• Weight-for-height z-scores will be compared in continuous and interrupted therapy arms at every monthly follow-up visits after randomization [ Time Frame: Over 18 months of post-randomization follow-up ] [ Designated as safety issue: No ]
  
• WHO stage of AIDS will be compared in continuous and interrupted therapy arms at every monthly follow-up visits after randomization [ Time Frame: Over 18 months of post-randomization follow-up ] [ Designated as safety issue: No ]
  
• Immunologic status (CD4%) will be compared in continuous and interrupted therapy arms at every 3-monthly follow-up visits after randomization [ Time Frame: Over 18 months of post-randomization follow-up ] [ Designated as safety issue: No ]
  

• Predictors of non-progression of HIV among infants: [ Time Frame: Over 24 months of treatment with HAART and 18 months of post-randomization follow-up ] [ Designated as safety issue: No ]
  
• Adherence to HAART [ Time Frame: Over 24 months of treatment with HAART and 18 months of post-randomization follow-up ] [ Designated as safety issue: No ]
  
• HIV-1 specific immune responses [ Time Frame: Over 24 months of treatment with HAART and 18 months of post-randomization follow-up ] [ Designated as safety issue: No ]
  
• Immune activation [ Time Frame: After 24 months of treatment with HAART and 18 months of post-randomization follow-up ] [ Designated as safety issue: No ]
  
• Correlates of virologic failure [ Time Frame: Over 24 months of treatment with HAART ] [ Designated as safety issue: No ]
  

Hypothesis: Deferring antiretroviral therapy in infants who have immune reconstitution and adequate growth following early therapy of primary infection (initiated HAART during primary infection at less than 4 months of age) will not compromise clinical status or growth and may spare antiretroviral toxicity.

Specific Aim/Primary Objective: To compare growth and morbidity in infants (who initiated HAART during primary infection at less than or equal to 4 months of age with subsequently normalized CD4% and growth following 24 months of HAART) randomized to deferred versus continuous therapy and followed for an additional 18 months.

Secondary Aim/Secondary Objective: To determine predictors of non-progression of HIV among the infants, including: age, adherence, HIV-1 specific immune responses, baseline HIV-1 RNA, CD4 percent and immune activation.

Design: Randomized clinical trial involving HIV-1 treatment of infants (0-4 months old) for 24 months, followed by randomization and 18 months follow-up of children randomized to continued versus deferred treatment. This trial is unblinded.

Population: HIV-1 infected infants (0-4 months) meeting eligibility will be enrolled. After 24 months of treatment follow-up, children with CD4% > 25% and normalized growth will be retained in the study and randomized.

Sample size: 150 infants will be enrolled of which 100 are expected to be eligible for randomization (50 in each arm).

Treatment: All infants will be treated with HAART according to WHO and Kenyan national guidelines. The specific regimens that will be used as a part of this study are:

First line regimen

• AZT/3TC/NVP (zidovudine/lamivudine/nevirapine)
• d4T/3TC/NVP (stavudine/lamivudine/nevirapine)
• AZT/3TC/ABC (zidovudine/lamivudine/abacavir)
• d4T/3TC/ABC (stavudine/lamivudine/abacavir)
• ABC/3TC/NVP (abacavir/lamivudine/nevirapine)

Second line regimen

- ddI/ABC/LPV/r (didanosine/abacavir/lopinavir-ritonavir (Kaletra))

For infants with prior exposure to nevirapine as part of PMTCT:

First line regimen - AZT/3TC/LPV/r (zidovudine/lamivudine/lopinavir-ritonavir (kaletra))

Second line regimen

- ABC/ ddI or TDF / NVP or EFV (abacavir/ didanosine or tenofovir / nevirapine or efavirenz)