Background:  While it has been shown that zidovudine (ZDV), lamivudine (3TC), and nevirapine (NVP) are present in the breast milk of lactating mothers receiving HAART, limited data describe ART concentrations in their nursing infants as a result of transfer via breast milk.

Methods:  The Kisumu Breastfeeding Study is a phase IIb open-label trial of prenatal, intra-partum, and post-partum maternal treatment from 34 weeks’ gestation to 6 months’ post-partum with ZDV, 3TC, and NVP for reduction of mother-to-child HIV transmission. Maternal plasma, breast milk, and infant dried blood spots (DBS) were collected on multiple study visits from consecutively recruited participants. These samples were assayed for ZDV concentrations by enzyme immunoassay (EIA) and for 3TC and NVP concentrations by liquid chromatography-mass spectrometry. Assay limits of quantitation for DBS, breast milk, and plasma samples were 30, 5, and 5 ng/mL for ZDV; 16, 17, and 7 ng/mL for 3TC; and 43, 17, and 40 ng/mL for NVP.

Results:  We assayed 45 plasma, 35 breast milk, and 42 DBS samples obtained longitudinally from 15 mother-infant pairs. ART DBS concentrations were quantifiable for ZDV in 7 of 42 (16.7%) infants, for 3TC in 38 of 42 (90.5%), and for NVP in 38 of 42 (90.5%). Median (IQR 25^th to 75^th) ARY concentrations (ng/mL) are presented in the table.

Conclusions:  Infant drug exposure from breast milk will be determined by the magnitude of breast milk drug delivery and by infant drug absorption, distribution, and elimination. For ZDV, maternal plasma and breast milk concentrations were low and infant concentrations nearly always below the quantitation limit, consistent with a short plasma half-life in both mother and infant. 3TC was concentrated in breast milk and has a longer plasma half-life than ZDV; median infant 3TC concentrations exceeded the mean 3TC trough concentrations of 40 ng/mL seen in adults with once-daily dosing. NVP breast milk concentrations were slightly below those in maternal plasma.  NVP plasma half-life is long, and in some infants NVP concentration exceeded the target concentrations of 3400 ng/mL used in therapeutic drug monitoring programs. These nursing infants whose mothers were receiving HAART demonstrated low ZDV concentrations but 3TC and NVP concentrations of sufficient magnitude to have potential biologic effects such as drug toxicity or, in those infants that do become HIV-infected, partial suppression of HIV replication or development of drug resistance.

bql = below the quantitation limit