To investigate the exposure–safety relationships of afatinib in Japanese population, we performed population pharmacokinetics (PK) analysis of afatinib in Japanese advanced non-small cell lung cancer patients harboring epidermal growth factor receptor mutation. Plasma samples were collected at 0.5–1, 2–3, 8–12, and 24 h after oral afatinib (40 mg) administration on day 1 and day 8. Plasma afatinib concentrations were determined using high-performance liquid chromatography. Data was analyzed following the population approach and using the software Phoenix® NLMETM Version 7.0 software (Certara USA, Inc., Princeton, NJ, USA). From 34 patients, a total of 354 afatinib plasma concentration values were available for the population PK analysis. Significant covariates in the population PK model included aspartate aminotransferase and creatinine clearance on CL/F, and age and body mass index on V/F. Results of simulation based on final PK model indicated that hepatic impairment had a significant effect on afatinib levels in plasma after multiple dosing. Afatinib trough plasma concentrations on day 8 were higher in patients with adverse events of grade 3 or higher. The population PK analysis showed that hepatic impairment affected afatinib PK parameters and contributed to the high inter-patient variability and high plasma concentrations of afatinib following multiple treatments.