The use of targeted viral vectors to localize gene transfer to specific cell types holds many advantages over conventional, non-targeted vectors currently used in gene therapy. The resulting improvements in gene localization from targeted adenovirus vectors are likely to reduce immunogenicity and toxicity, increase safety, and enable the systemic administration of these vectors for multiple indications including cancer, cardiovascular disease, and inflammatory disease. Recent advances in the biological understanding of adenovirus structure and adenovirus receptor interactions have fueled the rapid development of targeted adenovirus vectors. Two basic requirements are necessary to create a targeted adenovirus vector: interaction of adenovirus with its native receptors must be removed and novel, tissue-specific ligands must be added to the virus. Two general approaches have been used to achieve these basic requirements. In the 'two-component' approach, a bispecific molecule is complexed with the adenovirus. The bispecific component simultaneously blocks native receptor binding and redirects virus binding to a tissue-specific receptor. In the 'one-component' approach the adenovirus is genetically modified to ablate native receptor interactions and a novel ligand is genetically incorporated into one of the adenovirus coat proteins. Two-component systems offer great flexibility in rapidly validating the feasibility of targeting via a particular receptor. One-component systems offer the best advantages in producing a manufacturable therapeutic and in more completely ablating all native adenovirus receptor interactions. The coming challenges for targeted adenovirus vectors will be the demonstration that the technology performs in vivo. Ultimately, or in parallel, 'receptor-targeting' technology can be combined with improved adenovirus backbones and with 'transcriptional targeting' approaches to create adenovirus which deliver genes selectively, safely, and with little immune response.