tetano
Editor, Senior Moderator
For an Influenza Vaccine, Are Two Bs Better Than One?
Lindsey R. Baden, M.D.
December 11, 2013DOI: 10.1056/NEJMe1315317
Share:
Article
Every year, influenza causes epidemic disease. Periodically, antigenic drift and shift enable influenza viruses to evade human immunity from prior seasons, increasing the severity of the associated illnesses. Since the 1918 pandemic, circulating influenza A viruses have had several antigenic shifts, resulting, for example, in the emergence of the A/H2N2 (1957?1958) and A/H3N2 (1968?1969) strains, but influenza more often drifts with modest antigenic changes.1,2 A recent analysis has shown that a few specific point mutations near the hemagglutinin-receptor binding site of A/H3N2 and both circulating B lineages (B/Yamagata and B/Victoria) allow evasion of immunity.2 A central part of the strategy to prevent disease caused by influenza has been annual vaccination with a trivalent influenza vaccine that contains A/H1N1 and A/H3N2 plus a single B antigen, with the antigens determined on the basis of the likely, or ?best guess,? influenza strains that will circulate later in the year. In the Northern Hemisphere, the seed vaccine strains are typically selected in the first quarter of the year for the upcoming influenza season, given the time that it has traditionally taken to make the seasonal egg-based influenza vaccine. Developing the vaccine on the basis of the best-guess strains incurs the risk that the vaccine will not ?match? the influenza strains that will actually circulate.3-5
Influenza B typically causes fewer cases of disease each year than influenza A, with virologic surveillance indicating that 1 to 46% of the circulating viruses each season are B influenza viruses. Over the past 25 years, two main influenza B lineages have been circulating and causing seasonal illness in humans: B/Victoria and B/Yamagata.3 Notably, over the past 12 years, these two B lineages have cocirculated, and the primary circulating lineage varies from season to season. Each year, the annual trivalent influenza vaccine has incorporated only the single B-lineage antigen that is predicted to circulate that season. This choice has led to the B-lineage vaccine being mismatched to the actual circulating B-lineage virus half the time.3,4 A potential solution to minimize the chance of a B-lineage mismatch is to include both B lineages in the annual influenza vaccine, as has recently been suggested by the Advisory Committee on Immunization Practices of the Centers for Disease Control and Prevention and by the World Health Organization.3,4
It is difficult to assess the potential benefit of adding a second B lineage to the seasonal vaccine, since this benefit depends on several factors, including which B lineage is the dominant circulating B lineage in a given season. For example, in 2007?2008, there was a mismatch when the selected B antigen in the trivalent vaccine was against the B/Victoria lineage, whereas 98% of the circulating B strains were B/Yamagata, and about 29% of the circulating viruses that season were B viruses. It has been estimated that if the B/Yamagata antigen had been included in the 2007?2008 vaccine, there might have been 440 fewer influenza-associated illnesses, as well as 4.13 fewer hospitalizations and 0.22 fewer deaths related to influenza-associated illness, per 100,000 population.3
Jain et al.6 now report in the Journal on the efficacy of a quadrivalent influenza vaccine as compared with a control (hepatitis A) vaccine in 5168 children, 3 to 8 years of age, in eight countries. The quadrivalent vaccine that was studied had the same three antigens as the routine trivalent vaccine (A/H1N1, A/H3N2, and one B lineage) plus an additional fourth antigen ? thus containing both B/Yamagata and B/Victoria lineages. A total of 563 influenza-like illnesses occurred in the quadrivalent-vaccine group and 657 in the control group; 96% of these influenza-like illnesses were assessed for influenza infection by means of real-time polymerase chain reaction, and 62 children (2.4%) in the quadrivalent-vaccine group and 148 (5.7%) in the control group had confirmed influenza infection. The vaccine efficacy was estimated to be 59%, with evidence supporting an additional benefit ? a decrease in the incidence of moderate-to-severe illness among recipients of the quadrivalent vaccine. It is notable that 71% of the children in the control group with influenza-like illness did not have influenza identified as the cause of the illness. This has important implications for the development of improved diagnostics, as well as research in vaccine development.
The efficacy of the inclusion of the second B lineage in the vaccine cannot be assessed because only two cases of illness associated with the B/Yamagata-lineage influenza strain were identified in this study (both of which occurred in the control group in the Philippines). Thus, efficacy against infection and disease caused by the B/Yamagata strain was not tested. However, a hemagglutination-inhibition antibody (HAI) titer of at least 1:40 has typically been accepted as a marker of protection and has been used by the Food and Drug Administration for approval of the annual influenza vaccines. The data from this study and previous research show that HAI titers of 1:40 or more for both the B/Yamagata and B/Victoria lineages are elicited in most vaccinated children without evidence of interference from the increased vaccine valency.7-10 Given the additional B strain in the quadrivalent vaccine as compared with trivalent vaccines, there is concern that the increased vaccine antigen content (60 μg vs. 45 μg of hemagglutinin antigen) may lead to increased side effects; therefore, careful safety monitoring is warranted. To date, no safety signal has been observed.4,7-10 The effect that initial vaccination with both B lineages in young children will have on immunologic priming and subsequent boosting of the immune responses to both B lineages remains to be seen.
Novel influenza strains continue to emerge and cause severe disease in humans, as seen earlier this year in China with influenza A/H7N9.11 A vaccine is likely to be important to limit human illness from this pathogen if sustained human-to-human transmission develops. Over the 6 months since this pathogen was identified, a candidate vaccine antigen has been selected, manufactured, and clinically tested.12 Fries et al.12 recently showed in the Journal that two doses of a viruslike particle (VLP) vaccine (made with the use of cell-based recombinant baculovirus technology) that contained the influenza A/H7N9 antigen alone (15 μg or 45 μg) had limited immunogenicity (6 to 16% of the participants with HAI titers ≥1:40). However, when a saponin-based adjuvant with a lower amount of hemagglutinin antigen (5 μg or 15 μg) was used, the responses were improved (HAI titers ≥1:40 in 37 to 81% of the participants). No direct efficacy data are available, but the antibody titers elicited are encouraging, though it is unclear whether the protection usually associated with an HAI titer of 1:40 or higher applies to protection from A/H7N9. In addition, novel adjuvants raise important safety considerations that can be properly assessed only with larger clinical trials.
Constant reevaluation to identify the influenza strains that are the likeliest threats is important in determining the most effective strategy for preventing disease from influenza. Rapid development and wide use of vaccines with antigens that match the circulating strains are important for an effective response. Ongoing careful safety assessments of any new therapy, including these vaccines, is essential. In the United States, a quadrivalent influenza vaccine with both B-lineage antigens has been introduced this season as an alternative to the traditional trivalent vaccine. Over the next few influenza seasons we hope to see the value of the two Bs.
Disclosure forms provided by the author are available with the full text of this article at NEJM.org.
This article was published on December 11, 2013, at NEJM.org.
http://www.nejm.org/doi/full/10.1056/NEJMe1315317?af=R&rss=currentIssue&
Lindsey R. Baden, M.D.
December 11, 2013DOI: 10.1056/NEJMe1315317
Share:
Article
Every year, influenza causes epidemic disease. Periodically, antigenic drift and shift enable influenza viruses to evade human immunity from prior seasons, increasing the severity of the associated illnesses. Since the 1918 pandemic, circulating influenza A viruses have had several antigenic shifts, resulting, for example, in the emergence of the A/H2N2 (1957?1958) and A/H3N2 (1968?1969) strains, but influenza more often drifts with modest antigenic changes.1,2 A recent analysis has shown that a few specific point mutations near the hemagglutinin-receptor binding site of A/H3N2 and both circulating B lineages (B/Yamagata and B/Victoria) allow evasion of immunity.2 A central part of the strategy to prevent disease caused by influenza has been annual vaccination with a trivalent influenza vaccine that contains A/H1N1 and A/H3N2 plus a single B antigen, with the antigens determined on the basis of the likely, or ?best guess,? influenza strains that will circulate later in the year. In the Northern Hemisphere, the seed vaccine strains are typically selected in the first quarter of the year for the upcoming influenza season, given the time that it has traditionally taken to make the seasonal egg-based influenza vaccine. Developing the vaccine on the basis of the best-guess strains incurs the risk that the vaccine will not ?match? the influenza strains that will actually circulate.3-5
Influenza B typically causes fewer cases of disease each year than influenza A, with virologic surveillance indicating that 1 to 46% of the circulating viruses each season are B influenza viruses. Over the past 25 years, two main influenza B lineages have been circulating and causing seasonal illness in humans: B/Victoria and B/Yamagata.3 Notably, over the past 12 years, these two B lineages have cocirculated, and the primary circulating lineage varies from season to season. Each year, the annual trivalent influenza vaccine has incorporated only the single B-lineage antigen that is predicted to circulate that season. This choice has led to the B-lineage vaccine being mismatched to the actual circulating B-lineage virus half the time.3,4 A potential solution to minimize the chance of a B-lineage mismatch is to include both B lineages in the annual influenza vaccine, as has recently been suggested by the Advisory Committee on Immunization Practices of the Centers for Disease Control and Prevention and by the World Health Organization.3,4
It is difficult to assess the potential benefit of adding a second B lineage to the seasonal vaccine, since this benefit depends on several factors, including which B lineage is the dominant circulating B lineage in a given season. For example, in 2007?2008, there was a mismatch when the selected B antigen in the trivalent vaccine was against the B/Victoria lineage, whereas 98% of the circulating B strains were B/Yamagata, and about 29% of the circulating viruses that season were B viruses. It has been estimated that if the B/Yamagata antigen had been included in the 2007?2008 vaccine, there might have been 440 fewer influenza-associated illnesses, as well as 4.13 fewer hospitalizations and 0.22 fewer deaths related to influenza-associated illness, per 100,000 population.3
Jain et al.6 now report in the Journal on the efficacy of a quadrivalent influenza vaccine as compared with a control (hepatitis A) vaccine in 5168 children, 3 to 8 years of age, in eight countries. The quadrivalent vaccine that was studied had the same three antigens as the routine trivalent vaccine (A/H1N1, A/H3N2, and one B lineage) plus an additional fourth antigen ? thus containing both B/Yamagata and B/Victoria lineages. A total of 563 influenza-like illnesses occurred in the quadrivalent-vaccine group and 657 in the control group; 96% of these influenza-like illnesses were assessed for influenza infection by means of real-time polymerase chain reaction, and 62 children (2.4%) in the quadrivalent-vaccine group and 148 (5.7%) in the control group had confirmed influenza infection. The vaccine efficacy was estimated to be 59%, with evidence supporting an additional benefit ? a decrease in the incidence of moderate-to-severe illness among recipients of the quadrivalent vaccine. It is notable that 71% of the children in the control group with influenza-like illness did not have influenza identified as the cause of the illness. This has important implications for the development of improved diagnostics, as well as research in vaccine development.
The efficacy of the inclusion of the second B lineage in the vaccine cannot be assessed because only two cases of illness associated with the B/Yamagata-lineage influenza strain were identified in this study (both of which occurred in the control group in the Philippines). Thus, efficacy against infection and disease caused by the B/Yamagata strain was not tested. However, a hemagglutination-inhibition antibody (HAI) titer of at least 1:40 has typically been accepted as a marker of protection and has been used by the Food and Drug Administration for approval of the annual influenza vaccines. The data from this study and previous research show that HAI titers of 1:40 or more for both the B/Yamagata and B/Victoria lineages are elicited in most vaccinated children without evidence of interference from the increased vaccine valency.7-10 Given the additional B strain in the quadrivalent vaccine as compared with trivalent vaccines, there is concern that the increased vaccine antigen content (60 μg vs. 45 μg of hemagglutinin antigen) may lead to increased side effects; therefore, careful safety monitoring is warranted. To date, no safety signal has been observed.4,7-10 The effect that initial vaccination with both B lineages in young children will have on immunologic priming and subsequent boosting of the immune responses to both B lineages remains to be seen.
Novel influenza strains continue to emerge and cause severe disease in humans, as seen earlier this year in China with influenza A/H7N9.11 A vaccine is likely to be important to limit human illness from this pathogen if sustained human-to-human transmission develops. Over the 6 months since this pathogen was identified, a candidate vaccine antigen has been selected, manufactured, and clinically tested.12 Fries et al.12 recently showed in the Journal that two doses of a viruslike particle (VLP) vaccine (made with the use of cell-based recombinant baculovirus technology) that contained the influenza A/H7N9 antigen alone (15 μg or 45 μg) had limited immunogenicity (6 to 16% of the participants with HAI titers ≥1:40). However, when a saponin-based adjuvant with a lower amount of hemagglutinin antigen (5 μg or 15 μg) was used, the responses were improved (HAI titers ≥1:40 in 37 to 81% of the participants). No direct efficacy data are available, but the antibody titers elicited are encouraging, though it is unclear whether the protection usually associated with an HAI titer of 1:40 or higher applies to protection from A/H7N9. In addition, novel adjuvants raise important safety considerations that can be properly assessed only with larger clinical trials.
Constant reevaluation to identify the influenza strains that are the likeliest threats is important in determining the most effective strategy for preventing disease from influenza. Rapid development and wide use of vaccines with antigens that match the circulating strains are important for an effective response. Ongoing careful safety assessments of any new therapy, including these vaccines, is essential. In the United States, a quadrivalent influenza vaccine with both B-lineage antigens has been introduced this season as an alternative to the traditional trivalent vaccine. Over the next few influenza seasons we hope to see the value of the two Bs.
Disclosure forms provided by the author are available with the full text of this article at NEJM.org.
This article was published on December 11, 2013, at NEJM.org.
http://www.nejm.org/doi/full/10.1056/NEJMe1315317?af=R&rss=currentIssue&