Scolaris Content Display Scolaris Content Display

Cochrane Database of Systematic Reviews Protocol - Intervention

Antivirals for prevention of hepatitis B virus mother‐to‐child transmission in human immunodeficiency virus positive pregnant women co‐infected with hepatitis B virus

Authors' declarations of interest

Version published: 19 June 2020 Version history

https://doi.org/10.1002/14651858.CD013653

This is not the most recent version

view the current version 12 June 2023
Collapse all Expand all

Abstract

Objectives

This is a protocol for a Cochrane Review (intervention). The objectives are as follows:

To assess the benefits and harms of tenofovir‐based antiviral combination regimens for hepatitis B virus (HBV) for the prevention of mother‐to‐child transmission of HBV, in HIV‐positive pregnant women co‐infected with HBV.

Background

Description of the condition

Worldwide, hepatitis B virus (HBV) accounts for an estimated 370 million chronic infections, and HIV for about 40 million (Bosh 2018; Terrault 2018; Joe‐Ikechebelu 2019). Chronic HBV infection occurs commonly in HIV‐infected persons, because the two viruses have similar routes of transmission through sexual and percutaneous contacts (Benhammou 2018). The prevalence of HBV‐HIV co‐infection differs according to geographical settings, depending on various factors, such as predominant transmission modes, HBV vaccination rates, and prevalence of HBV in the general population (Andreotti 2014; Weitzel 2020). Globally, the estimated prevalence of hepatitis B among people living with HIV/AIDS is 5% to 20% (Tengan 2017; Sarmati 2019). Therefore, about two to four million HIV‐infected persons are co‐infected with HBV worldwide (Mave 2014; Sarmati 2019; Weitzel 2020). Estimates of HBV prevalence in HIV‐infected pregnant women are scarce, especially in Western countries (Benhammou 2018; Weitzel 2020), but co‐infection with HBV is common in HIV populations of sub‐Saharan Africa (Andreotti 2014; Frempong 2019). Available data suggest that in resource‐rich settings, such as Western Europe and the USA, about 4% to 16% of the HIV‐infected population have chronic HBV infection (Tengan 2017). Conversely, in resource‐poor settings, such as Asia and Africa, where there is HBV endemicity, the HBV‐HIV co‐infection rate is up to 20% (Mave 2014; Sun 2014). In a SWEN India trial, the prevalence of active maternal HBV‐HIV co‐infection in pregnant women was estimated at 4.6% (Mave 2014).

Chronic HBV infection is a chronic necro‐inflammatory liver disease, caused by persistent HBV liver infection (Ahn 2010; Price 2013; Omotowo 2018). Individuals with chronic HBV are at increased risk of having cirrhosis, hepatic decompensation, and hepatocellular carcinoma. HIV infection appears to have a negative impact on the natural history of HBV infection (Tengan 2017). For instance, in immunocompromised people, the chronicity rates of HBV infection are higher, and the rates of hepatitis Be antigen (HBeAg) and hepatitis B surface antigen (HBsAg) loss, and seroconversion to anti‐HBe and anti‐HBs are lower than those in immunocompetent subjects (Tengan 2017).

Co‐infection with HBV in people with HIV who are receiving a commonly used nevirapine‐based antiretroviral therapy is a major concern for clinicians, owing to its high prevalence, the infrequent testing and treatment of viral hepatitis, and the impact of liver disease on the tolerability and effectiveness of anti‐HIV treatment (Mbougua 2010). Limited information is available on the possible interactions between the presence of hepatitis B co‐infection and the administration of potentially hepatotoxic drugs, including the non‐nucleoside reverse transcriptase inhibitor, nevirapine (Mbougua 2010). Thus, data obtained during pregnancy are particularly relevant, because of the increased risk of drug‐related toxicity reported for HBV‐HIV co‐infected pregnant women (Ouyang 2009; Mpody 2019).

Generally, HBV‐HIV co‐infected women have a higher risk of transmitting HBV to their infants (Kourtis 2012; Uyoh 2018; Li 2020). The risk of transmitting HBV to the infant is higher for mothers who are positive for HBeAg, and have a higher HBV load (Tengan 2017). Even with active and passive immunisation, the rate of breakthrough transmission to the infant remains as high as 5% to 15% in these women (Hoffmann 2007). Consequently, most HBV infections are acquired during the perinatal period, especially in highly endemic areas (Yi 2016; Eke 2017). Even in areas with low endemicity, perinatal transmission may account for up to one‐third of chronic infections (Yi 2016; Djaogol 2019). For this reason, mother‐ to‐child transmission (MTCT), during pregnancy and childbirth, has been recognized as the most important phase for prevention of chronic HBV infection in infants. The progression of chronic HBV infection to cirrhosis, end‐stage liver disease, or hepatocellular carcinoma is more rapid in persons with HBV‐HIV co‐infection than in persons with chronic HBV mono‐infection (Sarmati 2019).

Pregnant women who are HBV‐HIV co‐infected are twice as likely to test positive for HBeAg, three times more likely to have detectable HBV DNA, and have higher HBV DNA serum concentrations than those who are HBV mono‐infected, thereby greatly increasing the risk of MTCT (Brown 2016). In one West African study, maternal HBV‐HIV co‐infection increased the probability of MTCT of HBV by 2.5 times (Sangaré 2009). A potential association with adverse HIV outcomes in individuals who are HBV‐HIV co‐infected has been reported, in which HIV‐associated immune deficiency was aggravated in those with active HBV replication, resulting in increased progression to AIDS‐related outcomes and all‐cause mortality (Eleje 2014; Brown 2016).

According to the American Association for the Study of Liver Diseases 2018 hepatitis B guidance, all people with HBV‐HIV co‐infection, regardless of CD4 count, should receive antiretroviral therapy that includes two drugs with activity against HBV: specifically, tenofovir (tenofovir alafenamide or tenofovir disoproxil fumarate) plus lamivudine or emtricitabine (Terrault 2018; Mu 2020). Pregnant women who are already receiving effective antiretroviral therapy that does not include a drug with antiviral activity against HBV should have their treatment changed to include tenofovir disoproxil fumarate or tenofovir alafenamide, with emtricitabine or lamivudine (Ba 2019). HBV‐HIV co‐infection promotes an aggressive disease course of hepatitis B by the following mechanisms: increasing replication of the virus and rates of HBV reactivation; increasing the risk of acute liver failure, chronicity of newly acquired HBV infections, and occult infection, characterised by the absence of HBsAg and low viral replication, as well as HBV DNA concentration (Sarmati 2019).

Description of the intervention

Tenofovir, lamivudine, and emtricitabine are nucleoside reverse transcriptase inhibitors, which are effective against both HIV and HBV (Benhamou 1999; Hoff 2001; Dore 2004; Boettiger 2016; Calcagno 2016; Naing 2018; Terrault 2018). The interventions could be in the form of tenofovir‐based antiviral combination regimens. The rationale for using at least two dually active antiviral agents is to prevent the emergence of HBV‐associated drug mutation and resistance (Sarkar 2018).

Tenofovir

Both tenofovir disoproxil fumarate and tenofovir alafenamide are prodrugs of tenofovir, which are further converted to the active metabolite called tenofovir diphosphate intracellularly (Eke 2020; Mu 2020).

Tenofovir disoproxil fumarate

Tenofovir disoproxil fumarate has a low bioavailability of 25%, a long plasma half‐life of 12 hours to 18 hours, and intracellular half‐life of 69 hours to 139 hours. It also has a low protein binding of < 0.7%, and a negative charge at physiological pH (Baheti 2011). Tenofovir is excreted in the urine, predominantly via the proximal tubules, mainly by the process of glomerular filtration and active tubular secretion (Moss 2014). The main target of toxicity of tenofovir disoproxil fumarate is the proximal tubule, where it can lead to renal Fanconi syndrome in severe cases of toxicity (Hall 2011).

Tenofovir alafenamide

Tenofovir alafenamide is a phosphonoamidate prodrug of the nucleotide analogue tenofovir, whose active metabolite, known as tenofovir diphosphate, has an intracellular half‐life of 150 hours to 180 hours (De Clercq 2018). Tenofovir alafenamide regimens have improved bone and renal safety compared with tenofovir disoproxil fumarate‐containing regimens, while still maintaining outstanding efficacy (DeJesus 2018; Hagins 2018). For example, a 90% decrease in tenofovir plasma concentration and a 2.41‐fold increase in cell‐associated tenofovir diphosphate concentration was seen after a regimen of tenofovir disoproxil fumarate was converted to one, containing tenofovir alafenamide (Podany 2018). Higher intracellular accumulation of tenofovir diphosphate allows a smaller dose of tenofovir alafenamide (25 mg) compared with tenofovir disoproxil fumarate (300 mg), associated with improved drug adherence (Mu 2020). In cases of tenofovir‐combined antiviral therapy for HBV‐HIV co‐infection, tenofovir alafenamide may be preferred to tenofovir disoproxil fumarate because of its improved safety profile, and improved uptake by the lymphoid tissue (Gallant 2016; Huhn 2017; De Clercq 2018; Eke 2020). Tenofovir alafenamide is a novel prodrug of tenofovir that is more stable in plasma, enabling a 10‐fold reduction in dose, and resulting in a 91% reduction in plasma tenofovir, and a four‐fold increase in intracellular levels of tenofovir diphosphate (Gallant 2016).

In terms of safety of a tenofovir‐based antiviral regimen (tenofovir, emtricitabine, and ritonavir‐boosted lopinavir) among HIV‐positive pregnant women, there was a greater risk for very premature birth, and death within 14 days post delivery than for those assigned to receive non‐tenofovir combined antiviral therapy (zidovudine, lamivudine, and ritonavir‐boosted lopinavir; Fowler 2016). However, another study on birth outcomes for HIV‐positive pregnant women receiving a tenofovir‐based antiviral regimen (tenofovir‐emtricitabine) found that the risk of adverse birth outcomes was not higher with tenofovir disoproxil fumarate‐emtricitabine‐lopinavir or ritonavir than with zidovudine‐lamivudine‐lopinavir or ritonavir, or tenofovir disoproxil fumarate‐emtricitabine‐atazanavir or ritonavir (Rough 2018).

Lamivudine

Lamivudine is the negative enantiomer of 2′‐deoxy‐3′‐thiacytidine (analogue of cytidine), a dideoxypyrimidine in which the 3′ carbon of the ribose ring is replaced by a sulphur atom (Seley‐Radtke 2018). It is a dideoxynucleoside analogue first approved for the treatment of chronic HBV (Johnson 1999; Tavakolpour 2018). It is rapidly absorbed after oral administration, with maximum serum concentrations usually attained 0.5 hours to 1.5 hours after the dose. The adult absolute bioavailability is approximately 82%. Lamivudine systemic exposure is not altered when it is administered with food. Lamivudine is widely distributed into total body fluid, with the mean apparent volume of distribution of approximately 1.3 L/kg following intravenous administration. In pregnant women, lamivudine concentrations in maternal serum, amniotic fluid, umbilical cord, and neonatal serum are comparable, indicating that the drug diffuses freely across the placenta. In postpartum women, lamivudine is secreted into breast milk (Johnson 1999). The dominant elimination half‐life of lamivudine is approximately five hours to seven hours, and the in vitro intracellular half‐life of its active 5′‐triphosphate anabolite is 10.5 hours to 15.5 hours in HIV‐1 and 17 hours to 19 hours in HBV cell lines (Johnson 1999).

Emtricitabin

Emtricitabin is an enantiomer of a cytidine analogue which has a fluorine in the position 5 (Mu 2020). It is primarily excreted in the urine, and its improved efficacy is achieved with combination treatment (Saag 2006). When taken orally, emtricitabin exhibits bioavailability in over 90% of recipients. The peak plasma concentrations of emtricitabin are reached within three hours post‐dose. Less than 4% of emtricitabin is bound to plasma proteins, with a blood‐to‐plasma ratio of 0.6 (Molina 2018). Emtricitabin is activated through intracellular phosphorylation to emtricitabin‐5ʹ‐triphosphate. Compared to healthy adults, people with HIV‐infected demonstrate similar peak plasma concentration and exposure (Uglietti 2012). Emtricitabin is predominantly eliminated in urine. Its median elimination half‐life is 10 hours, and that of emtricitabin‐5ʹ‐triphosphate is 39 hours (Molina 2018). Co‐administration of food does not effect the bioavailability of emtricitabin. Emtricitabin is metabolized through oxidation of the sulphur moiety to form the 3ʹ‐sulfoxide diastereomers, and conjugation with glucuronic acid, to form 2ʹ‐O‐glucuronide. According to in vitro studies, emtricitabin is not an inhibitor of cytochromes P450 enzymes (Xu 2013). At doses from 50 mg to 400 mg, emtricitabin significantly suppresses the HIV viral load; a maximum reduction occurs with emtricitabin ≥ 200 mg/day. Emtricitabine is potent, safe, efficacious, and well tolerated (Mu 2020). Emtricitabine is a pregnancy category B drug, and as such, is an acceptable medication to use during pregnancy (Saag 2006). The incidence of fetal variations and malformations did not increase in embryo‐fetal toxicity studies performed in mice (Saag 2006).

How the intervention might work

Tenofovir

Tenofovir works by reducing viral replication (Benhamou 1999; Hoff 2001; Dore 2004; Terrault 2018). It reduces HIV mother‐to‐child transmission, either by lowering plasma viral load in pregnant women, or through post‐exposure prophylaxis in their newborns (Siegfried 2011; Terrault 2018). Although tenofovir disoproxil fumarate shows high efficacy and tolerance, renal and bone toxicity has been noted following its use (Hall 2011). The predisposing factors for tenofovir‐associated renal damage include low CD4+ T lymphocyte count, low body weight, older age, hepatitis C virus co‐infection, diabetes comorbidities, and the concomitant use of protease inhibitors (Tourret 2013; Moss 2014). Tenofovir is generally preferred in cases of confirmed lamivudine resistance.

Lamivudine

Lamivudine works by inhibiting HIV‐1, HIV‐2, and hepatitis B virus reverse transcriptase. It is phosphorylated to active metabolites that compete for incorporation into viral DNA. Lamivudine undergoes anabolic phosphorylation by intracellular kinases to form lamivudine 5'‐triphosphate, the active anabolite that prevents HIV‐1 and HBV replication by competitively inhibiting viral reverse transcriptase, and terminating proviral DNA chain extension (Johnson 1999). Lamivudine exerts more potent antiviral activity, presumably because it is resistant to cleavage from the 3′ terminals of RNA/DNA complexes by 3′‐5′‐exonuclease, and is not subject to deamination (Johnson 1999). Lamivudine also works by promoting HBeAg seroconversion, HBV DNA suppression, normalisation of alanine aminotransferase, and in decreasing the progression of liver fibrosis (Tavakolpour 2018). Following its use in HBV‐HIV co‐infected individuals, lamivudine‐resistance mutations in HBV frequently develop (Stewart 2011). Lamivudine resistance occurs as a result of mutations occurring within the tyrosine‐methionine‐aspartate‐aspartate (YMDD) motif of the reverse transcriptase gene at position 204, with the substitution of methionine with valine (M204V) or isoleucine (M204I; Wang 2019). Lamivudine is presently in pregnancy category C, by the US Food and Drug Administration, based primarily on animal data, with no clear evidence of harm in sparse human data.

Emtricitabine

Emtricitabin works by exhibiting its activity against both HIV and HBV (Mu 2020). Like lamuvidine, and other nucleoside analogues, emtricitabine is a prodrug that must be phosphorylated intracellularly into its active triphosphate form, called emtricitabine 5‐triphosphate. Emtricitabine 5‐triphosphate is incorporated, by reverse transcriptase, into the elongating proviral DNA chain, leading to termination of DNA synthesis through the inability of the next nucleotide to bind to emtricitabine 5‐triphosphate (active form) at the 3‐prime position (Saag 2006).

Why it is important to do this review

HBV–HIV co‐infection in pregnancy represents a major global public health threat (Kourtis 2012; Uyoh 2018). Treatment of HBV alone, without addressing the HIV infection, will lead to the emergence of HIV strains that are resistant to non‐nucleotide reverse transcriptase inhibitors (NRTI; Mpody 2019). HIV infection impacts the course of hepatitis B, as higher rates of chronic carriage, lower seroconversion rates, and accelerated progression towards cirrhosis have been observed (Brown 2016; Uyoh 2018). Vaccination against hepatitis B is less effective in HIV‐infected individuals, and co‐infection is associated with a significantly reduced CD4 response in pregnancy (Floridia 2017). Because each virus affects the other's natural history and response to therapy, HBV‐HIV co‐infection requires dedicated evidence‐based research, and should be a priority for health researchers and policymakers. The use of antiretroviral agents with dual antiviral activity is a promising preventive approach. However, this is limited by a paucity of evidence on important agents (e.g. tenofovir) regarding safety during pregnancy, for both the fetus and the mother (Kourtis 2012; Eke 2020). The 2015 World Health Organization (WHO) guidelines for the management of chronic HBV infection did not recommend an antiviral agents approach, citing limited and low certainty evidence on the relative benefits or harm; but in 2016, the American Association for the Study of Liver Diseases recommended antiviral therapy in HBsAg–positive pregnant women who had an HBV DNA level of more than 200,000 IU/mL, despite a low certainty of evidence (Terrault 2016).

Prenatal antiretroviral or antiviral treatment may increase the risk for adverse birth outcomes among women with HIV infection or HBV, but whether the risk differs by antiretroviral therapy or antiviral regimen is unknown (Zash 2017). There remains a number of unanswered questions about disease pathogenesis and management in HBV‐HIV co‐infected pregnant women. Pregnancy itself can trigger elevation of liver enzymes (Kourtis 2012). During pregnancy, the administration of antiretroviral prophylaxis containing one agent with anti‐HBV activity may be associated with later development of HBV resistance. For pregnant women who take antiretroviral prophylaxis containing one or two agents with anti‐HBV activity to prevent perinatal HIV transmission, the safety of stopping treatment after delivery is unknown (Eleje 2018). The administration of antiretroviral agents without HBV activity in co‐infected pregnant women may leave their infants unprotected against HBV. As regimens including tenofovir become first‐line therapy for many HIV‐infected people (and are used as pre‐exposure prophylaxis for the uninfected), determining the safety of these medications during pregnancy becomes a critical research need. Finally, infection of the infant with HIV threatens the benefits of HBV immunisation for perinatal prevention (Kourtis 2012). Therefore, this review is timely in its aim to establish the availability of up‐to‐date evidence about the effects of antiviral agents for HBV, for the prevention of HBV mother‐to‐child transmission in HIV‐positive pregnant women, co‐infected with HBV. Although there are previous Cochrane Reviews on antenatal pharmacological interventions for the prevention of mother‐to‐child transmission of HBV (Eke 2017), or HIV (Siegfried 2011), none have been published on HBV‐HIV co‐infection. The only available systematic review and meta‐analysis evaluating the efficacy and maternal‐fetal safety of antiviral therapy during pregnancy, for prevention of mother‐to‐child transmission of HBV or HIV, is a non‐Cochrane review (Siemieniuk 2017). In the Siemieniuk 2017 review, none of the women studied had HBV‐HIV co‐infection, but were either HBV‐ or HIV‐seropositive. This justifies the need for high‐quality research to assess the feasibility, effectiveness, and safety of antiviral agents for HBV, for the prevention of mother‐to‐child transmission of HBV in HIV‐positive pregnant women co‐infected with HBV.

Objectives

To assess the benefits and harms of tenofovir‐based antiviral combination regimens for hepatitis B virus (HBV) for the prevention of mother‐to‐child transmission of HBV, in HIV‐positive pregnant women co‐infected with HBV.

Methods

Criteria for considering studies for this review

Types of studies

Randomised clinical trials, regardless of blinding, publication date, type and status of publication, language, and unit of randomisation, assessing the benefits and harms of hepatitis B antivirals as treatment for pregnant women with hepatitis B virus (HBV), who are co‐infected with HIV.

Types of participants

Pregnant women, 18 years of age or older, HIV‐positive and co‐infected with HBV, with pregnancy between 14 weeks and 28 weeks’ gestation (estimated on the basis of the dates of the last menstruation and ultrasonography findings, if available).

We will exclude women with hepatitis delta co‐infections, hepatitis C, or co‐existing malignancy.

Types of interventions

Experimental intervention

  • Tenofovir‐based antiviral combination regimens with anti‐retroviral regimen with lopinavir‐ritonavir therapy, or any other antiviral therapy, and two drugs with activity against HBV, specifically, tenofovir alafenamide (TAF) or tenofovir disoproxil fumarate (TDF), plus lamivudine or emtricitabine.

Control intervention

  • Placebo, or tenofovir alone, or non‐tenofovir‐based antiviral regimen (zidovudine, lamivudine, telbivudine, emtricitabine, entecavir, lopinavir‐ritonavir, or any other antiviral therapy) either alone or in combination with at least two other antivirals.

We will also include trials where one intervention, or combination of interventions was compared with the experimental intervention (tenofovir‐based antiviral combination regimen).

We will allow co‐interventions if women in both the experimental and control groups of the trials receive them equally.

Types of outcome measures

Primary outcomes
Infant outcomes

  • All‐cause infant mortality at longest follow‐up

  • Proportion of infants with serious adverse events (including infants with prematurity)

  • Proportion of infants with HBV mother‐to‐child transmission, assessed by either hepatitis B surface antigen (HBsAg)‐seropositivity at 6 months to 12 months, detectable HBV DNA at 6 months to 12 months, or anti‐HBc‐positivity at 6 months to 12 months

Maternal outcomes

  • All‐cause maternal mortality at the longest follow‐up

  • Proportion of mothers with serious adverse events. The adverse event, defined according to the International Conference on Harmonisation (ICH) Guidelines, may be any untoward medical occurrence that resulted in death, was life threatening, required inpatient hospitalisation or prolongation of existing hospitalisation, resulted in persistent or significant disability or incapacity, or was a congenital anomaly or birth defect (ICH‐GCP 1997). We will consider all other adverse events as non‐serious. We will assess the proportion of participants with one or more serious adverse events.

Secondary outcomes
Infant outcomes

  • Proportion of infants with adverse events not considered serious

Maternal outcomes

  • Proportion of mothers with detectable HBV DNA (before delivery)

  • Maternal hepatitis Be antigen (HBeAg) to HBe‐antibody seroconversion (before delivery)

  • Maternal adverse events not considered serious

Exploratory outcomes

  • Proportion of mothers undergoing a caesarean section

  • Proportion of mothers with postpartum haemorrhage

  • Proportion of mothers with detectable HBV DNA (at 28 weeks postpartum)

Search methods for identification of studies

We will search for papers in all languages, and translate them as necessary.

Electronic searches

We will search the Cochrane Hepato‐Biliary Group (CHBG) Controlled Trials Register (maintained and searched internally by the CHBG Information Specialist via the Cochrane Register of Studies Web); the Cochrane Central Register of Controlled Trials (CENTRAL; latest issue) in the Cochrane Library; MEDLINE Ovid (1946 to date of search); Embase Ovid (1974 to date of search); LILACS Bireme (Latin American and Caribbean Health Science Information database; 1982 to date of search); Science Citation Index Expanded Web of Science (1900 to date of search); and Conference Proceedings Citation Index‐ Science Web of Science (1990 to date of search; Royle 2003). We will apply no language or document type restrictions. Appendix 1 shows the preliminary search strategies with the planned time spans of the searches.

Searching other resources

We will identify additional references by manually searching the reference lists of included trials for any further potential trials that have not been identified via electronic searches. We will also search on‐line trial registries, such as ClinicalTrials.gov (clinicaltrials.gov), the European Medicines Agency (EMA; www.ema.europa.eu/ema), the WHO International Clinical Trials Registry Platform (www.who.int/ictrp), and the Food and Drug Administration (FDA; www.fda.gov) for ongoing or unpublished trials. We will contact experts in the field and pharmaceutical companies to enquire about additional trials.

Data collection and analysis

We will follow the available guidelines provided in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2019). We will use RevMan Web 2019 and Trial Sequential Analysis software to perform the analyses (Wetterslev 2008; Thorlund 2011a; Thorlund 2011b; TSA 2011; Wetterslev 2017).

Selection of studies

Following retrieval of potentially eligible publications for inclusion in this Cochrane Review, EOU and GUE will independently select the studies fulfilling the inclusion criteria detailed in the current protocol. We will include trials, regardless of their reporting on outcomes of interest to our review. EOU and GUE will contact study authors if trial reports selected for our review do not provide sufficient or clear information. Review authors will look for reports of harms in quasi‐randomised and other studies obtained during the searches for randomised clinical trials only. EOU, GUE, and AOU will resolve disagreements by discussion.

We will not specifically search for observational studies (i.e. quasi‐randomised studies, cohort studies, or patient reports) to include in this review, which is a known limitation of the review in terms of adverse events. We are aware that the decision to not search systematically for all observational studies, and to extract data on harm only from quasi‐randomised and controlled clinical studies might bias our review towards assessment of benefits, and might overlook certain harms, such as late or rare harms (Storebø 2018). If data demonstrate that antivirals prevent hepatitis B virus mother‐to‐child transmission in HIV‐positive pregnant women, co‐infected with hepatitis B virus, then a systematic review of the harms of this intervention in observational studies ought to be planned. We will report the screening and selection process in a PRISMA flow chart.

Data extraction and management

Two review authors (EOU and GUE) will independently complete a pilot‐tested data extraction form for all included trials. We will retrieve the following data.

  • General information: title, journal, year of publication, publication status, language of publication, country in which participants were recruited, and trial design

  • Sample size: number of participants meeting the criteria and total number screened

  • Baseline characteristics: baseline diagnosis, age, mode of infant feeding, race, disease severity, inclusion and exclusion criteria, type of hepatitis B or HIV treatment, and concurrent medications used

  • Intervention(s) and comparison: details of the intervention(s) and comparison

  • Funding of the trial or sources of support (see Appendix 2 for definition)

We will also record whether the included trials had pre‐published protocols, and we will add the protocol publications to the references to included studies.

Assessment of risk of bias in included studies

Two review authors (EOU and GUE) will independently assess the risk of bias of each included trial, as per the recommendations in the Cochrane Handbook for Systematic Reviews of Interventions and methodological studies (Schulz 1995; Moher 1998; Kjaergard 2001; Wood 2008; Higgins 2011; Savović 2012a; Savović 2012b; Savović 2018). We will use the following definitions in the assessment of risk of bias.

Allocation sequence generation

  • Low risk of bias: the study performed sequence generation using computer random number‐generation or a random number table. Drawing lots, tossing a coin, shuffling cards, and throwing dice will be considered adequate if an independent person, not otherwise involved in the study, performed them.

  • Unclear risk of bias: the study authors did not specify the method of sequence generation.

  • High risk of bias: the sequence generation method was not random. We will only include such studies for assessment of harms.

Allocation concealment

  • Low risk of bias: the participant allocations could not have been foreseen in advance of, or during enrolment. A central and independent randomisation unit controlled allocation. The investigators were unaware of the allocation sequence (as would be if, for example, the allocation sequence was hidden in sequentially numbered, opaque, and sealed envelopes).

  • Unclear risk of bias: the study authors did not describe the method used to conceal the allocation, so the intervention allocations may have been foreseen before, or during, enrolment.

  • High risk of bias: it is likely that the investigators who assigned the participants knew the allocation sequence. We will only include such studies for assessment of harms.

Blinding of participants and personnel

  • Low risk of bias: any of the following — blinding of participants and key study personnel ensured, and it was unlikely that the blinding could have been broken; or rarely, no blinding or incomplete blinding, but the review authors judged that the outcome was not likely to be influenced by lack of blinding, such as all‐cause mortality, or morbidity related to HBV or HIV, serious adverse events, or proportion of participants in need of antiviral agents for HBV.

  • Unclear risk of bias: any of the following — insufficient information to permit judgement of ‘low risk’ or ‘high risk’; or the trial did not address this outcome.

  • High risk of bias: any of the following — no blinding, or incomplete blinding, and the outcome was likely to be influenced by lack of blinding; or blinding of key study participants and personnel attempted, but likely that the blinding could have been broken, and the outcome was likely to be influenced by lack of blinding.

Blinded outcome assessment

  • Low risk of bias: any of the following — blinding of outcome assessment ensured, and unlikely that the blinding could have been broken; or rarely, no blinding of outcome assessment, but the review authors judged that the outcome measurement was not likely to be influenced by lack of blinding, such as all‐cause mortality, or morbidity related to HBV or HIV, serious adverse events, or proportion of participants in need of antiviral agents for HBV.

  • Unclear risk of bias: any of the following — insufficient information to permit judgement of ‘low risk’ or ‘high risk’; or the trial did not address this outcome.

  • High risk of bias: any of the following — no blinding of outcome assessment, and the outcome measurement was likely to be influenced by lack of blinding; or blinding of outcome assessment, but likely that the blinding could have been broken, and the outcome measurement was likely to be influenced by lack of blinding.

Incomplete outcome data

  • Low risk of bias: missing data were unlikely to make treatment effects depart from plausible values. The study used sufficient methods, such as multiple imputation, to handle missing data.

  • Unclear risk of bias: there was insufficient information to assess whether missing data, in combination with the method used to handle missing data, was likely to bias the results.

  • High risk of bias: the results were likely to be biased, due to missing data.

Selective outcome reporting

  • Low risk: the study reports at least one of the primary outcomes: all‐cause infant mortality at longest follow‐up, proportion of infants with serious adverse events (including infants with prematurity), proportion of infants with HBV mother‐ to‐child transmission, as assessed by either HBsAg‐seropositivity at 6 months to 12 months, detectable HBV DNA at 6 months to 12 months, or anti‐HBc‐positivity at 6 months to 12 months, all‐cause maternal mortality at the longest follow‐up, and proportion of mothers with serious adverse events. If the original trial protocol is available, the outcomes should be those called for in that protocol. If we obtain the trial protocol from a trial registry (e.g. clinicaltrials.gov), the outcomes sought should be those enumerated in the original protocol, if the trial protocol was registered before or at the time the trial was begun. If the trial protocol was registered after the trial was begun, we will not consider those outcomes to be reliable.

  • Unclear risk: the study authors do not report all predefined outcomes fully, or it is unclear whether the study authors recorded data on these outcomes or not.

  • High risk: the study authors report none of the predefined outcomes.

Other bias

  • Low risk of bias: the trial appears to be free of other bias domains that could put it at risk of bias.

  • Unclear risk of bias: the trial may or may not be free of other domains that could put it at risk of bias.

  • High risk of bias: there are other factors in the trial that could put it at risk of bias.

Overall bias assessment

  • Low risk of bias: all domains in a trial are classified at low risk of bias according to the definitions described above.

  • High risk of bias: one or more of the bias domains in a trial are classified at unclear or high risk of bias.

We will generate a 'Risk of bias' graph and 'Risk of bias' summary to show a summary of this assessment.

Measures of treatment effect

We will calculate the risk ratios (RRs) with 95% confidence intervals (CIs) for dichotomous variables. We do not anticipate continuous outcomes, nor time‐to‐event data.

Unit of analysis issues

We will only include participants according to the treatment group of the randomised clinical trials. If there are trials with more than two parallel intervention groups, we will split the control group into two or more groups with smaller sample size, if it is within the same comparison. Alternatively, we will combine groups to create a single pair‐wise comparison.

We do not expect to find cluster‐randomised or cross‐over trials. However, if we find cluster‐randomised trials, we will analyse and assess the risk of bias of cluster‐randomised trials separately from the randomised parallel group clinical trials included in our review (Higgins 2019). If we find cross‐over trials, we will only use data from the first trial period for analysis, in order to avoid the cross‐over effect of the intervention (Higgins 2019).

Dealing with missing data

If data are unclear or missing from a published report, we will contact the original study authors to request missing data, whenever possible.

We will seek to perform intention‐to‐treat analyses, and will include the missing data by considering trial participants as either treatment failures, or treatment successes, by imputing them according to the following scenarios.

  • Extreme case analysis that favours the experimental intervention ('best‒worst' case scenario): none of the dropouts or participants were lost from the experimental group, but all of the dropouts or participants lost from the control group experienced the outcome; we will include all randomised participants as per the intervention group in the denominator.

  • Extreme case analysis that favours the control ('worst‒best' case scenario): all dropouts or participants were lost from the experimental group, but none from the control group experienced the outcome; we will include all randomised participants as per the intervention group in the denominator.

Assessment of heterogeneity

We will use the I² statistic to measure heterogeneity among the trials in each analysis, and we will interpret the results following Deeks 2019:

  • 0% to 40%: might not be important;

  • 30% to 60%: may represent moderate heterogeneity;

  • 50% to 90%: may represent substantial heterogeneity;

  • 75% to 100%: considerable heterogeneity.

If we identify substantial heterogeneity (I² > 50%), we will report it, and explore the possible causes by prespecified subgroup analyses.

Assessment of reporting biases

Whenever we have 10 or more trials, we will draw funnel plots to assess reporting biases from the individual trials by plotting the risk ratio (RR) on a logarithmic scale against its standard error (Deeks 2019).

Data synthesis

Meta‐analysis

We will perform the meta‐analyses according to the recommendations of Cochrane (Higgins 2019). We will use the software package RevMan web provided by Cochrane (RevMan Web 2019). Where possible, we will analyse the data using the intention‐to‐treat principle, including all randomly assigned participants, regardless of how complete the data are. Otherwise, we will use the data provided by the trialists. We will use a random‐effects model, as trials are performed by different researchers, operating independently, and it would be unlikely that all the trials were functionally equivalent, with a common effect estimate. Therefore, the random‐effects model is more justified than the fixed‐effect model (DerSimonian 1986).

Subgroup analysis and investigation of heterogeneity

We will attempt to investigate:

  • trials at low risk of bias compared to trials at high risk of bias in the overall assessment, because trials at high risk of bias may overestimate or underestimate the intervention effects

  • trials without for‐profit support compared to trials with, or unknown for‐profit support (see Appendix 2 for definition), because trials with for‐profit support may overestimate or underestimate intervention effects (Lundh 2017)

  • triple highly active antiretroviral therapy (HAART) regimen without booster compared to triple HAART regimen with booster, because the different combinations may impact the treatment

  • lamivudine containing antiviral agent for HBV compared to non‐lamivudine containing antiviral agent for HBV, because the different combination may impact the treatment

  • treatment‐naïve compared to non‐treatment‐naïve participants, as the treatment may have different effects on the two group of participants

Sensitivity analysis

In addition to the two sensitivity analyses specified in Dealing with missing data, we plan to perform the following.

  • Assessing the robustness of our results by including only trials reporting intention‐to‐treat analyses

  • Assessing the effects of risk of bias by excluding trials classified at high risk of bias.

  • Comparing fixed‐effect and random‐effects estimates of the intervention effect, to assess the influence of small‐study effects on the results of our meta‐analysis

  • Comparing our assessment of imprecision with GRADE, to that performed with a Trial Sequential Analysis (see below).

Trial Sequential Analysis

We will conduct a Trial Sequential Analysis (TSA), as cumulative meta‐analyses are at risk of producing random errors due to sparse data and repetitive testing of the accumulating data (Wetterslev 2008; Thorlund 2011; TSA 2011; Imberger 2016; Wetterslev 2017). To minimise random errors, we will calculate the diversity‐adjusted required information size (DARIS; i.e. the number of participants needed in a meta‐analysis to detect or reject a certain intervention effect; Wetterslev 2008; Wetterslev 2017).

The DARIS calculation should also account for the diversity present in the meta‐analysis (Wetterslev 2008; Wetterslev 2009; Wetterslev 2017). In our meta‐analysis, we will base the required information size on the event proportion in the control group; assumption of a plausible RR reduction of 20%, or the RR reduction observed in the included trials at low risk of bias; a risk of type I error of 2.5% for the three primary infant outcomes and 3.3% for the two primary maternal outcomes, a risk of type II error of 10% (Castellini 2017); and the assumed diversity of the meta‐analysis (Wetterslev 2009). We will draw on alpha separately per group, newborns, and mother (see Types of outcome measures).

The underlying assumption of Trial Sequential Analysis is that testing for significance may be performed each time a new trial is added to the meta‐analysis. We will add the trials according to the year of publication. If more than one trial is published during the same year, we will add trials alphabetically, according to the last name of the first study author. On the basis of the required information size, we will construct trial sequential monitoring boundaries (Wetterslev 2008; Thorlund 2011; Wetterslev 2017). These boundaries will determine the statistical inference one may draw regarding the cumulative meta‐analysis that has not reached the required information size; if the trial sequential monitoring boundary for benefit or harm is crossed before the required information size is reached, firm evidence may perhaps be established and further trials may turn out to be superfluous.

On the other hand, if the boundaries are not surpassed, it is most probably necessary to continue adding trials in order to detect or reject a certain intervention effect. That can be determined by assessing whether the cumulative Z‐curve crosses the trial sequential monitoring boundary for futility.

We will report the Trial Sequential Analysis ‒ adjusted CI if the cumulative Z‐curve does not pass through any of the trial sequential monitoring boundaries for harm, benefit, or futility. We will make a comparison between our Trial Sequential Analysis and the GRADE assessment of imprecision as a sensitivity analysis. We plan to downgrade our assessment of imprecision in GRADE by two levels if the accrued number of participants is below 50% of the DARIS, and one level if between 50% and 100% of DARIS. We will not downgrade if the cumulative Z‐curve reaches futility or DARIS.

'Summary of Findings' tables

We will assess confidence in the evidence using GRADE criteria and the GRADEpro software (Atkins 2004; GRADEpro GDT). We will present the results in a ‘Summary of Findings’ table. We will present assessments of the evidence using five factors referring to limitations in the study design and implementation of included studies that suggest the quality of the evidence: risk of bias; indirectness of evidence (population, intervention, control, outcomes); unexplained heterogeneity or inconsistency of results; imprecision of results; and a high probability of publication bias. We will define the levels of evidence as 'high', 'moderate', 'low', or 'very low'. We will follow the recommendations of Section 8.5 and Chapter 13 of the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2019a; Page 2019). These grades are defined as follows.

  • High certainty: we are very confident that the true effect lies close to that of the estimate of the effect.

  • Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different.

  • Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect.

  • Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect.

We will include the quality of the evidence for these five outcomes in the 'Summary of findings' table.

  • all‐cause infant mortality at longest follow‐up;

  • proportion of infants with serious adverse events (including infants with prematurity);

  • proportion of infants with HBV mother‐to‐child transmission, assessed by either HBsAg‐seropositivity at 6 months to 12 months, detectable HBV DNA at 6 months to 12 months, or anti‐HBc‐positivity at 6 months to 12 months;

  • all‐cause maternal mortality at the longest follow‐up;

  • proportion of mothers with serious adverse events.