I've been saying this for years. Most of the tox studies or efficacy studies in mice have poor or no models that correlate with human biology. They are indicative at best but should never be used to give a black or white story. The pharma industry and academia is throwing every single year dozens of potentially active molecules because they see toxicity in animals. Note that animals models have also permeated the regulatory bodies, therefore making it into de facto requirements to develop drugs. And it is so rarely ever contested on scientific grounds, where the Science part is effectively very thin.