Study objective: To investigate the pharmacokinetics and pharmacodynamics of high-dose intravenous zidovudine (ZDV).
Design: Phase 1, dose-escalating, unblinded study in patients with cancer.
Setting: A university-affiliated cancer treatment center.
Patients: Fourteen patients (6 women) with solid tumors that were unresponsive to standard therapy received 31 courses of therapy.
Interventions: Intravenous ZDV was administered in doses of 2, 3, 4, 5.5, 7, 8.5, 10, 12, 15, or 20 g/m2/day as a continuous infusion over 48 hours. Patients also received fluorouracil plus leucovorin for 24 hours before the start of and during the ZDV infusion. If no dose-limiting toxicities were encountered, subsequent doses were escalated. Blood samples were collected at 24 and 48 hours after the start of the infusion, and hourly for 4 hours after stopping the infusion. Urine was collected in five patients during the infusion and for 24 hours after stopping it. Blood for measuring peripheral white blood cells was collected before and at the end of the infusion in seven patients to measure DNA chain breaks due to incorporation of ZDV.
Measurements and main results: Zidovudine was measured in plasma by high-performance liquid chromatography and in urine fluorescence polarization immunoassay. Its incorporation into DNA was measured by determining DNA strand breakage in peripheral white blood cells using fluorescence analysis. Pharmacokinetic models were fit to plasma ZDV concentrations using extended least squares regression. Short-term high-dose ZDV was generally well tolerated, with adverse effects related to large amounts of free water administered during the infusion. The mean (SD) ZDV pharmacokinetic values were total clearance 1.44 (1.09) L/hr/kg, volume of distribution 2.72 (2.97) L/kg, and half-life 1.2 (0.6) hours. There was considerable interpatient variability in total drug clearance. Although ZDV exposure increased proportionately with increasing dose, two of three patients receiving the highest dose (20 g/m2/day) had markedly low total drug clearances. The relation between the percentage of abnormal DNA in peripheral white blood cells and zidovudine area under the plasma ZDV versus time curve was described by the Emax pharmacodynamic model.
Conclusions: The pharmacokinetics of high-dose ZDV administered by continuous infusion to patients with cancer are similar to those reported with lower doses in patients with infection due to the human immunodeficiency virus. Further study of potential nonlinear pharmacokinetic behavior at doses above 20 g/m2/day is necessary. The high between-patient variability in ZDV clearance results in variable levels of exposure in vivo, and indicates the need for concentration- or effect-controlled study designs in the further evaluation of the agent's antineoplastic effects.