Enzyme replacement therapy (ERT) has become the standard of care for several lysosomal storage disorders (LSDs). Despite ERT's undisputed efficacy, the requirement for multiple and costly administrations as well as ERT's limited improvement of some LSD manifestations prompts the search for better therapies. Using a mouse model of mucopolysaccharidosis VI, we compared the efficacy of a single intravascular administration of an adeno-associated viral vector targeting liver to weekly infusions of human recombinant enzyme at the same doses used in mucopolysaccharidosis VI patients. While gene therapy results in increased and stable levels of circulating enzyme up to 1 year after vector administration, ERT has typical peak-and-drop serum kinetics. Both therapies similarly reduced glycosaminoglycan levels in urine and tissues including heart valves and myocardium, with gene therapy improving skeletal skull abnormalities slightly better, although not significantly, than ERT. Both therapies seem to similarly improve animal motor performance, with gene therapy possibly associated with less animal distress. Thus, a single vector administration that converts liver into a factory organ for systemic secretion of therapeutic proteins is at least as effective as ERT in a mouse model of LSD, potentially eliminating problems with compliance and costs. Only testing in humans will prove whether this holds true in a clinical setting.