Pediatric HIV-1 has a much more aggressive course than adult HIV-1 infection, with an over 50% risk of mortality in the first 2 years of life. It is important to determine ways to optimize pediatric highly active antiretroviral therapy (HAART), specifically in Africa, where -90% of the world's HIV-1 infected children reside. The hypotheses of this study are:1. Among nevirapine-exposed infants without detectable nevirapine resistance on population-based sequencing, nevirapine-containing HAART will be comparable in efficacy to nevirapine-sparing regimens that involve protease inhibitors, which are associated with heat-lability, toxicity, and poor palatability. 2. Sensitive genotypic resistance assays to detect low-level resistance may predict nevirapine-treatment failure. 3. Early HAART during primary infection will prevent infant mortality, restore immune function, and contain viremia, after which antiretroviral treatment can be deferred to later stages of infection, given age-related infant immune maturation and the resulting improved capacity to contain HIV-1; this approach will provide the survival benefits of early HAART without obligating life-long indefinite therapy (associated with cumulative toxicity, resistance, and treatment fatigue). We propose clinical trials among age-stratified HIV-1 infected infants in Nairobi. HIV-1 infected infants (6- 12 mos old)previously exposed to single-dose nevirapine will undergo sequencing to identify nevirapine resistance, after which infants without nevirapine resistance will be randomized to nevirapine-containing vs. nevirapine-sparing HAART (100 per arm) and compared for viral suppression, CD4%, toxicity, and clinical progression during 24 mos follow-up. In this group, genotypic resistance assays to detect low levels of nevirapine-resistance will be conducted to determine whether low-level genotypic resistance will predict treatment failure. Infants aged 0-3 mos old (300) will receive Pi-containing HAART for 24 months, after which those with normal growth and immune reconstitution will be randomized to continued vs. deferred treatment and compared for growth and clinical morbidity in an 18-month period. Children in the deferred arm will be maintained off therapy unless clinical or immune parameters require. Additional correlates, including compliance, age, immune activation markers, and HIV-1 specific immune responses, will be assessed for effect on HIV-1 progression. This combination of studies will provide critical information for improving pediatric HAART strategies in high HIV-1 seroprevalence regions.