Prevention of mother-to-child transmission of HIV type 1 through human breast milk remains a controversially discussed topic. World Health Organization guidelines recommend avoidance of breast-feeding (BF) and use of replacement feeding if it is “acceptable, feasible, affordable, sustainable, and safe.”1 Consequently, in many industrialized countries, BF is strongly discouraged or even prohibited to HIV-positive mothers and a drop of mother-to-child transmission below 2% has been observed in these regions.2,3 In many resource-limited countries, however, formula feeding (FF) encounters major problems, including high costs of the formula diets,4 reduced availability of clean drinking water, no stable supply of electricity, and low acceptance among mothers and family members. It has also been reported that in some resource-limited countries only a minority of women are able to breast-feed their children exclusively for more than 6 months.5 In theses regions, World Health Organization recommends exclusive BF during the first months of life and use of replacement feeding as soon as possible.1 Major advantages of BF versus FF in developing countries include an optimal composition of the milk, adequate water content even in dry regions, and the anti-infectious properties of human milk. As a result, BF is still widespread among HIV-infected women in sub-Saharan countries, and up to 15% of children born to HIV-positive mothers are infected through BF.6
Efavirenz (EFV)7 and NVP are the 2 non-nucleoside reverse transcriptase inhibitors commonly used in first-line highly active antiretroviral therapy (HAART) in resource-limited countries.1 Both are associated with important side effects. NVP has an increased risk of hepatotoxicity for women with a CD4 count of >250 cells/mm3 at treatment initiation.8-10 EFV has been linked to birth defects in animals and to neural tube defects in humans in the case of exposure during the first trimester of pregnancy. Although the prevalence of abnormalities in humans reported in literature is low,11,12 consensus recommendation is that EFV treatment should be avoided during the first trimester of pregnancy.13-15 No data about treatment of newborns with EFV are available; treatment of older children was generally well tolerated, and EFV showed a sustained antiviral effect.16-20
We here report the detection of EFV in mothers' plasma, infants' plasma, and mothers' breast milk in 13 HIV-positive women and their children taking part in the “AMATA” prospective feasibility study in Rwanda. AMATA stands for Allaitement Maternal sons Trithérapie Antirétrovirate and also means “milk” in the local kinyarwanda language. The determination of EFV blood levels of non infected breast-fed children may give important information not only about possible toxicity and side effects due to HAART with EFV but also about the possibility to provide supplementary protection.
All patients were part of the nonrandomized AMATA feasibility study. This study was conducted from May 2005 to November 2007 in Rwanda. It has been approved by the National Ethical Committee and was carried out in collaboration with Rwandan Ministry of Health. All patients had signed a “written informed consent” before enrolling into the study.
The main objective of the AMATA study was to compare exclusive BF with systematic maternal HAART for a maximum duration of 6 months to exclusive FF for the prevention of postnatal mother-to-child transmission of HIV. The study was open to all pregnant women with confirmed HIV infection. Advantages and disadvantages of BF versus FF were explained to the women before delivery. After receiving these information, they had to choose between FF and BF. If not already on HAART, all women received HAART during pregnancy starting from week 28 of gestation irrespective of the study arm or the CD4 cell count: national treatment regimen (D4T-3TC-NVP) if eligible for HAART or AMATA treatment regimen (AZT-3TC-EFV) if not eligible for HAART according to national guidelines (HIV stage 1 and CD4 cell count >350/mm3). If they chose to breast-feed, then they received systematic antiretroviral (ARV) triple therapy during the BF period; otherwise, formula diet was given free of charge. Thirteen women coming for their regular follow-up were chosen randomly for the determination of EFV levels. To be eligible, they had to be outside the perinatal period (>45 days) and in good general health condition (including breast condition). Adherence to the ARV treatment was controlled by questionnaires, pill counts, and education. Determination of the viral load of the breast milk has not been performed.
The mean age of the selected women was 27.8 years (SD: 4.0 years, range 21-34 years). All were in HIV clinical stage 1 and had a mean CD4 count of 631 cells/mm3 (SD: 261 cells/mm3, range 371-1232 cells/mm3). Viral loads were measured at delivery and after 6 months using polymerase chain reaction. Results were below detection level (<40 copies/mm3) or in the low titer region (Table 1). Of the 13 women, 12 received AZT (600 mg/d), 3TC (300 mg/d), and EFV (600 mg/d). One woman received D4T (60 mg/d) instead of AZT. No mother was eligible for HAART after weaning.
The newborns (9 boys, 4 girls) were checked for their HIV status immediately after birth (<48 hours of age), at day 45, at month 3, and at month 7 by DNA-polymerase chain reaction. Newborns received a single dose (2 mg/kg) of NVP in the 72 hours after birth and AZT (2 × 4 mg/kg/d) for 7 days. Physical examination of infants was systematic with notification of weight, height, head circumference, and neurological examination. The mean age of the infants at the moment of blood collection was 15.8 weeks (SD: 6.3 weeks, range: 6-25 weeks).
Sample Collection and Preparation
Blood and milk samples were collected at the health structure during the morning AMATA routine consultation and immediately after BF. All were collected 3-4 hours after the last drug intake. Plasma was obtained by centrifugation, frozen, shipped on dry ice to Luxembourg, and kept at −20°C until analysis. Measurement of CD4 cell counts was performed with a Facscalibur counter (Becton Dickinson, Franklin Lakes, NJ). HIV-1 levels were determined using Amplicor Monitor Assay (Version 1.5, Roche Molecular Systems, Pleasanton, CA). Sample preparation and operating conditions of the gas chromatography/mass spectrometry (GC/MS) system have been described earlier.21 About 40 mL of maternal milk was manually and cleanly extracted (disinfection of breast and use of sterile gloves). Milk was stored in sterile vials and kept in an icebox before being frozen at −20°C (maximum 3 hours after sampling). Acetonitrile and hexobarbital (internal standard) were added to the milk and calibration samples. Each solution was vortexed and centrifuged for 5 minutes of 5000 rpm. The supernatant (skim milk) was removed and used in the extraction steps. Extraction of free EFV consisted of 2 liquid-liquid steps followed by a solid phase extraction as described earlier.21 GC/MS operating conditions were the same as for plasma samples. Validation of the plasma samples has been described previously.21 Validation for the skim milk was performed using the same parameters and acceptation limits as for plasma samples. Statistical evaluation of the results was performed using GraphStat software (Version 3.06).
Concentrations of EFV in plasma samples and in skim milk are given in Table 1. Using Kolmogorov-Smirnov test, a Gaussian distribution of the 3 groups of results (skim milk, maternal, and infant plasma) was verified. The mean concentration of EFV was 6.55 mg/L (median 6.03 mg/L, range 1.62-14.43 mg/L, SD: 3.12 mg/L) in maternal plasma, 3.51 mg/L (median 3.45 mg/L, range 1.33-7.44 mg/L, SD: 1.72 mg/L) in skim milk, and 0.86 mg/L (median 0.87 mg/L, range 0.31-1.51 mg/L, SD: 0.41 mg/L) in infant plasma.
EFV concentrations were significantly higher in maternal plasma than in skim milk and higher in skim milk than in infant plasma (Wilcoxon signed ranked test, P = 0.0005 and P = 0.0002, respectively). The ratios were 0.54 (mean skim milk to mean maternal plasma concentration, M/P) and 4.08, respectively (mean skim milk to mean newborn plasma concentration). The newborn plasma concentration of EFV was an average 13.1% (range: 5.6%-26.8%) of the maternal plasma concentration.
Significant linear correlation between maternal plasma and skim milk (correlation coefficient r = 0.8666, P < 0.0001, Pearson correlation) and between infant plasma and skim milk (r = 0.6646, P < 0.02, Pearson correlation) was observed. However, no significant linear correlation was observed between maternal and infant plasma concentrations (r = 0.5441, P > 0.05) (Fig. 1).
After 6 months of BF, no child out of the 13 included in this report has been infected with HIV and no clinical side effects due to HAART with EFV treatment of their mothers have been detected. All children show a normal intellectual, psychomotorical (following with eyes, holding head), and growth (length, weight, and head circumference) evolution.
The mean concentration of EFV in maternal plasma was 61% above the Cmax measured in HIV-positive patients receiving a daily dose of 600 mg of EFV.22 This may be explained by parameters such as nutritional status, age, enzymatic activity, and genetic specificity of the local population. Ribaudo et al,23 for example, have shown a slower clearance rate of EFV in black and Hispanic patients than in white patients.
The M/P ratio of EFV is not documented in the literature, and M/P ratios of drugs often have high intra-and interindividual variability.24 Several antagonistic processes may explain the measured ratio of 0.54: Diffusion of EFV into milk is favored by its lipophilic properties but disfavored by a relatively high molecular weight (315 g/mol), a high pKa (10.2), and a high volume of distribution (252 L for a 600-mg dose).
EFV concentration in the infant plasma compared with the breast milk depends on elimination efficiency and protein binding in the newborn and on EFV protein binding in the milk. Several authors reported lower than expected area under the curve (AUC) values in young children treated with doses based on pharmacokinetic data of older children and concluded that these dose definitions may not be applicable.25,26 EFV is mainly metabolized in the liver by the CYP3A4 and CYP2B6 enzymes. No data about CYP2B6 increase or decrease during the post partum period or its activity in newborns have been published.27 Some studies imply that the liver from infants at 1 month of age possesses only 30% of the CYP3A4 activity associated with adult livers.28 This activity seems to approach adult levels by 6-12 months of age.29 It has also been reported that total plasma protein-binding levels are lower in newborns than in adults30 and that human serum albumin concentrations are only 75%-80% of adult levels. If these results suggest that EFV in newborns should be relatively high, then it is also known that pharmacokinetic processes in children undergo important changes during their growth, making ab initio estimation of the free versus bound ratio of EFV very difficult.31 No data are available about the binding interactions of breast milk proteins and EFV.
Giuliano et al32 have reported that triple therapy (AZT, 3TC and NVP) given during pregnancy and after delivery significantly reduces breast milk viral load, thus reducing viral exposure through breast milk in breast-fed newborns. Of the 13 newborns, 8 in this study have EFV concentrations below the lower therapeutic level recommended for adults. In case of HIV infection, subtherapeutic EFV levels may lead to development of resistance. However, no data about the minimum EFV concentrations necessary for effective protection in neonates for prevention of mother-to-child transmission have been published. Further studies on larger groups of women and children will be necessary to clarify if breast milk concentrations of EFV are sufficient to inhibit viral replication and if the dose given to the infants through breast milk is sufficient for protection against virus present in breast milk.
EFV passes into breast milk and is present in the blood of breast-fed infants. Mean infant plasma concentrations of EFV were 13.1% of maternal plasma concentrations and slightly below the lower therapeutic level generally considered necessary for effective treatment in adults. Because of the low number of patients included in this study, it was not possible to draw conclusions about protective effects or health risks for the children associated with the antiretroviral therapy treatment of the mothers. However, considering the good health status of the mothers and of the infants, our results suggest that EFV may be an alternative to NVP to women in the third trimester of pregnancy and during BF. No data about EFV in neonates are currently available, so additional studies on a larger group of patients will be necessary for a better understanding of long-term benefits and risks of EFV administration to neonates through breast milk.
The authors thank Dr Annemie Otto for helpful discussion and Dr Marc Schumann for review of this manuscript.
2. Palombi L, Marazzi M, Voetberg A, et al. Treatment acceleration program and the experience of the DREAM program in prevention of mother-to-child transmission of HIV. AIDS
3. Suksomboon N, Poolsup N, Ket-Aim S. Systematic review of the efficacy of antiretroviral therapies for reducing the risk of mother-to-child transmission of HIV infection. J Clin Pharm Ther
4. Ogundele M, Coulter J. HIV transmission through breastfeeding: problems and prevention. Ann Trop Paediatr
5. Magoni M, Okong P, Bassani L, et al. Mode of infant feeding and HIV infection in children in a program for prevention of mother-to-child transmission in Uganda. AIDS
6. UNAIDS. 2006 Report on the Global AIDS Epidemic
. Geneva, Switzerland: 2006.
7. Vrouenraets SM, Wit FW, van Tongeren J, et al. Efavirenz: a review. Expert Opin Pharmacother
8. Lyons F, Hopkins S, Kelleher B, et al. Maternal hepatotoxicity with nevirapine as part of combination antiretroviral therapy in pregnancy. HIV Med
9. Rivero A, Mira JA, Pineda JA. Liver toxicity induced by non-nucleoside reverse transcriptase inhibitors. J Antimicrob Chemother
10. Shibuyama S, Gevorkyan A, Yoo U, et al. Understanding and avoiding antiretroviral adverse events. Curr Pharm Des
11. Jeantils V, Khuong M, Delassus J, et al. Efavirenz (Sustiva) in pregnancy: a study about 12 HIV patients. Gynecol Obstet Fertil
12. Bussmann H, Wester CW, Wester CN, et al. Pregnancy rates and birth outcomes among women on efavirenz-containing highly active antiretroviral therapy in Botswana. J Acquir Immune Defic Syndr
13. Waters L, John L, Nelson M. Non-nucleoside reverse transcriptase inhibitors: a review. Int J Clin Pract
14. Watts D. Treating HIV during pregnancy: an update on safety issues. Drug Saf
15. Thorne C, Newell ML. The safety of antiretroviral drugs in pregnancy. Expert Opin Drug Saf
16. Funk MB, Notheis G, Schuster T, et al. Effect of first line therapy including efavirenz and two nucleoside reverse transcriptase inhibitors in HIV-infected children. Eur J Med Res
17. Engelhorn C, Hoffmann F, Notheis G, et al. Effect of antiretroviral combination therapies including efavirenz in heavily pretreated HIV-infected children. Eur J Med Res
18. Starr SE, Fletcher CV, Spector SA, et al. Combination therapy with efavirenz, nelfinavir, and nucleoside reverse-transcriptase inhibitors in children infected with human immunodeficiency virus type 1. Pediatric AIDS Clinical Trials Group 382 Team. N Engl J Med
19. Wamalwa DC, Farquhar C, Obimbo EM, et al. Early response to highly active antiretroviral therapy in HIV-1-infected Kenyan children. J Acquir Immune Defic Syndr
20. Scherpbier HJ, Bekker V, Pajkrt D, et al. Once-daily highly active antiretroviral therapy for HIV-infected children: Safety and efficacy of an efavirenz-containing regimen. Pediatrics
21. Lemmer P, Schneider S, Schuman M, et al. Determination of nevirapine and efavirenz in plasma using GC/MS in selected ion monitoring mode. Ther Drug Monit
22. EMEA. Stocrin, European Public Assessment Report, Annexe 1. Available at: http://www.emea
.europa.eu/humandocs/PDFs/EPAR/Stocrin/H-250-PI-fr.pdf. Accessed February 20, 2008.
23. Ribaudo H, Clifford D, Gulick R, et al. Relationships between efavirenz pharmacokinetics, side effects, drug discontinuation, virologic response, and race: results from ACTG A5095/A5097s. In: 11th Conference on Retroviruses and Opportunistic Infections
. San Francisco, CA; 2004.
24. Anderson G. Using pharmacokinetics to predict the effects of pregnancy and maternal-infant transfer of drugs during lactation. Expert Opin Drug Metab Toxicol
25. Brundage R, Fletcher C, Fenton T. Efavirenz (EFV) and nelfinavir (NFV) pharmacokinetics (PK) in HIV-infected children under 2 years of age. 7th Conference on Retroviruses and Opportunistic Infections
. San Francisco, CA; 2000.
26. Mirochnick M, Stek A, Acevedo M, et al. Safety and pharmacokinetics of nelfinavir coadministered with zidovudine and lamivudine in infants during the first 6 weeks of life. J Acquir Immune Defic Syndr
27. Mirochnick M, Capparelli E. Pharmacokinetics of antiretrovirals in pregnant women. Clin Pharmacokinet
28. Pearce R, Gotschall R, Kearns G, et al. Cytochrome P450 involvement in the biotransformation of cisapride and racemic norcisapride in vitro: differential activity of individual human CYP3A isoforms. Drug Metab Dispos
29. Lacroix D, Sonnier M, Moncion A, et al. Expression of CYP3A in the human liver-evidence that the shift between CYP3A7 and CYP3A4 occurs immediately after birth. Eur J Biochem
30. Alcorn J, McNamara P. Pharmacokinetics in the newborn. Adv Drug Deliv Rev
31. Back D, Gatti G, Fletcher C, et al. Therapeutic drug monitoring in HIV infection. AIDS
32. Giuliano M, Guidotti G, Andreotti M, et al. Triple antiretroviral prophylaxis administered during pregnancy and after delivery significantly reduces breast milk viral load: a study within the Drug Resource Enhancement Against AIDS and Malnutrition Program. J Acquir Immune Defic Syndr