1678P - Predictive Value of Two Phenotyping Probes for the Pharmacokinetics of Sunitinib in Cancer Patients

Abstract

ABSTRACT Background Cancer patients treated with sunitinib show a wide inter-patient variability in drug exposure. This may at least partly explain the variation in toxicity and efficacy related to sunitinib treatment. The primary aim of this study was to investigate whether the clearance of 2 phenotyping probes are related to the pharmacokinetics (PK) of sunitinib. Patients and methods A prospective multi-center study was performed in cancer patients treated with single agent sunitinib. PK of sunitinib and its active metabolite N-desethyl-sunitinib were measured at start of sunitinib intake and within 4 weeks of continuous daily dosing. A correlation analysis was performed between these PK data and midazolam clearance (as a CYP3A phenotyping probe) and hepatic 99mTc-MIBI scans (as a probe for ABCB1 [P-glycoprotein] activity). Results A total of 52 GIST and renal cell cancer patients were included in this study, of whom 46 were evaluable for analysis of sunitinib PK. Mean trough level of sunitinib, adjusted for dose, at day 1 of a new course was 14.7 ng/mL (range 7.2-28.9 ng/mL), and 56.2 ng/mL (range 19.1-151.1 ng/mL) in the 4th week of a treatment cycle. A significant correlation between N-desethyl-sunitinib trough levels and plasma clearance of midazolam (r = 0.58, P = 0.006) within 24 hours after the first intake of sunitinib was found. In addition, a trend was seen for the correlation between midazolam clearance and sunitinib trough levels at steady state (r = 0.26, P = 0.09). No correlation between sunitinib trough levels and hepatic 99mTc-MIBI clearance was seen in an interim analysis after 27 patients. Conclusions These observations show a correlation between the midazolam clearance test that serves as a probe for CYP3A and the exposure to sunitinib. Meanwhile, a lack of correlation was seen between sunitinib PK and the studied ABCB1 probe. These data could ultimately lead to a further personalization of sunitinib dosing. Additional investigations are in progress, including population PK modeling of sunitinib, and incorporation of pharmacogenetic parameters to further unravel the mechanisms behind these findings. Disclosure All authors have declared no conflicts of interest.