In 1994, the ACTG-076 protocol demonstrated that zidovudine prophylaxis during pregnancy reduced the vertical transmission of HIV from 22.6% to 7.6% . Since then, the additive prophylactic effects of lower maternal viral load, elective caesarean section and bottle feeding have reduced the risk of newborn infection to less than 2% [2–4]. Although the use of antiretroviral regimens during pregnancy is becoming common, few data are available for estimating in utero drug exposure. Some human trials indicate an excellent placental passage (newborn:mother blood concentration ratio) for zidovudine (0.85) and lamivudine (∼1.0) , nevirapine (NVP; 0.9)  and to a lesser extent for didanosine (0.35) . However, less is known for the protease inhibitors (PI). To date, most information originates from animal experiments (indinavir, saquinavir, ritonavir) or from human isolated placental perfusion models (ritonavir , saquinavir ). However, these data may not reflect the actual human fetal exposure to drugs. Several elements influence this exposure: the drug concentration profile in the maternal circulation, the kinetics of drug transfer across the placenta, and the placental and fetal drug metabolism. These elements depend, each in turn, on many related factors: gastrointestinal function, maternal protein level, total body mass, fat storage, cardiac output, glomerular filtration rate, hepatic metabolism, genetic and environmental influences on cytochrome P-450 enzymes expression, P-glycoprotein activity in both the digestive tract and the fetal barrier, etc. [10–12]. Recently, some human data, presented in abstract form, have shown extremely low cord blood levels for nelfinavir, indinavir, ritonavir and saquinavir [13–14]; however, no corresponding values for the mothers were given. Therefore, the extent of placental transfer in human has never been properly assessed .
This lack of data for PI prompted the present study, to evaluate the transplacental passage of PI drugs in order to assess in utero fetal drug exposure.
The Swiss Mother + Child HIV Cohort Study (MoCHiV) is a prospective survey with continuing enrolment of HIV-positive pregnant women and their newborn children. The MoCHiV, a substudy of the Swiss HIV Cohort Study, was formed in 1999 following a merger of the Swiss Neonatal HIV Study and the Swiss Collaborative HIV and Pregnancy Study, which had been active since 1986 and 1989, respectively [3,15]. This study was approved by the local Ethics Committees.
Eligible women, receiving PI and/or NVP, were enrolled in the protocol after giving their written informed consent just prior to elective delivery, performed by caesarean section at 38 ± 1 week of gestation. Two blood samples (10 ml) were collected into lithium heparin Monovettes (Sarstedt, Nümbrecht, Germany), one drawn from the mother within 15 min before delivery and one from umbilical cord (artery or vein) after cord section at delivery. The time separating these two samples was as short as possible (mean ± SD: 15 ± 12 min). Plasma was isolated by centrifugation, viro-inactivated in a waterbath at 60°C for 60 min and stored at −20°C until analysis. Plasma PI levels were determined by reverse-phase high-performance liquid chromatography (HPLC) according to a validated method , enabling the simultaneous quantification of various HIV PI drugs (limit of quantification 250 ng/ml for nelfinavir, ritonavir and lopinavir, and 100 ng/ml for saquinavir). NVP was analysed by a separate validated HPLC method . Each blood sample was provided with the exact sampling time and the hours of the last three antiretroviral dose intakes before delivery. This information allowed the exclusion of any bias owing to missing doses and to relate cord-to-mother blood ratios with the interval after maternal dosing.
Viral load was measured with Amplicor test (level of detection 400 copies/ml or < 20 copies/ml; Roche Diagnostics, Basel, Switzerland). The mother's virological status was evaluated by recording the last viral load value measured within 2 months before delivery. Detection of p24 antigen and HIV RNA in newborns was performed at 1, 2 and 6 months.
Twenty pregnant women aged 30 ± 4 years were included in the study (Table 1). Among them, seven were excluded from analysis for the following reasons: in four women, the time interval between maternal dosing and delivery was > 19 h, resulting in undetectable drug levels in both the mother's blood and the cord blood; in two women the time of the last drug intake was not available, and in one woman, the sample was inadvertently diluted. For two patients receiving drug combinations, only one drug could be assessed for placental transfer as the dosing intervals were too prolonged for the second drug.
The most frequently prescribed drug regimen during pregnancy was a combination of nelfinavir with zidovudine and lamivudine (15 patients). Two patients were taking two PI drugs with one or two nucleoside reverse transcriptase inhibitors and two were taking one PI with NVP and one or two nucleoside reverse transcriptase inhibitors. All these regimen were administered twice daily. As elective caesarean section was usually performed early in the morning, most samples were obtained at trough dosing interval. When available, viral load values measured within 2 months before delivery indicated that HIV was well controlled for most women (< 400 copies/ml). Viral load was elevated in one woman with a history of illicit drug use.
Among the 13 measurable maternal–cord blood sample pairs, transplacental passage was assessed for nelfinavir (n = 9), ritonavir (n = 2), saquinavir (n = 1), lopinavir (n = 1) and NVP (n = 2). The median cord (range) and maternal (range) concentrations, respectively, were nelfinavir < 250 (< 250–1005) and 1110 (505–2765), ritonavir < 250 (< 250) and 1113 (1095–1130), saquinavir < 100 and 350, and lopinavir < 250 and 3105 ng/ml and NVP 2072 (1983–2160) and 2546 (2432–2660). As cord PI concentrations were mostly under the limit of quantification, cord-to-mother ratio was 0–0.3 for nelfinavir, 0–0.2 for ritonavir, 0–0.3 for saquinavir, and 0–0.1 for lopinavir. The calculated ratio was 0.8 for NVP (Fig. 1). The ratio values were not affected by postdosing time intervals.
The PI and NVP concentrations measured at delivery in this particular population of pregnant women were globally lower than the median concentrations observed in a reference HIV-infected population at similar sampling times [18–20] (Table 1).
Results of infant HIV-1 serology, performed 1, 2, and 6 months after birth, were all negative.
In the present study, there was no evidence of HIV transmission to the newborns. Results of transplacental passage indicated that PI (nelfinavir, ritonavir, saquinavir and lopinavir) undergo incomplete transplacental transfer : they do not cross the placenta to an appreciable extent as cord concentrations were in general under the limit of quantification. Consequently, the fetal drug concentrations may not remain over the dose giving 50% inhibition (IC50) during the entire dosing interval. Most women presented good viral suppression during pregnancy; therefore, the extremely low PI cord concentrations cannot be explained by poor maternal adherence but must rather be a consequence of low placental transfer for this class of drug. Many factors may independently limit this passage: PI are highly bound to protein (> 98%, except indinavir ~60%) and are substrates for the efflux transporter pump P-glycoprotein, largely expressed in the placenta . In P-glycoprotein-deficient (mdr-1 knockout) mice, the fetal penetration of indinavir and saquinavir was, in fact, considerably increased (two and sevenfold, respectively) . The presence of P-glycoprotein in the placenta limits the exposure of the fetus to many xenobiotics, thus preventing their potential teratogenic action during embryonic and fetal development . Besides the backwards efflux through P-glycoprotein, a potential influence of placental metabolism on the antiretroviral drugs should be also considered, even though its relevance has not been established .
The transplacental ratio of NVP, observed in our study at maternal steady state, was 0.8 and comparable to the value reported after a single dose of NVP . Patient 20 exemplifies well the marked difference of placental transfer between PI and NVP. In this patient, the transfer of lopinavir was essentially zero even though lopinavir concentration measured simultaneously in the maternal circulation was satisfactory. On the contrary, the transfer of NVP was quite extensive. Several factors may account for NVP passage: the drug has a lower degree of protein binding (60%), a lower molecular weight (266), a relatively favourable degree of lipophilicity (log octanol:water partition coefficient 1.81)  and is reportedly not a substrate of P-glycoprotein . NVP is largely used in countries with limited resources in a single-dose schedule at the onset of labour  and active concentrations are maintained in the newborn during the first week of life because of the prolonged half-life of the drug [6,23]. This favourable profile may argue for its prescription during the third trimester of pregnancy. Its use may, however, be limited by the risk of hepatotoxicity, occurring mainly during the first 12 weeks of therapy  and with a higher incidence in women .
The results of HIV-1 detection test available in the newborns of the study (18/18 HIV negative) suggest that PI-containing regimens protect adequately against vertical transmission at delivery. This is probably a result of the reduction of maternal viral load. Moreover, PI were always combined with other antiretroviral drugs known to cross the placenta (zidovudine, lamivudine, NVP) and, therefore, contributing to fetal protection [1,5,6,26].
The maternal concentrations of PI and NVP measured at delivery tended to be lower than those observed in a general HIV-infected population. The decrease in plasma levels may result from the non-specific stimulation of general metabolism and enzyme expression levels during pregnancy . Since a good virological control in the mother is important to reduce the risk of vertical transmission , the low maternal PI concentration observed during pregnancy should prompt monitoring of drug levels in this special population. Cases of decreased exposure to nelfinavir resulting in virological failure during pregnancy have been recently observed in our centre and in another clinic .
In conclusion, the PI drugs do not cross the placenta to an appreciable extent, and are, therefore, unlikely to exert any direct protection for the newborn; their role is mainly to reduce the maternal viral load during pregnancy and they should probably not be used as single-dose (or limited-dose) peripartum prophylaxis. In contrast, the placental transfer of NVP, like nucleoside reverse transcriptase inhibitors in general, appears sufficient to ensure suppressive blood levels in the fetus at delivery time.
We are indebted to Dr Katja Wolf, Dr Irene Hug, and Dr Michal Repa for recruitment of patients and clinical data support and to Alexandre Beguin for excellent technical assistance in setting up the NVP analytical method.
1. Connor EM, Sperling ES, Gelber R. et al
. Reduction of maternal–infant transmission of human immunodeficiency virus type 1 with zidovudine treatment. N Engl J Med 1994, 331: 1173–1180.
2. Thea DM, Steketee RW, Pliver V. et al
. The effect of maternal viral load on the risk of perinatal transmission of HIV-1
. AIDS 1997, 11: 437–444.
3. Kind C, Rudin C, Siegrist CA. et al
. Prevention of vertical HIV transmission: additive protective effect of elective Cesarean section and zidovudine prophylaxis. AIDS 1998, 12: 205–210.
4. European Collaborative Study. Maternal viral load and vertical transmission of HIV-1
: an important factor but not the only one. AIDS 1999, 13: 1377–1385.
5. Moodley J, Moodley D, Pillay K. et al
. Pharmacokinetics and antiretroviral activity of lamivudine alone or when coadministered with zidovudine in human immunodeficiency virus type 1–infected pregnant women and their offspring. J Infect Dis 1998, 178: 1327–1333.
6. Mirochnik M, Fenton T, Gagnier P. et al
. Pharmacokinetics of nevirapine in human imunodeficiency virus type 1–infected pregnant women and their neonates.Pediatric AIDS Clinical Trial Group Protocol 250 Team.
J Infect Dis 1998, 178: 368–374.
7. Wang Y, Livingston E, Patil S. et al
. Pharmacokinetics of didanosine in antepartum and postpartum human immunodeficiency virus-infected pregnant women and their neonates: an AIDS Clinical Trials Group study. J Infect Dis 1999, 180: 1536–1541.
8. Casey BM, Bawdon RE. Placental transfer of ritonavir with zidovudine in the ex vivo placental perfusion model. Am J Obstet Gynecol 1998, 179: 758–761.
9. Peytavin G, Lamotte C, Mandelbrot L et al.Placental transfer of saquinavir studied in the human in vitro placental perfusion model.Seventh Conference on Retroviruses and Opportunistic Infections
. San Francisco, January–February 2000 [abstract 714].
10. Mirochnik M. Antiretroviral pharmacology in pregnant women and their newborns. Ann NY Acad Sci 2000, 918: 287–297.
11. Lobstein R, Lalkin A, Koren G. Pharmacokinetic changes during pregnancy and their clinical relevance. Clin Pharmacokinet 1997, 33: 328–343.
12. Huisman MT, Smit JW, Schinkel AH. Significance of P-glycoprotein for the pharmacology and clinical use of protease inhibitors
. AIDS 2000, 14: 237–242.
13. Bryson Y, Stek A, Mirochnick M et al.PACTG 353: a phase I study of safety, pharmacokinetics, and antiviral activity of combination nelfinavir, zidovudine, and 3TC in HIV-infected pregnant women and their infants.Seventh Conference on Retroviruses and Opportunistic Infections
. San Francisco, January–February 2000 [abstract 715].
14. Mirochnick M, Dorenbaum A, Cunningham-Schrader B et al.Cord blood protease inhibitor (PI) concentrations in infants born to mothers receiving PIs.Eighth Conference on Retroviruses and Opportunistic Infections
. Chicago, February 2001 [abstract 710].
15. The European Collaborative Study and the Swiss Mother plus Child HIV Cohort Study. Combination antiretroviral therapy and duration of pregnancy. AIDS 2000, 14: 2913–2920.
16. Marzolini C, Telenti A, Buclin T, Biollaz J, Decosterd LA. Simultaneous determination of the HIV protease inhibitors
indinavir, amprenavir, saquinavir, ritonavir, nelfinavir and the non-nucleoside reverse transcriptase inhibitor efavirenz by HPLC after solid-phase extraction. J Chromatogr B 2000, 740: 43–58.
17. Marzolini C, Beguin A, Telenti A et al
. Determination of lopinavir and nevirapine by HPLC after solid-phase extraction: application for the assessment of their transplacental passage at delivery
. J Chromatogr B Biomed Appl
2002, in press.
18. Marzolini C, Greub G, Decosterd LA, Biollaz J, Telenti A, Buclin T. Routine antiretroviral plasma levels of NNRTIs and PIs correlate with viral suppression and adverse events.Eighth Conference on Retroviruses and Opportunistic Infections
. Chicago, February 2001 [abstract 735].
19. Havlir D, Cheeseman SH, Mc Laughlin M. et al
. High-dose nevirapine
: safety, pharmacokinetics, and antiviral effect in patients with human immunodeficiency virus infection. J Infect Dis 1995, 171: 537–545.
20. Hurst M, Faulds D. Lopinavir. Drugs 2000, 60: 1371–1379.
21. Pacifici GM, Nottoli R. Placental transfer of drugs administered to the mother. Clin Pharmacokinet 1995, 28: 235–269.
22. Smit JW, Huisman MT, van Tellingen O, Wiltshire HR, Schinkel AH. Absence or pharmacological blocking of placental P-glycoprotein profoundly increases fetal drug exposure. J Clin Invest 1999, 104: 1441–1447.
23. Musoke P, Guay LA, Bagenda D. et al
. A phase I/II study of the safety and pharmacokinetics of nevirapine
–infected pregnant Ugandan women and their neonates (HIVNET 006). AIDS 1999, 13: 479–486.
24. Yazdanian M. Blood–brain barrier properties of human immunodeficiency virus antiretrovirals. J Pharm Sci 1999, 88: 950–954.
25. Greenblatt DJ, Störmer E, von Molke LL. Effects of non-nucleoside reverse transcriptase inhibitors on P-glycoprotein in tissue culture. Clin Pharm Ther 2001, 69: 16.16.
26. Guay LA, Musoke P, Fleming T. et al
. Intrapartum and neonatal single-dose nevirapine
compared with zidovudine for prevention of mother-to-child transmission of HIV-1
in Kampala, Uganda: HIVNET 012 randomised trial. Lancet 1999, 354: 795–802.
27. Martinez E, Blanco JL, Arnaiz JA. et al
. Hepatoxicity in HIV-1
–infected patients receiving nevirapine
-containing antiretroviral therapy. AIDS 2001, 15: 1261–1268.
28. Barlett J. Severe liver toxicity in patients receiving two nucleoside analogues and a non-nucleoside reverse transcriptase inhibitor.Eighth Conference on Retroviruses and Opportunistic Infections
. Chicago, February 2001 [abstract 19].
29. Angel BJ, Khaliq Y, Monpetit ML, Cameron W, Gallicano K. An argument for routin therapeutic drug monitoring of HIV-1 protease inhibitors
during pregnancy. AIDS 2001, 15: 417–419.
Members of the Swiss HIV Cohort Study and the Swiss Mother and Child HIV Study: C. Aebi, M. Battegay (Chairman of the SHCS Scientific Board), M.-C. Bernard, E. Bernasconi, K. Biedermann, H. Bucher, Ph. Bürgisser, J.-J. Cheseaux, G. Drack, M. Egger, P. Erb, W. Fierz, M. Flepp (Chairman of the SHCS Clinical and Laboratory Committee), P. Francioli (President of the SCHS, Centre Hospitalier Universitaire Vaudois, CH-1011–Lausanne), H. J. Furrer, M. Gorgievski, H. Günthard, P. Grob, T. Gyr, B. Hirschel, I. Hösli, O. Irion, K. Keller, C. Kind (Chairman of the MoCHiV Scientific Board Subcommittee), Th. Klimkait, U. Lauper (Chairman of the MoCHiV Clinical and Laboratory Subcommittee), B. Lederberger, D. Nadal, M. Opravil, F. Paccaud, G. Pantaleo, L. Perrin, J.-C. Piffaretti, M. Rickenbach, C. Rudin (Chairman of the MoCHiV Substudy, Basel UKBB, Römergasse 8, CH-4058 Basel), A. Schreyer, J. Schüpbach, A. Telenti, P. Vernazza, Th. Wagels, R. Weber, A. Wechsler, D. Wunder and C.-A. Wyler.