Pre-exposure Prophylaxis Use by Breastfeeding HIV-Uninfected Women: A Prospective Short-Term Study of Antiretroviral Excretion in Breast Milk and Infant Absorption

Abstract

Background: As pre-exposure prophylaxis (PrEP) becomes more widely used in heterosexual populations, an important consideration is its safety in infants who are breastfed by women taking PrEP. We investigated whether tenofovir and emtricitabine are excreted into breast milk and then absorbed by the breastfeeding infant in clinically significant concentrations when used as PrEP by lactating women.

Methods and findings: We conducted a prospective short-term, open-label study of daily oral emtricitabine-tenofovir disoproxil fumarate PrEP among 50 HIV-uninfected breastfeeding African mother-infant pairs between 1-24 wk postpartum (ClinicalTrials.gov Identifier: NCT02776748). The primary goal was to quantify the steady-state concentrations of tenofovir and emtricitabine in infant plasma ingested via breastfeeding. PrEP was administered to women through daily directly observed therapy (DOT) for ten consecutive days and then discontinued thereafter. Non-fasting peak and trough samples of maternal plasma and breast milk were obtained at drug concentration steady states on days 7 and 10, and a single infant plasma sample was obtained on day 7. Peak blood and breast milk samples were obtained 1-2 h after the maternal DOT PrEP dose, while maternal trough samples were obtained at the end of the PrEP dosing interval (i.e., 23 to 24 h) after maternal DOT PrEP dose. Tenofovir and emtricitabine concentrations were quantified using liquid chromatography-tandem mass spectrometry (LC-MS/MS) assays. Of the 50 mother-infant pairs enrolled, 48% were ≤12 wk and 52% were 13-24 wk postpartum, and median maternal age was 25 y (interquartile range [IQR] 22-28). During study follow-up, the median (IQR) daily reported frequency of infant breastfeeding was 15 times (12 to 18) overall, 16 (14 to 19) for the ≤12 weeks, and 14 (12 to 17) for the 13-24 wk infant age groups. Overall, median (IQR) time-averaged peak concentrations in breast milk were 3.2 ng/mL (2.3 to 4.7) for tenofovir and 212.5 ng/mL (140.0 to 405.0) for emtricitabine. Similarly, median (IQR) time-averaged trough concentrations in breast milk were 3.3 ng/mL (2.3 to 4.4) for tenofovir and 183.0 ng/mL (113.0 to 250.0) for emtricitabine, reflecting trough-to-peak breast milk concentration ratios of 1.0 for tenofovir and 0.8 for emtricitabine, respectively. In infant plasma, tenofovir was unquantifiable in 46/49 samples (94%), but emtricitabine was detectable in 47/49 (96%) (median [IQR] concentration: 13.2 ng/mL [9.3 to 16.7]). The estimated equivalent doses an infant would ingest daily from breastfeeding were 0.47 μg/kg (IQR 0.35 to 0.71) for tenofovir and 31.9 μg/kg (IQR 21.0 to 60.8) for emtricitabine, translating into a <0.01% and 0.5% relative dose when compared to the 6 mg/kg dose that is proposed for therapeutic treatment of infant HIV infection and for prevention of infant postnatal HIV infection; a dose that has not shown safety concerns. No serious adverse effects were recorded during study follow-up. The key study limitation was that only a single infant sample was collected to minimize venipunctures for the children. However, maternal daily DOT and specimen collection at drug concentration steady state provided an adequate approach to address the key research question. Importantly, there was minimal variation in breast milk concentrations of tenofovir and emtricitabine (respective median trough-to-peak concentration ratio ~1), demonstrating that infants were exposed to consistent drug dosing via breast milk.

Conclusion: In this short-term study of daily directly observed oral PrEP in HIV-uninfected breastfeeding women, the estimated infant doses from breast milk and resultant infant plasma concentrations for tenofovir and emtricitabine were 12,500 and >200-fold lower than the respective proposed infant therapeutic doses, and tenofovir was not detected in 94% of infant plasma samples. These data suggest that PrEP can be safely used during breastfeeding with minimal infant drug exposure.

Trial registration: ClinicalTrials.gov, Identifier: NCT02776748.

Fig 1. Box plot of maternal and infant tenofovir concentrations.

Non-fasting maternal blood and breast milk samples were obtained concurrently (i.e., within 30 min) at the seventh and tenth visits (corresponding to seventh and tenth maternal DOT PrEP doses). A single infant blood sample was obtained after the seventh maternal DOT PrEP dose. Peak maternal blood, breast milk, and infant blood samples were obtained a median (IQR) of 63 (60 to 68), 70 (65 to 77), and 80 (45 to 90) min after the maternal DOT PrEP dose, respectively. Trough samples were obtained at close of the dosing interval, a median of 23 h (IQR range 23 to 24) after the previous maternal DOT PrEP. One outlier peak maternal plasma tenofovir concentration (1,040 ng/ml) was out of the assay analytic range (0.31–1,000.0 ng/mL). This record was imputed to the upper limit of the assay analytic range and was included in the computation of the displayed summary estimate. Middle box line represents the median. Upper box line represents the 75th percentile and the lower box line represents the 25th percentile. The top whisker denotes the maximum data value or the third quartile plus 1.5 times the interquartile range, whichever is smaller. The lower whisker denotes the minimum data value or the third quartile plus 1.5 times the interquartile range, whichever is larger. The notches display the 95% confidence interval around the median. Small circles represent outlier data points (i.e., observations that are as extreme as ±1.5 of interquartile range). Only 3 of 49 infant plasma samples had quantifiable tenofovir concentration in plasma (infants aged 11 and 13 wk [both 0.9 ng/mL] and 17 wks [17.4 ng/mL]). NA, not applicable; BLQ, below assay limit of quantification for tenofovir: <0.31 ng/mL in plasma and <1 ng/mL in whole milk.

Fig 2. Box plot of maternal and infant emtricitabine concentrations.

Non-fasting maternal blood and breast milk samples were obtained concurrently (i.e., within 30 min) at the seventh and tenth visits (corresponding to seventh and tenth maternal DOT PrEP doses). A single infant blood sample was obtained after the seventh maternal DOT PrEP dose. Peak maternal blood, breast milk, and infant blood samples were obtained a median (IQR) of 63 (60 to 68), 70 (65 to 77), and 80 (45 to 90) min after the maternal DOT PrEP dose, respectively. Trough samples were obtained at close of the dosing interval, a median of 23 h (IQR range 23 to 24) after the previous maternal DOT PrEP. Middle box line represents the median. Upper box line represents the 75th percentile, and the lower box line represents the 25th percentile. The top whisker denotes the maximum data value or the third quartile plus 1.5 times the interquartile range, whichever is smaller. The lower whisker denotes the minimum data value or the third quartile plus 1.5 times the interquartile range, whichever is larger. The notches display the 95% confidence interval around the median. Small circles represent outlier data points (i.e., observations that are as extreme as ±1.5 of interquartile range). NA, not applicable; BLQ, below assay limit of quantification for emtricitabine: <0.31 ng/mL in plasma and <5 ng/mL in whole milk.