In the pre-partum phase the use of antiretroviral therapy for the mother during the last trimester of pregnancy is mandatory. The use of HAART during pregnancy, usually two nucleosides analogues and a protease inhibitor exposes the mother and the child to cumulate toxicities related to both families. The aim of this study is to assess the use of a boosted protease inhibitor without nucleoside analogue during the pre-partum phase for women with no indication of antiretroviral therapy for their own.

Recent data from the French perinatal cohort and others indicate that HIV-RNA levels at delivery correlate with risk of transmission among women treated with antiretroviral agents. Most of these treatments include zidovudine alone or in combination. Mitochondrial toxicity related to nucleoside analogues exposure (zidovudine and lamivudine) has been reported in adults and in infants with in utero exposure to these drugs. In addition, biological markers of genotoxicity on nuclear DNA have recently been shown in exposed newborn. These issues raised the concern of the risk/benefit of multiple therapy in the context of mother to child transmission for women who do not meet the standard criteria for antiretroviral therapy. In women with CD4≥350 and VL<30 000 copies/ml a treatment with lopinavir/ritonavir should achieve a rapid control of HIV1 viremia below 1000 copies/ml without harm in term of resistance. In this study we would like to assess under strict control, the safety and efficacy of such regimen compared to the same boosted PI + zidovudine and lamivudine as standard regimen. The treatment will start at 26 weeks of gestation, and the follow up will include safety and efficacy parameters as well as pharmacokinetics in plasma and genital tract for the women, blood/cord ratio, testing for ARV resistance. Women will stop their treatment after delivery. Infants will be closely monitored up to 24 months with HIV DNA and HIV.RNA-PCR for HIV testing and biochemical and haematology usual safety evaluation. In addition frozen samples will be collected for specific evaluation of nucleoside analogue foetal mitochondrial and nuclear DNA interactions.

In term of transmission safety, the end point would be to reach a viral load below 200 copies after 8 weeks of treatment. In case of failure, this would allow a sufficient delay for a treatment modification: i.e. addition of NRTI and an elective caesarian could be programmed.

Inclusion Criteria: Assessed between 20 and 24 months of pregnancy

• Pregnancy known before 24 weeks of gestation
• Documented HIV-1 infection without indication for ARV therapy
• CD4 count above or equal to 350 per mm3
• VL under 30 000 copies per ml
• Naïve for PI (except treatment during previous pregnancy)
• Informed consent signed

Exclusion Criteria:

• HIV2 infection or HIV1 group O infection
• Any pathology related to pregnancy
• Contra-indication to study drugs
• Unstable hypertension or diabetes
• Known risk of premature delivery
• In case of previous treatment with a protease inhibitor : presence of resistance mutations on the HIV-1 protease gene by genotyping analysis (1 mutation among V32I et I47A, I50V V82A/F/S/T, I84V, L90 M or more than 3 mutations among L10 F/I/R/V, K20/M/R, L24I, L33F, M46I/L, F53L, I54M/L/T/V, L63P, A71L/V/T,)