Population pharmacokinetics and pharmacodynamics of cisplatin in patients with cancer: analysis with the NONMEM...

The population pharmacokinetics and pharmacodynamics of cisplatin (CDDP) were evaluated based on a mixed-effect model using the NONMEM program. Unchanged CDDP in plasma was measured as a biologically active platinum species during CDDP chemotherapy, using high-performance liquid chromatography. Plasma concentration measurements (157) of unchanged CDDP from 26 patients with cancer receiving 80 mg/m2 CDDP by infusion over 2 hours, 3.5 hours, or 4 hours were analyzed according to a one-compartment model. The influences of individual characteristics such as body weight, dose schedule, course, and clinical laboratory values (renal function markers, albumin) on total body clearance (Cl) and volume of distribution (Vd) were examined. In the final pharmacokinetic model, body surface area and dose schedule affected Cl of unchanged CDDP. The Cl of CDDP was increased by 27.3% after the 2-hour infusion schedule compared with Cl after the longer infusions. The Vd was estimated as 13.4 L/m2. The interindividual variability for Cl and Vd and residual variability were 22.9%, 30.9%, and 35.5%, respectively. The relationships between maximum concentration (Cmax) of unchanged CDDP and maximum blood urea nitrogen (BUNmax), or minimum creatinine clearance (ClCr,min) over a 1-month period after CDDP administration were evaluated according to linear, exponential, or maximum response (Emax) models. The linear or Emax model described pharmacodynamics most successfully, with relatively large interindividual variability for both slope and EC50 (more than 25%). Residual variability was 15.3% and 17.1% in BUNmax and Clcrmin, respectively. The population means and interindividual and residual variability of pharmacokinetics and pharmacodynamics of CDDP were evaluated using the NONMEM program. The results of this study show that the population pharmacokinetic and pharmacodynamic approach could be useful to manage CDDP nephrotoxicity using sparse data in a clinical setting.