I can fathom the idea of PK being different b/w dogs and rodents, but why would I be seeing a significantly different PK profile for the same drug tested on rats and mice?
A rule of thumb is to assume that all animal species can (and do) have biological differences that affect drug ADME. At the risk of oversimplifying, these are based on differences in biochemistry and genetics (enzymes, for example) and "geometry" (size and weight, blood volume, organ sizes, basic metabolic rates, etc.). Even human subjects show significant variations in these factors.
I agree with Robert Bellantone. There are many factors that affect pharmacokinetic behaviour in animals. Even the pharmacokinetics of the same drug may differ in the same animals at different times or settings. In addition, I think mice and rats are not suitable animals for kinetic studies because of their small body weight and low blood volume for series blood sampling.
Cengiz Gokbulut - I think with modern algorithms you can overcome much of the problems of sparse sampling by applying population analysis techniques, which also better allow you to *quantify* the covariates that lead to the inter-individual differences. Whether they are continuous e.g. weight or categorical, e.g. sex, genotype, species etc. and they can then be built into the model to provide much more valuable simulations.
It is evident that the influence of species / gender/ strain / age / body weight of the animals will play a role in PK profile. This is mainly because of the difference between ADME profile (due to physiochemical property of the molecule, formulation, animal biology etc.) and contribution of metabolizing enzymes concentration and activity etc..
Collectively many factors contribute to the PK profile of a molecule or same molecule. Therefore conducting experiments in controlled conditions is essential to get reproducible results.
Further reading:
a) Sakai et al., J Drug Metab Toxicol 2014, 5:6
b) Maria et al. Pharmacokinetics of memantine in rats and mice. 2012 February 15. doi: 10.1371/currents.RRN1291