Multiple dose pharmacokinetics of erlotinib when combined with gastric acid reducing agents: A comparison with a physiologically based pharmacokinetic model

Abstract

Introduction: Erlotinib selectively inhibits the EGFR tyrosine kinase activity and consequently the tumour cell growth in patients. The drug is administered orally and being a weak base, its solubility is strongly dependent upon the acidic pH in the gastric fluid. Gastric acid reducing agents (ARAs), such as proton pump inhibitors (PPIs) and H2-receptor antagonists (H2RAs), increase the pH of the stomach (pH>4) and cause a physicochemical drug-drug interaction. The secretion ofH+ is drastically reduced by PPIs due to an irreversible binding to the H/K-ATPase pump. H2RAs have a shorter elimination half-life and competitively inhibit histamine action of H2 receptors, on gastric parietal cells. Our objective was to evaluate the plasma concentrations of erlotinib when given alone or in combination with different groups of ARAs (PPIs and H2RAs) and to simulate the erlotinib plasma concentration using a physiologically based pharmacokinetic (PBPK) model to evaluate possible physicochemical interactions from ARA co-medication. Methods: Three groups of each 8 patients, suffering from pancreatic cancer, received 100 mg erlotinib daily as fixed dose (control group), combined with PPI pantoprazole (PPI group) or combined with H2RA famotidine (H2RA group). Blood samples were collected on day 1 (pre-dose,1,2,3,4,6,8 and 24 hours after administration) and on days 2-7 (pre-dose and 4 hours after administration). Erlotinib samples were stored at -80-C and were quantified by a HPLC assay. Pharmacokinetic parameters were calculated by a noncompartmental method for extravascular input using WinNonlin 6.0(Phoenix Inc.). The PBPK model was built with the software Gastro PlusTM(Simulations Plus Inc.) to simulate plasma concentrations of erlotinib in a population of 25 Caucasian patients. Results: The PBPK model output corresponds well to the mean observed erlotinib plasma concentrations of the control group. The observed Cmax and AUC0-24 of erlotinib on day 1 were consistent with the predicted concentrations by Gastro PlusTM(Cmaxobserved=0.78 mg/mL vs. predicted=0.76 mg/ml, AUC0-24observed =10.7 hr∗mg/mL vs. predicted=11.8 hr∗mg/ml). The mean trough and peak concentrations showed a high inter-patient variability over the whole investigated period in all patient groups. The co-administration of PPIs decreased the erlotinib trough and peak concentrations as well as the AUC0-24 about 50% compared to the control and H2RA group. The mean trough concentration in the PPI group on day 7 was 0.36 mg/mL and therefore below the necessary threshold concentration of 0.5 mg/mL to inhibit the tyrosine kinase activity. On the contrary, the pharmacokinetic parameters of erlotinib did not differ significantly in presence of the H2RAs. The mean trough and peak concentration of erlotinib in the H2RA group on day 7 were 0.671 and 1.78 mg/mL and thus similar to the values of the control group (Ctrough=0.950 mg/mL, Cpeak=1.76 mg/mL). In the H2RA and control group all measured plasma concentrations exceeded the threshold. Conclusion: Co-administration of H2RA drugs instead of PPIs is strongly recommended during erlotinib treatment. H2RAs are given 12 hours before erlotinib administration and therefore show no influence on erlotinib plasma concentrations. PBPK is a useful tool to simulate plasma concentration-time curves of a drug and model possible interactions based on patient observations, physicochemical properties and drug classifications.