RESUMEN
BACKGROUND: Strategies to prevent postnatal mother-to-child transmission of HIV-1 in Africa, including infant prophylaxis, have never been assessed past 6 months of breastfeeding, despite breastfeeding being recommended up to 12 months after birth. We aimed to compare the efficacy and safety of infant prophylaxis with the two drug regimens (lamivudine or lopinavir-ritonavir) to prevent postnatal HIV-1 transmission up to 50 weeks of breastfeeding. METHODS: We did a randomised controlled trial in four sites in Burkina Faso, South Africa, Uganda, and Zambia in children born to HIV-1-infected mothers not eligible for antiretroviral therapy (CD4 count >350 cells per µL). An independent researcher electronically generated a randomisation schedule; we then used sequentially numbered envelopes to randomly assign (1:1) HIV-1-uninfected breastfed infants aged 7 days to either lopinavir-ritonavir or lamivudine (paediatric liquid formulations, twice a day) up to 1 week after complete cessation of breastfeeding or at the final visit at week 50. We stratified the randomisation by country and used permuted blocks of four and six. We used a study label on drug bottles to mask participants, study physicians, and assessors to the treatment allocation. The primary outcome was infant HIV-1 infection between age 7 days and 50 weeks, diagnosed every 3 months with HIV-1 DNA PCR, in the modified intention-to-treat population (all who attended at least one follow-up visit). This trial is registered with ClinicalTrials.gov, number NCT00640263. FINDINGS: Between Nov 16, 2009, and May 7, 2012, we enrolled and randomised 1273 infants and analysed 1236; 615 assigned to lopinavir-ritonavir or 621 assigned to lamivudine. 17 HIV-1 infections were diagnosed in the study period (eight in the lopinavir-ritonavir group and nine in the lamivudine group), resulting in cumulative HIV-1 infection of 1.4% (95% CI 0.4-2.5) and 1.5% (0.7-2.5), respectively. Infection rates did not differ between the two drug regimens (hazard ratio [HR] of lopinavir-ritonavir versus lamivudine of 0.90, 95% CI 0.35-2.34; p=0.83). Clinical and biological severe adverse events did not differ between groups; 251 (51%) infants had a grade 3-4 event in the lopinavir-ritonavir group compared with 246 (50%) in the lamivudine group. INTERPRETATION: Infant HIV-1 prophylaxis with lopinavir-ritonavir was not superior to lamivudine and both drugs led to very low rates of HIV-1 postnatal transmission for up to 50 weeks of breastfeeding. Infant pre-exposure prophylaxis should be extended until the end of HIV-1 exposure and mothers should be informed about the persistent risk of transmission throughout breastfeeding. FUNDING: INSERM/National Agency for Research on AIDS and Viral Hepatitis (including funds from the Total Foundation), European Developing Countries Clinical Trials Partnership, Research Council of Norway.