Skip to main content

Research: First, Make Sure The Test Works

Elie Dolgin wrote in Nature Medicine, an article titled: “Putting stem cells to the test.” (Dolgin 2010) The article describes how scientists inject induced pluripotent stem (iPS) cells into mice in order to determine whether they are truly pluripotent. (A pluripotent stem cell can give rise to any cell in the body.) Induced pluripotent stem cells could be the breakthrough that allows physicians to treat currently incurable diseases with stem cells. Dolgin:

The researchers took several of the cell lines, injected millions of cells from each between the shoulder blades of mice engineered to lack immune systems and waited for telltale cancerous lumps to appear. Although the appearance of cancer is usually a bad sign, but in this test it signals that the cells might have therapeutic potential. After seven weeks, the scientists sampled the human cell–derived tumors and performed staining to show that the tumors contained a mixture of differentiated cell types such as neurons, heart muscle and skin cells (Somers, A. et al. Stem Cells 28, 1728–1740 (2010)) . . . This final test, known as the teratoma assay, is currently the only established means of showing that human stem cells possess pluripotency when placed in a living creature.

Dolgin continues:

However, many stem cell biologists worry that the current metric for characterizing these cells—the teratoma test—might be setting a low bar for defining pluripotency. “It's the most ridiculous assay on the planet,” says Owen Witte, director of the Broad Stem Cell Research Center at the University of California–Los Angeles. “It's time consuming, cumbersome, expensive and hard to connect the information you get to the actual clinical utility of the cells you might put into patients.” In fact, some researchers have found that partially reprogrammed iPS cells can form teratomas even if they fail other tests of pluripotency, leading many in the field to call for a new gold standard.

I am sure that the above assay was approved by many Institutional Animal Care and Use Committees (IACUCs). When asked: “Is there another way of getting the data without using animals?” the researchers checked the no box and that was all there was to it. I seriously doubt if anyone on the IACUC or anywhere else asked whether or not the assay actually gave information that the scientists needed and, if not, then why was it being used. No one asked whether the test worked.

When addressing claims of the extraordinary, skeptics first ask whether the people quoted are in fact actually making the claim. Then, they can assess the validity of the claim. But there is no sense in refuting claims that were never made. For example, if someone claims that a village in Kenya (a place I have visited and love) witnessed the Virgin Mary going about healing the villagers on the night of such and such, the first question one should ask is whether the villagers did in fact make that claim. Most such stories can be squelched with that question as, upon questioning the villagers, one usually finds out that such healings are news to them.

This same question should be asked of animal-based research and testing in general. Animal models cannot predict human response to drugs and disease (see Animal Models in Light of Evolutionfor more) therefore mandating predictive research and testing before dropping the animal tests is simply nonsensical. Its like saying: “Lets keep performing trephinations (or blood-lettings or laetrile treatments or homeopathy or mesmerism) for AIDS until we find a cure.” In no field, other than animal-based research, would such nonsense go unchallenged.

This leads me to a second report by Dolgin in Nature Medicine.

Titled “Animal testing alternatives come alive in US,” (Dolgin 2010) Dolgin discusses attempts in the US to find alternatives to using animals. (Shanks and I cover alternatives in FAQs About the Use of Animals in Science: A handbook for the scientifically perplexed and I reproduced part of that in Alternatives: Facts and Fallacies and addressed it briefly in one of my first blogs: The 9 Ways Animals Are Used in Science.) Dolgin reports that there is a new organization in the US named “American Society for Cellular and Computational Toxicology.” The organization (funded by other organizations) has, as its goal, the finding of alternatives to using animals, mainly in the field of toxicology. Dolgin:

“It's an important step,” remarks Thomas Hartung, director of the Center for Alternatives to Animal Testing at Johns Hopkins University in Baltimore and a member of the new society's board of directors. He points to a “long-lasting tradition in Europe of alternative methods” for testing compounds, adding that “in the US there was no equivalent” until now.

(Just FYI, the European organization responsible for finding alternatives, ECVAM, has in reality found very, very few. Last time I checked, they had developed about one test for every year they have been in existence (Abbott 2005) and most were not animal-free.)

I find organizations like the above fascinating. When you consider the fact that there are thousands of animal tests and ways animals are used in research, together with the fact that new ways to use animals in research and testing come about almost daily, thinking you can find alternatives for them, on a one to one basis, is lunacy. Combine this with the fact that most of the aforementioned tests simply do not work in the first place and the madness is multiplied.

The pharmaceutical industry is well aware of the lack of predictivity of animal models. They are a huge and wealthy industry and are devoting resources to fixing the problem. So, a few small animal welfare organizations come along with their very meager and inadequate resources and announce to the world that they are going to help fix the problem. Looks good to the naïve observer: “Finally, an organization is going to get animals out of labs by developing alternatives! We should send them some money to help out.” Good for fundraising.

I am not suggesting that the pharmaceutical industry is a friend to animals, merely that they are losing money because of the lack of predictivity of animal tests and therefore are very motivated to fix the problem. In addition, they have the resources to actually accomplish this and, they are truthful; they actually admit the animal tests do not work. They are not looking for alternatives they are looking for predictive tests. But then again, they do not have to fundraise through direct mail.

I have said many times before, if so-called animal advocates really want to help animals (as opposed to raising more money) then they should focus their efforts on forcing the government to come clean on the fact that animal tests are not predictive for humans. This would essentially end the animal testing requirements currently in place by the FDA and EPA.

Instead, these so-called animal protection organizations are in fact a hindrance to ending animal use (see How animal protection groups are delaying the end of vivisection) because they reinforce the false notion that animal testing works; that animal models can predict human response to drugs and disease and thus alternatives must be developed before anything can be done. (Although, I must admit such a position does ensure job security.)

Inept organizations seeking to find replacements for tests that do not work in the first place are on a fool’s errand and those who donate to these charities are fools.


Abbott, A. 2005. Animal testing: more than a cosmetic change. Nature 438 (7065):144-6.

Dolgin, E. 2010. Animal testing alternatives come alive in US. Nat Med 16 (12):1348.

———. 2010. Putting stem cells to the test. Nat Med 16 (12):1354-7.


Popular Video