Advances with vaccination against Neisseria meningitidis

Authors


Corresponding Author Ray Borrow, Health Protection Agency, Manchester, UK. Email: ray.borrow@hpa.org.uk

Abstract

In the last decade, meningococcal serogroup C conjugate vaccination programs have been demonstrated to be hugely successful with a truly impressive public health impact. In sub-Saharan Africa, with the implementation of an affordable serogroup A conjugate vaccine, it is hoped that a similar public health impact will be demonstrated. Challenges still remain in the quest to develop and implement broadly protective vaccines against serogroup B disease. New, broad coverage vaccines against serogroup B are for the first time becoming available although little is known about their antibody persistence, effectiveness or effect on nasopharyngeal carriage. Enhanced surveillance following any potential vaccine introduction against serogroup B needs to be thoroughly implemented. The future now holds a distinct possibility, globally, for substantially decreasing meningococcal disease, regardless of infecting serogroup.

Abstract

Au cours de la dernière décennie, les programmes de vaccination conjugués avec le méningocoque du sérogroupe C ont démontrés un énorme succès avec un impact véritablement impressionnant sur la santé publique. En Afrique subsaharienne, avec l’implémentation d’un vaccin conjugué du sérogroupe A à prix abordable, il est à espérer qu’un impact similaire sur la santé publique sera démontré. Les défis restent encore dans la quête pour développer et implémenter des vaccins à large protection contre la maladie due au sérogroupe B. De nouveaux vaccins avec une large couverture contre le sérogroupe B deviennent pour la première fois disponibles quoique peu soit connu sur la persistance de leurs anticorps, leur efficacité ou leur effet sur le portage nasopharyngé. Une surveillance accrue après toute introduction d’un vaccin potentiel contre le sérogroupe B devrait être sérieusement implémentée. L’avenir nous réserve maintenant une possibilité distincte, globalement, pour diminuer sensiblement la méningococcie, quel que soit le sérogroupe infectant.

Abstract

Durante la última década, se ha demostrado que los programas de vacunación con la vacuna meningocócica conjugada frente al serogrupo C han sido muy exitosos, con un impacto en salud pública realmente impresionante. En África subsahariana, con la implementación de una vacuna conjugada frente al serogrupo A y asequible, se espera demostrar un impacto similar en salud pública. Los retos continúan en la búsqueda del desarrollo e implementación de vacunas que confieran protección amplia contra la enfermedad causada por el serogrupo B. Nuevas vacunas, con una amplia cobertura frente al serogrupo B, comienzan a estar disponibles por primera vez, aunque se conoce poco sobre la persistencia de anticuerpos y la efectividad o el efecto sobre los portadores nasofaríngeos. Es necesario implementar una vigilancia mejorada tras la introducción de cualquier vacuna potencial contra el serogrupo B. En el futuro ahora se vislumbra una posibilidad clara, a nivel global, de una disminución sustancial de la enfermedad meningocócica, independientemente del serogrupo.

Introduction

The first clinical description of meningococcal meningitis was over two centuries ago, in 1805, where an outbreak of ‘epidemic cerebrospinal fever’ was reported from Geneva, Switzerland (Vieusseux 1806). Unfortunately, to date, despite the availability of effective antibiotics and vaccination campaigns, Neisseria meningitidis still remains a leading cause of meningitis and septicaemia globally. In England and Wales, for 2009 alone 1031 laboratory confirmed cases of invasive meningococcal disease were reported, with 36% of these occurring in those younger than 2 years (Health Protection Agency 2009). In the absence of treatment, the mortality rate as a result of meningococcal meningitis can exceed 50% and, even with appropriate care, at least 10% of patients die, typically within 24–48 h of the onset of symptoms. Approximately 10–20% of survivors are left with permanent sequelae such as mental retardation, deafness, epilepsy or other neurological disorders. The main burden of disease remains in non-industrialised countries, for example, in the meningitis belt of sub-Saharan Africa. From 1 January to 6 March 2011, the Ministry of Health of Chad reported 923 suspected cases of meningococcal disease including 57 deaths, a case-fatality rate of 6.2% (WHO 2011).

Of the 12 meningococcal serogroups, the vast majority of infections are caused by only six serogroups, A, B, C, W135, X and Y (Harrison et al. 2009). The meningococcus commonly colonises the nasopharynx of up to 40% of the adult population but only occasionally causes invasive disease (Tan et al. 2010). The epidemiology of the virulent serogroups differs with serogroup A responsible for major epidemics in the sub-Saharan African meningitis belt, where the incidence of disease may approach 1000 per 100 000 population (Stephens et al. 2007). Outbreaks of serogroup X have also been reported from the meningitis belt; for example, in Niger from January to June 2006, 51% of the 1139 confirmed cases of meningococcal meningitis were serogroup X (Boisier et al. 2007). Serogroup W135 has been responsible for worldwide outbreaks associated with the Hajj pilgrimage in 2000–2002 (Borrow 2009) as well as being associated with the meningitis belt (Mueller et al. 2006). Serogroup B is the predominant cause of endemic disease in industrialised countries, and also prolonged epidemics in countries such as New Zealand in the 1990s (Baker et al. 2001). Serogroup C has been a causative agent in outbreaks in adolescents and young adults, mostly in industrialised countries (Morrow et al. 1990; Imrey et al. 1996). Increased rates of disease have been attributed to serogroup Y in the USA since the 1990s (Rosenstein et al. 1999) and from 2009 in some European countries (Hedberg et al. 2011; Ladhani et al. 2012).

For epidemiological purposes, meningococci have been further classified into types and subtypes on the basis of variation in PorB and PorA outer membrane proteins (OMPs), respectively (Tsai et al. 1981; Frasch et al. 1985; Maiden et al. 1999). The genetic relationships among meningococci are determined using multilocus sequence typing (MLST), which capitalises on the development of high-throughput nucleotide sequence analysis and is based on the sequence polymorphisms in housekeeping genes (Maiden et al. 1998). MLST has been used globally to monitor the epidemiology of meningococcal disease and to determine the relationships between isolates in localised outbreaks. Sequence type (ST)-32, ST-11, ST-8, ST-41/44, ST269 clonal complexes (cc) have been responsible for the majority of invasive serogroup B and C diseases since the 1960s, whilst serogroup A has historically been associated with ST-1, ST-4, ST-5 and ST-7 (Caugant 2008).

As meningococcal disease can be rapidly fatal within a few hours, early treatment with effective antibiotics is necessary. The meningococcus is sensitive to a range of antibiotics including benzylpenicillin, third-generation cephalosporins and fluoroquinolones (Stephens 2007). The most effective means of combating meningococcal infection is through vaccination. Early successful vaccines based on the capsular polysaccharide of serogroups A, C, Y and W-135 have been refined by the development of glycoconjugate vaccines, introduced in the late 1990s and now in widespread use (Jodar et al. 2002). No such polysaccharide-based vaccine is licensed for serogroup B because of structural similarities of the serogroup B polysaccharide and that found in certain human tissues (Wyle et al. 1972; Finne et al. 1983), and approaches have been largely based on subcapsular outer membrane antigens.

This review details the experience with the use of conjugate vaccines together with recent advances and progress towards a broad coverage vaccine against serogroup B disease.

Surrogates of protection against N. meningitidis

Clinical efficacy studies to support licensure of vaccines are the clearest way of demonstrating benefit but are not possible for N. meningitidis owing to low incidence and sporadic occurrence of cases of disease. Therefore, new meningococcal vaccines have been licensed on the basis of immunogenicity and safety. Serum antibodies confer protection against meningococcal disease by activating complement-mediated bacteriolysis. In the USA in the 1960s, Goldschneider et al. (1969) demonstrated the role of serum complement–mediated bactericidal antibodies in protection against meningococcal disease in military recruits. Recruits were followed for their first 9 weeks of training, and it was found that recruits with serum bactericidal antibody (SBA) titres of four or greater, who were exposed through a serogroup C meningococcal epidemic, did not develop disease. Conversely, virtually all cases occurred in individuals whose baseline SBA titres were <4 as measured with human complement. The importance of SBA activity in protection against meningococcal disease is also underscored by the greatly increased risk of acquiring meningococcal disease in persons with inherited complement deficiencies, thus lacking SBA activity (Figueroa et al. 1993; Fijen et al. 1999). Indirect evidence, also from Goldschneider et al. (1969), was the demonstration of an inverse relationship between the average age of acquisition of naturally acquired SBA and the incidence of meningococcal disease. Thus, the seminal study defining SBA activity as the surrogate of protection against developing meningococcal disease used human complement in the SBA assay (Goldschneider et al. 1969). As it is difficult to source sufficient volumes of human sera that lack antibodies against meningococci, many laboratories use baby rabbit serum as an exogenous source of complement. Suitable complement preserved baby rabbit sera are widely available and easier to standardise between different laboratories (Maslanka et al. 1997; Jodar et al. 2000). Using estimates of age-specific vaccine effectiveness, SBA titre of ≥8, as measured with baby rabbit complement, has been demonstrated to be the putative protective threshold following serogroup C conjugate vaccination (Borrow et al. 2001a; Andrews et al. 2003).

Experience with meningococcal conjugate vaccination

In the autumn of 1999, the UK was the first country to establish a national immunisation programme for meningococcal serogroup C conjugate (MCC) vaccines. This was the culmination of a rigorous clinical trial research programme involving partnership between the Department of Health (DoH), public bodies, academia and vaccine manufacturers (Miller et al. 2001). Pre-licensure studies commenced as early as 1994 when the UK DoH funded the initial MCC vaccine trials (Miller et al. 2001). Following promising immunogenicity and safety studies in infants, vaccinated under the UK 2-, 3- and 4-month schedules (Fairley et al. 1996; Richmond et al. 1999, 2001a; Borrow et al. 2000), further clinical trials to answer key policy–related questions were commenced. These trials determined the schedule to be used for catch-up immunisation of older age groups (Richmond et al. 2001b), the effect of prior meningococcal serogroup C polysaccharide vaccination on the response to MCC vaccines (Richmond et al. 2000; Borrow et al. 2001b,c), and the compatibility with other concomitantly administered childhood vaccines (Burrage et al. 2002). All three candidate MCC vaccines studied were found to be safe, elicit functional antibody response and immune memory in 1- to 2-year-olds after a single dose (Richmond et al. 2001b), justifying the use of a one dose catch-up schedule for those aged 1–18 years.

In those aged 20 years or younger serogroup C disease incidence fell from 5.34 to 0.03 per 100 000 population from 1998–1999 to 2008–2009, a dramatic reduction of 99% (Campbell et al. 2010). Vaccine effectiveness, however, declined significantly more than 1 year after the administration of a three-dose course in infancy. Despite this finding, the marked impact on serogroup C disease has been sustained with the lowest ever recorded incidence (0.02 per 100 000 population) in 2008–2009, mainly because of the indirect herd protection of the vaccine in reducing carriage (Maiden et al. 2002; Ramsay et al. 2003). Updated estimates of vaccine effectiveness to 30 June 2009 confirmed high short-term protection after vaccination in infancy at 97% (95% CI, 91–99%) falling to 68% (95% CI, −63 to 90%) more than a year after vaccination (Campbell et al. 2010). A similar fall in effectiveness was reported from Spain where MCC vaccines were introduced into their childhood immunisation schedule in the last quarter of 2000 using a 2-, 4- and 6-month schedule. Estimates of effectiveness during the first year following MCC vaccination was 98% (95% CI, 96–99%) but declined to 78% (95% CI, 3–95%) beyond 1 year (Larrauri et al. 2005). A similar decline in effectiveness was also noted in the province of Quebec, Canada, in those vaccinated below 2 years of age, 2 years out from vaccination (de Wals et al. 2011). EU-IBIS has published a listing of meningococcal conjugate serogroup C vaccination programmes in the EU, as at October 2007 (EU-IBIS 2007).

A decrease in the disease rate was observed in all age groups, including those not targeted for immunisation. Presence of immune memory was demonstrated not to be predictive of long-term protection as MCC vaccine failures occurred in children who had been primed for immune memory (Auckland et al. 2006), confirming the importance of circulating antibodies in the prevention of invasive disease.

To counteract the decline in effectiveness following the three dose primary schedule, the UK moved the 2-month primary dose of MCC to 12 months of age, combining this with a Haemophilus influenzae type b booster in the form of a combination MCC/Hib-TT vaccine (Chief Medical Officer 2006). A study examining the antibody persistence following this new schedule found firstly that the magnitude of the meningococcal SBA GMT was higher for those subjects primed with MCC-TT as opposed to either of the MCC-CRM197 vaccines up to 1 year following boosting. Surprisingly, 2 years after boosting, the percentages of subjects with SBA titres ≥8 for children primed with MCC-TT or MCC-CRM197 had declined to 43% and 22% or 23% (Menjugate and Meningitec), respectively (Borrow et al. 2010). Current studies in the UK have now focused on dropping to a single dose of MCC vaccine at 3 months of age with the booster at 12 months giving the possibility of a cost-neutral move of an infant dose to adolescence (Findlow et al. 2012). Antibody persistence 5 years following MCC vaccination in children aged 10 years or above has been demonstrated to be more prolonged as oppose to younger age groups, possibly owing to immunological maturation (Snape et al. 2008a).

To minimise the number of doses of MCC vaccine given, other countries, such as the Netherlands, opted for a single-dose strategy from 14 months of age, together with catch-up campaign to 19 years of age. In the Netherlands, up to April 2006, the number of serogroup C cases had fallen from 276 in 2001 to four in 2005 with no vaccine failures (de Greeff et al. 2006). In 2005, a quadrivalent meningococcal A, C, W135, Y diphtheria toxoid conjugated vaccine (MenACWY-D) was licensed in the USA. The US Advisory Committee on Immunization Practices (ACIP) recommended this quadrivalent vaccine for all adolescents of 11–18 years as well as for those at increased risk of meningococcal disease from 9 months of age (Centers for Disease Control and Prevention (CDC) 2011a,b). The effectiveness for the first 5 years was estimated, for serogroups C and Y, as 77% (14–94%) and 88% (−23 to 99%), respectively (MacNeil et al. 2010, CDC 2011a,b). When estimated by time from vaccination, the effectiveness was found to decline, that is, within the first year, 95% (10–100%); one to 2 years post-vaccination, 91% (10–101%); and 2–5 years, 58% (−72 to 89%). This was different from that observed in the UK with MCC vaccines, where for a comparable age group of 11–16 years, the overall effectiveness was 97% (93–99%) for the first decade since MCC introduction (Campbell et al. 2010). The decline in effectiveness and corresponding decrease in SBA levels in the USA have now led to the recommendation of booster doses of quadrivalent conjugate vaccine (CDC 2011a,b). The new recommendation, now approved by Advisory board on Immunization Practices (ACIP), was that routine vaccination of adolescents preferably at the age of 11 or 12 years should be boosted with a further dose at the age of 16 years (CDC 2011a,b).

Recent advances in conjugate vaccines

New developments with quadrivalent A, C, Y and W135 conjugate vaccines

Since the licensure of MenACWY-D (Menactra), in the USA in 2005, a further quadrivalent vaccine, utilising CRM197, has now been licensed in Europe, USA, the Middle East and Latin America (Bröker et al. 2011). MenACWY-CRM197 showed protective antibody titres in all age groups studied, with a safety profile comparable with that of a polysaccharide vaccine. MenACWY- CRM197 was the first quadrivalent conjugate vaccine to be shown to be immunogenic in young infants (Snape et al. 2008b; Perrett et al. 2009). A preliminary report of data from the USA cohort of a large Phase III trial conducted in infants in Latin America and the USA showed that MenACWY-CRM197 administered at 2, 4, and 6 months of age was safe and induced putatively protective immune responses (Bröker et al. 2011). Although not licensed for infants, the UK Department of Health ‘The Green Book’ advises MenACWY-CRM197 for infants under 1 year of age who are travelling to endemic areas as a two dose schedule, 1 month apart (Department of Health, 2007), whilst in the USA, from April 2011, the Food and Drug Administration approved the use of MenACWY-D as a two dose primary series, given 3 months apart, among children aged 9 through 23 months at increased risk of meningococcal (CDC 2011a,b). Because of their high risk for invasive pneumococcal disease, children with functional or anatomic asplenia in the USA are recommended to be vaccinated with MenACWY-D beginning at the age of 2 years to avoid interference with the immunological response to the infant series of pneumococcal conjugate vaccines (CDC 2011a,b).

A third meningococcal quadrivalent vaccine, conjugated to tetanus toxoid (MenACWY-TT), has now progressed through Phase II and Phase III studies and has been recently reviewed (Miller et al. 2011). These studies have shown that a single dose of MenACWY-TT was safe and induced putatively protective immune responses in toddlers, children, adolescents and adults. The SBA GMTs against the serogroup C portion were significantly higher following MenACWY-TT vaccination in toddlers in the second year of life when compared to vaccination with a licensed monovalent CRM197 conjugated MCC vaccine (Knuf et al. 2010, 2011; Vesikari et al. 2011).

Monovalent serogroup A conjugate vaccines

For over a century, major meningococcal serogroup A meningitis epidemics have occurred approximately every 10–12 years in the African ‘meningitis belt’ which stretches from Senegal to Ethiopia (Greenwood 1999; Molesworth et al. 2002). Recognising the need for better meningococcal vaccines for Africa, the Meningitis Vaccine Project (MVP), a partnership between the WHO and PATH, was established in 2001 with funding from the Bill and Melinda Gates Foundation with the goal of eliminating serogroup A meningococcal epidemics by the development, testing, licensure and introduction of a serogroup A meningococcal conjugate vaccine (LaForce et al. 2007). Affordability of this new conjugate vaccine by African countries was the key goal, and MVP, along with its partners, developed a new serogroup A meningococcal conjugate vaccine (MenAfriVac™, http://www.menafricar.org) utilising tetanus toxoid as its carrier protein, manufactured by the Serum Institute of India Ltd., Pune, India, at a cost of less than US$ 0.50 per dose (LaForce et al. 2007).

Following a Phase I trial in healthy adults in India (Kshirsagar et al. 2007), a pivotal clinical trial was conducted in 12- to 23-month-olds in West Africa (Mali and The Gambia) with an additional trial in healthy 2- to 29-year-old Africans (Mali, The Gambia and Senegal) to evaluate safety and immunogenicity of a single dose of serogroup A conjugate vaccine as compared to that of the serogroup A component of a licensed quadrivalent polysaccharide vaccine (Sow et al. 2011). Following a clinical Phase II/III trial, a lot of both consistency trial was performed in India and a large safety study was recently completed in Mali, a total of about 10 000 subjects had been vaccinated with no safety concerns in any age group evaluated (MVP 2011). After a single dose, in all age groups studied (1–29 years), the serogroup A conjugate vaccine elicited superior functional responses, as compared to the licensed polysaccharide vaccine, and in serogroup A–specific IgG, priming for immune memory, boosting of the anti-tetanus toxoid IgG was also demonstrated (Sow et al. 2011).

The vaccine was subsequently granted marketing authorisation for export by the Drugs Controller General of India in 2009 and gained WHO pre-qualification in 2010 with an indication for use in 1–29 years of age (MVP 2011). Licensure in Burkina Faso, Mali and Niger was also granted in 2010. Mass vaccination campaigns in the 1–29 years of age across these three countries resulted in a total of approximately 20 million vaccinated people between September and December 2010, with excellent vaccine coverage and good safety records. Initial data following the introduction of the vaccine are extremely promising with a dramatic fall in the cases of serogroup A disease in the three countries who report the lowest number of confirmed serogroup A cases ever recorded during a ‘meningitis epidemic season’, with no cases of serogroup A disease reported in the infant population (MVP 2011). Continued surveillance for the cases of meningitis and monitoring of vaccination coverage will be crucial to confirm the effects of this serogroup A conjugate vaccine as it is introduced across the meningitis belt. A stable fraction of non-serogroup A meningococcal isolates have been reported from those countries who have introduced to date. Immediate plans for 2011 comprise completion of country-wide vaccinations in Mali and Niger and introduction in three new countries of the meningitis belt such as Cameroun, Chad and Nigeria, whilst Senegal, Benin and The Gambia are planning to introduce the vaccine in 2012. Other countries in the meningitis belt are planning to introduce the vaccine before the end of 2015 (MVP 2011).

It is hoped that this conjugate vaccine will also have an effect on carriage and transmission as demonstrated with serogroup C conjugate vaccines in the UK (Maiden et al. 2002). To assess the population effects of the vaccination campaigns on carriage, studies are underway in a number of countries throughout the meningitis belt (African meningococcal carriage consortium 2011; Kristiansen et al. 2011). The main strategy for serogroup A conjugate vaccine introduction in Africa is to conduct mass campaigns of the entire target population aged 1–29 years of age, using a single dose of vaccine. Following this, to protect new birth cohorts and to maintain population immunity after the initial mass campaigns, single dose follow-up campaigns every 5 years that target all 1- to 4-year-olds could be performed, through strategies aiming at routine protection in infancy or in the second year of life. A dose-ranging clinical study commenced in November 2008 is currently being conducted in infants in Northern Ghana to evaluate the safety and immunogenicity of three different formulations of the serogroup A conjugate vaccine administered in a two-dose schedule at 14 weeks and 9 months concomitantly with EPI vaccines (MVP 2011). The introduction of this affordable serogroup A conjugate vaccine is a giant step towards achieving elimination of epidemic meningitis as a public health problem in sub-Saharan Africa. Widespread use of the vaccine throughout much of Africa may prevent more than a million cases of meningitis over the next decade. This serogroup A conjugate vaccine is expected to be cost-saving when compared to current expenditures on these epidemics; for example, an analysis shows that introducing it in seven highly endemic countries could save $350 million or more over a decade (LaForce & Okwo-Bele 2011).

Effect of polysaccharide vaccination on the immune response to subsequent conjugate vaccination

Since the 1970s, it has been known that a second dose of meningococcal serogroup C polysaccharide results in lower immune responses following the second dose as to the first dose (Gold et al. 1979). This has been observed in infants (Gold et al. 1975, 1979), toddlers (MacDonald et al. 1998) and adults (Granoff et al. 1998; Jokhdar et al. 2004). Similar lower responses for serogroup W135 have also been observed where, following a full dose of quadrivalent polysaccharide vaccine at 12–23 months of age and a 1/5 dose of quadrivalent polysaccharide vaccine 10 months later, lower responses were seen as compared to age-matched vaccine-naïve subjects receiving a 1/5 dose of quadrivalent polysaccharide vaccine (Findlow et al. 2011). Serogroup Y antibody levels were not measured in this study. Serogroup A polysaccharide appears to behave differently in that repeated doses can lead to boosting (Gold et al. 1979; Jokhdar et al. 2004).

Polysaccharide vaccination can also impair subsequent immune responses to conjugate vaccination. Infants and teenagers previously vaccinated with meningococcal polysaccharide vaccine have shown antibody hyporesponsiveness to a subsequent dose of meningococcal conjugate vaccine (MacDonald et al. 1998; Borrow et al. 2001c; Southern et al. 2004; Keyserling et al. 2005). If polysaccharide vaccine is administered following conjugate vaccination, this results in loss of the ability to mount a subsequent booster or immune memory antibody response (MacLennan et al. 2001). The potential mechanism is thought to be that the polysaccharide vaccination is depleting the memory B cell pool.

Advances in vaccines to target serogroup B meningococcal disease

Although polysaccharide and conjugate vaccines against serogroups A, C, Y and W135 have been available for a number of decades, no serogroup B polysaccharide-based vaccine is licensed owing to the poor immunogenicity in humans and concerns over safety owing to the possible induction of autoantibodies (Wyle et al. 1972; Finne et al. 1983). This has led to different strategies to develop a serogroup B meningococcal vaccine. These approaches include detergent-extracted outer membrane vesicle (OMV) vaccines, which have been implemented successfully during clonal outbreaks (Bjune et al. 1991; Sierra et al. 1991; de Moraes et al. 1992; Boslego et al. 1995; O’Hallahan et al. 2009). These OMV vaccines, although effective against strains with the same PorA subtype, only offer a relatively narrow breadth of protection against heterologous strains, particularly in infants and young children (Tappero et al. 1999; Wong et al. 2007; Galloway et al. 2009). Therefore, more recent approaches in the design of universal vaccines have made use of the genomic era together with conventional approaches (Rappuoli 2000; Fletcher et al. 2004). Progress of the four most advanced approaches in vaccine development against serogroup B disease is discussed below although other promising candidate antigens are at the pre-clinical stage. To fully understand the different serogroup B vaccines, the following section describes each of the potential vaccine antigens. It should be noted that only the potential antigens included in the four vaccines discussed are detailed and there are many other important candidates such as transferring-binding proteins (Tbp A and Tbp B) (Gorringe et al. 1995), ferric enterochelin receptor (FetA, formerly known as FrpB) (Ala’Aldeen et al. 1994), zinc uptake component D (ZnuD) (Stork et al. 2010) and macrophage infectivity potentiator protein (MIP) (Hung et al. 2011).

Porin A (PorA)

The PorA protein forms a protein trimer within the outer membrane and has eight surface-exposed loops with hypervariable regions (VR1 and VR2) at the apices of loops 2 and 4, the two most prominent and exposed (van der Ley et al. 1990; Poolman et al. 1995; Song et al. 1999). Antibodies against these loops are particularly effective in SBA assays and in animal protection studies (Saukkonen et al. 1989). SBA activity correlates with protection against systemic meningococcal disease as evidenced from trials of OMV vaccines (Borrow et al. 2006). Meningococci are classified into subtypes by their VR1 and VR2 regions. PorA VR epitopes are given in order after the prefix P1 and are separated from one another by a comma. PorA subtypes are further classified into variants by VR sequence typing (Maiden & Feavers 1994; Maiden et al. 1999). To date, 10 PorA VR1 families exist with a total of 220 variants, whilst for the VR2 there are 21 families with 616 variants (Neisseria PorA typing 2011). PorA proteins have long been perceived as potential serogroup B vaccine components, despite their intrinsic structural variability and variability in the level of their expression.

Neisserial adhesin A

NadA promotes adhesion to, and invasion of, epithelial cells, and deletion of the nadA gene has been shown to reduce invasion of meningococci into the epithelium (Comanducci et al. 2002; Capecchi et al. 2005). A total of five genetically distinct NadA variant groups have identified of which NadA1, NadA2 and NadA3 immunologically cross-reactive whilst these do not cross-react with NadA4 and NadA5 (Comanducci et al. 2002, 2004; Findlow et al. 2010; Snape et al. 2010). The level of expression varies between isolates by as much as or more than 100-fold (Donnelly et al. 2010) and is regulated by the repressor protein NadR (Metruccio et al. 2009) as well as phase variation (Martin et al. 2003). The gene encoding NadA is present in cc32, cc11, and cc213 but not associated with cc41/44 or cc269; however, cc213 are normally associated with NadA5 (Comanducci et al. 2002; Lucidarme et al. 2009). NadA has been shown to be immunogenic in animal models (Comanducci et al. 2002), and also, it has been demonstrated that specific antibody responses against NadA are induced following invasive disease (Litt et al. 2004).

Neisserial heparin-binding antigen

NHBA, initially designated GNA2132 prior to the understanding of its heparin-binding function, is present in most strains and expressed in vivo as documented by the detection of antibodies in serum samples from patients recovering from meningococcal disease (Serruto et al. 2010). Serum antibodies from mice immunised with recombinant NHBA have been shown to elicit SBA activity (Pizza et al. 2000; Giuliani et al. 2006). In addition, antibodies against NHBA may also act synergistically with antibodies against other meningococcal antigens (Biolchi et al. 2010).

Factor H binding protein

Hbp was first identified as a surface-exposed lipoprotein during the screening of the MC58 genome (Pizza et al. 2000). N. meningitidis evades complement-mediated killing by the binding of the host complement alternative pathway inhibitor, factor H by fHbp resulting in down regulation of the complement pathway and therefore increased bacterial survival (Schneider et al. 2009). Therefore, the inclusion of fHbp in a vaccine may have a dual effect, inducing SBA activity and also increasing the susceptibility of N. meningitidis to bactericidal activity by blocking factor H binding to the bacterium (Beernink et al. 2007). Whether the latter happens in vivo is questionable given that an investigational mutant fHbp vaccine that does not bind factor H was found to be superior in transgenic mice, eliciting higher SBA activity than that of fHbp vaccination itself (Beernink et al. 2011). In the case of mutant fHbp vaccination, the resultant antibodies may be directed more at epitopes in or near the fH binding site, which result in greater SBA activity; these epitopes may be obscured when factor is bound to the wild-type fHbp vaccine.

There are currently three classification schemes for fHbp, firstly that housed at http://pubmlst.org (Neisseria Factor H binding protein sequence typing 2011) which classifies peptide variants with a sequential number on submission to the database (fHbp pubmlst.org website). A second scheme classifies fHbp into three variants, 1, 2 and 3 (Masignani et al. 2003; Bambini et al. 2009), each variant is then further divided into related subvariants, for example 1.1 and 1.14 [71]. A third scheme is where fHbp is divided into two subfamilies (A [corresponding to variants 2 and 3] and B [corresponding to variant 1]) (Fletcher et al. 2004). Among serogroup B invasive disease isolates from England and Wales from 2008, 64, 20 and 16% had variant 1, 2 and 3 genes, respectively (Lucidarme et al. 2010), whilst in serogroup B isolates collected from California, USA, in 2003/2004, 83, 13 and 4% had variant 1, 2 and 3 genes, respectively (Beernink et al. 2006). In a global study of 1837 invasive serogroup B isolates from the Europe, New Zealand, South Africa and the USA, 70% harboured variant 1 (subfamily B) genes and 30% harboured variant 2/3 (subfamily A) genes (Murphy et al. 2009). From pre-clinical studies, it was determined that fHbp variant 1 elicited SBA activity against diverse variant 1 subvariants but was poorly cross-reactive with subvariants of variants 2 and 3 (Masignani et al. 2003; Fletcher et al. 2004). It has also been shown that the levels of expression of fHbp can affect SBA activity (Pajon et al. 2010).

Trivalent native outer membrane vesicle vaccine

As detergent-extracted OMVs are depleted of most of the surface lipoproteins and lipooligosaccharide (LOS), antigens that are capable of eliciting cross-protective SBA, the Walter Reed Army Institute of Research (WRAIR) took the approach to investigate native OMVs in which three genes, siaA (also known as synX), lpxL1 or lpxl2 and lgtA have been knocked out to improve vaccine safety (Keiser et al. 2010, 2011; Zollinger et al. 2010; Pinto et al. 2011). High expression of OpcA was stabilised and expression of both fHbp and NadA increased (Zollinger et al. 2010). A second PorA gene, of a different subtype than that of the native porA, was also inserted (Zollinger et al. 2010). The final vaccine composition was based on NOMVs from three vaccine strains as depicted in Table 1. Preliminary studies with combined NOMVs from the three strains confirmed the capacity of the vaccine to induce a broad-based SBA response against serogroup B strains as well as serogroups A, C, Y, W135 and X (Zollinger et al. 2010; Pinto et al. 2011).

Table 1. Profile of WRAIR NOMV vaccine strains (Zollinger et al. 2010; Pinto et al. 2011)
StrainParent strain designationGenetic modificationClonal complexPorA subtypesOver expressed antigensStabilised antigens
OpcALOS
A44/76ΔsiaA, ΔLpxL1, ΔlgtAST32P1.7,16 and P1.7-1,1NadA allele 3YesL8-3
B8570ΔsiaA, ΔLpxL1ST32P1.19,15 and P1.22,14Fhbp variant 1YesL8-5
CB16/B6ΔsiaA, ΔLpxL1, ΔlgtAST11P1.5,2 and P1.22-1,4Fhbp variant 2NoL8-2

Vaccine preparations of the two of the strains (strains A and B) have undergone Phase I trials (Keiser et al. 2010, 2011) as a three dose schedule. Both preparations were deemed safe and immunogenic showing a good cross-reactive antibody response. The main antigens contributing to the SBA activity were shown by using depletion assays (Zollinger et al. 2009) to be for both strains A and B to be OpcA, LOS, whilst for strain B, fhbp also contributed.

Bivalent fHbp (rLP2086) vaccine

An investigational vaccine containing a single subvariant from each subfamily, A (fHbp variants 2/3) and B (fHbp variant 1), a bivalent formulation (LP2086), is being developed by Pfizer (Fletcher et al. 2004). A Phase 1 study of three doses of the rLP2086 vaccine administered in a 0-, 1- and 6-month schedules demonstrated that 22–100% of 18- to 25-year-olds elicited SBA responses, depending on the dose formulation and target strain (Nissen et al. 2008). A further Phase 1 study of a 0-, 1- and 6-month schedules of different formulations of rLP2086 was performed in 127 children, aged 8–14 years (Richmond et al. 2008); again, a broad SBA response was measured against all fHbp variant groups. Most subjects reported only mild or moderate self-limiting adverse events. A Phase I/II safety and immunogenicity study of a 120 μg dose of rLP2086 was then carried out in healthy adults aged 18–40 years under a three dose schedule, demonstrating 94% of subjects having putatively protective SBA titres against strains with either homologous or heterologous fHBP variants. Local reactions were generally mild or moderate; though, 3 and 4 cases of severe induration and severe erythema, respectively, were reported (Marshall et al. 2011). A Phase I study of rLP2086 in toddlers aged 18–36 months under a 0-, 1- and 6-month dose schedule has been completed, but detailed results are not published to date (Arora 2009).

The specificity and vaccine potential of rLP2086 was demonstrated to be dependent on the level of fHbp surface expression as evaluated by flow cytometry in 100 invasive serogroup B invasive disease isolates (Jiang et al. 2010). Encouragingly, the rabbit sera, raised against bivalent fHbp, killed 87 of 100 invasive serogroup B isolates that expressed fHbp, whilst human sera killed 36 of 45 isolates (Jiang et al. 2010). The surface expression assay for fHbp has now been refined as the meningococcal antigen surface expression (MeASurE) assay, which utilises a unique monoclonal antibody that binds to all fHbp variants for (McNeil et al. 2011).

Engineered multivalent PorA vaccine

A recombinant hexavalent PorA OMV vaccine was produced by the National Institute of Public Health and Environment (RIVM) (Tommassen et al. 1990; van der Ley et al. 1995). Following a Phase I study in healthy adults in Utrecht in 1993 (Peeters et al. 1996), HexaMen was trialled in a Phase II study in UK infants in a 2-, 3- 4- and 12- to 18-month schedule in 1995 (Cartwright et al. 1999). The vaccine was well tolerated, and three doses induced putatively protective immune responses to two of six meningococcal strains expressing PorA OMPs in the vaccine whilst after a fourth dose, more than 77% of subjects demonstrated ≥4-fold rises in SBA activity to all of the strains tested. Hexamen was also studied in Phase II in toddlers and 7- to 8-year-old school children (de Kleijn et al. 2000) where the vaccine was well tolerated and ≥4-fold rises in SBA were demonstrated in 28–98% of toddlers and 16–100% of school children. The RIVM have now added a third OMV to their vaccine to produce NonaMen (van den Dobbelsteen et al. 2007), and the PorAs contained within are listed in Table 2. NonaMen has progressed through a Phase I trial in Spain in 2007 and has a potential coverage of 80% based upon European PorA subtype data from 1999 to 2004 (Trotter & Ramsay 2007).

Table 2. PorA proteins contained with NonaMen (van den Dobbelsteen et al. 2007)
VesiclePorA subtype
1P1.7,16
P1.5-1,2-2
P1.19,15-1
2P1.5-2,10
P1.12-1,13
P1.7-2,4
3P1.22,14
P1.7-1,1
P1.18-1,3,6

Recombinant proteins combined with outer membrane vesicles

Reverse vaccinology enabled the design of vaccines starting from the prediction of antigens in silico. This is in contrast to conventional approaches, which relied on biochemical, immunological and microbiological methods (Rappuoli 2000). Reverse vaccinology was first successfully used by Novartis Vaccines to develop a novel serogroup B vaccine using the genome from strain MC58 (Pizza et al. 2000). Initial screening identified 570 open reading frames that encoded surface-expressed proteins; of these, 350 were expressed and purified for pre-clinical immunisation studies. Twenty-eight novel proteins were identified which induced SBA activity and were also conserved amongst serogroup B strains. Five of the most promising antigens, fHbp, NadA, NHBA, GNA2091 and GNA1030, were selected to be included in an investigational vaccine. To investigate ways of increasing immunogenicity and coverage, this preparation was also formulated with OMVs derived from the New Zealand strain NZ 98/254, (B:4:P1.7-2.4) and progressed satisfactorily through Phase I in healthy adults (Toneatto et al. 2011). The final vaccine formulation, which has progressed through clinical trials and was submitted in December 2010 to the European Medicines Agency for a marketing authorisation, has been assigned the trade name of Bexsero. Prior to submission, this vaccine was also termed four component meningococcal serogroup B vaccine (4CMenB).

Each dose (0.5 ml) of Bexsero contains 50 μg of each of NadA (subvariant 3.1), GNA2091-fHbp fusion protein (fHbp subvariant 1.1) and NHBA-GNA1030 fusion protein (NHBA subvariant 1.2) adsorbed onto 1.5 mg aluminium hydroxide, 3.25 mg NaCl and 10 mm histidine as well as 25 μg detoxified OMVs from N. meningitidis strain NZ98/254 (B:4:P1.7-2.4). The inclusion of OMVs from NZ98/254 was to provide additional protection against the cc41/44 strains, particularly strains with PorA VR2 P1.4. OMVs also provide increased immunogenicity when combined with other antigens possibly because of the OMV having an adjuvant property or by induction of antibodies that act in a synergistic manner (Giuliani et al. 2010).

To assess the potential coverage of Bexsero, with regard to antigen expression and the genetic variation, it is not possible to test large panels of strains in the SBA assay. Consequently, the Meningococcal Antigen Typing System (MATS) has been developed to assess the potential coverage of Bexsero against individual meningococcal strains (Donnelly et al. 2010). MATS is performed on a detergent extract of each of the recombinant proteins and takes into account variations in both the immunological recognition of each antigen and the level of expression. The relative potency (RP) of each unknown isolate is calculated against a reference serogroup B strain with high and consistent expression of protein. The PorA P1.4 component is assessed by conventional subtyping methodologies. A relationship between MATS RP and SBA titre was determined using pooled post-immunisation sera from 13-month-old toddlers who had received four Bexsero doses. This relationship enabled the establishment of a positive bactericidal threshold (PBT) for each antigen which can be used to predict if isolates will be covered by Bexsero. Validation of the MATS revealed that isolates exceeding the PBT for one antigen had a ±80% chance of being killed by sera in the SBA assay. This probability increased to 96% for isolates with ≥2 antigens with RP above the PBT. Invasive serogroup B isolates were collected by national reference laboratories of England and Wales, France, Germany, Norway and Italy from epidemiological year 2007/8; totally, 1052 isolates were studies by MATS and PorA subtyping (Donnelly et al. 2011). Valid results were obtained for 96% of these isolates, and potential coverage was estimated at 78% (95% CI 66–91%). Of interest 64% of the covered isolates were positive for more than one antigen.

To measure the SBA activity of the main components of Bexsero, target strains expressing each of the antigens but not expressing sufficient quantities of the other components were selected. Strains 44/76-SL, 5/99, NZ 98/254 and M10713 were used to assess SBA responses against fHbp, NadA, PorA and NHBA, respectively. These indicator strains have been used in Phase II, II/III and III trials that have been recently reviewed (Bai et al. 2011).

Bexsero has been shown to induce immune responses in clinical trials reported across all age groups (Findlow et al. 2010; Snape et al. 2010; Gossger et al. 2012; Santolaya et al. 2012). Latest data from infant trials showed that the large majority of infants displayed a robust immune response against the serogroup B reference strains and that the vaccine has acceptable reactogenicity rates when given in conjunction with other infant immunisations (Gossger et al. 2012). Ongoing work will produce further supporting evidence and address antibody persistence, if acquisition of carriage affected young adults and if introduced on a wide scale, the ability to induce herd protection.

The future

Meningococcal conjugate vaccines will eventually replace the plain polysaccharide vaccines as conjugates have many clear advantages in being immunogenic in infants, able to prevent acquisition of carriage and booster dose responses. Over the forthcoming years, if successfully introduced, a better understanding of the effectiveness of the new vaccines for the prevention of serogroup B disease should be gained together with knowledge of any impact on nasopharyngeal carriage. Neither polysaccharide nor conjugate vaccines currently exist for serogroup X; thus, research should be targeted against this serogroup. Subcapsular vaccines for serogroup B have the potential to impact against other serogroups; thus, this is may also be an important approach.

Ancillary