Progressive multifocal leukoencephalopathy (PML) is a rare but frequently fatal disease caused by the uncontrolled replication of JC virus (JCV), a polyomavirus, in the brains of some immunocompromised individuals. Currently, no effective antiviral treatment for this disease has been identified. As a first step in the identification of such therapy, we screened the Spectrum collection of 2,000 approved drugs and biologically active molecules for their anti-JCV activities in an in vitro infection assay. We identified a number of different drugs and compounds that had significant anti-JCV activities at micromolar concentrations and lacked cellular toxicity. Of the compounds with anti-JCV activities, only mefloquine, an antimalarial agent, has been reported to show sufficiently high penetration into the central nervous system such that it would be predicted to achieve efficacious concentrations in the brain. Additional in vitro experiments demonstrated that mefloquine inhibits the viral infection rates of three different JCV isolates, JCV(Mad1), JCV(Mad4), and JCV(M1/SVEDelta), and does so in three different cell types, transformed human glial (SVG-A) cells, primary human fetal glial cells, and primary human astrocytes. Using quantitative PCR to quantify the number of viral copies in cultured cells, we have also shown that mefloquine inhibits viral DNA replication. Finally, we demonstrated that mefloquine does not block viral cell entry; rather, it inhibits viral replication in cells after viral entry. Although no suitable animal model of PML or JCV infection is available for the testing of mefloquine in vivo, our in vitro results, combined with biodistribution data published in the literature, suggest that mefloquine could be an effective therapy for PML.