Cellular glutathione (GSH) levels are related to the resistance of tumor cells to platinum and alkylating agents, and depletion of GSH may enhance the activity of these drugs. The pharmacodynamic effects of mesna on depleting plasma cysteine, a GSH precursor, were evaluated in 22 patients as part of a Phase I study. Escalating doses of ifosfamide and mesna were administered; carboplatin was administered to achieve an AUC of 4 mg x min/mL. Plasma samples were collected and assayed by reverse-phase high-performance liquid chromatography (HPLC) for total mesna and total cysteine concentrations at 0, 1, 3, 6, 24, 25, 28, and 48 hours. A one-compartment pharmacokinetic model was fit to the mesna plasma concentrations, using M.A.P. Bayesian estimation (ADAPT II). Pharmacodynamics were evaluated by fitting an inhibitory Emax model to the cysteine concentration data. Both the pharmacokinetic (median R2 = 0.95; range = 0.85-0.98) and pharmacodynamic (median R2 = 0.96; range = 0.74-1.0) models fit the data well. Mean (coefficient of variation [CV%]) mesna pharmacokinetic parameter estimates were as follows: Vss of 15.3 (29) L/m2, CL of 4.6 (29) L/h/m2, and half-life of 2.2 (37) hours. Mean (CV%) pharmacodynamic parameter estimates were as follows: Emax of 31.7 (19) microg/mL and EC50 of 10.3 (52) microg/mL. Mesna produced a rapid, concentration-dependent reduction in plasma cysteine concentrations that could be adequately characterized by an inhibitory Emax pharmacodynamic model. The depletion of plasma cysteine was facilitated by ifosfamide, suggesting a pharmacodynamic interaction between these two agents. Further increases in mesna doses beyond those administered in this study would be unlikely to provide additional benefit.