Interspecies Scaling of Therapeutic Monoclonal Antibodies: Initial Look

The authors evaluated interspecies scaling for the prediction of human clearance of 18 therapeutic monoclonal antibodies (mAbs). Human and monkey/chimpanzee data of 14 mAbs were classified based on the targeted antigens (soluble or membrane bound). Simple allometry and/or a time-invariant method (elementary Dedrick plot) were performed. Results indicate that human clearance might be accurately predicted from monkey data for mAbs targeting soluble receptors or membrane-bound receptors with limited tissue distribution using simplified allometry. The optimal exponents were estimated to be 0.85 or 0.90. If nonlinearity is anticipated at the human efficacious dose, pharmacokinetic parameters obtained at high doses in animals might not be sufficient for full pharmacokinetic characterization and prediction. Using prespecified criteria, including predicted human clearance (< or = or > 10 mL/d/kg), simplified allometric scaling might be helpful in predicting the effect of receptor-mediated clearance for mAbs targeting membrane-bound antigens. Furthermore, simplified allometry and an elementary Dedrick plot provide similar results in predicted clearance. Given the significant advantages offered by simplified allometry, it should be used when data are available from only 1 species. When reasonable data from > or =3 species are available, traditional allometry should be explored. Overall, clearance prediction is useful for human dose prediction in drug discovery and development.