Universal influenza vaccines, science fiction or soon reality?

Abstract

Currently used influenza vaccines are only effective when the vaccine strains match the epidemic strains antigenically. To this end, seasonal influenza vaccines must be updated almost annually. Furthermore, seasonal influenza vaccines fail to afford protection against antigenically distinct pandemic influenza viruses. Because of an ever-present threat of the next influenza pandemic and the continuous emergence of drift variants of seasonal influenza A viruses, there is a need for an universal influenza vaccine that induces protective immunity against all influenza A viruses. Here, we summarize some of the efforts that are ongoing to develop universal influenza vaccines.

Keywords:

• immunity
• influenza
• universal
• vaccine
• virus

Seasonal influenza is an annually recurring acute respiratory disease caused by influenza A/H1N1, A/H3N2 or B viruses. A recent study revealed influenza attack rates of 18% in unvaccinated individuals [1], confirming that it constitutes a major public health problem. People within the so-called high-risk groups are more likely to develop severe disease and complications, leading to increased morbidity and mortality. Therefore, vaccination of these high-risk groups is recommended to reduce the seasonal influenza disease burden.

Influenza viruses are able to perpetuate in the human population by the continuous accumulation of mutations in the surface glycoproteins, hemagglutinin (HA) and neuraminidase (NA), which allow them to evade recognition by virus-neutralizing antibodies that are induced after each influenza virus infection. This continuous antigenic drift complicates the production of effective vaccines and necessitates regular updates of the vaccine composition to maintain their effectiveness. To ascertain the best possible recommendation for the vaccine composition each year, the WHO coordinates a worldwide surveillance system that monitors and characterizes circulating seasonal influenza viruses to identify the most likely virus candidates for use in the vaccine for the coming season.

In addition to infections caused by antigenic drift variants, antigenically distinct influenza viruses, often of a novel subtype and originating from animal reservoirs, are occasionally introduced into the human population. The seasonal influenza vaccines afford little or no protection against infection with these novel influenza viruses. These viruses may cause influenza pandemics when they acquire the ability to be transmitted efficiently from human to human. It is difficult to predict which subtype of influenza A virus will cause the next pandemic and when. The last four pandemics have been caused by influenza A viruses of the H1N1 (1918 and 2009), H2N2 (1957) and H3N2 (1968) subtypes. But in recent years also avian influenza viruses of various other subtypes, including H5N1 [2], H7N9 [3] and H10N8 [4], have caused severe disease in humans and posed a pandemic threat.

Seasonal versus universal vaccines

Attempts to produce vaccines against avian viruses have taught us an important lesson. Vaccines produced according to the recipe for seasonal influenza vaccines are poorly immunogenic and fail to induce protective antibody responses. These observations have spurred the development of adjuvants that can be used in (pre-) pandemic vaccines. The pandemic of 2009 taught us another lesson: it may take too long to prepare sufficient doses of a tailor-made vaccine in the face of an emerging pandemic. For example, in some countries, the vaccination campaign started after the peak of the pandemic. This is of course an unwanted scenario and clearly there is a need for universal influenza vaccines; vaccines capable of inducing protective immunity against not only intrasubtypic drift variants but also various subtypes of influenza A virus that have pandemic potential. The availability of such vaccines will buy time in a pandemic scenario and reduce morbidity and mortality until a tailor-made vaccine becomes available. In addition, technology that would allow pandemic vaccines to become available more rapidly is highly desirable.

Correlates of universal protection

Currently used influenza vaccines largely aim at the induction of virus neutralizing antibodies directed to the globular head domain of hemagglutinin molecules. However, this domain is highly variable and harbors mutations that define antigenic drift [5]. Furthermore, HA-head-specific antibodies fail to neutralize viruses of other subtypes. Obviously, an universal influenza vaccine should target more conserved proteins or regions. Furthermore, both humoral and cell-mediated arms of the immune system should be activated to confer broadly protective immunity. The current dogma is that only antibodies preventing infection of cells induce sterile immunity. Other correlates of protection will not confer sterile immunity and allow low-level virus replication and potential development of clinical signs.

M2-specific antibodies

One of the targets for protective antibodies is the M2 protein [6], which is a small membrane protein of influenza A viruses with a relatively conserved ectodomain. Upon natural infection, the protein is poorly immunogenic. However, after hyperimmunization of experimental animals or passive administration of M2-specific antibodies, protective immunity to challenge infection with influenza A viruses of various subtypes was achieved. Of interest, protection was dependent on the expression of Fcγ-receptors, which suggests that antibody dependent cellular cytotoxicity and/or phagocytosis played a role [7].

HA-stem-specific antibodies

The identification of antibodies specific for the HA-stem region with broad neutralizing activity has boosted research in this area enormously and renewed the hope for the development of universal influenza vaccines [8]. The HA-stem region is relatively conserved, and human antibodies have been identified that recognize epitopes shared by either group I subtypes of HA (including H1, H2, H5, H6 and H9), group II subtypes (H3, H4, H7, H10, H14 and H15) or even both. Upon natural infection with influenza viruses, the HA-stem-specific antibody response is subdominant compared with the response to the globular head region. The current challenge is to develop strategies that allow strong antibody responses to this less immunogenic region of the HA molecule. Several approaches have already been tested, including the use of ‘headless’ HA molecules, HA molecules with hyperglycosylated heads or sequential immunization with chimeric HA molecules, carrying head domains of various subtypes of influenza virus but with the same stem region to boost the antibody response to the latter [9]. All of these approaches were more or less successful in animal models, but the biggest challenge is maintaining the structure of the stem region.

The stem-specific antibodies do not neutralize the virus in the classical way. For example, they fail to prevent binding of virus to its receptor, which normally can be assessed by hemagglutination inhibition assays. Instead they probably exert neutralizing activity at a post-entry step, for example, by inhibiting fusion of the viral envelope with the endosomal membrane. Alternatively, antibody dependent cellular cytotoxicity and antibody-dependent phagocytosis may play a role as has been described for M2e-specific antibodies [7].

NA-specific antibodies

The other major influenza virus surface glycoprotein, NA, has been underappreciated as a vaccine immunogen. It has been shown that NA-specific antibodies contribute to protective immunity. Although the cross-reactivity between various NA subtypes of influenza A virus may be limited, substantial cross-reactivity within a subtype has been observed, which may be useful in case of an mismatch in HA subtype (e.g., H1N1 vs H5N1) [10,11].

Virus-specific T cells

In addition to protective antibodies to HA, NA or M2, cross-reactive virus-specific T cells correlate with protection. In animal models, it was shown that heterosubtypic immunity is induced by influenza virus infection and that this largely correlated with the induction of cross-reactive T cells [12]. The majority of virus-specific T cells, especially CD8⁺ cytotoxic T lymphocytes, is directed against relatively conserved proteins like the nucleoprotein and the matrix 1 protein (M1) [13]. Indeed, human CD8⁺ T cells induced after infection with seasonal influenza A viruses display considerable cross-reactivity with viruses of other subtypes, including the avian H5N1 and H7N9 viruses and the swine-origin pandemic H1N1 [14–16]. Thus, the induction of cross-reactive CD4⁺ and CD8⁺ T-cell responses seems a promising target for universal influenza vaccines. Indeed, during the pandemic of 2009, subjects with a high frequency of pre-existing virus-specific CD8⁺ T cells were less likely to develop severe disease [17]. Although T-cell-mediated immunity induced by either vaccination or natural infection may not induce sterile immunity, it could afford a significant degree of clinical protection against both seasonal and pandemic influenza viruses.

Universal vaccine delivery

For the efficient induction of virus-specific CD8⁺ T cells, vaccine delivery systems are required that allow de novo synthesis of viral proteins. Recombinant adenoviruses and poxviruses are promising vaccine candidates that could be used for this purpose. A candidate vaccine vector is modified vaccinia virus Ankara, originally developed as a safe smallpox vaccine with many favorable properties. It has been used to express influenza virus nucleoprotein, M1 and HA. These vaccine candidates have been successfully tested in several mouse, macaque and ferret vaccination-challenge experiments [18]. Furthermore, the first clinical studies with modified vaccinia virus Ankara candidate vaccines have been performed, and it was shown that this vector is immunogenic and elicited virus-specific CD8⁺ T-cell responses or H5-specific antibodies [19,20].

Conclusion

The research activities indicated above suggest that the development of a universal influenza vaccine is not science fiction, but may soon become reality. Clearly, correlates of protection have been defined other than the conventional antibodies to the head domain of HA, although it is not yet clear what the minimum requirements for protection are. The biggest challenge now is to find ways to efficiently induce those immune correlates of protection. Novel antigen delivery systems and vaccination strategies may be required for their efficient induction. One could think of the use viral vector systems, special adjuvants or use of chimeric proteins or synthetic proteins that are optimized for the induction of specific humoral or cell-mediated immune responses. These vaccines may not afford sterile immunity but may dampen the severity of disease caused by seasonal or pandemic influenza viruses. Tremendous efforts are being made at present to develop universal influenza vaccines and they may become reality, rather sooner than later.

Financial & competing interests disclosure

The authors receive financial support from EU grant FLUNIVAC (602604). RD de Vries receives financial support from the European Research Council grant FLUPLAN (250136). GF Rimmelzwaan is employed as a consultant to Viroclinics Biosciences B.V. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.