Gene Therapy (2002) 9, 256–262  2002 Nature Publishing Group All rights reserved 0969-7128/02 $25.00 www.nature.com/gt RESEARCH ARTICLE Effective single chain antibody (scFv) concentrations in vivo via adenoviral vector mediated expression of secretory scFv WO Arafat 1,4 , J Go ´ mez-Navarro 1,4 , DJ Buchsbaum 5 , J Xiang 1,4 , M Wang 1,4 , E Casado 1,4 , SD Barker 1,4 , PJ Mahasreshti 1,4 , HJ Haisma 6 , MN Barnes 3 , GP Siegal 2,4 , RD Alvarez 3,4 , A Hemminki 1,4 , DM Nettelbeck 1,4 and DT Curiel 1,4 1 Division of Human Gene Therapy, Departments of Medicine, Pathology and Surgery, University of Alabama at Birmingham, Birmingham, AL, USA; 2 Departments of Pathology, Cell Biology and Surgery, University of Alabama at Birmingham, Birmingham, AL, USA; 3 Department of Obstetrics and Gynecology, University of Alabama at Birmingham, Birmingham, AL, USA; 4 Gene Therapy Center, University of Alabama at Birmingham, Birmingham, AL, USA; 5 Department of Radiation Oncology, University of Alabama at Birmingham, Birmingham, AL, USA; and 6 Department of Therapeutic Gene Modulation, University Centre for Pharmacy, University of Groningen, The Netherlands Single chain antibodies (scFv) represent powerful inter- ventional agents for the achievement of targeted thera- peutics. The practical utility of these agents have been lim- ited, however, by difficulties related to production of recombinant scFv and the achievement of effective and sus- tained levels of scFv in situ. To circumvent these limitations, we have developed an approach to express scFv in vivo. An anti-erbB2 scFv was engineered for secretion by eukaryotic cells. The secreted scFv could bind to its target and specifi- cally suppress cell growth of erbB2-positive cells in vitro. Adenoviral vectors expressing the cDNA for the secretory Keywords: single chain antibody; gene therapy; adenoviral vector; erbB-2; immunotherapy Introduction The development of antibody-based therapies has allowed the achievement of the first valid targeted ther- apies for neoplastic disease. 1 Specifically, the direct administration of anti-tumor monoclonal antibodies (mAbs) has shown efficacy in a variety of model systems, and currently has demonstrated definitive benefit in human clinical trials in the context of various lymphoma tumor targets. 2 Based on these promising results, a num- ber of strategies have endeavored to optimize the per- formance of mAbs for clinical application. In this regard, immune response against murine antibody domains not involved in antigen recognition/binding (HAMA) has led to the development of single chain antibody (scFv) agents. 3 These antibody components, derived via molecu- lar engineering or phage display methods, can embody the full anti-tumor potency of the parent antibody. Fur- thermore, their size/structure offer some advantages in Correspondence: DT Curiel, Division of Human Gene Therapy, Gene Therapy Center, University of Alabama at Birmingham, 1824 6th Avenue South, Room WTI 620, Birmingham, Alabama 35294, USA Received 15 June 2001; accepted 20 November 2001 scFv likewise could induce target cells to produce an anti- tumor anti-erbB2 scFv. In vivo gene transfer via the anti- erbB2 scFv encoding adenovirus also showed anti-tumor effects. Thus, by virtue of engineering a secreted version of the anti-tumor anti-erbB-2 scFv, and in vivo expression via adenoviral vector, effective concentrations of scFv were ach- ieved. In vivo gene transfer clearly represents a powerful means to realize effective scFv-based approaches. This method will likely have applicability for a range of disorders amenable to targeted therapeutic approaches. Gene Therapy (2002) 9, 256–262. DOI: 10.1038/sj/gt/3301639 terms of biodistribution, tumor penetration and immuno- genicity. Despite the potential advantages offered by scFvs, full employment of these agents has been limited by several key factors. In this regard, the principle means for realiz- ation of preparative amounts of scFv has been via pro- duction in recombinant systems in vitro. Unfortunately, only a minority of candidate anti-tumor scFvs has proven amenable to large scale, high yield production via these methods. Further, biodistribution and metabolism aspects frequently limit the achievement of effective and sustained levels of anti-tumor scFv in the target tissue. 3,4 On this basis we have explored a novel strategy of employment of scFv for anti-tumor applications. Taking advantage of the capacity of adenovirus vectors to achi- eve efficient in vivo gene transfer, we have endeavored to express anti-tumor scFvs as secretory molecules in situ. This method appears to offer the means to produce func- tional scFv via eukaryotic cell synthesis, and to allow for high level and sustained concentrations of scFv, which can achieve anti-tumor efficacy. This in vivo secretory scFv approach may thus allow wider application of the anti-tumor scFv approach via circumventing the key technical and biological limits noted in this context.