907P - Impact of CYP3A4*22 on pazopanib pharmacokinetics in cancer patients

Abstract

Background: Pazopanib is characterized by a large interpatient variability in systemic drug exposure. As pazopanib trough levels (>20.5 mg/L) are correlated with clinical outcome (Suttle et al, BJC 2014) in metastatic renal cell carcinoma (mRCC) patients, it is vital to identify factors that influence pazopanib pharmacokinetics (PK). The objective of the current analysis was to evaluate if single nucleotide polymorphisms (SNPs) in the metabolic pathway of pazopanib (i.e. CYP3A4, ABCB1 and ABCG2) affect systemic pazopanib concentrations. Methods: We analyzed 97 patients who participated in 3 pazopanib PK studies. Starting point of the current analysis was a population PK model for pazopanib (Yu et al, Clin Pharmacokinet 2017). Four SNPs located on 3 genes, that were associated with decrease of function were analyzed using real time PCR: CYP3A4 15389 C>T (*22), ABCB1 3435 C>T, and the ABCG2 SNPs 421 C>A, and 34G>A. The influence of these SNPs on pazopanib bioavailability and clearance (CL) was explored with NONMEM. Statistical significance was determined with the likelihood ratio test using the objective function value (OFV). Trough concentrations (Ctrough) at 6 weeks after start with doses of 400 to 800 mg once daily (OD), were simulated. A threshold Ctrough of 20.5 mg/L was used as reference. Results: From 3 patients, insufficient DNA was isolated to run a PCR analysis. All SNPs were in Hardy-Weinberg equilibrium. Eleven patients (12%) had a variant allele at CYP3A4*22, all of whom were heterozygous. Incorporation of CYP3A4*22 in the NONMEM model resulted in a 35% lower CL for the variant carriers (0.18 L/h vs 0.27 L/h; ΔOFV = -7.8; P < 0.01). Simulated median Ctrough of patients with CYP3A4*22 with 400 mg OD, 600 mg OD or 800 mg OD were 16 mg/L, 25 mg/L and 33 mg/L, respectively. Simulated Ctrough for the population excluding the CYP3A4*22 heterozygotes after 800 mg OD was 21 mg/L. No effect of the ABCB1 or ABCG2 SNPs on systemic concentrations were found. Conclusions: Our analysis shows that CYP3A4*22 carriers have a clinically relevant lower pazopanib CL. Prospective analysis should point out whether CYP3A4*22 carriers are at risk for more toxicity and require a lower pazopanib starting dose. Legal entity responsible for the study: Erasmus MC, Rotterdam, The Netherlands Funding: None Disclosure: All authors have declared no conflicts of interest.