The antitumor activity of irinotecan in vitro primarily results from its hydrolysis by carboxylesterase to the active metabolite SN-38. The present study was conducted to evaluate the effect of human neuroblastoma xenografts on irinotecan and SN-38 disposition after i.v. and oral irinotecan administration. Non-tumor-bearing mice and mice bearing three different human neuroblastoma xenograft lines (NB1691, NB1643, and NBEB) were given irinotecan (10 mg/kg) by short i.v. injection into the tail vein or by oral gavage. Serial plasma samples were obtained, processed to isolate irinotecan and SN-38 lactone, and assayed with a sensitive and specific high-performance liquid chromatography assay. Noncompartmental and compartmental pharmacokinetic analyses were performed. A four-compartment model was used for analysis of irinotecan and SN-38 concentration-time data after i.v. administration. The presence of tumor increased irinotecan systemic exposure (1.2-3.8-fold; P < 0.05) after i.v. and oral administration in mice bearing neuroblastoma xenografts compared to non-tumor-bearing mice. Moreover, SN-38 systemic exposures were higher (1.3-3.8-fold; P < 0.05) in mice bearing human neuroblastoma xenografts as compared to non-tumor-bearing mice, with the greatest effect observed after oral administration of irinotecan. A schematic model is presented to provide a mechanistic basis for our observations. These results emphasize the need to perform preclinical pharmacokinetic studies to evaluate the influence of tumor on drug disposition.