INTRODUCTION — Group B streptococcus (GBS or Streptococcus agalactiae) is an encapsulated gram-positive bacterium that colonizes the human gastrointestinal and genital tracts. It causes invasive infections in three populations:
●Neonates, particularly in the first few days of life, and infants up to three months of age (see "Group B streptococcal infection in neonates and young infants")
●Pregnant women (see "Group B streptococcal infection in pregnant individuals")
●Adults over the age of 65 years or those immunocompromised by underlying medical conditions (see "Group B streptococcal infections in nonpregnant adults")
Vaccination strategies and the composition of potential GBS vaccines are reviewed here. The microbiology and epidemiology GBS infections are discussed separately. (See "Group B Streptococcus: Virulence factors and pathogenic mechanisms".)
RATIONALE FOR MATERNAL VACCINATION — GBS is the most frequent cause of sepsis and meningitis in neonates and young infants. It also causes substantial perinatal morbidity and mortality (eg, stillbirths and pregnancy-related chorioamnionitis, bacteremia, postpartum endometritis). Maternal antibody against GBS capsular polysaccharides, and perhaps some surface proteins, appears to be protective against perinatal disease. However, many GBS-colonized pregnant women do not have sufficient serum antibody concentrations to these putative antigens at delivery, and infants born before 28 to 32 weeks of gestation may not benefit from optimal placental transport, even if available maternal GBS-specific immunoglobulin (Ig)G levels are present in moderate concentrations at birth.
Although intrapartum (IP) penicillin G or ampicillin four or more hours before delivery can prevent at least 80 percent of early-onset GBS infections, it has no impact on late-onset (age 7 to 89 days) invasive disease. Also, concern has been raised that if strains of GBS resistant to beta-lactams emerge, rates of early-onset disease in neonates would increase to the pre-IP antibiotic prophylaxis era. Maternal IP antibiotic prophylaxis also must be given parenterally with each pregnancy, is expensive when compared with maternal immunization, and has the attendant risk of allergic reactions. Finally, maternal screening and IP antibiotics are not feasible in most middle- and low-income countries worldwide. (See "Prevention of early-onset group B streptococcal disease in neonates".)
TARGETS OF PROTECTIVE IMMUNITY — Our current understanding of GBS antigens and their potential roles in inducing protective immunity stems from the studies of Rebecca Lancefield and colleagues [1-5]. GBS possesses two distinct saccharides, the group B carbohydrate (common to all strains) and the surface capsular polysaccharide (CPS). The group B carbohydrate is more closely associated with the cell wall of GBS than are CPS or protein antigens . The GBS non-CPS antigen is not a vaccine candidate as antibodies to this carbohydrate are not protective [7,8].
GBS capsular polysaccharide antigens — The CPS antigens confer serotype specificity for GBS. Ten distinct CPS antigens have been characterized (Ia, Ib, II-IX).
Composition — With only two exceptions, all GBS CPS are composed of 150 or more oligosaccharide repeating units of galactose, glucose, N-acetylglucosamine, and N-acetylneuraminic acid. Type VI CPS lacks N-acetylglucosamine, and serotype VIII is the simplest structure, containing only four sugars including rhamnose, a saccharide previously only found in the group B carbohydrate . Each CPS has a unique structure containing backbone saccharides and side chains comprised of mono-, di-, or trisaccharides. Despite containing similar sugars, each CPS is antigenically distinct, even types Ia and Ib in which the sole difference resides in a single linkage in the attachment of galactose to N-acetylglucosamine [10,11].
Sialic acid — The presence of N-acetylneuraminic (sialic) acid is preserved among all GBS CPS whose structures have been elucidated. This sugar is a virulence factor for GBS; GBS mutants with CPS that does not contain sialic acid lose their virulence in animal models of GBS infection .
GBS protein antigens — Nearly all GBS also express surface proteins [13-17]. The best characterized of them are the historically designated C proteins, present in most GBS strains that are not serotype III [17,18]. Lancefield and others initially showed that antibodies to the two components of C protein protected against lethal infection in a mouse model . The components of C proteins have been delineated and shown to be expressed independently in different strains [15,17,18,20].
Beta C protein — The 130-kDa beta C protein binds specifically to human IgA via the Fc portion of the immunoglobulin [17,18]. This antigen is present in nearly all strains of GBS type Ib, and some type Ia, II, and IV strains [14,15].
Observations in animal models suggest that a beta C protein vaccine might protect against GBS infection:
●Antibody to the cloned beta C protein protected mice from lethal infection with GBS that expressed the protein .
●Purified beta C protein actively immunized adult female mice and protected their neonatal pups against challenge with GBS .
●Recombinant beta C protein and a variant lacking the IgA binding domain were effective vaccines and carrier proteins for GBS CPS in mice .
Invasive disease, but not colonization, elicits IgM and IgG antibodies to beta C protein in humans .
Alpha C protein — The alpha C protein is present in approximately 70 percent of nontype III GBS strains, primarily in Ia, Ib and type II strains [14,15,17]. A monoclonal antibody directed at the alpha C protein identified a series of regularly spaced bands on Western blots of extracts from alpha-C-protein positive GBS [17,20]. Moreover, the monoclonal antibody protected mice challenged with an alpha-positive strain of GBS. The cloned alpha C protein from the prototype Ia/C strain A909 elicited antibody that protected mice from lethal challenge with alpha-positive GBS .
Deletions in the repeat region of the alpha C protein have been noted in both human isolates and an immune mouse model and appear to reflect a form of antigenic variation resulting in escape from host immunity . In contrast to antibodies raised to the wild-type, 9-repeat alpha C protein (antibodies raised to the 1- or 2-repeat species) bind with high affinity to both 1-repeat and 9-repeat alpha C protein . In addition, alpha C protein with one or two repeats is more immunogenic and more protective than is the 9-repeat protein. A 2-repeat alpha C protein conjugated to type III CPS has been reported to confer significant protection against GBS strains expressing either antigen . Thus, like the beta C protein, the alpha C protein is able to function both as a carrier protein for the CPS and as a protective immunogen.
Other proteins — Other GBS surface proteins share the properties of trypsin resistance and a ladder-like pattern on Western blot with the alpha C protein. The R proteins are present in many type II and III strains and may be involved in protective immunity in animals and in humans [16,27]. One group reported the presence of a novel protein termed "Rib" in most type III GBS strains . This protein seems to be analogous to the alpha C protein. It is a protective antigen that appears variable in molecular size in different strains and produces a ladder on Western blot. Nucleotide sequencing of the gene for Rib protein reveals extensive homology to the alpha C protein, and it contains a series of identical tandem repeats . Type V and type VIII clinical GBS isolates contain alpha-like proteins; nearly all GBS strains appear to express a surface protein that is a member of this family, regardless of serotype . Recombinant alpha C protein 3, found on most GBS types V and VIII isolates, has been shown to be a viable vaccine candidate and an effective carrier protein for CPS antigens . BibA, a surface-associated adhesin found predominantly on type III clinical isolates, confers active protection in mice against GBS challenge . Finally, surface pilus island (PI) proteins, such as PI 1 and PI 2a, are protective in a murine model and also may be protective against invasive perinatal or even adult disease GBS disease .
Genomic approaches to vaccine development — The availability of complete genome sequences for GBS has led to the discovery of novel vaccine candidates. Analysis of genomes from eight GBS strains representing five serotypes (Ia, Ib, II, III, and V) identified 589 potential surface-exposed and secreted proteins . Of these, 312 could be expressed in recombinant form and used for production of antibodies. Many of the antibodies recognized a single predominant protein on Western blot of GBS extracts or could detect a surface exposed protein of GBS by fluorescence-activated cell sorting.
A combination of four of these candidate proteins given to female mice provided protection to their offspring against GBS strains of the above five serotypes plus serotype VIII . These proteins, some of which are GBS pili antigens , may lead to the development of a combination vaccine capable of providing protection against several serotypes . These candidate proteins can be screened for protective activity against GBS in animal models and could lead to the development of new vaccines.
VACCINE FORMULATIONS — Numerous different GBS antigens have been considered for inclusion in potential vaccines (table 1) . Capsular polysaccharides (CPS) were the initial GBS vaccines tested in humans. Conjugation to protein antigens such as tetanus toxoid or CRM197 enhances the immunogenicity of these vaccines. Use of proteins from GBS in conjugate vaccines also is being studied.
Polysaccharide vaccines — In 1975, Lancefield summarized years of work demonstrating that antibodies directed at CPS antigens of GBS protect mice against lethal challenge . These results in animals were followed by the observation that susceptibility of newborns to GBS infections correlated strongly with a lack of maternal antibody to the GBS CPS antigen . Baker and Kasper first suggested that the CPS antigen might be a good vaccine candidate . A subsequent prospective, case-controlled study of mothers who delivered neonates with early-onset GBS disease and matched mothers colonized with the same CPS types but who delivered healthy neonates showed that naturally-occurring maternal delivery IgG antibody of ≥1 mcg/mL specific to CPS types Ia, III, and V was associated with a decrease in disease risk by 70 (V) to 90 percent (Ia and III) . A subsequent study in South African mothers and neonates confirmed these findings using a different serologic assay that resulted in better prediction of “protective” antibody serum concentrations (3 to 5 mcg/mL) against type Ia and III disease .
Clinical trials — The first phase I GBS vaccine clinical trial used two native type III CPS vaccines that differed in the method of CPS extraction (ethylenediaminetetraacetic acid [EDTA] compared to trichloroacetic acid [TCA] extraction) . Both vaccines passed safety tests in animals before administration to 40 healthy adults, 33 of whom had low levels of antibody to III CPS before vaccination. Although both vaccines elicited type III-specific antibodies, the EDTA-extracted preparation was superior in both magnitude of the antibody response and rate of response . This trial suggested that GBS CPS were safe and immunogenic in humans and paved the way for evaluation of other clinically relevant GBS serotypes.
“Native” EDTA-extracted type Ia CPS  and II CPS vaccines [41,42] also were shown to be safe in healthy adults and to elicit type-specific IgG in their sera higher than baseline values. However, the magnitude of the response and the percent of the volunteers who achieved a >1 mcg/mL increase in CPS-specific antibody varied considerably from 40 to 88 percent . Thus, the CPS vaccines were suboptimal for use in women of childbearing age with low levels of baseline antibody to GBS CPS, the target population for vaccination.
However, a pivotal study with the "first generation" GBS vaccines in pregnant women demonstrated that GBS type III unconjugated CPS vaccine was well tolerated and immunogenic when administered early in the third trimester to 40 pregnant women . The direct correlation (r = 0.913) between maternal delivery and infant cord sera type III CPS-specific IgG levels provided evidence that a strategy of maternal vaccination to prevent infant GBS disease would be feasible if the immunogenicity of the vaccine was improved.
Polysaccharide-protein conjugate vaccines — The observation that humoral responses to variably immunogenic polysaccharide antigens could be augmented by covalent attachment to proteins was first reported in 1931 . GBS type III oligosaccharide and CPS conjugate vaccines were developed and shown to be more immunogenic in rabbits and mice compared to uncoupled CPS [45-48]. These GBS CPS conjugate vaccines were generated by first creating aldehydes on a selected number of sialic acid sugars on the CPS, then using these aldehydes as coupling sites to proteins by a reaction known as reductive amination .
Animal models — An animal model was developed in which pregnant female mice were actively immunized with GBS conjugate vaccines to induce IgG responses. Their newborn pups were subsequently challenged with a lethal dose of the homologous GBS strain. As hypothesized, conjugated but not uncoupled GBS CPS was highly immunogenic in adult mice, and the type-specific antibody provided protection against lethal GBS challenge in pups [50,51].
A novel hexavalent CPS conjugate vaccine comprised of the six GBS types (Ia, Ib, II, III, IV, and V) responsible for a majority of invasive GBS disease in humans globally has been found to be immunogenic in mice and rhesus macaques, to induce opsonophagocytic killing, and to protect mouse pups from GBS lethal infection following immunization of their mothers .
Clinical trials — The first phase I clinical study with the "second generation" of GBS vaccines was performed in healthy women ages 18 to 40 years. Type III CPS coupled to tetanus toxoid (III-TT) was safe and substantially more immunogenic than was uncoupled type III CPS . Antibodies elicited to the III-TT conjugate vaccine were predominately (99 percent) IgG class, were functionally active in vitro, and recognized III CPS, indicating that the reductive deamination coupling method did not alter critical conformational epitopes on the type III CPS. Tetanus toxoid was chosen as the carrier protein because of its safety and efficacy as a vaccine when administered to pregnant women worldwide to prevent maternal and neonatal tetanus infections.
In a subsequent phase 2 randomized study in 650 women, GBS type III CPS-tetanus toxoid resulted in a fourfold increase of concentrations of GBS type III CPS-specific IgG and significantly delayed acquisition of vaginal and rectal type III GBS but not other CPS types . If confirmed in future trials, these findings suggest that a multivalent GBS conjugate vaccine could provide passive maternal antibody protection for infants as well as reducing direct neonatal exposure to GBS at maternal vaginal and rectal sites during delivery.
Additional GBS CPS-TT conjugates have been evaluated. Type Ia-TT and Ib-TT conjugate vaccines were well tolerated by healthy, childbearing age women, and the serotype-specific IgG response was significantly greater in recipients of the conjugate compared to the uncoupled CPS vaccine (figure 1) . In addition, Ia and Ib CPS-specific antibodies elicited by the conjugate vaccines were predominantly IgG (99 percent), functionally active in vitro  and in vivo . Similar findings were reported for the type II-TT conjugate and unconjugated type II CPS vaccines  and for the type V conjugate and unconjugated type V CPS vaccines, but substantial amounts of IgM antibodies also were elicited [58,59].
Nearly 90 percent of the cases of invasive GBS disease in the United States occur in adults, and the highest risk is in those older than 65 years of age. The safety and immunogenicity of a type V conjugate vaccine was assessed in 32 healthy elderly individuals . The vaccine elicited specific antibodies that were opsonically active, suggesting that a type V conjugate vaccine may benefit this age group.
A successful formulation for an effective GBS vaccine will need to include multiple CPS antigens to provide protection against invasive GBS disease in mothers, infants, and perhaps nonpregnant adults. Toward that goal, a small clinical trial was conducted in healthy adults to study the safety and immunogenicity of a GBS conjugate vaccine containing both types II and III CPS . The bivalent vaccine was well tolerated and postvaccination sera contained functionally active antibodies against both serotypes.
A trivalent vaccine consisting of GBS Ia CPS, Ib CPS, and III CPS, each coupled to CRM197, was evaluated as a maternal vaccine . The vaccine was found to be well tolerated and immunogenic. The ratio of mother-to-infant transfer of specific IgG antibody ranged from 0.68 to 0.81. The trivalent vaccine elicited lower geometric mean concentrations of antibodies to each GBS type in pregnant women with HIV than in those without HIV, but placental transfer ratios were similar across groups . This candidate vaccine, however, has been discontinued from further development given the large proportion of maternal and infant disease caused by serotypes II and V and to a less extent by CPS IV.
A phase 1/2 dose-escalation study in healthy, nonpregnant adults evaluated the safety and immunogenicity of a hexavalent vaccine comprised of GBS Ia, Ib, II, III, IV, and V CPS, each coupled to CRM197 and administered with or without aluminum phosphate . This vaccine was well tolerated and elicited robust CPS-specific IgG responses at all doses tested with or without alum. These favorable results led the same group to initiate a clinical trial with the same hexavalent vaccine in healthy pregnant and nonpregnant women .
Potential GBS protein vaccines — The inclusion of proteins in future GBS vaccines offers several potential advantages over vaccines containing irrelevant proteins such as tetanus toxoid or CRM197. Since any GBS vaccine will need to be multivalent, the inclusion of GBS surface proteins in the vaccine could simplify a vaccine by covering strains not among the CPS serotypes included. As an example, the inclusion of the beta C protein could eliminate the need for the Ib CPS in a multivalent vaccine. Likewise, the alpha C protein could provide protection against many Ia serotype strains. The inclusion of protein antigens might also enhance the protective efficacy of the vaccine against some strains. A growing number of conjugate vaccines are employing a limited number of carrier proteins, which could lead to increased reactogenicity of a vaccine and the potential for suppression of immune response to the CPS . Thus, future clinical trials may well involve vaccines containing one or more GBS surface proteins.
Other GBS surface proteins have the potential to be effective vaccine candidates. Streptococcal C5a peptidase is located on the surface of the organism and conserved among GBS serotypes . Rabbit antisera to this enzyme promoted killing of GBS by macrophages in vitro. C5a peptidase has also been shown to function as an effective carrier protein for GBS polysaccharide . Surface immunogenic protein (Sip) is a 53 kDa protein that is also highly conserved among GBS isolates [68,69]. Active vaccination of mice with purified Sip resulted in protection against lethal challenge with GBS serotypes Ia, Ib, II, III, V, and VI .
The recombinant form of another GBS surface protein, called LrrG, protected mice against lethal GBS challenge . This observation suggests that this highly conserved protein may be a viable vaccine candidate.
Discovery of GBS pili  and of the high prevalence of pilus islands in clinical GBS isolates  suggests a potential role for these surface structures as viable vaccine antigens. Compared with controls, sera from mothers whose babies developed early onset GBS disease had significantly lower levels of IgG specific to pilus 1 and pilus 2a, but not to pilus 2b .
Finally, antibodies against a combination of unique proteins have been reported to promote opsonophagocytosis in vitro and protection in an animal model of GBS sepsis. This candidate vaccine (GBS-NN, a fusion of two GBS protein domains, Rib and Alpha C) was given to 240 healthy women in a randomized, placebo-controlled, double-blind phase 1 trial . It was well tolerated, immunogenic, and elicited antibodies that were opsonophagocytic in vitro. This trial evaluated four different doses, some with or without alum adjuvant, given 28 days apart. A phase 2 trial of this vaccine was initiated in 240 pregnant women (with or without HIV infection) and their newborn infants.
Vaccination for adults at risk of GBS disease — Vaccination is likely to be effective in pregnant women based on numerous observational studies indicating that women with low serum concentrations of IgG directed at CPS are those whose infants develop invasive infection [37,74]. While low concentrations of maternal IgG specific for GBS CPS is clearly a risk factor for infant GBS disease, the immune defects leading to adult GBS infection are poorly characterized, creating challenges to defining the role of GBS vaccines in nonpregnant adults.
In a surveillance study of adults with GBS bacteremia, acute serum samples collected within 48 hours of the first positive blood culture from 7 of 12 patients contained levels of type-specific antibody thought to be protective in neonates . It is difficult to determine whether the levels of antibody were high at the time bacteremia developed or whether they rose rapidly in response to infection. Since these patients may have developed bacteremia despite apparent protective levels of antibody, other factors could account for increased susceptibility to GBS infection in adults with malignancy, diabetes, liver disease, or age 65 years or older without underlying illness.
An alternative explanation is that higher levels of antibodies to CPS might be needed for protection in adults. In a cross-sectional study of adults ≥65 years of age, the serum level and functional activity of type V GBS CPS-specific IgG was measured . Thirty-six of 40 adults studied (90 percent) lacked GBS antibody in amounts sufficient to be effective against GBS . There was no apparent impairment of neutrophil function, which is another possible risk factor for susceptibility to GBS.
One study explored the immune response following GBS bacteremia in 100 adults . Substantial immune responses to both the CPS and pilus islands were observed in convalescent sera, but many individuals had relatively high concentrations of CPS-specific IgG in their acute sera.
●Group B streptococcus (GBS or Streptococcus agalactiae) is the most frequent isolate from neonates with bacterial sepsis, pneumonia, or meningitis in the first few days of life, and also causes invasive infection in infants 7 through 89 days of age. GBS also is a major cause of stillbirth and maternal morbidity and mortality worldwide. Many pregnant women who are colonized with GBS do not have sufficient serum levels of capsular polysaccharide (CPS)-specific IgG at delivery to provide passive protection from infection for the neonate and infant less than age 90 days. The current prevention approach of antenatal culture-based screening and maternal intrapartum antibiotic prophylaxis for those women who are colonized with GBS is effective but has several downsides, including the requirement for intravenous administration, selection for beta-lactam resistant organisms, lack of effectiveness for late-onset disease (infants age ≥7 days), potential allergic reactions, and cost. Thus, development of an effective vaccine is an attractive strategy for the prevention of maternal and young infant GBS disease. (See 'Rationale for maternal vaccination' above.)
●There is no commercially available GBS vaccine. A hexavalent Ia, Ib, II, III, IV, and V glycoconjugate vaccine has been developed, has completed a phase 1 clinical trial in healthy adults where it was shown to be immunogenic and well tolerated, and is undergoing a phase 2 trial. Further, a recombinant protein vaccine has been shown to be immunogenic and well tolerated in healthy nonpregnant women and is being further evaluated in a phase 2 trial. (See 'Vaccine formulations' above.)