Safety review of tenofovir disoproxil fumarate/emtricitabine pre-exposure prophylaxis for pregnant women at risk of HIV infection

Abstract

Introduction:

Pregnancy is a period of elevated HIV risk in high-burden settings, motivating the need for prevention tools that are both safe for use and effective during pregnancy. Oral pre-exposure prophylaxis (PrEP) containing tenofovir disoproxil fumarate (TDF) is recommended by the World Health Organization, including for pregnant and postpartum women at substantial risk of HIV infection. Although TDF use during pregnancy appears generally safe, data on PrEP use during pregnancy remain limited.

Areas covered:

We provide an overview of the clinical pharmacology and efficacy of daily TDF-based PrEP and summarize current evidence on the safety of PrEP use by pregnant HIV-uninfected women. We synthesize relevant studies assessing pregnancy outcomes among pregnant women who are living with HIV (WLHIV) and using TDF-based therapy. Finally, we make comparison to the safety profiles of other emerging HIV prevention options.

Expert opinion:

The current evidence indicates that TDF/FTC PrEP use is not associated with increased risk of adverse pregnancy and early infant growth outcomes. While safety data are generally reassuring, there is need for continued accrual of data on growth and pregnancy outcomes in PrEP research, implementation projects, and controlled pharmacokinetic studies to support current evidence and to understand concentration-efficacy relationship in pregnant women.

1. Introduction

For women living in high HIV prevalence settings, pregnancy is a period of elevated HIV risk with estimated average HIV incidence of 3.6 infections per 100 person-years among pregnant and postpartum women in sub-Saharan Africa [1]. Longitudinal data demonstrate a nearly two-fold increase in HIV risk for pregnant women in HIV-serodiscordant relationships and the per-condomless-sex-act probability of acquiring HIV rises steadily throughout pregnancy and into the post-partum period [2,3]. These findings underscore the critical need for safe and effective HIV prevention options for pregnant women.

In 2012, a fixed-dose combination oral tablet of co-formulated emtricitabine (FTC) 200 mg and tenofovir disoproxil fumarate (TDF) 300 mg (Truvada, Gilead Sciences) became the first medication approved for HIV prevention by the US Food and Drug Administration after efficacy was demonstrated in multiple clinical trials [4,5]. Global scale-up of oral TDF/FTC for HIV pre-exposure prophylaxis (PrEP) has expanded rapidly. By the end of 2020, Truvada had been registered in over 40 countries and nearly 800,000 individuals were enrolled in PrEP programs around the world [6]. Though some countries include recommendations for PrEP use during pregnancy in their national HIV/AIDs guidelines, the paucity of safety data is still an impediment to PrEP integration into national antenatal and postnatal care in many settings [7]. In this review, we summarize the current safety data on TDF-based PrEP among pregnant women and highlight key evidence gaps.

2. Mechanism of action and clinical pharmacology

TDF is a prodrug of tenofovir (TFV), an antiviral agent with potent anti-HIV and anti-hepatitis B virus (HBV) activity (Box 1) [8,9]. As an acyclic nucleoside phosphonate analogue of adenosine 5’-monophosphate, TFV inhibits HIV reverse transcriptase activity by competing with the endogenous nucleotide deoxyadenosine 5’-triphosphoate for incorporation into viral DNA. Once incorporated into the viral DNA sequence, TFV then halts further DNA chain elongation due to its lack of a ribose ring [9]. The prodrug TDF, with two additional methyl carbonate esters, was formulated to overcome TFV’s poor oral bioavailability and limited intestinal absorption [10,11]. Upon oral administration, TDF is converted to TFV by removal of the additional ester groups through esterase hydrolysis and cellular phosphorylation to its active metabolite, TFV diphosphate (TFV-DP) [12]. TFV reaches maximum concentrations within one hour after oral dosing and the terminal elimination half-life of ~17 hours and intracellular half-life >60 hours make it conducive to once-daily dosing [13]. Studies among pregnant women living with HIV (WLHIV) show that TFV readily accumulates in amniotic fluid and crosses the placenta [14-16] with umbilical cord TFV concentrations approximately 70-100% of maternal plasma concentrations [17-19]. However, excretion of TFV into breastmilk and subsequent transfer to infants has been shown to be minimal among lactating HIV-uninfected women using TDF/FTC PrEP [17]. Pharmacokinetics studies of TDF-based PrEP among pregnant women have observed significantly lower (~30-40%) plasma TFV and intracellular TFV-DP concentrations during pregnancy versus postpartum or non-pregnancy periods [20,21].

Box 1. Drug Summary.

Drug name	Tenofovir Disoproxil Fumarate (TDF), alone or formulated with emtricitabine (FTC)
Phase	US FDA approval of emtricitabine FTC/-TDF for prevention of HIV infection: 2012
Indication	Prevention and Treatment of HIV infection
Pharmacology description	TDF is an oral prodrug of tenofovir, an acyclic nucleotide (nucleoside monophosphate) analogue with activity against retroviruses, including HIV, and hepadnaviruses. FTC is a synthetic nucleoside analogue of cytidine with activity against HIV reverse transcriptase.
Route of administration	Oral
Chemical structure
Pivotal trial for safety and efficacy against HIV acquisition in heterosexual women	The Partners PrEP Study was a randomized, double-blind, placebo-controlled 3 arm trial of daily TDF alone or in combination with FTC in 4747 serodiscordant heterosexual couples in Kenya and Uganda [Reference 4].

Oral TDF can be coformulated with FTC, a nucleotide reverse transcriptase inhibitor that is also an effective anti-HIV and anti-HBV agent. FTC has a shorter terminal elimination half-life in plasma (10 hours) compared to TFV [13] and is readily transferred across placenta and into breastmilk [14,22]. Both TDF and FTC are eliminated by glomerular filtration and tubular secretion [13]. Use of both TDF and FTC for PrEP are well tolerated, with the most common side effects being diarrhea, nausea, headache and fatigue. TDF/FTC PrEP is associated with small declines in estimated glomerular filtration rate (eGFR) but the risk of clinically relevant declines in eGFR (≥25%) is very rare (<2%) and quickly resolves within weeks of product discontinuation [23]. Small reductions in bone mineral density (BMD) have also been observed with TDF-based PrEP [24-26], with recovery of BMD observed months after PrEP discontinuation [25,27].

3. Clinical application of TDF-based PrEP for HIV prevention among women

The US Centers for Disease Control and Prevention (CDC) and the World Health Organization (WHO) recommend PrEP for individuals at substantial risk for HIV acquisition [28,29], including at-risk pregnant and lactating women. PrEP efficacy was demonstrated in three clinical trials involving women: the Partners PrEP Study, the TDF2 Study, and the Bangkok Tenofovir Study (BTS) [5,17,30]. However, two trials among sub-Saharan Africa women, the Preexposure Prophylaxis Trial for HIV Prevention among African Women (FEM-PrEP) and the Vaginal and Oral Interventions to Control the Epidemic (VOICE) trial, failed to demonstrate PrEP efficacy due to low adherence to the study product [31,32]. The key results from these five trials are summarized in Table 1. Study protocols for these trials required women to discontinue PrEP upon detection of pregnancy. Therefore, relative efficacy of PrEP during periods of pregnancy compared to periods of non-pregnancy could not be evaluated.

Table 1.

Summary of completed PrEP efficacy studies involving cisgender women living without HIV

Study	Location	Participants	No. of pregnancies Reported (no. women)	Main Findings
Partners PrEP Study[ 5 ]	Kenya and Uganda	4747 HIV serodiscordant, heterosexual couples randomized 1:1:1 to TDF/FTC, TDF, placebo pill use (included 1785 couples with women living without HIV)	288 (267 women)	Overall, TDF/FTC was shown to be 75% (95% CI: 55-87%) effective and TDF was 67% (95% CI: 44-81%) effective relative to placebo Effectiveness of 49% (95% CI: −22-81%) among couples where HIV-uninfected partners are women
TDF2 Study [ 30 ]	Botswana	1219 heterosexual men and women randomized 1:1 to TDF/FTC or placebo pill use (included 557 women)	107 (101 women)	The overall efficacy of TDF/FTC demonstrated in the trial was 62% (95% CI: 22-83%). Among women, TDF/FTC was 49% (95% CI: −22-81%) efficacious relative to placebo
Bangkok Tenofovir Study (BTS) [ 17 ]	Thailand	2413 men and women who inject drugs randomized to TDF or placebo pill use (included 489 women)	58 (58 women)	TDF was 49% (95% CI: 10-72%) effective at preventing HIV overall TDF effectiveness of 79% (95% CI: 17-97%) among women (p=0.03)
Vaginal and Oral Interventions to Control the Epidemic (VOICE) [ 32 ]	South Africa, Uganda and Zimbabwe	5029 women randomized to use of oral TDF/FTC, oral TDF, vaginal 1% TFV gel, oral placebo or vaginal placebo	452 (428 women)	TDF was −49% (95% CI: −129-3%) effective and TDF/FTC was −4% (95% CI: −49-27%) effective Adherence to study product was low: plasma TFV was detected in 29-30% of random sample of oral PrEP users
Preexposure Prophylaxis Trial for HIV Prevention among African Women (FEM-PrEP) [ 31 ]	Kenya, South Africa, Tanzania	2120 women randomized to TDF/FTC or placebo pill use	117 (115 women)	Overall TDF effectiveness of 6% (95% CI: −52 to 41%) Adherence to study product was low: plasma TFV concentrations were above target concentrations in <40% of women at seroconverters’ infection windows

TDF: tenofovir disoproxil fumarate, FTC: emtricitabine; TFV: tenofovir; CI: confidence interval

4. Safety evaluation

4.1. HIV uninfected women using TDF-based PrEP during pregnancy

Safety of PrEP in pregnant women was evaluated in four of the major PrEP efficacy trials: the Partners PrEP Study, the TDF2 Study, FEM-PrEP, and VOICE [33-35]. Of 288 pregnancies that occurred during the Partners PrEP Study, the frequencies of pregnancy loss, preterm birth or pregnancy-related complications did not significantly differ between the active PrEP and placebo arms [35]. There were no significant differences in weight, length or head circumference between infants born to women in either arm.

Frequencies of fetal loss was similar between pregnant women assigned to TDF/FTC and placebo pill use in the TDF2 Study [30]. A total of 107 pregnancies from 101 women were reported in the trial.

In the FEM-PrEP trial, 119 total pregnancies from 115 women were observed [34]. Complete data on pregnancy complications were not reported but the authors note that vaginal bleeding was the most common complication and frequency of complications did not differ between active and placebo arms.

The VOICE trial observed 452 pregnancies from 428 women [32]. Report of adverse pregnancy outcomes, including premature birth, stillbirth, spontaneous abortion and ectopic pregnancy, did not differ between women in the active and placebo arms [33]. Subsequent follow-up of infants born to enrolled mothers found no difference in growth indicators in the first year of life between active and placebo arms [36]. However, the VOICE and FEM-PrEP data are limited due to poor adherence to trial product.

Safety data from women who used PrEP throughout pregnancy came from the Partners Demonstration Project, a prospective, open-label evaluation of a combined strategy of PrEP and antiretroviral therapy (ART) use for HIV prevention in serodiscordant couples [37]. Among 30 women who become pregnant and who chose to continue on PrEP during pregnancy, the frequency of live births and adjusted infant weight, length and head circumference z-scores across the first year post-delivery were similar between PrEP users and non-users. No preterm deliveries or congenital abnormalities were observed among women using PrEP [38].

The PrEP Implementation for Young Women and Adolescents (PrIYA) program in Kenya integrated PrEP into maternal and child health and family planning services [39,40]. The frequency of experiencing pre-term birth or low birthweight did not differ and 6-weeks postpartum, growth measures were similar between PrEP-exposed and PrEP-unexposed infants. There was no evidence of difference in frequency of underweight, stunted, or wasting. Among 721 women exposed to PrEP in a sister cluster randomized trial called PrEP Implementation for Mothers in Antenatal Care (PrIMA) [41,42], the frequencies of miscarriage, stillbirth, preterm birth, congenital malformations, low birth weight, or small for gestational age were no different from the 3520 PrEP-unexposed women. Ongoing extended follow-up in this study will provide additional important evidence.

Finally, a recent pharmacokinetic study (IMPAACT 2009) of daily directly observed doses of TDF/FTC for 12 weeks among pregnant and post-partum women in sub-Saharan Africa found that PrEP use was well tolerated and none of the maternal or infant adverse events reported were found to be related to women’s PrEP use [21].

4.2. Safety data from WLHIV using TDF-based HIV treatment during pregnancy

TDF/FTC is indicated for treatment of individuals living with HIV and is commonly used as a nucleoside reverse-transcriptase inhibitor (NRTI) ‘backbone’ in combination antiretroviral therapy (ART) regimens. For pregnant and breastfeeding women, the WHO recommends daily TDF/FTC with efavirenz (EFV) as a first-line regimen [43]. While evaluations of the safety of TDF in studies of pregnant WLHIV may be confounded by concurrent use of other classes of antiretroviral medications and exposure to HIV infection, they can provide useful corollary data to supplement the limited PrEP data.

Multiple reviews have synthesized the studies of TDF use among pregnant WLHIV extensively [44,45]. A meta-analysis that estimated pooled risk ratios using data from 17 prospective studies and randomized controlled trials found that rates of preterm delivery and stillbirth were significantly lower among women who received TDF-based ART compared to women who did not receive TDF-based ART. No differences in risk of Grade 2, 3, or 4 adverse events, small for gestational age, miscarriage, congenital abnormalities or low birth weight were observed between groups. In the PROMISE trial among pregnant WLHIV ≥14 weeks gestation and CD4≥350 cells/mm³, women randomized to TDF-based ART were more likely to experience preterm delivery (<34 weeks) and neonatal mortality relative to those assigned zidovudine (ZDV)-based ART, but not when compared to women using ZDV alone [46], hypothesized to potentially be due to interactions between TDF and lopinavir/ritonavir. Most studies report normal infant growth for children of women who have used TDF-based ART, including z-scores for weight-for-age, length-for-age and head circumference at birth [47-49]. In a study of HIV-exposed, uninfected infants in the US, adjusted mean length-for-age z-score and head circumference z-score at 1-year were smaller among children whose mothers used TDF during pregnancy [49]. Studies that have assessed bone and kidney outcomes among infants generally did not find significant changes with in utero TDF exposure [47,50-52]. In two studies, bone mineral content was found to be lower among neonates exposed to TDF compared to TDF unexposed infants [48,53].

4.3. Comparison with safety of potential alternative PrEP drugs

While TDF/FTC is currently the only medication approved for use by HIV-uninfected cisgender women for HIV prevention, including pregnant women, there are several promising HIV prevention products in late stages of development or regulatory approval. As potential alternative HIV prevention options, it is important to consider how their safety profiles compare to oral TDF/FTC.

4.3.1. Oral tenofovir alafenamide with emtricitabine

Tenofovir alafenamide (TAF) is a newer TFV prodrug that has been shown to achieve 4.4-7 times higher intracellular concentrations of active TVF-DP with oral dosing compared to TDF despite ~90% lower drug exposure with 25 mg dosing [54-56] with potential to result in better renal and bone health outcomes [57]. For PrEP, current FDA approval for F/TAF (Discovy) oral PrEP only covers at-risk men and transgender women; cisgender women were excluded from the indication due to lack of trial data [58]. Clinical trials and pharmacokinetic studies of Discovy in women are ongoing in Uganda and South Africa [58]. Similar to how safety data on prenatal TDF use for ART among WLHIV helped inform guidelines in support of using oral TDF-based PrEP in pregnancy [44], accruing data on prenatal TAF use among WLHIV may help advance the evidence base for eventual TAF use as PrEP in pregnancy. Currently, data on F/TAF’s safety during pregnancy comes from WLHIV. The IMPAACT P1026s study, a pharmacokinetic study conducted among pregnant women using TAF 10mg with cobicistat or TAF 25 mg, identified two preterm deliveries and hepatic steatosis that were potentially associated with treatment [59]. Five infant abnormalities were observed among women using TAF 10 mg with cobicistat versus two among women using 25 mg TAF alone. In the IMPAACT 2010 study, dolutegravir (DTG)+FTC/TAF was associated with significantly fewer adverse pregnancy outcomes (driven by lower preterm and small-for-gestational-age rates) and less neonatal deaths than either DTG/TDF/FTC or EFV/FTC/TDF [60]. Mechanisms for these observations are unknown but higher exposure to TFV in the TDF-containing arms (vs. TAF) could be one possible mechanism. TAF is not currently recommended in pregnancy due to limited safety and pharmacokinetic data but additional evidence from ongoing phase 3 trials will become available soon [61].

4.3.2. Dapivirine vaginal ring

Use of a vaginal ring containing the non-nucleoside reverse transcriptase inhibitor, dapivirine, has been shown to reduce the risk of HIV acquisition among cisgender women by 27-31% in two major trials [62,63]. In July 2020, the European Medicines Agency (EMA) provided positive benefit-risk opinion of the dapivirine vaginal ring which, with recent prequalification by WHO, will bring this new HIV preventive modality closer to public introduction and approval by individual countries. Women who became pregnant had similar frequencies of pregnancy outcomes (preterm birth, stillbirth, spontaneous abortion, ectopic pregnancy) and early childhood outcomes (congenital abnormality and infant growth metrics) when assigned to the dapivirine ring relative to women assigned to a placebo ring [64]. Two phase IIIb Microbicide Trials Network studies to evaluate the safety of dapivirine ring and oral PrEP in pregnant and breastfeeding women are underway in Malawi, South Africa, Uganda and Zimbabwe [65].

4.3.3. Injectable cabotegravir

Recently, the HPTN 084 study of long-acting injectable cabotegravir (CAB-LA), an HIV integrase strand transfer inhibitor, demonstrated that CAB-LA given once every eight weeks was safe and superior to daily oral TDF/FTC for HIV prevention among cisgender women [66]. CAB-LA, when it gets regulatory approval, could be instrumental to overcoming the challenges observed with daily adherence to TDF/FTC. Limited data from 13 women who became pregnant after exposure to oral or injectable cabotegravir during Phase 2/3 studies show that four resulted in live births, two resulted in spontaneous abortion, two in medical or induced abortion and one possible early miscarriage [67]. It is unclear how these reported frequencies compare to those in the underlying population.

5. Conclusions

The data collected to date on TDF-based PrEP use during pregnancy are reassuring, indicating that the frequency of adverse pregnancy, birth and infant growth outcomes among mother infant exposed to TDF-based PrEP in pregnancy is no more common than that experienced by mothers or infants without TDF exposure. This evidence is further supported by extensive corollary data on TDF use for treatment among pregnant WLHIV. While current data are reassuring on safety in pregnancy, additional robust data from prospective studies and implementation projects that actively recruit women prior to conception or during pregnancy are still essential to better understand the pharmacokinetics, safety, and efficacy of TDF-based PrEP in pregnancy.

6. Expert opinion

In settings with high HIV burden, cisgender women face elevated risk of acquiring HIV in pregnancy and postpartum with the risk of HIV acquisition per-condomless-sex-act as high as 3- and 4-fold during late pregnancy and in the postpartum, respectively, compared to non-pregnant periods [3]. Oral TDF/FTC PrEP is recommended and is an attractive strategy for HIV prevention during periods of elevated HIV risk including during pregnancy and lactation.

In first-generation PrEP clinical trials, women discontinued PrEP when pregnancy was diagnosed resulting in limited pregnancy time on PrEP. However, there is a growing body of reassuring evidence from women with brief in utero TDF exposure from these trials as well as from real-world PrEP implementation projects. In trials among women, those who became pregnant and exposed to PrEP before protocol-mandated discontinuation, pregnancy outcomes were similar between placebo and TDF-based PrEP arms. Similarly, no reported discernable differences in weight or height z-scores or head circumference between infants with in utero PrEP exposure vs. placebo. Additional data from the Partners Demonstration Project among 30 women who become pregnancy and elected to continue PrEP antenatally are also reassuring – with no increases in pregnancy loss, preterm birth, or congenital anomalies [38]. Similar data from real-world PrEP implementation project in Kenyan women (PrIYA and PrIMA), lend credence to these findings [39,40,42]. While safety data are generally reassuring, there is need for continued accrual of data on growth and pregnancy outcomes in PrEP research studies and implementation projects.

The pharmacokinetics of TDF/FTC PrEP in pregnancy is a critical knowledge gap for the HIV prevention field. Recent clinical studies in African pregnant women using TDF/FTC PrEP suggest that blood (TFV) and cellular (TFV-DP) concentrations may be as low as 45-58% during pregnancy than non-pregnant periods even when adjusted for adherence, with the reduction more pronounced in the third trimester [20,21]. In the IMPAACT 2009 directly observed therapy study of TDF/FTC PrEP in pregnancy and postpartum, DBS TFV-DP levels in pregnancy were nearly 30-40% lower than in postpartum [21]. These data have raised important questions about the appropriate TDF/FTC PrEP dosing strategy in pregnancy. Whether these reductions in TFV-DP concentrations translate into sub-optimal protection against HIV during pregnancy is not clearly known. At the root of this confusion is the lack of clinical data that link cumulative dosing thresholds and concentrations with PrEP efficacy in women.

New PrEP options are nearing the market, which will allow women to select a method that best suits their needs and preferences. Longer-acting methods, such as the dapivirine vaginal ring and CAB-LA, hold potential to increase PrEP adherence and persistence by reducing the frequency that women will need to return to the clinic. Additionally, these methods can be used more discreetly than pills, which may help women to cope with stigma associated with PrEP in their communities [68]. However, to provide women with a well-informed choice, more evidence on the safety and efficacy of these products during pregnancy is needed.

The paucity of research on the use and safety of PrEP during pregnancy highlights wider barriers and disincentives to involving pregnant women in clinical studies of investigational biomedical products. While restrictions on enrollment of pregnant women are taken to protect the mother and fetus, this creates harm by perpetuating existing evidence gaps. As new PrEP agents come through the development pipeline, it will be important to actively recruit pregnant women in pivotal studies and provide women who experience incident pregnancies during follow-up with an informed option to remain on study product to address these gaps and promote equity [69].

In summary, TDF/FTC PrEP is a potent and recommended HIV prevention option, with a demonstrably good safety profile, including in pregnancy. While safety data are generally reassuring, there is need for continued accrual of data on growth and pregnancy outcomes in PrEP research studies and implementation projects. Lastly, pharmacologic studies that link TFV-concentrations to efficacy in cisgender women are needed to definitively understand the clinical relevance of observed reduction in cellular TFV-DP concentrations during pregnancy.

Article highlights:

• Safety of TDF/FTC use during pregnancy has been evaluated in multiple completed PrEP efficacy trials and real-world PrEP implementation projects

• Women who have been exposed to TDF/FTC experience similar frequencies of pregnancy, perinatal and early childhood outcomes compared to unexposed women

• Studies of women living with HIV who have been given treatment regimens containing TDF/FTC support these findings

• Safety data among pregnant women are limited for other PrEP products nearing the market, including longer-acting vaginal ring and injectable options

• Active inclusion of pregnant women in future evaluations of oral TDF/TDF use and other PrEP products is critical to addressing evidence gaps and promoting equity

Abbreviations

ART

antiretroviral therapy

CDC

Centers for Disease Control and Prevention

DTG

dolutegravir

EFV

efavirenz

eGFR

estimate glomerular filtration rate

EMA

European Medicines Agency

FTC

emtricitabine

HBV

hepatitis B virus

HIV

human immunodeficiency virus

NRTI

nucleoside reverse-transcriptase inhibitor

PrEP

Pre-exposure prophylaxis

TAF

tenofovir alafenamide

TDF

tenofovir disoproxil fumarate

TFV

tenofovir

TFV-DP

tenofovir diphosphate

WHO

World Health Organization

WLHIV

women living with HIV

ZDV

zidovudine