The strange case of smoking animals, tobacco companies and research - Part I

British Prime Minister Tony Blair recently said that he would sign a public petition in support of animal testing in research. And the ex-Beatle Sir Paul McCartney, almost as if in reply, claimed: “Animal Tests Are Unreliable and Cruel”.

The debate over animal research has never been as public and centre stage as it is now.

There are both medical and ethical reasons to ban animal testing. Most people are mainly interested in the medical aspect, so I’ll focus on that here.

First of all, it’s no good to say that animal research is essential because researchers say so. People in the biomedical field are part of animal research, have a vested interest in it, and therefore their views are biased in favour of it. So it’s important that the public, who after all have to bear the consequences of it (and in many cases pay for it through taxation as well as by buying its products), make up their own mind.

Animal research is unreliable in the extreme. Let’s analyze one particularly important and significant example, before giving general arguments to back it up.

Smoking and cancer

The link between smoking and lung cancer was discovered by the British scientist Sir Richard Doll in the 1950s, by means of a study of human lung cancer patients in twenty London hospitals. He used the epidemiological method, i.e. the statistical study of diseases in human populations. The practical application to medicine of this all-important discovery had been hindered and delayed by animal research. Why?

First, because animal experiments had previously failed to demonstrate this link and they had ruled it out. Medical researchers were dismissive of Doll’s discovery. They had already tried to trigger cancer in animals using tobacco tar but had failed, they pointed out. Only later was it shown that their experimental procedures contained serious errors.

And then, even after Doll’s theory was published, animal researchers tested it by trying to reproduce the carcinogenic effects of smoking in animals. Some of this animal research was funded (you may guess) by tobacco manufacturers. Look at what they said: "No clearcut case against cigarette smoking has ever been made despite millions spent on research ...The longer these tests go on, the better our case becomes." (Phillip Morris, 1968)



It took 50 years to induce lung cancer in laboratory animals forced to breathe tobacco smoke (remember those pictures of rows of smoking beagles with cigarettes fixed in their masks?), thus delaying the health warning to humans and resulting in millions more unnecessary deaths.

In the end, it was the so-called British doctors study, a prospective clinical trial which ran from 1951 to 2001, that provided convincing statistical proof that tobacco smoking increased the risk of lung cancer. Doll’s study has provided the foundation for all other research into the impact of smoking cigarettes on health. It has arguably helped to save millions of lives.

This was a particularly important discovery, because it shows the role of lifestyle in cancer prevention (it actually paved the way to other discoveries linking cancer and diet) and, as we know, prevention is better than cure. There is here an analogy between the animal research lobby and the tobacco manufacturer's lobby, an analogy that goes beyond the fact that in this case they formed an “unholy alliance”.

Some smoking experiments on animals, intended to disprove its link with lung cancer and to give smoking the okay, were financed by cigarette companies. The analogy goes beyond this “unholy alliance” because both lobbies (animal experimentation and tobacco manufacturers) put the interest in their own self-preservation above public health.

The strange case of smoking animals, tobacco companies and research - Part II

Why animal experiments are unreliable

At this point people might say that yes, this was a case in which animal research retarded science and misled medicine, but it is an isolated case, the odd exception. But it is not an exception: it is indeed the rule, for obvious and necessary reasons, which will become apparent.

Animal experimentation is by its own very nature unreliable for application to humans. One of the reasons is that minuscule differences in biochemistry and miscroscopic composition of organisms produce enormous effects, hence the huge differences existing between not only species, but also individuals.

It’s easy to see that it is so. Think of how an incredibly tiny alteration at the microscopic level in the DNA can give rise to so many dramatic changes between individuals. In fact, it can give rise to a new species if the DNA change is a mutation.

Moreover, the differences between species will be greater and more difficult to compensate for exactly in those areas which interest animal researchers. This is the case, for example, of metabolic differences between species, which are centrally important in toxicological (effects of drugs and chemicals) and teratological (effect on the fetus) investigations.

Researchers refer to animals as models. But models which are useful to inspire hypotheses are not necessarily good to test them, in fact they often are terribly and disastrously bad at that.

For instance, to see what this means in practice, think of the planetary model of the atom in physics, according to which in atoms the electrons orbit the nucleus the way planets orbit the sun. In the early stages of the atomic theory, when knowledge was limited, the solar system has indeed served as a useful tool, something known to help understand the unknown.

But no one would dream of testing on the solar system a hypothesis about electrons’ behaviour, for example, even if it were physically possible.

If you are inclined towards logic and formal arguments, read Two Models of Models in Biomedical Research. Otherwise, what was just said would suffice.

The fundamental mistake of animal research is to transfer an experimental technique (replication of phenomena) from physics to biology. The problem is that biological entities like animals, including humans, are extremely more complex than physical objects. This technique which is valid in physics is not valid in biology, where such regularities as “effect B always follows cause A” do not apply.

Biological beings behave in a probabilistic way, not in a deterministic one. This is a major factor of great uncertainty in predictions. Even when animal research results were later confirmed, we only know that they were valid because they were confirmed by their application to humans. So, the latter was, in the end, the final test of the former, not the other way around.

It’s a classical case of post hoc propter hoc fallacy (mistaking time sequence with causation), as Prof. Pietro Croce, MD, an eminent pathologist, puts it in his brilliant book Vivisection or Science - A Choice to Make, which anyone who would like to develop a serious position on the subject, fruit of thought rather than pre-conception, should read. You can also find a short version of it published online.

Here’s an analogy that helps picture the problem of unreliability. Suppose I like mushrooms but I cannot recognize the good ones from the poisonous ones. I go and pick them in the wild. Then I call a mushroom expert, and he says they’re okay. It does not mean that I did the right thing by choosing them myself, I could just as easily have picked up the poisonous mushrooms. And it was only after, not before, the mycologist gave his opinion that I knew if the mushrooms were edible or not.

So it is with animal research: it’s only after, not before, its results are applied to humans that we know whether they are valid (for us humans) or not. And believe you me, when they are not, some of us are going to pay for it with our health. That’s why so many drugs are quietly and surreptitiously withdrawn from the market after they have caused some disaster. Remember that a new drug being introduced will make headlines, but a drug being withdrawn will likely be ignored by the media.

The strange case of smoking animals, tobacco companies and research - Part III

Alternatives to animal experiments

There are many methods and techniques which could replace animal experimentation, and are more precise, cost-effective, and humane.
The main ones are:

• the epidemiological method, the study of human populations, which was used to discover the smoking-lung cancer link discussed above;
  
• in vitro techniques, ie cultures of cells and tissues on which to conduct tests - penicillin and streptomycin are historical examples of in vitro discoveries;
  
• clinical research, ie careful observations and analyses of patients;
  
• computer and mathematical modelling, a relatively new branch of medical research using complex software to simulate biochemical reactions by recreating our body components structurally and in terms of healthy and diseased chemical reactions, then submitting them to chemical and curative substances;
  
• genetic research, often used alongside epidemiological evidence;
  
• autopsies - practically every disease has either been discovered or clarified as a result of autopsy, which also indicates aspects of illness missed in diagnoses;
  
• post-marketing drug surveillance (PMDS), the reporting of effects and side effects of a medication after its release, which unfortunately is not required at present and only relies on voluntary and infrequent reporting. The current situation therefore makes it impossible to maintain comprehensive data on any drug’s potential for negative reactions.
  
• technology, for example ultrasound, blood-gas analysis machines, monitoring devices, DNA sequencing, gene chips, combinatorial and solid phase syntheses, bio-compatible materials, polymerase chain reaction, separation and purification methods, the Fast Fourier transforms used in spectroscopy and CAT scans, fast sequence alignment and database methods used in genomics, conformational search and optimization methods used in protein folding.
  

These methods would represent an improvement if they replaced animal research now, even considering the little money and time that have been spent on them in comparison to the gigantic resources invested in animal experiments. The main problem is not that there are no alternatives, but that there is no or little political will to make that choice. If funds and energy were devoted to these other methods, great progress could be made.

And, to go back to the fungal analogy above, in case you were thinking that I could have fed the mushrooms to some animal to test them, don’t. It is well known that a mushroom can be eaten by squirrels, rabbits, or other animals and still be dangerous for humans.

The Illinois Mycological Association, for example, says: “According to Dr. John Rippon, an IMA member and world expert on fungal diseases, squirrels have an interesting adaptation that allows them to eat mushrooms containing deadly amanita toxins without being affected.”