Thank for the post Erich. In relation to Anisha's point' If we didn't use animal models even fewer drugs would succeed in human trials', those who advocate for human-relevant drug development would argue that using non-animals methods reduces the time and expense and increase safety and effectiveness. More drugs, not fewer, would success in human trials because irrelevant or conflicting animal models would be eliminated. Additionally, we don't know how many drugs effective in humans have not been approved because they were ineffective or unsafe in animals.
A couple of developments since the original post.
The US drug regulator no longer requires animal testing as a mandatory requirement: https://en.protothema.gr/drugs-can-now-be-tested-on-humans-for-fda-approval-without-animal-trials-first/
In a landmark study published in Communications Medicine, researchers measured how effective the Emulate human Liver-Chip is at predicting human response to toxic drugs compared to animal and hepatic spheroid models. The results were staggering: the Emulate human Liver-Chip was able to correctly identify 87% of drugs that caused drug-induced liver injury to patients despite passing through animal testing. https://go.emulatebio.com/nature-publication-emulate-liver-chip-predictive-toxicology
Considering how much funding is allocated to medical research, I would like to see far greater scrutiny as to the research method used.