Pharmacokinetic and toxicity scaling of the antitumor agents amsacrine and CI-921, a new analogue, in mice, rats, rabbits, dogs, and humans

Abstract

The aim was to investigate interspecies relationships between body weight (W) (kg) and various pharmacokinetic parameters for the anti-tumor agents amsacrine and its 4-methyl-5-(N-methylcarboxamide) analogue, CI-921, and examine which pharmacokinetic parameter, if any, might be used to predict the toxicity of these agents. Pharmacokinetic, plasma protein binding, and toxicity data were available for CI-921 in mice, rats, rabbits, dogs, and humans. For amsacrine, similar interspecies pharmacokinetic data were available but toxicity and protein-binding data were available for only 3 species. Significant linear relationships were obtained for CI-921 between log W and log Vss (liters) (r = 0.971, P = 0.006), and log W and log Cl (liters/h) (r = 0.911, P = 0.031) resulting in the allometric equations Vss = 1.22W0.68 and Cl = 0.91W0.51. For amsacrine these corresponding equations were Vss = 3.37W0.81 (r = 0.996, P less than 0.001), and Cl = 2.28W0.46 (r = 0.952, P = 0.012). When interspecies differences in plasma protein binding were taken into account, the allometric relationships improved and the exponents of the power equations increased. For CI-921 the allometric equations for the kinetic parameters calculated from plasma "free" concentrations were: Vssfu (liters) = 247W0.93 (r = 0.984, P = 0.002) and Clu (liters/h) = 186W0.76 (r = 0.961, P = 0.009). The dog was a noticeable outlier in the relationship between the log maximum tolerated dose (MTD) (mg/kg) of CI-921 and log W. Omission of the latter resulted in a highly significant allometric relationship, MTD = 23.6W-0.14 (r = -0.988, P = 0.012). For amsacrine there was no significant allometric relationship between MTD and W. CI-921s prolonged t1/2 in the dog and the dog's increased susceptibility to CI-921 toxicity suggested a relationship between MTD and t1/2 (h). A significant linear relationship was observed between in MTD and t1/2 (r = -0.994, P less than 0.001), from which the following equation was developed MTD = 47.5e-0.51t1/2 Combining the amsacrine toxicity data in the latter relationship yielded a similar equation MTD = 44.7e-0.51t1/2 (r = -0.933, P less than 0.0001). It was concluded that allometric equations may be developed for CI-921 and amsacrine from animal pharmacokinetic data which allow a reasonable prediction of Cl and Vss in patients, despite these agents being eliminated mainly by biotransformation. However, similar relationships between toxicity and body weight were susceptible to variation between individual species. Species differences in the toxicity of these agents were predictable from the t1/2. This study emphasized the importance of pharmacokinetic data in preclinical toxicity and efficacy testing of antitumor agents.