Comparative pharmacokinetics of a murine monoclonal antibody to a rat colon tumor in rats and nude mice.

The pharmacokinetics of the 131I-labeled murine anti-rat colon carcinoma monoclonal antibody, E4, and its F(ab')2 fragments were compared in normal Sprague-Dawley rats as well as in syngeneic BDIX rats and nude mice bearing the tumor to which the monoclonal antibodies had been generated. 125I-labeled irrelevant antibody of the same IgG2a subclass or its labeled F(ab')2 fragments were used as controls. Results of labeled antibody uptake after i.v. administration were analyzed in terms of accumulation and localization indices for normal tissues and tumor. Whole E4, which is tumor specific by immunoperoxidase staining, bound to a variety of normal tissues, in addition to the tumor. These tissues included liver, stomach, colon, and lung of rats bearing s.c. tumors. Targeting to the tumor was better in rats bearing i.p. tumors, but targeting to other organs was also high. The use of F(ab')2 fragments in rats bearing s.c. tumors showed results similar to those found with the whole antibody and did not improve tumor localization. These observations demonstrate a lack of correlation between in vivo tissue uptake of injected murine monoclonals, or their fragments, and in vitro histochemical studies. In contrast to observations in rats, tumor alone was specifically targeted with whole E4 or F(ab')2 fragments in tumor-bearing nude mice. Trichloroacetic acid precipitation of rat and mouse tissue radioactivity indicated that labeled antibody, and not catabolites or free iodine, was being measured. These data suggest that studies on the xenogeneic nude mouse model may not necessarily be relevant to the choice of monoclonal antibodies for clinical diagnostic imaging or therapy.