Biological factors that may impair transplacental transfer of RSV antibodies: Implications for maternal immunization policy and research priorities for low- and middle-income countries.

Respiratory syncytial virus (RSV) is the leading viral cause of acute lower respiratory tract infection (ALRI), including bronchiolitis and pneumonia, in infants and children worldwide. Protection against RSV is primarily antibody mediated and passively acquired RSV neutralizing antibody can protect infants from RSV ALRI. Maternal immunization is an attractive strategy for the prevention of RSV in early infancy when immune responses to active immunization may be suboptimal and most severe RSV disease and death occur. However, several biologic factors have been shown to potentially attenuate or interfere with the transfer of protective naturally acquired antibodies from mother to fetus and could therefore also reduce vaccine effectiveness through impairment of transfer of vaccine-induced antibodies. Many of these factors are prevalent in low- and middle-income countries (LMIC) which experience the greatest burden of RSV-associated mortality; more data are needed to understand these mechanisms in the context of RSV maternal immunization. This review will focus on what is currently known about biologic conditions that may impair RSV antibody transfer, including preterm delivery, low birthweight, maternal HIV infection, placental malaria, and hypergammaglobulinemia (high levels of maternal total IgG). Key data gaps and priority areas for research are highlighted and include improved understanding of the epidemiology of hypergammaglobulinemia and the mechanisms by which it may impair antibody transfer. Key considerations for ensuring optimal vaccine effectiveness in LMICs are also discussed.

Introduction

Respiratory syncytial virus (RSV) is the leading viral cause of acute lower respiratory tract infection (ALRI) in infants and children worldwide [1]. Globally, RSV causes an estimated 33.1 million episodes of illness, 3.2 million hospitalizations, and 59,600–118,200 deaths in children under five years each year, 99% of which occur in low- and middle-income countries (LMICs) [1]. Over 40% of all RSV-associated hospitalizations and 45% of in-hospital deaths due to RSV occur among infants under 6 months of age [1].

Protection against RSV is primarily antibody-mediated, and passively-acquired RSV neutralizing antibody can protect infants against RSV ALRI [2], [3], [4], [5], [6], [7], [8], [9], [10], [11]. While single-dose extended half-life monoclonal antibodies are in clinical development, the only product currently available for RSV prevention is the monoclonal antibody palivizumab (Synagis®). Palivizumab requires weight-based monthly injections in infancy, and while effective for preventing hospital admissions for specific high-risk groups, broader use, particularly in LMICs, is either logistically infeasible, cost-prohibitive, or both.

Despite decades of research, no vaccines for active or passive prevention of RSV in neonates, infants, or adults have demonstrated sufficient efficacy and safety in phase III clinical trials to support licensure to date. Nonetheless, multiple promising vaccine candidates are in the pipeline and may be licensed in the next 2–5 years [12], [13], [14], [15]. Several of the leading products employ passive immunization approaches (e.g., vaccines for antenatal administration) to protect children during the first few months of life, when risk of severe RSV ALRI and death are high and immunologic responses to active immunization are often inadequate.

Maternal immunization has been used widely for the prevention of maternal and neonatal tetanus, pertussis, and influenza [16], [17], [18]. It is a promising approach for the prevention of RSV, and likely a cost-effective strategy for use in LMICs; however, protection of infants via this mechanism requires both robust immune responses to immunization among pregnant women and efficient transplacental transfer of vaccine-induced antibodies. The latter requires adequate levels of pathogen-specific immunoglobulin G (IgG) in the mother’s blood during the period of pregnancy when transfer is most significant, a functionally intact placenta, and, ideally, term or near-term delivery of the infant [11], [19]. Maternal antibodies for RSV are actively and efficiently transported across the placenta. However, in many geographic settings where the need for a maternal vaccine against RSV is greatest, conditions associated with impaired transfer of antibodies for RSV or other antigens, such as HIV, placental malaria, and hypergammaglobulinemia, are present [8], [20], [21], [22], [23], [24], [25], [26], [27]. There is a paucity of data on factors that may impair transplacental transfer of RSV antibodies specifically; understanding of how infections, chronic medical conditions, and environmental factors may impact effectiveness of maternal immunization for RSV in different geographic settings will be critical to planning for vaccine use and estimating potential vaccine effectiveness and impact.

This review assesses what is currently known about biological factors that may impair RSV antibody transfer, what relevant knowledge might be gleaned from studies involving other pathogens and makes recommendations for future research priorities to address knowledge gaps in preparation for future availability of maternal RSV vaccines.

Background

Mechanism and timing of transplacental antibody transport

Among the five classes of human immunoglobulin, only IgG is transported across the placenta in meaningful quantities. Significant progress in understanding the mechanisms of transplacental transport of maternal antibody has been made in recent years [19], [28], [29], [30]. Briefly, maternal antibodies are transferred via an active, receptor-mediated process across all three layers of the placenta maternal-fetal interface: the multinucleated syncytiotrophoblasts, the villous stroma containing placental fibroblasts and Hofbauer cells, and the endothelial cells of the fetal capillaries [31]. IgG molecules are endocytosed from extracellular fluid on the maternal side syncytiotrophoblast layer. The acidic pH inside the endosome allows binding of IgG to the membrane-bound FcRn (Fragment crystallizable receptor, neonatal) and prevents degradation. Then, IgG is then transcytosed to the fetal surface, where a return to physiological pH allows disassociation of the IgG from the FcRn [19], [28], [29], [32], [33].

Of the four IgG subclasses, affinity for the FcRn receptor is greatest for IgG1, which are transferred more efficiently than IgG4 and IgG3. IgG2 have the lowest transfer efficiency [34], [35], [36]. RSV-specific antibodies have been shown to efficiently cross the placenta in healthy pregnancies. Similarly, RSV-specific antibodies detected in infants are mostly IgG1; a minority are IgG3. IgG4 and IgG2 are not common [37], [38], [39], [40]. Recent studies have indicated that differences in subclass transport may also be governed by glycosylation patterns [28], [29], [34], [41] and that antibodies that activate natural killer cells may transfer more efficiently than antibodies that activate neutrophils or monocytes [35], [42], [43]. Furthermore, other Fc gamma receptors (FcγRI, FcγRII, FcγRIII) are expressed in the layers of the placental interface in various combinations and may be important in transplacental antibody transfer [35], [44]. Other studies, however, suggest that involvement of other Fcγ receptors is limited [45], [46]. Improved understanding of the stages of IgG transfer and the underlying mechanisms is needed and may enable the design of improved maternal vaccines with enhanced transfer efficiency [28], [31].

Timing of transplacental antibody transfer

Transplacental antibody transfer begins in the first trimester but is minimal; approximately 10% of the maternal concentration of antibodies is believed to be present in cord blood by 17–22 weeks’ gestation [19], [28]. Transfer begins to increase during the second trimester as the expression of FcRn increases on the growing surface area of the placental interface. In general, fetal concentrations approach approximately 50% of maternal concentration by 30 weeks’ gestation and continue to increase markedly until term delivery, although there is some variation by subtype (Fig. 1). Notably, the majority of IgG is transported during the last four weeks of a full-term (40-week) pregnancy [19], [28], [34], [47], at which time fetal concentrations may surpass 100% of maternal concentration.

Fig. 1

Transplacental antibody transfer ratios by gestational week and IgG subtype. Percentages of placental transfer ratios of IgG subclasses delivered to preterm and term newborns in different gestational weeks. IgG1 shows a peak transfer ratio at 37 weeks of pregnancy. IgG2 transfer ratios are always lower than other IgG [19].

Assessing transplacental transfer efficiency

Neonatal antibody levels are highly correlated with maternal levels [19]. Several studies have documented cord-to-maternal titer ratios (CMR) by measuring maternal peripheral blood on or near the day of delivery and blood taken from the fetal side of the umbilical cord shortly after birth or directly from the neonate. The concentration of any specific maternal IgG in the blood of full-term infants from healthy pregnancies typically exceeds the concentration in maternal blood, resulting in a CMR > 1.0. This metric can be helpful for assessing the efficiency of transplacental antibody transfer, although the observed ranges can vary substantially by antigen and method of measurement. Absolute values (cord titers) of antibody may be a more appropriate metric for understanding presence of passive immunity when pathogen-specific correlates of immunity are known. RSV antibody CMRs vary and have been documented to range from 0.58 to 1.33 in a variety of settings (Table 1) [8], [21], [23], [24], [25], [26], [27], [48], [49], [50], [51].

Table 1

Factors evaluated for impairment of RSV transplacental Ab transfer.

First author, year	Location	N pairs	HIV	PM	HyG	Pre-maturity	Birth-weight	Other factors	Observed RSV CMRs	MatAb Half-life	Key findings
Okoko, 2001 TMIH [50]	The Gambia	213	n/a	n/a	n/a	Yes	Yes	n/a	1.18 (term, ABW), 1.01 (term, LBW), 0.76 (preterm, ABW), 0.86 (preterm, LBW)	n/a	In The Gambia, prematurity and LBW were associated with reduced RSV Ab transfer
Okoko, 2001 JID [49]	The Gambia	213	n/a	Yes	Yes	n/a	n/a	Parity, maternal age, weight, height, not associated with transfer	0.62 PM+, 1.18 PM-	n/a	In The Gambia, PM and HyG were associated with reduced RSV Ab transfer
Chu, 2014 [26]	Bangladesh	149	n/a	n/a	n/a	n/a	No	Mode of delivery, infant sex, parity, season of birth, maternal age, maternal education not associated with transfer	1.01	38 days	RSV Ab transfer was efficient in Bangladesh; higher cord titers were associated with protection from RSV infection
Atwell, 2016 [25]	Papua New Guinea	300	n/a	No	Yes	n/a	n/a	n/a	1.20; 1/3 of infants with titer < 1:200, 1/3 of infants with CMR < 1.0	n/a	RSV Ab transfer was impaired in ∼ 1/3 of pairs; HyG was associated with impairment, but not PM
Chu, 2017 [27]	Nepal	310	n/a	n/a	n/a	No	No	Increasing maternal parity and female sex associated with increased Ab transfer	1.03	n/a	RSV Ab transfer was efficient in Nepal; higher cord titers were not associated with protection from RSV infection
Jallow, 2019 [48]	South Africa	240	No	n/a	Yes	n/a	No	n/a	0.74 overall; 0.58 HyG+, 0.80 HyG-	n/a	HEU infants and those born to mothers with HyG had lower cord titers; only HyG was associated with impaired transfer
Atwell, 2019 [24]	Papua New Guinea (two cohorts)	157; 143	n/a	No	Yes	n/a	n/a	Primigravity, maternal age, maternal malnutrition not associated with transfer	1.19 overall (1.09 HyG+, 1.31 HyG-);1.22 overall (0.96 HyG+, 1.24 HyG-)	n/a	HyG,but not PM, was associated with impaired transfer and lower cord titer
Patel, 2019 [72]	Botswana	316	Yes	n/a	n/a	n/a	Yes	n/a	1.02 for HEU infants; 1.15 for HUU infants	n/a	Higher newborn birthweight and undetectable maternal antenatal viral load were associated with improved transfer
Pou, 2019 [56]	Sweden	78	n/a	n/a	n/a	Yes	n/a	n/a	n/a	81.6 term; 66.3 preterm	Preterm and term infants had similar repertoires of maternal IgG, but lower RSV-specific MatAb and shorter half-lives
Chu, 2020 [54]	US:Seattle; Alaska	57; 75	n/a	n/a	n/a	Yes	Yes	n/a	1.15 Seattle,1.04 Alaska	n/a	RSV Ab transfer was lower in Alaska compared to Seattle; in Alaska, pairs with CMR < 1.0 were more likely to be LBW or preterm
Yildiz, 2020 [51]	Turkey	127	n/a	n/a	n/a	n/a	Yes	Mode of delivery, infant sex, gestation week, parity, gravidity not associated with transfer	1.22 overall; 1.10 (LGA), 1.25 (SGA), 1.29 (AGA);	n/a	Median CMR was higher among AGA infants compared to SGA and LGA infants
Alonso, 2021 [70]	Mozambique	341	Yes	No	n/a	No	n/a	P. falciparum exposure, independent of PM, not associated with transfer	n/a	n/a	Maternal HIV infection was associated with impaired transfer of total RSV IgG, but increased transfer of IgG2

Abbreviations, RSV: respiratory syncytial virus, Ab: antibody, N pairs: number of mother-infant pairs, HIV: Human immunodeficiency virus, PM: placental malaria, HyG: hypergammaglobulinemia, CMR: cord to maternal titer ratio, MatAb: maternal antibody, ABW: adequate birthweight (≥2500 g), LBW: low birthweight (<2500 g), HEU: HIV-exposed uninfected, HUU: HIV-unexposed uninfected infants, LGA: large for gestational age, SGA: small for gestatinoal age, AGA: appropriate for gestational age.

Potential inhibitors of transplacental transfer of maternal antibodies

RSV is one of only a handful of diseases for which purpose-built maternal vaccines are in late-stage development. However, only a few published studies have specifically evaluated transplacental transfer of naturally-acquired maternal RSV antibodies in various settings, and in particular, the factors that may impair transfer. These studies are summarized in Table 1 and described in the sections that follow.

Optimal transplacental transfer of maternal antibody requires a healthy, full-term pregnancy (≥37 weeks’ gestation) and a functionally intact placenta. Gestational age (GA) at delivery, birthweight, maternal infections, chronic conditions, and environmental exposures have all been shown to have a negative impact on this process (Box 1). The following sections assess what is known about such factors and what gaps in knowledge should be prioritized for future research.

Note: Impairment of transfer for any IgG, not RSV antibody-specific.

Note: As reported in the scientific literature and not in order or importance or magnitude of effect; as many factors listed above occur simultaneously, challenges remain in understanding independent effects.

a>1 study in scientific literature confirmed findings.

bIndependently, or through inducing hypergammaglobulinemia.

cInconsistent results in scientific literature.

dAddressed in too few studies, more research needed.

Pre-term birth and birthweight

Many studies have demonstrated that infants born early have reduced maternal antibody levels at birth compared to full-term infants in the same setting for a variety of antigens. Given that the majority of maternal IgG is transferred during the final weeks of normal gestation, preterm infants may be at a substantial disadvantage in terms of their maternal antibody concentrations and passive protection from disease. Cord blood antibody levels between 28 and 33 weeks may be only half of maternal levels [19]. Furthermore, insufficient protection from maternal antibody could compound the increased risk of infectious disease conferred by any physiologic underdevelopment in preterm infants.

Preterm birth has been associated with reduced transplacental transfer of antibodies for measles, rubella, varicella zoster, mumps, Haemophilus influenzae B (Hib), diphtheria, tetanus toxoid, pertussis, and polio [52]. Few published studies have evaluated this for RSV, but one study of 213 infants found that those born < 37 weeks (both preterm low birthweight and preterm adequate birthweight) had lower CMRs for RSV compared to term infants (both term-low birthweight and term-adequate birthweight) in The Gambia [53]. A more recent study assessing RSV CMR among 75 Alaska Native mother-infant pairs found that infants with CMR < 1.0 were more likely to be lower birth weight or preterm than infants with CMR > 1.0 [54]. Similar findings were noted in a recent study of pregnant women in Kenya [55]. In contrast, among a prospective cohort of 310 infants in rural Nepal, they did not find a statistically significant difference in RSV antibody transfer by GA in weeks, or between term and pre-term infants [27].

A recent analysis of 32 early preterm (<30 weeks GA) and 46 full-term mother-infant pairs in Sweden evaluated maternal antibody repertoire and virus-specific antibody concentrations and neutralization activity. Findings differed by virus, but this study found that although preterm infants had a similar full maternal antibody repertoire to full-term infants, they had reduced concentrations of RSV-specific antibodies. Furthermore, infant RSV antibody concentrations were associated with GA but not maternal titers [56]. Despite reduced RSV antibody concentrations in preterm infants, authors noted that neutralization activity in the first three months of life appeared to be similar in both term and preterm infants. This work indicates further research is needed to better understand specific determinants of antibody specificity for the FcRn and underscores that quality and specificity of the transferred antibodies may be more important than CMRs or composition of the full antibody repertoire [56]. Importantly, most studies of RSV antibody transfer and gestational age categorize infants as pre-term or term, using only two categories. More granular data evaluating transfer by week of gestational age would be highly valuable, not only to better understand the biology, but also to help inform optimal timing of RSV immunization in pregnancy.

Though the mechanisms are likely multifactorial, low birthweight may also be associated with reduced RSV antibody transfer. Several studies have shown that term infants born underweight or small tend to have reduced transplacental antibody transfer compared to adequate birthweight infants for a variety of bacterial and viral pathogens. However, as with GA, few studies have evaluated the associations between low birthweight and small for gestational age with impaired transplacental transfer of RSV-specific antibodies. In the Gambia study described above, pre- and term-low birthweight infants demonstrated decreased transplacental antibody transfer for RSV, as well as other vaccine-preventable pathogens, compared to term-adequate birthweight infants [50], [53]. A recent study conducted in Turkey among 127 infants born ≥ 37 weeks GA also found that the median CMR among appropriate for gestational age weight infants was higher than in small for gestational age infants, as well as large for gestational age infants [51]. Conversely, reduced transplacental transfer of RSV antibody in term-low birthweight infants or small for gestational age infants was not observed in Nepal [27]. Similarly, a study in Dhaka, Bangladesh, in a more affluent and urban setting, found RSV antibody transfer was not lower in small for gestational age infants compared to adequate birthweight infants [26]. In both Nepal and Bangladesh, the observed proportion of small for gestational age infants was high; 50% of study participants in Nepal and 34% in Bangladesh were categorized as small for gestational age. The reasons for these contradictory findings are unclear, but it is possible that global standards for infant weight/length are not sufficiently representative in South Asia; methodological differences could also contribute. Additionally, in both settings, underlying reasons for low birthweight or small for gestational age may be relevant to transfer efficiency.

HIV, placental Malaria, and hypergammaglobulinemia

Three maternal conditions have been associated most frequently in the literature with impaired transfer of antibodies: maternal HIV infection, malaria in pregnancy, and excessive levels of maternal total IgG (hypergammaglobulinemia) [19], [22], [24], [25], [29], [41], [49], [52], [57], [58], [59], [60], [61], [62], [63], [64], [65], [66], [67], [68], [69], [70]. In addition to being prevalent in many regions of the world where RSV morbidity and mortality are high and prevention products are needed, these three conditions frequently occur together, which can complicate the assessment of each factor’s independent associations with antibody transfer. Recent studies have begun to disentangle the relative contributions of each.

: Numerous studies in LMICs, including Bangladesh [71], Botswana [72], Brazil [58], [73], Cameroon [74], Kenya [48], [55], [60], [61], [75], Malawi [59], Mozambique [70], Nigeria [76], and South Africa [77], [78], [79], [80], [81] have evaluated differences in maternal antibody levels and CMRs for various pathogens in cohorts of women with and without HIV. While there is some variation in findings across study settings, most show HIV-positive pregnant women have lower specific antibody titers during pregnancy compared to their HIV-negative peers and demonstrate impaired transplacental transfer, resulting in lower mean CMRs, lower mean antibody concentrations, and/or higher proportion of infants born with antibody levels below protective thresholds among their infants compared to those born to HIV-negative women.

Several studies of RSV-specific antibody transfer have found evidence of impaired transplacental transfer in HIV-positive women and their exposed but uninfected infants compared to HIV-negative women and their HIV-unexposed, uninfected infants. This was the case in Botswana, where HIV exposed but uninfected infants had both reduced RSV CMR and lower RSV cord titers compared to HIV unexposed, uninfected infants [72]. In Brazil and Kenya, CMRs were also lower among HIV exposed but uninfected infants, even when maternal antibody levels for RSV in HIV-infected women were the same or higher than antibody levels in HIV-uninfected women [48], [73]. Not all studies have evaluated antibody subclass; this may also be relevant. In Mozambique, maternal HIV infection was associated with a reduction in total RSV-specific IgG transfer, but an increase in IgG2 transfer [70].

Although not the focus of this review, an understanding of the mechanisms by which maternal HIV infection may impair transplacental transfer of antibodies for naturally acquired and vaccine-induced antibodies is critical to informing ongoing research. One proposed mechanism is through the impairment of maternal immune responses, leading to lower maternal concentrations of specific antibodies and increased susceptibility to infectious diseases [73], [82], [83], [84], [85]. In addition to modulation of B-cell activity and maternal immune response to infection and vaccination, HIV has also been shown to cause inflammation of the placenta, which may further compromise antibody transfer [19]. Further studies are needed to better understand the immunologic underpinnings of these phenomena and how they relate to RSV antibody generation and transfer.

In addition to impairing the maternal immune response, HIV infection may also lead to inefficient transplacental transfer through the induction of hypergammaglobulinemia. Many pathogens, including HIV and malaria, are known to cause B-cell immunologic defects, including polyclonal B-cell activation leading to an overabundance of total IgG (discussed below) [82], [83].

Several studies have also evaluated associations between malaria in pregnancy, specifically placental malaria, and impairment of transplacental transfer of antibodies for measles, varicella, Hib, S. pneumoniae, and tetanus in the Gambia, Malawi, Kenya, and Papua New Guinea, but findings have been less consistent than those for HIV [22], [24], [25], [49], [59], [61], [62], [70], [86].

Placental malaria may impair transplacental transfer due to direct effects of parasitic infection of the placental interface, including inflammation and (depending on GA and duration of infection) alteration of placental architecture, resulting in reduction in surface area and, therefore, transfer of antibodies, potentially leading to intrauterine growth restriction and/or pre-term birth [87]. However, growing evidence supports that, as with HIV, induction of hypergammaglobulinemia may also play a leading role in the observed associations between malaria and impaired transplacental antibody transfer [19], [24], [25], [29], [88], [89].

: High maternal total IgG levels have been associated with reduced CMR for both total and antigen-specific antibodies. Various thresholds developed with data from high income countries (HICs) have been used to define hypergammaglobulinemia (concentrations > 1500, >1600 or > 1700 mg/dL are commonly used) [19], [24], [25], [89]. In hypergammaglobulinemic women, high levels of non-specific and often low-quality IgG are produced and thought to either saturate the finite number of Fc receptors at the placental interface (in effect causing a ‘traffic jam’), or outcompete other IgG with lower affinity for the FcRn [19], [22], [25], [49], [66], [89] (Fig. 2).

Fig. 2

Mechanisms by which hypergammaglobulinemia may impair transplacental transfer. Created with BioRender.com. Panels illustrate normal placental interface (left) and placental interface with hypergammaglobulinemia (right); in the presence of hypergammaglobulinemia, high levels of non-specific and often low-quality IgG are produced and thought to either saturate the finite number of Fc receptors at the placental interface (in effect causing a ‘traffic jam’), or outcompete other IgG with lower affinity for the FcRn.

While hypergammaglobulinemia has been recognized as an important factor in the impairment of transplacental antibody transfer for some time [22], [48], [49], [58], [59], a growing body of work has been able to elucidate its significance more clearly in recent years. Several studies now indicate that the reduced transplacental transfer historically documented as occurring in women infected with malaria and HIV may be due, at least in large part, to hypergammaglobulinemia, whether independent of or induced by those infections [24], [25], [48], [74].

Understanding the independent effects of HIV, placental malaria, and hypergammaglobulinemia is complicated by the geographic prevalence of these conditions. Previous studies of HIV and placental malaria did not measure total maternal IgG [62] or were unable to distinguish the conditions due to high prevalence of both conditions in the study populations (overall and within individuals) [22], [49]. This was not the case in a more recent study of 300 mother-full-term-infant pairs in Papua New Guinea [25]. In contrast to previously noted studies in The Gambia, where malaria and hypergammaglobulinemia were highly prevalent and overlapping (95% of hypergammaglobulinemic women also had placental malaria), in the more recent Papua New Guinea studies, only 56% of women with hypergammaglobulinemia also had placental malaria. This permitted evaluation of the relationships between impaired antibody transfer and the two exposures of interest independently of one other. In two temporally distinct cohorts of pregnant women in the same geographic area, the proportion of infants with impaired transplacental transport of RSV antibody was similar, despite substantial reduction in the prevalence of placental malaria. In multivariable models, high maternal total IgG was associated with impaired RSV antibody transfer, but placental malaria was not [25]. An additional analysis of 325 mother-infant pairs from Papua New Guinea corroborate these findings [24]. More research is needed to better understand the causes and prevalence of hypergammaglobulinemia, as well as the mechanism by which it interferes with antibody transfer and how that may vary across antibody subclass or specificity.

COVID-19 and acute infections

Emerging data from pregnant women with COVID-19 may advance our understanding of how acute infections during pregnancy affect transplacental transfer dynamics. At least one study has described both impaired SARS-CoV-2-specific antibody transfer and elevated total IgG (though not clinical hypergammaglobulinemic) among women with SARS-CoV-2 infection in pregnancy [42]. However, the impact of such infection-induced increases in IgG was unclear, as total and non-SARS-CoV-2-specific antibody transfer was intact in women with COVID-19 during pregnancy, and SARS-CoV-2-specific transfer was impaired only among those with COVID-19 in their third trimester. This study also found that differences in Fc glycosylation across the SARS-CoV-2 antibody repertoire may be related to transfer efficacy, and interestingly, that potentially compensatory changes in placental Fc receptors may occur in the context of acute maternal infection. It is unclear what the long-term implications of SARS-CoV-2 infection in pregnancy will be, particularly in settings where vaccination coverage is high. However, findings from COVID-19 during pregnancy may shed light on the impacts of other acute infections (e.g., Zika, influenza) that have been previously understudied or underappreciated.

Maternal age, weight, parity, type of delivery, malnutrition, and smoking status

Other factors, including maternal age, weight, parity, type of delivery (vaginal or caesarian), nutritional status, and smoking status have also been evaluated for potential associations with impaired transplacental antibody transfer. There is a lack of compelling evidence for such relationships, except for one study in Nepal that noted increased RSV antibody transfer ratios along with increasing parity [27]. Additionally, one study of immune response to Hib vaccine administered during pregnancy to malnourished and non-malnourished women documented reduced vertical antibody transfer among the former group, even though there was no difference in maternal antibody response between groups [90]. The reasons for this observation are unclear, and associations of impairment and such variables are likely not independent of one another or other factors, including those discussed above. If malnutrition was associated with impairment, it could be due to placental size, morphology, or vascular development [29], [68]. Apparent associations with malnutrition might also be confounded by the presence of unmeasured infection or immune dysregulation.

The number of studies evaluating the effects of maternal cigarette smoking during pregnancy on transplacental transfer is limited [27]. However, several studies have demonstrated detrimental impacts of smoking on placental growth and development, which has important implications for transfer efficiency. Maternal smoking and nicotine exposure have been shown to lead to premature descent of the placenta, increased trophoblastic basal membrane thickness, dysregulation of cytotrophoblasts, and reduced vascularization [91], [92]. Importantly, the latter can result in placental insufficiency, which would impact the surface area of the placenta, thereby limiting transfer between the mother and fetus. Considering these morphological consequences, there is a need to include maternal smoking in future assessments of transplacental transfer impairment.

Diabetes

Diabetes has been evaluated for its potential to impair antibody transfer, but findings are unclear. Two studies demonstrated increased IgG transfer for multiple bacterial pathogens, including GBS, among hyperglycemic mothers [29], [93], [94], but a more recent study found that total IgG, expression of FcRn, and transfer of IgG1, IgG3, and IgG4 were lower in women with pre-existing type II diabetes compared to women without diabetes [95]. To further complicate understanding of these relationships, women with mild gestational hyperglycemia were shown to have increased expression of FcRn and higher transfer of IgG3. It has been proposed that alterations in placental structure that facilitate passage of glucose among women with gestational hyperglycemia may also facilitate binding or transport of IgG, and that the decreased expression of FcRn in patients with diabetes may explain the reduced transfer of some IgG subclasses [29]. Furthermore, high levels of glycosylated IgG have been detected in the plasma of patients with diabetes, which may have important implications given that a growing body of literature indicates that glycosylation is involved in antibody avidity and binding to FcRn as discussed above [29].

Priorities for future research

Maternal immunization products for the prevention of RSV infection are expected to be available within the next two to five years [12], [13], [14]. The only candidate to complete phase III trials (though not licensed), was shown to elicit efficient transplacental transfer of binding, palivizumab competitive, and neutralizing antibodies to the fetus [15], [96]. CMRs among infants born to RSV vaccine recipients in the study were consistently > 1, although they were reduced among recipients from LMICs compared to HICs [15]. In countries with high prevalence of health conditions discussed in this review that may result in differential efficiency of transplacental antibody transfer, these differences may be even more pronounced. Post-licensure studies will be critical to assess vaccine effectiveness in multiple geographic settings and among representative and generalizable populations of pregnant women to guide vaccine policy decision-making and to ensure optimal effectiveness globally. While RSV is an immediate priority, given that additional vaccines for use in pregnancy are likely to be developed in the future (e.g., GBS), additional epidemiologic studies of these mechanisms that are not pathogen specific and not focused on vaccine-acquired antibodies are also prudent.

Among the discussed above, the biggest data gaps that appear critical to ensuring optimal transfer relate to understanding of hypergammaglobulinemia, the conditions that can cause it, and the mechanism(s) by which it results in impaired transfer. To inform future RSV and other maternal immunization programs and strategies in LMIC settings, several key questions about this condition should be addressed.

First, future studies should prioritize confirming hypergammaglobulinemia’s role in reduced transplacental antibody transfer in general and specifically for RSV. Such studies should be conducted across a range of geographic locations and be capable of investigating hypergammaglobulinemia both independent of and along with other factors previously shown to be associated with reduced transfer (e.g., HIV, PM, preterm birth, etc.). Sub-analyses of data from ongoing and future clinical trials could also be informative. However, given the stringent inclusion and exclusion criteria for participants in such clinical trials and that enrollment is largely from high- and middle-income countries, very few participants would be expected to have the conditions of interest, therefore, such studies are unlikely to provide all the necessary data to fully address these questions.

Secondly, more research is needed to understand the epidemiology of hypergammaglobulinemia, including geographic prevalence and what factors or comorbidities are associated with development of excessive IgG, and therefore, how the condition may be addressed via public health action. Few data exist describing how common hypergammaglobulinemia is, especially in LMICs. Studies assessing the range of total IgG among adults, and particularly among pregnant women across multiple settings and in relation to underling conditions, chronic infections, or comorbidities would be valuable. Such assessments could be added to large prospective or cross-sectional studies for other or similar outcomes. While vaccine clinical trial data are unlikely to provide much insight, measuring total IgG within epi studies and post-licensure analyses would be both feasible and valuable. There may also be opportunities to evaluate banked specimens previously untested for total IgG.

More research is also needed to understand the underlying factors that are associated with and/or cause hypergammaglobulinemia. Chronic infections such as HIV and placental malaria have been associated with elevated total IgG levels in some studies [22], [24], [25], [49], [50], [74], [83], but these and other infections are not the only (or perhaps even the most significant) conditions or exposures that may lead to hypergammaglobulinemia at the population level. Studies conducted within environmental health sciences have demonstrated clear associations between arsenic exposure and high levels of total IgG, and increased susceptibility for arsenic toxicity during pregnancy [97], [98]. It is possible that failure to appreciate and measure environmental factors potentially associated with hypergammaglobulinemia has limited our understanding of observed differences in CMRs between populations. These exposures may be highly relevant in LMICs, and perhaps also in HICs, particularly in rural areas. The maternal immunization field may benefit from increased collaboration with researchers in adjacent fields, such as environmental epidemiology.

In addition to improved understanding of the epidemiology and mechanisms of action, more research is needed to assess if prevention or treatment of chronic infections that are associated with the development of hypergammaglobulinemia could reduce or potentially mitigate its effects on transplacental antibody transfer. In the case of HIV, treatment with antiretroviral therapy (ART) may be helpful, but the magnitude and duration of effects are unclear [72], [83], [84], [99], [100]. One study in Malawi showed that 24 months of ART in HIV-positive pregnant women was associated with reduced total IgG over time but may not result in reduction of total IgG levels below the threshold of hypergammaglobulinemia (in this case, 1500 mg/dL). Furthermore, HIV is associated with altered IgG subclass distributions. HIV has been associated with IgG1 concentrations exceeding 90% and extremely low IgG2 concentrations (∼5%, compared to a more typical proportion of ∼ 20%). ART only partially restored that balance [83]. Shorter courses (7–10 months) had a similar effect, but the reduction in IgG production was not maintained over time [83]. Other studies indicate that these effects may vary geographically; studies of how ART might improve B-cell-related immunologic dysfunction have shown greater benefit in studies of HIV + women in developed countries than they have in African cohorts [83].

Perhaps most importantly, future observational and post-licensure epidemiologic studies will be needed to assess the extent to which reduced transplacental transfer of antibodies translates to diminished vaccine effectiveness and/or duration of protection among infants, for RSV (overall and subtype-specific) and other pathogens. The current lack of clear correlates of protection for some pathogens (including RSV) will be a challenge to such research and may necessitate disease outcome endpoints rather than immunobridging, but given that the absolute concentration of maternal antibody that reaches the infant may be of more practical importance that the ratio, and thus the presence of sufficiently high maternal titers could potentially mitigate the impact of factors that impair transfer, this should be thoroughly investigated. Studies should be conducted to understand the potential for reduced efficacy among various antigens, and in the case of RSV specifically, for naturally acquired and vaccine-induced antibody separately, as future maternal vaccines may elicit antibody repertoires with different overall or subtype-specific neutralizing activity, higher absolute titers, or differential half-lives compared to naturally acquired antibody, any of which could significantly attenuate the impact of reduced CMRs. While establishment of a clear immunologic correlate of protection for RSV has been elusive, ongoing clinical trials and associated efforts in the field may provide more clarity in the coming years.

Summary

Transplacentally-transferred maternal RSV antibodies can protect newborns from RSV in early life, and vaccines for maternal RSV immunization are in clinical development. Achieving maximum benefit from maternal immunization hinges on optimizing both the maternal immunological response to immunization and the transplacental transfer of effective antibodies to the newborn. In this review we have explored the available evidence on biological factors that are or may be associated with impaired transplacental antibody transfer and highlighted critical knowledge gaps that must be addressed to ensure optimal use of maternal immunization for RSV globally. We note that hypergammaglobulinemia may be an especially important and under researched factor.

Disclaimer

The authors alone are responsible for the views expressed in this article and they do not necessarily represent the views, decisions or policies of the institutions with which they are affiliated.

Funding

This work was supported by the Bill & Melinda Gates Foundation, Seattle, WA [grant number OPP1114766].

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Definitea

Maternal HIV infectionb

Maternal hypergammaglobulinemia

Maternal malaria infection, notably placental malariab

Pre-term birth

Probablec

Acute maternal infectionsb

Low infant birthweight

Maternal malnutrition

Parity

Possibled

Environmental exposuresb

Maternal age

Maternal diabetes

Maternal smoking status

Maternal weight