It is easier to kill cancer cells in culture than in solid tumors, in part because therapeutic agents have poor microscopic access in the latter. For a number of years, we have been exploring ways to integrate macroscopic and microscopic aspects of the pharmacology of biologically interesting ligands, principally to get at the problem of poor access. That work has centered on monoclonal antibodies but with an eye to correlates in the pharmacology of other biological ligands and low molecular weight agents. Theoretical: We developed a program package (PERC; operating on the CRAY supercomputer) that solves differential equations for macroscopic and microscopic pharmacology. We then formulated the """"""""binding site barrier"""""""" hypothesis - i.e., that the very fact of successful binding to target antigen or receptor can limit penetration into the substance of a tumor. Calculations suggested that (1) the barrier effect could prevent penetration even 100-200 microns from a blood vessel; (2) paradoxically, high affinity and high target density are expected to produce lower concentrations of ligand a few hundred microns from a vessel. Experimental: We validated the binding site barrier hypothesis experimentally in bulk tumor an micrometastases of L10 carcinoma in guinea pigs. A combination of double-label autoradiography and double-chromophore immunohistochemistry was used to detect simultaneously the distribution of antibody, control IgG, antigen, and blood vessels. We have now extended these calculations and experiments to two-step therapy, in which a slowly distributing ligand is chased by a fast-distributing ligand-effector chimera. We believe that the """"""""binding site barrier"""""""" has been a factor in the evolution of autocrine-paracrine molecules and other biological ligands. As a corollary, the micropharmacology must be considered when designing ligands for exogenous administration or for secretion in vivo by genetically modified cells. Current work focuses on the access problem in a new hollow fiber model for solid tumors that we have developed (with Hollingshead, Mayo, et al., see Z01CM07349-02).
