Does this translate to humans though? At the least it should be tested with something living longer than a few years.

Mice have a very short window to reproduce compared to humans so n optimisation to live longer in absence of sufficient protein makes more sense.

We’re optimising quite different I would assume.