To retrieve information for this review, a PubMed based research was conducted using the medical subject heading database terms “vaccination” OR “immunization” AND “pregnancy.” In addition, maternal, fetal, immune system, placenta, neonate were used as search terms. We also included the currently allowed, contraindicated and in development vaccines for pregnant women, as well as the recent evidences on COVID-19. Filters for humans, any publication date and articles in English language were applied. For this narrative review, evidence was included from randomized clinical trials, original research and observational studies, case series, position statements, systematic reviews, meta-analysis studies, and selected reviews addressing the covered questions or cross-references from these publications.

Early in pregnancy, the maternal immune system undergoes a timely regulated remodeling to allow the implantation, preservation, and growth of the semi-allogenic fetus while protecting against pathogens (13–15). Maternal and fetal immune systems establish a cooperative status (16) depending on a delicate balance between anatomic, endocrine, metabolic, and microbiome factors (17). Of note, in pregnant women, the ability to mount an adequate antibody response and immunologic memory following an infection or vaccination is not affected (18–20).

Main changes in the maternal immune system can be summarized into three phases: the 1st trimester requires a strong proinflammatory state to guarantee the implant of the blastocyst in the uterus. This process implies the break of the epithelial lining of the uterus, damage of the endometrial tissue and a rearrangement of the endothelium and vascular smooth muscle of the maternal blood vessels to establish an adequate placental–fetal blood supply (21). All these activities require an inflammatory environment to secure the adequate repair of the uterine epithelium and the removal of cellular debris; in the 2nd and 3rd trimester, the anti-inflammatory response prevails, allowing tolerance of the semi-allogenic fetus, its rapid growth and development. Finally, the switch to a new proinflammatory state favors the cascade of events that synergistically lead to parturition (22, 23). Specific cell populations mediate the transition from the pro- to the anti-inflammatory state and viceversa. First, T lymphocytes take part through a shift from a type 1 T helper lymphocytes (Th1) response, oriented toward cell-mediated immunity, toward a type 2 T helper lymphocytes (Th2) response, which favors humoral immunity by stimulating B cells to increase the production of maternal antibodies (24). Th2 lymphocyes are fundamental for the mother-child immunological tolerance up to the time of delivery, suppressing cytotoxic T cells activity (25). Moreover, T regulatory cells (Tregs), a sub-type of CD4+ T cells that express CD25, increase in pregnancy and contribute to the maintenance of immunological tolerance to the fetus by actively suppressing self-reactive lymphocytes (26, 27).

In pregnancy, the maternal immune barrier to infections enlists a combination of signals and immune modulators originating from the fetoplacental dyad (28), including the placenta and decidua. In fact, the placenta is an active site of innate immune response capable of reacting to pathogens by the release of antimicrobial peptides and cytokines (13, 16). Nevertheless, a second important site of innate immune responses is localized at the decidual layer. The maternal-fetal interface represents the direct contact between the embryo and the mother and it is populated by fetal trophoblast cells, maternal decidual stromal cells (DSCs), and decidual immune cells (DICs) (29). The decidua is an active site of chemokine synthesis throughout gestation, attracting neutrophils, natural killer cells, dendritic cells, and macrophages (17). However, the transfer of innate inflammatory mediators from the mother to the fetus is less well-understood (16). Finally, early in the second trimester, the maternal immune system protects the fetus by passive transfer of maternal antibodies (IgG) and maternal immune cells across the placenta (30).

Recent technological advances have provided critical information in the field of early life human immunology with its timely and spacely development (31). The human immune system starts developing after the first 2–3 weeks of fetal life, when hematopoiesis and generation of pluripotent and self-renewing hematopoietic stem cells (HSC) begin in the yolk sac and the aorta-gonad- mesonephros region of the embryo (32). Subsequently, HSC migrate to the fetal liver, representing the major hematopoietic site during fetal life, as localization into the bone marrow occurs by week 20 (33). T cell differentiation and maturation occur in the thymus. During embryogenesis, immune cells also inhabit peripheral organs, including skin, intestine, kidney, and lung, and participate in the respective organ environment (34).

Innate immune cells are the first to emerge: granulocytes, NK cells and lymphocyte precursors are detected in the fetal circulation between gestational week (GW) 8–10 (35). From GW 8, HSCs migrate to the thymic rudiment and, by GW 16–20, mature T cells are released into the peripheral blood (36). Lymphoid precursor cells develop into B-lymphocytes with subsequent functional maturation in secondary lymphoid tissue (e.g., lymph nodes and spleen) toward the end of the first trimester (37). Thus, during the second trimester, immune cells, predominantly lymphocytes, grow in number, mature, and differentiate. Toward the end of the third trimester, the fetal immune system is functional and capable of producing a response (38).

Diverse immune cell types develop and mature at various gestational stages, which is necessary to establish feto-maternal tolerance and functional responses according to specific demands. The fetal and neonatal T-cell response is shifted toward the suppression of cytotoxicity: with a strong Th2 polarization, a dominant anti-inflammatory cytokine profile (39) and Tregs dominating the fetal circulation, suppressing reactivity to non-inherited maternal antigens (40).

At birth, the infant's response to Toll-like receptor (TLR) (i.e., innate immune response) is characterized by high production of IL-10, IL-6 and IL-23, which induce IL-17-producing helper T cells (Th17 cells) (41). The predominance of a Th17-like pattern, combined with considerable IL-10 production, may contribute to diminished T helper type 1 (Th1) responses, resulting in greater susceptibility to intracellular infections and diminished vaccine responses during infancy (42).

Since delivery, exposure to antigens and environmental challenges during infancy and childhood will promote further adaptation and expansion of the immune system (43). The exposure to the antigen rich world during the first weeks after birth leads to a massive increase in lymphocytes in a way that is largely independent of gestational age (GA) at birth. B and T cells are abundant at birth, followed by a gradual decrease over the first years of life until adulthood. As expected, the newborn presents with a lower number of memory B cells compared to older ones, reflecting the lack of exposure to foreign antigens during fetal life (33). The frequency of switched and non-switched memory B cells increases slowly with age and reaches adult levels in children 10–15 years of age (44).

Briefly, the human immune system follows a precise developmental trajectory with extraordinary plasticity in response to the different demands. However, after antigen exposure, specific protective immunity takes time; thus maternal immunization represents a great opportunity to reduce the risk of disease in the mother, fetus, and infant (30).

Immunization of the mother during pregnancy increases vaccine-specific antibodies to ensure effective protection not only of the mother, but also the offspring in first months of life (19) (Figure 1A). The concentration of maternal antibodies in the serum of the neonate determines the effectiveness of protection. Understanding the mechanisms by which IgGs are transferred across the placenta is important to develop optimal maternal and infant immunization strategies and make decisions about the timing of vaccination in pregnancy (45, 46).

The transfer mechanisms depend on various factors, including the syncytiotrophoblast and the capillary barrier. IgGs are selectively transported from the maternal blood to the fetus via neonatal Fc receptor (FcRn) expressed in syncytiotrophoblast of the placenta (47). FcRn is a major histocompatibility complex (MHC) class I-related Fc receptor involved in the bidirectional transcytosis of IgG and IgG immune complexes across various human epithelia. It binds the Fc fragment of Ig antibodies at acidic pH (48). In maternal blood, with a physiological pH of 7.4, IgG cannot bind the FcRn at the apical side of the syncytiotrophoblasts and need to be transported by endocytosis. In the acid environment of endosomes, FcRn-IgG complexes are then carried to the fetal side of the syncytiotrophoblast, where IgG is released upon exposure to normal pH (30). IgG subsequently crosses the villous stroma and fetal capillary endothelium and enters fetal circulation, although this mechanism is not fully understood (49).

Several factors affect the IgG transfer from the mother to the fetal circulation, i.e., the amount of FcRn expressed by syncytiotrophoblasts and total maternal IgG levels. Indeed, it was observed that when the amount of maternal IgG reaches a level of 15 g/L, FcRn is saturated and maternal antibodies stop to be transferred (50).

Placental transfer of IgG also depends on gestational age. The process starts during the first trimester of pregnancy and increases exponentially over time (51). Most of the IgG transfer occurs after 28 weeks of gestation (52). As pregnancy progresses and the placenta grows in cell mass, the expression of FcRn might increase with relatively higher antibody transport rate (53). Although in preterm infants, the reduced period of gestation leads to a lower and less efficient IgG placental transfer, several studies support the efficacy of early maternal vaccination in reducing the risk of infections at birth (54, 55). Recent UK guidance recommended pertussis immunization from 20 GW to optimize protection of preterm infants (56).

Different specificities of IgG antibodies are passively transferred to the newborn, depending on maternal immunological experience, and on the type of vaccine administered in pregnancy (57). The transfer efficacy through the placenta depends on the IgG subtype affinity for the FcRn receptor (58). IgG1 is the subtype of antibodies transferred most efficiently to the fetus, followed by IgG4, IgG3, and IgG2 (52). This differential ability of immunoglobulin to be transferred is important in the context of vaccination in pregnancy. While IgG1 and smaller amounts of IgG3 and IgG4 are induced predominantly by vaccines containing protein antigens, such as tetanus, dipheteria, pertussis (Tdap), IgG2 with some IgG1, are induced predominantly by vaccines containing polysaccharide antigens, such as those against Haemophilus influenzae type b or Neisseria meningitidis (59, 60).

Previous maternal infections are another factor that can impact the specific and amount of maternal IgG antibodies to be transported across the placenta (45), as reported in mothers infected by human immunodeficiency virus-1 (HIV-1) or malaria (61–63). To provide greater protection to the infant, specific immunization programs should be developed in childbearing age or in pregnant women with chronic infections (60).

Breastfeeding represents another mechanism by which the mother protects the neonate against pathogens in early life. Indeed, maternal secretory IgA (sIgA) and, to a lesser extent IgG and IgM, are secreted in the colostrum and breast milk and seed the luminal tract of the newborn, conferring an immunological benefit (64). In particular, sIgA provides defense at the mucosal level against pathogens of the gastrointestinal and respiratory tract through direct neutralization, inactivation, and prevention of adherence to epithelial cells of toxins and other virulence factors (65).

Several studies on the protective role of breast milk in women vaccinated during pregnancy identified a high amount of vaccine-specific sIgA in the breast milk samples up to several weeks postpartum (66). Not surprisingly, a lower incidence of respiratory illness with fever episodes in young infants of influenza-vaccinated mothers was also reported (67).

Of note, different maternal immune cell populations, including dendritic cells, macrophages, natural killers (NK), memory B and T cells are also detected in breast milk. It has been suggested that these cells may modulate the developing neonatal and infant immune system (68) (Figure 1B).

Despite evidences of the protective role of maternal IgG antibodies, there are concerns about their possible interference with the infant antibody response to the vaccines administered in the first months of life. It has been suggested that the presence of maternal antibodies might result in a decreased antibody production by the infant after vaccination and, consequently, in decreased protection for the child (11). This phenomenon, called “blunting,” has been associated not only to the antibodies transferred by the mother to the fetus after vaccination, but also to those generated in response to natural infection before or during pregnancy. Immune blunting has been observed in several studies regarding tetanus, diphtheria, pertussis, influenza, measles, and mumps maternal vaccination (69–71), but results are still controversial (72–74), etiopathogenesis and clinical relevance are still to be clarified (50, 75). Indeed, low antibody levels post-vaccination do not necessarily imply a low degree of protection (76). Children born to immunized mothers and vaccinated against diphtheria, tetanus, Polio, Hep B, and Hib had lower antibodies levels than children of non-vaccinated mothers. However, specific antibodies reached the seroprotective levels for all antigens (73, 77). In addition, no evidence of a clinically significant effect of blunting has been found in the United States (US) and the United Kingdom (UK), where maternal Tdap vaccination has been implemented since 2011 (US) and 2012 (UK) (78, 79).

Protection induced by vaccines may be related not only to antibody concentration, but also to their functional characteristics, such as avidity and neutralization potential, and to the generation of memory B cells, all aspects not extensively assessed yet (77). Finally, the effect of maternal antibodies on cellular immune responses after infant vaccination is another interesting aspect to consider. A recent review reporting five studies on humans concluded that pre-existing maternal antibodies do not affect infants' cellular immune response after primary vaccination (70). Specific T cell responses have been observed after immunization in the presence of maternal antibodies (71, 80). Also, maternal antibodies could prime the infant immune system to develop a stronger and long life cellular response, the so-called “prime-boost” effect (81, 82).

The fetal immune system can be primed and responds to foreign antigens during pregnancy. Increasing evidence suggests that maternal infection or vaccination during gestation may shape and train the fetal immune system even in the absence of fetal infection, affecting infants' immune responses to pathogens and vaccines during the extrauterine life (83) (Figure 1A).

The exact mechanisms by which the fetal immune system can be primed by antigen in the absence of fetal infection remain unclear. A possible explanation is the vertical transmission of a low level of antigens from the mother to the fetus. A second hypothesis is that maternal cells carrying the antigen or antigen-loaded microvesicles cross the placental barrier (83, 84). Actually, maternal cells are found in human fetal tissues from the second trimester of pregnancy onwards (85, 86). This natural phenomenon is referred to as microchimerism (Mc). It is defined as the presence in an organism of a small number of cells or amount of DNA deriving from a genetically different individual. The Mc origins from bi-directional cell exchange during pregnancy (87).

Among the accepted functions of Mc, that is worth mentioning is promoting feto-maternal tolerance and improving the outcome of future pregnancies. Feto-maternal tolerance is the process that allows fetal and maternal cells and tissues to avoid immune rejection and coexist. Fetal immune components, exposed to maternal tissues, tolerate foreign non-inherited maternal antigens (NIMAs). On the other hand, mothers during pregnancy, encompass genetically foreign paternal antigens expressed by the developing fetus. This coincides with a bidirectional transfer across the placenta of cells that seed both in maternal and fetal tissues. The presence of maternal microchimeric cells in the fetus favors the generation of fetal Tregs, which are generated in utero against NIMAs expressed by maternal cells (85) and assure fetal tolerance suppressing fetal T cells functions.

Besides promoting tolerance, Mc might be crucial to prepare the fetal immune system to face the pathogens after birth. Despite the predominant tolerogenic context, the existence of fetal memory T cells has been reported in the fetal spleen (88) and gut (89), as well as in the cord blood (90). The generation of memory T cells requires fetal exposure to foreign antigens presented by antigen-presenting cells (APC). It has been supposed that maternal APC loaded with antigen may pass the placenta and prime the fetal immune system.

Fetally memory T cells have been also found in uninfected children born to mothers infected with human immunodeficiency virus (HIV) (91), hepatitis B virus (HBV) (92), hepatitis C virus (HCV) (93), or plasmodium (94). In endemic regions, it has been observed that maternal infection by helminths, including filariasis (95), schistosomiasis (96), onchocerciasis (97), and ascariasis (98) was associated with fetal lymphocytic responses and the consequent production of specific immunoglobulins. If fetal immune cells respond to antigens of infectious agents carried by the mother, it seems reasonable that the fetal immune system may also respond to maternal antigen exposure to pathogens. Interestingly, maternal tetanus toxoid immunization led to IgM anti-tetanus antibodies in the blood of children before neonatal immunization with tetanus vaccine, showing that vaccine antigen is available and that it activates the fetal immune system (99). Of note, influenza virus-specific T cells were detected in cord blood cells of neonates whose mothers received vaccination during pregnancy (100).

The exposure to antigens mediated by maternal cells in utero might affect the development of fetal innate immune cells as well, such as macrophages and monocytes. Not surprisingly, fetal programming of monocytes is functionally different from the adult one. It has been demonstrated that in fetal monocytes stimulation with IFN-γ, IL-6, or IL-4 generate distinct JAK/STAT signaling responses, thus triggering innate responses and antimicrobial activities rather than promoting a strong adaptive and potentially harmful inflammatory response (101). Fetal monocytes can mount a more primitive, but potentially protective, innate antimicrobial response, which may also optimize the fetus and newborn's chance of successfully combating a microbial invader.

These observations add important information to understand the potential of vaccination in pregnancy including the ability to modulate the development of the fetal innate immune system (see paragraph “Future directions” on live-attenuated vaccines).

The timing of maternal vaccination can influence the transplacental transfer of vaccine antigens and the subsequent fetal immune response. In other words, the gestational age when the vaccine is administered may have an impact on priming. A study on allergen exposure in pregnancy suggested that T-cell priming with inhalant and food allergens may occur starting from 22 weeks' gestation (102).

Fetal exposure to pathogens-related antigens during pregnancy might lead to acquired protective immune responses in the offspring. However, there is still concern that it may also cause immune tolerance, increasing infants' susceptibility to both homologous and unrelated pathogens (103, 104). Non-replicating modified vaccinia Ankara was associated with reduced responses to unrelated pathogens, indicating some degree of induced tolerance (102); Diphteria-Tetanus-Pertussis (DTP) has also been shown to induce immune tolerance to unrelated antigens 3 months after vaccination, with immuno-tolerance partly restored by concurrent or subsequent BCG vaccination (105). T cell anergy and a rise in Tregs might be responsible for immune tolerance (106). Tregs have been shown to suppress antigen-specific immune responses to malaria in infants born to mothers with infection during pregnancy (107).

Additional vaccines can be considered to protect against infectious diseases in the case of traveling to at-risk areas or close positive contacts. Anti-hepatitis A and B, pneumococcus and meningococcus vaccines are an example of this condition (25). In the same way, further vaccines such as those against tick-borne or Japanese encephalitis should be administered if the risk of these infections is high and/or in endemic areas during disease re-surgence (6, 25).

Live attenuated vaccines (LAV), including measles, mumps, rubella and varicella vaccine (MMR or MMRV), are generally contraindicated in pregnant women, as a precautionary measure, because of the theoretical risk of perinatal infection. LAV are burdened by the risk of transplacental vaccine-virus/bacterial infection. Furthermore, data regarding LAV safety during pregnancy are limited due to the exclusion of pregnant women from clinical trials (134).

However, data from accidental vaccination with MMRV in pregnant women have shown that the risk of adverse pregnancy-related events is rare (109). Evidence regarding other LAVs is lacking (135).

SARS-CoV-2 pandemic is the major global health emergency of our century.

The infection might lead to COVID-19 disease, described mainly by upper and lower respiratory symptoms, fever and in some cases by a systemic inflammatory syndrome, with high mortality (154). Pregnant women are also susceptible to the infection with a higher risk of preterm labor and pregnancy related complications (155). Little is known on the long-term consequences of the perinatal infection in the newborn.

As COVID-19 vaccination programs have been adopted by the states to deal with the health emergency, many concerns are rising regarding the safety of the licensed COVID-19 vaccines for pregnant women.

Two mRNA based-technology vaccines Pfizer/BioNTech and Moderna, and two viral vector vaccine-type, AstraZeneca and Johnson & Johnson, have been approved by most countries worldwide including European Union, United States and United Kingdom. Others are under rolling review.

Current recommendations suggest that pregnant women should be vaccinated when the benefits outweigh the potential risks (156). Thus, pregnant workers on the frontline and those with pre-existing conditions should receive the vaccine. About 20,000 pregnant women have been vaccinated for COVID-19 in USA and a smaller number in the UK (157, 158). The data is encouraging, as no major adverse effects have raised. Moreover, one case study described specific maternal IgG anti-spike protein in a newborn whose mother was previously vaccinated (159).

While awaiting more data about the risks and benefits of the vaccine during pregnancy, these encouraging studies suggest that maternal vaccination could provide newborns with some level of protection against infection.

Beyond maternal vaccines currently available, several new vaccines are in development to protect against neonatal pathogens, including respiratory syncytial virus (RSV) and group B streptococcus (GBS).

Live-attenuated vaccines (BCG, measles, mumps, rubella, smallpox, chickenpox, yellow fever) generally have the advantage of single-dose, rapid induction and durable immunity compared to inactivated vaccines. Epidemiological studies have also suggested that live vaccines extend protective effects toward unrelated pathogens (172). This unexpected pathogen-agnostic protection has been demonstrated for the measles, oral polio vaccine (OPV), BCG, and smallpox vaccines (3, 173–176). Heterologous immunity and trained immunity might favor the occurrence of an agnostic immunity that broadens beyond the specific immune response to vaccine antigens (177).

However, LAV are currently contraindicated during pregnancy because of the potential risk of causing infection and subsequent transplacental transmission of the pathogen to the developing fetus, recent systematic review and meta-analysis on maternal immunization safety with LAV mostly demonstrated no or very low risk of adverse pregnancy outcomes (135). Randomized clinical trials (RCTs) and non-RCTs studies should be performed to investigate the potential of vaccination with live vaccines in pregnancy as a tool to protect the mother and the newborn from infectious diseases.

Although national and international health authorities have defined tetanus, pertussis, and flu maternal immunization schedules, vaccine uptake remains unsatisfactory. In fact, apart from few studies in the UK and US reporting a Tdap and flu uptake between 40 and 70% (refer to these studies), in most European countries vaccine coverage rarely reaches >10% (165). Several studies have investigated potential factors affecting pregnant women's awareness and behavior on seasonal flu and pertussis vaccine compliances (178). Different cultural, social, economic, and structural barriers need to be overcome (45, 179). Within our Italian project called VaxInPerMam, at Tor Vergata University Hospital, ~1,000 childbearing Italian women have been surveyed to evaluate flu and pertussis vaccine attitude and uptake. Less than 40% of women knew maternal immunization recommendations, and, mainly due to safety concerns, <10% would receive pertussis and flu vaccines (manuscript in preparation). Misinformation represents a key factor affecting vaccine uptake. Specific intervention programs directed at both the healthcare provider and the mother-to-be are required to overcome barriers and prejudices (180).