This study evaluated the population pharmacokinetics of dasatinib in Japanese subjects with imatinib-resistant or -intolerant chronic myeloid leukemia (CML) and Philadelphia chromosome positive acute lymphoblastic leukemia (Ph⁺ALL) enrolled in 3 Japanese clinical trials, and compared individual PK estimates to that of non-Japanese subjects in global database. A full Markov Chain Monte Carlo Bayesian analysis method in NONMEM 7 was utilized for the estimation of model parameters. In terms of covariate model selection, baseline demographic and clinical laboratory covariates were assessed on oral clearance (CL/F). The food effect and the dose effect were also tested on the relative bioavailability (F_R) and CL/F, respectively. A total of 706 observations were obtained from 63 Japanese subjects who received twice daily administration of dasatinib at 50, 70 and 90 mg, and once daily administration at 100 mg. Consistent with a PPK model in non-Japanese subjects, plasma concentration-time data were well described by a linear two-compartment model with the inter-occasion variability (IOV) on the relative bioavailability (F_R), which is to account for between-visit difference of dasatinib exposure within a subject apparently observed in a phase 1/2 study in Japanese subjects. Comparable exposures of individual Japanese subjects with that of non-Japanese subjects were obtained. Investigation of covariates revealed neither marked trends nor clinically relevant effect on CL/F or F_R. These results indicated that no dose adjustment is warranted for Japanese CML and Ph⁺ALL patients, based on their body size, age, or gender.