> Although exciting, we also need to remember that translatability rates from animal models to humans is notoriously low, usually in the single digits for most therapeutic areas.
This is true of humans as well -- there are plenty of programs that get interesting and clear results in Phase 2 (the dose ranging phase: essentially "what dosage is most efficacious") that die in Phase 3 (roughly: "OK, how does that dose really work on a statistically significant population that also represents the demographics of the country"). These studies are so expensive that nobody goes into Phase 3 unless they believe Phase 2 pretty certainly demonstrated that the drug works, and well.
It's also hard to figure out just how the animal's hearing is improving (you can't simply ask). I'm sure they have some experiments, but growing the hairs back may be necessary but not sufficient. Look at all those Alzheimers programs aimed at removing the plaque that haven't demonstrated any value in the clinic. The plaque might not even be alzheimers itself -- it could merely be the body's response to some different underlying effect of the disease.
The choice of animal is very important, and the FDA cares a lot. Mice are popular because they are cheap. I worked on a program years ago that used guinea pigs because mice couldn't get the disease. We didn't use rats as the compound caused cancer in rats, and someone else had had their program derailed because the FDA required a separate analysis and study to demonstrate that the cancer was specific to rats and not other species (rats get lots of cancers). For our program the FDA required some studies in (non-guinea) pigs before they were willing to allow any human trials.