Temperature-sensitive (ts) recombinants of influenza A virus were evaluated for use in a live virus vaccine. Evidence from several sources suggested that the ts lesions were responsible for attenuation of these mutants. Specification of attenuation by defined genetic lesions which can be assayed for in the laboratory offers an advantage to the use of ts viruses for vaccination. This means that ts recombinants can be assessed for genetic stability during vaccine development, production and later during usage in man. One ts virus, influenza A/Hong Kong/68-ts-1[E], with a 38°C shut-off temperature, had the following properties desirable for a live virus vaccine: (1) satisfactory infectivity for seronegative (serum HI antibody titre ≤ 1:8) adults; (2) satisfactory attenuation for adults; (3) capacity to stimulate local and serum anti-haemagglutinin and anti-neuraminidase antibodies in seronegative volunteers; (4) stimulation of resistance to virulent, wild type virus; (5) relative genetic stability in vivo; (6) lack of communicability in man; (7) replication to high titre in avian leucosis virus-free eggs; and (8) localization of ts lesions to genes that do not code for the haemagglutinin and neuraminidase. The ts lesions of influenza A/Hong Kong/68-ts-1[E] virus were transferred to more current viruses within the H3N2 subtype (influenza A/Udorn/307/72 and influenza A/Georgia/101/74). These recombinant Udorn/72 and Georgia/74 ts viruses, which possessed the same shut-off temperature and the same ts lesions as the influenza A/Hong Kong/68-ts-1[E] parent virus, exhibited a pattern of infection and attenuation in hamsters and man similar to their ts parent. These data suggest that ts mutants which are sufficiently attenuated for man, could serve as donors of ts lesions for the rapid production of an attenuated vaccine when new antigenic variants arise.
When the influenza A/Hong Kong/68-ts-1[E] virus was administered to children who lacked both anti-haemagglutinin and anti-neuraminidase antibody the virus replicated for a longer period than in adults and mild fever developed in some of the young vaccinees. A minority of children shed wild type revertant virus. The emergence of wild type virus in children, but not in adults, probably reflected the more extensive replication of the virus in the doubly seronegative children. The implications of these findings to the development and testing of live influenza A virus vaccines were discussed.