A Universal Influenza Vaccine: How Close Are We? | ASM.org (original) (raw)

This article was originally published on asm.org Aug. 23, 2019, but has since been updated by the author.

Seasonal influenza vaccinations currently provide narrow protection against select strains of the virus. There are now several "universal" flu vaccine candidates, using a variety of technologies, in Phase 2 and Phase 3 clinical trials that aim to provide broader and longer-lasting influenza protection. While much progress has been made to develop these vaccines, researchers still have to clear several more hurdles to improve vaccine efficacy.

Seasonal Influenza and the Need for Yearly Flu Vaccinations

There are 2 main reasons why we need seasonal influenza vaccinations:

1. Strains of influenza change annually.
2. Flu vaccine efficacy is narrow and short-lived.

The flu is primarily caused by the influenza A virus (IAV), but can also be caused by the influenza B virus (IBV). Both are enveloped RNA viruses, with IAV having several different strains. A study that analyzed patient data from Glasgow, United Kingdom from 2003 to 2013 found the prevalence of IAV and IBV to be 30% and 15%, respectively, in those with respiratory illness. Both influenza virus membranes contain proteins known as hemagglutinin (HA) and neuraminidase (NA), which are important for entry and release (respectively) of the virus from infected cells. Other structural components of the virus, such as the RNA-binding matrix protein M1, the nucleoprotein (NP) that coats the viral RNA or the ion channel M2 protein, can be recognized by our immune systems. While our immune system can mount a response to an influenza infection, it is usually specific to the strain of influenza causing infection because most neutralizing antibodies generated by either infection or vaccination target what is known as the globular head of HA. As shown in the graphic below, HA and NA are more numerous and accessible on the viral envelope and are therefore more accessible antibody targets than other components.

The viral membrane contains several membrane proteins, such as hemagglutinin (HA) and neuraminidase (NA). Source: CDC Seasonal Influenza Resource Center.

The reason we need yearly flu vaccinations is partially due to changes in the sequence of the HA protein. A person exposed to 1 strain of IAV will develop neutralizing antibodies specific to the HA globular head of that strain. Random mutations in IAV make the globular head of HA highly variable over time. This process is known as antigenic drift. Antibodies that recognize a previous strain will no longer protect against the new variant. Another mechanism of evasion by the virus is known as antigenic shift, or recombination. Recombination of 2 different strains of viruses in the same infected host can yield a completely new HA that has never been seen by our immune system. Such antigenic shifts have caused pandemic strains of influenza, such as the 2009 H1N1 outbreak. New seasonal vaccinations must be developed to provide protection against strains predicted to be common in the upcoming flu season. This is why the U.S. Centers for Disease Control and Prevention (CDC) analyzes the data on circulating strains to predict what strains to vaccinate against in the coming year. The predictive nature of vaccine design is one of the primary reasons why vaccine efficacy varies widely from year to year.

The other reason annual flu vaccination is necessary is because the antibody response to current flu vaccines is fleeting. A systemic review and meta-analysis of various influenza vaccination studies found that vaccine effectiveness waned 180 days post vaccination compared to 15-90 days post vaccination, suggesting a fading immune response within 6 months of vaccination. Due to both variation in the virus and a temporary immune response, most influenza vaccinations have short-lived efficacy and narrow protection. Efficacy, in this scenario, refers to broad protection against multiple strains of influenza.

Strategies for a Hemagglutinin (HA)-Based Universal Vaccine

The variability of the globular head of HA and the immune system’s preference for these epitopes has led to the search for more conserved epitopes across multiple influenza strains. If vaccines can focus the immune response against viral regions that undergo less mutation, there is a greater probability of protection on a near-universal level. Two of these strategies center on conserved regions found within HA: recombinant stalk-specific HA and chimeric recombinant HA.

The stalk, which is the domain of HA that anchors the globular head to the membrane of the virus, is relatively similar across IAV strains. This means that despite the vast number of different IAV strains, the stalk remains conserved. Conserved regions for viruses typically correspond to a preserved enzymatic activity, such as a polymerase or protease, or structural features that cannot be easily changed without deleterious effects. This is why the HA stalk is a target for universal vaccine candidates. However, recent research has highlighted the need to consider childhood and previous exposure to influenza and the ability to generate anti-stalk antibodies. The response of an individual to generating protective anti-stalk antibodies may be dependent on the influenza subtype that they were exposed to as a child.

One strategy to target the stalk involves a recombinant HA protein that lacks the globular head and contains only the stalk domain. To date, most vaccine development using this approach remains in the preclinical stage of research.

The second strategy for a universal vaccine uses reverse genetics to make viruses expressing recombinant, chimeric HA proteins. These constructs typically have the same stalk (the H1 clade of widely circulating IAV strains) fused with the globular head of non-human IAV strains. Sequential vaccine doses against these chimeric HAs that share the same stalk aim to generate stalk-specific antibodies that provide universal protection against IAV. Unfortunately, this approach suffers from the possibility that vaccine-generated stalk-specific antibodies may target regions inaccessible during an actual infection. Therefore, epitope mapping of the most immunogenic sites of HA stalk can help determine the availability of those sites during infection and aid in vaccine design.

One advantage of the chimeric HA approach is that it has the potential to protect against novel pandemic IAV strains. Since non-human influenza HA globular heads are used, our immune system will generate strain-specific antibodies to the head along with the conserved stalk domain. Therefore, if a pandemic strain ever expressed that same HA, vaccinated people would be protected from an otherwise potentially lethal infection.

GlaxoSmithKline (GSK) started clinical trials of 2 different chimeric HA-based vaccines, one of which is a collaborative vaccine between GSK, Icahn School of Medicine at Mount Sinai and Duke University. This candidate completed Phase I trials in May 2020, though its fate is unknown, as it is not listed in GSK’s development pipeline. Novavax’s Nanoflu is currently in Phase 3 trials, testing the efficacy of the vaccine in an older adult population. Nanoflu utilizes a quadrivalent approach, with recombinant HA from 4 IAV strains that have been predicted to circulate during the 2019-2020 season. Nanoflu successfully demonstrated efficacy in its clinical trials, demonstrating non-inferiority against the current seasonal vaccine (which is a major hurdle for Food and Drug Administration (FDA) approval), as well as generating comparable hemagglutination assay inhibition (HAI) against the 4 influenza strains. Additionally, Nanoflu had higher seroconversion rates when compared to the seasonal influenza vaccine.

Vaxart, Inc. is currently in Phase 2 clinical trials utilizing an adenovirus vector-based vaccine expressing the HA protein of H1N1. While not strictly a universal flu vaccine candidate, the VXA-A.1 vaccine is a proof-of-concept for the use of an oral tablet-based vaccine versus the standard intramuscular injection. Recently published data from the clinical trials suggests that the vaccine was well-tolerated and provided protection against homologous H1N1. The hope is that Vaxart can establish this platform as a jumping off point for a universal influenza vaccine.

VXA-A.1 is not the only adenovirus-vector HA influenza candidate under development. Altimmune has generated a nasal spray-administered vaccine composed of a replication-deficient adenovirus vector expressing an H1N1 HA. NasoVax completed Phase 2a clinical trials in 2019, revealing robust antibody protection against H1N1, as well as detectable increases in mucosal antibodies, suggesting induction of mucosal immunity. There is some trepidation about nasal sprays as influenza vaccine delivery systems, with evidence pointing to reduced efficacy with FluMist, an approved nasal influenza vaccine. NasoVax must clear that trepidation if it is to establish itself as an efficacious influenza vaccine.

Another promising candidate is a quadrivalent HA virus-like particle (VLP) vaccine from Medicago, Inc. This vaccine candidate completed Phase 3 trials in June 2020. This candidate is currently a proof-of-concept vaccine for the plant-based VLP technology, which uses plants to manufacture recombinant virus-like particles. These particles can be engineered to express HA proteins from influenza or spike (S) protein from coronaviruses, such as SARS-CoV-2. If this vaccine successfully protects against seasonal flu, Medicago could repurpose the vaccine using new HA antigens.

While many IAV researchers still believe that stalk-specific antibodies will ultimately be the most protective strategy, using this chimeric approach allows the immune system to develop antibodies against the non-human HA globular heads. Whether these will actually confer protection against novel IAV strains remains to be seen.

Recombinant M2 and Other Non-HA Vaccine Strategies

Other conserved regions in IAV include the exposed surface domain of the M2 structural protein. The M2 protein has not been a major target previously due to the lack of accessibility to antibodies. HA and NA-specific antibodies are more common than M2-specific antibodies in natural infections.

A front runner non-HA candidate is FLU-v, a synthetic peptide-based vaccine created by Imutex. This synthetic peptide has conserved sequences for M1, M2 and NP from IAV and IBV. FLU-v completed Phase 2 trials in 2019, and has since published data on the results of the trial. Unlike most of the vaccines listed, FLU-v was designed to promote cellular (T-cell) immune responses over humoral (antibody) immunity, and demonstrated successful protection against intranasal challenge with H1N1.

Bioinformatics analysis of conserved peptides across IAV strains offers another approach. BiondVax’s M-001, created in collaboration with the National Institute of Allergy and Infectious Disease (NIAID), is a vaccine created using this method and is undergoing clinical trials. M-001 completed Phase 3 in February 2020. M-001 is a linear polypeptide with 3 repetitions of 9 conserved sequences from M1, NP and HA from IAV and IBV influenza strains.

Osivax has a nucleoprotein nanoparticle-based vaccine known as OVX836 that has completed Phase 2 clinical trials. OVX836 induced CD4 and CD8 T-cell NP-specific responses in mice during preclinical studies, and similar results are expected from the clinical trials. Another vaccine candidate that is in Phase 2 is MVA-NP+M1, sponsored by Vaccitech. This vaccine combines NP and M1 from IAV in an adenoviral vector platform.

Ultimately, this is not an exhaustive list of current vaccine candidates, as there are numerous ongoing Phase 1 trials and several promising pre-clinical strategies for a universal influenza vaccine. One preclinical animal study utilized an H1N1 HA stem trimer that was stabilized and formulated into capsid-like particles. They were able to show protection in mice against heterologous challenge with a different strain of IAV 28 days post vaccination. Additionally, protection against homologous H1N1 was observed 34 weeks post vaccination. Another group utilized chimeric HA composed of an H5 strain globular head and an H1 stem. The monoglycosylated chimeric HA produced stem-specific antibodies in mice. When challenged with a panel of IAV strains, they found broad protection against the tested IAV strains.

Current universal influenza vaccine candidates in late clinical trials.

How "Universal" Are Universal Influenza Vaccines?

While many of the above strategies protect against a variety of IAV strains and some even target IBV, researchers do not yet know how protective these vaccines will be in people. For example, dosage effects of the chimeric HA vaccines differ in mouse model experiments compared to preliminary results from human trials. Mouse models can never fully capture the immunological history of people who experience both bouts of flu and receive influenza vaccinations. Therefore, human trials are critical to test vaccine performance in people.

There is also a concern regarding the length of protection conferred by a universal vaccine. Most seasonal influenza vaccinations only provide a short-term period of efficacy. While the efficacy of several universal vaccine candidates is being assessed, it appears that many of these vaccines may also only provide short-term efficacy. Thus, instead of being a one-time replacement to the seasonal vaccine, these new strategies may only serve to replace it with a more efficacious vaccine.

It is clear that while significant progress has been made to develop a broadly protective universal vaccine, there is still more work to be done to achieve a long-term solution to influenza. While influenza continues to strike year after year, promising work on broadly effective vaccines may ultimately break our never-ending cycle of annual influenza vaccinations.