Inborn errors of metabolism are collectively common, frequently severe and in many instances difficult or impossible to treat. Accordingly, there is a compelling need to explore novel therapeutic modalities, including gene therapy, and examine multiple phenotypes where the risks of experimental therapy are outweighed by potential benefits to trial participants. Among available gene delivery systems recombinant AAV shows special promise for the treatment of metabolic disease given the unprecedented efficiencies achieved in transducing key target tissues, such as liver and muscle, in small animal models. To date over 30 metabolic disease phenotypes have been investigated in small animal studies with complete phenotype correction being achieved in a substantial proportion. Achieving adequately widespread transduction within the central nervous system, however, remains a major challenge, and will be critical to realization of the therapeutic potential of gene therapy for many of the most clinically troubling metabolic disease phenotypes. Despite the relatively low immunogenicity of AAV vectors, immune responses are also emerging as a factor requiring special attention as efforts accelerate toward human clinical translation. Four metabolic disease phenotypes have reached phase I or I/II trials with one, targeting lipoprotein lipase deficiency, showing exciting early evidence of efficacy.