The formation of platinum (Pt)-DNA adducts is thought to be crucial to the antitumor activity of cisplatin, and relationships between adduct formation in peripheral blood leukocytes (PBLs) and response to cisplatin therapy have been reported. The current study directly tests, for the first time, whether pharmacokinetic or other factors predominantly determine the drug-target interaction of cisplatin in a pediatric patient population. Cisplatin pharmacokinetics and Pt-DNA adduct formation in PBLs were determined in 10 children in parallel with measurement of adduct levels after incubation of pretreatment blood samples with cisplatin in vitro. Total and unbound plasma Pt concentrations were determined by atomic absorption spectrophotometry and adduct measurements performed by competitive ELISA. Pt-DNA adduct levels determined after cisplatin treatment showed considerable interindividual variation (peak levels at 24 h ranged from 0.15 to 1.31 nmol/g DNA) and correlated strongly with adduct levels determined after incubation of pretreatment whole blood with cisplatin (r = 0.92; P = 0.0002). No significant correlation was observed between in vivo adduct formation and either unbound or total cisplatin plasma concentrations (r = 0.14 and 0.18, respectively). A correlation was also observed between the degree of myelosuppression, as determined by WBC nadirs measured over a 14-day period after cisplatin treatment, and the extent of adduct formation, with greater WBC toxicity observed in patients with higher levels of Pt-DNA adducts (P = 0.010). These preliminary results provide evidence that interpatient variation in formation of Pt-DNA adducts in PBLs of children is determined by host-specific factors other than cisplatin pharmacokinetics. These results imply that analysis of adducts in PBLs after incubation of pretreatment blood samples with cisplatin may be used to predict in vivo adduct levels, leukopenia, and, potentially, response to cisplatin therapy.