The New Statesman contains an op-ed piece by gay activist, human rights champion and animal rights advocate Peter Tatchell entitled: Why animal research is bad science. It's a fairly standard AR piece, and I encourage you to read it - judge for yourselves the extent to which I accurately summarize his main arguments and style.
My bottom line - I believe Mr. Tatchell is either blinded to facts and logic by his ideology, or he is deliberately obtuse and misleading. You can decide for yourselves who of the two of us is closer to the mark.
The main thrust of Tatchell's argument is this: testing drugs on animals can (and has) produced both "false positives" and "false negatives" with respect to their effects on humans. Further, because animals are unsuited to predict drug effects, they are therefore unsuited for all other scientific endeavors and animal studies should be abolished forthwith.
In the case of using animals to test drugs, a "false positive" would be a case in which the drug produced no adverse affects during animal tests, but did produce adverse affects in humans (safe in animals, unsafe in humans).
A "false negative" would be a case in which a drug produced dangerous side effects during animal tests, but did not when used by humans (unsafe in animals, safe in humans).
The concept of false positives and false negatives is a fact of life for scientists and physicians - the terms are used all the time in science and medicine (one of the main goals of science and medicine is to correctly distinguish between true and false positives, and true and false negatives).
Still, Tatchell makes much of the existence false negatives and positives, and he uses several examples of things gone awry in his efforts to discredit the use of animals in science altogether.
Thus, he claims that animals are just too different from humans to give reliable results - pointing out that strychnine kills people but not monkeys, and belladonna kills humans but not rabbits. He says two drugs (Opren and Eraldin) passed animal tests but proved unsafe in humans. And, of course, he is disturbed that the release of protease inhibitors - known now to be effective in the treatment of HIV/AIDS - was delayed for several years when research stopped because they were proving unsafe in dogs and rats (the effects being so severe that the developers thought it would be harmful to humans and unethical to release their product for human trials).
Has Mr. Tatchell proven his case, viz., that the response of humans and animals to the same drugs is so different that animal use in biomedical research is worthless - or worse, actually dangerous to human health - and should be abandoned altogether?
Absolutely not. Consider this.
Tatchell, like many AR advocates, cites numbers of people killed and made ill by drugs that were believed safe because animal tests failed to reveal their side effects: in the case of Opren (an arthritis drug) he points to 76 deaths and 3,500 made "seriously ill" (this would be a "false positive" - animal studies say it's safe, but it turns out not to be safe for everyone).
It doesn't take a rocket scientist to see Mr. Tatchell's logical faux pas: in the real world, where there is an objective reality, "safe" is a relative term. A drug can be safe and efficacious for a great many people, but a relatively small subpopulation (a small percentage ... ) may suffer adverse side effects, some very severe. No drug reaching the market can be guaranteed to be completely without side effects (e.g. no drug is perfect - drugs are "of this world" and things that are "of this world" are imperfect).
By focusing on absolute safety, Mr. Tachell has engaged in a little slight-of-hand - he condemns the "very good" of the real world, the world we must live in, the world where there is an objective biological reality, with the "perfectness" of a fantasy world, a world that does not exist other than in the wishful minds of some very passionate and romantic human beings. (In passing, using a utopian "perfect" to bash a real-world "very good" is a classic AR tactic.)
If Mr. Tatchell - and his fellow AR extremists - seeks perfection, he should look to the spiritual world where it may exist, not to the material world of man where it most certainly does not.
Still, what to make of the numbers? They are troubling ... are they not?
To understand them, you need to understand 3 things: 1) how drugs reach the market; 2) the nature of clinical trials; 3) (gulp ...) some elementary statistics.
1. Drugs are not introduced en masse to the market exclusively on the basis of having been proved safe in animals. They are introduced to the market gradually, over several years, as they pass through 3 phases of clinical trials. In other words, in spite of the fact that a given drug was used safely in animal tests, it is assumed unsafe for widespread dissemination until it has successfully passed clinical trials.
Only after a drug has passed the clinical trials, and it's safe dosage, side effects and efficacy have been established to the satisfaction of a governmental authority (yet another couple of years), is it released for general use. And then, it is still monitored (post-release monitoring is sometimes considered the 4th phase of clinical trials).
2. You can read about clinical trials here (more complete), and here (nice self explanatory graphic). Bear in mind that clinical trials are carefully designed and the patients very carefully monitored (this means objective as well as subjective evaluation - blood tests, urine tests, rash, cardiac output, blood pressure, etc., as well as subjective patient complaints of nausea, shortness of breath, taste abnormalities, fatigue, etc.). Here's the system:
Preclincial trials - animal trials - of 3 - 4 years duration, perhaps several hundred animals used.
Phase I: 20 - 80 volunteers, to determine safety and dosage - of 1 year duration
Phase II: 100 - 300 volunteers, to evaluate effectiveness and identify obvious side effects - of 2 years duration
Phase III: 1000 - 3000 volunteers, to further characterize effectiveness, and identify and monitor adverse effects - of 3 years duration
If the drug passes these three phases, and it satisfies a governmental review - 2 years - it is released.
Phase IV: Post release monitoring of the entire patient population receiving he drug.
3. And now ... numbers. They rule.
Typically, drugs that reach the market will have been tested on fewer than a combined total of 4500 animals and human volunteers (think of this number as being a rule-of-thumb cost/benefit tipping point). So - for a drug that has an actual adverse side effect in only 1:100,000 cases , you will be unlikely to identify it during the preclinical and clinical trials, even if the response of your animal and human cohorts perfectly mirrors the general population. In novel drugs (many are not) it can be really hard to predict the frequency of side effects, or what they might be, until the drug has been on the market.
Once a drug is released, and if it is prescribed for a rare condition, it may take years before you see the adverse reaction (it's the numbers - how long will it take to find that 1 person in 100,000 who will have an adverse reaction? And then, how many people over what length of time in what size patient population should react badly, and how badly, before you issue warnings, restrict use of the drug or withdraw it?).
On the other hand, if the drug is intended for a fairly common disease - like rheumatoid arthritis (as in Opren) - and it is widely prescribed to large numbers of patients (say 3,000,000 over 3 years), and the incidence of adverse side effects is 1:100,000, the odds are you're going to see any unsuspected adverse reaction sooner rather than later.
There really isn't a way around the numbers. Drugs can appear perfectly fine throughout the clinical trials, only to show their darker side when a very large population is exposed to them. But that's one cost of modern medicine: the drug may benefit hundreds of thousands or even millions of people, but it may also make some people sick or even kill some folks. In the real world, it's impossible for this not to be the case.
My challenge to Mr. Tatchell and like thinkers is this: find a way to beat the numbers. If you're going to demand every drug be completely without hazardous side effects, you'll have to withdraw a great many that are already on the market, and that are benefitting millions of people. If you do not demand perfection, what level of imperfection would you find acceptable, and how would you identify the most lethal side effects without using animals?
Now, admittedly, Mr. Tatchell is most concerned with drugs intended for HIV/AIDS patients, and these patients may have been something of a special case owing to matters of urgency (without a more effective treatment than then available, many were doomed). In this and similar other special cases, it might be justifiable to rush some drugs to market with abbreviated trials and expedited government review - throw caution to the winds and increase the risk of adverse - perhaps lethal - side effects.
But how should other non-HIV/AIDS drugs be brought to market? Should we extrapolate directly from the petri dish to humans, and increase the number of humans in clinical trials?
It is disingenuous of Mr. Tatchell to ignore the fact that the relatively unsafe drugs that entered the market went through exactly the same pre-clinical and clinical trials as drugs that proved safe. In other words, it's not as though removing animals from the equation would have made any difference whatsoever: the drugs - those with acceptable side effects and those with unacceptable side effects - were tested on humans!
So again there is the question: what would Mr. Tatchell do differently, how would he propose to beat that pesky "false positive" number problem?
Mr. Tatchell is incensed that false negative results - the adverse reaction of protease inhibitors in dogs and rats - slowed the PI's development because they were deemed (incorrectly) unsafe for humans. As a result, some HIV/AIDS patients were deprived of a potentially life-saving treatment, which Mr. Tatchell views as scandalous.
If I'd been faced with the facts that Mr. Tatchell presented in his article, I'd have made the same decision as the drug guys: let's slow down for awhile and get our bearings. Let's figure out what the problem is, and find a way to work around it rather than release a drug that showed every indication of being dangerous to every human who took it.
What is the alternative? Apparently, Mr. Tatchell would have us ignore the animal results, on the theory that animals and humans are too different for the animal results to be valid.
But if we're to follow Mr. Tatchell's logic, we should go back and test every agent that proved dangerous or lethal to animals on human beings!
Is that really what Mr. Tatchell is advocating? Is he prepared to accept full responsibility - legal, moral and financial - for doing so, or would he prefer for drug company executives and scientists to do that?
The fact of the matter is that animals are a very good, though imperfect, indicator of how drugs will affect humans. If you doubt that, consider that all drugs that reach market and are proven safe in humans were found to be safe in animals. And agents that prove unsafe for humans far more often than not will prove unsafe for animals, passionate assertions to the contrary notwithstanding.
To ignore these facts is to ignore a lot of "objective-reality" biological history in favor of a faith-based ideology.
As an aside, keep in mind that work on protease inhibitors didn't stop altogether with the negative dog and rat tests - it just slowed until further investigatons suggested that the cost/benefit risk was tilting towards benefit, rather than cost. Unlike ideology, science is self-correcting, and this is an example of science doing just that.
The introduction of protease inhibitors may not have happened quickly enough for Mr. Tachell, but it is, again, disingenuous of him to suggest that the process of bringing drugs to market is a failure because it did not work according to the timetable of his dreams. Once again, Mr. Tatchell has made the "perfect" of his fantasy world the enemy of the real world's "very good."
Finally, Mr. Tatchell focusses on drug testing as if it is the only scientific endeavor that uses animals: he argues that since the response of animals to drugs can't predict drug effects in humans, the use of animals in science should be abolished altogether.
Mr. Tatchell should know that animals are used for far more than just testing drugs: they are used for understanding basic biological mechanisms. And understanding these mechanisms in health is the basis for understanding them in disease, and understanding them in disease is the basis for developing pharmacological agents to treat them. Again - this isn't rocket science ...
If Mr. Tatchell were familiar with the biology of the objective world, he'd know that in countless respects, animal systems are virtually indistinguishable from human systems (how mammalian hearts work, nerves conduct impulses, muscle contracts, bone grows, gas exchange occurs in the lungs, blood carries oxygen, kidneys operate, what the immune response is, how the nervous and immune systems interact, hormones act, how pancreases create and release digestive juices, how eyes see, noses smell, fetal development happens, wounds heal, etc. etc. etc.).
Of course, if Mr. Tatchell really does understand biology, he has willfully chosen to ignore the facts and logic of an objective biological reality in preference to a fantasy world.
There is one final point: regardless of "why" Mr. Tatchell doesn't accept biological reality, it is nevertheless true that even if you accept everything he says about animal tests being useless for predicting the actions of drugs on humans (he is in fact wrong), that says nothing about the usefulness of animals in other areas of science. In other words, his logic fails: his conclusion cannot be supported by his data, even if his data were not false.
You make the call: who is closer to real-world truth: is Mr. Tatchell, or am I?
UPDATE: Edited for conciseness and clarity 8/7/04