PURPOSE: Following phase I studies of docetaxel and cisplatin in patients with non-small-cell lung cancer, the recommended doses of docetaxel were different for elderly (> or = 75 years) and non-elderly (< 75 years) patients. To elucidate the mechanism of the difference, the pharmacokinetics of docetaxel and cisplatin were investigated in two phase II studies separately conducted in elderly and non-elderly patients. PATIENTS AND METHODS: Twenty-seven elderly and 25 non-elderly patients were treated with three weekly administrations of docetaxel and cisplatin every 4 weeks. Doses of docetaxel were 20 and 35 mg/m(2) for elderly and non-elderly patients, respectively. All patients received 25 mg/m(2) of cisplatin. The pharmacokinetics and pharmacodynamics of docetaxel and cisplatin were compared in elderly and non-elderly patients. RESULTS: There were no differences in pharmacokinetics of docetaxel or cisplatin between elderly versus non-elderly patients with regard to clearance and volume of distribution. In the pharmacodynamic analysis, neutropenia was positively correlated with the area under the concentration-time curve for docetaxel but not for cisplatin. In evaluating the relationship between neutropenia and the area under the concentration-time curve of docetaxel, elderly patients experienced greater neutropenia than those predicted by a pharmacodynamic model developed in non-elderly patients; the residual for prediction of the percent change in neutrophil count was -11.2% (95% CI, -21.8 to -0.5%). CONCLUSION: The pharmacokinetics of docetaxel and unchanged cisplatin were not different between elderly and non-elderly patients. The elderly patients were more sensitive to docetaxel exposure than the non-elderly patients, resulting in the different recommended doses for the phase II studies.