|Home | About | Journals | Submit | Contact Us | Français|
Reflecting on the 2009 H1N1 pandemic, we summarize lessons regarding influenza vaccines that can be applied in the future. The two major challenges to vaccination during the 2009 H1N1 pandemic were timing and availability of vaccine. Vaccines were, however, well-tolerated and immunogenic, with inactivated vaccines containing 15μg of HA generally inducing antibody titers ≥1:40 in adults within 2 weeks of the administration of a single dose. Moreover, the use of oil-in-water adjuvants in Europe permitted dose- reduction, with vaccines containing as little as 3.75 or 7.5μg HA being immunogenic. Case-control studies demonstrated that monovalent 2009 H1N1 vaccines were effective in preventing infection with the 2009 H1N1 virus, but preliminary data suggests that it is important for individuals to be re-immunized annually.
Influenza viruses are enveloped RNA viruses belonging to the Orthomyxoviridae family. There are three types of influenza viruses that cause disease in people: A, B and C. Type A viruses infect a wide range of hosts and pose the most significant public health risk. They are further subtyped based on the antigenicity of the hemagglutinin (HA) and neuraminidase (NA) surface proteins. Sixteen HA and nine NA subtypes have been identified in the natural hosts, waterfowl (order Anseriformes) and shorebirds (order Charadriformes) and several subtypes have transmitted to domestic poultry and mammalian species including pigs, horses, and humans [1,2].
The genome is comprised of 8 segments of negative-stranded RNA that encode at least 11 proteins. Reassortment of gene segments between influenza viruses is a frequent occurrence in nature. This may lead to the emergence of a novel subtype of influenza capable of infecting humans. There have been three influenza pandemics in the 20th century. The first, caused by an H1N1 virus, occurred in 1918. The virus was maintained in humans until reassortment events led to the emergence of an H2N2 pandemic virus in 1957, which circulated until it was replaced by an H3N2 virus in 1968 [1,2]. In 1977, H1N1 viruses re-emerged in humans and co-circulated with H3N2 viruses, causing seasonal influenza epidemics until 2009, when a new H1N1 virus emerged (2009 H1N1) (reviewed by ). In addition to these pandemics, people have also been infected with H5, H7 and H9 viruses. Since 2003, almost 500 cases of human infection with highly pathogenic avian influenza (HPAI) H5N1 viruses have been reported, mostly due to direct transmission from poultry to humans. Although there has not yet been sustained transmission from person to person, there is concern that continued transmission from birds to humans may increase the probability of an H5N1 virus emerging with the capacity to transmit efficiently.
Vaccination remains the primary strategy for the prevention and control of influenza [4,5] and both inactivated and live attenuated vaccines are licensed. Vaccine production begins with the generation of vaccine reference strains, reassortant viruses with the HA and NA derived from the vaccine target, combined with internal protein genes from a strain which, for inactivated vaccines, specifies high yield in eggs (A/Puerto Rico/8/34; PR8)  and, for live attenuated influenza vaccines (LAIV), confers temperature sensitivity, cold-adaptation (ca) and attenuation that limits viral replication to the cooler, upper respiratory tract (A/Ann Arbor/6/60 ca; A/Leningrad/47/57 ca) .
We are now in the post-pandemic period following the 2009 H1N1 virus outbreaks. Reflecting on the pandemic, this review summarizes lessons regarding influenza vaccines that can be applied in the future.
The 2009 pandemic H1N1 virus first emerged in Mexico and California in April and quickly spread globally. On June 11, 2009, a pandemic was declared by the World Health Organization . The virus was antigenically unrelated to previously circulating seasonal H1N1 viruses, and molecular studies revealed that it had gene segments from several viruses that were circulating in pigs: the North American H3N2 triple-reassortant and the Eurasian ‘avian-like’ swine H1N1 viruses [9–11].
The majority of cases occurred in younger age-groups, with the median age estimated to be 12–17 years in Canada, USA, Chile, Japan and the UK . In most individuals, infection led to a mild, self-limiting, upper respiratory tract illness, but 2–5% of cases required hospitalization  and pregnant women were at a higher risk for severe disease [13–15]. The overall case-fatality rate was 0.15–0.25%, and deaths mostly occurred in middle-aged adults (median age 40–50 years) . Deaths amongst hospitalized patients mostly occurred in those with underlying health conditions, but nearly one third who died were previously healthy .
National authorities identified the following high-priority groups for vaccination: health care workers, pregnant women, young children, and individuals with underlying cardiovascular or respiratory medical conditions, autoimmune disorders and diabetes. Production of seasonal influenza vaccine for the 2009–2010 season was, however, already under way (reviewed in ) and the decision was made to produce a separate monovalent pandemic H1N1 vaccine.
A major challenge to pandemic vaccine production was timing. In spite of a rapid response, vaccines were neither available for widespread use during the 2009 winter season in the Southern Hemisphere , nor until after the autumn/winter wave had nearly peaked in the Northern Hemisphere  (Figure 1). Another challenge was availability. As of June 2009, the global capacity for seasonal vaccine production was only 876 million doses a year, with 7 manufacturers providing 64% of this capacity . In addition, pandemic vaccine virus yields in eggs and cell culture were lower than expected (reviewed by [16,18]), the optimal quantity of HA per dose was unknown, nor was it clear whether adjuvants should be utilized for dose sparing, as had been demonstrated for H5N1 vaccines (reviewed in ). It was also assumed that two doses of vaccine would be needed to induce protection, as clinical trials conducted in 1977 indicated that two doses of vaccine were needed to successfully immunize naive individuals against a novel virus [20–24].
Since then, new manufactures have emerged in China, India, Thailand and South America and there are now 26 manufacturers making pandemic 2009 H1N1 vaccines  using various platforms in many countries, but challenges remain for vaccinating people living in under-resourced countries that lack the infrastructure for mass vaccination.
All registered pandemic H1N1 vaccines were evaluated for safety and immunogenicity in clinical trials and met criteria required for licensure of seasonal influenza vaccines by regulatory authorities, including the Committee for Medicinal Products for Human Use (CHMP) in Europe and the Food and Drug Administration (FDA) in the United States . This includes an assessment of the proportion of participants with either seroconversion or significant increase in antibody titer, the fold-increase in geometric mean antibody titer (GMT) and the proportion of participants with antibody titers of 1: 40 or greater, which is associated with a 50% reduction in the risk of illness in a susceptible adult population . Typically hemagglutination inhibition (HAI) antibody titers are determined as a surrogate for neutralizing antibodies.
The results of monovalent 2009 H1N1 vaccine trials are summarized in Table 1. A single dose of unadjuvanted inactivated split-virion 2009 H1N1 vaccine containing 15μg HA led to antibody titers ≥1:40 in 95–98% of healthy adults [27–30]. In children, one study found that 92.5% of 6 month-9 year olds had antibody titers ≥1:40 after a single dose of 15μg HA in an inactivated split-virion unadjuvanted 2009 H1N1 monovalent vaccine , however another study found that only 47–50% of 6–35 month-olds and 65–75% of 3–9 year olds achieved these titers . After two doses, however, 90–100% children had titers ≥1:40 in both studies [31,32] (Table 1). Therefore, a single dose of monovalent 2009 H1N1 vaccine was recommended in adults, but young children were recommended to receive 2 doses (reviewed by ). It is likely that a single dose was sufficient to induce immunity in adults because prior exposure to seasonal H1N1 viruses had immunologically primed the population. This hypothesis is supported by studies in a mouse model of infection, where prior infection with seasonal influenza virus or seasonal LAIV primed mice for a robust response to a single dose of pandemic LAIV while unprimed mice were only partially protected against 2009 H1N1 infection with a single dose of pandemic LAIV .
The use of oil-in-water adjuvants, MF59 (Focetria, Novartis Vaccines, Italy) and AS03 (GlaxoSmithKline (GSK) Biologicals, Dresden, Germany) in 2009 H1N1 vaccines was approved by the European Medicines Agency (EMEA). The adjuvants had previously been approved in Europe for use in seasonal influenza vaccines and adjuvanted inactivated H5N1 vaccines had enabled dose sparing and induced broad cross-clade humoral responses [34,35]. Their mechanism of action remains incompletely understood, however they may amplify immune responses by enhancing antigen presentation and recruiting inflammatory cells to the area of antigen deposition (reviewed by ).
Adjuvanted monovalent 2009 H1N1 vaccines permitted dose sparing (Table 1). Antibody titers ≥1:40 were induced in 92% of adults given a single dose of MF59-adjuvanted vaccine  and 94–96% given the AS03-adjuvanted vaccine each containing only 3.75μg HA [38,39]. In children, antibody titers ≥1:40 were induced in 100% of 6–35 month olds given 2 doses of 7.5μg or 3.75μg HA [40,41] and 99.3% of 6 month-13 year olds given 1.875μg HA in adjuvanted vaccine . Moreover, in the UK, adjuvanted split-virion vaccines induced seroconversion and antibody titers ≥1:40 in a larger percentage of adults and children than non-adjuvanted whole virion vaccines [38,42]. Adjuvanted influenza vaccines are, however, not licensed in the USA . Fortunately, adjuvants were not necessary to achieve adequate immunity to the 2009 H1N1 virus, though they may prove to be important in the future.
Live attenuated influenza vaccine (LAIV) viruses were also licensed for use against the 2009 H1N1 virus, however, it was necessary to introduce point mutations in the HA to optimize the yield in eggs . Intranasal vaccination with 2 doses of 107 fluorescent focus units of virus, 28 days apart, led to antibody titers ≥1:32 in 26.4% children 2–17 years old and 13.5% of adults 18–49 years old (Table 1) . While these responses were modest, HAI titers are not predictive of protective efficacy of LAIV and vaccine efficacy is seen in the absence of seroconversion [46,47]. Determining meaningful immune correlates of protection for LAIV is an active area of research.
The safety of the monovalent 2009 H1N1 vaccines was assessed in clinical trials (Table 1). All vaccines were well-tolerated and elicited only mild or moderate transient side-effects, such as local pain, swelling and redness at the injection site, temporary fever, headache, fatigue or myalgia. Moreover, in the USA, data from the Vaccine Adverse Event Reporting System (VAERS) indicated that the adverse event profile was consistent with that of seasonal influenza vaccines and Guillian-Barre syndrome, anaphylaxis and death were rare . However, there are reports of narcolepsy being a rare adverse event following vaccination with an AS03-adjuvanted 2009 H1N1 influenza vaccine, Pandemrix [49–51]. At the time of writing, this is under investigation .
In the 2010–2011 influenza season, trivalent vaccines incorporated an A/California/07/2009 H1N1-like virus, an A/Perth/16/2009 (H3N2)-like virus and a B/Brisbane/60/2008-like virus. While the majority of these vaccines are well tolerated, in Australia, an increased risk of febrile convulsions following immunization with an inactivated, split-virion, unadjuvanted vaccine, Fluvax, led to the suspension of seasonal influenza immunization of healthy children in April 2010. The cause of this association remains unknown, and an investigation is underway. A similar vaccine was utilized in the UK, until the Department of Health advised against its use in younger children and enhanced surveillance for febrile convulsions. Other vaccine products were, however, not associated with an increased risk of febrile convulsions .
Robust immunogenicity does not always result in greater vaccine effectiveness (VE) . This may be, in part, because the 1:40 HAI titer used to measure immunogenicity is extrapolated to all influenza viruses. It is, therefore, important to estimate the effectiveness of vaccines at the population level. Published effectiveness data from 2009–2010 is available from studies conducted in Europe, UK, Canada, China and Korea (Table 2), with VEs ranging from 66.0% to 93.0% [17,54–57]. These studies demonstrate that the monovalent 2009 H1N1 vaccines were effective in preventing infection with 2009 H1N1 virus, in contrast to the 2009–2010 trivalent vaccines, which did not contain the pandemic H1N1 vaccine antigen and showed no evidence of effectiveness against 2009 H1N1 infection, with an adjusted VE of only 9.9% (95% CI, −65.2–50.9%) ([17,54–57]).
Vaccine efficacy studies for the 2010/2011 trivalent influenza vaccines based on mid-season data analysis (Table 3) show evidence of some protection against confirmed 2009 H1N1, however the VEs are lower than the monovalent vaccines administered in 2009/10 (44.1 to 52%) [58–60]. One study also reported that if individuals were vaccinated with monovalent vaccine in the 2009/2010 season, but not with seasonal vaccine in 2010/2011, the VE against current circulating H1N1 viruses was 34%, compared to 46% if individuals were only vaccinated in the 2010/11 season, and 63% if individuals were vaccinated with both monovalent vaccine in the 2009/2010 and the trivalent vaccine in 2010/11 . This suggests that the longevity of protection is poor and that repeat vaccination boosted protective immune responses, reinforcing the recommendation for annual re-immunization.
The two major challenges to vaccination during the 2009 H1N1 pandemic were timing and availability of vaccine. The vaccines were well tolerated and surprisingly immunogenic, with inactivated vaccines containing 15μg of HA generally inducing antibody titers ≥1:40 in adults within 2 weeks of the administration of a single dose. This suggests that exposure to H1N1 viruses ‘primed’ the majority of the population, such that only one dose of 2009 H1N1 vaccine was needed for protection. Moreover, the use of an oil-in-water adjuvant permitted dose-reduction, with vaccines containing as little as 3.75 or 7.5μg HA being immunogenic. These factors enabled more vaccine doses to be distributed than was first expected. In future pandemics, we may not be so lucky, and priorities have been set for accelerating vaccine production, including a shift towards cell-based vaccines (Reviewed by ). Additionally, the use of adjuvanted vaccines may become more widespread in the future. Case-control studies demonstrated that monovalent 2009 H1N1 vaccines were effective in preventing infection with the 2009 H1N1 virus, but preliminary data suggests that it is important for individuals to be re-immunized annually.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.