Population Pharmacokinetics of Erlotinib in Patients With Non–small Cell Lung Cancer: Its Application for Individualized Dosing Regimens in Older Patients

Purpose Erlotinib is an oral first-generation epidermal growth factor receptor (EGFR) tyrosine kinase inhibitor approved for non–small cell lung cancers (NSCLC) with EGFR-activating mutations. Older patients experience more toxicities compared with younger patients at the standard recommended dose of 150 mg once daily. The aims of this study were to describe the pharmacokinetic profile of erlotinib in unselected patients with NSCLC, to quantify and explain its variability, to challenge the standard recommended dose in older patients, and to propose clinical recommendations for the therapeutic management of patients taking erlotinib. Methods : A population pharmacokinetic model was developed using erlotinib plasma concentrations collected from patients with NSCLC participating in a routine therapeutic drug monitoring program (with the nonlinear mixed effect modeling program NONMEM). Relevant demographic characteristics, clinical factors, and co-medications were tested as potential covariates. An independent dataset was used for model validation. Simulations based on the final model allowed comparison of expected erlotinib concentrations under standard and alternative dosing regimens for smokers and for several age groups. Findings A total of 481 erlotinib plasma concentrations from 91 patients with NSCLC were used for model building and 239 plasma drug concentrations from 107 patients for model validation. A one-compartment model with first-order absorption and elimination provided the best fit. Average erlotinib CL/F with interindividual variability (%CV) was 3.8 L/h (41.5%), and V/F was 166 L (53.8%). The absorption rate constant was 1.48 h−1. The external validation showed a negligible bias of −4% (95% CI, −7 to −1) in the individual predictions, with a precision of 23%. Current smoking and use of proton pump inhibitors were associated with higher CL/F, whereas age was associated with lower CL/F. Simulations suggest that a lower dose in older patients would decrease the risk of overexposure. Implications This large cohort study confirms the substantial interindividual variability in erlotinib plasma exposure and the impact of smoking and proton pump inhibitor intake. This large variability in erlotinib pharmacokinetics indicates that the standard recommended dose of 150 mg once daily is likely not appropriate to reach the expected concentrations in each patient. Concentration monitoring should be performed to individually adjust the erlotinib dosing regimen. The observed decrease in erlotinib CL/F with age suggests that a lower starting daily dose of 100 mg with concentration-guided dose adjustment would prevent overexposure and potential toxicity in older frail patients with co-morbidities.