Objective: Overexpression of the tyrosine kinase receptor erbB-2 is important in the pathogenesis of a variety of neoplasms including ovarian cancer. As a strategy to selectively eradicate erbB-2-overexpressing tumor cells, an anti-erbB-2 single-chain immunoglobin (sFv) gene was constructed to direct expression of intracellular anti-erbB-2 antibody. The purpose of this study is to establish tim antitumorigenicity of this strategy in in vitro and in vivo models. Methods: An anti-erbB-2 sFv construct containing an endoplasmic reticulum (ER)-directed leader sequence was transiently expressed in the human ovarian carcinoma cell line SKOV3 using the adenovirus-polylysine vector. SKOV3 cells transfected with a non-ER form of an anti-erbB-2 sFv construct or an irrelevant plasmid DNA served as controls. Antitumorigenicity, as measured by anchorage-independent growth in soft agar and subcutaneous (sc) tumor formation, was analyzed. The ability to achieve a biological effect with relevant sFv in murine orthotopic xenograft models and in primary human ovarian cancer cells was also evaluated. Results: The ER form of anti-erbB-2 sFv was found to exert a marked antineoplastic effect on the SKOV3 cell line resulting in an arrest of anchorage-independent growth. A significant increase in sc tumor volume was noted in animals challenged with control constructs. In marked contrast, complete tumor eradication was noted at necropsy 80 days after sc transplantation in the group challenged with the ER-directed anti-erbB-2 sFv gene. Intraperitoneal treatment of malignant ascites in human tumor xenograft models with the ER form of the anti-erbB-2 sFv gene resulted in profound downregulation of cell surface erbB-2 in retrieved ovarian cancer cells. The ER-directed anti-erbB-2 sFv also elicited a significant cytotoxic effect in transfected primary ovarian cancer cells obtained from a patient with malignant ascites. Conclusion: The ability to selectively "knock out" erbB-2 demonstrates that this strategy can induce a significant antineoplastic effect in ovarian cancer cells overexpressing this growth factor receptor. In addition, the ability to accomplish selective abrogation of erbB-2 expression in animal treatment models and to transfect and eradicate primary ovarian cancer cells justifies further investigation of this novel strategy in ovarian cancer patients.