Time to Change Dosing of Inactivated Quadrivalent Influenza Vaccine in Young Children: Evidence From a Phase III, Randomized, Controlled Trial

Abstract Background. Children under 3 years of age may benefit from a double-dose of inactivated quadrivalent influenza vaccine (IIV4) instead of the standard-dose. Methods. We compared the only United States-licensed standard-dose IIV4 (0.25 mL, 7.5 µg hemagglutinin per influenza strain) versus double-dose IIV4 manufactured by a different process (0.5 mL, 15 µg per strain) in a phase III, randomized, observer-blind trial in children 6–35 months of age (NCT02242643). The primary objective was to demonstrate immunogenic noninferiority of the double-dose for all vaccine strains 28 days after last vaccination. Immunogenic superiority of the double-dose was evaluated post hoc. Immunogenicity was assessed in the per-protocol cohort (N = 2041), and safety was assessed in the intent-to-treat cohort (N = 2424). Results. Immunogenic noninferiority of double-dose versus standard-dose IIV4 was demonstrated in terms of geometric mean titer (GMT) ratio and seroconversion rate difference. Superior immunogenicity against both vaccine B strains was observed with double-dose IIV4 in children 6–17 months of age (GMT ratio = 1.89, 95% confidence interval [CI] = 1.64–2.17, B/Yamagata; GMT ratio = 2.13, 95% CI = 1.82–2.50, B/Victoria) and in unprimed children of any age (GMT ratio = 1.85, 95% CI = 1.59–2.13, B/Yamagata; GMT ratio = 2.04, 95% CI = 1.79–2.33, B/Victoria). Safety and reactogenicity, including fever, were similar despite the higher antigen content and volume of the double-dose IIV4. There were no attributable serious adverse events. Conclusions. Double-dose IIV4 may improve protection against influenza B in some young children and simplifies annual influenza vaccination by allowing the same vaccine dose to be used for all eligible children and adults.

Influenza has a high incidence and burden in children [1][2][3][4]. In particular, influenza B is reported to cause a disproportionate number of influenza-related deaths in children [5]. Routine vaccination of children against influenza is recommended in the United States [6] and other countries. Quadrivalent influenza vaccines containing 2 influenza A strains and 2 influenza B strains are increasingly used in vaccination programs to replace trivalent vaccines.
Inactivated influenza vaccines (IIVs) are administered to adults and children from 3 years of age at a dose of 0.5 mL, containing 15 µg of hemagglutinin (HA) per virus strain. In children under 3 years of age, the United States-licensed standard-dose is 0.25 mL, containing 7.5 µg of HA per virus strain. Both the 15 µg and 7.5 µg doses are available for this age group in some countries, including Canada, Brazil, Mexico, Finland, and the United Kingdom. The 7.5 µg dose was introduced in the 1970s to reduce reactogenicity, including febrile convulsions, associated with the whole virus vaccines available at the time [7][8][9][10][11]. However, young children mount a variable immune response to the 7.5 µg dose [12][13][14]. Currently available split virus vaccines are better tolerated than whole virus vaccines [10,15,16], questioning the practice of using the 7.5 µg dose with IIVs.
The inactivated quadrivalent influenza vaccine (IIV4) manufactured in Quebec, Canada by GSK Vaccines is licensed at a double-dose (15 µg per antigen) for children from 6 months of age in Canada and Mexico, but it is currently only licensed for children 3 years of age and older in the United States. The only IIV4 licensed for use in children 6-35 months of age in the United States is Sanofi Pasteur's Fluzone Quadrivalent in a standard-dose (7.5 µg per antigen). No other IIV is approved in the United States in this age group either because immunogenic noninferiority to Fluzone could not be demonstrated [17,18] or because of excessive reactogenicity [19,20].
If the double-dose vaccine could be administered in young children without adverse effects on tolerability, this age group may benefit from potentially improved immunogenicity. In this study, we describe a phase III study that compared the safety and immunogenicity of a double-dose IIV4 manufactured by GSK Vaccines with the United States-approved standard-dose IIV4 in children 6-35 months of age.

METHODS
This was a phase III, randomized, controlled, observer-blind, multicenter trial in children 6-35 months of age (ClinicalTrials. gov Identifier NCT02242643). The trial was approved by independent ethics committees or institutional review boards, conducted in accordance with the Declaration of Helsinki, the International Conference on Harmonisation Good Clinical Practice (ICH-GCP) guidelines, ICH Harmonised Tripartite guideline for pediatric populations, and US regulatory requirements. Parents or legally acceptable representatives provided written informed consent. Children were randomized 1:1 to double-dose or standard-dose IIV4. Allocation to a study group at the investigator site was performed using an internet-based randomization system (SBIR). The randomization algorithm used a minimization procedure to balance the composition of treatment groups, accounting for age (6-17 and 18-35 months), center, and influenza vaccine priming status. The study aimed to enroll 40%-50% of children in the 6-17 months age group. Children were considered vaccine-primed if they had received 2 or more doses of influenza vaccine since July 1, 2010 or at least 1 dose of the

Study Endpoints
Blood for serologic testing was obtained on days 0 and 28 from primed children and on days 0 and 56 from unprimed children. The following parameters were derived from hemagglutination inhibition (HI) titers: (1) geometric mean titer (GMT), (2) seroconversion rate (SCR), (3) seroprotection rate (SPR), and (4) mean geometric increase (MGI). Seroconversion rate was defined as the percentage of participants with either (1) prevaccination reciprocal HI titer <1:10 and a postvaccination reciprocal titer ≥1:40 or (2) prevaccination reciprocal titer ≥1:10 and at least a 4-fold increase in postvaccination reciprocal titer. Seroprotection rate was defined as the percentage of participants who attained reciprocal HI titers of ≥1:40. Mean geometric increase was defined as the geometric mean of the within-subject ratios of the postvaccination/prevaccination reciprocal HI titer.
Parents recorded solicited injection site and general symptoms on the day of vaccination and for the next 6 days. Spontaneously reported symptoms were recorded until 28 days after vaccination. Serious adverse events (SAEs), potential immune-mediated diseases, and medically attended adverse events were recorded until the final study contact on day 180. Monitoring for febrile seizures was carried out throughout the study.

Study Objectives
The primary objective was to demonstrate immunogenic noninferiority of the double-dose versus the standard-dose IIV4 28 days after completion of the vaccination course. Noninferiority criteria were met if, for each of the 4 vaccine strains, the upper limit of the 95% confidence interval (CI) of the GMT ratio (standard-dose/double-dose) was ≤1.5 and the upper limit of the 95% CI of the difference in SCR (standard-dose minus double-dose) was ≤10%.
If the primary objective was achieved, the secondary objective was to evaluate whether double-dose IIV4 produced an immune response against each of the vaccine strains that met Center for Biologics Evaluation and Research (CBER) criteria, ie, the lower limit of the 95% CI of the SCR was ≥40% and the lower limit of the 95% CI of the SPR was ≥70%. Additional secondary objectives were to (1) evaluate GMT, SPR, SCR, and MGI at 28 days after completion of the vaccination course, (2) describe the safety and reactogenicity of the vaccines, and (3) evaluate the relative risk of fever with double-dose versus standard-dose during the 2-day postvaccination period.
A post hoc evaluation was conducted to compare the immune response of the double-dose versus the standard-dose using CBER criteria conventionally applied to establish vaccine lot-to-lot consistency. Immunogenic superiority of the double-dose was concluded if the lower limit of the 95% CI of the GMT ratio (double-dose/standard-dose) was >1.5 and the lower limit of the 95% CI of the difference in SCR (double-dose minus standard-dose) was >10%.

Statistics
Enrollment of 1200 children per group (1020 evaluable subjects assuming an attrition rate of 15%) was planned to allow a global statistical power of 99% for the primary objective evaluation. The immunogenicity analysis was based on the per-protocol cohort and the safety analysis was based on the intent-totreat cohort (Supplementary Appendix). Subgroup analyses according to age and priming status were conducted on the per-protocol cohort. The overall type I error for the study was 5%. If the primary objective was met, the secondary objective of CBER criteria evaluation was tested to provide supportive evidence of immunogenicity. Calculation of 95% CIs is described in the Supplementary Appendix. The group GMT ratio was computed using an analysis of covariance model on the log-transformed titers. Analyses of immunogenicity excluded participants with missing or nonevaluable measurements at the postvaccination time point. Study power was calculated using PASS 2005 (Supplementary Appendix).

RESULTS
A total of 2424 and 2041 children were included in the intentto-treat cohort and per-protocol cohort, respectively ( Figure 1). Demographics were similar in both vaccine groups (Table 1). In the per-protocol cohort, 57.5% of children were vaccine-primed; mean age was 24.6 and 13.3 months for primed and unprimed children, respectively. Other demographic characteristics were similar in primed and unprimed children (Table 1).

Immunogenicity
Both vaccines were immunogenic against all vaccine strains in terms of GMT values ( Figure 2). Immunogenic noninferiority of the double-dose IIV4 versus the standard-dose IIV4 was demonstrated for all vaccine strains ( Figure 3). Seroconversion rate, SPR, and MGI values were higher in the double-dose group compared with the standard-dose group in the whole study population (6-35 months of age, regardless of priming status; Table 2). The lower limit of the 95% CI for SCR was ≥40% for the double-dose IIV4 against all vaccine strains (Table 2), meeting CBER criteria for demonstration of adequate immunogenicity. For SPR, the lower limit of the 95% CI was ≥70% for all strains except B/Victoria (Table 2).
Immunogenicity was higher in the double-dose group compared with the standard-dose group, particularly against vaccine B strains in children 6-17 months of age and unprimed children (Table 3; Figure 2). When the unprimed group was further evaluated by age, it could be seen that the main difference between vaccines occurred in children 6-17 months of age. These observations prompted us to perform the post hoc evaluation comparing the immune response elicited by the vaccines in the whole study population and according to age group and priming status. The analysis indicated superior immunogenicity of the double-dose IIV4 against both vaccine B strains in children 6-17 months of age and all unprimed children. In children 6-17 months of age, the GMT ratio was 1.89 (95% CI, 1.64-2.17) for B/Yamagata and 2.13 (95% CI, 1.82-2.50) for B/Victoria (Figure 4 and Supplementary Table 1).    Corresponding values in all unprimed children were 1.85 (95% CI, 1.59-2.13) and 2.04 (95% CI, 1.79-2.33). Superior immunogenicity of the double-dose was also observed for the same groups in terms of SCR difference (Figure 4 and Supplementary Table 1).

Safety and Reactogenicity
Pain was the most common solicited injection site symptom, occurring in approximately 40% of children in both vaccine groups; severe (grade 3) pain occurred in 2.9% (95% CI, 2.0-4.1) and 1.7% (95% CI, 1.0-2.6) of children with the double-dose and standard-dose, respectively (Table 4). Fever (≥38.0°C) was reported in approximately 8% of children up to 7 days postvaccination; fever >39.0°C occurred in approximately 2% of children (Table 4). During the 2-day postvaccination period (days 0-1), the incidence of fever (≥38.0°C) was similar in both groups (Table 4), and the relative risk (double-dose/standard-dose) was 0.97 (95% CI, 0.62-1.52; P = .9777). Twenty-two SAEs occurred in the double-dose group and 21 in the standard-dose group (Table 4), none considered related to vaccination. Febrile seizure was reported in 5 children in the double-dose group and in 4 children in the standard-dose group (Table 4).
There was a modest increase in reactogenicity with regard to general symptoms in children 6-17 months of age compared with those aged 18-35 months with both the double-dose and standard-dose vaccines. With the double-dose vaccine, the fold-difference between the younger and older age groups ranged from 1.3 for loss of appetite to 2.7 for fever ≥38.0°C. With the standard-dose, the fold-difference ranged from 1.2 for loss of appetite to 1.6 for drowsiness and fever ≥38.0°C. The difference between age groups was unlikely to be due to chance because, in general, 95% CIs did not overlap. However, there were overlapping 95% CIs and thus no apparent age group differences with the standard-dose vaccine for fever ≥38.0°C and loss of appetite.

DISCUSSION
The introduction of IIV4 provides an opportunity to review long-accepted practices in administration of influenza vaccines. Since the 1970s, the standard-dose of IIVs in children less than 3 years of age has been 7.5 µg per antigen, half the dose given to older children and adults. The lower dose was intended to reduce reactogenicity and febrile convulsions observed with the whole virus vaccines that were in use at the time [7][8][9][10][11]. However, young children mount a variable immune response to this lower dose, especially against vaccine B strains [12][13][14]. In particular, vaccine-naive children less than 3 years of age mount a lower immune response compared with older or  vaccine-primed children [14,[21][22][23]. The immune response in this vulnerable group could be improved by a change in practice to administer the double-dose, ie, same dose as used for children 3 years of age and above, and for adults. Increasing the immunogenicity of IIVs for young children is expected to improve their effectiveness, because the postvaccination HI antibody titer is inversely related to the risk of illness [24,25]. However, there is controversy regarding the HI antibody titer necessary to offer high-level effectiveness [24,25].
In the present study, both the double-dose and the standard-dose IIV4s were immunogenic against all vaccine strains in primed and unprimed children 6-35 months of age. The primary objective of the study-to fulfill US licensure criteria by demonstrating immunogenic noninferiority of the investigational IIV4 to a licensed IIV4 and acceptable safety of the investigational IIV4-was achieved. Most children receiving the double-dose IIV4 seroconverted (SCRs, 64.9%-85.5%), and most children achieved seroprotection (SPRs, 66.0%-97.0%). Similar immune responses have been achieved with the double-dose IIV4 in children of the same age in small studies conducted in 3 prior seasons [26][27][28].
Greater antibody responses were observed with the double-dose IIV4 compared with the standard-dose, prompting us to perform a post hoc analysis to evaluate whether the double-dose elicited a superior immune response in terms of the CBER criteria usually applied to establish lot-to-lot consistency of influenza vaccines. In this analysis, the double-dose IIV4 did not reach superiority to the standard-dose in the overall population, the older age group (18-35 months), or previously primed children. However, in the younger age group (6-17 months) and in all unprimed children, the double-dose IIV4 met the applied superiority immune response criteria compared with the standard-dose against the B strains. It should be noted that the unprimed group was predominantly 6-17 months of age.
Several previous studies have compared the HI antibody response elicited by a double-dose versus a standard-dose IIV. The results of the present large phase III study contrast with those of a phase II study comparing GSK's double-dose IIV4 with the United States-approved standard-dose inactivated trivalent influenza vaccine (IIV3) (the corresponding IIV3 to the licensed IIV4 comparator used in the present study) [26]. In the phase II study, the immune response with the double-dose and the standard-dose was similar against the strains common to both vaccines, but the sample size was too small to reliably detect differential immunogenicity against the vaccine B strains, especially in children 6-17 months of age [26]. Two other small studies compared the immunogenicity of the United States-approved IIV3 administered as a standard or double-dose to young children in different years, with contrasting results [21,29]. A 2008-09 trial found that a double-dose IIV3 elicited a higher immune response than a standard-dose in vaccine-unprimed children 6-23 months of age, reaching statistical significance in children 6-11 months of age for 2 of 3 vaccine strains [21]. However, a 2010-12 trial found no difference in immunogenicity between standard-dose versus double-dose IIV3s in unprimed children [29]. This prior experience highlights the necessity for trials of adequate size to reliably establish treatment benefit, and it suggests that observations made in 1 year may not be repeated in other years, because the baseline All solicited injection-site symptoms were considered related to vaccination. c Grade 3 events were defined as follows: pain: child cried when the limb was moved or the limb was spontaneously painful; redness and swelling: >100 mm surface diameter; drowsiness and irritability/fussiness: prevented normal activity; loss of appetite: did not eat at all; spontaneously reported symptom: prevented normal activity. d In the double-dose group, seizures occurred 5, 50, 88, 106, and 168 days after the first vaccine dose. In the standard-dose group, 1 seizure occurred 178 days after the first vaccine dose and the others 39, 74, and 80 days after the second vaccine dose. All children recovered, and none of the seizures was considered by the investigator to be related to vaccination. e Hospitalization, emergency room visit, medical practitioner visit. f Autoimmune diseases and other inflammatory and/or neurologic disorders that may or may not have an autoimmune etiology, according to a protocol-specified list or investigators' judgment. g Kawasaki's disease in the double-dose group and erythema multiforme in the standard-dose group, neither related to vaccination. h Serious adverse events were defined as any untoward medical occurrence that results in death, is life-threatening, requires hospitalization or prolongs hospitalization, or results in disability or incapacity.
immunity of young children may vary. Furthermore, the dose effect on immunogenicity among IIVs may differ according to their manufacturing process [23,26].
In the present study, the double-dose and standard-dose IIV4s had a similar reactogenicity profile despite the higher antigen content and volume of the double-dose. Injection site symptoms, including pain, occurred at a similar rate in both groups. There was no difference in the rate of fever over the 2-day postvaccination period between the 2 groups. Febrile seizures occurred at a similar rate in both groups, none were reported within 2 days of vaccination, and none were considered related to the vaccine. The finding that the higher antigen dose and volume in this study did not adversely affect tolerability in children confirms previous findings from studies comparing reactogenicity and safety of double-dose versus standard-dose IIVs [21-23, 26, 29], and IIV4s versus IIV3s [26,27,[30][31][32].

CONCLUSIONS
In conclusion, a double-dose IIV4 may afford greater protection in young children against influenza B. Increased protection against influenza B, a potentially serious and life-threatening illness particularly in young children [33], would be a beneficial clinical outcome. Use of the same vaccine dose for all eligible ages would also simplify the annual influenza vaccine campaign and reduce cost [34] and logistic complexity. This study provides evidence to support a change in clinical practice to use a double-dose IIV4 (15 µg per antigen) in all children 6 months of age and older, once that dosing for a vaccine product has been approved.

Supplementary Data
Supplementary materials are available at Journal of the Pediatric Infectious Diseases Society online.