Previous Antiretroviral Therapy for Prevention of Mother-to-Child Transmission of HIV Does not Hamper the Initial Response to PI-Based Multitherapy During Subsequent Pregnancy

Background: Few data are available on the possible long-term negative effects of a short exposure to antiretroviral therapy (ART) for prevention of mother-to-child transmission (PMTCT). Objective: To determine whether ART for PMTCT, discontinued after delivery, affects the virological response to highly active antiretroviral therapy (HAART) administered during subsequent pregnancies. Methods: All current pregnancies of HIV-1-infected women enrolled in the French Perinatal Cohort (ANRS CO-01 EPF) between 2005 and 2009 and not receiving ART at the time of conception were eligible. We studied the association between history of exposure to ART during a previous pregnancy and detectable viral load (VL) under multitherapy at current delivery (VL >= 50 copies/mL). Results: Among 1116 eligible women, 869 were ART naive and 247 had received PMTCT during a previous pregnancy. Previous ART was protease inhibitor (PI)-based HAART in 48%, non-PI-based HAART in 4%, nucleoside reverse transcriptase inhibitor bitherapy in 19% and zidovudine monotherapy in 29% of the women. At current pregnancy, women with or without prior exposure to ART had similar CD4 cell counts and VL before ART initiation. PI-based HAART was initiated in 90% of the women. VL was undetectable (<50 copies/mL) at delivery in 65% of previously ART-naive women, 72% of women previously exposed to HAART, 62% previously exposed to bitherapy, and 67% previously exposed to monotherapy for prophylaxis (P = 0.42). Detectable VL was not associated with previous exposure in multivariate analysis (adjusted OR for previous versus no previous exposure to ART: 0.92; 0.95% confidence interval: 0.59 to 1.44). Conclusions: Efficacy of PI-based combinations is not decreased in women previously exposed to various regimens of antiretroviral PMTCT.