Influenza vaccine immunogenicity in 6- to 23-month-old children: are identical antigens necessary for priming?

OBJECTIVES: Immunoprophylaxis with influenza vaccine is the primary method for reducing the effect of influenza on children, and inactivated influenza vaccine has been shown to be safe and effective in children. The Advisory Committee on Immunization Practices recommends that children 6 to 23 months of age who are receiving trivalent inactivated influenza vaccine for the first time be given 2 doses; however, delivering 2 doses of trivalent inactivated influenza vaccine > or = 4 weeks apart each fall can be logistically challenging. We evaluated an alternate spring dosing schedule to assess whether a spring dose of trivalent inactivated influenza vaccine was capable of "priming" the immune response to a fall dose of trivalent inactivated influenza vaccine containing 2 different antigens. PATIENTS AND METHODS: Healthy children born between November 1, 2002, and December 31, 2003, were recruited in the spring and randomly assigned to either the alternate spring schedule or standard fall schedule. The 2003-2004 licensed trivalent inactivated influenza vaccine was administered in the spring; the fall 2004-2005 vaccine had the same A/H1N1 antigen but contained drifted A/H3N2 antigen and B antigen with a major change in strain lineage. Reactogenicity was assessed by parental diaries and telephone surveillance. Blood was obtained after the second dose of trivalent inactivated influenza vaccine for all of the children and after the first dose of trivalent inactivated influenza vaccine in the fall group. The primary outcome of this study was to demonstrate noninferiority of the antibody response after a spring-fall dosing schedule compared with the standard fall dosing schedule. Noninferiority was based on the proportion of subjects in each group achieving a hemagglutination-inhibition antibody titer of > or = 1:32 after vaccination to 2 of the 3 antigens (H1N2, H3N2, and B) contained in the 2004-2005 vaccine. For each antigen, the antibody response was proposed to be noninferior if, within the upper bound of 95% confidence interval, there was < 15% difference between the proportion of children in the fall and spring groups with postvaccination titers > or = 1:32. RESULTS: A total of 468 children were randomly assigned to either the spring (n = 233) or fall (n = 235) trivalent inactivated influenza vaccine schedule. Excellent response rates to A/H1N1, as measured by antibody levels > or = 1:32, were noted in both the spring (86%) and fall groups (93%). The A/H1N1 response rate of the spring group was noninferior to that of the fall group. Noninferiority of the spring schedule was not met with respect to the other 2 influenza antigens: for A/H3N2 the response was 70% in the spring group versus 83% for the fall group, and the response to B was 39% in the spring group versus 88% for the fall group. After 2 doses of vaccine, the geometric mean antibody titers also were less robust in the spring group for both A/H3N2 and B antigens. For each of the 3 vaccine antigens, the respective geometric mean antibody titers for the spring group versus the fall group were: A/H1N1, 79.5 +/- 3.3 and 91.9 +/- 2.6; A/H3N2, 57.1 +/- 4.1 and 77.8 +/- 3.7; and B, 18.0 +/- 2.4 and 61.6 +/- 2.5. However, a significantly higher proportion of children in the spring group achieved potentially protective levels of antibody to all 3 antigens after their first fall dose of trivalent inactivated influenza vaccine than children in the fall group after receiving their first fall dose. For influenza A/H1N1, there was an antibody level > or = 1:32 in 86% of children in the spring group versus 55% of children in the fall group. Likewise, for influenza A/H3N2, 70% of children in the spring group and 47% of children in the fall group had antibody levels > 1:32; for influenza B, the proportions were 39% of children in the spring group and 16% of children in the fall group. Reactogenicity after trivalent inactivated influenza vaccine in both groups of children was minimal and did not differ by dose. CONCLUSIONS: Although the immune response to the identical A/H1N1 vaccine antigen was similar in both groups, priming with different A/H3N2 antigens and B antigens in the spring produced a lower immune response to both antigens than that shown in children who received 2 doses of the same vaccine in the fall. However, approximately 70% of children in the spring group had a protective response to the H3N2 antigen after 2 doses. Initiating influenza immunization in the spring was superior to 1 dose of trivalent inactivated influenza vaccine in the fall. The goal of delivering 2 doses of influenza vaccine a month apart to vaccine-naive children within the narrow flu vaccination season is a challenge not yet met; thus far, only about half of children aged 6 to 23 months of age are receiving influenza vaccine. By using the spring schedule, we were able to administer 2 doses of trivalent inactivated influenza vaccine to a higher proportion of children earlier in the influenza vaccination season. In years when there is an ample supply of trivalent inactivated influenza vaccine, and vaccine remains at the end of the season, priming influenza vaccine-naive infants with a spring dose will lead to the earlier protection of a higher proportion of infants in the fall. This strategy may be particularly advantageous when there is an early start to an influenza season as occurred in the fall of 2003. A priming dose of influenza vaccine in the spring may also offer other advantages. Many vaccine-naive children may miss the second dose of fall trivalent inactivated influenza vaccine because of vaccine shortages or for other reasons, such as the potential implementation of new antigens at a late date. Even with seasonal changes in influenza vaccine antigens, by giving a springtime dose of trivalent inactivated influenza vaccine, such children would be more protected against influenza than would children who were only able to receive 1 dose in the fall. In summary, our data suggest that identical influenza antigens are not necessary for priming vaccine-naive children and that innovative uses of influenza vaccine, such as a springtime dose of vaccine, could assist in earlier and more complete immunization of young children.

Duke Scholars

Author Emmanuel Benjamin Walter Jr. Pediatrics, General Pediatrics and Adolescent Health