key: cord-302082-aaokc182
authors: Stanberry, Lawrence R.; Strugnell, Richard
title: Vaccines of the future
date: 2011-08-31
journal: Perspectives in Vaccinology
DOI: 10.1016/j.pervac.2011.05.006
sha: 
doc_id: 302082
cord_uid: aaokc182

nan

The advances made in vaccine technology since Edward Jenner vaccinated the young James Phipps against smallpox have had a spectacular impact on human health over the last two centuries (see Chapter 1 e Vaccine evolution). Vaccines have been fundamental in the control and elimination of many debilitating and lethal diseases, and more diseases are currently targeted for eradication by vaccination. Recent major breakthroughs in immunology, molecular biology, genomics, proteomics, biochemistry and computing sciences have driven vaccine technology forward, and will continue to do so.

Many challenges remain, however, including persistent or latent infections, pathogens with complex life cycles, antigenic drift and shift in pathogens subject to selective pressures, challenging populations and emerging infections. To address these challenges researchers are exploring many avenues: novel adjuvants are being developed that enhance the immune response elicited by a vaccine while maintaining high levels of tolerability; methods of protective antigen identification are iterated with every success; vaccine storage and transport systems are improving (including optimising the cold chain and developing temperature-stable vaccines); and new and potentially more convenient methods of vaccine administration are being pursued.

High priority targets include life-threatening diseases, such as malaria, tuberculosis (TB) and human immunodeficiency virus (HIV), as well as problematic infections caused by ubiquitous agents, such as respiratory syncytial virus (RSV), cytomegalovirus (CMV) and Staphylococcus aureus. Non-traditional vaccines are also likely to become available for the management of addiction, and the prevention, treatment and cure of malignancies.

This chapter is not meant as a compendium of all new-generation vaccines, but rather as an outline of the modern principles that will likely facilitate the development of future vaccines. As shown in Figure 6 .1, there are several key elements that are likely to be 152 UNDERSTANDING MODERN VACCINES the foundation for the development of future vaccines. This chapter will illustrate these elements and provide examples that show promise.

Since the first use of an adjuvant in a human vaccine over 80 years ago, adjuvant technology has improved significantly with respect to improving vaccine immunogenicity and efficacy. Over The advances in adjuvant design have been driven by parallel advances in vaccine technology as many modern vaccines consist of highly purified antigens e with low non-specific reactogenicity which require combination with adjuvants to enhance the immune response. Future developments in adjuvant technology are expected to provide stronger immune priming, enhance immune responses in specific populations, and lead to antigen sparing. Adjuvants to date have demonstrated an ability 

New adjuvants must aim to drive the immune response that is associated with lifelong protection. New adjuvants and adjuvant combinations will play many roles in future vaccines as illustrated in Figure 6 .2. Adjuvants will need to be individually selected for specific vaccine targets in order to achieve the desired goal (ie enhanced immunogenicity, induction of specific immune profile etc). To deliver this aim, some adjuvants will be mixed with free antigens, while others will need to be covalently linked to the antigenic moiety as part of a complex molecule.

Some examples of new adjuvants that have been evaluated in humans or that are in clinical trials are listed in Table 6 .1 (also see Chapter 4 e Vaccine adjuvants).

Modern approaches to antigen design tend to eschew classical trial and error techniques in favour of identifying the type of pathogenic structures (ie antigens) that are most likely to be important immunogens based on their structural signature or physical location within the pathogen (Table 6 .2) (see Chapter 3 e Vaccine antigens).

The T or B cell immune responses to an antigen are targeted to precise regions of the antigen (ie epitopes e either linear or three-dimensional conformational structures; in the case of protein antigens these are specific peptide epitopes). Historically, simple, linear, synthetic peptide epitope vaccines have been poorly immunogenic because they lack a specific conformation and are easily degraded by a variety of extracellular and cell-surface proteases that serve to limit epitope presentation to T cells and/or result in destruction of the B-cell epitope. Peptide vaccines need to Union. Every effort has been made to verify the information in the above table. The information included is not meant to be exhaustive but is intended to provide an overview of the subject matter. Subunit and individual epitope vaccines need to be optimised to ensure adequate immunogenicity. Novel strategies are being developed and exploited in order to identify antigens recognised by T and B cells, thus facilitating a more knowledge-based vaccine design. One of the most common ways to identify these antigens is to measure cellular proliferation (T or B cells) upon in vitro stimulation with antigen. High-throughput screening assays of candidate synthetic peptides that drive cellular proliferation help speed the rate of antigen discovery.

Reverse vaccinology combines knowledge of the pathogen's genome sequence with known protein sequences via computer analysis, to predict protein expression and post-translational modifications and identify likely vaccine candidates (see Chapter 3 e Vaccine antigens; Figure 3 .5).

The development of epitope-based vaccines is one example of reverse vaccinology where computer software combines prediction algorithms to suggest sequences similar to those for pathogenic components. Epitope mapping, combined with the creation of more stable poly-epitope vaccines, may lead to the successful translation of this technology into products.

MHC molecules exhibit widely varying binding specificities; a vaccine expressing a single peptide antigen would therefore only target a few MHC molecules and thus only be recognised by the T cells of individuals carrying a specific MHC phenotype.

Poly-epitope technology could be used to generate a synthetic protein carrying antigenic epitopes from multiple strains or pathogens. This would overcome the MHC restriction and afford protection in individuals carrying different MHC types.

The screening of pathogen peptide libraries is another example of new approaches to antigen discovery. Screening methods are used to identify antigens that can stimulate CD4 þ or CD8 þ T cells, or which bind to antibodies from humans known to have been infected with the relevant pathogen. Where peptide screening uses antibodies, an additional consideration is the synthesis of antigens that contain the tertiary (folding/three-dimensional) structure of the native immunogen, since vaccine efficacy can be impacted by infidelities in the structure of the final product. Incorrect protein folding may result in a less immunogenic antigen or an antigen that induces an immune response that differs from that of the native immunogen. The mimicking of the three-dimensional structure of the native immunogen is important during the synthesis of antigens that are being used to target B-cell responses. Conversely, the requirement for folding is reduced for T cells since T cells bind only processed peptides, from degraded proteins. Likewise, DNA expression libraries using the pathogen genomic DNA have been screened using animal model systems to identify genes encoding proteins that afford protection against infection or disease caused by the pathogen.

One example is Genocea's vaccine development programmes that are built around a broad platform for the rapid discovery of T-cell antigens. The process is explained in Figure 6 .3. T-cell antigens, specifically antigens that stimulate CD4 þ and CD8 þ T cells, are critical to generating disease-specific cellular immune responses and long-term T-cell memory.

Stability of the final product is another important consideration. Adverse environmental conditions can result in degradation of the vaccine, rendering it non-immunogenic. In order to maintain product integrity many vaccines (particularly live vaccines) must be stored at cold temperatures ( 4 C). The maintenance of the vaccine at this temperature from production site to distribution site, and medical office or clinic, is referred to as the 'cold chain'. Maintaining the cold

chain is much less of a challenge in resource-rich countries, but can be a major barrier to vaccine implementation in resource-limited areas. Ongoing research designed to increase our understanding of vaccine degradation may address the problems associated with cold chain management and lead to the development of thermostable vaccines.

The level of antigen presentation which occurs with some current vaccines may sometimes be insufficient to drive long-lasting immune responses of high quality (see Chapter 3 e Vaccine antigens). This may be due to inadequate exposure of the antigen to immature antigen-presenting cells (APCs) rapid or subimmunogenic degradation or sequestration of antigens, or lack of immunogenicity due to the physical presentation of the antigen. The discovery and 

Modifying vaccine formulations to increase tolerance to temperature fluctuations is likely to increase the shelf-life of the product and reduce transport and wastage issues.

UNDERSTANDING MODERN VACCINES refinement of new and varied options for antigen presentation is expected to allow the design of vaccines to produce specific immune profiles. Some of these technologies have been shown to facilitate oral delivery to target mucosal immune responses and also trigger both innate and adaptive immune systems, including T-and B-cell effector and memory responses.

Candidate viral vector vaccines utilise a non-pathogenic virus to carry and subsequently induce expression of genes that produce immunogenic foreign proteins at high levels in the host. These are taken up by immature APCs, and have been shown to lead to a robust, long-lasting immune response to the target antigen ( Figure 6 .4). Viral vector vaccines, eg recombinant poxvirus vaccines, can be administered mucosally to stimulate mucosal immune responses. The attenuated modified vaccinia virus Ankara (rMVA) vectors are showing promise as mucosal delivery vectors.

Pre-existing immunity to the viral vaccine vector is an impediment to successful use of this approach. As ways to avoid anti-vector immunity, viruses can be attenuated or inactivated, by deleting or replacing pathogenic genes. Figure 6 .4 demonstrates how viral vaccine vectors are made. DNA expressing an immunogenic transgene (the vaccine antigen) is inserted into the viral vector genome for expression following administration into the recipient; expression of the vaccine antigen can be boosted by using a variety of DNA promoters. If the viral vector is no longer able to grow and replicate, the virus is grown using a cell line (a so-called complementing cell line) that has been engineered to produce the missing viral product.

Often, viral genes are removed in an effort to reduce or eliminate the pathogenicity of the vector and in some cases viral genes are removed to make the vector itself less immunogenic; an anti-vector immune response would greatly reduce the ability of the vector to induce an antigen-specific response. Examples of viral vector candidate vaccines in clinical development are listed in Table 6 .3.

Non-pathogenic bacterial vectors have many features that make them an attractive vaccine platform. Bacterial vectors can be engineered for maximum safety (eg deletion of two or more genes from the same metabolic pathway), and to express large numbers of foreign antigens ( Figure 6 .5).

Two key issues affecting bacterial vaccine vectors are: a) to decide whether the optimal platform should be a bacterial vaccine in its Figure 6 .4 Viral vectors for vaccines. Viral vector vaccines exploit the natural ability of viruses to infect or otherwise enter (in the case of disabled viral vectors) host cells, and then deliver pathogen-specific antigens. Antigen-encoding genes are isolated from the pathogen and inserted into the viral vector genome. The viral vector can then be used as a factory for production of large quantities of pathogen antigen in vivo, following introduction of the vector into the vaccine recipient, with the pathogen antigen then expressed on the surface of the infected/transduced host cells or exported out of the producer cell.

MHC, major histocompatibility complex. own right or a bacterial vector system to deliver exogenous antigens; and b) to determine whether re-administration of the vector, either with the same or different target antigens, will fail because of the immune response to the bacterial vector vaccine at the time of its initial administration.

Initial assessments of the feasibility of using attenuated bacterial vectors for the delivery of foreign antigens have focused on Salmonella species. Bacterial vaccine vectors for humans, however, have been disappointing so far. It may be necessary to develop unique bacterial vaccine vectors for delivering exogenous antigens, in which case the vectors can be modified to allow for re-use. For example, if immunity against the vector, which is a major impediment to vaccine re-use, is determined by antibodies against the surface structures of the bacterium (such as lipopolysaccharide [LPS]), the dedicated vaccine vector could be developed to lack expression of LPS or to express truncated/ different forms of LPS to the target, thereby avoiding priming of the immune response and allowing for re-use of the vector and/or vaccine. Some potential options for live, attenuated bacterial vectors are shown in Table 6 .4. 

DNA vaccines are the result of the discovery in the early 1990s that the gene, rather than the encoded protein, if delivered in an 'expressible' form, could induce an immune response (see Chapter 1 e Vaccine evolution). The principle behind DNA vaccines is that the antigenic molecule is produced within the host from the DNA or RNA that is injected, in contrast to more traditional vaccination where the antigen is supplied in the vaccine formulation. The gene(s) for target antigen(s) is/are usually encoded in a circular plasmid expression vector under the control of promoter sequences that direct gene expression in mammalian cells, which is achieved after injection into mammals. 

The DNA vaccine process can circumvent some of the major issues resulting from recombinant protein administration. The construction and production of the plasmids carrying the gene of interest together with the promoter sequences is relatively simple; antigens expressed from plasmids retain their native conformation, the gene can be readily modified to produce tailored antigens, and bacterial plasmid DNA is intrinsically immunogenic (subsequently shown to result from the pathogen-associated molecular patterns [PAMPs] it carries). Additional desirable features include the ability to engineer and deliver genetic adjuvants in tandem or parallel with the antigen, the potential to deliver multiple antigen genes in one construct or within other constructs that encode adjuvanting protein(s), and the ability to induce both cellular and humoral immune responses.

Despite promising data in pre-clinical testing, DNA vaccine candidates have shown only limited success in clinical settings so far. One of the current drawbacks of DNA vaccines is the inefficiency of conventional delivery methods for the plasmid DNA; however, 

UNDERSTANDING MODERN VACCINES emerging proprietary particle-mediated delivery technology or electroporation technology seeks to improve this situation. With the electroporation method, brief electrical pulses are applied at the site of immunisation which causes a transient disruption of cell membranes. This results in an enhancement in uptake of the DNA vaccine between 10e100-fold. Examples of DNA candidate vaccines in clinical development are presented in Table 6 .5.

Dendritic cell (DC) vaccines typically use monocytes harvested from the blood (in most cases from the individual who will receive the vaccine) to produce immature DCs in vitro. The monocytes are antigen-loaded and treated to induce their maturation into APCs and infused back into the patient. The first Food and Drug Administration (FDA)-approved DC vaccine, designed for the treatment of prostate cancer, was licensed in 2010 (Sipuleucel-T); examples of other targets for DC vaccine therapy are presented in Table 6 .6. DC vaccines offer an individualised approach to therapeutic vaccine development, but represent a specialised method of vaccination that is currently limited to aggressive cancers, and the treatment of serious, intractable infections.

A comparison between the strengths and weaknesses of selected new vaccine platforms is presented in Table 6 .7.

Developing administration techniques that place the vaccine directly at the site(s) where pathogens are most likely to initiate an infection (eg mucosal or respiratory sites) is likely to improve vaccine efficacy and safety.

Traditional methods of vaccine administration can potentially pose a number of limitations with respect to reactogenicity, immunogenicity, convenience, efficacy, safety and cost-effectiveness. The information included is not meant to be exhaustive but is intended to provide an overview of the subject matter. Ongoing research on alternative experimental administration strategies includes ballistic delivery to skin (the gene gun), the transdermal patch and other intradermal methods, plus sublingual, aerosol, rectal and vaginal mucosal vaccines. The main advantages of alternative delivery strategies are the potential to induce immune responses at the common portals of pathogen entry (eg oral polio vaccine replicating in the gut), potential convenience (eg ease of use of the transdermal patch), potential combination of vaccines to reduce or simplify the vaccination schedule, and reduction or elimination of administration via standard hypodermic needle injection. Despite the intuitive value of these approaches, few vaccines today are administered via non-IM routes. This is for several reasons including feasibility, lack of proven efficacy and limited safety data. Some problems have been observed with new routes of delivery, for example, after the 2000 launch of an inactivated intranasal influenza vaccine (a virosome formulation adjuvanted by heat labile enterotoxoid of Escherichia coli), post-licensure data indicated a significantly increased risk of Bell's palsy in vaccinees and forced its withdrawal from the market. This experience led to a higher level of caution in the development of intranasal vaccines. Today, the only example of a licensed vaccine against a latent infection is the zoster vaccine; the vaccine formulation is the high potency (about 15-fold) version of the live, attenuated varicella zoster virus (VZV) vaccine. This vaccine has been used to boost the anti-VZV cell-mediated immune response in older subjects and has been shown to reduce the overall incidence of zoster by 50% in subjects aged 60 years or older (Oxman et al., 2005) .

Future vaccines may control persistent infections either by preventing the initial infection or disease (ie prophylactic vaccines)

Aerosol delivery: 'Mass immunization of almost all susceptible children in a short period of time, has the potential of rapidly eliminating measles as a public health problem. Immunization by inhalation of aerosolised measles vaccine provides a procedure that could make such a mass programme possible, especially in parts of the world where measles continues to be a serious problem.' (Sabin et al., 1983) .

Administering the measles vaccine as an aerosol, either as nebulised vaccine or as An increased understanding of human immunology and of hostepathogen interactions should enable the identification of the type(s) of immunity required to effectively prevent or control persistent infections (see Chapter 2 e Vaccine immunology). Some examples of persistent infections are shown in Table 6 .8.

Mycobacterium tuberculosis can persist in a latent state within the human host for years without causing disease (latent TB). Protection against miliary (disseminated) TB in children is provided by the bacille CalmetteeGuérin (BCG) vaccine, developed through culture attenuation of Mycobacterium bovis early in the 20th century, which is routinely given in many countries. The vaccine, however, provides only modest and often temporary protection against pulmonary TB, and provides lower efficacy in resource-limited regions closer to the equator. In addition, vaccination with live, attenuated Mycobacterium bovis is a particular concern in HIV-positive individuals, especially those with advanced immune suppression; this population would particularly benefit from TB vaccination as TB is a leading cause of death worldwide for people with HIV/acquired immunodeficiency syndrome (AIDS). However, a recent Phase III trial demonstrated that protection against TB can be provided to individuals with HIV by using an inactivated whole-cell mycobacterial vaccine (von Reyn et al., 2010) . The current state of TB vaccine development has been summarised in reviews by Walker et al. (2010) and Lambert et al. (2009) and examples of vaccines in development are shown in Table 6 .9.

Cytomegalovirus, a herpes virus, establishes latent infection in cells in the bone marrow and peripheral blood. Primary infection during pregnancy is associated with congenital infection that 

frequently causes a well-characterised spectrum of abnormalities and disabilities, which may be severe or fatal. Reactivation in pregnancy is common, but is unlikely to cause severe congenital infection, although some manifestations, especially hearing loss, remain common. Reactivation of CMV is of special concern in immunocompromised individuals, where severe and fatal pulmonary, hepatic and central nervous system infections are common. Gastrointestinal disease and retinitis are common in association with HIV. A successful CMV vaccine has proved elusive for more than 30 years. Based upon the observation that antibodies to the CMV envelope glycoprotein B (gB) could Pass et al., 1999) . A recent Phase II clinical trial in CMV-seronegative women 1 year post-partum has shown the potential of gB/MF59 in decreasing incident cases of maternal and congenital CMV infection (Pass et al., 2009) . This is the first evidence that a CMV vaccine can protect against infection.

An alternative approach to the development of a CMV vaccine has been to utilise DNA vaccination to induce host responses to CMV gB and phosphoprotein 65 (pp65 is another viral target). Recent studies have shown that injection of combinations of plasmids, formulated with an adjuvant, can induce vaccine-specific immune responses, and can prime for effective memory responses.

The hallmark of herpes simplex virus types 1 and 2 (HSV-1 and HSV-2) is their ability to establish and maintain latent infection in to be exhaustive but is intended to provide an overview of the subject matter.

sensory ganglion neurons. Periodic reactivation of the latent infection results in recurrent infections. Both HSV-1 and HSV-2 can cause myriad diseases but the greatest public health problem is genital herpes. Genital HSV-2 infection increases the risk of HIV acquisition and transmission, and control of genital herpes has been predicted to significantly impact the HIV epidemic. Given the complex natural history of HSV infections, vaccines could have a variety of possible risks and benefits (Table 6 .10).

An effective HSV vaccine has been sought for more than 80 years. Recently, an HSV-2 glycoprotein D (gD2) candidate vaccine containing the AS04 adjuvant (see Chapter 4 e Vaccine adjuvants), was tested in three large, double-blind, Phase III controlled trials. The first two studies recruited volunteers with a partner with genital herpes disease and found the candidate vaccine was 73% effective against genital herpes disease in women seronegative for both HSV-1 and HSV-2 (Stanberry et al., 2002) . Trends towards protection against infection were also observed, but were not statistically significant. The candidate vaccine was not effective in HSV-1 seropositive women; or in 

UNDERSTANDING MODERN VACCINES men, regardless of their HSV seropositivity status. These were the first studies to report a significant difference in vaccine efficacy between men and women. This finding could have important implications for other vaccines targeting sexually transmitted diseases. The basis for this difference could relate to differences in how men and women respond to novel adjuvants or may reflect differences in the acquisition and natural history of genital herpes in men and women. A third Phase III efficacy trial of the gD2 candidate vaccine in HSV-1 and HSV-2 negative women who thought themselves possibly at risk of acquiring genital herpes (a different risk population than in the original two trials) has been completed and is being analysed. An initial assessment of the results of the third trial showed that the vaccine had an acceptable safety profile but the primary trial endpoint, prevention of genital herpes disease, was not met (NIAID, 2010) . Although the development of the vaccine has been stopped, further analyses and comparison of the trials may guide researchers as they continue seeking vaccines to control HSV infections.

As discussed in Chapter 2 e Vaccine immunology, some pathogens have complex life cycles. One specific example is parasites, sometimes using more than a single host, where each development phase is marked by differential expression of major proteins, meaning that possible antigen targets are host-and development-phase specific. Taenid worms aside, vaccines against parasites have been extremely difficult to develop and only a limited number have performed well in later-stage clinical trials.

The protozoan parasite Plasmodium falciparum, the most common cause of malaria, has a complex life cycle, as shown in Figure 6 .7.

The Plasmodium parasite has a genome encoding more than 5000 proteins, and presents different allelic and immunogenic Table 7) . One of the furthest advanced of these new candidate vaccines is RTS,S/AS01. The vaccine targets the pre-erythrocytic stage of the parasite (Figure 6.7) . To be protective, a vaccine targeted at this phase needs to induce humoral immunity, to prevent parasites from invading the liver, and cell-mediated immunity to destroy hepatocytes that become infected in the face of the humoral immune response. The RTS,S antigen, produced in Saccharomyces cerevisiae, contains sequences of the P. falciparum circumsporozoite protein, linked to the hepatitis B surface antigen (HBsAg). This chimeric protein spontaneously assembles into mixed polymeric particulate structures. In Phase II studies, the RTS,S/AS01 candidate vaccine induced a strong neutralising antibody response and cell-mediated immunity, and afforded protection against malaria (Bejon et al., 2008; Abdulla et al., 2008) . RTS,S/AS01 has been selected to proceed to Phase III clinical testing due to its higher efficacy compared with alternative formulations. If successful, the RTS,S/AS01 candidate vaccine could be the first licensed human vaccine against a parasite. Other malaria candidate vaccines in development are shown in Appendices, Supplementary Table 7 .

Pathogens may mutate or recombine to change their antigenic profile. Antigenic drift refers to a gradual process whereby point mutations in genes encoding antigenic proteins change the antigen sufficiently so that over time previously effective antibodies and vaccines no longer effectively control the pathogen and hence new vaccines need to be created. Antigenic shift is a more dramatic event where there is a recombination of genes between different pathogen strains that gives rise to a new strain with a unique antigenic profile.

In theory, pathogens are susceptible to selective pressure and an immunological environment that provides strong selective pressures should provide the 'bottleneck' that drives selection. This occurs with influenza viruses, where the high mutation frequency allows for the selection of mutants that are not neutralised. The risk of vaccine-mediated immune selection of pathogens, though certainly present, is difficult to demonstrate. Moreover, peptide vaccines only use the antigenic epitope so the risk of pathogen evolution is theoretically increased. However, this phenomenon has not been regularly observed in experimental studies and may reflect the complex nature of most vaccine antigens and the presence of immune responses against multiple antigens and multiple epitopes within antigens.

Serotype replacement, where the distribution of specific microbial serotypes within communities changes after the introduction of vaccines, has occurred for some bacterial pathogens and may be a consequence of the use of capsular vaccines that address only a limited number of serotypes.

Similarly, since their introduction in the 1940s, the use of antibiotics has exerted a selective pressure on bacterial strains leading to selection for common resistance alleles (eg the extended-spectrum beta-lactamase [ESBL] resistance of enteric bacteria and beta-lactamase resistance in gonococci). To date, there has been no requirement to remodel a vaccine because of vaccine-mediated immune escape; however, new vaccines against the pneumococcus have been licensed, including additional capsular types, to expand the geographical coverage of most frequent types and, in part, to counter the observed phenomenon of serotype replacement. Annual seasonal influenza infections are subject to natural antigenic drift which requires the reformulation of the vaccine when drifts occur, but there is no evidence that the deployment of the vaccine accelerates this drift.

Antigenic shift, while not the result of selective pressure, gives rise to viral strains containing a mixture of the surface antigens from the parent strains. Pathogens that can undergo antigenic shift,

including influenza viruses (Figure 6 .8), present major challenges for vaccine developers.

Chapter 4 e Vaccine adjuvants, there has been progress in the Another approach to the problem of influenza genome shifts has been to target weakly immunogenic conserved antigens such as the influenza M2e protein. One approach to addressing the weak immunogenicity of the antigen has been to link it to a potent Toll-like receptor adjuvant such as flagellin, an approach developed by VaxInnate Inc.

During primary infection of a single individual with HIV, mutations in surface proteins of the virus lead to selection of a 'cloud' of antigenic variants that can evade the cell-mediated immune responses complicating the development of broadly effective vaccines. This propensity for mutation has given rise to many strains of HIV (Figure 6 .9). Two types of HIV, HIV-1 and HIV-2, have been identified, with HIV-1 being the most common. On a global scale, HIV-1 strains are differentiated according to their respective group and subtypes (or 'clades') within groups. The amino acid sequence of the viral envelope glycoprotein gp120 shows 25e35% divergence between clades and up to 20% divergence within any given clade, which constitutes a formidable hurdle to vaccine development. This is made worse by recombination between clades of HIV-1, which has produced circulating recombinant forms (CRFs) which differ in antigenicity depending on the geographical region. 

Since the initiation of HIV vaccine programmes, more than 30 candidate vaccines have been tested in over 80 Phase I/II clinical trials involving more than 10,000 healthy human volunteers. Regrettably, all attempts to date have failed to yield a licensed HIV vaccine.

Questions remain concerning the immune mechanisms behind vaccines that achieve partial protection. Regardless of the unknowns, the ability to prevent infection in at least some individuals still offers real hope that a globally effective HIV vaccine might be possible. Current research is comparing the immune responses of subjects who are naturally protected against HIV with those who were infected, seeking to find the elusive immunological mechanisms of protection to help guide the design of future T-cell vaccines against the virus.

Infections of group A streptococcal serotypes (ie Streptococcus pyogenes) account for approximately 85% of cases of 

uncomplicated bacterial pharyngitis and streptococcal invasive infections in North America. The M protein of group A streptococci is a major virulence determinant of these organisms and also functions as a major target for protective antibodies. One of several strategies for vaccine prevention of these infections is based on type-specific M protein epitopes. However, group A streptococcal vaccine development faces many obstacles: i) the widespread diversity of circulating M protein types; ii) immunological cross-reactivity between epitopes in the M protein and several human tissues introducing an autoimmune risk; and iii) animal models are of limited value because humans are the only hosts for group A streptococci. In an attempt to partially overcome some of these obstacles, a design strategy akin to that of the pneumococcal polysaccharide vaccines has been employed to generate a group A streptococci multivalent M protein-based vaccine containing type-specific determinants from 26 different M serotypes. This multivalent vaccine is currently in clinical development.

The 'prime-boost' approach

The term 'prime boost' (or heterologous boosting) describes an approach to vaccination where one type of vaccine, such as a live-vector vaccine, is administered followed by a second type of vaccine, such as a recombinant subunit vaccine. This is in contrast with the traditional method of homologous boosting in which two or more doses of the same vaccine are given successively. The intent of prime-boost vaccination is to induce different types of immune responses and enhance the overall immune response, a result that may not occur if only one type of vaccine were to be given for all doses. This approach has been employed in trials with, for example, TB, CMV, malaria and HIV candidate vaccines. For example, in studies on new TB vaccines, subjects already primed with the live, attenuated BCG vaccine have been boosted with a subunit adjuvanted vaccine (see Tuberculosis). 

Respiratory syncytial virus is a common cause of bronchiolitis and pneumonia in infants, and exacerbations of chronic obstructive pulmonary disease in the elderly. The development of an effective vaccine has been challenging; natural immunity to RSV infection is incomplete and re-infections occur in all age groups. Moreover, the primary target population for vaccination is newborns and young infants, and they are a challenging population as they have relatively immature immune systems and the presence of maternal antibodies may interfere with vaccination of the young infant (see Chapter 2 e Vaccine immunology). The initial efforts to develop a formalin-inactivated cell culture-derived RSV vaccine resulted in an unanticipated enhancement of natural RSV disease in some of the RSV-naïve infants who received the vaccine in a clinical trial and subsequently were exposed to RSV. The exacerbated disease is thought to be due to an exaggerated T helper type 2 cell immune response (see Chapter 2 e Vaccine immunology). Safety concerns regarding the potential of vaccines to trigger or prime for immunopathological responses has resulted in a cautious approach to the development of RSV vaccines. The vaccine candidates most advanced in clinical development use two different approaches e one uses a live, attenuated virus with a gene deletion deliberately targeted to minimise immunopathological responses. The other approach uses a live viral vector to deliver only a key RSV surface antigen, thereby avoiding the risk of an immunopathological response arising from exposure to the RSV virus itself.

Infectious illnesses exert a major burden of disease in developing countries. The greatest burden is caused by diseases for which we currently have no vaccines, eg taeniid cestode parasites are associated with high human morbidity and losses in livestock.

Global efforts to reduce these infections in humans are ongoing through the use of antihelminthics and the implementation of lifestyle changes, but this is having little effect. However, substantial progress has been made towards developing veterinary vaccines which encourages investigation of the potential use of similar vaccines in humans to prevent, for example, hydatid disease (arising from infection with Echinococcus granulosus) and cysticercosis (from infection with Taenia solium).

Relative to their burden on society, such diseases have a low priority for funding. Unless comprehensive measures are taken to address the gaps in funding, research and global immunisation coverage, developing countries will continue to be overwhelmed by some of the most devastating diseases. In order to improve the situation, collaborative schemes are underway that bring together academic institutions, industry and public/charitable financing organisations. 

Microbiome Project is a National Institutes of Health initiative that seeks to determine the relationship between human health and changes in the human microbiome. By using revolutionary sequencing technologies to characterise the microbiology of five body sites e oral cavity, skin, vagina, gut and nasal tract/lung e an association may be made between the microbiomes associated with either the healthy body state or disease. Characterising microbes associated with disease-related pathogens may allow for the development of new vaccines that preserve or protect the healthy microbiome and hence could protect human health. Some of the areas of current research are outlined in the box, right.

Some conditions traditionally thought of as non-infectious may in fact have infectious origins (Table 6 .12); therefore, vaccination could be a strategy to prevent these diseases. Other diseases may result from an interaction between the host's genetic background and a particular microbe (a so-called gene-environment interaction). Some diseases have an established link with an identified infectious agent. For example, primary CMV infection is a known cause of congenital mental retardation; similarly the link between bacterial vaginosis and foetal prematurity is widely accepted. While some links have been established, others remain speculative (Table 6 .12).

Candidate vaccines are in development for the prevention and treatment of various types of addiction. The basic concept is to induce the production of antibodies which will bind the drug and impede its crossing the bloodebrain barrier to exert its psychoactive effects. Several nicotine candidate vaccines have now entered clinical trials. A cocaine candidate vaccine has also shown some benefit in a Phase IIb clinical trial. The key issue to date for both nicotine and cocaine candidate vaccines has been to induce 

Continued on next page high immunoglobulin (Ig)G anti-drug antibody levels, which appear to be critical in achieving some degree of efficacy. Candidate vaccines against methamphetamine addiction are also in early development.

To date, the approach to developing prophylactic cancer vaccines has been to target infectious diseases that cause or contribute to the development of cancer such as HPV (cervical cancer) and HBV (hepatocellular carcinoma). Examples of infectious diseases associated with cancer are shown in Table 6 .13. The successful development of a nicotine vaccine would be expected to reduce cigarette smoking-related lung cancer.

Some cancers express tissue-specific antigens that can be targeted by the immune system. Therapeutic cancer vaccines aim to target tumour-associated antigens (TAA) with T-cell mediated immune responses. TAA can be related to the genetic changes that drive the cancer (eg Ras oncogene), or inappropriate up-regulation/ expression of genes (eg carcinoembryonic antigen). With such TAA targets, vaccines aim to maximally stimulate a cytotoxic T-cell response and their design often includes adjuvants to enhance antigen presentation. Tumours develop in a multistep process in the face of the host immune response and frequently evolve to escape immune control. Mechanisms of evasion include genetic changes (loss of human leukocyte antigen/TAA expression) and induction of immune regulatory systems (T-cell anergy due to the activity of 

T reg cells) which limit anti-tumour immunity. The key approach for therapeutic cancer vaccines is resetting the immune response to deliver anti-tumour immunity that alters or destroys cancer cells and hence eliminates or reduces the tumour. One strategy uses the patient's own tumour as the immunogen, thereby providing all the potential idiotypic changes that might act as TAA, in conjunction with antigen-presenting DCs harvested from the same patient and activated in vitro (see Dendritic cell vaccines). There are different types of therapeutic candidate vaccines currently undergoing clinical trials for numerous types of cancer (Table 6 .14). The most advanced candidates currently in Phase III are described in Chapter 4 e Vaccine adjuvants.

There has been some success in the development of therapeutic cancer vaccines, with the FDA approval of the first DC vaccine Oxford BioMedica MUC1, Mucin 1 cell surface; hIL2, human cytokine interleukin-2; ASCI, antigen-specific cancer immunotherapeutics; MAGE, melanomaassociated antigen; CTL, cytotoxic T lymphocyte. Every effort has been made to verify the information in this table. The information included is not meant to be exhaustive but is intended to provide an overview of the subject matter.

vaccine approaches. However, this presents opportunities for the application of novel technologies and adjuvants. Some of the considerations for vaccines designed for use in special populations include: immunosenescence in the elderly; the poor immunological response to traditional vaccines seen in immunocompromised individuals (patients with HIV, transplant recipients); the crossing of vaccine components into the foetal bloodstream when vaccines are administered to pregnant women; and the safety and immunogenicity concerns surrounding vaccines for neonates due to their naïve and immature immune system.

Cell-mediated immunity is depressed in pregnant women, leaving them at high risk of infection from pathogens, including those harmful to the foetus. Most live, attenuated vaccines are contraindicated during pregnancy because of the theoretical risk of foetal infection from the vaccine. However, inactivated viral or bacterial vaccines can be administered. Pregnant women can, therefore, be vaccinated against some infections, including several that pass from mother to foetus (such as hepatitis A and B), and against infections acquired by the infant in the first few months of life (often from close contact with the mother). In the latter case, the infant can be protected by transfer of maternal antibodies during late gestation. Examples of diseases that can be prevented in pregnant women include influenza, tetanus, diphtheria and probably pertussis. Other diseases, such as those caused by the so-called TORCH pathogens (toxoplasma, others including syphilis, CMV and HSV), are not yet preventable through vaccination though encouraging Phase II results have been presented for a vaccine to prevent Group B streptococcus carriage in pregnant women (Hillier et al., 2009; Smith, 2009 ).

The 2009 H1N1 pandemic influenza outbreak posed an increased risk to pregnant women and vaccination was specifically recommended in pregnant women as one of the high-risk groups.

The pandemic example has emphasised once more the importance of protecting pregnant women against influenza.

Seasonal influenza vaccination in pregnancy is well tolerated and the benefiterisk profile when administered to pregnant women supports its use during pregnancy. Many public health authorities worldwide recommend seasonal influenza vaccination in pregnant women and this recommendation is motivated not only by the potentially severe course of influenza during pregnancy, but also by the need to protect vulnerable infants against influenza during their first months of life. Boosting RSV immunity in pregnant women through vaccination may be another approach to protecting the newborn against RSV infection during the most vulnerable early months after birth.

Neonatal immunisation is a strategy to protect infants against infections during a particularly vulnerable period. A recent study showed that immunisation with an acellular pertussis vaccine at birth and 1 month of age induces high IgG anti-pertussis antibody titres by 2 months of age (Wood et al., 2010) . It is hoped that this approach may reduce death and morbidity from Bordetella pertussis infection in the first 3 months of life.

The elderly respond poorly to vaccination as the immune system becomes more senescent with increasing age and, therefore, new vaccine technologies are needed to improve the response to vaccination in this population. In the late 1990s, an influenza vaccine adjuvanted with the oil-in-water emulsion, MF59Ô, was shown to be more effective at inducing high immune responses in the elderly (Minutello et al., 1999) . Alternative vaccine administration techniques have also been studied in the elderly. Research showed that in subjects 60 years of age or older, an influenza vaccine administered with an intradermal microinjection system induced significantly higher antibody titres compared with IM vaccination (Arnou et al., 2009 ). Subsequently, a microinjection system influenza vaccine was licensed for use in Europe, and a high antigen dose formulation has been licensed for the elderly in the USA.

Individuals with cancer, HIV infection or who are asplenic can be immunocompromised as a result of their condition. Patients can also be immunocompromised as a result of therapy, eg when receiving an organ transplant, radiation therapy or immunosuppressive medication. Such patients are therefore at an elevated risk of infection from pathogens such as herpesviruses (particularly CMV and EpsteineBarr virus), HBV, HCV, pneumocystis and coinfections and represent a special population regarding immunisation. Despite a likely reduction in the efficacy of vaccinations in immunocompromised individuals, immunisation remains a frequent recommendation in the hope that at least partial immunity will be achieved. Eliciting a response from vaccination in immunocompromised patients may require an increase in the dose and/or number of doses; altering the dosing interval; selecting a different vaccine formulation; or administration via an alternative route. Evidence in this patient population is lacking and guidelines are often based on theoretical assumptions.

Live vaccines are generally contraindicated in immunocompromised or immunosuppressed individuals due to the risk of an active and symptomatic infection resulting from the vaccine itself (non-controlled replication process). Encouragingly, vaccine formulations with highly purified antigens and novel adjuvants or alternative deliveries have been shown to induce more effective immune responses than the classical inactivated vaccines in immunocompromised hosts, including patients with end-stage renal diseases in pre-haemodialysis and haemodialysis (see Chapter 4 e Vaccine Adjuvants), patients with HIV and those who have received haematological stem cell transplants.

The future of vaccine development can build on the knowledge and experience gained over the last 200 years, and at the same time can take advantage of the most cutting-edge technologies and 194 UNDERSTANDING MODERN VACCINES research. New approaches to antigen selection and production, antigen delivery, adjuvantation and vaccine administration will allow us to target established and emerging diseases, and populations with complex needs. Vaccination has been one of the most successful and cost-effective health interventions ever conceived and is now expanding further into cancer and chronic diseases. This expansion of scope and the subsequent impact on human disease is likely to continue into the future in currently unforeseen ways, further increasing the importance of vaccine science and engineering in improving human health.

Safety and immunogenicity of RTS,S/AS02D malaria vaccine in infants

Intradermal influenza vaccine for older adults: a randomized controlled multicenter phase III study

Efficacy of RTS,S/AS01E vaccine against malaria in children 5 to 17 months of age

Women receiving Group B Streptococcus serotype III tetanus toxoid (GBS III-TT) vaccine have reduced vaginal and rectal acquisition of GBS Type III

New vaccines against tuberculosis

Safety and immunogenicity of an inactivated subunit influenza virus vaccine combined with MF59 adjuvant emulsion in elderly subjects, immunized for three consecutive influenza seasons

A vaccine to prevent herpes zoster and postherpetic neuralgia in older adults

A subunit cytomegalovirus vaccine based on recombinant envelope glycoprotein B and a new adjuvant

Vaccine prevention of maternal cytomegalovirus infection

Vaccination with ALVAC and AIDSVAX to prevent HIV-1 infection in Thailand

Successful immunization of children with and without maternal antibody by aerosolized measles vaccine. I. Different results with undiluted human diploid cell and chick embryo fibroblast vaccines

Associate Clinical Professor of Medicine

GlaxoSmithKline Herpes Vaccine Efficacy Study Group. Glycoprotein-D-adjuvant vaccine to prevent genital herpes

Prevention of tuberculosis in Bacille Calmette-Guérin-primed, HIV-infected adults boosted with an inactivated whole-cell mycobacterial vaccine

The second Geneva Consensus: recommendations for novel live TB vaccines

Acellular pertussis vaccine at birth and one month induces antibody responses by two months of age

Genocea technology

Developing cell culture-derived pandemic vaccines

Seeking new pathways for HIV vaccine discovery

Infectious diseases and global cancer control

Intrauterine infection and preterm delivery

The NIH human microbiome project

Alliance for Case Studies for Global Health. Case studies for global health: Building relationships

Alliance for Case Studies

International AIDS Vaccine Initiative

VACCINES OF THE FUTURE National Institute of Allergy and Infectious Diseases (NIAID). Statement: Study finds genital herpes vaccine ineffective in women

World Health Organization's Special Programme for Research and Training in Tropical Diseases

AIDSVAXÔ is a trademark of Global Solutions for Infectious Diseases Corporation; ALVACÔ is a trademark of Connaught Technology Corporation

MF59Ô is a trademark of Novartis; MontanideÔ is a trademark of Seppic; NanoStatÔ is a trademark of NanoBio Corporation. For details of trademarks not listed above please see the manufacturer's website (see section Internet resources)

VACCINES OF THE FUTURE