Background and objective: The wide inter-patient variability in drug exposure partly explains the toxicity and efficacy profile of sunitinib treatment. In this prospective study cytochrome P450 (CYP) 3A and adenosine triphosphate binding cassette (ABC) B1 phenotypes were correlated to the pharmacokinetics of sunitinib and its active metabolite N-desethylsunitinib.
Methods: A correlation analysis was performed between sunitinib pharmacokinetics and 1'OH-midazolam/midazolam ratio and parameters derived from technetium-99m-2-methoxy isobutyl isonitrile ((99m)Tc-MIBI) scans, respectively. A population pharmacokinetic model using non-linear mixed-effects modeling software NONMEM was built, which included the phenotype tests as covariate.
Results: In 52 patients, the mean trough concentration of sunitinib plus metabolite increased from 21.4 ng/mL at day 1 of a cycle to 88.1 ng/mL in the fourth week of treatment. A trend for a correlation was observed between (99m)Tc-MIBI elimination constant and trough concentrations of N-desethylsunitinib; however, this was not significant. Correlations were found between 1'OH-midazolam/midazolam ratio and sunitinib clearance (P = 0.008) and day 1 N-desethylsunitinib trough concentrations (P = 0.005), respectively. Moreover, patients suffering from grade 3 toxicities had significant lower clearance of sunitinib than patients without grade 3 toxicities (34.4 vs. 41.4 L/h; P = 0.025).
Conclusions: Phenotype tests for ABCB1 and CYP3A4 did not explain inter-individual variability of sunitinib exposure sufficiently. However, the correlation between sunitinib clearance and the occurrence of severe toxicity suggests a direct exposure-toxicity relationship.