Dr. Ray Greek: Toxicity Prediction


Jeremy P. Jacobs of Greenwire published the article Agencies Hope Robot Can Speed Toxics Evaluations, End Animal Testing on May 13, 2011. The article is about, “ ‘Tox21,’ a robot capable of assessing the toxicity of a chemical at mind-blowing speed.” The article states that there are 80,000 chemicals in consumer products like food and cleaners and so forth the risks of which are largely unknown. Jacobs:

"The testing of environmental chemical for toxicology was expensive, inefficient, not terribly predictive of toxicity and slow," said Christopher Austin, director of the National Institutes of Health Chemical Genomics Center. "Because of those things, the vast majority of chemicals in the environment have no data." While U.S. EPA must test one chemical at a time and can complete only a couple dozen assessments a year, Tox21 is capable of screening thousands of chemicals multiple times in a week, its creators say. "It is going to completely revolutionize the way chemical testing is done," Austin said . . . It could also lead to the end of animal testing.

This is very interesting news and I sincerely hope that it replaces the use of animals in testing, however, there are some facts and caveats that some news organizations and animal protection organizations seem to be missing when reporting on this story.

The technology of interest is here is both the robot—that is the part making the news—and the microtesting that the robot is actually performing. We wrote extensively about this in What Will We Do If We Don't Experiment On Animals? Medical Research for the Twenty-first Century. The actual testing will take place on human and rodent cells. Which leads us to our first problem. Jacobs:

But researchers caution that a chemical reacting with a liver cell in tests does not necessarily mean that it would have the same effect when a person is exposed to that chemical.

If human liver cells in vitro cannot predict how a liver cell in an intact human will respond, what do you think the chances are for rodent cells accomplishing this?

The knowhow to predict human response to drugs does not currently exist. Animal tests do not predict human response but neither do in vitro and or in silico tests. Everyone in this discussion wants to ignore this. Animal protection groups want to ignore it and focus on the apparent “progress” being made as the “progress” makes it look like they are doing their job and getting animals out of labs. They are not.

The testing, regulatory, and pharmaceutical industries want to ignore the lack of predictive ability in general because that would mean they would have to admit that:

1. the first humans in clinical trials are very vulnerable since animal studies and other technologies cannot predict what is going to happen to them;
2. humans that take the drugs after approval are vulnerable as the human testing phase is inadequate (adequate human testing would cost much more money);
3. if animals cannot predict human response to drugs then using them in other endeavors such as to predict human response to disease is going to be scientifically very suspect thus basic researchers are not going to be happy with the reporting of this news;
4. the regulatory bodies that have been assuring people that medications are more or less safe have been lying to them all these years.

    Let me state very strongly: I am pro-medication! I pro science-based medicine aka Western Medicine aka traditional medicine and am even pro-pharma. But! I am also in favor of honesty—both scientific and the plain old-fashioned kind—and proper government regulation and these are parts of the equation that have been missing for decades.

    Another problem with the reporting on concepts like Tox-21 is the notion that we are now testing for “toxicity.” Toxicity for whom or for what? What is toxic for me may not be toxic for you and what is toxic for a specific species of plant may not be toxic for another species much less a mammal or a human. There is no such thing a test for toxicity. ALL chemical are toxic at a given dose. Even water and oxygen. Paracelsus stated in the 16th century that all things were poisons and that the dose determines the poison. It is disingenuous to report that a new system, whatever it is, will predict whether a chemical will harm the “environment.” In all likelihood some chemicals will be toxic to some plants but not humans and others will kill a few humans but not others and so on. The only way society will ever know the toxicity of a chemical is to test it on the genomes of all the species (plant, animal, fungus, bacteria and so forth) that will be exposed to it. Even the possibility for doing that is a century away.


    "We want to make better predictions of human toxicity and we also want to use fewer animals in our clinical studies," said David Jacobson-Kram, FDA's associate director for pharmacology and toxicology. "That's the ultimate goal here from our point of view."

    Don’t get me wrong. Tox-21 is a great project! It is an example of what government and industry can do when they work together and someday it might be one part that contributes to an end to animal testing. But animal testing for human toxicity should end now, regardless of Tox-21 and other projects, because it simply is ineffective.

    The spin about all this from the animal protections groups is nauseating. I see absolutely no reason for their claims of “progress for animals” except as a means to pad their own pocketbooks. Claiming, directly or indirectly, that society must continue to use animals in drug testing because it works serves only to prolong the suffering of animals and humans and to ensure job security for themselves. I strongly suggest that anyone considering contributing money to a group that sells itself as an animal protection organization carefully read what they have to say on topics like this one. There are many examples of groups that want to sell themselves to society as animal advocates that make misleading statements.

    AltTox.org “is a website dedicated to advancing non-animal methods of toxicity testing, both to better protect the health of humans, animals, and the environment and to reduce the numbers and suffering of animals used in current toxicology assessments.”

    AltTox.org was originally supported by HSUS and Procter & Gamble Company. In their April 2011 E-newsletter they address the need for an "elevator speech" to give to people in order to highlight the problems of animal testing and the need to replace animals in testing. The “speech” states in part:

    Chemicals are the basic ingredients in a wide variety of important consumer and industrial products, including soaps, detergents, drugs, pesticides, and fuels. The standard approach to assessing the potential health effects of chemicals on people has changed little over the past fifty-plus years, despite extraordinary advances in science and technology. This process involves exposing animals to high doses of chemicals in attempts to elicit any harmful effects (e.g., tumor growth). This approach is expensive and time-consuming, doesn't always predict what happens in people, and is controversial for using large numbers of animals. (Emphasis added.)

    Saying that animal testing doesn't always predict what happens in people is like saying that fortunetellers don’t always get it right. In reality, predicting the future is impossible and the positive predictive value and negative predictive values of animal testing is so low as to be better equated with fortunetelling than science. (See Are animal models predictive for humans? for more)

    Before you contribute to or endorse any charity or advocacy organization, find out what they really believe and how they really act.


    Popular Video