MENINGOCOCCAL DISEASE AND VACCINATION

Five serogroups of meningococci, A, B, C, Y, and W135, defined by the biochemistry of their polysaccharide capsule, are responsible for almost all meningococcal disease, although the overall proportions of cases caused by each serogroup vary widely around the globe. Serogroup A meningococci are the cause of cyclic epidemic meningitis in Africa² and Asia,³ and occasional outbreaks have been associated with population movements and overcrowding in other regions over the past half century.⁴ In industrialised nations serogroup B meningococci cause 30–70% of cases of sporadic meningococcal disease^5–7 and have been responsible for pockets of persistently increased rates of disease.^8–10 Serogroup C meningococci are particularly associated with small outbreaks of disease among teenagers and young adults and sporadic disease in individuals of other ages.⁶ Serogroup Y disease is uncommon in the United Kingdom, but accounts for up to 30% of cases in the United States⁶; rates of Y disease may also be on the increase in parts of Canada.⁷ Occasional sporadic disease caused by W135 meningococci had been largely ignored until a recent large outbreak among pilgrims to the Hajj in 2000.¹¹ Although the epidemiological characteristics of disease caused by each serogroup are intriguingly different, the clinical features of invasive disease are mostly indistinguishable.

Although it is likely that antibiotics and specialist intensive care can significantly reduce the mortality from meningococcal septicaemia,¹² it is clear that timely, universal delivery of such care cannot be guaranteed for all children, particularly in the context of this fulminant disease that kills in hours. Widespread control of meningococcal disease is an important goal in paediatric practice and immunisation seems the obvious solution.

MenC, the protein–polysaccharide conjugate vaccine that was introduced into the UK primary infant schedule in November 1999, has already had a major impact on disease caused by serogroup C meningococci in childhood in this country.¹³ The vaccine has also been licensed in other countries around Europe,¹⁴ and in Canada.⁷ Furthermore, development of combination A, C, Y, and W135 protein–polysaccharide conjugate vaccines is underway, and there is an expectation that these vaccines will be available within a few years, and will provide effective protection against these serogroups from infancy; combination with pneumococcal protein–polysaccharide conjugate vaccines is planned.¹⁵ There is also optimism, boosted by the announcement of significant funding from the Bill and Melinda Gates Foundation in July 2001, that further development of a serogroup A/C conjugate vaccine under the direction of the Global Alliance for Vaccines and Immunization (GAVI) will halt the cycle of epidemic disease in Africa.¹⁶

Optimism that global control of disease caused by A, C, Y, and W135 meningococci is within reach must be tempered by the absence of an imminent solution to the problem of serogroup B disease in children. The highest attack rate of meningococcal disease is in children under 5 years of age; at this age, 50% of disease is caused by serogroup B meningococci in the USA (1992–1996),⁶ 39% in Canada (1985–2000),⁷ and more than 65% in the UK (1999–2000).⁵

Vaccines against the non-B serogroups (A, Y, and W135) of meningococci that contain the capsular polysaccharide of the organism conjugated to a protein carrier are currently under development, using the same technology as was used in the development of Haemophilus influenzae type b (Hib) vaccines.¹⁷ MenC, the new serogroup C protein–polysaccharide conjugate vaccine, is already in routine use in parts of Europe and Canada. MenC is highly immunogenic from infancy,¹⁸ and it is expected that similar immunogenicity and protection will be afforded by A, Y, and W135 conjugate vaccines.

However, the polysaccharide capsule of serogroup B meningococci is a homopolymer of sialic acid, which is chemically identical to polysaccharides found in human tissues during development.¹⁹ Hence, the B capsule is seen by the immune system as a self antigen and is thus poorly immunogenic, even after conjugation to a protein carrier.²⁰ Therefore, the “simple” strategy of polysaccharide–protein conjugation (described above for protection against non-serogroup B meningococci) cannot easily be translated to serogroup B meningococci and various alternative strategies are in active development.

STRATEGIES FOR SEROGROUP B MENINGOCOCCAL VACCINE DEVELOPMENT

A number of different strategies for prevention of serogroup B disease are considered below. Evaluation of the potential utility of each vaccine candidate is difficult because of the lack of an accepted laboratory surrogate of protection. There is no defined level of circulating antibody to serogroup B meningococci that is known to relate to protection against the disease, but some indirect evidence suggests that a certain titre of bactericidal antibody (antibody that can kill meningococci in the laboratory in the presence of complement) might be correlated with protection at the population level.²¹ However, the lack of an established laboratory correlate of protection has hampered the development of serogroup B vaccines.

Polysaccharide vaccines

While the lack of immunogenicity of the serogroup B sialic acid capsule is a major problem in development of a capsule based B vaccine, research efforts in this area continue because of the attractiveness of a vaccine antigen that is, by definition, shared across this group of meningococci. Chemical modification of the polysaccharide (N-propionylation), in an attempt to induce immunogenic epitopes, has resulted in development of a protein–polysaccharide conjugate vaccine that elicits functional (bactericidal) antibody in both mice and non-human primates.²² Some of the antibodies elicited have activity against polysialic acid and therefore have the potential to be autoreactive in humans, although no deleterious effects have been noted in early human trials (personal communication, P Fusco, Baxter, 2001). However, other antibodies that arise after immunisation with a conjugate N-propionylated serogroup B polysaccharide vaccine do not cross react with human tissues and might therefore be used in the development of molecular mimetics of the non-autoreactive epitopes, and result in the production of a safe serogroup B vaccine.²³

Outer membrane protein vaccines

The difficulties in production of an effective serogroup B polysaccharide vaccine have led to investigation of other bacterial surface structures as vaccine constituents. Beneath the polysaccharide capsule, meningococci are surrounded by an outer membrane, rich in protein, lipoprotein, and lipopolysaccharide.

Since the 1970s the outer membrane proteins (OMPs) of serogroup B N meningitidis have been extensively studied as potential vaccine constituents. However, the exposed regions of these structures are almost all antigenically highly variable among different serogroup B isolates, such that a vaccine containing a protein from a single organism is unlikely to provide cross protection to all other serogroup B bacteria. Furthermore, the antigenic regions of many of these protein structures appear to evolve rapidly within bacterial populations, perhaps as a result of pressure from acquired immunity in human populations so that a vaccine may become ineffective against the evolved strain. Despite these problems, several vaccine candidates have reached trials and one is in routine use in Central and South America,²⁴ though none of these vaccines has proven effectiveness in young children, those most affected by serogroup B disease.

Early OMP vaccines consisted of insoluble aggregates of outer membrane proteins and, although immunogenic in animals, were poorly immunogenic in humans.²⁵ Another OMP vaccine containing purified OMPs non-covalently complexed to meningococcal C polysaccharide was tested in a trial in Chile and provided some protection in older children and young adults (70%), but was poorly protective in children less than 5 years of age.²⁶ In order to improve immunogenicity by presenting the proteins in a more natural conformation, several developers have produced OMP vaccines in solution that contain OMPs in spheres of bacterial outer membrane, known as outer membrane vesicles (OMVs).^24,27–30 These OMV vaccines have now been evaluated in large scale trials (see table 1), and shown to be immunogenic and partially protective in older children.^24,26,29 Unfortunately, again, there is no evidence of protection against serogroup B disease in any of these studies in children under 4 years of age.^24,26,31,32 Furthermore, bactericidal antibody appears to be largely confined to the serogroup B meningococcal “type” included in the vaccine.^33,34

The major protein immunogen in these OMV vaccines is a porin protein named PorA, which is antigenically variable. In an attempt to overcome this variability and provide cross strain protection, a hexavalent OMV vaccine was produced containing six different PorA proteins (see table 1). Phase II trials in various age groups have shown variable immunogenicity to each PorA type.^30,35 Unfortunately, minor mutations in the genes encoding this protein are common and result in evasion of complement mediated killing of the organism, reducing the likelihood of success with this approach.³⁶

The observation that some protection and induction of bactericidal antibody directed against the vaccine strain is induced with monovalent OMV vaccines suggests that these vaccines may be of use in outbreaks of serogroup B disease caused by a single “type” of bacteria. In New Zealand, rates of meningococcal disease 4–20 times higher than other industrialised countries have been observed for the past decade,⁸ with a majority of cases caused by closely related serogroup B meningococci. In July 2001, the Ministry of Health in New Zealand announced plans for further development of a monovalent OMV vaccine with Chiron Corporation in conjunction with National Institute for Public Health (NIPH), Norway in the hope of halting serogroup B disease.³⁷

Other outer membrane protein vaccine candidates

The antigenic variability of the major OMP, PorA, the disappointing efficacy of OMV vaccines, and concerns over safety and immunogenicity of B polysaccharide vaccines have directed research to other surface exposed OMPs. Prior to evalauation in human trials, the ideal candidate(s) should be immunogenic (induction of serum bactericidal antibody following immunisation), protective in animal models, and have limited antigenic variability. Table 2 lists a number of possible vaccine candidates, but there is likely to be a problem with antigenic variability for a number of these. Nevertheless, some of these vaccine candidates have shown promise in preclinical studies. Of these, only TbpB has been evaluated in preliminary human trials, but immunogenicity was disappointing with the vaccination formulation used.³⁸

The recent availability of two meningococcal genome sequences^39,40 has provided a new resource for the search for vaccine candidates.⁴¹ Using computer software to scan the meningococcal genome, Pizza and colleagues⁴¹ have identified 570 open reading frames (sequences of DNA), which encode proteins that are likely to be surface exposed or secreted by meningococci. Following cloning into an E coli expression system, animal immunogenicity studies and selection of the most promising candidates, seven proteins were chosen for further study. These proteins were all shown to be surface exposed; five of these novel proteins were found to be widely expressed among pathogenic Neisseria (with >99% sequence homology), and two were able to induce bactericidal antibody in mice (GNA33 and GNA1946). Several of these proteins have been used in animal studies and found to be less immunogenic than OMVs,⁴² though combinations of proteins appear to be more immunogenic than single proteins.⁴³ Further development of vaccines using this strategy is anticipated.⁴⁴

A similar approach using genome mining has been described by Poolman and Berthet^44a using sequence from serogroup B strain ATCC13090 to produce recombinant vaccine candidates in E coli. Preclinical evaluation of these proteins led to the identification of 10 promising vaccine candidates. Upregulated expression of some of these OMPs has been achieved by either gene delivery (replacement of a dispensable gene with the vaccine candidate by homologous recombination) or promoter replacement (delivery of a strong promoter upstream of the gene of interest), prior to production of candiate OMV vaccines containing the upregulated OMPs.^44a

In another molecular approach, genetic manipulation of meningococci was recently used to identify genes that are involved in pathogenesis of the organism in an animal model.⁴⁵ Some of the proteins encoded by these genes are likely bacterial surface structures and could be evaluated for vaccine potential.

Whole cell and exotoxin vaccines

One approach to the problem of heterogeneity of surface structures is to develop vaccines that contain a wide range of surface antigens rather than a single cross protective antigen. Indeed, the history of meningococcal vaccine development began almost a century ago with trials of killed whole cell vaccines, but it seems that there was limited efficacy and a high incidence of adverse reactions to these vaccines.⁴⁶ Trials of exotoxin vaccines in the 1930s also showed some promise but were not pursued.⁴⁶ More recently, nasal immunisation with a live commensal organism such as N lactamica⁴⁷ or with a live attenuated N meningitidis vaccine⁴⁸ have been considered. There is currently no evidence that these approaches will be successful in humans and it is unlikely that live attenuated N meningitidis vaccines would ever receive regulatory authority approval, but the concept of presentation of multiple antigens in a single vaccine is likely to be pursued.

Lipopolysaccharide

Lipopolysaccharide (LPS; endotoxin), a major component of the meningococcal outer membrane, is believed to be the bacterial factor responsible for the dramatic clinical presentation of meningococcal disease through stimulation of widespread inflammatory cytokine production. Like the polysaccharide capsule, LPS is highly conserved among invasive isolates, making this an attractive vaccine candidate for serogroup B meningococci and more broadly across the serogroups. Bactericidal anti-LPS antibodies are present in convalescent human serum⁴⁹ and after immunisation with an LPS containing OMV vaccine in non-human primates.⁵⁰ Moreover, a single murine antibody directed at the inner core of LPS appears to have bactericidal activity against 70% of meningococci tested (including isolates from serogroup A and B),⁵¹ and also enhances opsonophagocytic activity of human phagocytes.⁵² A protein–lipopolysaccharide conjugate inner core LPS vaccine has been produced that induces bactericidal antibody in animals and has activity against an array of meningococcal isolates.⁵³

CONCLUSION

In the next few years, it is likely that effective multivalent protein–polysaccharide conjugate meningococcal vaccines will become available for the control of disease caused by serogroup A, C, Y, and W135 meningococci, either alone or in combination with pneumococcal conjugate vaccines.

Outer membrane vesicle (OMV) vaccines for serogroup B meningococci are already in use for endemic disease in parts of Central and South America, but these vaccines appear to offer no protection for young children and provide limited cross protection to non-vaccine serogroup B meningococci. Further large trials of monovalent OMV vaccines are expected imminently in New Zealand and offer the possibility of control of hyperendemic clonal serogroup B disease.

Perhaps inspired by the availability of genomic data, efforts to produce an effective vaccine for protection against serogroup B meningococci are today more intense than ever before, and several novel candidates are already in preclinical or early clinical trials. However, serogroup B meningococci must not be underestimated. The success of these bacteria as colonisers of the human nasopharynx probably lies in their ability to vary their immunogenic surface structures and evade immunologic attack, whether natural or vaccine induced. Vaccines that are directed at a single variable surface structure are unlikely to be successful, but this problem might be overcome by use of multicomponent protein vaccines. Highly conserved structures such as the capsular polysaccharide (despite the developmental problems noted above), inner core lipopolysaccharide, or conserved proteins that have limited variability are the most attractive vaccine candidates, and it is likely that several of these candidates will be advancing through clinical trials within the next five years.

The meningococcus has not yet been tamed but the momentum in vaccine development today is far greater than ever before, with involvement from many research groups, government institutions, and at least five vaccine manufacturers. There is reason for optimism that control of endemic meningococcal disease through infant immunisation against all five serogroups may be within reach.