The Flu Fighters

The influenza vaccine pipeline is diverse and crowded with candidates that have the potential to transform the field

By Regina McEnery

In what could be described as a 21st century Flu Rush, influenza researchers have been aggressively pursuing dozens of candidates over the past decade that could possibly work better, last longer, and be easier and quicker to manufacture than the current crop of nearly 30 licensed influenza vaccines.

The impetus for this surge in vaccine development is two-fold: to reduce the global impact of seasonal influenza, which results in an estimated three to five million severe illnesses and 25,000-50,000 deaths annually, and to dramatically improve emergency preparedness for the unpredictable nature and sometimes tsunami-like strength of pandemic influenza strains.

The hemagglutinin (HA) molecule that juts out from the surface of the influenza virus (see image, below) continually mutates, requiring vaccine developers to update the seasonal influenza vaccines annually to adapt the vaccine to the two to three dominant strains they predict will be in circulation that year. Therefore, finding better approaches is a global public health priority.

An Illustration of the Influenza Virion  

Shown on the surface are three proteins: hemagglutinin (in purple), neuraminidase (in yellow), and matrix protein 2 extracellular domain (in red).

Whether scientists will succeed in building a better flu trap remains to be seen, but there are certainly plenty of contestants vying for the prize. Eighty-five candidates are in various stages of preclinical or clinical development (see Figure 1, below). Ten of the vaccine candidates are considered universal vaccine candidates and are designed to protect more people against a broader range of influenza viruses for a longer period of time. More than half of those in all stages of development employ innovative strategies not yet available in any licensed seasonal or pandemic influenza vaccine. And at least 60 of the 85 candidates are in various stages of clinical development—about double the number of AIDS vaccine candidates in clinical testing.

Most of the candidates (72%) are in early stages of clinical development, so odds are many will fail. Still, vaccine manufacturers find the upsurge in research exciting. “Influenza used to be the sleepy backwater town of vaccine manufacturing,” says Alan Shaw, the chief scientific officer of VaxInnate, a New Jersey-based biotech whose seasonal virus-like particle (VLP) and universal vaccine candidates have both been tested in Phase II trials. “It’s really remarkable what is happening.”

Figure 1: The Flu Vaccine Pipeline  

Source: Biomedical Advanced Research and Development Authority

Frank Arnold, senior program manager of the influenza division at the Biomedical Advanced Research and Development Authority (BARDA), a branch of the US Department of Health and Human Services (HHS), agrees. He says while vaccine manufacturers have long recognized the need to advance the influenza vaccine technology, the swine flu H1N1 virus that appeared in 2009 and triggered a global pandemic really solidified why this goal is so important. “It took manufacturers six months to get the product out the door,” notes Arnold. “We missed the first two waves and if it was a 1918-like epidemic it would have been a disaster.”

Arnold’s reference to the novel 1918 avian influenza strain that sparked the mother-of-all influenza pandemics is every public health person’s worst nightmare—a lethal, fast-spreading H1N1 strain that claimed 50 million lives between 1918 and 1920.

When a novel, highly pathogenic avian strain of H5N1 was first detected in humans in 1997 in Hong Kong, some feared a repeat, but the virus, while claiming 300 of its 500 victims, wasn’t transmitted efficiently from person to person and in nearly every case appeared to have been transmitted through contact with infected poultry rather than through human contact. More worrisome though, says Arnold, would be if the circulating H5N1 strain, largely confined to a handful of countries in Asia, was able to genetically re-assort with a pandemic strain like H1N1 to form a new strain that is both swift and violent. “You would have massive issues,” he says.

This fear of being blindsided by a crippling pandemic drove HHS in 2004, mostly through the work of BARDA, to devote US$2.2 billion to the development of vaccines, antiviral drugs, diagnostics, respirators, and ventilators to prevent and treat pandemic influenza, with about $1.8 billion appropriated for vaccine candidates employing cell-based technologies, antigen-sparing adjuvants, and recombinant and molecular technologies. This financial windfall drove many large pharmaceutical companies and smaller biotechs to create or expand their influenza vaccine research and development programs and is the primary reason why the pipeline is now so active.

The global push to make influenza vaccination more affordable in developing countries is also contributing to the crowded pipeline. Developing countries can’t afford to buy flu vaccines, so they are taking their own candidates through the development process using the egg-based technology that is now used to make most seasonal flu vaccines. Arnold says the difference in cost per shot—$5 vs. 50 cents—says it all. “This is the cheapest and fastest route for the developing world.”

A universal solution

Not surprisingly, a universal vaccine candidate capable of providing protection against all influenza type A strains, which comprise 66% of all human strains, would be the ultimate achievement, though the approach, for now, is considered a long shot. At a World Health Organization meeting in November, researchers and public health leaders discussed the possibility of generating universal immunity against influenza, and the panel is expected to release a position paper soon summarizing how researchers are trying to achieve this goal.

Though not as variable a virus as HIV, influenza presents significant challenges for universal vaccination because of the slight but rapid antigenic shifts on the virus’ surface from season to season. Developers of universal vaccine candidates are attempting to overcome this obstacle by targeting the more conserved areas of the HA molecule or influenza’s other proteins. So far, this approach has provided mixed results.

One candidate, a recombinant fusion protein called VAX102 developed by VaxInnate, targets the matrix protein 2 extracellular domain (M2e), a short, stubby surface protein that is more conserved than HA across influenza subtypes, but which doesn’t induce much of an immune response on its own. VaxInnate’s approach genetically fuses the ectodomain located on the outer region of the M2e protein’s cell membrane to bacterial flagellin, the long, hair-like tails that enable bacteria to swim, and which interact with toll-like receptors (TLRs), a class of proteins that play a key role in innate immunity. A Phase I study in 60 adults showed that VAX102 alone induced a strong immune response to the M2e protein when it was joined with the TLR agonist (1), and when tested in combination with a trivalent seasonal flu vaccine it appeared it might enhance the immune responses to the standard vaccine, as well as add a secondary immunity to the M2e component (2). However, VaxInnate’s VAX102 program was put on hold over problems with the M2e protein sequence, putting the whole concept of developing this universal influenza vaccine candidate in doubt.

The University of Oxford has adopted another strategy with its universal vaccine candidate, a recombinant modified vaccinia Ankara (MVA) viral vector vaccine candidate containing sequences that code for matrix 1 (M1) and nucleoprotein (NP), internal proteins that are highly conserved and associated with cell-mediated immunity.

Sarah Gilbert, who heads up the project at Oxford’s Jenner Institute, recently tested the MVA-NP-M1 vaccine candidate in a dose-escalation Phase I study involving 28 men and women ages 18-50 from the UK and found the vaccine boosted interferon-γ secreting, antigen-specific CD8+ and CD4+ T cells (3). Gilbert also recently completed a Phase IIa study in 22 men and women from the UK, half of whom received the MVA vaccine candidate and half who received no vaccine. The volunteers were then quarantined for 7-10 days and challenged nasally with an A/Wisconsin strain of the H3N2 flu strain isolated in 2005 that is still in circulation. Gilbert described the results of the study in vague terms at the Influenza Congress USA conference Nov. 8-10 in Virginia, saying fewer vaccinees developed influenza or moderate-to-severe influenza symptoms compared to the controls.

Researchers are also attempting to build a universal influenza vaccine candidate by focusing on the stalk of the HA protein, rather than its mushroom-like head targeted by existing flu vaccines. Peter Palese, the Horace W. Goldsmith Professor and chair of the department of microbiology at Mt. Sinai School of Medicine in New York, is one of several researchers trying to generate cross-reactive stalk-specific antibodies to flu.

Palese demonstrated in mice that he could elicit the production of a handful of rare broadly neutralizing monoclonal antibodies against an array of H3N2 viruses responsible for most of the influenza morbidity and mortality in the last 43 years. His lab determined that the antibodies worked by inhibiting viral fusion, and identified the binding site of one of the monoclonal antibodies on the stalk of the HA as a continuous region that was 100% conserved between the H3 viruses used in the study (4).

In a separate experiment, Palese’s laboratory developed a method of chemically treating purified influenza virus to behead the HA protein. They then immunized mice with the truncated HA and found that the headless HA was sufficiently immunogenic to induce antibodies against multiple subtypes, and though it was not as protective as the full-length HA immunogen used in conventional vaccines, it protected against lethal challenge (5). “The next step is trying to get this vaccine into people,” says Palese.

However, it remains to be seen if it is possible to design immunogens that can elicit these stalk-specific antibodies, says John Treanor, head of the infectious diseases division at the University of Rochester Medical Center and a longtime influenza vaccine researcher who is not involved in the Palese project. “It’s absolutely possible to get the immune system to make those antibodies, but there are still lots of obstacles.”

Faster production

Another goal, now within closer reach, is being able to manufacture vaccines for seasonal and pandemic flu more quickly. Since the start of the Cold War, pharmaceutical manufacturers have used fertilized chicken eggs to grow live strains of three dominant influenza viruses, which are then inactivated and inserted into a vaccine.

Global health authorities and flu vaccine manufacturers have remained loyal to the egg because it is cheap and safe, and because the vaccines produced using this method have, for the most part, been moderately effective in preventing seasonal influenza. But the 8-10 months required to produce a typical trivalent seasonal vaccine using this approach can lead to delays in seasonal vaccine production. Even more problematic is when a single-strain or monovalent vaccine is suddenly needed to attack pandemic flu strains, as was the case in 2009 when the pandemic H1N1 virus surfaced in April of that year. The novel strain was first detected in Mexico and the US, and though the percentage of those who became ill or died from causes related to the H1N1 pandemic turned out to be far lower than the previous three pandemics—only 18,000 people are estimated to have died from it worldwide—it still raced around the globe. By the time a vaccine to combat the pandemic was ready for distribution, it was October and the second wave of cases was already peaking.

Global health leaders and flu manufacturers recognize the limitations of growing the seasonal flu vaccine in chicken eggs, just as they are aware that the seasonal shots, now available at so many different venues that even protestors at Zuccotti Park lined up to get them during the New York City Occupy Wall Street protest, have not been well studied in high-risk populations. This point was captured in a recently published meta-analysis of 31 influenza vaccine studies conducted over the past decade. The study noted that “evidence for consistent high-level protection is elusive for the present generation of vaccines, especially in individuals at risk of medical complications or those aged 65 years or older,” (6).

But researchers are optimistic about overcoming these challenges with new approaches. The next-generation influenza vaccine candidate closest to regulatory approval is a recombinant vaccine that is grown in insect cells. The trivalent vaccine candidate representing strains of H1, H3, and B influenza strains, known as FluBlok, is made by cloning HA genes from target flu viruses and splicing them into baculoviruses that are then used to infect ovary cells of caterpillars. Baculovirus is found on vegetables and infects some insects, but not mammals, birds, fish, or other species. This process allows the vaccine to be produced in a matter of weeks, says Daniel Adams, executive chairman and global head of business development at Protein Sciences, the developer of FluBlok.

A randomized placebo-controlled study of nearly 5,000 adults ages 18-49 conducted during the 2007-2008 influenza season found FluBlok was 45% effective in preventing culture-confirmed influenza meeting the US Centers for Disease Control and Prevention case definition for influenza-like illness, despite significant antigenic mismatch between the vaccine antigens and circulating viruses (7). The study was led by Treanor, who is also a medical advisor at Protein Sciences.

In a separate study of 601 healthy adults ages 50-65 comparing FluBlok with a trivalent seasonal influenza vaccine, FluBlok was found to be safe and immunogenic, but those who received FluBlok had a higher seroconversion rate with the H3N2 strain than those in the seasonal vaccine group (8). The company has also developed a pandemic influenza vaccine candidate called PanBlok against an H5N1 strain using the same baculovirus platform and tested it in Phase I and II trials.

Still, it’s been an uphill battle to licensure. Two years ago, an advisory panel declined to recommend to the US Food and Drug Administration (FDA) that Protein Sciences be granted a license for FluBlok because of insufficient safety data, which the company disputed. The FDA sided with Protein Sciences and is not requiring the company to conduct any additional larger trials or resubmit its application. Adams says Protein Sciences is hoping for FDA approval in early 2012.

Cell-based technology

Another non-egg-based approach that has garnered significant public and private investment is cell-based technology, which involves growing the influenza virus in mammalian cells, usually kidney cells. The virus is injected into the cells where it multiplies, then the cells’ outer walls are removed, harvested, purified, and inactivated.

Because it is not made inside a chicken egg, it is safe for people with egg allergies, but the entire production is only slightly shorter than egg-based methods and is not without risk. The amount of vaccine produced depends upon the strain—some grow faster than others—making it difficult to accurately predict how long it will take to meet inventory. Also, the manufacturing process requires expensive biosafety measures to ensure that the vaccine stock isn’t contaminated.

Regulators from Europe and India have licensed three different cell-based vaccines, and the FDA could be ruling soon on two others developed by Baxter Sciences and Novartis Vaccines & Diagnostics, the maker of the European flu vaccines.

The development of cell-based technology for influenza vaccination has been inching forward since the 1990s, but recent BARDA investments have greatly accelerated the process. Novartis, for instance, was awarded a $487 million, multi-year contract from HHS two years ago to construct the first facility in the US for cell-based seasonal and pandemic influenza vaccines. The contract required that Novartis pay 60% of the construction costs of the $1 billion North Carolina complex, which was completed this year and is designed to produce 150 million doses of monovalent vaccine or 50 million doses of trivalent vaccine within six months.

BARDA had similar partnerships with GlaxoSmithKline Biologicals, Sanofi Pasteur, MedImmune, and CSL Biotherapies to increase US pandemic vaccine manufacturing capacity. “We’re reaching a point where we can pick that fruit,” says Robert Huebner, BARDA’s deputy director of the influenza division.

1. Vaccine, 29, 5145, 2011
2. PLoS One 5, e14442, 2010
3. Clin Infect Dis. 52, 1, 2011
4. PLoS Pathog. 6, e1000796, 2010
5. mBio 1, e00018-10, 2010
6. Lancet Infec. Dis. 2011, doi: 10.1016/S1473-3099(11)70295-X
7. Vaccine 29, 7733, 2011
8. Vaccine 29, 2272, 2011