The influence of liver biochemistry tests on epirubicin pharmacokinetics has been investigated in 52 women with advanced breast cancer, 27 of whom had radiologically proven liver metastases. Patients received epirubicin 12.5-120 mg m-2 given as an i.v. bolus. Epirubicin levels were measured by HPLC following the first cycle of treatment. Epirubicin elimination, expressed as clearance (dose/AUC), in the 22 patients with normal AST and bilirubin was compared with that of 30 patients with a raised AST +/- raised bilirubin. Epirubicin clearance was significantly reduced in the patients with a raised AST, whether their serum bilirubin was normal (22 patients) or elevated (eight patients). In the 30 patients with a raised AST +/- raised bilirubin, epirubicin clearance correlated strongly with the level of AST (r = -0.72) but not with serum bilirubin, alkaline phosphatase, albumin or creatinine. Using a multiple regression analysis, AST was the only one of these biochemical variables predictive of epirubicin clearance (r2 = 0.47, P = 0.0006). We conclude that a raised serum AST is a more sensitive and reliable measure of abnormal epirubicin pharmacokinetics than increased bilirubin. These findings have implications for anthracycline treatment in patients with abnormal liver biochemistry.