You are correct there are indeed regulatory requirements to have animal testing, but that isn't the same thing as requiring animal testing to develop drugs. 95 percent of all drugs that are shown to be safe and effective in animal tests fail in human trials because they don’t work or are dangerous.
A survey of 4,451 experimental cancer drugs developed between 2003 and 2011 found that more than 93 percent failed after entering the first phase of human clinical trials, even though all had been tested successfully on animals.
Meanwhile many drugs that are widely used by people have failed animal trials. Penicillin kills guinea pigs. Aspirin kills cats and causes birth defects in rats, mice, guinea pigs, dogs, and monkeys. And morphine, a depressant in humans, stimulates goats, cats, and horses.
Despite this overwhelming failure rate of developing drugs for humans by testing them in non-humans, billions of dollars are spent on it, including 40 percent of all research funding from the US National Institutes of Health.
Its time for a change of thinking..
"The most significant trend in modern research is the recognition that animals rarely serve as good models for the human body. Human clinical and epidemiological studies, human tissue- and cell-based research methods, cadavers, sophisticated high-fidelity human-patient simulators, and computational models have the potential to be more reliable, more precise, less expensive, and more humane alternatives to experiments on animals. Advanced microchips that use real human cells and tissues to construct fully functioning postage stamp–size organs allow researchers to study diseases and also develop and test new drugs to treat them. Progressive scientists have used human brain cells to develop a model “microbrain,” which can be used to study tumors, as well as artificial skin and bone marrow. We can now test skin irritation using reconstructed human tissues (e.g., MatTek’s EpiDermTM), produce and test vaccines using human tissues, and perform pregnancy tests using blood samples instead of killing rabbits."
https://www.peta.org/issues/animals-used-for-experimentation/animal-testing-bad-science/
You have conflated two things - poor experimental design and analysis, and genuine biological differences between species, to reach a false conclusion. Most animal species respond the same way to most drugs. Most of the work that failed to translate was early stage work done badly on drugs that truly don't 'worl' in animal disease models or humans, but which were published owing to the general ignorance in the sector on what constitutes good desugn and analysis. The journal Nauture prioritises the punlication of work that would have massive impact if true. The problem is the l;ast bit and the journal has more retractions for false findings than any. And this is not just in drug research but in all research.
Most attempts on goal do not result in a goal. Therefore the best way to win more games would be to take fewer shots on goals. Right?