I heard it was because the species of rat they used tend to get cancers anyway when they got older, and shouldn't be used in any long-term study.
I love scientists, loved working with them, best clients i ever had (if you're a devops or automate dev/scientists tools, and have to regularly explain stuff to scientists and to other devs, we don't compare advantageously at all), but sometimes there is too much noise.