Thalidomide has been used for the treatment of refractory multiple myeloma, the dosage in Japan is lower than in other countries; however, there is little information on the pharmacokinetics and their relationship with the drug response. The aim of this study was to characterize the pharmacokinetics of low-dose thalidomide in Japanese patients with refractory multiple myeloma, and to examine the relationship between pharmacokinetics and adverse events. On the first and second days, a 100 mg capsule was administered to 8 Japanese patients after breakfast and blood samples were obtained. The plasma concentrations were measured using HPLC and analyzed based on a one-compartment model. If intolerable adverse events were not observed for 14 d, the dose was increased to 200 mg. The average apparent volume of distribution (Vd/F), apparent total clearance (CL/F) and area under the plasma concentration-time curve from 0 to infinity (AUC0-infinity), which were 45.3 l, 5.5 l/h and 21.7 microg.h/ml, respectively, with smaller Vd/F and CL/F and larger AUC0-infinity than in Caucasian populations. This pharmacokinetic difference may explain the dose difference between Japan and other countries. Adverse events were associated with AUC0-infinity, which was best correlated with plasma concentration at 12 h after administration. The 12-h time point was suggested to be a capable indicator for "safety-oriented" therapeutic drug monitoring of thalidomide.