There is considerable evidence that DNA intercalating drugs fail to penetrate tumor tissue efficiently. This study used the multicellular layer (MCL) experimental model, in conjunction with computational modeling, to test the hypothesis that a DNA intercalator in phase II clinical trial, N-[2-(dimethylamino)-ethyl]acridine-4-carboxamide (DACA), has favorable extravascular transport properties. Single cell uptake and metabolism of DACA and the related but more basic aminoacridine 9-[3-(dimethylamino)propylamino]acridine (DAPA), and penetration through V79 and EMT6 MCL, were investigated by high-performance liquid chromatography. DACA was accumulated by cells to a lesser extent than DAPA and was metabolized to the previously unreported acridan by V79 but not EMT6 cells. Despite this metabolism, flux of DACA through MCL was much faster than that of DAPA. Modeling MCL transport as diffusion with reaction (metabolism and reversible binding) showed that the faster flux of DACA was due to a 3-fold higher free drug diffusion coefficient and 10-fold lower binding site density. The MCL transport parameters were used to develop a spatially resolved pharmacokinetic model for the extravascular compartment in tumors, which provided a reasonable prediction of measured average tumor concentrations from plasma pharmacokinetics in mice. Area under the curve was essentially independent of distance from blood vessels, although the combined pharmacokinetic/pharmacodynamic model predicted a modest decrease in cytotoxicity (from 1.8 to 1.1 logs of cell kill) across a 125-microm region. In conclusion, this study demonstrates that it is possible to design DNA intercalators that diffuse efficiently in tumor tissue, and that there is little impediment to DACA transport over distances required for its antitumor action.