Despite the growing number of cancer cases and cancer surgeries around the world, the pharmacokinetics (PK) and pharmacodynamics (PD) of anesthetics used in this population are poorly understood. Patients operated due to cancer are usually in severe state and often require chemotherapy. It might affect the PK/PD of drugs used in this population. Therefore, in this study we explored the PK/PD of propofol in cancer patients having a major lung surgery. 23 patients that underwent a propofol–fentanyl total intravenous anesthesia were included in the analysis. A large set of demographic, biochemical and hemodynamic parameters was collected for the purpose of covariate analysis. Nonlinear mixed effect modeling in NONMEM was used to analyze the collected data. A three-compartment model was sufficient to describe PK of propofol. The anesthetic effect (AAI index) was linked to the propofol effect site concentrations through a sigmoidal Emax model. A slightly higher value of clearance, a lower value of distribution clearance, and a decreased volume of peripheral compartment were observed in our patients, as compared with the literature values reported for healthy volunteers by Schnider et al. and by Eleveld et al. Despite these differences, both models led to a clinically insignificant bias of −8 and −1 % in concentration predictions, as reflected by the median performance error. The Ce50 and propofol biophase concentration at the time of postoperative orientation were low and equaled 1.40 and 1.13 mg/L. The population PK/PD model was proposed for cancer patients undergoing a major lung surgery. The large body of studied covariates did not affect PK/PD of propofol significantly. The modification of propofol dosage in the group of patients under study is not necessary when TCI-guided administration of propofol by means of the Schnider model is used.Electronic supplementary materialThe online version of this article (doi:10.1007/s10928-015-9404-6) contains supplementary material, which is available to authorized users.