would sort out in most drug developments substances which are damaging to DNA, because this is possibly leading to cancer.  But the tests we are using would be failed by kitchen salt or sugar, so our standards are as high that we could not develop this to become a drug.  Or you could say another way around, aspirin, the oldest synthetic drug on the market, synthesised in 1897, it would fail essentially all of the drug tests of today and would be very unlikely to get to the market.  So for example, it's producing embryonic malformations in mice, rats, rabbits, cats, dogs and monkeys, but not in humans.  We are actually sure there is meta-analysis on 89,000 pregnancies and there's no indication of malformations by aspirin.

DYANI LEWIS 
This is just purely down to the subtle differences in biology between organisms?

THOMAS HARTUNG 
Yes, it is a matter of this and doses.  So we are trying to be on the safe side by applying much, much higher doses of a given substance to the animals and very often you see then things which are not observed when you're using the actual treatment doses.  Let's take another example; most recently Coca-Cola had to change their recipe in California for the first time in 140 years because one of the colourants, a caramel colourant, turned out to be producing cancer in rats at doses where you would have to drink 850 cans of Coke per day.

DYANI LEWIS 
Right, so not really matching up what the reality would be anyway?

THOMAS HARTUNG 
Not matching up the reality, but as soon as a property is found of a substance which indicates it is possibly leading to cancer, it cannot be used and added to food anymore.

DYANI LEWIS 
Is that because we can't go through tests to find out what the long term consequences are of low doses of those substances are?

THOMAS HARTUNG 
Yes, exactly.  If you want to see long term effects it gets costly, because you really have to treat the animals for long and observe many, many possible effects in each and every organ.  If you're lowering the doses, the probability of seeing an effect goes down, so you need more and more animals.  So it gets prohibitively costly to test something for long term low dose.

DYANI LEWIS 
You're listening to Up Close.  I'm Dyani Lewis and in this episode we're talking about alternatives to animal testing with toxicologist and pharmacologist, Professor Thomas Hartung.  Thomas, are there some classes of chemicals where we already know enough to say that no further animal testing is required?

THOMAS HARTUNG 
For some substances we know their mode of action, the way they work and we have established tests which represent this effect and we can test whether a similar substance is actually producing this effect.  Sometimes we can even do this by computer simulation. So we can use structural similarity to help up, but this requires understanding of mechanism and that's the revolution which is taking place at the moment in the science of safety sciences.

DYANI LEWIS 
So how do you go about working out mechanisms of action and then how can that help with safety testing?

THOMAS HARTUNG 
I mean this is taking place in all life sciences at the moment.  We have taken our animal models, or our cell culture models, at the beginning mainly as black boxes.  So we put something in and we measure what comes out.  But we did not elucidate what is happening within the organism, what is happening within the cell.  Over the last 20 years, I would say, we have seen a tremendous change, not only sequencing of the human genome and understanding of gene function in both humans and animals, but also biochemistry, understanding the molecular targets of substances.  And this is the principal paradigm shift away from models which are taken as a black box and which deliver reasonably similar results, but we don't know what is happening, to something where we understand this is the way the damage propagates and that's how we can test also very specifically whether a substance has this property.

DYANI LEWIS 
And so it really comes down to sharing of information, pooling of information from different resources.  Is this how we could eventually rely more on information rather than these blind tests of add the substance to the animal and see what happens?

THOMAS HARTUNG 
It is one of the visions.  The problem is no individual can oversee the scientific literature anymore.  The number of publications in any given field - and this includes toxicology, the safety of substances - is hundred thousands per year and it's very difficult to sort the good ones from the bad ones.  So we hardly have any quality measures.  What is a good publication?  Is it because it's published in a good journal?  Or is it because it is documented more precisely?  Or is it something which is following certain quality regimes which we have established?  All of this comes together in the end to form quality and it is very important that we have means of evaluating existing information and condensing this information.

DYANI LEWIS 
I guess this kind of information wouldn't take into account combinations of chemicals, as you might have in a makeup formulation, for example.

THOMAS HARTUNG 
No chance at the moment.  You have to imagine, given the current costs of testing, to test one substance for a possible carcinogenic effect, so producing cancer in rats or mice is about $1 million per substance, which allows us to test three concentrations.  This is not giving us really the opportunity to test combinations of substances, far too expensive, because it's not only about putting two substances together, you need to find the different mixes.  You can also have different time courses, you could treat with one before the other, you can modulate during the lifetime treatment to create windows of vulnerability.  And this is absolutely impossible in tests which are that costly.

DYANI LEWIS 
You work on a project called the Human Toxome Project.  Could you tell me a bit about what the vision for that is?

THOMAS HARTUNG 
The idea is there's 50 ways to leave your lover and there's a couple of hundred ways to harm a cell, but not an endless number.  Our biology is pretty robust and there's very few let's say Achilles' heels of the cell, the points where a cell really can be damaged, because most of it goes on the self repair, adaptive responsiveness, otherwise we would not turn 100 years old.  So it is extremely important that we identify what are the critical infrastructures, what are these vulnerable points of a cell.  There's most probably only a number of ways which are converging to these targets which can actually produce harm.  This is at this stage a hypothesis, nobody can say whether we can define something like a pathway of toxicity because we are talking actually about networked systems.  It's a rather complicated thing, but we believe that there are certain ways of perturbing these networks which are characteristic and unique.  The Human Toxome Project aims for producing a process of how you can deduce such pathways, how you can annotate them and share them in a public database.  

DYANI LEWIS 
What about in vitro methods such as tissue culture that might be able to replace whole animals?

THOMAS HARTUNG 
I mean it's very optimistic, let's say it carefully, that a single cell type in a culture can actually reproduce the complexity of an organism and say something about its vulnerability.  You can do so if this single cell is happening to be the point of vulnerability in the organism if the mechanism is crucial and that without this mechanism nothing is going to happen.  We have some cases like this and some 50 methods have been internationally validated to do such a job, but these are typically the very simple cases.

DYANI LEWIS 
What sort of cases are used with tissue culture then?

THOMAS HARTUNG 
Well we have been successful with acute toxicities, so direct poisoning where we at the moment are killing dead animals, because these substances typically also damage cells very effectively.  Or we have been quite effective with topical toxicities, which means things we can observe directly on the skin, in the eye, because here we can reproduce skin and eye tissues and can observe exactly the same local reactions.  But it's getting difficult when you're talking about long term chronic effects, when we are talking about the effect on the reproductive cycle, including embryonic malformations, when we are talking about cancer, which as we know in humans lasts often 10 to 15 years to develop.

DYANI LEWIS 
So these sorts of systems, have they replaced tests that the cosmetic industry, for example, would have used?

THOMAS HARTUNG 
Yes, they've been very successful, there is international harmonisation, so OECD for chemicals and the International Conference on Harmonisation for drugs have adopted quite a few of these methods, especially those on human skin are quite popular.  But we have to admit these are a small percentage of all animals used because we could do this with one to three rabbits in the past, so you're not saving a terrible number of animals by using them.

DYANI LEWIS 
What about in terms of cost?  How do they compare - how does tissue culture, for example, compare with just using animals?

THOMAS HARTUNG 
It depends very much how you calculate.  I mean tissue culture is first of all quite an investment, you need a high infrastructure, you need a high level of training of people, more compared to animal experiments.  On the other hand, you can automate things, you can do thousands of substances in parallel, you can miniaturise systems and the cost for evaluating the safety of a drug, a high production chemical or a pesticide, is in the range of $5 million to $20 million.  So you can do quite a bit of cell culture for testing one substance.  On the other hand, it is not really the actual cost, it is the cost of the wrong decisions which is really driving us.  If you sought out the wrong substances in drug development and they never make it to therapy because of a misleading animal experiment, then this is far more costly than any animal experiment you could possibly have done.

DYANI LEWIS 
How do we go about validating an alternative to animal testing, given that what we are comparing it to, the status quo, is testing on animals, and you've already mentioned that there are quite a few failings in that method?

THOMAS HARTUNG 
That's the big question.  I mean first of all we want to do two things; we want to show that a method is reproducible and we want to show that the method is relevant.  Reproducible is pretty easy.  It still costs because you have to do it at different labs under different conditions, but it's a very doable thing, but it does not tell us that we get a result we can rely on, so we need to show that it is relevant.  Our approach so far has been that we are comparing to the animal experiment which is in place, but this means also that we inherit all of its flaws, we will always judge the method only as approximating the old standard.  
To give you one example, I was responsible for a validation body for six years; officially in Europe, responsible for doing this exercise.  Once I saw a test which identified thalidomide, the most notorious embryotoxicant, which identified it in the mouse cell system, but people wanted to call it a false positive because mice don't show this effect.  They said the mouse cell culture which shows this effect is a false one.  It gets really ridiculous and we will never become better.  So what is important for us is to find a better point of reference to look to.  And here the only thing which we can offer at the moment in the absence of large data sets for humans is scientific mechanism.  So we could show that our cell systems are giving results in line with the scientific literature.  We call this mechanistic validation but this is an astonishingly young proposal which dates back from last spring (2013).  So far we have really been doing things most on the basis of correlation do we get results similar to the ones we expect.

DYANI LEWIS 
But again, a lot of the scientific literature would be based on mechanisms that have been worked out in animal systems.

THOMAS HARTUNG 
Yes, but mechanism is not necessarily wrong, mechanism can very often be translated and we can test this, because we can demonstrate whether a cell system based on human cells is replicating the mechanism.  The difficulty is more that the complexity of an entire organism does obscure mechanism, so you don't see the tree because of the forest, as we would say in Germany.

DYANI LEWIS 
I'm Dyani Lewis and my guest today is pharmacologist and toxicologist, Professor Thomas Hartung.  We're talking about animal testing and its alternatives here on Up Close.  Thomas, what about the scientific community itself?  How does it play a role in reducing animal testing?

THOMAS HARTUNG 
The scientific community is usually following the money.  The funding for alternatives to animal testing has made a tremendous difference, first in Europe where some, I would say, 100 million is spent per year for the development and validation of alternative methods and now most recently in the US, with advancements towards new regulatory sciences, human-on-a-chip type of approach has received unseen amounts of money, the latter $200 million in the last two years, for example.  At the same time, it is a few institutions like ours, the Centre for Alternatives to Animal Testing, which are let's say trying to be an opinion leader and organise the information which is available.  Our centre has been created in the US 32 years ago and is first of all an information hub, trying to organise our knowledge and make it accessible to people in a let's say quality controlled way that every researcher can find information on what is possible, but also lay audiences and others.  
We are driving this by entertaining a think tank for toxicology, developing new concepts, because we as academia are outside of the pressure of regulating or being regulated and we can let's say toy around, we can develop the new ideas with the idea of if I was king what I would do.  So this is a type of approach which only academia can actually pursue and we are very unusual in the sense that we are also developing some pilot cases.  We demonstrate that some of the things can be done we are talking about as proof of principles, or that have even run policy programs, both on Capitol Hill in the US and in the European Parliament.  So we are trying to show policymakers where in their legislation they can open the doors for new approaches.

DYANI LEWIS 
Many would argue that there have been a great deal of medical advances that would not have been possible without the use of animals in research and also in drug safety testing.  Would removing animals from research hamper that medical progress?

THOMAS HARTUNG 
I would say at the moment yes.  I have no doubt that animal models have contributed, but they also have helped to confuse us and have helped to these big failure rates which we are seeing at the moment.  It is incredible that we have a permanent decrease in the number of drugs making it to the market, given the investment over the last four decades, to an extent that the model of pharmaceutical industry is less and less sustainable.  The idea of having a blockbuster which brings back all of the money of investment is very, very difficult and can only be done for some major advances in major diseases.  So situations where everybody switches instantly to the new drug, because you have only a few years under patent, and where the market is big enough so you have many patients.  This means that all of the pharma companies are focusing on the same set of diseases at the moment and many things are never tackled because they are no longer an attractive market.

DYANI LEWIS 
And they're also still all going through the same process using animals in their testing.  So what role do regulatory bodies play in changing the status quo for animal testing?

THOMAS HARTUNG 
That's actually the interesting situation, that we have an atmosphere of change at the moment which comes from agencies.  The agencies have over decades been very conservative, as they most probably should be if safety's at stake.  They have been adding to the toolbox of toxicology but they have not been exchanging very much, because who dares to take away a safety test and then be responsible for the first scandal to come?  So all of the new stuff, if it was added at all, would be added as nice additional information.  
But it is time to rethink and the agencies become more and more aware that the system is not delivering, the drug agencies are not happy with less and less drugs being approved, the chemical agencies see that the throughput of the system has allowed us only to test small percentages of all of the substances which come to the market and which are on the market already.  So we really need a different type of tool, especially in the US we are seeing at the moment strong investment by FDA, by EPA, by the National Institutes of Health and its different sub-agencies to revamp regulatory toxicology.

DYANI LEWIS 
So to rely more on these computational methods and those types of pieces of information?

THOMAS HARTUNG 
Computational cell systems for high throughput testing and not necessarily yet for taking the ultimate decision on whether a substance can make it to the market or not, but to prioritise our testing.  You have to imagine, we have about 100,000 substances in our products, only about eight per cent of them have any type of test, only three per cent of them, mostly pesticides and drugs, have been intensively tested.  So most of the things you daily use have no information and the agencies would love to know which of these 100,000 they should request information on.  They can't ask it for all, it gets simply too costly.  
So the prioritisation is the first step, but there's also another interesting movement, both in the drug industry and the chemical industry at the moment, which is to use such type of test systems, cheap cell tests, fast computer-based models, which are not perfect, but to use them in a very early stage of development.  So let's talk to the chemist to develop substances which have less of a burden of toxicological liability later, because toxicologists know what smells like a problem.  Certain structural features are something you would avoid.  In the early stage you very often can choose amongst many, many substances and you simply develop those further or synthesise them at all, which would less likely fail in a late stage when it's costly and when the companies want to avoid this.

DYANI LEWIS 
Taking safety into account earlier on rather than just going for properties that will be good for the product? 

THOMAS HARTUNG 
Yes, safe chemical design, we call this green toxicology.  This is a very important movement where we have teamed up with EPA.  Green toxicology is a real new way of approaching the problem and simply taking care that you don't have to test as extensively.

DYANI LEWIS 
We live in a very globalised marketplace where products are developed in one place and manufactured in another and then sold in a third place.  Does this introduce duplications in the regulatory process if each country requires safety testing of their own?

THOMAS HARTUNG 
To some extent yes, but actually we have overcome this problem to a large extent.  The most important thing here was the process by OECD, the Organisation for Economic Collaboration and Development, who run a chemical safety testing program.  This has two components.  We on the one hand have a set of tests which are standardised by OECD and which are used in the different countries in the world.  It does not say that let's say a given test has to be used in Japan or in Australia, but it says if you use it, refer to this guidance and then it can be done everywhere in the world and there's not a tiny difference which needs duplication when redoing it.  
And the second component to it is good laboratory practice, so a quality scheme where countries also control each other and control the controllers so that the tests are done in a standardised way in each country.  Together this leads to what we call MAD, the Mutual Acceptance of Data, that in the end a product produced in Germany and tested in Germany can be brought to Australia and comes with test data which doesn't require any replication.  This was probably the biggest saver of animals ever, because before we really had small differences, a four-week study here and an eight-week study there, different concentrations to be used, different stems or threads, or whatever leading to duplication.

DYANI LEWIS 
Does this also mean that there are no longer, I guess, safe havens for companies who might want to be a bit unscrupulous in their practices?  If these are global regulations is there any way for companies to get around the regulations?


THOMAS HARTUNG 
You can always cheat on the system, but this is not my experience; companies don't want to poison their customers.  They have some tests to do anyway, they like to spend their money on the right stuff.  Global companies, as we are seeing this now and global markets actually require harmonised regulations.  It is actually sometimes the other way around, that countries which are not adopting because they are, for example, not part of the OECD process or the equivalent for drugs of the International Conference on Harmonisation, that these counties are obstructing the introduction of new technologies.  So if you are owning a company, or deciding on leading a company, you will use the traditional test as long as there is still an important market which has not converted to the new one.  So countries like Brazil, Russia, China, which are getting more and more important as a customer market, they can be obstacles for introducing alternative methods if they rely still on the old schemes, because they're not part of the discussions.

DYANI LEWIS 
And what role does the global consumer have then in changing those practices?

THOMAS HARTUNG 
There is enormous pressure by consumers and the consumer expectations are more powerful than we often think.  Acceptance of animal experimentation, for example, is a very, very strong driving force for policymakers.  We see in all countries over the world that there is a decrease in acceptance, especially for vanity products like cosmetics, but there is also the new role of retailers.  We sometimes say that the new regulator is Walmart, meaning that companies of this buying power and deciding on what's put on the shelf, what is not, sometimes react faster and more vigorously to perceived health threats and simply don't buy certain products, or label them as let's say Bisphenol A-free.  And by doing so, they force the industries to react more than or faster than any regulation could do.

DYANI LEWIS 
So if we are to reduce our reliance on animal testing, what are the main things, the key first steps that we really need to take?

THOMAS HARTUNG 
I think the first step is really to understand the limitations of the current paradigm, because we pretend too much.  We believe and overestimate the value of some of these animal tests, because we have forgotten about the limitations having done them for many years, and Petr Skrabanek and James McCormick very nicely said that learning from experience might be nothing more than making the same mistakes with increasing confidence.  So we have introduced these assays sometimes decades ago and we're using them and we believe in them, but we don't have a scientific discussion on their merits and their limitations, because far too many processes are positive feed forward.  If I try to write a scientific paper, I will praise the model I've been doing, I will not criticise it.  If I want to get money for funding, I will just do the same.  It needs a lot of frustration before we start to turn around our tools and criticise them.  And that's a fundamental problem we have.  
Secondly, we see that a lot of the most recent developments in biotechnology, the omics technologies which allow us biological phenotyping with data-rich, information-rich type of approaches, combined with bioinformatics tools which make sense out of this data, which help us to model, which help us to move into systems biology approaches, understanding how the system really works.  That these models necessarily are based on mechanistic type of information we can't get out of animals, we only can get from systems where we can make lots of measurements which we do understand because it's sufficiently simple.  
There's also an exciting development taking place to move cell culture into a second generation.  The first one was really this just making cells survive and grow.  They look quite odd, they look like pan-fried eggs sunny-side up, with the nucleus being the egg yolk, and these cell cultures have a density which is less than one per cent of normal tissue, they have cell contacts which are less than one per cent of the normal tissues, and we have chosen conditions which are completely artificial.  By making them all more organotypic, by producing really small organs in vitro, we have the impression that we are getting to results which are much more relevant.  So to improve the quality of the alternative while opening the door by showing the limitations of the existing system is a big step.  Last, it is all about mechanism, about understanding what I'm looking for.  The more targeted I can be in my research and can use tailored systems, if I'm looking for unknown unknowns then perhaps a living organism is the best to check.

DYANI LEWIS 
Thomas Hartung, thank you very much for being with us on Up Close today.

THOMAS HARTUNG 
My pleasure. 

DYANI LEWIS 
Professor Thomas Hartung is director of the Centre for Alternatives to Animal Testing at the Johns Hopkins Bloomberg School of Public Health in Maryland.  If you'd like more information or a transcript of this episode, head to the Up Close website.  Up Close is a production of the University of Melbourne, Australia, created by Eric van Bemmel and Kelvin Param.  This episode was recorded on 17 February 2014.  Producers were Eric van Bemmel, Kelvin Param and myself, Dr Dyani Lewis.  Audio engineering by Gavin Nebauer.  Until next time, goodbye.  

VOICEOVER
You've been listening to Up Close.  We're also on Twitter and Facebook.  For more information visit upclose.unimelb.edu.au, copyright 2014, the University of Melbourne.


show transcript | print transcript | download pdf