Comparison of Various Approaches to Translate Non-Linear Pharmacokinetics of Monoclonal Antibodies from Cynomolgus Monkey to Human

The prediction of pharmacokinetics of monoclonal antibodies (mAbs) exhibiting non-linear pharmacokinetics in preclinical species to human is challenging, and very limited scientific work has been published in this field of research. Therefore, we have conducted an elaborate comparative assessment to determine the most reliable preclinical to clinical scaling strategy for mAbs with non-linear pharmacokinetics. We have compared three different scaling approaches to predict human pharmacokinetics from cynomolgus monkey. In the first approach, cynomolgus monkey pharmacokinetic parameters estimated using a two-compartment model with parallel linear and non-linear elimination were allometrically scaled to simulate human pharmacokinetics. In the second approach, allometric exponents were integrated with a minimal physiologically based pharmacokinetic (mPBPK) model to translate human pharmacokinetics. In the third approach, we have employed a species time-invariant method, wherein a two-compartment model with parallel linear and non-linear elimination was used as a framework model for simulation of the human profile. Human exposure parameters projected by an integrated allometric method were only within two fold for approximately 45–70% of predictions at different doses of five mAbs evaluated, while approximately 70–80% of Cmax and AUC predictions by integrated mPBPK modelling as well as the species time-invariant method were within two-fold error. The average fold error for clearance predictions by the integrated mPBPK method was 1.10–1.45 fold, whilst for the species time-variant and integrated allometric methods, the average fold error was between 1.04 and 1.37 fold and 1.24 and 2.13 fold, respectively. Our findings suggest that the species time-variant method and mPBPK proposed by us can be employed to reliably translate non-linear pharmacokinetics of mAbs from cynomolgus monkey to human.