Influenza viruses are cunning. as soon as we develop immunity to one strain, another strain comes along and attempts to infect us. It's a seasonal battle and--in the case of pandemic strains--a global threat.
The influenza vaccines that are currently available provide protection, but not for all of us, and when it comes to the next pandemic flu, not for any of us.
Influenza viruses belong to the Orthomyxoviridae family. Three types, distinguished by the virus's antigenic proteins, infect humans. Types A and B are widespread and cause clinical disease, whereas type C is rare.
Unlike type B, which is primarily a childhood pathogen that mostly causes mild disease, type A has a well-established track record of undergoing major genetic changes that result in the global pandemics that occur several times each century. The most recent ones were the Asian influenza pandemic in 1957 and the Hong Kong influenza pandemic in 1968, both of which caused significant morbidity and mortality globally, according to the World Health Organization (WHO). The Spanish flu in 1918–19 is thought to have infected at least 200 million people throughout the world and killed around 40 million people.
"The Spanish flu pandemic occurred when the world population was 1.7 billion," observes infectious disease epidemiologist Michael T. Osterholm, a professor at the University of Minnesota, Twin Cities. "Today, we have a population of 6.4 billion and a much greater opportunity for rapid transmission around the world due to modern transportation systems. New flu strains emerge, evolve, and replace old strains continuously, making influenza a complex and dangerous disease. The question is not if, but when, the next pandemic will occur."
The greatest pandemic threat comes from type A strains that infect animals, such as pigs and birds, and can transfer to humans. One type, known as H5N1, currently circulating in birds in Asia, is highly pathogenic to humans. Waterfowl, particularly wild ducks, are a natural reservoir for avian influenza virus strains such as H5N1. Since January 2004, the avian virus has resulted in 42 human deaths: one in Cambodia, 12 in Thailand, and 29 in Vietnam, according to WHO.
"We now have a situation where the avian virus H5N1 is binding to human cells, albeit inefficiently," says virologist John M. Wood of the National Institute for Biological Standards & Control (NIBSC), in the U.K. "The virus is so pathogenic that, once it gets into cells and replicates, it kills over 70% of the people diagnosed with the infection. If the virus acquires the ability to easily transmit from human to human, we have the potential of a global devastation that could eclipse the Spanish flu pandemic. It could be catastrophic."
INFLUENZA VIRUSES are defined by two spike-like antigenic proteins--hemagglutinin (HA) and neuraminidase (NA)--on the surface of the virus particle. The spikes protrude from a lipoprotein envelope that encloses a single-stranded RNA genome comprising eight separate ribonucleoprotein segments.
"The hemagglutinin spikes are responsible for attaching the virus to our respiratory epithelial cells," Wood explains. "The neuraminidase spikes help the release of the progeny virions from the cells so that the virus can infect other cells."
The 16 known subtypes of HA (H1–H16) and nine subtypes of NA (N1–N9) occur in many different combinations, according to Wood. "Of those, we just have H1N1 and H3N2 circulating in humans," he says. "In the past, we also had H2N2, but this subtype has now disappeared."
As influenza viruses circulate, minor changes in the antigenic proteins occur by a process known as "antigenic drift." "This phenomenon allows the virus to at least partially escape the human immune responses primed by vaccination or exposure to earlier versions of circulating viruses," notes Anthony S. Fauci, director of the National Institutes of Health's National Institute of Allergy & Infectious Diseases (NIAID). The institute is the lead agency in the U.S. for conducting research on all infectious diseases, including influenza.
The segmental nature of the influenza virus genome also contributes to the high immunological variability of the HA and NA proteins. When two different influenza viruses infect the same cell, the viruses can exchange genes. This process, known as reassortment, is particularly dangerous when avian and human strains recombine.
"If the virus that has jumped species or the new reassorted virus evolves to be efficiently transmitted between people, a deadly influenza pandemic can result," Fauci says. "As the population acquires immunity to the new strain over the next several years, the pandemic strain fades into the routine background of circulating viruses."
The emergence of a major pandemic strain with new antigenic proteins is known as antigenic shift. The 1918, 1957, and 1968 pandemics were caused by the evolution of H1N1, H2N2, and H3N2 strains, respectively.
THE CONSTANTLY CHANGING antigenic nature of the influenza virus poses a significant challenge to vaccine manufacturers and agencies engaged in predicting suitable strains for the vaccines.
In February and September of each year, the WHO Global Influenza Surveillance Network assesses the epidemiological behavior of currently circulating virus strains and selects the three strains that are likely to be most virulent in the forthcoming influenza seasons starting in November–December and May–June in the Northern and Southern Hemispheres, respectively. The network, a partnership of 112 national influenza centers in 83 countries, is responsible for monitoring the influenza viruses circulating in humans and for identifying new strains.
The three strains are then matched in the vaccines that are produced for the next flu seasons. Flu vaccines for the current season in the Northern Hemisphere and the forthcoming 2005 season in the Southern Hemisphere both contain antigens from influenza A subtypes H3N2 and H1N1 and one influenza B virus. The H1N1 and B strains are the same for both hemispheres, whereas the H3N2 strains are different.
Earlier this month, WHO recommended that vaccines to be used in the Northern Hemisphere for the 2005–06 influenza season should contain an A/New Caledonia/20/99(H1N1)-like virus; an A/California/7/2004(H3N2)-like virus; and a B/Shanghai/361/2002-like virus.
The process of determining which strains are to be selected for vaccine production in the U.S. is undertaken by WHO, the Food & Drug Administration (FDA), and the Centers for Disease Control & Prevention (CDC). CDC is also one of four WHO Collaborating Centres for Reference & Research on Influenza, the others being in Australia, Japan, and the U.K.
According to WHO, more than 250 million doses of influenza vaccine containing the WHO-recommended strains are produced annually. Most influenza vaccines are grown in fertilized chicken eggs. Seeds of the three strains are injected separately into the eggs. In the U.S., CDC provides new strains of the seed virus to FDA, which then distributes the three seed viruses to manufacturers.
"The WHO strains are not selected on the basis of their growth properties," Wood tells C&EN. "If a WHO-recommended strain doesn't grow well, it can be reassorted with a laboratory virus that does grow well in eggs. The reassorted virus has the same surface proteins as the recommended strain and internal components of the laboratory strain that govern the high growth."
After injection, the virus is allowed to grow in the egg for several days, after which the egg white is separated and the virus is harvested. The live virus is then killed, usually with formaldehyde or ß-propiolactone.
The inactivated virus particles are treated with a detergent to produce split vaccines that contain the antigens HA and NA and internal protein subunits. Most subunit vaccines are further purified, leaving only HA and NA. "The virions are split and further purified to lower the incidence of adverse reactions on injection," Wood explains.
Finally, the three individual monovalent pools of vaccines are blended to form the trivalent vaccine.
INACTIVATED FLU vaccine prevents influenza in 70–90% of healthy adults who receive the vaccine. However, among the elderly, the level of protection falls significantly. More than 90% of influenza-related deaths occur in people over 65 years old. Children under five years old and women in the first and third trimesters of pregnancy are also at higher risk from infection. WHO notes that annual influenza epidemics, although difficult to assess, are thought to result in between 3 million and 5 million cases of severe illness and between 250,000 and 500,000 deaths every year around the world. It adds that antiviral drugs for influenza are an important adjunct to flu vaccine for the treatment and prevention of the disease. Neuraminidase inhibitors, such as zanamivir and oseltamivir, are particularly effective, but they are relatively expensive compared with vaccines and not available for use in many countries, according to WHO.
Three companies--Chiron, Sanofi Pasteur (formerly Aventis Pasteur), and MedImmune--manufacture influenza vaccines for the U.S. market. Chiron's manufacturing license was suspended for three months last October because of possible microbial contamination of Fluviron, a trivalent inactivated vaccine manufactured at its plant in Liverpool, England. The suspension was renewed for another three months starting last month. Sanofi Pasteur produces a trivalent inactivated influenza vaccine called Fluzone.
The MedImmune product, FluMist, is produced in chicken eggs, but unlike the Chiron and Sanofi Pasteur products, it contains live influenza A and B viruses that are still capable of replication. The viruses are genetically attenuated, or weakened, to stimulate immunity without causing influenza. The vaccine is administered as a fine mist by a sprayer into the nose. The product, which was approved by FDA in June 2003, is "the first innovation in flu vaccine technology in over 50 years," MedImmune says.
"Live, attenuated influenza vaccines give a better immune response, and you need to give a much lower dose," remarks Wendy S. Barclay, reader in virology at the University of Reading, in England. "The difficulty I have with live vaccines is the genetic instability of the influenza virus. The vaccines can be genetically engineered to be safe, but the influenza virus has the capacity to mutate a lot. So how can we tell that it's not going to acquire back some of the virulence traits that were engineered away from it and become a new virus?
"With inactivated vaccines, on the other hand, you inject proteins into somebody and stimulate their antibodies to recognize these foreign proteins," she continues. "As there is no RNA present, the vaccine is incapable of replication and of infecting people."
Jamie P. Lacey, director of media and public relations at MedImmune, points out that MedImmune has found no evidence that the vaccine strains used in FluMist mutate. "The strains in the vaccine appear to retain their attenuation," she says.
Fauci remarks that although egg-based technology in general has served us well for more than 40 years, it is cumbersome and relatively inflexible with regard to the ability to surge up production once the process has begun. He points out that the virus strains to be used in the vaccines have to be forecast at least six months in advance of the influenza season, and the manufacturing process is lengthy. In addition, hundreds of millions of fertilized chicken eggs are needed each year to manufacture the vaccine.
THE PROBLEM of having eggs available on a year-round basis is already solved, says James Matthews, director of external research and development at Sanofi Pasteur, Shiftwater, Pa. He points to an agreement that the Department of Health & Human Services (HHS) has with the company to make flocks and eggs available 365 days a year. The real issue, Matthews says, is manufacturing capacity.
"The decisions about which viral strains to include in the vaccine may not always be correct, but the long lead time required to acquire the eggs for vaccine production makes midcourse corrective action virtually impossible," Fauci says. "Additionally, some people are allergic to eggs and therefore cannot receive the classic vaccine. Also, some influenza viruses do not grow well in chicken eggs and may be virulent for the eggs, a circumstance that may result in delays bringing the vaccine to market and a possible decrease in the total number of doses available."
"There is a finite amount of vaccine that can be produced from any one facility," he says. "That capacity now is dedicated to manufacturing regular, seasonal influenza vaccine. One viable way to boost capacity is to increase vaccination rates during the regular influenza season. In turn, companies would invest in new manufacturing facilities. In the event of a pandemic, there would be adequate capacity as the plants were converted to making pandemic vaccine.
"There is also another challenge," Matthews continues. "We cannot stockpile trivalent vaccines, because strains change each year, although it is rare that all three strains change in a given year. But we can produce some of the strain that is unlikely to change from the previous year to get a head start on next year's production. As we now have eggs to match our capacity that are available 365 days of the year, we are able to make vaccines" all year.
He points out that egg-based technology is well established and that further developments that significantly improve its efficiency are always possible but unlikely. "We've pretty much maximized what can be done with the egg platform," he says. "Even so, we continue to make efforts to upgrade and streamline the process by, for example, finding better ways to prepare the eggs, to move them through the facility, to incubate them, and to harvest them."
To overcome some of the limitations of egg-based technology, a number of companies are now developing mammalian-cell-culture technologies for producing inactivated influenza vaccines. "Cell culture is a more scalable system and can be manipulated much more readily to accommodate enormous fluctuations in demands," comments virologist Maria Zambon at the U.K.'s Health Protection Agency, in London. "Eggs have a long lead time, and variations in demand cannot easily be accommodated, as was evident in the vaccine supply crisis to the U.S. market this winter."
Cell-culture preparation of influenza vaccines works on the same principle as the egg-based technology. "Instead of injecting the virus into an egg, you put the virus in contact with the cells, and it grows in the cells," Wood says.
Influenza vaccine manufacturers are exploring the use of a variety of cell lines. Netherlands-based Solvay Pharmaceuticals, for example, has developed a method for producing vaccines using a dog kidney cell line known as MDCK (Madin Darby canine kidney). It recently announced that it had obtained marketing approval from the Dutch regulatory authorities for the method. The company is constructing a large-scale facility and hopes to produce the vaccine from the cell line in time for the winter 2005–06 vaccination season. The company also manufactures an egg-based influenza vaccine, Influvac, which is sold in more than 35 countries.
Chiron and ID Biomedical are also developing influenza vaccines produced from MDCK cells.
Baxter International has developed a product, known as PreFluCel, which is prepared using the Vero (African green monkey kidney) cell line. However, in December 2004, the company announced that it had voluntarily suspended enrollment in the Phase II–III clinical study in Europe. "Based on the preliminary data we've seen, the rate of fever and associated symptoms observed with the current formulation of PreFluCel is higher compared with other vaccines on the market," says Norbert G. Riedel, Baxter's chief scientific officer.
The Dutch biotechnology company Crucell has a strategic agreement with Sanofi Pasteur to develop and commercialize influenza vaccine products based on Crucell's PER C6 cell-line technology. "The technology uses a human embryonic retinal cell line," Matthews explains. "The cells are highly susceptible to all the different strains of influenza viruses. We've had projects over a period of about 10 years trying to identify better cell lines. PER C6 is the best that we've found. At the moment, it's in preclinical development. We hope to be in the clinic by the end of this year with a Phase I study."
The use of cell culture to produce human influenza viruses has limitations, observes Manon M. J. Cox, chief operating officer at Meriden, Conn.-based vaccine company Protein Sciences. "The process still requires the production of a high-yield reassortment virus, this time not an egg-adapted but a mammalian-cell-line-adjusted reassortment virus," Cox notes. "This process may introduce cell-line-specific mutations in the genes that can lead to the selection of variants characterized by antigenic and structural changes in the HA protein, potentially resulting in less efficacious vaccines."
Protein Sciences has used recombinant DNA technology to develop a trivalent vaccine, known as FluBlØk, that contains three recombinant HA proteins corresponding to WHO-recommended virus strains. The HA antigens are grown in cell culture using immortal insect cell lines developed by the company from the ovaries of Spodoptera frugiperda, commonly known as the fall armyworm moth. "With this technology, you make the insect cells synthesize the HA protein simply on its own, rather than as part of the virus," Barclay explains.
FluBlØk is currently undergoing a Phase II–III proof of principle/field study in 480 healthy adults to firmly establish the final commercial dose. Market introduction is expected in 2007.
"Unlike the licensed vaccines and many cell-culture vaccines in development, no live influenza vaccines, biocontainment facilities, or harsh chemicals such as formaldehyde are used in manufacturing," Cox observes. The vaccine consists solely of the three antigenic proteins stored in sterile buffered saltwater.
The company has also used the technology to produce a subunit vaccine containing the HA antigen of the avian H5N1 virus. Cox points out that, with this technology, new vaccines can be developed quickly and safely to address late-appearing influenza viruses and emerging natural or man-made pandemic viruses.
In a separate development, PowderMed, a therapeutic DNA vaccine company based in Oxford, England, announced last November the results of a Phase I clinical trial of a DNA influenza vaccine. At the maximum dose, 100% of the subjects achieved a seroprotective level of antibodies, demonstrating that the vaccine is a candidate for further trials, the company says.
THE VACCINE is administered using a particle-mediated epidermal delivery device known as PowderJect, developed by the company. With the device, DNA is precipitated onto microscopic gold particles that are then propelled by pressurized helium gas at high speed into immunologically active antigen-presenting cells of the epidermis. Once inside the nuclei of the cells, the DNA elutes off the gold and becomes transcriptionally active, producing the encoded protein that leads to immune responses.
"I am convinced that a PowderMed DNA flu vaccine could become a viable solution for the current threat from pandemic flu," says Clive Dix, chief executive officer of PowderMed. He points out that bulk DNA for threatening influenza strains and formulated DNA on gold in cassettes can be stockpiled. "Biohazards are absent from the manufacturing process, and we have the ability to make enough DNA in less than one month to vaccinate 500,000 individuals," he adds.
A process known as reverse genetics is also being extensively explored to improve the yields and enhance the safety of viruses used for vaccines. The technique enables exact copies of the eight RNA segments of the viral genome and four other viral proteins that are required for replication to be generated. The genetically engineered virus can then be grown in chicken eggs or mammalian cells.
"With this technique, we can now clone, mutate, and rescue almost any influenza virus, including an unmodified wild-type virus or a reassortment in which the genome segments originate from different viruses," Wood explains.
The technique is particularly useful for developing pandemic vaccines. Wood and NIBSC colleague James S. Robertson note that most conventional vaccine viruses are reassortments that are not virulent in humans [Nat. Rev. Microbiol., 2, 842 (2004)]. Attempts to produce large quantities of vaccine from a highly pathogenic avian virus, such as H5N1, in conventional vaccine production facilities would be disastrous, they say. The virus would kill chicken embryos and therefore lower the yield and quality of the vaccine.
"More seriously, production staff would be immunologically naive to an avian virus and potentially susceptible to infection," they write. "Furthermore, wild and domestic animals, including birds, in the vicinity of the production facility would be at considerable risk of infection from such a lethal virus."
Safe production of pandemic vaccines, therefore, requires either high levels of biocontainment or the development of a safe vaccine virus.
Wood, Robertson, and coworkers have used reverse genetics to generate "depathogenized" vaccine strains of H5N1 viruses that are suitable for use in vaccine production. "We have the technology and the containment facilities at NIBSC to make H5N1 vaccine strains by reverse genetics," Wood says. "An H5N1 vaccine is being made now which will hopefully go into Phase I clinical trials later this year."
Sanofi Pasteur is one of several manufacturers that are working with WHO to produce clinical lots of an egg-based vaccine using the genetically engineered attenuated strain of the H5N1 virus provided by NIBSC.
In May of last year, NIAID awarded contracts to Chiron and Sanofi Pasteur to produce 10,000 and 8,000 investigational doses, respectively, of vaccines from a seed of the H5N1 virus provided by NIAID. The seed was a depathogenized virus developed by reverse genetics using a virus taken from a Vietnamese patient in February 2004. In addition, HHS contracted Sanofi Pasteur to produce up to 2 million doses of the vaccine to be stockpiled.
"We took the seed that we were given, which had been attenuated, and used it to make a conventional vaccine in eggs," Matthews says. "We have prepared the vaccine, and it's ready for shipment immediately for clinical trials."
Osterholm observes that no one knows how fast H5N1 will evolve. "There is no guarantee that the vaccines now in development will be effective against the exact strain that emerges to infect humans in large numbers," he says.
Matthews remarks that cell-culture technology offers the best potential for creating the next generation of faster and more efficient methods for manufacturing seasonal and pandemic influenza vaccines.
"But, at the same time, Sanofi Pasteur believes that egg-based production is currently the best and most reliable way to produce large quantities of influenza vaccine," he says. "Given the hurdles facing cell culture, it will be many years before cell culture can take its place as a mainstream manufacturing method for influenza vaccine production."