The swine flu pandemic has thrown the importance of vaccines into the spotlight. Dr Sarah Houlton reviews the challenges faced by vaccine producers
Swine flu has thrown the importance of flu vaccination into sharp focus. Although, as yet, the symptoms of this form of flu are less severe than normal seasonal flu, this may change over time, and if it does it will become an entirely more dangerous proposition. Vaccination does not prevent all cases of flu, but it does substantially reduce the risk, and pharma companies are working on creating a vaccine against this new strain.
The new flu virus first appeared in April, and the WHO raised the pandemic threat level of H1N1 to 6 on 11 June; the different stages of pandemic are explained in Table 1. In the past, declaring a global pandemic has been associated with a jump in pathogenicity of the virus; this is not the case this time, and the decision is based purely on the geographic spread of the virus.
The virus took everyone by surprise when it first emerged, as did the fact that it continued to circulate widely during the summer season when there are normally few cases of flu. However, the plans that have been made in the past few years because of the threat of H5N1 avian flu mutating into a form that is transmissible to humans meant that vaccine manufacturers were already getting up to speed with new manufacturing technologies and facilities.
The flu virus is a moving target as it constantly mutates to form new strains. This antigenic drift means that a new preventative vaccine needs to be developed every year to provide protection against the strains that are most likely to be circulating in the coming season. Current seasonal flu vaccines are trivalent, with antigens against two forms of influenza A, and one influenza B. Influenza A drifts much more rapidly than the B form, and the current season's vaccine contains an H3N2 virus first isolated in Brisbane in 2007, and an H1N1, also first isolated in Brisbane in the same year, plus a B strain.
The makeup of the seasonal flu vaccine is determined by the WHO's Influenza Surveillance Network, and announced nine to 12 months before the flu season to allow manufacturers sufficient time to make vaccines. More than 250 million doses of seasonal flu vaccine are currently produced annually, and it is estimated that it saves more than 8 million lives.
The H and N codings refer to the haemag-glutinin and neuramidinase glycoproteins on the surface of the viral envelope. Influenza is an RNA virus and contains eight pieces of RNA that code for 11 different proteins, including H and N. These eight genes are prone to reassortment, which causes an antigenic shift, creating a virus that is substantially different from those that had previously been circulating. The new H1N1 swine flu virus contains two genes from North American avian flu, one from human H3N2, three from classic swine flu and two from Eurasian swine flu.
There were three pandemics in the 20th century: in 1918, 1957 and 1968. The 1918 pandemic happened before the influenza virus had been discovered, but it has since been shown that it was H1N1. This continued to circulate within the population, and was still doing so when the first flu virus was isolated in 1933. The 1957 pandemic resulted from an antigenic shift in the circulating flu virus to H2. "This was probably of avian origin," explains Peter Palese, of the department of microbiology at Mount Sinai School of Medicine in New York. "Those who had previously been infected with the H1 form of the virus had no immunity to H2, resulting in the pandemic." The 1968 pandemic happened when the H3 form appeared. Then in 1977, H1 re-emerged, and it is this and the H3 form that are circulating in the population now.
While the current swine flu is, like the 1918 pandemic, a form of H1N1, there is an important difference - it does not express the virulence gene PB1-F2 that made the 1918 virus so dangerous. This gene was also present in the 1957 and 1968 pandemic viruses. "The virus could acquire this gene by mutating with something that expresses it, but I think it is not very likely," Palese says. "It would take several mutations to get it."
pressure on capacity
The emergence of this new pandemic strain of flu has put a great deal of pressure on vaccine manufacturers. Vaccines take months to manufacture, and the capacity is limited - and that capacity is already being used to make the standard seasonal flu vaccine. The vaccines are traditionally made in eggs via a very labour- intensive process, and only one dose is made per egg. Cell culture is beginning to be used to make flu vaccines, which will be much more efficient in the long run, but capacity remains limited and so vaccines will continue to be produced using eggs for some time to come.
Cell culture is carried out in closed bioreactors, which also reduces the chances of cross-contamination. The vaccines do not need additives such as thimerosal or antibiotics, and they are suitable for people with egg allergies. Typically, the cells for the culture are first propagated in stainless steel fermenters before the virus strain is added. This process takes several days, with the virus being released from the cells into the medium. To purify the virus, it is first separated from the cell debris, and then captured and the medium removed. It then needs to be inactivated - influenza vaccine uses inactivated virus, which is one of the reasons it is so safe - and then split to separate the fractions of the viral surface proteins that are used in the vaccine. For seasonal flu vaccine, this process needs to be carried out three times, making antigens for each of the different strains of virus it contains.
Another advantage of cell culture vaccines is that there is no risk of egg adaptive mutations, where the virus mutates while it is in the egg. "With some years" strains, it is impossible to grow a suitable matched strain for the vaccine in eggs, but it can be grown in cells," claims Theodore Tsai of Novartis Vaccines in the US.
However, it is not just vaccine capacity that is an issue - it is the ability to grow sufficient antigen. Current estimates are that there is enough to make maybe a billion doses of H1N1 vaccine, but if the vaccine has to be delivered in two doses to give full protection - this is not known yet - then only 500 million people can be vaccinated. If adjuvants are used in the vaccine, then the antigen could go much further, but the US FDA is currently refusing to licence flu vaccines that use adjuvants, despite the fact that they have been used safely in many millions of patients elsewhere in the world.
Adjuvants are chemicals - generally aluminium or surfactant - that boost the activity of a vaccine. "Traditional vaccines are given without adjuvants because they boost the immune response in subjects who have been previously primed, whether by infection or vaccine," explains Giuseppe del Giudici, global head of translational medicine at Novartis Vaccines & Diagnostics. "They are very efficacious in immunocompetent adults who need them less, but suboptimal in those who need them more, notably children and immunocompromised adults. Adjuvants increase the potential of vaccines."
The first use of adjuvants was published in 1961, he says. These were mineral oils, but they were not used for safety reasons as they caused abscesses, although they did increase the effectiveness of the vaccine. Since then, various vaccine manufacturers have developed safer oil-in-water adjuvants, some of which have been used extensively.
Novartis" MF59 is a combination of squalene, polysorbate and sorbitan trioleate, and has been used commercially in Novartis" seasonal flu trivalent vaccine Fluad since the 1990s. Not only is it antigen sparing but the oil-in-water adjuvant has been shown to help elicit broad cross-reactive immune responses against a wide range of seasonal influenza strains, including some that are not contained within the trivalent seasonal flu vaccine. It also gives a strong immune memory and sustained antibody response in both seasonal and pre-pandemic vaccines, which can help the immune system produce a protective response that can be boosted several years after an initial vaccination.
While squalene-based adjuvants were claimed to be a potential cause of Gulf War syndrome, the WHO has determined that this was not the case, and the continuing refusal of US Food & Drug Administration (FDA) to approve adjuvanted flu vaccines is a cause of frustration. "It's not clear what the FDA wants to see!" del Giudici says. "MF59 has had 45 million real-use exposures, 111,000 of which were in clinical trials."
Other companies have also developed adjuvants for use in vaccines. GlaxoSmithKline's AS03, for example, contains squalene, DL-a-tocopherol and polysorbate, and Sanofi Pasteur's AF03 is an emulsion that contains squalene. More than 45,000 subjects have now been given GSK's adjuvanted flu vaccine. Studies on Prepandrix, the company's pre-pandemic H5N1 vaccine, showed that by adding the adjuvant to the vaccine, a higher immune response was stimulated using a smaller amount of antigen. This antigen-sparing property of adjuvants will be essential if a global vaccination programme is to be feasible.
unresolved questions
There are other issues besides capacity that have to be addressed in getting vaccine out to the general population. For example, ampoules or prefillable syringes need to be sourced in huge quantity, and there's the issue of distribution. And, of course, there are questions about who should get the vaccine. "Which groups should be immunised? How much vaccine is needed? Which groups should be immunised first, and in what order should other priority groups follow?" asks David Salisbury, chair of WHO's Strategic Advisory Group of Experts on Immunisation. "Should the 2010 seasonal flu vaccine be trivalent or tetravalent [including the new variant of H1N1]? And what will the impact be of H1N1 vaccine manufacture on the 2010 seasonal flu vaccine?"
Clearly, top of the list are key workers such as those in healthcare who are more likely to be exposed to the flu virus and whose absence from work would be problematic, and also groups with underlying medical conditions who are at most risk of complications. These include asthmatics, people with heart disease and diabetes, and pregnant women. Then, according to Julia Walsh of the University of California at Berkeley, US, the elderly and small children should be a priority. "The highest proportion of all influenza related deaths occurs in the over-65s," she says. "They are three times more likely to be hospitalised, and those over 85 are 16 times more likely to die than those aged 65-70."
H1N1 vaccines are now being developed by all the manufacturers - Novartis, GSK, Sanofi Pasteur, CSL and Medimmune - and clinical trials are being carried out to ensure their safety. The vaccine should start to become available this month, with doses continuing to be delivered until about January. The number of patients who can be immunised depends on whether the FDA accepts adjuvanted vaccines, and also the results of the clinical trials, which will indicate whether one or two doses are required to confer immunity.
EMEA and FDA have both required clinical trials of the new vaccine to give two doses 21 days apart, with the subject's immune response being tested before the second dose is given. It is thus possible that the second dose may not be needed, which would, at a stroke, double the number of patients who can be immunised. The US does have the power to allow an unlicensed product to be used under its emergency use authorisation procedure, so it may be that adjuvanted vaccines will be used there without full approval. The pandemic authorisation for vaccines in the EU has already been triggered.
The decision has already been taken not to combine pandemic vaccine with seasonal flu vaccine, for this season at least, because the new H1N1 virus appeared too late, and as the required antigen dose for H1N1 was unknown it would have delayed the production of the standard seasonal vaccine. It may be that in future the seasonal flu vaccine becomes quadrivalent.
Capacity remains an issue, however, and if there is insufficient vaccine for western countries, what about developing ones? While some companies have said they will give some doses of the pandemic vaccine free to the developing world, Novartis will not. "We will definitely reduce the price for the developing world, but we will not give it completely free," says Andrin Oswald, chief executive of its Vaccines & Diagnostics division.
"We have made a loss on flu vaccine over the past three years - there's no money in it! There has been an increase in capacity, but only because governments have supported it with orders for vaccine. But if WHO or non-governmental organisations want to finance a discount or a lower price, we are happy to work with them."