The interruption of HIV transmission from mother to child is important. Prevention strategies, including antiretroviral agents administered to the mother and/or child, can successfully prevent such transmission. Avoidance of breastfeeding or the administration of antiretroviral agents to the mother while she continues breastfeeding is another strategy. Based on the successful model of hepatitis B prevention by treatment of the newborn, research protocols have been designed to prevent mother-to-child transmission of HIV by the administration of passive HIV-specific antibody preparations and HIV vaccines to the mother and/or child. A number of animal studies using active and passive products have shown efficacy in similar settings, providing hope that a similar strategy will work in humans. Clinical trials conducted through the US National Institutes of Health AIDS Clinical Trials Programs have focused on such an approach. This article summarizes the results of a number of these trials, including AVEG (AIDS Vaccine Evaluation Group) 104, a phase I study of active immunization of HIV-infected pregnant women with an HIV gp120 subunit vaccine, and PACTG (Pediatrics AIDS Clinical Trial Group) 185, a phase III efficacy trial of HIV immunoglobulin administered to HIV-infected pregnant women and their newborns. A number of HIV vaccine trials have been performed in HIV-exposed and -infected children. These include PACTG 218, a vaccine immunotherapy phase I trial of HIV subunit vaccines administered to asymptomatic HIV-infected children, and two phase I vaccine trials in HIV-exposed children, PACTG 230 and 326. While the results of PACTG 230 were encouraging, the gp120 subunit vaccines were not capable of generating cytotoxic T-cell (CTL) responses. A subsequent phase I study, PACTG 326, utilized a canarypox vectored HIV vaccine (ALVAC vCP205, Sanofi Pasteur), which was previously shown to generate CTL responses in HIV-uninfected adults. The vaccines were safe but the immunogenicity was poor when compared with results of adult studies. Specifically, no antibody responses were found, lymphoproliferative responses to HIV-specific antigens were found in <50% of vaccinees, and CTL responses, modest in nature, were seen in approximately 50% of vaccinees.Future planned vaccine studies are focused on prime-boost approaches, using live recombinant vectors or DNA vaccines combined with subunit vaccines to stimulate both cellular and antibody responses. HIV vaccines may have special utility in newborns, in infants who continue to breastfeed, and in pre-adolescent children before they become sexually active. However, to date, candidate HIV vaccines capable of generating robust immunologic responses in these populations have been disappointing.