Previous: Accomplishing the Impossible Task of Taking Animals out of Meat
Next: No, Phones Aren't Giving Kids Horns, Seriously



View count:74,404
Last sync:2022-11-26 04:00
Testing new treatments in other animals can help us spot complications or potential pitfalls, but the results don’t always carry over to humans, which means that safely going from animal to human trials is a lot more complicated than you might think.

Hosted by: Olivia Gordon
Camera Operator: Hiroka Matsushima

SciShow has a spinoff podcast! It's called SciShow Tangents. Check it out at
Support SciShow by becoming a patron on Patreon:
Huge thanks go to the following Patreon supporters for helping us keep SciShow free for everyone forever:

Adam Brainard, Greg, Alex Hackman, Sam Lutfi, D.A. Noe, الخليفي سلطان, Piya Shedden, KatieMarie Magnone, Scott Satovsky Jr, Charles Southerland, Patrick D. Ashmore, charles george, Kevin Bealer, Chris Peters
Looking for SciShow elsewhere on the internet?


When we talk about cool new research here on SciShow, you might hear us say something like: “This study was done on rats. And rats aren't people.” It's nothing against them personally.

Animal studies allow scientists to test new treatments in living, breathing organisms before they try them out in people, which helps spot complications or potential pitfalls. But animals aren't mini humans, so what happens in a mouse may not happen the same way in a person. And that means two things.

For one, it means we can't always assume a cure in mice will be effective in human patients. But it also means that safely going from animal to human trials is a lot more complicated than you might think. Before any animals even get involved, pharmaceutical researchers run a bunch of what are called in vitro tests, basically anything done in dishes and tubes rather than animals.

What they're doing has to have a lot of potential since, as you can imagine, the use of animals in research can be unsettling for a lot of people. Many countries have formal institutions in place which try to limit the use of animals for cases where they're really needed, and they have regulations which specifically ensure these animals are treated in a humane way. Now, a lot of medical research is conducted in rodents.

That's because rodents are small, easy to breed, and relatively cheap to care for in large numbers. And rodents have enough similarity to humans in a lot of areas, including genetics, behavior, and biology, that they can help scientists suss out whether a drug has real potential. Also, we can breed them to be optimized for research.

Large numbers of rats can have almost identical genomes, for example, allowing for more uniform and reliable results. And scientists know how to manipulate their genomes to get what we call knockout specimens: ones that lack specific genes. This lets them dig deeper into what a treatment is actually doing to the body, which helps minimize any surprises that might happen when a drug is given to humans instead.

You might think it's strange that so many trials are done in mice or rats when we'd get the most reliable information from animals that are closer relatives, especially other primates. But while primates are the right choice in some cases, there are ethical and logistical concerns with conducting research on them, and for the most part, rodents are pretty good substitutes. Besides, testing a drug in primates doesn't guarantee a smooth transition to human trials, because the only animals with all the right cells and proteins to perfectly predict what happens in a human body are, well, humans.

That's why, even when a drug has passed animal tests with flying colors, researchers build in a number of safety measures to minimize the risk of adverse reactions in people. It's considered important to use as little of the drug as possible, for example. That usually means starting with the smallest amount that seemed to work in animals, called the minimum anticipated biological effect level, or MABEL.

If that doesn't cause any problems, but doesn't work, either, then researchers can dial up the dosage. They also lay out exactly what their definition of “working” is ahead of time. These are what are called primary and secondary outcomes.

If they're too lofty or unfocused, like if there are dozens of things the drug could do, then you're likely to perceive a benefit that isn't really there. This is what's called multiplicity in a trial, and it arises because the statistics used to determine whether something has made an impact or not have a certain margin of error. On the flip side, if your trial outcomes are too narrow or impossible to obtain, you might not realize a drug is actually doing something good.

And at the same time as they're determining primary and secondary outcomes, researchers also set guidelines for adverse event outcomes, basically, what reactions mean it's time to call it quits, and what reactions, though perhaps unpleasant to experience, aren't really that big a deal. The good news is that they already have an idea what might happen from those pre-clinical animal experiments. But they also have to make plans for things that come out of left field.

So, they make guidelines to determine how to react if things go wrong, which take into account everything from the number of participants affected to the severity of their reactions. Of course, it doesn't help matters if companies are overzealous about their hopeful cures. In many cases, when a failed clinical trial makes headlines, investigators later discover corners that were cut along the way.

But sometimes, researchers do everything right and the worst still happens. Take the 2006 trial for TeGenero, an antibody that researchers hoped would be the next cure for a set of autoimmune diseases. For the first human test, 6 of the otherwise healthy participants were given 1/500th the dose used during primate trials, but within an hour, all experienced severe inflammatory reactions requiring hospitalization.

The drug was supposed to activate certain cells in the immune system, so in a sense, it was doing its job, it just did it a little too well. The human cells the drug targeted turned out to be way more sensitive than any of the animals tried. And that kind of uniquely human reaction to a drug is really hard to predict.

Nowadays, trials build in a lot more safety measures than they did a couple decades ago, like taking more time between dosing each participant to see if something bad happens. And they can take advantage of newer technologies like computer models that can better predict how human proteins interact with potential drugs. But researchers and regulators are still working on ways to make the clinical testing process even safer.

And a big part of that is making better protocols for working with animals. That's because, despite all the amazing things we can do with computers, bodies are really complicated. So we still need animal testing to see how things work in real living things.

So even though rats aren't people, we still need their help to keep people safe. Thanks for watching this episode of SciShow! If you enjoyed learning more about the importance of animals in research, you might like our episode about 5 times researchers gave animals drugs and what we learned from those experiments. [♩OUTRO].