Prevention of Fetal and Early Life Infections Through Maternal−Neonatal Immunization

Safety Concerns

The Hippocratic principle primum non nocere , first do no harm, is the supreme driving principle in all vaccination programs, but even more so in the development of neonatal vaccines. Many events and factors that occur during the neonatal period can complicate the interpretation of vaccine safety at this age. The population at risk for birth defects is estimated to be about 3% or 4%, and not all of these defects are fully apparent at the time of birth. Many neonates in the United States are discharged from hospitals and birthing centers before 24 hours of age. If a defect was present at birth but not detected until later, it might be falsely linked to a vaccine given to the neonate. Many, if not most, congenital immunodeficiencies do not declare themselves this early in life. In some areas of the world, a high percentage of infants are infected with human immunodeficiency virus (HIV), but this status is not known at the time of birth. Live-virus vaccines and the live Mycobacterium bacillus Calmette-Guérin (BCG), although generally effective, are usually contraindicated in immunodeficient individuals. Other types of congenital defects first become apparent during early infancy, a time in which many vaccinations are given. For instance, many neurologic disorders, including seizure disorders and neurodegenerative diseases, manifest in the first few months of life. Many cardiopulmonary disorders, for instance, cyanotic heart disease or cystic fibrosis, do not cause symptoms in the neonatal period. Sudden infant death syndrome (SIDS) also occurs early in life during the target period of vaccination. The cause of this fatal disorder is not well understood; therefore it is very difficult to assess the risk of exacerbation of SIDS by infection or vaccination. A large number of physiologic changes occur during the first few months of life. Young infants possess airways with very small diameters, which exhibit high resistance, especially during airway inflammation. Therefore young infants are prone to wheezing with relatively minor provocation. Infants are obligate nose breathers. Therefore a vaccine that has the potential to cause an increase in nasal secretions leading to nasal obstruction, such as live-attenuated respiratory virus vaccines, can interfere with feeding, which is a significant problem at this age. Also, infants are susceptible to intestinal intussusception during inflammation of the gastrointestinal tract, which complicates approaches such as live-attenuated rotavirus vaccines. However, careful epidemiology studies and experimental trial designs can overcome these obstacles.

There are also medical factors that complicate the evaluation of vaccines in this age group. Infants in the first 2 months of life who exhibit fever without localizing symptoms generally undergo a complete medical workup for sepsis and meningitis. A vaccine that causes even a low rate of fever in this age group will be associated with a large number of expensive and unnecessary medical workups.

Immunologic Immaturity

Neonates are clearly in transition in their immunologic development as they move from a sterile environment enveloped in the placenta, through the birth canal, into a world with vast numbers of environmental and microbial exposures. In general, it is thought that fetal immune systems are regulated in utero to avoid robust innate and adaptive immune responses to self-antigens or to maternal antigens that cross the placenta. Mouse models suggest that fetuses can be tolerized to antigens after in utero exposure, whereas human studies are more limited. Suppressive T-regulatory (Treg) cells in the fetus are generated against noninherited maternal antigens, and these cells establish functional tolerance to foreign antigens present during development in utero. For instance, one study observed Treg cells in the lymph nodes of fetal products of conception at 18 to 22 weeks of gestation that promoted maternal microchimerism (presence of maternal cells in the fetus) in 15 of 18 lymph node samples. Foreign antigens clearly do cross the placenta. Some human studies suggest that fetal adaptive immune responses to non–self-antigens are relatively intact. Overall, the evidence suggests that the fetal immune system promotes a relatively high level of tolerance, but it is not devoid of functional activity.

After parturition, a rapid transition must be made to deal with new antigens from food, the environment, and commensal bacteria, and to differentiate them from harmful microorganisms. It is likely that this transition takes time and that human neonates still exhibit some features of predisposition to tolerance, associated with persistence of long-lived Treg cells and significant evidence of B-cell tolerance. Theoretical concerns are sometimes raised that exposure to antigens early in life during a phase when the immune milieu exhibits a residual tolerogenic status might result in the infant becoming less well able to respond to the antigen rather than achieving immunologic priming for memory.

Neonates tend to make poor immune responses after infection or vaccination in terms of quantity, quality, and durability. The magnitude of antibody responses, as measured by serology, is reduced. The time to peak titer of serum antibodies is often delayed by a month or more compared with the response of older children. The level of function of neonatal antibodies often is low, for example, the neutralizing activity of antiviral antibody responses, suggesting that neonates secrete antibodies that bind but do not kill viruses. The durability of the antibody response to particular microbes made early in life also is poor. Young infants who are demonstrated to be infected with a virus early in life, as evidenced by disease and virus shedding, may seroconvert in the months after infection but then appear to be seronegative the following year. It is likely that this observation suggests a neonatal B-cell response characterized by differentiation of naïve antigen-specific cells to antibody-secreting plasma cells without induction of long-lived plasma cells or significant numbers of memory B cells. The B cells of neonates are markedly predisposed to apoptosis after stimulation, compared with adult cells, because of reduced expression of interleukin-4 (IL-4) receptor and higher levels of gene expression related to pro-apoptotic programs.

Neonatal mice exhibit skewed antibody gene-segment usage compared with adult mice; however, most evidence in human infants suggests that the antibody variable gene repertoire is very similar to that of adults, including microbial-specific B-cell repertoires. The antibody sequences of B cells of infants exhibit mature levels of junctional diversity, including nontemplated and palindromic types of additions at the V-D and D-J gene-segment junctions and the lengths of the antibody variable loops (complementarity-determining regions) are similar to those of adults. The distinguishing molecular difference between adult and infant antibodies is the striking lack of somatic mutations in infant antibody sequences. The use of germline sequences to encode antibodies to microorganisms early in life leads to the generation of low-affinity antibodies. Somatic mutations, which occur in the germinal center during antigen exposure, are the driving molecular force behind antibody affinity maturation and increases in antibody function. It is not clear currently whether the lack of mutations stems from the fact that neonatal B cells are encountering antigen for the first time (as opposed to the secondary responses made by older previously exposed individuals) or whether there are intrinsic B-cell defects in affinity maturation. After stimulation with the CD40 ligand and cytokines (mimicking helper T-cell interaction), human cord blood B cells do upregulate the transcription of genes involved in somatic hypermutation, including activation-induced cytidine deaminase and error-prone DNA polymerases.

There are also factors extrinsic to B cells that affect antibody responses early in life. T-cell responses, although perhaps more robust than B-cell responses early in life, exhibit some altered features compared with adult T cells. In particular, neonatal responses generally appear to be reduced in the magnitude of T-helper 1 (Th1)-type cytokines, with relatively preserved levels of Th2 cytokines, leading to an overall Th2-biased response. The model of Th1 versus Th2 biases in this age group is an oversimplification of very complex and highly regulated responses that are skewed early in life in various ways. Studies in recent years of Th-cell differentiation have revealed further complexity in Th-cell subsets, such as Th17, Th9, and Th22 cells. In the presence of IL-1 and IL-23 in humans, the naïve Th cell expresses retinoic acid–related orphan receptor (ROR)γt and differentiates into a Th17 cell that produces a host of cytokines, including IL-17. These cells appear to contribute protection against extracellular bacteria and fungi, but they also may have potential for autoimmune effects. In the presence of IL-4 and transforming growth factor-β, the Th2 cell can further differentiate into Th9 cells, which produce IL-9 and IL-10. In the presence of tumor necrosis factor-α and IL-6, naïve Th cells can differentiate into Th22 cells that express the aryl hydrocarbon receptor and secrete IL-22. The level of maturity of regulation of such subsets of Th cells in neonates is poorly understood. Recent work suggests that fetal T cells arise from different populations of hematopoietic cells than adult T cells and that the fetal T-cell lineage is biased toward immune tolerance. In addition, professional antigen-presenting cells, such as dendritic cells and macrophages, may exhibit developmental programs that affect the outcome of humoral responses.

The basis for the profile of the neonatal response likely stems not only from relative deficiencies, compared with adults, but also from active suppression or regulation of immune responses. For example, neonates exhibit high numbers of circulating CD71 ⁺ red blood cells, which express the enzyme arginase-2, whose activity mediates immunosuppressive properties. These cells appear to afford some cell-mediated protection against unwanted immune cell activation in the intestine during transition to postnatal life but also inhibit active responses to pathogens such as Listeria monocytogenes .

There are also extrinsic factors that affect the function of antibody proteins. For instance, complement protein levels are low in neonates, especially terminal elements of the complement cascade. Even if complement-fixing antimicrobial antibodies are induced, they may not be able to induce effective formation of membrane attack complexes in neonates when terminal complement components are in short supply. Complement also is necessary for optimal antigen presentation in many cases.

Infants exhibit a particular deficiency in responding to capsular polysaccharides, such as those of pathogenic bacteria, including Neisseria meningitidis , Haemophilus influenzae , Streptococcus pneumoniae , and others. A functional response to the small repeating units of these carbohydrates is usually not observed until the age of 2 years, although this deficiency has been overcome with conjugate vaccines, discussed later. For a more detailed discussion of fetal and neonatal immunity, see Chapter 4 .

Maternal Antibodies

Antibodies from the mother cross the placenta into the fetus, beginning at about 32 weeks, and increasing until term. The transfer is an active process, mediated by a receptor (FcRn) that specifically transports immunoglobulin G (IgG), especially IgG1, but not other immunoglobulin isotypes. In many cases, the IgG titer of antibodies to particular pathogens at birth exceeds those of the mother. If the mother has malaria or HIV infection, the transplacental transfer of specific IgG often is reduced, likely resulting from placental dysfunction. Transfer of IgG is beneficial because the infant effectively becomes passively immunized against all of the pathogens to which the mother had mounted an effective response. Acquisition of these antibodies affords protection against severe disease in many cases, but the antibodies are lost over time. The specific half-life of transferred IgG in older subjects is typically about 3 weeks, but some reports in neonates suggest that maternal antibodies may possess a longer half-life in infants. Passively acquired maternal antibodies may, however, interfere with the response of neonates to infection or immunization, a phenomenon termed antibody-mediated immune suppression.

Interference

In some cases, the suboptimal responses exhibited by neonates may be due to interference caused when multiple exposures, infections, or immunizations occur simultaneously. Combination vaccines have been carefully developed, with an eye toward adding new vaccine antigens in such a way as to maintain effective responses to existing vaccines. However, in some cases, investigators have observed that addition of new antigens can affect the quantitative response to other components of the vaccine. Forcing multiple exposures early in life could lead to interference. It should be remembered that vaccines are not the only exposures because neonates are also exposed to a myriad of naturally acquired infections, food and environmental antigens, and allergens. Vaccine antigens, in fact, represent a very small component of the antigen exposures early in life.

Logistics of Immunization Programs

Immunization programs have to be implemented in the context of an overall public health approach. In many areas of the world, children have the highest rate of access to medical interventions in the newborn period. Investigators interested in global health have often dreamed of a single efficacious vaccination given at birth for all major childhood infections because access to children is highest at birth. However, this goal is not really realistic, given the obstacles outlined earlier. Therefore each country has to develop an approach to vaccinating infants that achieves the highest feasible coverage based on local resources, infrastructure, financial commitments, cold chain, and other practical considerations. The efficacy of vaccines tested in clinical trials can vary widely in different settings and populations. Even if a vaccine has been shown in definitive clinical trials to be efficacious, the effectiveness of a vaccine in the field often is determined by practical considerations of cost-benefit, adverse-event profile, and the clinical relevance to the experience of the practitioner and of parents. Although pediatricians in the outpatient setting in the United States generally are strong advocates of proper vaccination, hospital physicians and staff who manage the peripartum period may be less acculturated to routine vaccination of neonates. Administration of multiple vaccines to all newborns before discharge after birth would require an infrastructure and culture that is still not currently present. Parents play a major role in decision making too, and appropriately so. Parents need assurance of the clinical benefit and the safety of any vaccines offered shortly after birth.

Maternal Immunization

Immunization of mothers during pregnancy is an attractive strategy for several reasons. Pregnant women typically are easy to identify, and in many areas of the world there is a high level of access to prenatal care. The principle of maternal immunization is to induce or boost the levels of antibodies against microorganisms in the mother’s serum, causing a quantitative or qualitative enhancement of the IgG isotype antibodies that cross the placenta and circulate in the blood of the fetus. Maternal immunization has been shown to be safe and effective for several diseases, especially tetanus and influenza, has been tested for RSV, and pertussis. Current recommendations for use of vaccines in women before, during, or after pregnancy are provided by the CDC ( http://www.cdc.gov/vaccines/pubs/downloads/f_preg_chart.pdf ) and are shown in Table 38-2 .

In 1979, the FDA introduced a classification of fetal risks resulting from pharmaceuticals given during pregnancy. Pregnancy category A is applied when adequate and well-controlled studies have failed to demonstrate a risk to the fetus in the first trimester of pregnancy and there is no evidence of risk in later trimesters. Pregnancy category B pertains when animal reproduction studies have failed to demonstrate a risk to the fetus and there are no adequate and well-controlled studies in pregnant women or animal studies have shown an adverse effect but adequate and well-controlled studies in pregnant women have failed to demonstrate a risk to the fetus in any trimester. Pregnancy category C is assigned when animal reproduction studies have shown an adverse effect on the fetus and there are no adequate and well-controlled studies in humans but potential benefits may warrant use of the drug in pregnant women despite potential risks. All vaccines that have been licensed by the FDA are categorized as pregnancy category C, except for the quadrivalent human papillomavirus vaccine, which is category B.

Maternal Immunization to Prevent Tetanus, Diphtheria, and Pertussis

In developed countries, tetanus and diphtheria are essentially controlled or eliminated. In the 1970s, there was a worldwide push to deliver tetanus-diphtheria (Td) vaccine to a broad segment of the population by targeting susceptible pregnant women. These campaigns were highly effective in markedly reducing or eliminating tetanus of the mother and infant in some areas. Still, in the 47 poorest countries in Africa and Asia, an estimated 128,250 babies and up to 30,000 mothers died of tetanus in 2004, according to the United Nations Children’s Fund. That agency set a goal to deliver 1 million doses of vaccine to mothers and infants in an effort to eliminate the disease in these groups by 2012. The World Health Organization (WHO) estimated that in 2010 (the latest year for which estimates are available), 58,000 newborns died from neonatal tetanus. As of November 2012, 31 countries still had not reached maternal and neonatal tetanus elimination status. There does not seem to be any problem with performance or safety of the vaccine in these groups; the shortfall stems simply from implementation.

Pertussis has been much more difficult to address. Major reductions in numbers of pertussis cases were accomplished by childhood pertussis vaccination through the middle of the 20th century; however, the disease has been rising in incidence for several decades. It is estimated that hundreds of thousands of cases occur in adults in the United States each year, which places infant contacts at risk. Many feel that the durability of solid vaccine-induced immunity may not extend beyond early adolescence, leaving a susceptible adult and older adolescent population.

Pregnant mothers can be infected with Bordetella pertussis and suffer symptomatic respiratory tract disease, spanning from mild to severe. Surprisingly, there is little evidence in the literature that pregnant women are more susceptible to severe disease than other healthy adults. Also, there is little evidence that infection of pregnant mothers is associated with adverse outcomes for the fetus, such as fetal demise or altered development. Therefore the focus of maternal immunization against pertussis is on preventing severe disease in young infants after birth.

Most of the deaths caused by pertussis occur in infants younger than 2 months at a time before routine immunization is initiated. The optimal strategy to protect these young infants is not entirely clear. Immunization of women postpartum could maximize immunity in the mother to prevent the mother from acquiring a new infection and transmitting it to newborns. However, data suggest that mothers are the source of pertussis in less than a quarter of cases. Therefore this strategy may have minimal effect on reducing risk in the first months of life. Immunization in the second or third trimester of pregnancy is of potential benefit to the mother by prevention of disease in her and induces higher levels of antibodies that can be transferred across the placenta. In the first half of the 20th century, pregnant mothers were commonly immunized with whole-cell pertussis vaccine, and it was clear that third trimester vaccination raised the level of antibodies in infants. Efficacy was not studied or proven in a rigorous way, however. There is the suggestion in clinical and epidemiology surveillance studies that maternal immunization reduces the incidence of disease in infants. Tetanus–reduced diphtheria–acellular pertussis vaccine (Tdap) was licensed for use in adolescents and adults in the United States in 2005. This vaccine is commonly given to pregnant women without evidence of harm to mothers or fetuses. Recent changes in policies of the ACIP recommend that adolescents and adults should receive a single dose of Tdap (instead of a single dose of Td), if their last dose of Td was greater than 2 years ago (instead of >10 years ago). The American College of Obstetricians and Gynecologists (ACOG) and the AAP recommend that women should receive Tdap before pregnancy if possible. In 2012, the ACIP voted to recommend use of Tdap during every pregnancy. Vaccinating during the third trimester (optimally at weeks 30-32 of pregnancy) provides the highest concentration of maternal antibodies to be transferred to the newborn. If not administered during pregnancy, Tdap should be given immediately postpartum. The issue of pertussis and pertussis immunization strategies for pregnant women and infants is discussed in detail in Chapter 21 .

Maternal Immunization to Prevent Influenza

Immunization of pregnant women against influenza is important for the health of both the mother and the infant. The risk of influenza during pregnancy continues to be underappreciated. Seasonal influenza poses a significant risk to the health of pregnant women during annual winter seasonal epidemics. Both pregnant women and their infants are at increased risk of morbidity from influenza, and both benefit from maternal immunization. During the 2009 H1N1 pandemic, rates of stillbirth and prematurity were greater in pregnant women with influenza infection. Both infants and pregnant women were at greater risk of hospitalization and severe medical complications from the 2009 pandemic H1N1 infection, including death. Both the ACIP and ACOG recommend that all pregnant women be immunized during the influenza season. Vaccination is recommended at any gestational age. The WHO also recommends that all pregnant women should receive the vaccine regardless of trimester. The indicated vaccine is conventional trivalent inactivated vaccine, given by the intramuscular route in the deltoid muscle. Although an intranasal live-attenuated trivalent vaccine is available, that vaccine is not recommended during pregnancy. Immunization during pregnancy has been shown to be safe for both the infant and the mother. Large numbers of pregnant women have been immunized with influenza vaccine; for example, during the 2009 to 2010 influenza season, according to the CDC, 51% of pregnant women in 10 states received the seasonal influenza vaccine.

Pregnant mothers appear to respond well to inactivated influenza virus vaccination in a similar manner to nonpregnant women, achieving elevated antiviral antibody titers in both maternal serum and umbilical cord serum. A careful prospective trial in 158 mother-infant pairs suggested that immunization of pregnant women could delay onset or reduce severity of disease in infants. A 2004 to 2005 randomized study of 340 mothers receiving either inactivated influenza vaccine or the 23-valent pneumococcal polysaccharide vaccine showed that the influenza vaccine reduced proven influenza illness by 63% in infants up to 6 months of age and prevented about a third of all febrile respiratory illnesses in mothers and young infants. This striking result reignited enthusiasm in the maternal immunization research community for testing this strategy more broadly for other respiratory virus infections, such as RSV. Influenza antibody titers in umbilical cord blood of immunized mothers do achieve protective levels; in fact, they can be higher than those of the mother. Higher levels of maternal antiinfluenza antibodies are associated with greater and longer protection of infants.

Maternal Immunization to Prevent Respiratory Syncytial Virus Infection

Respiratory syncytial virus causes hospitalization of infants for wheezing, pneumonia, or apnea, with a peak incidence at about 6 weeks of age. There appears to be some relative sparing of disease in the first weeks of life, possibly associated with maternal antibodies. It is difficult to contemplate inducing immunity in neonates before this age; therefore investigators have investigated maternal immunization against RSV to increase the titer of virus-neutralizing maternal antibodies that cross the placenta. The rationale is that, for every twofold rise in maternal antibodies that could be achieved, infants might be protected for an additional 3 weeks if a conventional IgG antibody half-life of 21 days is observed. A small experimental trial of an RSV subunit protein vaccine has been conducted in pregnant women. The vaccine, an immunoaffinity-purified protein isolated from infected cell culture–designated purified protein-2 (PFP-2), was safe but minimally immunogenic in a small trial. New experimental vaccines for RSV based on the surface fusion (F) protein, which is the protective antigen, are being developed for maternal immunization using various technologies. Structure-based designs for RSV F subunit vaccines, including a stabilized prefusion form of F and a computationally designed epitope vaccine, have been described and were enabled by the determination of the structure of prefusion and postfusion F protein at atomic resolution. A method for reproducibly isolating aggregates of postfusion F protein expressed in insect cells, termed F nanoparticles, has been developed. This vaccine candidate has been tested in human phase I trials, and maternal immunization studies are planned.

Measles-Mumps-Rubella (MMR) Vaccine During Pregnancy

Women are advised not to receive the MMR vaccine during pregnancy because all components are live viruses. Rubella virus is of particular concern because there is the theoretical possibility of this live-virus vaccine causing congenital rubella syndrome. A number of women have inadvertently received this vaccine while pregnant or soon before conception. The CDC collected data about the outcomes of their births. From 1971 to 1989, 324 infants were born to 321 women who received rubella vaccine while pregnant and continued pregnancy to term, and no cases of congenital rubella syndrome were identified. Given that the risk to the fetus appears to be negligible, a recommendation suggesting termination of pregnancy after inadvertent immunization is not warranted.

Maternal Vaccines in the Setting of Special Risk Factors

The ACIP recommends some other vaccines during pregnancy if particular risk factors are present— specifically, meningococcal and pneumococcal polysaccharide, and hepatitis A and B virus vaccines. If meningococcal conjugate vaccine (MCV4/MenACWY) is indicated, pregnancy should not preclude vaccination. There are two types of licensed pneumococcal vaccines: pneumococcal polysaccharide (PPSV23) and conjugate vaccine (PCV13). The safety of pneumococcal polysaccharide vaccine during the first trimester of pregnancy has not been evaluated formally; however, no adverse consequences have been reported among newborns whose mothers were inadvertently vaccinated during pregnancy. The use of PCV13 has been very limited among women of childbearing age, and the ACIP has not published pregnancy recommendations for PCV13. The ACIP recommends that all persons between the ages of 1 and 18 years should receive the hepatitis A virus (HAV) vaccine, as well as persons of any age who have risk-associated conditions. Pregnancy is not a contraindication to HAV vaccination. Studies reveal no apparent risk for adverse events in the developing fetus when the current vaccine is administered to pregnant women. The hepatitis A inactivated vaccine therefore is recommended if another high-risk condition or other indication is present. The ACIP recommends that the hepatitis B virus (HBV) vaccine should be administered to all patients before age 19 years, including unvaccinated pregnant women. Hepatitis B vaccine also is not contraindicated in pregnancy. This inactivated vaccine should be given to pregnant women who are at elevated risk for acquiring HBV infection during pregnancy because they (1) have had more than one sex partner during the previous 6 months, (2) have been evaluated or treated for a sexually transmitted disease, (3) have recent or current injection drug use, or (4) have had a hepatitis B surface antigen (HBsAg)-positive sex partner.

Maternal Vaccines for Group B Streptococci Under Development

Group B streptococci (GBS) are desirable targets for maternal immunization because these bacteria mainly cause disease in infants younger than 3 months, especially early-onset disease in newborns. GBS are a significant cause of infant disease and mortality worldwide. There is no licensed GBS vaccine to date, but vaccine candidates are under development, and several clinical trials suggest that a capsular polysaccharide conjugate vaccine is well tolerated and immunogenic. There are 10 GBS serotypes, but the most likely candidate for a vaccine would be a multivalent conjugate vaccine incorporating capsular polysaccharides from the five most common serotypes. A capsular polysaccharide conjugate vaccine was given to pregnant women in a double-blind, randomized, controlled trial that showed that the vaccine was immunogenic in mothers and caused transplacental transfer of IgG to infants. Additional vaccine studies are active and listed at http://Clinicaltrials.gov .

Respiratory Syncytial Virus Immune Globulin and Monoclonal Antibodies

Respiratory syncytial virus immune globulin was partially effective in preventing hospitalization caused by severe RSV infection ; however, the intravenous (IV) route and large volumes needed brought challenges for administration. Subsequently, a neutralizing, humanized mouse monoclonal antibody to the RSV fusion protein (palivizumab) was developed and licensed and allowed intramuscular administration. The efficacy of palivizumab was assessed in a randomized, double-blind, placebo-controlled trial (designated the Impact-RSV Study) in high-risk infants, with a 55% reduction in hospitalizations. This antibody is the only monoclonal antibody licensed to date to prevent virus infection. An affinity-matured, second-generation RSV monoclonal antibody (motavizumab) was tested in large clinical trials, but skin events were increased in motavizumab recipients, raising safety concerns, and the development of this antibody has been halted.