Streptavidin distribution in metastatic tumors pretargeted with a biotinylated monoclonal antibody: theoretical and experimental pharmacokinetics.

We have developed a pharmacokinetic model for the analysis of a protocol that involves injection of a biotinylated monoclonal antibody followed at a later time by radiolabeled streptavidin. Three distinct physiological spaces are described: an avascular tumor nodule, the normal tissue surrounding the tumor, and the plasma. The model incorporates processes such as plasma kinetics, transcapillary transport, interstitial diffusion, binding reactions, and lymphatic clearances. We have modeled cases in which antigen turnover does not occur, in which antigen turnover does occur (24-h time constant), and in which circulating antibody is cleared from the plasma immediately prior to injection of streptavidin. We have calculated the spatial and temporal distributions of a tumor-specific antibody and of streptavidin in the tumor nodule using parameter values that simulate conditions of recent experiments on metastatic nodules in the guinea pig lung. The theoretical distribution of streptavidin in the tumor nodule shows an initial localization at the periphery that progresses to a fairly uniform distribution throughout the nodule, a temporal sequence that is very similar to experimental observation. This finding indicates that, in a tumor pretargeted with biotinylated antibody, streptavidin can encounter significant retardation in its penetration as a consequence of the high affinity interaction between these two species. Tumor:blood and tumor:lung ratios were calculated and compared to experimental results. In addition, the calculated tumor:blood ratios, tumor:lung ratios, and relative exposures were compared to values obtained from a model of one-step antibody delivery. The two-step protocol yielded an approximately 2- to 3-fold enhancement in these pharmacokinetic indices compared with the one-step method.