A potent anti-West Nile virus (anti-WNV)-neutralizing humanized monoclonal antibody, hE16, was previously shown to improve the survival of WNV-infected hamsters when it was administered intraperitoneally (i.p.), even after the virus had infected neurons in the brain. In this study, we evaluated the therapeutic limit of hE16 for the treatment of WNV infection in hamsters by comparing single-dose peripheral (i.p.) therapy with direct administration into the pons through a convection-enhanced delivery (CED) system. At day 5 after infection, treatments with hE16 by the peripheral and the CED routes were equally effective at reducing morbidity and mortality. In contrast, at day 6 only the treatment by the CED route protected the hamsters from lethal infection. These experiments suggest that hE16 can directly control WNV infection in the central nervous system. In support of this, hE16 administered i.p. was detected in a time-dependent manner in the serum, cerebrospinal fluid (CSF), cerebral cortex, brain stem, and spinal cord in CSF. A linear relationship between the hE16 dose and the concentration in serum was observed, and maximal therapeutic activity occurred at doses of 0.32 mg/kg of body weight or higher, which produced serum hE16 concentrations of 1.3 μg/ml or higher. Overall, these data suggest that in hamsters hE16 can ameliorate neurological disease after significant viral replication has occurred, although there is a time window that limits therapeutic efficacy.