Due to the complexity of brain function and the difficulty in monitoring alterations in neuronal gene expression, the potential of lentiviral gene therapy vectors to treat disorders of the CNS has been difficult to fully assess. In this study, we have assessed the utility of a third-generation equine infectious anemia virus (EIAV) in the Brattleboro rat model of diabetes insipidus, in which a mutation in the arginine vasopressin (AVP) gene results in the production of nonfunctional mutant AVP precursor protein. Importantly, by using this model it is possible to monitor the success of the gene therapy treatment by noninvasive assays. Injection of an EIAV-CMV-AVP vector into the supraoptic nuclei of the hypothalamus resulted in expression of functional AVP peptide in magnocellular neurons. This was accompanied by a 100% recovery in water homeostasis as assessed by daily water intake, urine production, and urine osmolality lasting for a 1-year measurement period. These data show that a single gene defect leading to a neurological disorder can be corrected with a lentiviral-based strategy. This study highlights the potential of using viral gene therapy for the long-term treatment of disorders of the CNS.