Greenberg, David P. MD*†; Robertson, Corwin A. MD, MPH*; Landolfi, Victoria A. MS, MBA*; Bhaumik, Amitabha PhD*; Senders, Shelly D. MD‡; Decker, Michael D. MD, MPH*§
Despite widespread availability of vaccines, influenza remains a serious health risk for children in the United States. During the 2008−2009 through 2011−2012 influenza seasons, national laboratory-confirmed influenza hospitalization rates were between 14.2 and 72.8 per 100,000 children ≤ 4 years of age and between 4.2 and 27.3 per 100,000 children 5−17 years of age.1 During the same period, the total number of pediatric deaths ranged from a low of 26 during the 2011−2012 season to a high of 348 during the 2009−2010 pandemic season.1
To help prevent influenza virus infection and disease, the Advisory Committee on Immunization Practices of the US Centers for Disease Control and Prevention recommends routine annual influenza vaccination for all children ≥6 months of age.2 Seasonal trivalent inactivated influenza vaccines (TIVs) contain 2 influenza A subtype strains, 1 A/H1N1 strain and 1 A/H3N2 strain and 1 influenza B strain, from either the Victoria or Yamagata lineage. These 2 B lineages have circulated globally since the mid-1980s;3 however, before 2001, only 1 lineage predominated in the United States each season. Since the 2001−2002 season, both B lineages have cocirculated with varying frequencies.4
In February of each year, advisory committees of the World Health Organization and US Food and Drug Administration meet to choose which A/H1N1, A/H3N2 and B strains should be included in TIVs for the upcoming influenza season in the Northern Hemisphere. The decision is based on global virologic and epidemiologic surveillance and serologic studies in which circulating strains are tested against ferret and human antibodies. Despite extensive analysis of these and other data, it is difficult to predict, 8−14 months in advance, which B lineage will predominantly circulate in the upcoming influenza season.
The B lineage selected by the World Health Organization and US Food and Drug Administration for vaccine formulation has matched the predominant circulating lineage in only 6 of the past 12 seasons.5 During this period, influenza B viruses caused a yearly average of approximately 25% of all influenza cases in the United States, with yearly proportions as high as 44%.5 Even when the predominant B-lineage strain is correctly chosen, because cocirculation occurs annually, some proportion of influenza is caused by an opposite B-lineage strain. For example, during the 2011−2012 season, 51% of circulating B strains were from the Yamagata lineage and 49% were from the Victoria lineage.
While vaccination enhances immunity against the influenza strains contained in TIV, the degree of protection depends in part on how well the vaccine strains match those actually circulating during the influenza season. When the vaccine and circulating A strains are not well-matched or the B strains are of different lineages, the effectiveness of vaccination is reduced.6 During the 2006−2007 influenza season, among persons ≥9 years of age in Canada, the effectiveness of TIV against a well-matched A/H1N1 strain was 92% [95% confidence interval (CI): 40–91%], whereas the effectiveness of the same TIV, containing a Victoria-lineage B strain, against the circulating Yamagata B-lineage strain was only 19% (95% CI: 112–69%).7
Quadrivalent inactivated influenza vaccine (QIV) containing a strain from each of the 2 B lineages, in addition to the standard A/H1N1 and A/H3N2 strains, should resolve the issue of B-lineage mismatch. A modeling study conducted by the US Centers for Disease Control and Prevention found that replacing TIV with QIV has the potential to reduce annual influenza cases, hospitalizations and deaths.8 Before a manufacturer’s QIV can be approved for general use, regulatory authorities must be assured that there are no safety concerns compared with standard-of-care TIVs, and there is no immunologic interference caused by the addition of the second B-lineage strain. We report here the results of a Phase III multicenter study conducted in the United States during the 2010−2011 influenza season to assess the safety and immunogenicity of QIV compared with control TIVs among children 6 months to <9 years of age.
MATERIALS AND METHODS
This was a Phase III, randomized, observer-blinded, active-controlled, 3-arm, multicenter trial to assess the safety and immunogenicity of QIV compared with control TIVs in a pediatric population (NCT Registry No.: NCT01240746). The primary objective was to demonstrate that for each A and B strain QIV induced noninferior antibody responses compared with those induced by 2 control TIVs, each containing the same A strains and either a Yamagata- or Victoria-lineage B strain. The secondary objective was to demonstrate that for each B strain QIV induced superior antibody responses compared with those of each respective TIV not containing the same B-lineage strain. The study was performed at 69 centers in the United States during the 2010−2011 influenza season. It was approved by all relevant institutional review boards and was carried out in accordance with International Conference on Harmonization Guidelines for Good Clinical Practice and the Declaration of Helsinki. Written informed consent was obtained from parents or legal guardians before children were included in the trial. In addition, for children 7 to <9 years of age, signed assent was obtained. Enrollment took place between November 11, 2010, and June 20, 2012.
Study Population and Sample Size Calculation
Children included in the study had to be 6 months to <9 years of age and generally in good health. In addition, those who were 6 months to <24 months of age had to be born at full term (≥37 weeks) and with a birth weight ≥2.5 kg (5.5 lbs). Children were excluded if they had a history of allergy to egg proteins, latex or any constituents of the vaccine; a history of serious adverse reactions to any influenza vaccine; received any vaccine in the 4 weeks preceding the first study vaccination (or scheduled between study visits) or influenza vaccine after August 1, 2010; a history of Guillain-Barré syndrome; a known or suspected congenital or acquired immunodeficiency; received immunosuppressive therapy within the preceding 6 months or long-term systemic corticosteroid therapy within the past 3 months; a history of developmental delay, neurologic disorder or seizure disorder; known seropositivity to human immunodeficiency virus, hepatitis B or hepatitis C; or received blood or blood-derived products in the past 3 months.
Enrollment was stratified by age so that approximately half of the participants at each site were 6 months to <36 months of age and half were 3 years to <9 years of age. Enrollment targets were 3340 children in the QIV group and 800 children in each of 2 control TIV groups. This was estimated to yield 95% overall power to demonstrate that the immunogenicity of QIV was noninferior to TIV for all 4 strains and 90% power for each age group separately (6 months to <36 months and 3 years to <9 years) assuming that 90% of those enrolled would be evaluable (approximately 3000 for QIV and 720 for each TIV). This was also estimated to result in at least 99% power to demonstrate immunologic superiority of QIV compared with each respective TIV not containing the corresponding B strain for each age group separately and overall.
Vaccine Formulation and Administration
QIV contained A/California/07/2009 (H1N1), A/Victoria/210/2009 (H3N2), B/Brisbane/60/2008 (B Victoria lineage) and B/Florida/04/2006 (B Yamagata lineage) strains. A licensed 2010–2011 formulation of TIV (Fluzone, Sanofi Pasteur, Swiftwater, PA) contained B/Brisbane/60/2008 and an investigational TIV contained B/Florida/04/2006. Each TIV contained the same A/H1N1 and A/H3N2 strains as QIV. Each vaccine was formulated to contain 30 μg hemagglutinin/strain/mL. Doses were provided in prefilled 0.25- or 0.5-mL, single-dose syringes or in 0.5-mL, single-dose vials. Vaccine potency was assessed periodically through a routine stability monitoring program.
Participants were randomized at a ratio of approximately 4:1:1 to be immunized with QIV, licensed TIV or investigational TIV, using a programmed interactive voice response system. Participants were immunized with the appropriate dose of vaccine based on age at the time of enrollment; 0.25 mL for children 6 to <36 months of age and 0.5 mL for children 3 to <9 years of age.2 Participants received 1 or 2 doses of study vaccine 4 weeks (window, 28–35 days) apart based on their influenza vaccine history, as recommended by the Advisory Committee on Immunization Practices for the 2010–2011 season.2 All immunizations were administered by intramuscular injection into the anterolateral thigh or the deltoid region. The vaccinees, family members and all site personnel except the vaccination nurse (who did not collect safety data) were blinded to the administered vaccine.
Hemagglutination Inhibition Assay and Immunogenicity Endpoints
Blood samples were collected on day 0 (prevaccination) and day 28 (window, days 28−35) after the final vaccination. A validated hemagglutination inhibition assay, approved by the Food and Drug Administration for clinical trial testing, was used to quantify antibody titers against study vaccine antigens. Assays were performed by Sanofi Pasteur personnel who were blinded to vaccine assignment. Control and participant sera were incubated with type III neuraminidase to eliminate nonspecific inhibitors. Spontaneous antispecies agglutinins were adsorbed by incubating the sera with a suspension of Turkey red blood cells. Ten 2-fold dilutions (starting at 1:10) of the treated sera were incubated with a previously titrated influenza virus solution at a concentration of 4 hemagglutination units/25 μL. The reported hemagglutination inhibition titer corresponded to the highest serum dilution resulting in complete inhibition of hemagglutination and was determined in 2 independent assay runs.
The titer for each sample was calculated as the geometric mean of the reciprocal of the 2 independent values. The lower limit of quantitation was a titer of 1:10; samples with titers below this level were assigned a titer of 1:5. The seroprotection rate for each group was defined as the percentage of vaccinees with a titer ≥1:40. The seroconversion rate for each group was defined as the percentage of vaccinees with either a prevaccination titer <1:10 and a postvaccination titer ≥1:40 or a prevaccination titer ≥1:10 and a ≥4-fold increase in titer postvaccination.
Solicited injection-site and systemic reactions were recorded for 7 days after each vaccination. Solicited reactions differed by age: for children 6 months to <24 months of age, injection-site tenderness, erythema and swelling; systemic fever, vomiting, abnormal crying, drowsiness, loss of appetite and irritability; for children 2 years to <9 years of age, injection-site pain, erythema and swelling; systemic fever, headache, malaise and myalgia. Unsolicited adverse events (AEs) and serious adverse events (SAEs) were collected according to International Committee for Harmonization Guideline (E2A) for Clinical Safety Data Management: Definitions and Standards for Expedited Reporting. Unsolicited AEs were collected from day 0 to day 28 (window, days 28−35) after the final vaccination. SAEs and AEs of special interest (Guillain-Barré syndrome, Bell’s palsy, encephalitis/myelitis, optic neuritis, Stevens-Johnson syndrome, toxic epidermal necrolysis and febrile seizure) were collected for 6 months after the final vaccination.
All analyses were carried out using SAS version 9.1 or higher (SAS Institute, Cary, NC). Missing and incomplete data were not replaced. Immunogenicity was assessed in all participants who were randomized and received the study or a control vaccine, had a valid postvaccination serology result and completed the study according to protocol (per-protocol analysis set). Noninferiority was assessed for all 4 viral strains in QIV compared with the control TIVs. For comparison of A/H1N1 and A/H3N2 responses, data were pooled among the 2 TIVs. For comparison of B-strain responses, QIV was compared with the respective TIV containing the same B-lineage strain. Noninferiority of geometric mean titers (GMTs) was achieved if the lower limit of the 2-sided 95% CI of the GMTQIV/GMTTIV ratio was > 0.66, and noninferiority of seroconversion rates (SCRs) was achieved if the lower limit of the 2-sided 95% CI of the SCRQIV − SCRTIV difference was >−10%.
Superiority was assessed for each B-lineage strain in QIV compared with each respective TIV not containing the same B-lineage strain. Superiority of GMTs was achieved if the lower limit of the 2-sided 95% CI of the GMTQIV/GMTTIV ratio was >1.5, and superiority of SCRs was achieved if the lower limit of the 2-sided 95% CI of the SCRQIV − SCRTIV difference was >10%.
Safety was assessed in all randomized participants who received a study or control vaccine (safety analysis set) and reported descriptively. The 95% CIs of point estimates were calculated using the normal approximation for quantitative data and the exact binomial distribution (Clopper-Pearson method) for proportions.
Disposition and Demographics
A total of 4363 children were randomized, of whom 4348 received study vaccine: 2893 in the QIV group, 734 in the licensed TIV group and 721 in the investigational TIV group (Fig. 1). Approximately three-quarters of the children in each group received a second dose of vaccine, based on Advisory Committee on Immunization Practices recommendations. A total of 350 participants did not complete the vaccination phase of the study (comparable proportion of each group), primarily because of loss to follow up (n = 145), noncompliance (n = 111) and voluntary withdrawal not due to an AE (n = 82). Approximately equal numbers of males and females were enrolled and the mean age and race/ethnic distributions were similar in each group (Table 1).
Prevaccination GMTs were similar across the 3 vaccine groups (see Table, Supplemental Digital Content 1, http://links.lww.com/INF/B802). For A/H1N1 and A/H3N2, QIV induced postvaccination GMTs and seroconversion rates that were noninferior to those induced by TIV (Table 2). For each B-lineage strain, QIV induced GMTs and seroconversion rates that were noninferior to those induced by TIV containing the same B strain and superior to those induced by TIV not containing the same B strain. Further, all noninferiority and superiority criteria were met for the 2 age subgroups, 6 to <36 months and 3 to <9 years of age (see Tables, Supplemental Digital Content 2, http://links.lww.com/INF/B803: Table A, for 6 months to <36 months of age and Table B, for 3 to <9 years of age). The postvaccination seroprotection rates were similar among the 3 vaccine groups (Table 3).
Some children received investigational TIV in which the antigen content decreased during the study to below a prespecified level (28 μg hemagglutinin/strain/mL), but antibody responses were similar among children administered in-specification (n = 211) and out-of-specification (n = 388) vaccine, and all noninferiority and superiority criteria were met for all participants combined (Table 2) as well as when those who received out-of-specification lots were excluded (see Table, Supplemental Digital Content 3, http://links.lww.com/INF/B804).
The proportions of children reporting solicited injection-site and systemic reactions were similar across the vaccine groups (Table 4). The most common reactions among children 6 to <24 months of age were irritability and injection-site tenderness, and the most common reactions among children 2 to <9 years of age were myalgia, malaise and injection-site pain (Fig. 2). In all vaccine groups, most reactions were grade 1 or 2 in intensity, began within 3 days of vaccination and resolved within 3 days of onset.
In all vaccine groups, <1% of participants reported unsolicited immediate (ie, within 20 minutes of vaccination) AEs, none of which were considered grade 3 in intensity. Proportions of children reporting any vaccine-related or grade 3 nonserious unsolicited AEs were similar for all 3 vaccine groups. The most common nonserious unsolicited AEs were cough, upper respiratory tract infection, fever and vomiting, most of which were grade 1 or 2 in intensity.
Five children receiving QIV (0.2%) discontinued the study because of 1 or more AEs thought to be related to vaccination, including fever, malaise, irritability and abnormal crying, injection-site erythema, swelling and pruritus and hives on face, hands and feet. None of these were considered SAEs. No children receiving a TIV discontinued due to a related AE.
Three SAEs were reported as being related to vaccination by investigators: croup in a 13-month old 3 days after the first dose of QIV, febrile seizure in an 11-month old 8 hours after the second dose of investigational TIV and febrile seizure in a 4-year old 1 day after the first dose of licensed TIV. All resolved and did not result in early discontinuation. One death (not vaccine related) was reported during the study: a case of drowning in a 19-month old that occurred 43 days after the second dose of licensed TIV.
Thirteen participants experienced an AE of special interest, all of which were febrile seizures. Only 2 of these were considered vaccine related and are described above.
Influenza causes substantial illness, complications, hospitalizations and deaths among children.1,2 For example, during the 2002−2003 and 2003−2004 influenza seasons, laboratory-confirmed influenza accounted for between 50 and 95 clinic visits per 1000 children <5 years of age presenting with acute respiratory tract infection or fever and between 6 and 27 emergency department visits in the same cohort.9
Influenza B occurs in persons of all ages, but it appears to affect older children and young adults more than other age groups.10,11 In addition, type B causes a substantial proportion of influenza-related deaths in the pediatric age group. Based on data provided by US Centers for Disease Control and Prevention, Ambrose and Levin12 calculated that type B caused some 34% of reported pediatric deaths attributable to influenza during the past 7 influenza seasons (2004−2005 through 2010−2011, excluding the 2009–2010 A/H1N1 pandemic).
TIVs provide substantial protection against influenza illness in children. For example, in a randomized study conducted over a period of 5 seasons among children 1−15 years of age in the United States, TIV reduced symptomatic culture-positive influenza by 77% against A/H3N2 strains and 91% against A/H1N1 strains.13 In a case-control study conducted in 4 states during the 2010−2011 season among persons presenting to hospitals, emergency departments and medical clinics with laboratory-confirmed influenza illness, the adjusted vaccine effectiveness of TIV was 71% (95% CI, 58−78%) in children 2−8 years of age.14
Despite widespread vaccination of children with influenza vaccines, 1 reason that TIV offers suboptimal protection is because it contains a B strain from only 1 of the 2 cocirculating B lineages and often not the most common B-lineage strain in circulation that season. QIV containing a B strain from each lineage offers a promising solution to this problem. In the study reported herein, we showed that QIV was as immunogenic as licensed TIV in children 6 months to < 9 years of age. There was no evidence of immunologic interference as a result of adding the alternate B-lineage strain. In addition, QIV induced superior antibody responses to the B strain not covered by each respective control TIV.
QIV was well-tolerated by the study children and had a safety profile similar to that of licensed TIV, the safety of which has been well-documented.15–18 As with TIV, injection-site and systemic reactions to QIV were generally mild and short-lived. Of note, the rate of fever during the 7 days postvaccination was no higher among QIV recipients than TIV recipients, including in the youngest children 6 to <24 months of age. There was no unusual pattern of unsolicited nonserious or serious AEs. A febrile seizure was reported as related to vaccination in 1 child after licensed TIV and in 1 child after investigational TIV, but none among QIV recipients. Overall, QIV did not present any safety concerns.
The safety and immunogenicity of a quadrivalent live attenuated influenza vaccine (Q/LAIV) have been reported in adults and children.19,20 In both age groups, Q/LAIV induced noninferior antibody responses compared with trivalent LAIV (T/LAIV) for all corresponding influenza strains. As in our study, the addition of a second B-lineage strain did not interfere with antibody responses. The safety profile of Q/LAIV was comparable to T/LAIV, except for a higher rate of fever after Q/LAIV than after T/LAIV (5.1% vs. 3.1% with fever ≥ 38.0°C and 1.2% vs. 0.3% with fever ≥39.0°C) in the pediatric study.
In a separate study conducted among children 3–17 years of age, the safety profile of another manufacturer’s QIV was comparable to 2 control TIVs.21 Antibody responses to each strain in the QIV were noninferior to those of the same strains in each control TIV and the responses to the B strains in the QIV were superior to those induced by each TIV containing the alternate B-lineage strain.
We have recently published our results of QIV administered to adults ≥18 years of age.22 As in the pediatric study, QIV induced noninferior antibody responses compared with 2 control TIVs for all corresponding strains and higher responses compared with each TIV that contained the B strain from the opposite lineage. Rates of solicited and unsolicited AEs were similar between groups and there were no safety concerns. Similar results for QIV versus control TIVs were noted in a third study of QIV in elderly persons 65 years of age and older, which will be published separately.23
A limitation of this study was that we restricted enrollment to generally healthy children. Antibody responses would likely be diminished among children with congenital or acquired immunodeficiencies. However, given the similar immune responses shown herein between QIV and the control TIVs, we would expect QIV to perform as well as TIV in high risk populations. We did not evaluate children 9 through 17 years of age, but our data, demonstrating similar immunogenicity and safety profiles between QIV and control TIVs in both younger children and in adults, provide reassurance that QIV would perform comparably among 9- to 17-year olds.
In conclusion, the safety and immunogenicity of QIV was comparable with licensed TIV in a healthy pediatric population. By inducing antibody responses to both B lineages simultaneously, QIV should help overcome the limitations of TIV, namely, its inability to protect against both cocirculating B lineages simultaneously and that it often does not contain the B-lineage strain predominating in a given influenza season.
The authors would like to thank the investigators and research staff at the 69 clinical study sites for performing this study and to the many families for their participation. Medical writing assistance in the preparation of this article was provided by Drs. Phillip Leventhal and Kurt Liittschwager of 4Clinics (Paris, France). Support for this study and medical writing assistance was provided by Sanofi Pasteur.
1. Centers for Disease Control Update: influenza activity - United States, 2011–12 season and composition of the 2012–13 influenza vaccine. MMWR Morb Mortal Wkly Rep. 2012; 61:414–420
2. Fiore AE, Uyeki TM, Broder K, et al. Centers for Disease Control and Prevention (CDC) Prevention and control of influenza with vaccines: recommendations of the Advisory Committee on Immunization Practices (ACIP), 2010. MMWR Recomm Rep. 2010; 59:(RR-8)1–62
3. Rota PA, Wallis TR, Harmon MW, et al. Cocirculation of two distinct evolutionary lineages of influenza type B virus since 1983. Virology. 1990; 175:59–68
4. Glezen WP, Schmier JK, Kuehn CM, et al. The burden of influenza B: a structured literature review. Am J Public Health. 2013; 103:e43–e51
6. Belongia EA, Kieke BA, Donahue JG, et al. Marshfield Influenza Study Group Effectiveness of inactivated influenza vaccines varied substantially with antigenic match from the 2004–2005 season to the 2006–2007 season. J Infect Dis. 2009; 199:159–167
7. Skowronski DM, De Serres G, Dickinson J, et al. Component-specific effectiveness of trivalent influenza vaccine as monitored through a sentinel surveillance network in Canada, 2006-2007. J Infect Dis. 2009; 199:168–179
8. Reed C, Meltzer MI, Finelli L, et al. Public health impact of including two lineages of influenza B in a quadrivalent seasonal influenza vaccine. Vaccine. 2012; 30:1993–1998
9. Poehling KA, Edwards KM, Weinberg GA, et al. New Vaccine Surveillance Network The underrecognized burden of influenza in young children. N Engl J Med. 2006; 355:31–40
10. Grant KA, Carville K, Fielding JE, et al. High proportion of influenza B characterises the 2008 influenza season in Victoria. Commun Dis Intell Q Rep. 2009; 33:328–336
11. Olson DR, Heffernan RT, Paladini M, et al. Monitoring the impact of influenza by age: emergency department fever and respiratory complaint surveillance in New York City. PLoS Med. 2007; 4:e247
12. Ambrose CS, Levin MJ. The rationale for quadrivalent influenza vaccines. Hum Vaccin Immunother. 2012; 8:81–88
13. Neuzil KM, Dupont WD, Wright PF, et al. Efficacy of inactivated and cold-adapted vaccines against influenza A infection, 1985 to 1990: the pediatric experience. Pediatr Infect Dis J. 2001; 20:733–740
14. Treanor JJ, Talbot HK, Ohmit SE, et al. US Flu-VE Network Effectiveness of seasonal influenza vaccines in the United States during a season with circulation of all three vaccine strains. Clin Infect Dis. 2012; 55:951–959
15. Mitchell DK, Ruben FL, Gravenstein S. Immunogenicity and safety of inactivated influenza virus vaccine in young children in 2003–2004. Pediatr Infect Dis J. 2005; 24:925–927
16. Lina B, Fletcher MA, Valette M, et al. A TritonX-100-split virion influenza vaccine is safe and fulfills the committee for proprietary medicinal products (CPMP) recommendations for the European Community for Immunogenicity, in Children, Adults and the Elderly. Biologicals. 2000; 28:95–103
17. Gonzalez M, Pirez MC, Ward E, et al. Safety and immunogenicity of a paediatric presentation of an influenza vaccine. Arch Dis Child. 2000; 83:488–491
18. France EK, Glanz JM, Xu S, et al. Safety of the trivalent inactivated influenza vaccine among children: a population-based study. Arch Pediatr Adolesc Med. 2004; 158:1031–1036
19. Block SL, Falloon J, Hirschfield JA, et al. Immunogenicity and safety of a quadrivalent live attenuated influenza vaccine in children. Pediatr Infect Dis J. 2012; 31:745–751
20. Block SL, Yi T, Sheldon E, et al. A randomized, double-blind noninferiority study of quadrivalent live attenuated influenza vaccine in adults. Vaccine. 2011; 29:9391–9397
21. Domachowske JB, Pankow-Culot H, Bautista M, et al. A randomized trial of candidate inactivated quadrivalent influenza vaccine versus trivalent influenza vaccines in children aged 3–17 years. J Infect Dis. 2013; 207:1878–1887
22. Greenberg DP, Robertson CA, Noss MJ, et al. Safety and immunogenicity of a quadrivalent inactivated influenza vaccine compared to licensed trivalent inactivated influenza vaccines in adults. Vaccine. 2013; 31:770–776
23. Greenberg DP, Robertson C, Talbot HK, et al. Safety and immunogenicity of a quadrivalent inactivated influenza vaccine (QIV) containing two A and two B strains among persons ≥ 65 years of age (Abstract 544, presented). 49th Annual Meeting of the Infectious Diseases Society of America (IDSA). 2011; Boston, MA Infectious Diseases Society of America