Pharmacokinetics and placental transfer of dolutegravir in pregnancy Free

Abstract

Dolutegravir is currently recommended by the WHO as the preferred first-line treatment for all people with HIV, including pregnant women. Estimates indicate that, by 2024, nearly 22 million adults in low- and middle-income countries will have transitioned to dolutegravir-based ART. It is therefore critical that there is a clear appreciation and understanding of the risks that may be associated with in utero exposure to dolutegravir. In this review we consolidate data from studies on dolutegravir and the placenta. The studies have largely focused on the pharmacokinetics and placental transfer of dolutegravir in pregnancy. These include studies on transplacental transfer of dolutegravir, ex vivo placenta perfusion models, physiologically based pharmacokinetic (PBPK) models and animal studies. The data available clearly demonstrate that placental transfer of dolutegravir occurs in moderate to high concentrations. Intracellular placental dolutegravir has been demonstrated in the placental villous tissue. There are limited data suggesting that pregnancy is associated with decreased maternal dolutegravir levels. In addition, PBPK models have great potential in predicting the passage of drugs through the placenta and further contributing towards the elucidation of fetal exposure. The animal studies available demonstrate that in utero dolutegravir exposure can be associated with neural tube defects. Taking into consideration that antiretroviral exposure may be associated with poor placental development or function and increased risk of adverse effects to the fetus, it is crucially important that these risks are evaluated, especially with the rapid scale up of dolutegravir-based ART into national treatment programmes.

Introduction

The HIV epidemic continues to disproportionately affect women. In 2019, of the 28.2 million adults aged 15–49 years old living with HIV, 15.6 million were women and, of these, 12.5 million were in sub-Saharan Africa.¹ With the high rate of both planned and unplanned pregnancies within the region,² there has been a significant scale up of access to ART for the prevention of vertical transmission of HIV. Dolutegravir is an integrase strand transfer inhibitor that was first approved by the US FDA in 2013 and is currently recommended by the WHO as the preferred first-line treatment for all people with HIV, including pregnant women.^3,4 In women of childbearing potential who wish to become pregnant, the WHO recommends that dolutegravir-based ART is initiated or continued during pregnancy for women who identify pregnancy after their first trimester.⁵

Most ART regimens include a backbone of two NRTIs; therefore, dolutegravir is recommended at a standard dosing of 50 mg once daily for pregnant and non-pregnant individuals, as a triple fixed-dose combination of tenofovir disoproxil fumarate, lamivudine or emtricitabine and dolutegravir and tenofovir disoproxil fumarate, lamivudine and efavirenz as the alternative first-line therapy.⁵ Dolutegravir has a favourable adverse effect profile, has few drug–drug interactions and confers a low risk of developing resistance, reducing the need for second-line PI-based therapy.⁶ In addition, with generic versions of the drug now available to ensure affordability, the roll-out of dolutegravir-based ART continues worldwide. By 2019, 82 low- and middle-income countries (LMICs) had transitioned to dolutegravir-based ART, providing access to nearly 7 million people living with HIV.^4,7 It is estimated that, by 2024, nearly 22 million adults in LMICs will have transitioned to dolutegravir-based ART.⁸

Studies have shown that placental transfer of dolutegravir occurs in moderate to high concentrations, with cord plasma/maternal plasma ratios ranging from 0.6 to 1.38.^9–12 The transplacental transfer of ART is an unwarranted consequence of maternal treatment, because an ideal antiretroviral (ARV) regimen would optimize maternal viral suppression while eliminating the risk of vertical transmission without any detrimental effects on the fetus. In cases of late initiation of ART during pregnancy, transplacental transfer of ARVs can be beneficial to the HIV-exposed uninfected (HEU) infant, by providing pre-exposure prophylaxis (PrEP), in addition to reducing maternal viral load to reduce the risk of vertical transmission. Data relating to the transplacental and fetal effects of dolutegravir-based ART regimens remain sparse, highlighting the need for a clearer understanding of the effects of dolutegravir-based ART on the placenta and fetus, as this will inform potential risks associated with in utero exposure.

In this review, we aim to consolidate data on existing human and animal studies on dolutegravir and the placenta. The studies available have examined the pharmacokinetics and placental transfer of dolutegravir, including ex vivo placental perfusion models, physiologically based pharmacokinetic (PBPK) models and studies of transplacental transfer of dolutegravir, including data on intracellular placental dolutegravir.

Role of the placenta in linking dolutegravir exposure to birth outcomes

The importance of investigating the role of dolutegravir-exposed placentae in mediating adverse birth outcomes can be justified by findings from previous studies on placentae from pregnant women with HIV on ART. ART regimens including PIs have been associated with poor placental vascular development and impaired decidualization.^13,14 We recently published data showing that timing of ART initiation can be associated with maternal vascular malperfusion (MVM) and adverse birth outcomes, including preterm birth (PTB), small for gestational age (SGA) and low birth weight (LBW) deliveries.¹⁵ ART regimens may also alter the delicate placental angiogenic balance, which can underlie poor placental development and functioning, resulting in poor health outcomes in the infant.^16–19 It is therefore important to understand how the placenta links in utero dolutegravir exposure to the risk of poor pregnancy, birth and infant health outcomes.

The initial dolutegravir preclinical reproductive toxicology studies on rats and rabbits found no dolutegravir-related adverse effects on fertility and embryonic or fetal development.²⁰ There have also been numerous observational clinical studies evaluating the safety of dolutegravir during pregnancy and these findings have been consolidated in extensive reviews.^21–26 The data available show no evidence of increased risk of adverse birth outcomes associated with dolutegravir use during pregnancy, including PTB, SGA, stillbirths and congenital anomalies.^22,24,26 Perhaps the most concerning remains the potential association of peri-conceptional dolutegravir exposure with neural tube defects (NTDs). Preliminary analysis from the clinical study Tsepamo²⁷ demonstrated a potential association of NTDs with the use of dolutegravir at conception, prompting a note of caution with the initial recommendation of dolutegravir use for women of childbearing age. However, in 2019, WHO downgraded the risk of NTDs, following new evidence demonstrating that the benefits of dolutegravir use (which include a favourable adverse effect profile, a low risk of developing drug resistance and rapid reduction of viral load) substantially outweighed the risks.^6,28,29

Most of the studies that have investigated the effect of dolutegravir on pregnancy or birth outcomes do not include data on placental investigations in relation to birth outcomes. This therefore limits the understanding of the potential placental origin of any reported outcomes in pregnant women with HIV on dolutegravir (including PTB, SGA, stillbirths and congenital anomalies). We recently published findings from a sub-study of the Dolutegravir in Pregnant HIV Mothers and Their Neonates (DolPHIN)-2 trial, which was designed to assess the efficacy and safety of dolutegravir compared with efavirenz in a cohort of women initiating ART in late pregnancy.³⁰ The study investigated the T cell profile and histopathology in placentae from the parent study and reported no dolutegravir-specific associated effects on the placental T cell landscape or pathology.³⁰ The above studies demonstrate there are limited data on the subject, warranting further research on dolutegravir-exposed placentae and associated birth outcomes. In the next section, we discuss studies that have examined the pharmacokinetics and placental transfer of dolutegravir.

Ex vivo perfusion of term placentae

Until recently, a number of clinical trials in pregnant women were considered ethically difficult. Therefore, alternative preclinical approaches have been considered to obtain data on ARVs in/across the placenta and subsequent fetal exposure. One such approach is the use of an ex vivo human placenta perfusion model, which can be used to study and interpret the transplacental transfer of drugs.^31,32 This model closely resembles the situation in vivo in terms of structural and cellular integrity. To yield the best results, perfusions are commenced immediately after delivery on fresh placentae.³³ An intact placental cotyledon is selected and then a fetal chorionic plate artery and its associated vein are cannulated. The intervillous space on the maternal side below the decidual surface is also perfused. Perfusates are often prepared in balanced salt solutions and the drug is then introduced into the system after an initial control period where baseline parameters of different measures are obtained.³¹ Maternal and fetal samples are collected at intervals over a period of time, until the perfusion model equilibrates, and the concentration of the drug is determined. The results from the perfusion model can then be compared with in vivo cord blood/maternal blood drug concentration ratios if available and these data provide information on how much of the drug crosses the placenta.

To investigate transplacental transfer of dolutegravir, Schalkwijk et al.¹⁰ used term placentae collected after uncomplicated pregnancies from women without HIV and HCV. The study was designed to provide active drug concentrations equivalent to the standard dolutegravir dosing. The authors demonstrated transplacental transfer of dolutegravir with a fetal-to-maternal ratio of 0.6 after 3 h of perfusion. Mandelbrot et al.³⁴ used a similar experimental design, on term placentae collected after uncomplicated pregnancies from women without HIV, HBV and HCV. The study was also designed to provide active drug concentrations equivalent to the standard dosing. Here, the authors demonstrated transplacental transfer of dolutegravir with a fetal-to-maternal ratio of 0.34 after 3 h of perfusion. Both these studies were able to provide evidence for transplacental transfer of dolutegravir, which is important for fetal PrEP. This provides a baseline for further investigations on the impact of in utero exposure on fetal development and any associated toxicity-related studies in HEU infants.

Dual perfusion of placental tissue ex vivo is an experimental model that has been used to not only study transplacental transfer of drugs but also endogenous and exogenous substrates, such as hormones, amino acids and viruses.^35–37 One key limitation, however, is that the model cannot fully mimic physiological conditions, as there are numerous endogenous factors that could influence transplacental transfer. The addition of proteins to the perfusates, for example albumin, in dolutegravir perfusion experiments can be considered, because dolutegravir is highly bound to plasma proteins, especially albumin.³⁸ An additional limitation of the perfusion model is the use of mostly term placentae, whose data may not be extrapolated to earlier gestational ages, when placental function may differ. Lastly, there are now data demonstrating that ART can alter vascular development in the placentae^13,14 and therefore the use of healthy placentae may not be representative of the microenvironment in vivo. Despite these limitations, when applied appropriately and taking into consideration placental pharmacokinetic factors as well as physiochemical properties of the drug, the perfusion model can be an invaluable tool to help assess the benefits versus the risks of drugs administered during pregnancy.

Pharmacokinetic models of dolutegravir exposure

An alternative approach to overcoming the limitation of evaluating transplacental transfer of ARV regimens at term, as discussed in the previous section, involves including data for the pregnant patient in a broader population PBPK model. PBPK models combine data on human anatomy, physiology and drug-specific parameters to simulate concentration–time profiles in plasma and various tissues.^32,39 They can therefore be used to also predict fetal exposure levels following maternal dosing, as there is a delicate balance between beneficial versus adverse effects of fetal PrEP in utero.

Freriksen et al.⁴⁰ combined a pregnancy PBPK modelling approach with data on placental drug transfer, based on ex vivo placental perfusion experiments. The aim was to simulate maternal and fetal exposure to dolutegravir. Simulations were performed following the oral administration of the standard dose of dolutegravir (50 mg once daily) to a virtual pregnant woman at 34 weeks of gestation. Interestingly, simulated cord plasma/maternal plasma concentrations ranged between 0.57 and 1.51, and these data were comparable to clinical data available at the time from the Pharmacokinetic Properties of Antiretroviral and Related Drugs During Pregnancy and Postpartum (IMPAACT P1026s) study,⁴¹ from the European Pharmacokinetics of Antiretroviral Agents in HIV-Infected Pregnant Women (PANNA) study⁴² and from Rimawi et al.¹¹ Here, the integration of data from the ex vivo perfusion placental experiments into the models allowed the study of maternal and fetal exposure to dolutegravir with accurate parameterization based on human physiology.

The limitation here, which is discussed by the authors, is the use of term placentae for the ex vivo perfusion experiments compared with pharmacokinetic data from women with HIV at 34 weeks of gestation. In addition, as mentioned in the previous section, the use of healthy term placentae may not be a representative model for pregnant women with HIV. Despite this, the integration of data from ex vivo experiments, for example placenta perfusion data, into PBPK models has great potential to enable the prediction of the passage of drugs through the placenta and further contributing towards the elucidation of fetal exposure following maternal dosing of dolutegravir. Future studies should also take into account drug metabolism and include experimental data on intracellular unbound dolutegravir.

Dolutegravir concentration ratios in maternal plasma, placenta and cord blood

In clinical settings, the transplacental transfer of ARV regimens is often determined by measuring maternal plasma and cord plasma concentrations at the time of delivery. The in-between, which is the intracellular concentration in placental cells, remains under-investigated. Rimawi et al.¹¹ evaluated the pharmacokinetics and placental transfer of 50 mg of dolutegravir administered within a combined ARV regimen in maternal blood, cord blood and placental samples from pregnant women with HIV and showed that dolutegravir readily crossed the placenta into the fetal circulation. The authors reported on three placentae from the dolutegravir group and were able to detect dolutegravir in the placental villous tissue and reported a cord plasma/maternal plasma ratio of 1.06. However, they also reported a cord plasma/placenta ratio of 19.7, suggesting that dolutegravir accumulates in the placenta, although the authors did not discuss this further and additional studies would be warranted to examine the accumulation of dolutegravir in the placenta. There are additional studies that have evaluated dolutegravir pharmacokinetics and reported on maternal blood and cord blood plasma concentration ratios, but with no data on placental tissue concentrations. However, despite the lack of data on intracellular placental dolutegravir, these studies provide additional insight into transplacental transfer of dolutegravir and are thus discussed here.

The IMPAACT study evaluated dolutegravir pharmacokinetics with standard dosing administration (50 mg once daily) in the maternal blood and cord blood, but not in the placenta.⁴¹ The study reported a median cord plasma/maternal plasma concentration ratio of 1.25. The authors further showed that pregnancy was associated with a decrease in dolutegravir maternal levels, particularly in the second and third trimester compared with post-partum, but that the lower exposure levels were still sufficient to suppress viral replication. The postulation here is that decreased exposure may be explained by a progesterone-induced increase in enzyme UDP-glucuronosyltransferase 1A1 (UGT1A1) expression, resulting in a net increase in metabolism. UGT1A1 is an enzyme that is found in a number of anatomical sites, including the liver and placenta, and is known to participate in the metabolism and detoxification of clinical drugs and xenobiotics.^43,44 Dolutegravir metabolism is primarily dictated by UGT1A1 activity and progesterone, a hormone that is essential for the maintenance of pregnancy, has been shown to induce UGT1A1 expression.^43,45

The DolPHIN-1 study enrolled ART-naive pregnant women with HIV who were initiated on dolutegravir during the third trimester and also demonstrated high in utero transfer of dolutegravir.¹² This is important, because late initiation of ART in pregnancy, after 28 weeks of gestation, has been associated with a 7-fold higher risk of vertical transmission of HIV.⁴⁶ Here, the authors reported a cord plasma/maternal plasma ratio of 1.21.^11,41 In contrast to the IMPAACT study,⁴¹ pregnancy did not appear to alter the drug exposure levels in DolPHIN-1, although the time of sampling was different; DolPHIN-1 was earlier during the post-partum period at a median of 10 days post-delivery compared with 6–12 weeks post-delivery in the IMPAACT study. Therefore, maternal physiology may have not yet been restored to a non-pregnant state in the women in DolPHIN-1. More recently, the European PANNA study⁴² reported a cord plasma/maternal plasma ratio of 1.38, similar to other reports.^11,12,41 The post-partum period was 6 weeks after delivery and the dolutegravir plasma concentrations were similar between the third trimester and post-partum.

Cord plasma/maternal plasma ratios provide data at a single timepoint—at term. This is clinically useful; however, there is likely to be variability between last-dose administration and delivery. Placental pharmacokinetic studies investigating the transplacental transfer of dolutegravir are therefore important, as they allow for quantification of drug passage across the placenta, which can provide insight into fetal exposure and the resultant fetal systemic drug levels that could potentially have adverse effects on the infant.^47,48 Here, drug metabolism comes into play, because the placenta is involved in the metabolism of certain drugs. For dolutegravir, the enzyme UGT1A1, which dictates the metabolism of the drug, has been shown to be expressed and active in third-trimester placental villous tissue.^45,49

Knowledge of drug transporter expression and function is also necessary to completely understand dolutegravir absorption, distribution, elimination and effects. For example, dolutegravir is a substrate for the efflux transporters human breast cancer resistance protein (BCRP) and P-glycoprotein (P-gp), both of which have been found in the placenta.^50,51 BCRP is expressed on the basolateral membranes and fetal blood cells.⁵² P-gp is expressed on cytotrophoblasts and on the apical membrane of syncytiotrophoblasts.⁵² Transporters on apical membranes are thought to selectively allow substrates to the fetus and this may have a protective role, by excluding harmful substances.⁵¹ As reviewed in detail by Han et al.,⁵² the expression of these drug transporters changes with gestational age. P-gp has been shown to decrease with advancing gestation and this is postulated to be important to protect the fetus during the developmentally sensitive first-trimester period.^53,54 For BCRP, the data have been inconsistent, with studies showing an increase, a decrease or no change in expression levels during gestation.^54–56

Pregnancy physiology, particularly hormonal changes and infections, may also alter drug metabolism.^21,57 UGT1A1 and CYP3A4, which are enzymes involved in dolutegravir metabolism, are increased during pregnancy.^58,59 P-gp and BCRP expression increase with an increase in progesterone levels in trophoblast cell lines.^54,55,60 Animal studies have also shown that treatment of trophoblast cells with cytokines or simulated infections results in decreased P-gp and BCRP expression.^61,62 In addition, P-gp and BCRP levels are decreased in placentae from SGA preterm deliveries and in preeclampsia when compared with uncomplicated pregnancies.^63,64 Overall, the evaluation of placental dolutegravir pharmacokinetics integrating data on drug metabolism can provide a deeper understanding, while exploring the aspect of fetal exposure following maternal dosing, with a particular focus on pre-conception and first-trimester use of dolutegravir when the risk of NTDs is highest.

Animal studies

Animal studies enable the investigation of in utero ART exposure in vivo at multiple timepoints during gestation with variations in regimen doses and can therefore provide preclinical information regarding toxicity.⁶⁵ An additional advantage of animal models is the ability to perform mechanistic studies, which are not possible in humans. With dolutegravir, the initial preclinical reproductive toxicology studies on rats and rabbits found no dolutegravir-related adverse effects on fertility and embryonic or fetal development.²⁰ These included pre- and post-implantation loss, fetal resorption, spontaneous abortions and fetal malformations.²⁰ However, in these studies, dolutegravir was tested as a single drug regimen, which is not representative of combination ART. In addition, the animals were exposed to dolutegravir during specific times in pregnancy, providing a limited window of dolutegravir exposure during gestation.

Recently, a large prospective mouse dolutegravir fetotoxicity study evaluated gross fetal anomalies in pregnant mice on dolutegravir-based ART.⁶⁶ Doses were administered to the animals to yield therapeutic or supra-therapeutic dolutegravir levels. The authors reported a 2-fold increase in fetal anomalies as well as lower fetal and placental weights in the mice that received therapeutic levels of dolutegravir compared with controls (water). The mice in the supra-therapeutic dolutegravir arm had higher fetal and placental weights, no fetal growth restriction and anomalies similar to the control group. There were five NTDs reported, all in the therapeutic-level dolutegravir arm, and this NTD rate (0.47%) was similar to those reported in the Tsepamo study in the 2019 surveillance study (0.3%) and later in 2020 (0.19%).^27,29

NTDs have been associated with reduced folate (vitamin B9) levels and pregnant women and women of childbearing age are advised by the WHO to take supplemental folic acid.⁶⁷ Disruption of folate uptake or metabolism is a common pathway through which drug regimens exert their teratogenic effects.⁶⁸ In line with this, a recent study by Cabrera et al.⁶⁹ demonstrated that dolutegravir is only a partial antagonist of the folate receptor FOLR1. The authors further showed that dolutegravir-associated developmental toxicity in zebrafish can be rescued by supplemental folic acid. However, in the study, the dolutegravir concentration used in the zebrafish experiments was higher, selected to increase the risk/incidence of developmental toxicity, and was above the reported clinical plasma concentrations. Despite this, dolutegravir-induced zebrafish phenotypes can be explored further and be used more specifically to investigate precise cell populations that underlie dolutegravir-associated pathologies.

Animal models are useful in the study of ARV safety and toxicity during pregnancy. They enable exploration of the underlying mechanisms. However, it is important that the utility and translatability of the animal models is considered. This is especially important with the placenta, as this is a species-specific organ. Therefore, animal models should aim to test clinically relevant drug concentrations based on optimal dosing of the different regimens, with the aim of informing ART safety.

Conclusions

In this review we consolidate data from studies on dolutegravir and the placenta. The studies have largely focused on the pharmacokinetics and placental transfer of dolutegravir in pregnancy. These include studies on transplacental transfer of dolutegravir, ex vivo placenta perfusion models, PBPK models and animal studies. It is not surprising that the data available are quite limited, given that until recently dolutegravir was not the recommended first-line treatment for women of childbearing potential with HIV, owing to the initial concerns about dolutegravir’s safety during pregnancy. The data available clearly demonstrate that placental transfer of dolutegravir occurs in moderate to high concentrations, as demonstrated by perfusion models. However, data are lacking on dolutegravir transport and metabolism in the placenta. There are limited data suggesting that pregnancy is associated with decreased dolutegravir maternal levels; however, this warrants further investigation, particularly because pregnancy physiology, particularly hormonal changes, may alter drug metabolism.^21,57 The data available from animal studies demonstrate that dolutegravir exposure can be associated with NTDs, although at low rates.

There is a need for further studies designed to contribute to a better understanding of placental pharmacokinetic factors as well as physiochemical properties of dolutegravir. In this regard, placenta perfusion models, both ex vivo perfusion models and PBPK models, can be invaluable tools to help assess the benefits versus the risks of drugs administered during pregnancy. These models have great potential to enable the prediction of the passage of dolutegravir through the placenta and further contribute towards the elucidation of fetal exposure following maternal dosing. Studies are also required on the metabolism of dolutegravir, not only systemically but also in the placenta, taking into consideration key metabolic enzymes, such as UGT1A1 and CYP3A4.

Placentae from women with HIV on different ART regimens have been examined and there is evidence of HIV/ART-associated placental injury. Dolutegravir-based studies should also integrate the placental arm of research, including pathology and immunology-based investigations, and how these may link to poor health outcomes in the infant. More recently dolutegravir has been associated with weight gain.^70,71 Maternal obesity during pregnancy is associated with an increased risk of adverse pregnancy and birth outcomes and poor infant health outcomes.^72–74 Every kilo gained above the internationally recognized Institute of Medicine recommendations for gestational weight gain is linked to a 10% increased risk of adverse pregnancy outcomes, including preeclampsia, gestational diabetes, spontaneous pregnancy loss and complications during labour and delivery.^75,76 Therefore, this is an additional avenue worth exploring.

Animal models are useful in the study of ARV safety and toxicity during pregnancy, as they enable exploration of the underlying mechanisms during gestation, which may not be possible to investigate in human pregnancy. However, while this is important, the utility and translatability of the animal models should be considered and the models should aim to test clinically relevant drug concentrations based on optimal dosing of the different regimens, with the aim of informing ART safety.

These efforts will collectively contribute to a better understanding of the effect of dolutegravir-based ART on the placenta. This is of huge public health importance, considering the rapid scale up of dolutegravir-based ART into national treatment programmes.

Funding

N.M.I. is a research associate supported by the National Institute for Health Research (NIHR) (17/63/26) using UK Government aid to support global health research.

Transparency declarations

None to declare.

Disclaimer

The views expressed in this publication are those of the authors and not necessarily those of the NIHR or the UK Department of Health and Social Care.

References