Adjuvanted influenza vaccines

Abstract

The search for adjuvants has been stimulated by the need to ensure greater protection against influenza among subjects who show a reduced immune response to conventional influenza vaccines. Aluminum salts have long been used but are not considered satisfactory. This has led to the development of other possible compounds, sometimes on the basis of new knowledge concerning the mechanisms regulating the immune response to infections. Some of the new adjuvants (emulsions and virosomes) have been widely evaluated, and the apparently good results have led to the registration of adjuvanted influenza vaccines for use in humans, at least in some countries and in some subjects. In other cases, the adjuvants have been mainly or exclusively studied in experimental animals, and are unlikely to be used in humans in the near future. However, even in the case of those for which a considerable amount of data are available, assessments of their superiority over conventional influenza vaccines have mainly been based on immunogenicity studies, and have not been confirmed by comparative, randomized, double-blind clinical trials. Moreover, the very few human data comparing different adjuvants are frequently conflicting. The aim of this review is to discuss the characteristics and advantages of the adjuvants that have so far been used and to describe some of the new adjuvants that are still in the development phase.

Introduction

For a number of years, the yearly administration of a specific vaccine has been recommended throughout the world in order to reduce the risk related to influenza infection among the elderly and subjects of any age with chronic underlying diseases.¹ Furthermore, given the medical and socioeconomic impact of influenza on healthy children,^2,3 many (but not all) health authorities have more recently included all healthy children and adolescents in the list of subjects to whom influenza vaccination should be administered.^4-6

Influenza vaccines prepared for intramuscular use and based on inactivated viruses (TIVs) have been available for more than 50 years, although the first preparations based on whole viruses have been replaced by subunit and split-virus vaccines in order to reduce the incidence of adverse events. However, the immune response to all of these preparations among subjects at greater risk of severe influenza, such as old people, younger children and immunocompromised patients, is considered not completely satisfactory by some experts.⁷

In addition to their sometimes inadequate immunogenicity, TIVs have two other limitations that can further reduce their global efficacy. The first is that they are not very capable of evoking high levels of cross-reactive antibody to heterovariant viruses in the case of a mismatch between circulating strains and the strains included in the vaccine⁸ and, as antigenic drifts are common, this means that even people with a perfectly functioning immune system may not be adequately protected by conventional TIVs. The second is the amount of antigen required to obtain a protective immune response. A number of studies have clearly shown that no less than 15 µg of hemagglutin (HA) for each of the viral strains included in the vaccine are required to obtain seroconversion and the production of seroprotective antibody concentrations.⁹ As the production of influenza vaccines is currently limited because it is based on the availability of chicken eggs, the need for a large amount of antigen to prepare a single dose can significantly affect the total number of vaccinated subjects, and this may have serious consequences in the case of a pandemic when universal vaccination is recommended.

Various attempts have been made to overcome these problems: the composition of TIVs has been modified by adding adjuvants, the amount of antigen has been increased, and other routes of administration have been tried. Moreover, completely new vaccines have been prepared, including some containing preserved antigens.¹⁰ The addition of adjuvants was the first method used to improve TIVs, and has been the most widely studied. The aim of this review is to discuss the characteristics and advantages of the adjuvants that have so far been used and to describe some of the new adjuvants that are still in the development phase.

Currently Licensed Adjuvanted Influenza Vaccines

Very few adjuvants for influenza vaccines have been licensed for use in humans and, in some cases, licenses have only obtained in some countries, and some subjects (mainly children) have been excluded for fear of adverse events (Table 1). Aluminum salts have been used as adjuvants to vaccines for more than 80 years. Until recently, were the only adjuvants approved in the US Examples of vaccines containing these adjuvants are hepatitis A, herpatitis B, human papillomavirus and pediatric diphtheria, tetanus and pertussis vaccines.¹¹ However, none of the licensed influenza vaccines are adjuvanted. with aluminum salts. Virosomes and oil-in-water emulsions (such as MF59 and AS03) are licensed in Europe but, although the former can be used in subjects of all ages, the latter cannot be administered to children.

Table 1. Adjuvants licensed for use in the preparation of influenza vaccines for humans

Adjuvant	Immunostimulatory component(s)	Mode of action
Alum	Aluminum salts	Increased attraction and uptake of APCs; activation of inflammasome complex with increased efficiency of innate immunity
Oil-in-water emulsions MF59 AS03	Squalene Squalene + α tocopherol	Innate inflammatory responses; increased attraction and uptake of APCs; increased antigen persistence at injection site and presentation to immune-competent cells
Virosomes	Virosomes	Stronger interactions with B lymphocytes and APCs with a strong Th1 response

APCs, antigen presenting cells

It needs to be underlined that most of the data concerning adjuvanted vaccines regards their ability to induce antibody production, and their safety and tolerability; there are very few data concerning their real effectiveness and efficacy in comparison with conventional TIVs. This may be important when judging the use of adjuvanted vaccines as a means of preventing influenza. Nauta et al. found that an increase in antibody levels does not necessarily mean greater clinical protection.¹² In particular, it has been shown that mean protection titer rises from 0% for negative subjects to 75% when the titer is 40, but the increase in efficacy is less substantial if the titer is ≥40.¹³ This means that, if a conventional TIV evokes antibody titers of ≥40 in most treated subjects, the increase in absolute antibody concentration offered by an adjuvanted vaccine may not significantly modify global protection. However, although this is probably true of healthy adults who have a good immune response to conventional TIVs, adjuvanted vaccines may be really more protective in subjects who do not often respond to TIVs but show adequate antibody production when immunized with adjuvanted products. Moreover, neither the importance of cell-mediated immune response in conditioning defenses against influenza viruses, nor the type and extent of the cellular immunity evoked by different adjuvants have yet been precisely defined. Consequently, as there is a lack of comparative studies of different adjuvanted vaccines in humans, it is not possible to say what is the best adjuvant or which adjuvanted vaccine should preferred to immunize subjects at risk. A similar conclusion can be drawn in the case of safety and tolerability because, once again, there are very few comparative studies.¹⁴

Aluminum salts

Although aluminum salts (i.e., aluminum hydroxide, aluminum phosphate or potassium aluminum sulfate) have been largely used since long time, their precise mechanism of action is not completely understood. It was long thought that they acted mainly because they and their adsorbed antigen formed a depot at the injection site that improved the attraction of, and uptake by antigen presenting cells (APCs).¹⁵ It has recently been demonstrated that engulfment of the salt, or the uric acid arising from necrotic cells in response to the presence of the salt, leads to the lysosomal disruption and activation of NALP3, a component of the inflammasome complex that induces a number of pro-inflammatory cytokines and increases immunogenicity.¹⁶ It has also been suggested that the antigens absorbed on the aluminum salts are presented in a multivalent form, making them more efficiently internalized by APCs.¹⁵ However, one drawback of aluminum salts is that they primarily enhance Th₂-driven antibody responses and have little effect on Th₁-type responses, which are critical for protection against intracellular pathogens.¹¹

The efficacy of aluminum salts in increasing immune responses to TIVs has been widely demonstrated in healthy subjects of all ages, as well as those suffering from severe chronic underlying diseases. However, there few and sometimes conflicting data regarding younger children: the increased response is relatively small and, in the few comparative studies, significantly less than that induced by other adjuvants such as MF59 and AS03.¹⁶

The global experience with all the vaccines containing aluminum salts clearly indicate that serious adverse events attributable to these compounds are rare, even if local reactions such as redness, swelling and/or tenderness at the injection site can be relatively frequent, although more severe local reactions. such as large areas of swelling, sterile abscesses, subcutaneous nodules and allergic responses are much less common.¹⁷ A recent review found no evidence that exposure to aluminum-containing vaccines causes any serious or long-lasting adverse event.¹⁸

Emulsions

MF59

MF59 is a low oil-in-water emulsion of microvescicles of squalene, a natural substance in plants, animals (particularly in the liver), and humans as a natural precursor of cholesterol, and as a component of cell membranes and sebaceous gland secretions. The emulsion consists of a drop of oil surrounded by a monolayer of two non-ionic surfactants found in many foods (polysorbate 80 and sorbitan trioleate) that is used to stabilize the emulsion droplets. Citrate buffer is used to stabilize pH.¹⁹

Although the mechanism of action of MF59 is not fully understood, most experts believe that it acts by directly involving the innate inflammatory response, recruiting and activating APCs, enhancing antigen persistence at the injection site and ensuring its presentation to immune-competent cells and eliciting various patterns of cytokines.²⁰ Globally, MF59-adjuvanted vaccine seems to enhance T cells and antibody responses without altering the Th₁/Th₂ cell balance of the specific antigen.²¹

The immunogenicity of seasonal MF59-adjuvanted TIV (MF59-AV) has been widely evaluated in open and controlled studies involving the elderly and adult patients with chronic underlying diseases, such as severe respiratory and/or cardiovascular diseases, cancer, metabolic disorders, renal transplantation or HIV infection. All of these studies showed that seasonal MF59-AV evoked an immune response that was consistently superior to that evoked by conventional TIVs, and led to higher seroconversion and seroprotection rates and geometric mean titers (GMT) even in subjects with no pre-existing immunity against the influenza strains included in the vaccine.^22-26 Moreover, the addition of MF59 significantly increases the ability of TIV to induce cross-immunity against heterovariant viruses in the elderly and subjects with chronic disease. Given the frequency with which genetic drifts in influenza viruses occur, this seems to be the most important immune advantage arising from the use of MF59. Similar results have been obtained in younger children: MF59-AV is not only more immunogenic than conventional TIVs, but also capable of inducing significantly greater cross-reactivity at least against mismatched A/H3N2 and A/H1N1 strains.^27,28

The greater immunogenicity of MF59-AV in healthy subjects and subject with a known reduced immune response to conventional TIVs was also demonstrated in the case of the vaccine specifically prepared to face the 2009 pandemic.²⁹ This made it possible to reduce the effective vaccine dose and thus increase the total number of doses available to meet the global demand. A number of studies found that adequate seroconversion and seroprotection rates could be reached at lower antigen doses than the 15 µg of HA for each influenza strain usually needed to obtain protection with conventional TIVs.^30-32 One of the best examples in this regard is given by our data regarding prematurely born infants and young children:³³ The administration of a single dose of only 7.5 µg of HA together with MF59-adjuvant was sufficient to achieve seroprotection in more than 90% of children aged six to 23 months who were born before the end of the 32nd week of gestation. Khurana et al. used genome fragment phage display libraries and surface plasmon resonance to study the effects of MF59 on the quantity, diversity, specificity and affinity maturation of human antibody responses to the A/H1N1 vaccine in different age groups.³⁴ They found that MF59 selectively enhanced antibody responses to the HA1 globular head (relative to the more conserved HA2 domain) in terms of increased antibody titers and a more diverse antibody epitope repertoire in both adults and children. Moreover, antibody affinity was significantly increased in toddlers and children who received the pandemic MF59-AV, and this was associated with greater virus-neutralizing capacity.

Substantially similar immunogenicity was found when antigens derived from the potential pandemic A/H5N1 virus were used to prepare a pre-pandemic MF59-AV.^35,36 In this case, the importance of the increased immunogenicity due to the presence of MF59 seemed to be even greater than that related to seasonal or the 2009 pandemic virus because it was demonstrated that the HA of A/H5N1 virus is a very weak antigen and that only larger and/or multiple doses were needed to obtain a protective immune response.³⁷

Although the increased immunogenicity of influenza vaccines adjuvanted with MF59 seems to be clear, there are very few data concerning the effectiveness of MF59-AV. Two studies have found that seasonal MF59-AV is more protective than conventional TIVs:^38,39 however, the endpoint of one was mortality without virological confirmation³⁸ and the other was not randomized,³⁹ and so the conclusions of both are debatable. Consequently, further studies are needed to establish what can be expected from the immune protection offered by the use of the adjuvant.

The safety and tolerability of MF59-AV was satisfactory in all studies involving the elderly, adults and young children^33,40-42 although, in comparison with unadjuvanted preparations, MF59 was associated with slightly higher rates of local adverse events (particularly erythema and induration at the injection site accompanied by pain). However, systemic adverse events such as fever, general malaise, headache and myalgia were very uncommon, and usually mild and transient.⁴³ The good safety profile shown in clinical trials has been confirmed by post-marketing pharmacovigilance. Analysis of the data regarding seasonal MF59-AV, which has been administered to more than 27 million subjects, indicates that serious adverse events were diagnosed in 1.4/100,000 doses: This is no greater than that expected in the general population and lower than that recorded in countries in which conventional TIVs are used. Regarding Guillan-Barré syndrome (the most common of the severe adverse events following influenza vaccination), the incidence was 0.03 cases/100,000 recipients, less than that in the general population.⁴⁴ Furthermore, Italian post-marketing surveillance of adverse events after the administration of A/H1N1 pandemic MF59-AV revealed a safety and tolerability resembling that of seasonal MF59-AV.⁴⁵

It has been discovered that the so-called Gulf war syndrome and the emergence of narcolepsy initially attibuted to MF59 do not have any relationship with the adjuvant. Gulf war syndrome is typified by a constellation of unexplained symptoms including fatigue, rashes, headache, arthralgias, myalgias, lymphadenopathies, diarrhea, memory loss, autoimmune thyroid diseases, increased allergies, sensitivities to environmental elements and neurological abnormalities.⁴⁶ Nearly a decade ago, it was thought that squalene was the experimental anthrax vaccine ingredient that caused the syndrome in many veterans, because antibodies to squalene were detected in the blood of most of the patients. This raised widespread concern about the safety of squalene-containing adjuvants of TIVs, especially MF59. However, subsequent clinical evidence has clearly suggested that squalene is poorly immunogenic, that low titers of squalene antibodies can also be detected in the serum of healthy subjects, and that neither the presence of anti-squalene antibodies nor their titer is significantly increased by immunization with vaccines containing squalene (or MF59) as an adjuvant.⁴⁷

In Scandinavian countries, narcolepsy has been associated with the use of vaccines containing another squalene-based adjuvant but has never been diagnosed in subjects receiving MF59. One study that analyzed 5,305 adverse events reported after immunisation with more than 23 million MF59-adjuvanted pandemic vaccine doses showed no case of narcolepsy and no evidence of an increase in the risk of sleep-related problems.⁴⁸

AS03

AS03 is an oil-in-water emulsion containing 5% DLα-tocopherol and squalene in the oil phase and 2% of the non-ionic detergent polysorbate 80 in the aqueous phase. Given its similarities to MF59, it is thought that its mechanism of action is also likely to be similar. The association of AS03 with influenza antigens was widely evaluated after the preparation of a vaccine against the avian A/H5N1 influenza virus, which was considered to be the most probable cause of the expected pandemic and proved to be an infectious agent with very poor immunogenic properties.^49-51 Additional data were gathered when AS03 was associated with the A/H1N1 pandemic virus.^52-56

All of the data collected in clinical trials involving the elderly, children of different ages and immunocompromised patients have clearly shown that the addition of AS03 significantly enhances immune responses to the HA of influenza viruses. In comparison with unadjuvanted preparations, AS03-adjuvanted vaccines (AS03-AV) induce a satisfactory immune response even though fewer doses are administered with significantly less antigen in each dose. Unfortunately, there are very few data regarding the clinical advantages of the use of AS03-AV in clinical practice and so, once again, it is not known whether increased immunogenicity is always accompanied by a real reduction in the incidence of influenza infection. However, data collected in Canada during the recent pandemic indicate that AS03-AV was effective in preventing influenza in both children and adults.⁵⁷

However, it has to be pointed out that the incidence of adverse events was not marginal. In infants and toddlers: despite the lower antigen content, the administration of a first dose was followed by pain at the injection site in 35.6% of cases, and the incidence of fever was 20.2%. The frequency of all of the adverse events increased after the second dose, reaching values of more than 40%.⁵⁵ Moreover, a significant increase in the occurrence of narcolepsy (a very rare neurological disorder) was associated with the use of AS03-AV in Finland and Sweden.⁵⁸ The same problem was not observed in other countries in which the vaccine was widely used, and so the World Health Organization (WHO) decided to continue investigations in order to establish what other factor(s) may be involved.⁵⁹ As narcolepsy has a strong genetic linkage, it has been surmised that genetic factors associated with AS03 could have favored the development of an autoimmune disorder leading to the neurological disease.⁵⁹

Virosomes

Virosomes consist of phospholipids that spontaneously form virus-like vescicles to which influenza virus surface glycoprotein HA and neuraminidase (NA) are anchored.⁶⁰ As they are structurally similar to the native virus, virosomes retain their cell binding and membrane fusion properties. Consequently, like the native virus, they avidly interact with B lymphocytes and are taken up by APCs with a stronger Th₁ response than that induced by a traditional split or subunit TIV.⁶¹

A number of studies performed since 1997 have demonstrated that virosome-adjuvanted vaccines (VAVs) are immunogenic, safe and well tolerated in healthy subjects and immunocompromised patients of any age.⁶² In studies comparing a VAV with a conventional TIV,^63,64 more subjects in the VAV group achieved protective antibody levels but, when MF59-AV was the comparator, there was no difference in antibody titers.^65,66 Data have also been collected regarding the effectiveness of the vaccine in the elderly and children. In comparison with untreated controls, all of these studies found a significant reduction in the incidence of influenza-like illness, even when children with chronic underlying disease were included. However, as in the case of other adjuvanted vaccines, it is not possible to evaluate the real importance of adding virosomes to TIVs in clinical practice because there is a lack of comparative controlled efficacy trials. How, it does seem that VAV is better tolerated than MF59-AV. In comparative studies involving the elderly, the local and systemic adverse events following vaccine administration were significantly less frequent in the VAV than in the MF59-AV group.^65,66

Adjuvants for the Future

Over the past 20 years, a greater understanding of innate and adaptive immunity and their close molecular interactions in host responses to a pathogen has enabled researchers to identify possible new adjuvants to increase the efficacy of vaccine prevention. However, most of the studies so far available have involved experimental animals and the paucity of human data means that the use of these adjuvants in influenza vaccines for humans is still very far.

Liposomes are synthetic nanospheres consisting of lipid layers that can encapsulate antigens and act as antigen delivery vehicles. They have recently been used to prepare a TIV adjuvant that combined cationic liposomes of dimethyldiocatadecyl ammonium as the delivery vehicle with trehalose 6,6’-dibehenate as both an immunomodulator and liposome-stabilizing compound.⁶⁷ In mice, the administration of a TIV with the cationic liposome adjuvant system was followed by a significant increase in humoral immune response as measured by HA inhibition and influenza-specific serum antibody titers and a strong Th₁ response with increased levels of interleukin(IL)-1β, IL-2, IL-12, IFNγ and TNFα. Furthermore, high levels of IL-17 were detected and, importantly, the Th₁/Th17 cytokine profile was maintained 20 weeks after the last vaccination, thus indicating the long-lasting effect of the adjuvant. Finally, the vaccine reduced weight loss and decreased influenza-related body temperature, and also increased the survival of mice challenged with a drift A/H1N1 pandemic influenza strain. Another cationic liposome for which promising results have been obtained in the experimental animals is the N-palmityol-derythro-sphingosyl-Carbamoyl-Spermine lipid (CCS/CHA). The optimized CCS/CHA split virus vaccine, when administered intramuscularly (i.m.), has been found significantly more immunogenic in mice, rats and ferrets than split virus HA vaccine alone. Moreover it provides for protective immunity in ferrets and mice against live virus challenge that exceeds the degree of efficacy of the split virus vaccine. Similar adjuvant effects of optimized CCS/C are also observed in mice for H1N1 swine influenza antigen.⁶⁸

Virus-like particles (also called nanoparticles) derived from the expression of papaya mosaic virus coat protein (PaPMV) in bacteria and consisting of several hundred recombinant coat proteins organized in a repetitive and ordered manner have been tried as TIV adjuvants.⁶⁹ They are used as an epitope display system that leads to the production of antibodies against surface-exposed peptides, thus providing protection.^70,71 Savard et al. treated mice and ferrets with PaPMV nanoparticles and observed an increase in the global humoral response to TIV by measuring the increase in total IgG titers, particularly those of the IgG2a subclass.⁷² This seems to be important because IgG2a is more effective in preventing intracellular virus replication as it more efficiently activates complement activation and antibody-dependent cell immunity.^73,74 It has also been found that the addition of PaPMV nanoparticles to TIV can lead to the long-lasting protection of experimental animals against heterosubtypical strains as a result of the production of antibodies against a highly conserved pocket in the stem region of the viral HA containing the fusion peptide and a cell-mediated immune response to highly conserved influenza proteins within different subtypes, such as NP and M1. Finally, unlike other adjuvants, the PaPMV nanoparticles did not cause any significant local adverse events, probably because their administration was not associated with the secretion of TNF-α, which induces a strong inflammatory response.

The addition of cytosine guanine di-nucleutide-containing oligodeoxynucleotides (CpG-ODN) or recombinant chicken interferon-γ (rChiIFNγ) has been tried as a means of further increase the immunogenicity of VAV.⁷⁵ These adjuvants were chosen because previous studies had clearly shown that they may act as immune inductors by stimulating B-cell proliferation, Ig production and monocyte cytokine secretion and activating natural killer (NK) cytotoxicity and IFNγ release.⁷⁶ The results of a study of chickens showed that the addition of CpG-ODN was associated with significantly higher systemic and neutralizing antibody production than that elicited by VAV alone whereas, despite its clear ability to enhance antigen-specific IgG and IgA responses, rChiIFNγ induced antibodies with less neutralizing capacity.⁷⁷ Antibodies generated in chickens immunized with CpG-ODN are more capable of cross-reacting with several viral strains than those included in VAV, even in the presence of significant genetic differences, which suggests that this kind of adjuvant can protect against heterotypical influenza viruses. Moreover, CpG-ODN VAV enhanced the cell production of INFγ when cells from immunized animals were re-stimulated in vitro, thus showing that a VAV containing CpG-ODN can induce cell-mediated responses.

Several years ago, it was found that micelles of saponin Quil A extracted from the bark of Quillaja saponaria may act as an antigen delivery system with powerful immunostimulating activity.⁷⁸ A further development of this adjuvant led to the formulation of particular antigen delivery systems consisting of antigen, cholesterol, phospholipid and saponin-defined immunomudulatory complexes (ISCOMs), and finally the production of ISCOMATRIX^TM, a particular adjuvant consisting of cholesterol, phospholipid and saponin without antigen. However, antigens can be formulated with ISCOMATRIX^TM to produce ISCOMATRIX^TM vaccines that can provide similar antigen presentation and immunomodulatory properties to those of ISCOMs, but with much broader application as they are not limited to hydrophobic membrane proteins. Saponin-based adjuvants enhance antigen uptake and prolong retention by dendritic cells in draining lymph nodes, induce the activation of dendritic cells and lead to strong antibody and T cell responses.⁷⁹ Moreover, unlike most other adjuvants, they enable substantial MHC class I presentation and induce both CD8⁺ and CD4⁺ T cell responses to a variety of soluble protein antigens.⁸⁰ One study evaluated the dose of HA needed to obtain a protective immune response in ferrets with the use of three A/H5N1 influenza vaccines (one unadjuvanted, one adjuvanted with aluminum phosphate and the third adjuvanted with ISCOMATRIX^TM), and found that aluminum allowed a significant dose reduction. However, the saponin-based adjuvant induced a more balanced immune response (potentially including cytotoxic T lymphocytes) and protected against death and viral disease at even lower doses and by mechanisms that might not be strictly related to the viral HA.⁸¹ The current applications of ISCOMATRIX^TM in humans include the development of a nasal influenza vaccine.⁸²

Toll-like receptor (TLR) agonists are considered possible new adjuvants because they are capable of recognizing pathogen-associated molecular patterns and initiating an immune response. Alone or in combination with other immunoenhancers, they are currently under evaluation in a number of vaccine trials and may also be considered in the preparation of influenza vaccines. One good example is flagellin. Song et al. first demonstrated that the globular head of HA fused to flagellin of Salmonella typhimurium fljB (a TLR5 ligand) elicits protective immunity to lethal A/H1N1 and A/H5N1 influenza infections in mice.⁸³ It was subsequently discovered that vaccine candidates in which the HA globular head of A/H1N1 2009 pandemic virus was fused to STF2 at the C-terminus instead of domain 3, or in both positions, significantly reduced the titer of lung virus on day 4, offered full protection against a lethal virus challenge and led to a neutralizing antibody titer that lasted at least eight months.⁸⁴ Recently a novel influenza vaccine with these characteristics has been evaluated in humans, particularly in subjects ≥65 y who received different amounts of the antigen with a single administration. A 5 μg dose of VAX125 was demonstrated to be safe and able to induce a greater than 10-fold increase HA antibody levels and nearly complete seroprotection confirming the possible use of this new vaccine to prevent influenza in humans.⁸⁵

Alum and monophosphoryl lipid A (MPL), an agonist of TLR4, is another combination (AS04) for which there are data suggesting its possible use as an adjuvant for human influenza vaccines. Although it has so far been mainly used in experimental animals to increase the immune response to HPV vaccine, this combination has characteristics that may also be useful for influenza vaccines because it has been shown that it enhances immune responses by rapidly triggering a local cytokine response leading to the optimal activation of APCs.⁸⁶

Conclusions

The need for increased protection against influenza in subjects who may show a reduced immune response to conventional influenza vaccines has largely stimulated adjuvant research. Although aluminum salts have been used for a good number of years, they are not considered satisfactory. This has led to the development of other possible compounds, sometimes on the basis of new knowledge concerning the mechanisms regulating the immune response to infections. Some of the new adjuvants (emulsions and virosomes) have been widely evaluated and the apparently good results have led to the registration of adjuvanted influenza vaccines for use in humans, at least in some countries and in some subjects. In other cases, the adjuvants have been mainly or exclusively studied in experimental animals, and are unlikely to be used in humans in the near future. However, even in the case of those for which a considerable amount of data are available, assessments of their superiority over conventional influenza vaccines have mainly been based on immunogenicity studies and have not been confirmed by comparative, randomized, double-blind clinical trials. Moreover, the very few human data comparing different adjuvants are frequently conflicting. Consequently, it is still not possible to say definitely whether the new adjuvanted vaccines should replace conventional preparations in the prevention of influenza. Further studies of these products in humans are therefore urgently needed.

Acknowledgments

None of the authors has any commercial or other association that may pose a conflict of interest. This study was supported in part by grants from the Italian Ministry of Health (Bando Giovani Ricercatori 2007) and Amici del Bambino Malato Onlus.