Isolation and characterization of an anti-CD16 single-chain Fv fragment and construction of an anti-HER2/ neu/anti-CD16 bispecific scFv that triggers CD16-dependent tumor cytolysis

Bispecific antibody (bsAb)-based clinical trials of cancer have been conducted primarily using intact murine monoclonal antibody (mAb)-derived molecules. In some of these trials, toxicity resulting from the interactions of antibody Fc domains with cellular Fc receptors has limited the doses of antibody (Ab) that can be employed. Furthermore, human anti-mouse Ab responses prohibit multiple therapy courses. These factors have decreased the efficacy of the bsAb 2B1, which targets the extracellular domains (ECD) of the HER2/ neu protooncogene product and the human Fc γRIII (CD16). To address these obstacles, we have constructed and characterized a fully human gene-fused bsAb from single-chain Fv (scFv) molecules specific for HER2/ neu and CD16. The human anti-CD16 scFv component, NM3E2, was isolated from a human scFv phage display library. As binding of NM3E2 to human neutrophil-associated CD16 decreased in the presence of plasma IgG, we have concluded that NM3E2 recognizes an epitope in the vicinity of the Fc binding pocket. Furthermore, the NM3E2 scFv was found by surface plasmon resonance-based epitope mapping to share an overlapping epitope with the Leu-11c mAb. The human anti-HER2/ neu scFv component, C6.5, which was previously isolated from a human scFv phage display library, was employed as fusion partner for the creation of a bispecific scFv (bs-scFv). In the presence of the C6.5×NM3E2 bs-scFv, peripheral blood lymphocytes promoted significant lysis of human SK-OV-3 ovarian cancer cells overexpressing HER2/ neu. Biodistribution studies performed in SK-OV-3 tumor-bearing scid mice revealed that 1% ID/g of 125I-labeled C6.5×NM3E2 bs-scFv was specifically retained in tumor at 23 h following injection. These results indicated that both scFv components of the bs-scFv retained their function in the fusion protein. This bsAb should overcome some of the problems associated with the 2B1 bsAb. C6.5×NM3E2 bs-scFv offers promise as a platform for multifunctional binding proteins with potential clinical applications as a result of its human origin, lack of an Fc domain, ease of production, high level of in vitro tumor cell cytotoxicity and highly selective tumor targeting.