Neisseria meningitidis B vaccines

Abstract

Invasive infections caused by Neisseria meningitidis are a serious public health problem worldwide and have a heavy economic impact. The incidence of invasive disease due to Neisseria meningitidis is highly variable according to geographical area and serogroup distribution. Since the introduction of vaccination programs with conjugated vaccine C in children and adolescents, most cases of invasive meningococcal disease in developed countries have been caused by meningococcus B. It is important to underline that invasive meningococcal disease will not be controlled until safe and effective vaccines for meningococcal B are available and widely used. The aims of this article are to describe the most recent developments in meningococcal B vaccines and to discuss how these vaccines can contribute to containing meningococcal disease.

Keywords::

• meningococcal B vaccines
• meningococcus B
• Neisseria meningitides
• prevention of meningococcal disease
• vaccination

Neisseria meningitidis is a Gram-negative, oxidase-positive, aerobic diplococcus. Meningococcus is acquired through contact with respiratory droplets. It infects only humans and the usual ecologic niche of the bacterium is the mucosa of the human oropharynx and nasopharynx. Encapsulated strains cause the great majority of cases of invasive disease. The meningococcal polysaccharide capsule is the main virulence factor, allowing evasion of opsonization, and phagocytic and complement-mediated killing [1,2]. Unencapsulated strains are frequently found in the upper respiratory tract of asymptomatic carriers and have rarely been the cause of invasive disease. The biochemical composition of the polysaccharide capsule determines the serogroups of meningococcal strains. There are usually 13 serogroups described [2–7]; however, the WHO reports that there are 12 serogroups [201]. Of the 13/12 different polysaccharide capsular types, only six (A, B, C, W135, Y and X) frequently cause disease globally. Serogroup X determines substantially invasive disease in sub-Saharan Africa, rarely in other parts of the world [8].

Invasive infections caused by N. meningitidis are a serious public health problem worldwide and have a heavy economic impact, not only in epidemic areas but also in areas where sporadic forms occur. Indeed, it is estimated that approximately 500,000 cases and 50,000 deaths occur worldwide every year [9].

Meningococcal disease usually develops within 1–14 days following acquisition [10]. Initial symptoms may sometimes be nonspecific and affect the upper respiratory tract. Subsequently, in the case of meningitis, the symptoms may be severe nausea, vomiting, stiff neck, intense headache and photophobia; in the event of bloodstream infection, the patient presents maculopapular, purpuric or petechial rash. Delirium and coma often ensue. It is generally accepted that septicaemia precedes meningitis [11]. In spite of timely and efficacious antibiotic therapy, the case–fatality rate ranges from 5 to 10% [12]. In addition, from 12 to 19% of survivors suffer permanent disability (neurologic sequelae such as deafness and paralysis, and amputations) [1].

The incidence of invasive disease by N. meningitidis is highly variable according to geographical area and serogroup distribution. The reasons for the different serogroup distribution throughout the world are unknown, but possible mechanisms include differences in population immunity and environmental factors [2].

Other factors also influence meningococcal epidemiology, such as age, low socioeconomic status, immunodeficiency status and behavioral risk factors (e.g., active and passive smoking, kissing, and bar and pub patronage) [2,8].

The highest incidence of meningococcal disease occurs in the ‘meningitidis belt’ of sub-Saharan Africa, where serogroup A is responsible for the largest and most devastating meningococcal epidemics. In Africa, serogroup C and serogroup X diseases occur, and serogroup W-135 disease has recently been reported. Serogroup A disease is also registered in China and Russia, but is rare in other parts of the world [13]. Serogroup B causes a substantial proportion of the meningococcal disease endemic in many areas of the world, including the USA and Europe, and it may cause prolonged epidemics. The annual mean incidence of meningococcal disease in the USA was 0.53 cases per 100,000 population from 1998 to 2007 [13].

The European Surveillance System has revealed considerable variability from one country to another in the incidence of meningococcal disease. The serogroups mostly associated with invasive cases are B and C, but serogroups W-135 and Y are also present, while serogroup A is only responsible for sporadic cases [202]. In recent years, the incidence of invasive disease caused by serogroup C has declined owing to the introduction of vaccination programs with conjugated vaccine C (MenC) in children and adolescents in some countries [1,8,13].

The incidence of meningococcal disease is highest in infants under 1 year of age followed by adolescents.

Meningococcal C conjugate vaccines, tetravalent conjugate vaccines (serogroups A, C, Y and W-135) and meningococcal A conjugate vaccine are currently on the market. These vaccines have determined a reduction in the incidence of invasive meningococcal disease in the countries where large-scale vaccination programs have been implemented. Since the introduction of the conjugate vaccines, meningococcus B has become the serogroup most frequently involved in invasive disease in European countries, North America and South America, where nearly one-half of all cases of meningococcal disease are caused by capsular group B strains; in many European countries the proportion is even higher (90%) [14,15].

It is important to underline that the control of invasive meningococcal disease will not be achieved until safe and effective vaccines for meningococcal B (MenB) are available and widely used.

MenB vaccine development has followed a different approach from that used for the preparation of conjugate vaccines for serogroups A, C, Y and W-135. Meningococcal serogroup B differs from the other pathogenic serogroups in that it has a capsular polysaccharide identical to the polysialic acid (α[2–8]N-acetylneuraminic acid) present in many human glycoproteins [16]; in particular, it is similar to carbohydrates found in fetal brain tissue. This mimicry does not allow the use of polysaccharide protein conjugate vaccine.

The aims of this article are to describe the most recent developments in meningococcal B vaccines and to discuss how these vaccines can contribute to containing meningococcal disease.

Pathogenesis & mechanisms of avoidance of host defences by N. meningitidis

The capacity of N. meningitidis to colonize humans efficiently and to cause high levels of bacteraemia depends on its ability to evade the immune system. Indeed, the wide variation of surface antigens among meningococcal strains has proved to be one of the main obstacles to the development of new vaccines.

Regarding the evasion mechanisms of immunodefense, N. meningitidis hinders, in addition to other process of the immunity, the dendritic cell actions. Dendritic cells are crucial immune cells that populate peripheral tissue, mucosal surfaces and the circulation [17]. N. meningitidis is exposed to dendritic cells during colonization of the nasopharynx. It has been shown that encapsulated, wild-type Neisseria display less adherence to dendritic cells than unencapsulated strains [18]. This finding is consistent with the fact that the ratio between IL-10 (which plays a crucial role in Th1 cell differentiation) and IL-12 (which plays an important role in the adaptive immune response) has been seen to be altered in dendritic cells infected with live encapsulated bacterium [19].

Furthermore, when N. meningitidis passes through the mucosa and into the bloodstream, it is exposed to the full force of the complement system, which is a major component of the immune system. There are three pathways of activation of the complement cascade: classical (initiated by the antigen–antibody complex), the lecitin pathway (initiated by microbial carbohydrates) and the alternative pathway. All lead to cleavage of C3 to C3b, which is involved in the effector functions of the complement (e.g., opsonization resulting in phagocytosis). N. meningitidis is relatively resistant to killing by complement, thanks to its molecular mimicry of human structures (particularly relevant for meningococcus B), which enables the bacterium to avoid antibody recognition, thereby blocking the classical pathway [20,21]. Capsule is the main factor that functions to prevent complement-mediated lysis and phagocytosis. Furthermore, lipopolysaccharide is a universal component of the outer membrane of Gram-negative bacteria, and is required for the resistance of N. meningitidis to complement [22]. In addition, N. meningitidis produces a surface protein that binds factor H, which is important in activating the alternative complement pathway (fHbp). Recently, it has been demonstrated that another surface protein, called Neisserial surface protein A (NspA), is able to bind human factor H (fH) [23]. Other factors which mimic or bind host molecules also function to prevent complement-mediated lysis and phagocytosis are known. For instance, within the bloodstream, N. meningitidis sheds outer membrane blebs containing proteins and lipopolysaccharide (LPS). The outer membrane vesicles are able to initiate complement activation in human whole-blood assay, and might thereby redirect complement activation away from meningococci in the circulation, hindering the bactericidal effects of complement [24].

The findings of research on the ability of N. meningitidis to evade the mechanisms of the human immune system are crucial to the design of efficacious vaccines. Indeed, some new vaccines already take into account some of this recent knowledge.

Bactericidal activity & serological correlate of protection against meningococcal infections

The bactericidal assay measures the interaction of antibody and complement at the bacterial surface, which results in bacterial lysis. The pioneering Goldschnedier study, which demonstrated a correlation between serum bactericidal activity and defense against developing meningococcal disease, used human complement in the assay [25]. However, it could be difficult to find a sufficient quantity of human sera without antimeningococcal antibodies that could be used as a source of exogenous complement. Therefore, many laboratories utilize infant rabbit serum as an exogenous source of complement, as suitable infant rabbit sera are widely accessible [26]. As already mentioned, fH is a protein of the alternative complement pathway. Since it has been shown that encapsulated strains of N. meningitidis bind fH, their resistance to complement-mediated bacteriolysis is enhanced.

The binding of fH to meningococci is specific to human fH [27]. Thus, this fact may explain why serum bactericidal titers measured with human complement are lower than those measured with rabbit complement [28–30].

When assessed by means of human complement a bactericidal titer of 1:4 or greater is considered protective. However, the assay lacks sensitivity and the results probably underestimate the level of protection. With regard to rabbit complement, the ‘protective’ threshold is controversial [31]. Titers as high as 1:128 are needed to ensure that a titer of 1:4 is present if measured with human complement [32,33].

It is difficult to assess the ability of recombinant multicomponent serogroup B vaccines to provide protection against meningococcal infections as it is not always possible to carry out serum bactericidal antibody (SBA) testing against a sufficient number of strains to achieve adequate statistical power. To address this problem, Donnely et al. assessed meningococcal antigens qualitatively and quantitatively in order to evaluate the potential strain coverage of protein-based vaccines [34]. They found that SBA outcomes for serogroup B vaccines could be predicted on the basis of relative potency determined by meningococcal antigen typing system ELISA. SBA is generally acknowledged to be a surrogate marker of resistance to meningococcal disease; however, as it measures only the total of functional antibodies in a sample, it is not, on its own, indicative of antibody responses to specific antigens or subcomponents of the vaccine [35].

Outer membrane vesicle vaccines

Development of an effective MenB capsular polysaccharide vaccine is hindered by the poor immunogenicity of the polysaccharide [36] and concerns over the possible induction of autoimmune antibodies [37].

Therefore, the development of MenB vaccines has focused on subcapsular antigens either as outer membrane vesicles (OMVs) or as individual antigens [38].

Outer membrane vesicles can be separated from meningococcal bacteria [39] or isolated as membrane blebs, which are released into media during bacterial growth. The OMVs are treated with detergents to extract lipooligosaccharide and reduce endotoxin activity [40]. On using sodium dodecylsulphate-polyacrylamide gel electrophoresis, the detergent-treated vesicles have been shown to contain four or five major outer membrane proteins (OMPs). On using more sensitive proteomic methods, the vesicles have been shown to contain many other periplasmic and cytoplasmic proteins [41,42]. The role of these proteins in safety or immunogenicity is unknown.

Several candidate OMV vaccines have been developed and tested in large-scale efficacy studies in Norway, Cuba, Brazil, Chile and New Zealand (Table 1). The Norwegian experience [43], conducted in 1988 on 171,800 students in secondary schools who volunteered to take part in a double-blind, placebo-controlled, efficacy trial with the school as the randomization unit, demonstrated a rate of protection of 57.2% (p = 0.012; one-sided test). These findings suggested that, although the vaccine conferred protection against group B meningococcal disease, the effect was insufficient to justify a public vaccination program.

Good results were reported by Sierra et al. in the Cuban study of OMPs from B meningococci [44]. These authors carried out a randomized, double-blind, controlled trial of the vaccine. Efficacy was tested in 1987–1989 on 106,000 10–14-year-old students from 197 boarding schools in seven provinces. The efficacy obtained was 83% (χ²: p < 0.002; Fisher exact test: p < 0.001). Efficacy and safety were confirmed in another study comparing vaccinated and nonvaccinated populations, in which a total of 133,600 subjects aged 5 months to 24 years were followed-up for 2.5 years in Ciego de Avila province (30 cases per 105 inhabitants, the highest incidence in Cuba) [44]. The results of the trial and the fact that no severe or long-lasting reactions to the vaccine were observed prompted the Cuban Ministry of Public Health to vaccinate all children aged between 3 months and 6 years in the most severely affected provinces. The efficacy of vaccination varied from 83 to 94% among the provinces and the age groups, which ranged from 3 months to 20 years of age. After 3 years of massive application, no severe reactions occurred, and one of the most severe epidemics was practically eradicated.

Unfortunately, these results were only partially confirmed in other epidemics occurring in other nations. Thus, in an attempt to control epidemic serogroup B meningococcal disease in Sao Paulo, Brazil, in 1989 and 1990, a Cuban-produced OMP-based serogroup B meningococcal vaccine was given to approximately 2.4 million children aged from 3 months to 6 years [45]. A case–control study was performed to estimate the efficacy of the vaccine in Sao Paulo. Microbiologically confirmed cases of serogroup B meningococcal disease were identified through hospital-based surveillance. Between June 1990 and June 1991, 112 patients and 409 matched controls with confirmed vaccine status were enrolled. The estimated vaccine efficacy varied by age: 48 months or older = 74% (95% CI: 16–92%), 24–47 months = 47% (95% CI: -72–84%), and less than 24 months = -37% (95% CI: < -100–73%). Since the vaccine used by De Moraes was strain specific, the effectiveness evaluation could be misleading. Indeed, the correlation between the vaccine strain and the isolates (in particular the correlation of PorA) is important [46]. However, De Moraes et al. reported that even though only 44% of serogroup B meningococcal isolates corresponded with the vaccine type strain (B:4:P1:15), many isolates were of the same serotype or subtype antigens as the vaccine type strain. Thus, the vaccine was able to protect against some serogroup B meningococcal strains other than the vaccine type strain [45].

In 1995, Boslego et al. published the results of a vaccination program promoted by the Chilean National Committee for Meningococcal Disease [47]. A meningococcal group B (15:P1.3) OMP vaccine was tested for efficacy in a randomized, double-blind, controlled study in Iquique, Chile. This vaccine was developed by the Walter Reed Army Institute of Research. A total of 40,811 volunteers, aged 1–21 years, were enrolled in the study. Volunteers received two doses of vaccine 6 weeks apart by jet injector. Both the experimental vaccine and the control vaccine (Menomune, A, C, Y and W135 meningococcal polysaccharide vaccine) were well tolerated and caused only minor side-effects. Active surveillance for suspected cases of meningococcal disease was conducted for 20 months in Iquique. A total of 18 cases of group B meningococcal disease were confirmed during the 20 months. Efficacy was estimated to be 51% (p = 0.11) for all ages combined. In children aged 1–4 years, no protection was evident, but in volunteers aged 5–21 years vaccine efficacy was 70% (p = 0.045). The bactericidal antibody response was characterized by the lack of a significant booster response and higher responses in the older children. The Walter Reed Army Institute of Research and the NIPH/Chiron developed another OMV meningococcal vaccine (MENGOC-BC). Tappero et al. carried out a clinical trial with this vaccine in a particular setting, namely during an epidemic which broke out in 1993 in Santiago, Chile. The vaccine contained a strain that was heterologous with the epidemic meningococcus strain. After the administration of three doses, a response to the vaccine was recorded in 31–35% of children (vs 5% to placebo) and 37–60% of adults (vs 4% to placebo; overall, p < 0.05 vs controls). However, no response was seen in infants. By contrast, against homologous vaccine type strains, the response rate was 67% or higher among children and adults and 90% or higher among infants [48]. Therefore, the efficacy of these vaccines varied from 57 to 83%; however, children under the age of 2 years on a two-dose schedule proved to be unprotected.

Although PorA OMP-specific antibodies have been seen to correlate with SBAs [43,49,50], immunologic responses to OMV vaccines have proven difficult to assess, owing to the variability of responses observed in recipients. Since these vaccines are based on a single meningococcal isolate, they can provide only partial protection against virulent heterologous meningococci. However, protection depends on age [51–54].

These findings may be particularly relevant in countries where MenB disease is of a multiclonal nature, such as The Netherlands and the UK. However, OMV vaccination has successfully controlled clonal MenB epidemics in Cuba [44,55] and Brazil [45].

Furthermore, it is clear that the OMV vaccines can be useful in curtailing localized epidemics through the administration of a specific ‘tailor-made’ vaccine, as the New Zealand experience subsequently demonstrated [56]. Since 1991, an epidemic of serogroup B meningococcal infection has afflicted New Zealand [57], causing more than 4700 cases and over 200 deaths [58]. The overall incidence peaked in 2001 at 17.4 cases per 100,000 people. In 2002, Maori and Pacific Island children <1 year of age displayed incidence rates of 286 and 368 per 100,000, respectively [58]. On the advice of the WHO, an international advisory group was set up in 1993 and several options were considered; in 2000, the New Zealand health authorities accepted a proposal by the Norwegian Institute of Public Health (NIPH) and Chiron Vaccines (Siena, Italy) to develop a vaccine and to implement clinical trials [59]. This approach involved preparing a protein-based, OMV vaccine from a wild-type strain typical of the one responsible for the epidemic [60].

A new group B strain-specific meningococcal vaccine, referred to as ‘MeNZB’, was developed by Chiron Vaccines in collaboration with the NIPH. Its effectiveness was assessed in a prospective observational study following a nationwide vaccination program in New Zealand. The vaccination program began in July 2004, and the study used data from January 2001 to June 2006. To estimate vaccine effectiveness, a generalized estimating equation model was used that included potential confounding variables, such as disease progression over time, age, ethnicity, socioeconomic status, seasonality and geographic region [61]. The model provided strong statistical evidence for a vaccine effect (p < 0.0001), with estimated disease rates 3.7 times higher in the unvaccinated group than in the vaccinated group (95% CI: 2.1–6.8) and a vaccine effectiveness of 73% (95% CI: 52–85%).

In 2009, Galloway et al. published the results of an observational cohort study on the effectiveness of the New Zealand vaccine in subjects less than 5 years of age [62]. Compared with unvaccinated children, fully vaccinated children were five to six times less likely to contract epidemic strain meningococcal disease in the 24 months after they became eligible to receive a full vaccination series, corresponding to an estimated vaccine effectiveness of 80.0% (95% CI: 52.5–91.6) for children aged 6 months to <5 years and 84.8% (95% CI: 59.4–94.3%) for children aged 6 months to <3 years. The authors concluded that, with over 3 million doses administered to individuals aged under 20 years throughout New Zealand, combined evidence from the Phase I and II clinical trials, the descriptive epidemiology of meningococcal disease and their results supported the effectiveness of the vaccine.

An OMV vaccine has also been prepared from a strain of Neisseria lactamica that shares a number of antigens with N. meningitidis but lacks an antigenically related PorA. The hypothesis is that, when PorA is present in an OMV, the molecule is immunodominant. Immunization with an OMV vaccine that lacks PorA may shift the antibody responses to other antigens that are poorly immunogenic in the presence of PorA but capable of eliciting protective antibodies in its absence. Mice immunized with an N. lactamica OMV vaccine did not develop serum bactericidal antibodies, but appeared to be protected from lethal challenge by group B N. meningitidis. A clinical lot of this vaccine has been prepared for testing in humans [63–66].

A vaccine based on native outer membrane vesicles (NOMVs) has recently been developed with the aim of providing safe, broad protection against group B strains of N. meningitidis. Three antigenically different group B strains of N. meningitidis were genetically modified in order to improve safety and enhance expression of the desired antigens. Safety was improved by disabling three genes: synX, lpxL1 and lgtA. The vaccine strains were genetically configured to have three sets of antigens, each of which would induce antibodies against a variety of group B strains. Preliminary trials using combined NOMV from the three strains demonstrated that the vaccine was able to elicit a broad bactericidal antibody response [67,68].

One limitation of conventional detergent-treated OMV vaccines is that SBA responses in children are largely directed against surface-accessible loops on a porin protein, PorA [53], which is antigenically variable [69]. The use of OMV vaccines is therefore best suited to controlling epidemics caused by a predominant strain [70,71]. To broaden protection, OMV vaccines have been prepared from >one strain [72,73], or from mutant strains engineered to express >one PorA molecule [74–79]. Indeed, a hexavalent PorA-based OMV was developed by the Netherlands Vaccine Institute. The Netherlands Vaccine Institute used OMVs prepared from two N. meningitidis strains that were each engineered to express three different PorA proteins each [74,75,80]. The vaccine has been evaluated in adults, children and infants [78,79,81]. In infants, the bactericidal antibody response after a primary three-dose series was only modest. Moreover, the immunogenicity of certain serotype antigens was poor. After a fourth dose in toddlers, higher serum bactericidal antibody responses were observed, which suggest that this vaccine might be efficacious after a four-dose regimen. In order to provide even broader protection, a nonavalent PorA OMV vaccine (NonaMen) was subsequently developed by adding a third trivalent OMV to cover the nine most frequently occurring subtypes in the industrialized countries. Preclinical studies in mice immunized with NonaMen showed high SBA responses against P1.7,16; P1.5-1,2-2; and P1.5-2,10, intermediate responses against P1.7-1,1; P1.22.14; P1.18-1,2,6 and P1.7-2,4 and low responses against P1.19,15-1 and P1.12-1,13. NonaMen also proved immunogenic in rabbits [82].

In the USA, however, meningococcal disease is caused by strains with considerable PorA antigenic diversity (120 PorA variable region types) [69,83,84]. Therefore, OMV vaccines that predominantly target PorA are unlikely to confer broad protection in infants and young children. These vaccines may be more useful in older subjects, as SBA responses in OMV-vaccinated adults are broader than those in vaccinated infants [48]. One possible reason for this finding is that most adults are naturally primed by exposure to neisserial organisms, and even small quantities of residual non-PorA antigens in the detergent-extracted OMV vaccines may be sufficient to boost SBA memory responses with broad activity. The quantity of these antigens, however, may be insufficient for immunogenicity in unprimed infants [85]. The potential problem of vaccines based on OMVs is the ability of meningococcal OMPs to undergo antigenic shift or gene deletion, as seen with PorA [86], thereby rendering the vaccine ineffective.

Furthermore, another limitation of OMV is the rapidly waning elicited immunity. Indeed, on studying a vaccine administered to students (aged 13–14 years) in Norway in 1988–1991, Holst et al. found a substantial reduction in immunity over time, with an efficacy of 87% in the first 19 months after vaccination declining to only 30% after 21–29 months [60].

Thus, the need to find highly conserved antigens for a universal Neisseria meningitidis B vaccine has led to alternative strategies (e.g., reverse vaccinology). These alternative strategies are illustrated in the next section.

Table 1 shows a summary of studies of outer-membrane vesicle vaccines against N. meningitidis.

Reverse vaccinology & universal vaccine for serogroup B meningococcus (rMenB)

To develop a universal vaccine against MenB, the researchers of Novartis Vaccines have used the results obtained by Tettelin et al., who sequenced the MenB bacterium genome (N. meningitidis serogroup B strain MC58) [87]. Thus, MenB became the prototype for the use of genomics for vaccine development, a process called ‘reverse vaccinology’. This approach predicted 600 novel antigens. Candidate sequence was expressed in Escherichia coli and used to immunize mice. Analysis of sera revealed more than 90 previously unknown surface-located proteins, 29 of which were able to induce bactericidal antibodies [88,89]. In the following years, five antigens were selected and included in the new vaccine, named 5CVMB (five component vaccine against MenB) [90]. Four of the five antigens were expressed as fusion proteins: GNA2132 or neisserial heparin binding antigen with GNA1030, GNA2091 with GNA1870 (factor H-binding protein), while the fifth antigen is NadA (an antigen which promotes adhesion to and invasion of epithelial cells) [91] or GNA1994 [90]. This new vaccine uses aluminium hydroxide as an adjuvant. As the functional activity of N. menigitidis is still under investigation, the role of GNA2132 is not yet completely known; however, this antigen induces protective immunity in humans. The protein binds heparin in vitro through an arginine-rich region and increases survival of the unencapsulated bacterium in human serum [92].

To evaluate the bactericidal activity of the 5CVMB vaccine, the sera obtained from immunized mice were tested in a bactericidal assay against a panel of 85 strains of meningococcal B isolated from clinical cases. The vaccine induced bactericidal titers for 66 strains (77.7%). The killing of 77.7% of the strains by a meningococcal B vaccine is very encouraging [90] and suggests that this vaccine has the potential to provide broad coverage against serogroup B disease.

After successful preclinical studies, clinical trials were started (Table 2). In 2008, Novartis Vaccines initiated clinical trials of 5CVMB (rMenB), both with and without OMV of the strain NZ98/254, in adolescents and infants. In a Phase II trial, 150 infants were vaccinated with three doses at 2, 4 and 6 months of age. Safety and immunogenicity were good, with SBA titers of 1:4 or higher against reference strains in 89, 96 and 85% of subjects after the third dose; these values increased to 100, 98 and 93%, respectively, after a booster dose administered at 12 months [93]. This finding is indicative of an immunological memory response. The vaccine displayed good safety and tolerability. The addition of OMV of the strain NZ98/254 increases strain coverage because the PorA antigen is present within the outer-membrane vesicle. After three doses of rMenB + OMV, 90% or more of participants had an human serum bactericidal antibody (hSBA) titer >four for five MenB strains, with 70% of participants displaying an hSBA titer >four for six strains. rMenB alone was immunogenic for only three strains. Both vaccines were well tolerated [94].

The Phase III study indicated that this multicomponent vaccine (4CMenB) might be the first to provide broad coverage against dynamic and deadly meningococcal B disease. These data were presented at the International Pathogenic Neisseria Conference in Banff, Canada [95]. The trial involved more than 3600 infants, and the vaccine was administered at 2, 4 and 6 months of age; the vast majority of the subjects vaccinated displayed a vigorous immune response against all vaccine meningococcal B antigens [95]. Indeed, 1 month after the third dose, tests using human complement (hSBA) ≥1:5 revealed serum bactericidal antibodies against the MenB strains 5/99, NZ98/254 and H44/76 in 100, 84 and 100% of subjects, respectively. All three lots of 4CMenB elicited highly consistent immune responses [95].

Additionally, although co-administration of 4CMenB with other routine infant vaccines has determined some systemic reaction in infants, an acceptable tolerability profile has been found [96], providing support for the use of this vaccine in the first 12 months of life, when it is most needed. The incidence of typical vaccine-associated events elicited for 7 days after each administration has proved to be similar (83% after routine vaccine alone vs 87% after co-administration with 4CmenB). In the aforementioned study, <1% of infants dropped out on account of reactogenicity, with no difference between groups. The incidence of serious adverse events was also similar in the study groups [96].

These Phase III data emerge from a comprehensive clinical program and show that 4CMenB can be used in every age class and can be co-administered with other routine vaccines or as part of a flexible vaccination schedule.

The immunogenicity and safety of 4CMenB have also been evaluated in laboratory workers (18–50 years of age) who received three doses of 4CMenB at 0, 2 and 6 months, followed by a single dose of Menveo^® 1 month later. Sequential administration of 4CMenB and Menveo^® provided robust evidence of an immune response against all serogroups. Both vaccines were well tolerated [97].

One limitation of the Novartis vaccine could be the variability of fHbp. Indeed, one analysis on a representative sample of meningococcal isolates has demonstrated that the fHbp gene and the encoded protein vary in two or three major groups of meningococcal isolates, each displaying several alleles that have some association with meningococcal clonal complexes and serogroups [98]. Furthermore, Murphy et al. performed nucleotide sequencing of fHbp genes obtained from 1837 invasive N. meningitidis serogroup B strains from the USA, Europe, New Zealand and South Africa [99]. Multilocus sequence typing analysis was performed on a subset of the strains. Every strain contained the fHbp gene. All sequences fell into one of two subfamilies (A or B), with 60–75% amino acid identity between subfamilies and at least 83% identity within each subfamily. One fHbp sequence may have arisen via inter-subfamily recombination. Subfamily B sequences were found in 70% of the isolates, and subfamily A sequences were found in 30% [99].

On 23 December 2010, Novartis Vaccines and Diagnostics applied to the EMA for authorization to market the multicomponent vaccine for Meningococcus B. Their dossier contained clinical and epidemiological data supporting the safety profile and immunogenicity of the vaccine and its probable broad coverage.

In the battle against invasive meningococcal disease, one of the greatest challenges is to come up with a vaccine that can provide protection against most of the meningococcal B strains that circulate worldwide. The antigens present in this vaccine have been identified in most of these strains [91,100,101].

Factor H-binding protein (rLP2086)

At the same time as Novartis, Wyeth (currently Pfizer) independently identified fHbp and developed a vaccine containing two protein variants (rLP2086). Indeed, in 2004 Fletcher et al. of the Wyeth Vaccines Research group published their results on a potential vaccine for N. meningitidis constituted by the lipidated 2086 protein from N. meningitidis[102]. The authors described a lipoprotein called LP2086, which was first observed within a complex mixture of soluble OMPs following a series of fractionation, protein purification and proteomics steps. They identified two different subfamilies of LP2086. Flow cytometry analyses and electron microscopy indicated that LP2086 was localized on the outer surface of N. meningitidis. Antiserum produced against a single protein variant was capable of eliciting bactericidal activity against meningococcus strains expressing different sero-subtype antigens. The authors concluded that one recombinant lipidated 2086 (rLP2086) variant from each subfamily with rPorA variants could be a good combination for further vaccine development. This LP2086 protein is the fHbp (also known as GNA1870) contained in the Novartis vaccine. Previously, Fletcher et al. had presented the characterization, cloning and expression of different subfamilies of the ORF 2086 gene from N. meningitidis during the 13th International Conference on Pathogenic Neisseria, which was held in Oslo in 2002 [103].

In 2005, Zhu et al. published the results of a preclinical study, performed in a murine model, on a recombinant lipidated group B meningococcal OMP rLP2086 [104]. These authors demonstrated that mice immunized with rLP2086 displayed significantly less nasal colonization after challenge with two different strains of group B N. meningitidis than did nonimmunized mice. Furthermore, immunized mice produced a strong systemic IgG response, and the serum antibodies were cross-reactive with heterologous strains of meningococcus B.

At the 26th Annual Meeting of the European Society for Pediatric Infectious Diseases, Nissen et al. presented the results of a randomized, placebo-controlled, double-blind, Phase I trial of ascending doses of meningococcal group B rLP2086 vaccine [105]. The authors recruited 103 subjects aged 18–25 years, who were subdivided into three cohorts which received ascending doses of rLP2086 vaccine or placebo on a 0-, 1- and 6-month schedule. By means of the SBA test, the authors ascertained that more than 87.5% of the subjects who received the highest dose (20 µg) of the antigens responded against five or six of the six meningococci tested.

In 2010, Richmond et al. presented an abstract at the 17th IPNC (International Pathogenic Neisseria Conference) concerning three clinical trials of a serogroup B N. meningitidis vaccine in adults and adolescents [106]. The vaccine contained antigenic components from subfamilies A and B of meningococcal fHbp (rLP2086). The authors concluded that the bivalent vaccine elicited hSBA titers >1:4 against subfamily A and B strains in a high proportion of adults and adolescents, and that no significant safety concerns were identified.

Molecular epidemiology has made a valuable contribution to the development and implementation of meningococcal vaccines [43,55,107,108]. In choosing protein components, it is important to determine the number of variants needed and to pick out those that can provide the broadest protection, if possible before vaccine testing in humans. While functional assays play a major role in this process, the diversity of nucleotide and peptide sequences constitutes an important indicator. The fact that fHbp displays many variants and the evidence for particular epitopes under immune selection indicate that it will be necessary to use formulations containing multiple components in order to achieve broad coverage, not least as cross-protection between the two main subfamilies and within subfamily A, variants 2 and 3, is limited [102,109,110]. An alternative approach is to create chimeric proteins containing domains from the different subfamilies/variants [111]. In any case, combined molecular epidemiological and functional analyses will be required in order to establish the optimum number of variants to use. Moreover, how long such vaccines remain usable will depend on how fHbp evolves in populations of meningococci and on the impact of vaccination on this evolution.

While fHbp is present in all meningococci, its degree of expression may differ among isolates [102,110]. In most meningococcal strains, its epitope surface exposure is less than that of other vaccine antigens [112]. Moreover, fHbp expression is a prerequisite to the survival of meningococci, especially of high-expressing strains, in ex vivo human blood and serum [98,113,114].

Table 2 shows a summary of human clinical trials of serogroup B meningococcal vaccines with recombinant proteins.

Expert commentary

Conjugate vaccines against meningococcus A, C, W 135 and Y are currently licensed and used. There are three monovalent conjugate vaccines for serogroup C, one monovalent conjugate vaccine for serogroup A and two quadrivalent conjugate vaccines. These vaccines generate T-cell-dependent responses and have several immunologic advantages over polysaccharide vaccines: they can induce herd immunity and stimulate immunologic memory, and do not induce immunologic hyporesponsiveness [115].

In recent decades, the incidence of invasive disease caused by serogroup C has declined in developed countries as a result of the introduction of vaccination programs with conjugated vaccine C in children and adolescents; these age-groups constitute the highest risk class in some countries. Since the introduction of the conjugate vaccines, meningococcus B has become the serogroup most frequently involved in invasive disease in Europe, North America and South America, where nearly half of all cases of meningococcal disease are caused by capsular group B strains; in many European countries the proportion is even higher (90%) [14,15].

The development of novel vaccines against meningococcus B disease could constitute an important step forward. N. meningitides B genome sequencing has enabled reverse vaccination and to discover new antigens, which could be very useful in optimizing the vaccine against N. meningitidis. Genomics provides a great opportunity for vaccine development, particularly in the case of microorganisms, for which the traditional approaches have failed [88]. Vaccine coverage, or rather the spectrum of vaccine efficacy, is a critical issue for protein-based vaccines, since any serogroup B disease that occurs following their introduction will be treated as a vaccine failure. However, while it is too early to forecast the precise impact of new vaccines on the natural history of the disease, many data are being accumulated. For instance, with regard to the Novartis vaccine, in 2006 Giuliani et al. published research that demonstrated that the vaccine adjuvated by aluminum hydroxide induced bactericidal antibodies in mice against 78% of a panel of 85 meningococcal strains, representative of the diverse global population. The strain coverage could be increased to 90% or higher by the addition of CpG oligonucleotides or by using MF59 as an adjuvant [90]. Findlow et al. conducted a study on 535 MenB isolated in England and Wales, and genetically characterized the component of the Novartis vaccines. These authors concluded that, on the basis of genotypic data and potential cross-reactivity, the 4CMenB vaccine has the potential to protect against a significant proportion of MenB disease in England and Wales [116]. Other encouraging results have been reported by other authors with regard to genotypic data on upcoming vaccines from both Novartis and Pfizer [117–119]. Regarding possible limitations in the efficacy of these more recent vaccines, it should also be said that potential cross-protective immunity may be lacking within fHbp variant families, as has been observed in clinical trials in infants [120].

Meningococcal B vaccines are designed to provide an optimal immune response against the majority of hypervirulent MenB strains, while at the same time taking into account the constantly changing nature of the bacterium. Future strategies for the development of meningococcal vaccines should be global, that is, aimed at designing a ‘universal vaccine’ effective against meningococcal disease due to any strain, regardless its phenotype and genotype [121].

Finally, because of the need for a broad immunoresponse against N. meningitis B, it could be useful to empower the multivalent vaccines by means of adjuvants, such as MF59, which is able to stimulate the Toll-like receptors, as, for instance, TLR-4.

Five-year view

Vaccines currently available on the market can only immunize against meningococcal disease caused by serogroups A, C, W 135 and Y. Infection due to N. meningitidis B, which has an important worldwide burden, has not yet been covered.

Recent advances in molecular genetics and the availability of the meningococcal genome sequence, have also allowed the development of new vaccine preparations against serogroup B meningococcal infection. Development of the Novartis vaccine (currently in European Medicines Agency submission) is now complete, and the Pfizer vaccine is at an advanced stage of clinical trial.

In the battle against invasive meningococcal infection, a major challenge is to come up with a vaccine that provides protection against most of the meningococcal B strains in circulation. These vaccines will constitute an important milestone in modern vaccinology. If their effectiveness against invasive disease is confirmed, the excellent immunogenicity results obtained in clinical trials will enable cases of invasive disease to be drastically reduced.

Vaccines against all forms of N. meningitidis will be licensed in the coming years. However, some important questions remain open: their effectiveness on the different geographic distribution of pathogenic strains, their impact on carriers and the timing of the appearance of escape mutants once herd immunity has been achieved remain to be verified. These vaccines could also provide protection against nonserogroup B strains.

The most important issue appears to be the impact on carriers. Asymptomatic carriage of N. meningitidis is common (5–35% of individuals) [122]. The prevalence of carriers is low in infancy, increases during childhood, and reaches its peak in adolescents and in young adults; thereafter, it declines in adults and the elderly [123]. N. meningitidis encounters a number of challenges during transmission and colonization. The bacterium expresses a number of surface structures and produces many substances during colonization. Although considerable advances have been made in research on the adhesion capacity of the bacterium to host cells, the exact mechanism is not completely known. At the same time, it is very important to ascertain the effect of these novel vaccines on mucosal immunity. Indeed, natural protective immunity depends on both mucosal immunity and T-cell memory. It is therefore essential to establish whether the new vaccines can re-program pre-existing naturally acquired mucosal immunity, in order to prevent disease and carrier status [124]. Only future studies will be able to clarify this key issue.

Furthermore, national policies for the use of the new vaccines will play a vital role in limiting the reservoir of the bacterium. In this regard, the priority targets of an efficacious and safe MenB vaccine appear to be adolescents and infants.

Finally, it is strongly advisable to program economic studies in order to determine the correct vaccination strategy. As yet, however, economic evaluations, such as cost–benefit or, better still, cost–utility analyses cannot be made. Indeed, the cost of the Novartis and Pfizer vaccines is not known. Nevertheless, it is possible, and indeed important, to develop decisional models to guide decision makers in planning future vaccination programs.

Table 1. Summary of studies of outer membrane vesicle vaccines against Neisseria meningitidis.

Location	Study period	Vaccine name and serologic classification^†	Age group	Disease incidence before vaccination (no. of cases per 100,000 population)	Study design	Doses (n)	Control vaccine or placebo	Persons vaccinated (n)	Vaccine efficacy or SBA fourfold antibody increase (%)^‡	Ref.
Cuba (Finlay Institute)	1987–1989	VA-MENCOG-BC B:4:P1.15	10–14 years	14.4	Prospective, randomized double-blind	Two	Placebo	106,000	83	[44]
	1987–1989	VA-MENCOG-BC B:4:P1.15	5 months to 24 years	30	Prospective, randomized double-blind	Two	Placebo	133,000	83–94
Iquique, Chile (Walter Reed Army Institute of Research)	1987–1989	B:15:P1.3	1–21 years	20	Prospective, randomized double-blind	Two	A, C, Y, and W135 polysaccharide vaccine	40,811	51	[47]
Norway (Norwegian Institute of Public Health)	1988–1991	MenBvac B:15:P1.7,16	13–21 years	25 (13–21 year)	Prospective, randomized double-blind	Two	Placebo	171,800	57.2	[43]
São Paulo, Brazil	1989–1990	VA-MENCOG-BC B:4:P1.15	3 months to 6 years	2.07 (1–6 year)	Retrospective, case–control	Two	Not applicable	2.4 million	47–74	[45]
	1990–1991	VA-MENCOG-BC B:4:P1.15	3 months to 6 years	2.07 (1–6 year)	Retrospective, case–control	Two	Not applicable	621	0–74
Santiago, Chile	1994	MenBvac B:15:P1.7,16 VA-MENCOG-BC B:4:P1.15	<12 months	5,9	Prospective, randomized double-blind	Three	Placebo	187	Homologous strain >90 Heterologous strain 0	[48]
			2–4 years					183	Homologous strain >67 Heterologous strain 31–35
			7–30 years					173	Homologous strain >67 Heterologous strain 37–60
Rotterdam (The Netherlands)	1996	Exavalent PorA-OMV	2–8 years		Open randomized, case–control	Three	HBV	337	16–100	[79]
New Zealand (Novartis Vaccines and Norwegian Institute of Public Health)	2001–2006	MeNZB B:4:P1.7–2,4	6 months – 20 years	17	Vaccination campaign^§	Three (four in infants after 2006)	Not applicable	905,507	73	[62]
	2004–2006	MeNZB B:4:P1.7–2,4	6 months – 5 years	17,4	Vaccination campaign	1–3	Not applicable	258,421	80
			6 months – 3 years					143,625	84, 8
Normandy, France	2006–2009	MenBvac B:15:P1.7,16	2 months – 19 years	18–30	Vaccination campaign	2 + 1 3 + 1	Not applicable	26,014	After 2 dose 37 After 2 + 1 dose at 6 week 88 After 2 + 1 dose at 15 months 56	[46]
The Netherlands	2007	NonaMen PorA OMV	Non human model (mice and rabbits)		Experimental study					[82]
UK	2009		Adults		Prospective, randomized double blind	Three	Placebo	97	8–31	[63]
USA (Walter Reed Army Institute of Research)	2011	NOMV	Adults		Phase I trial study	Three		36	41–81	[67]

^†The serologic classification of the Neisseria meningitidis strain used in the vaccines includes the serogroup (capsular polysaccharide), the serotype (PorB outer-membrane protein) and the serosubtype (PorA outer-membrane protein).

^‡Vaccine efficacy is based on the ideal conditions that characterize clinical trials and compares the incidence of meningococcal disease in vaccinated persons with that in unvaccinated persons or in those administered placebo. Vaccine effectiveness is based on the field conditions of vaccine-rollout programs and compares the percent reduction in attack rates among vaccinated persons as compared with those not vaccinated.

^§There were no Phase III trials of MeNZB; instead, a vaccination campaign was launched on the basis of data on immunogenicity and safety from Phase I and II trials and on data from studies of other, similar vaccines (mainly OMV vaccines). Thus, an estimate of vaccine effectiveness is provided instead of vaccine efficacy.

NOMV: Native outer membrane vesicle; OMV: Outer membrane vesicle; SBA: Serum bactericidal activity.

Table 2. Summary of human clinical trials of serogroup B meningococcal vaccines with recombinant proteins.

Vaccine composition	Clinical status	Age group	Number doses	Results	Ref.
5CVMB (five component vaccine against MenB) (rMenB) GNA1030-GNA2132 GNA2091-fHbp NadA	Phase II	Infants	Three at 2, 4 and 6 months of age	The percentage of infants with hSBA titers of 1:4 or higher against reference strains were 89, 96 and 85% after the third dose	[93]
5CVMB	Phase II	Infants	Three doses + a booster dose at 12 months	The percentage of infants with hSBA titers of 1:4 or higher against reference strains were 100, 98 and 93% after the booster dose	[93]
5CVMB (rMenB) + OMV strain NZ98/254	Phase II	Infants	Three at 2, 4 and 6 months of age	Three doses of rMenB + OMV in infants induce bactericidal antibodies against more strains than rMenB alone, demonstrating the potential for broader vaccine prevention of MenB disease	[94]
4CMenB GNA1030-GNA2132 GNA2091-fHbp NadA	Phase III	Infants	Three doses at 2, 4 and 6 months of age	The percentage of subjects with hSBA ≥1:5 against three MenB strains (5/99, NZ98/254 and H44/76) were 100, 84 and 100%, respectively, after the third dose	[95]
4CMenB + routine infant vaccines	Phase III	Infants	Three doses at 2, 4 and 6 months of age	4CMenB has acceptable tolerability profile when coadministered with other routine infant vaccines	[96]
4CMenB + Menveo^®	Phase II	18–50 years	Three doses at 0, 2 and 6 months	Bactericidal immune responses were evident after each dose of 4CMenB	[97]
rLP2086 (subfamilies A and B of meningococcal factor-H binding protein)	Phase I	18–25 years	0, 1, 6 months schedule with ascending doses	>87.5% of the subjects that received the highest dose (200 µg) of the antigens responded against five or six of the six meningococci tested by the SBA test	[105]
rLP2086 (subfamilies A and B of meningococcal factor-H binding protein)	Phase I	Adults and adolescent	0, 1, 6 months	hSBA titers >1:4 against subfamily A and B strains were present in a high proportion of adults and adolescents	[106]

hSBA: Human serum bactericidal antibody; OMV: Outer membrane vesicle; SBA: Serum bactericidal activity.

Key issues

• • Invasive infections caused by Neisseria meningitidis are a serious public health problem worldwide, entailing a high mortality rate and severe sequelae in survivors and generating a heavy economic impact, not only in epidemic areas but also in areas where the disease is sporadic.

• • Meningococcal C conjugate vaccines, tetravalent conjugate vaccines (serogroups A, C, Y and W-135) and a meningococcal A conjugate vaccine are currently on the market.

• • Meningococcal serogroup B differs from the other pathogenic serogroups in that it has a capsular polysaccharide identical to the polysialic acid (α(2–8)N-acetylneuraminic acid) present in many human glycoproteins; in particular, it is similar to carbohydrates found in fetal brain tissue.

• • Serogroup B N. meningitidis causes a substantial proportion of the meningococcal infections endemic in many areas of the world, including the USA and Europe, and may cause prolonged epidemics.

• • Several outer membrane vesicle vaccines have been developed and tested in large-scale efficacy studies in Norway, Cuba, Brazil, Chile and New Zealand. The potential problem of vaccines based on outer-membrane vesicles is the ability of outer-membrane proteins of the bacteria to undergo antigenic shift or gene deletion, as seen in the case of PorA.

• • The sequence of the MenB bacterium genome has been determined and novel antigens have been discovered with a view to developing a universal vaccine against N. meningitidis B.

• • Universal vaccines that are also active against serogroup B infections are being developed and will soon appear on the market. These vaccines are expected to drastically reduce cases of invasive meningococcal infections.

Financial & competing interests disclosure

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

No writing assistance was utilized in the production of this manuscript.