In studies with polioviruses (PV) isolated from infected children, Sabin, seeking suitable strains to attenuate for live oral PV vaccine (OPV), derived substrains with exceptionally low virulence for monkeys inoculated intraspinally-the species and route of inoculation most sensitive to residual neurovirulence. Because OPV strains tend to undergo partial genetic reversion, it is important to verify that live vaccine preparations retain all markers of attenuation, especially lack of neurovirulence. For many years US federal regulations have required that each monovalent virus pool be tested for neurovirulence by injection into the spinal cords of monkeys before incorporation into trivalent OPV. Monkey neurovirulence testing (MNVT) is expensive and uses very large numbers of animals-testing for a single lot of trivalent OPV requiring 88 monkeys or more. We have studied three tests that might eventually reduce the reliance on MNVT if not replace it: 1) A molecular assay to detect increased content in PV vaccine of those virions with revertant mutations at sites in the viral genome thought to be responsible for neurovirulence (“mutant analysis by polymerase chain reaction and restriction enzyme cleavage”/MAPREC). 2) Intraspinal inoculation of transgenic (Tg) mice expressing the human PV-receptor gene and susceptible to infection with PV. 3) Increased replication of neurovirulent strains of type-3 PV in cultures of interferon-treated human neuroblastoma cells. MAPREC and Tg-mouse tests successfully detected preparations of type-3 OPV (the least stable type of OPV most often incriminated in vaccine-associated paralytic poliomyelitis) that had failed MNVT. The neuroblastoma test recently showed promise in detecting revertant type-3 OPV. Both MAPREC and Tg-mouse tests are currently subjects of multi-center studies coordinated through the World Health Organization.