There are a ton of studies that show side effects in non-human animals are not a good predictor of side effects in humans; and this becomes more true with rarer side effects.
There is not good correlation of side effects between different animals and failures in phase 1 studies often show that safety data in animals did not predict side effects in humans that became apparent in very small human populations.
So, widespread use in animals can actually harm humans by bubbling up side effects that are not relevant to humans.
Among some safety people, there is even a question of whether there are drugs shelved for animal side effects that would be 100% safe in humans. But if a drug killed half your rats, you cannot ethically start dosing humans with it.
There is not good correlation of side effects between different animals and failures in phase 1 studies often show that safety data in animals did not predict side effects in humans that became apparent in very small human populations.
So, widespread use in animals can actually harm humans by bubbling up side effects that are not relevant to humans.
Among some safety people, there is even a question of whether there are drugs shelved for animal side effects that would be 100% safe in humans. But if a drug killed half your rats, you cannot ethically start dosing humans with it.