Influenza: The Quest for a Universal Vaccine

ACCORDING TO THE World Health Organization (WHO), worldwide annual influenza (flu) epidemics are estimated to result in three million to five million cases of severe illness and 250,000 to 500,000 deaths. WHO continuously monitors the flu viruses circulating in humans and twice a year updates the composition of flu vaccines that targets the three most representative virus types in circulation — two subtypes of influenza A virus and one influenza B virus — for the development of trivalent vaccines, as well as a second influenza B virus to support quadrivalent vaccine development.¹ However, because of antigenic drift, current flu vaccines only contain antibodies for circulating viruses and can only prevent some illnesses and deaths, not all.

But researchers believe they have found the key to eradicating seasonal and pandemic flu. That key is a new vaccine design that can protect against all flu viruses, both circulating viruses and mutated viruses that have not yet occurred. This next generation of flu vaccines is being touted as the precursor to a universal flu vaccine and is expected to dramatically reduce the number of cases of illness and the high morbidity rates that result from the flu.

More than 40 universal influenza vaccine designs are currently in development.² Some of the more promising research is focusing on how to elicit an immune response that targets the whole hemagglutinin (HA) protein (found on the surface of influenza viruses), just the hemagglutinin head (where flu viruses mutate constantly) or just the hemagglutinin stalk (where viruses do not mutate as often).

The COBRA Vaccine

An academic research lab and a vaccine manufacturer are working together to design a synthetic flu vaccine that targets the whole hemagglutinin protein, both the head and stalk, by looking at the entire sequence holistically.

It all started in 2012, when Ted Ross, PhD, director of the Center for Vaccines and Immunology at the University of Georgia, began collaborating with Sanofi Pasteur on a new type of broadly protective influenza vaccine (BPIV) called computationally optimized broadly reactive antigen, or COBRA. The COBRA method focuses on neutralizing epitopes (the part of an antigen that is recognized by the immune system) of the hemagglutinin, building off of the current standard of care.

Between 2011 and 2013, two studies demonstrated that a novel hemagglutinin for H5N1 influenza, derived using the COBRA method, elicited a broad antibody response against H5N1 isolates from different clades in mice and nonhuman primates.³ The researchers then conducted a joint study in which they designed nine independent COBRA hemagglutinin genes to elicit antihemagglutinin antibodies directed at the head domain of the H1N1 hemagglutinin protein.² The antibodies were assessed as vaccines used alone, in cocktails or in prime-boost combinations. The most effective regimens elicited the broadest hemagglutinin-inhibition response against a panel of H1N1 viruses isolated over the past 100 years, even against viruses whose sequences were not included in the design strategy. The study represents the first demonstration of a COBRA-based hemagglutinin vaccine strategy that elicits a broadly reactive response against both seasonal and pandemic H1N1 isolates.³

According to study results, the monoclonal antibody binds to conformational epitopes in the conserved stem domain of H1N1 hemagglutinin proteins. In contrast, the two hemagglutinin head-specific monoclonal antibodies bound to some, but not all, COBRA hemagglutinin proteins. Overall, the pattern of monoclonal antibody binding indicates that each of the COBRA hemagglutinin proteins has differently exposed epitopes and, therefore, may have different antigenic properties. Even so, a COBRA hemagglutinin vaccine has the potential to recall a broader repertoire of memory responses to protect against more antigenic H1N1 variants.²

According to Harold Kleanthous, PhD, head of research (North America) and associate vice president at Sanofi Pasteur, the differences between the yearly flu shot and the COBRA vaccine is that the “annual seasonal influenza vaccines are preselected and matched to circulating viruses. Sometimes, though, viruses drift prior to rollout of these vaccines, and they are not as well-matched, leading to reduced vaccine effectiveness.” When an antigenic shift occurs, making it even less possible to predict the circulation of a new virus in humans, explains Dr. Kleanthous, today’s vaccine strategies are not able to reliably protect against the strains that emerge, as was the case with the introduction of the A/California H1N1 strain in 2009. This, he says, is “the advantage of a synthetic approach that is designed to represent many more circulating viruses. The COBRA vaccine … is able to represent several influenza viruses within a subtype — influenza A H1, H3 and both influenza B viruses — isolated over several years, preferentially displaying to the immune system only the most important domains of influenza viruses, that have the potential to protect against strains yet to emerge into circulation. [It offers] greater breadth of protection, and when coupled with suitable immune stimulants, may eliminate the need for an annual flu shot.”

Another benefit of the COBRA vaccine, says Dr. Ross, is that it can be tested using the U.S. Food and Drug Administration’s approved hemagglutinin-inhibition assay, which detects vaccine-elicited antibodies that block the receptor binding sites on the hemagglutinin head to bind to the hemagglutinin receptor on human and animal cells (sialic acid). These epitopes are found on the globular head. “Other universal vaccine designs, of which there are many and some that do not focus on the hemagglutinin at all, do not have this benefit,” he says. “An assay with correlation to human protection may have to be established.”

The Chimeric Vaccine

Researchers at two North American labs are working together to develop a new vaccine design that targets only the stalk of the hemagglutinin protein, which is highly conserved and is not prone to mutation. The vaccine focuses only on influenza A viruses because they are the ones that cause pandemics.

Normally, upon exposure to an influenza A virus, humans generate high quantities of strain-specific antibodies that bind to the head domain of hemagglutinin — the viral protein that mediates attachment to human cells, which is extremely variable among different strains of flu and is prone to mutation. Thus, the immunity these antibodies provide is very short-lived. Because yearly flu vaccines seek to elicit antibodies that bind to the head domain of hemagglutinin, they must be reformulated annually to keep pace with the rapidly mutating virus, and they do not provide protection against the emergence of new pandemics.

However, during the 2009 H1N1 swine flu pandemic, the Palese Laboratory at the Icahn School of Medicine at Mount Sinai in New York, along with several other labs around the world, discovered infected humans generate unusually high amounts of antibodies that target the highly conserved stalk domain of the hemagglutinin protein. Unlike the strain-specific antibodies that humans normally generate, these stalk-binding antibodies have the ability to neutralize diverse strains and subtypes of influenza A virus. According to Matthew Miller, PhD, assistant professor in the department of biochemistry and biomedical sciences at McMaster University, “It was quickly recognized that a vaccine capable of generating similarly high levels of hemagglutinin stalk-binding antibodies might then be capable of providing universal protection against influenza A viruses.”

Researchers at Dr. Miller’s lab, as well as at other institutions, demonstrated that these broadly neutralizing antibodies were boosted most efficiently anytime the immune system thought it was being exposed to a pandemic-like strain of flu. So, they developed chimeric hemagglutinin proteins that are capable of tricking the immune system into thinking it has been exposed to a pandemic virus and thereby efficiently boosts levels of broadly neutralizing antibodies in animal models.

“In the context of this chimeric vaccine, we refer to the hemagglutinin proteins as ‘chimeric’ because they are, in fact, a ‘chimera’ of two proteins,” explains Dr. Miller. “Basically, we take the stalk domain from one influenza virus subtype (an H1, for instance) and then fuse it to the head domain of a different influenza virus subtype (an H5, for example). The nomenclature we then use to describe this chimeric hemagglutinin protein would be ‘cH5/1’ where ‘c’ equals chimeric, ‘H’ equals hemagglutinin, ‘5’ equals head domain of H5 virus and ‘1’ equals stalk domain of H1 virus.”

Extensive preclinical work from the Palese Laboratory demonstrated that vaccines containing these chimeric hemagglutinin molecules are able to provide broad, or universal, protection against influenza A virus infection in animals and are now being developed clinically. However, work is also being conducted to target influenza B viruses using chimeric hemagglutinin, so they are working to define the optimal types and specificities of antibodies capable of generating “universal immunity” to optimize their chimeric vaccine design.

“An important benefit of this chimeric vaccine platform relative to other vaccines that are being developed, such as the headless hemagglutinin vaccine or viral-vectored vaccines that target M2e [influenza matrix protein 2], is that these chimeric hemagglutinin molecules retain functionality,” says Dr. Miller. “As a result, they can be incorporated into the existing influenza virus vaccine platforms, either inactivated vaccine or live-attenuated vaccine. This is not true of many other platforms which are subunit-based or rely on viral-vectored delivery of synthetic genes. Such platforms will need to undergo much more rigorous preclinical testing in animals, and Phase I clinical trials must be completed to test safety and immunogenicity of the vaccine in humans.”

Next Steps

Dr. Kleanthous says the University of Georgia’s and Sanofi Pasteur’s intent is to move the most promising approaches for a BPIV into the clinic as expeditiously as possible after all preclinical proof-of-concept criteria have been met: “We expect to take an iterative approach to early clinical studies because of the complex history of influenza exposure and pre-existing immune response all humans have.”

According to Dr. Miller, the next steps for the chimeric vaccine in the next couple of years is the completion of the preclinical and clinical trials: “We also need to establish a correlate of protection for the broadly neutralizing antibodies elicited by this universal vaccine, since we still do not know what levels of these antibodies will be required to protect humans from infection.”