We performed in vitro and in vivo experiments of the anti-epidermal growth factor receptor (EGFR) x anti-CD3 bispecific diabody (hEx3-Db) with the IgG-like bispecific antibodies (BsAbs) (hEx3-scFv-Fc and hEx3-scDb-Fc) and the anti-EGFR therapeutic antibody cetuximab to assess the effect of BsAbs on cancer growth inhibition. In vitro, efficacy of the BsAbs and cetuximab were compared by growth inhibition assays of human cell lines of bile duct (TFK-1, HuCC-T1, OCUCh-LM1), epidermoid (A431), gastric (Kato-III), colon (DLD-1, SW480), and breast (SK-BR-3, MCF-7) cancer. In vivo, in three mouse models, we evaluated the anti-tumor activity of hEx3-Db and cetuximab, assessed the effect of hEx3-Db alone, and compared the antitumor activity of hEx3-Db with the IgG-like BsAbs. In vitro, hEx3-scFv-Fc showed nearly 100% killing activity for all cell lines. Both in vitro and in vivo, hEx3-Db needed CD3-positive phenotypes to induce a growth inhibitory effect. In contrast, IgG-like BsAbs showed monotherapeutic effects in vivo by inducing antibody-dependent cellular cytotoxicity (ADCC) similar to cetuximab. However, enhancement was not observed when lymphokine-activated killer cells with the T-cell phenotype were co-injected. Results suggest that IgG-like BsAbs could not efficiently direct T lymphocytes toward tumor cells to induce ADCC due to steric hindrance on binding to CD3- and Fc-receptor-positive phenotypes. Although hEx3-scFv-Fc showed high cytotoxicity in vitro, its high molecular weight limits its usefulness. With an in vivo effect comparable to hEx3-scFv-Fc and its realistic molecular weight, hEx3-scDb-Fc shows promise as a novel recombinant therapeutic antibody and may be modified to enhance its potency by prevention of steric hindrance.