Am I right in my understanding: that the reason things are easily cured in mice is that mice have "throw away" disposable soma metabolisms that don't do anywhere near as much error correction as human ones, and so there's a lot of low hanging fruit. Something badass enough to harm humans, on the other hand, has already dodged many error correction mechanisms and is intrinsically harder to fight.
To be frank, I don't know. My work with mice was rather limited so I don't have the depth of knowledge of mice as a model organism that I would need to answer this good question.
In general, toxicities are identified in humans that weren't observed in mice, the delivery mechanism may be intrinsically unsafe or unknown to be safe, the genetic architecture differs in a way that changes expression of relevant genes, or the metabolism differs in some way.
