OBJECTIVE With increasing interest in pretargeting procedures for improving the delivery of radionuclides for cancer imaging and therapy, this investigation was undertaken to examine how to optimize a bispecific monoclonal antibody (bsMAb) pretargeting procedure for therapeutic applications. METHODS The model system examined was a bsMAb composed of two Fab' fragments, one from a humanized anti-carcinoembryonic antigen antibody (hMN-14), and the other a murine antibody (679) against histamine-succinyl-glycine. These Fab' fragments were chemically conjugated to form a F(ab')(2) that is joined by a stable thioether bond. The peptide used for these studies (IMP-241) contained two histamine-succinyl-glycine moieties for binding to the 679 portion of the bsMAb and a single 1,4,7,10-tetra-azacyclododecane N,N',N",N"'-tetraacetic acid chelate for radiolabeling with (111)In. RESULTS The bsMAb cleared rapidly in nude mice bearing the GW-39 human colonic cancer xenograft. Administration of a radiolabeled peptide 1 day after the bsMAb, using a bsMAb/peptide mole injection ratio of 10:1, allowed for higher tumor accretion than if delayed by 2 days. Tumor uptake measured 3 h after the peptide injection given 1 day after the bsMAb was 11.3 +/- 2.2% percentage of injected dose/gram (%ID/g), with just 2.9 +/- 0.4% ID/g of the bsMAb in the tumor at this time. Tumor/blood ratios were 8.1 +/- 2.1. Peptide uptake was highest in the kidneys, but even so, the tumor/kidney ratio was 2.5 +/- 1.9 just 3 h after the peptide injection. Although low bsMAb/peptide mole injection ratios allow for greater concentrations of the peptide in the tumor, kidney uptake is increased at a proportionally higher amount than in the tumor. Therefore, a bsMAb/peptide injection ratio of 10:1 with a 24-h interval was preferred for pretargeting. Increasing the bsMAb dose, and thereby increasing the bsMAb/peptide injection ratio, further enhanced the delivery of the radiolabeled peptide to the tumor, but the interval spacing between the bsMAb and peptide had to be increased. Despite having a lower %ID/g of the bsMAb in the tumor, with a bsMAb/peptide injection ratio of 50:1 and a 48-h interval, tumor uptake of the (111)In-peptide was nearly 30% ID/g, a 1.6-fold improvement over that seen with the 10:1/24-h interval pretargeting group, and tumor/blood was 35:1, and tumor/kidney ratio was 8:1. Two fractionation strategies were also examined. Giving two equal fractions of peptide after a single injection of bsMAb loaded more moles of peptide into the tumor but would not permit higher radioactivity delivery than what could be achieved with a single injection. However, area under the curve analysis indicated that giving repeated cycles of the bsMAb followed by the peptide would enable improvements in the amount of radioactivity delivered to the tumor without increasing the amount delivered to normal tissues, but the timing of the bsMAb/peptide cycles was important to optimize this process. Finally, it was noted that larger tumors (e.g. those > 0.3 g) were more likely to have higher peptide uptake in a pretargeting procedure than smaller tumors (e.g., those of approximately 0.1 g), perhaps due to the greater mass of the bsMAb localized in the larger tumors, but also possibly because of better blood supply in these tumors. CONCLUSIONS These studies reveal principles that might be applied generally to other pretargeting procedures and demonstrate how a bsMAb pretargeting method could potentially exceed a directly radiolabeled antibody in its ability to deliver radionuclides for cancer therapy.