Vietnam indicates avian influenza A/Vietnam/1203/2004(H5N1); Anhui indicates avian influenza A/Anhui/01/2005(H5N1).
Anhui indicates avian influenza A/Anhui/01/2005(H5N1). Participants received a priming vaccination on day 0 and a booster vaccination on day 28.
aThe 90-μg dose was administered as 2 injections of 0.75 mL 2 cm apart in the same arm.
bParticipants who were randomized to receive placebo received 1 injection of 0.5 mL, 1 injection of 0.75 mL, or 2 injections of 0.75 mL to serve as blinded controls for the 3.75-, 7.5-, and 15-µg 0.5 mL doses, the 45-µg 0.75 mL dose, or the 90-µg dose administered as two 0.75-mL injections.
Anhui indicates avian influenza A/Anhui/01/2005(H5N1); NA, not applicable (no 90-µg dose group with adjuvant). Error bars indicate 95% CI.
eTable 1. Demographic Data by Study Group
eTable 2. Frequency of Previously Primed Participants Reporting Solicited Local or Systemic Reactions Within Seven Days After Either Dose 1 or Dose 2 of the Indicated Dose of Vaccine With or Without Adjuvant
eTable 3. Frequency of H5N1 Naïve Participants Reporting Solicited Local or Systemic Reactions Within Seven Days After Either Dose 1 or Dose 2 of the Indicated Dose of Vaccine With or Without Adjuvant
eFigure 1. HAI and Neutralizing Antibodies After Low Dose Boosting One Year After Vaccine Priming
eFigure 2. HAI and Neutralizing Antibodies in Naïve Participants After Two Doses of Vaccine
Customize your JAMA Network experience by selecting one or more topics from the list below.
Belshe RB, Frey SE, Graham IL, et al. Immunogenicity of Avian Influenza A/Anhui/01/2005(H5N1) Vaccine With MF59 Adjuvant: A Randomized Clinical Trial. JAMA. 2014;312(14):1420–1428. doi:10.1001/jama.2014.12609
The need to respond quickly to potential influenza pandemics is important. Immunologic priming (initial presentation of an antigen to allow antibody responses on revaccination) with vaccine directed toward an older avian influenza H5 strain might lead to secondary antibody responses to a single dose of more current H5 avian influenza vaccine.
To assess priming with the older avian influenza A/Vietnam/1203/2004(H5N1) (Vietnam) vaccine and to conduct dose-response studies with vaccine directed against the more contemporary H5N1 avian influenza virus, influenza A/Anhui/01/2005 (Anhui).
Design, Setting, and Participants
Multicenter US randomized clinical trial beginning in June 2010 with follow-up continuing through October 2011 enrolling 72 healthy adults who were vaccinated 19 to 25 months previously with the Vietnam vaccine and 565 vaccine-naive adults.
Participants who were previously vaccinated with 90 µg of unadjuvanted Vietnam vaccine were randomly assigned to receive 3.75 µg of avian influenza Anhui vaccine with or without MF59 adjuvant, stratified by 1 vs 2 previous doses (1 dose: n = 18 with MF59 and n = 17 without; 2 doses: n = 19 with MF59 and n = 18 without). Vaccine-naive individuals were randomly assigned to receive Ahnui vaccine with or without MF59 adjuvant in 1 of 5 doses (3.75 µg [n = 55 with MF59 and n = 59 without], 7.5 µg [n = 51 with MF59 and n = 57 without], 15 µg [n = 48 with MF59 and n = 44 without], 45 µg [n = 47 with MF59 and n = 47 without], or 90 µg [n = 57 without adjuvant]) or placebo (n = 100) given at days 0 and 28.
Main Outcomes and Measures
The primary immunogenicity outcome was hemagglutination inhibition assay (HAI) titer against each vaccine antigen 1 month (day 28) and 6 months (day 180) after last vaccination. The primary safety outcomes were local and systemic adverse events on days 0 to 7 after each vaccination and serious adverse events.
Previously vaccinated participants manifested secondary antibody responses after receipt of low-dose Anhui vaccine (“boosting”); by day 28, 21% to 50% developed HAI responses of 1:40 or greater. Use of adjuvant was not associated with increased HAI responses. Among vaccine-naive participants (n = 565), the optimum dose was 7.5 µg of antigen with adjuvant (geometric mean titer [GMT], 63.3; 95% CI, 43.0-93.1). The greatest response to unadjuvanted antigen was seen at the highest dose, 90 µg (GMT, 28.5; 95% CI, 19.7-41.2). Local or systemic reactions occurred, respectively, in 40 (78%) and 25 (49%) of 51 participants who received 7.5 µg plus adjuvant vs 50 (88%) and 29 (51%) of 57 who received 90 µg of unadjuvanted vaccine. In general, antibodies were short-lived, and by day 180, HAI titers had decreased to less than 1:20 in all treatment groups.
Conclusions and Relevance
Previous receipt of a single dose of influenza A(H5N1) Vietnam vaccine was associated with sufficient immunologic priming to facilitate antibody response to a different H5N1 antigen using low-dose Anhui (booster) vaccine. In participants who had not previously received H5 vaccine, low-dose Anhui vaccine plus adjuvant was more immunogenic compared with higher doses of unadjuvanted vaccine.
clinicaltrials.gov Identifier: NCT00680069
Planning for influenza pandemics is of vital importance, with vaccine being a cornerstone of control efforts.1,2 The avian influenza A/Vietnam/1203/2004(H5N1) (Vietnam) vaccine has been approved by the US Food and Drug Adminstration for use as a 2-dose schedule in unadjuvanted form with 90 µg per dose.1 However, the relatively low frequency of immune response seen in previous studies of inactivated unadjuvanted influenza A(H5N1) vaccines in humans,3-8 the continued antigenic drift of H5N1 in birds, and the likely need for 2 doses of vaccine pose significant obstacles to effective approaches for control during an actual pandemic.
Previously, we evaluated the effect of vaccine schedule and antigenic variant on the immunogenicity of the subunit influenza H5 antigen Vietnam, combined with or subsequent to an antigentic variant H5, A/Indonesia/05/2005.8 With mixed vaccine schedules, the initial vaccination helped improve response to a second, similar antigen (“priming” for a heterologous boost). An accelerated vaccine schedule at days 0 and 14 was as immunogenic as vaccination on days 0 and 28, but both were inferior to vaccination times 0 and 6 months for inducing antibody responses after the second dose of vaccine. The 6-month interval between doses was associated with superior immunogenicity for subsequent vaccinations (boosters) with the same or different H5 antigens.8
The objective of the present study was to evaluate the immunogenicity and safety of inactivated influenza vaccine made from a more recent avian influenza A(H5N1) variant, A/Anhui/01/2005 (Anhui). Participants vaccinated 19 to 25 months previously with 1 or 2 doses of Vietnam vaccine received low-dose Anhui vaccine with or without MF59 adjuvant. We also studied the antibody response of 2 doses of Anhui vaccine at 5 different dose levels with or without MF59 in H5N1 vaccine–naive participants.
The institutional review board of each participating institution approved the protocol (Supplement 1) and written informed consent was obtained from each participant.
The rgA/Vietnam/1203/2004 batch 04-067 clade 1 vaccine was previously administered as a monovalent subunit influenza vaccine (previously vaccinated or “primed” participants) consisting of 90 µg of hemagglutinin in each dose; some volunteers had received 1 dose and others 2 doses separated by 1 month; the previous vaccine was given a mean of 22 months (range, 19-25 months) prior to boosting in the current study.
A monovalent inactivated surface antigen influenza A(H5N1) (modified hemagglutinin and neuraminidase of A/Anhui/01/2005) vaccine was used for the current study. This vaccine was manufactured by Novartis Vaccines and Diagnostics. Vaccine contained purified hemagglutinin and neuraminidase surface antigens from the influenza A/Anhui/01/2005(H5N1)-PR8-IBCDC-RG6 virus.
Anhui vaccine was supplied in 5-mL multidose vials with thimerosal (0.01% weight-volume ratio) as a preservative. At each dose level (Figure 1 and Figure 2), both unadjuvanted and MF59C.1-adjuvanted formulations were manufactured; 8 formulations were used: 7.5, 15, 30, or 60 µg/mL of Ahnui H5N1 hemagglutinin without adjuvant and 7.5, 15, 30, or 60 µg/mL of Anhui H5N1 hemagglutinin with MF59 adjuvant. Placebo consisted of sterile normal saline for injection (0.9% sodium chloride injection, USP preservative-free) supplied by Fisher Bioservices. The MF59 adjuvant was an oil-in-water emulsion composed of squalene as the oil phase, stabilized with the surfactants polysorbate 80 and sorbitan trioleate, in citrate buffer.
Vaccine doses of 3.75, 7.5, 15, and 45 µg were delivered by intramuscular injection (0.5 mL for 3.75-, 7.5-, and 15-µg doses and 0.75 mL for 45-µg dose). A vaccine dose of 90 µg was delivered by 2 intramuscular injections in the same arm of 45 µg/0.75 mL each. Participants randomized to receive saline placebo received intramuscular injections in 1 of 3 different volumes (1 injection of 0.5 mL, 1 injection of 0.75 mL, or 2 injections of 0.75 mL) depending on randomization for blinding purposes. Because study products differed in appearance and were administered in different volumes, syringes were concealed and participants were asked to look away during injection. Clinical research staff and participants who were assigned to receive the 90-µg dose or saline placebo as 2 injections in the same arm were partially unblinded.
We conducted a randomized trial at 8 clinical sites in the United States, with enrollment beginning in June 2010. Healthy adult participants aged 18 through 49 years (age at initial priming vaccination) who had previously received either 1 or 2 doses of vaccine were reenrolled for the present study.8 These previous study participants were randomized in a 1:1 fashion to receive a single dose of 3.75 µg of Anhui vaccine with or without adjuvant, stratified by number of previous Vietnam priming doses (1 vs 2 doses) to ensure balanced study groups (Figure 1).
Additionally, healthy influenza A(H5N1) vaccine–naive participants aged 18 through 49 years were recruited using identical entry criteria as the previous study.8 These H5N1 vaccine–naive participants were randomly assigned with equal allocation to receive 2 doses of 3.75 µg, 7.5 µg, 15.0 µg, or 45.0 µg of Ahnui vaccine with or without adjuvant, 90.0 µg of Ahnui vaccine without adjuvant, or placebo (in 1 of 3 different volumes) on day 0 (priming dose) and day 28 (booster dose) (Figure 2).
The randomization sequence was generated by the trial statistician using blocked randomization. On enrollment, each participant was assigned a randomization number from the electronic data entry system corresponding to a treatment on a randomization list available to the unblinded vaccine administrator.
Race and ethnicity were assessed to track diversity of participants and permit ad hoc group-specific assessments of outcomes if needed. The participants’ race and ethnicity were classified according to the participants’ responses using options provided by study staff. Race options were American Indian/Alaskan Native, Asian, native Hawaiian or other Pacific Islander, black or African American, and white. Ethnicity options were Hispanic or Latino and non-Hispanic or non-Latino.
Participants were followed up for safety, reactogenicity, and immune responses after vaccination. Participants were sent home with a memory aid on which they recorded their daily oral temperature, systemic or local reactions, and any other adverse events or serious adverse events that occurred within the week following vaccination. Serum samples were collected prior to vaccination and on days 7, 14, 21, and 28 after each vaccination and on day 180 after the last vaccination. All participants were followed up for 12 months after the last vaccination, with the last follow-up visit on October 21, 2011.
Hemagglutination inhibition assay (HAI) antibody titers and serum neutralizing antibody titers (assessed by plaque reduction) were measured against the influenza A/Vietnam/1203/2004 and influenza A/Anhui/1/2005 viruses by the Southern Research Institute as previously described.1,2,9,10 For analysis, values below the limit of detection (1:10 for each assay) were imputed as half the limit of detection.
The primary immunogenicity outcome was HAI titer against both Vietnam and Anhui antigens 1 month (28 days) and 6 months after last vaccination. A secondary outcome was neutralizing antibody titers at 28 days after last vaccination. The primary safety outcomes were local and systemic adverse events on days 0 to 7 after each vaccination and serious adverse events throughout the trial. The sample size for each group was chosen based on logistic considerations of available participants previously vaccinated with Vietnam vaccine and also for the purpose of gathering information on the safety, dose, and optimal schedule of vaccination strategies; the sample size was not based on testing a formal hypothesis.
For both HAI and neutralizing antibody titers at each time point, the proportion of participants with a postvaccination titer of 1:40 or greater, the geometric mean titer, and the geometric mean fold increase are presented, with 95% confidence intervals. Exact (Clopper-Pearson) confidence intervals are reported for all proportional end points. Immunogenicity analyses were performed for the per-protocol analysis subset.
Several post hoc exploratory analyses of adverse events were considered. Logistic regression models were fit to examine the association of antigen dosage, receipt of adjuvant, and their interaction with the occurrence of any systemic symptom or the occurrence of any moderate or severe systemic symptom among vaccine-naive participants. The rates of systemic reactions were compared between previously vaccinated and previously unvaccinated participants within each adjuvant group using the Fisher exact test. For exploratory analyses, all tests were 2-sided with a significance threshold of α = .05. All analyses were generated using SAS software, version 9.3 (SAS Institute Inc).
A total of 637 participants were enrolled in the study (Figure 1 and Figure 2); 250 (39.2%) were male. The mean age was 33 years (range, 18-51 years). The groups were well balanced by sex, age, and race and ethnicity (Table 1).
Participants vaccinated in the earlier study with either 1 or 2 priming doses of Vietnam vaccine received a single dose of Ahnui vaccine with or without MF59 adjuvant (Table 2 and eFigure 1 in Supplement 2). By day 7 after Ahnui vaccination, antibody titers had increased in the previously vaccinated participants. Antibody titers peaked by day 14 after Ahnui vaccination and declined thereafter. By day 28, 21% to 50% of participants who had previously received Vietnam vaccine and were revaccinated 19 to 25 months later with a single dose of Anhui vaccine with or without MF59 adjuvant developed HAI responses of 1:40 or greater. Use of adjuvant was not associated with increased HAI responses relative to unadjuvanted vaccine.
However, as indicated by neutralizing antibody, a higher proportion of participants responded to receipt of adjuvanted vaccine, although this difference did not achieve statistical significance; at day 28, among participants previously vaccinated with at least 1 dose of Vietnam vaccine, 85% (95% CI, 71%-95%) who received Anhui vaccine plus MF59 adjuvant compared with 71% (95% CI, 54%-85%; P = .12 by χ2 test) who received Anhui vaccine without MF59 adjuvant manifested a 1:40 or greater neutralizing antibody response. There did not appear to be an immunologic advantage among participants who had previously received 2 doses vs 1 dose of Vietnam vaccine (Table 2 and eFigure 1 in Supplement 2); 83% (95% CI, 59%-96%) and 89% (95% CI, 67%-99%; P = .59 by χ2 test) of participants previously vaccinated with 1 or 2 doses of Vietnam vaccine, respectively, followed by receipt of 3.75 µg of Ahnui vaccine plus MF59 adjuvant manifested a 1:40 or greater neutralizing antibody response.
All influenza A(H5N1) vaccine–naive participants were antibody negative prior to enrollment. Little or no HAI antibody response was observed after the first dose of Anhui vaccine with or without MF59 adjuvant regardless of antigen dose among vaccine-naive participants (eFigure 2 in Supplement 2). However, modest neutralizing antibody titers were induced by 1 dose of adjuvanted vaccine or a high dose of unadjuvanted vaccine (eFigure 2). Participants developed a more vigorous immune response after the second dose of vaccine, with the antibody titers generally reaching their peak at 14 days after the second vaccination (eFigure 2).
Among vaccine-naive participants, by day 28 after the second vaccination (Table 3 and Figure 3), there was a dose response to both preparations, with or without MF59 adjuvant. The optimal dose of Anhui vaccine with adjuvant was 7.5 µg, with peak immunogenicity responses at 14 days after receipt of the second dose (eFigure 2 in Supplement 2); the HAI geometric mean titer against Anhui vaccine was 83 (95% CI, 55-125) at the peak response 14 days after dose 2. Neither the magnitude of antibody response nor the proportion of participants responding to vaccine increased with antigen doses higher than 7.5 µg of Anhui with MF59 adjuvant.
In contrast, among participants who received unadjuvanted vaccine, increasing antigen dose was associated with higher antibody responses, and at none of the doses tested was there a plateau in antibody response (Table 3 and Figure 3). Geometric mean HAI titer at 14 days after the second dose (time of peak response) of 90 µg was 36 (95% CI, 24-54); this response to the highest antigen dose without MF59 adjuvant was lower than the antibody response to 7.5 µg of antigen with adjuvant. In general, antibodies were short-lived, and by day 180 after the second dose, HAI geometric mean titers had declined to less than 1:20 across all treatment groups.
Participants previously vaccinated with Vietnam vaccine and then given Anhui vaccine developed both HAI antibody and neutralizing antibody to both antigens (Table 2 and eFigure 1 in Supplement 2). In contrast, influenza A(H5N1) vaccine–naive participants who received 2 doses of Anhui vaccine developed little or no antibody to Vietnam hemagglutinin (Table 3 and eFigure 2 in Supplement 2).
Overall, Anhui vaccine preparations were well tolerated at all dose levels with and without MF59 adjuvant. When adverse events occurred, the majority were mild in nature (Table 4 and eTables 2 and 3 in Supplement 2). There were no serious adverse events attributed to the vaccine.
Adverse reactions that were moderate or severe in nature were few in number; 1 severe local reaction was reported after dose 1 and 2 after dose 2. The proportion of influenza A(H5N1) vaccine–naive participants reporting adverse events is shown in Table 4. Results of the post hoc exploratory logistic regression model fit for these 465 participants suggest no association of antigen dose or adjuvant with the proportion of participants reporting systemic adverse events. A similar model fit for moderate or severe systemic reactions found that the percentage of participants reporting such events who received adjuvanted vaccine (10%; 95% CI, 7%-16%) was less than those who received unadjuvanted vaccine (18%; 95% CI, 14%-23%; P = .03) but did not identify an association with antigen dose.
Exploratory analyses also suggested that among participants who received adjuvanted vaccine (any dose), a larger portion of previously vaccinated participants (with Vietnam vaccine) reported at least 1 systemic event (primarily mild malaise) after receiving a single dose of Anhui vaccine compared with influenza A(H5N1) vaccine–naive participants after 1 or 2 doses of Anhui vaccine (67% [95% CI, 50%-82%] vs 49% [95% CI, 42%-56%]; P = .048) (eTables 2 and 3 in Supplement 2). The percentage of previously vaccinated participants who received adjuvanted vaccine and reported malaise was twice that of vaccine-naive participants (40% [95% CI, 25%-58%] vs 20% [95% CI, 15%-27%]; P = .01), but the differences observed between these groups for all other systemic symptoms were not statistically significant. Among participants who received unadjuvanted vaccine, there were no statistically significant differences in the rates of systemic events between groups.
Previous receipt of vaccine containing 90 µg of influenza A/Vietnam/1203/2004(H5N1) antigen was associated with secondary antibody responses after vaccination with low-dose influenza A/Anhui/01/2005(H5N1) antigen 19 to 25 months later; antibody responses were observed to both the original Vietnam antigen and the subsequent Anhui boosting antigen. This phenomenon has previously been termed original antigenic sin, and we observed it with the 2 H5N1 antigens; participants who are exposed to a related antigen later in life respond with antibody to the initial (priming) antigen as well as to the second (boosting) antigen. The secondary antibody responses were brisk (responses seen by day 7) and peaked at 14 days after the boosting vaccination. Although most participants did not achieve titers of 1:40 or greater, higher dose boosting may have produced higher titers.
Our previous study demonstrated the value of a long interval (6 months) relative to shorter intervals (≤28 days) for priming for heterologous antibody responses.8 These findings were confirmed in the present study. A prime-boost study with various quantities of unadjuvanted Vietnam antigen after a 6-month interval was performed by Zangwill et al.11 Booster doses of 90-μg of Vietnam were associated with the best responses; 78% achieved 1:40 or greater HAI titers and lower booster doses were associated with less frequent responses of 1:40 or greater. Adjuvanted boosts were not evaluated in that study.11
In the present study, significant dose sparing was associated with the addition of MF59 adjuvant among vaccine-naive participants; among these participants, the advantage of adding an adjuvant to novel influenza vaccine antigens became apparent. The optimum vaccine dose was 7.5 µg of antigen combined with adjuvant. In the absence of adjuvant, 90 µg of unadjuvanted antigen was the most immunogenic antigen dose studied, but this did not achieve the same levels of immunogenicity as the adjuvanted product at 7.5 µg. Higher quantities of vaccine antigen (≥15 µg) in combination with MF59 adjuvant were associated with lower antibody responses compared with the 7.5-µg dose, although this did not achieve statistical significance in the present study. In other studies with other influenza antigens, the same phenomenon was observed, with significant decreases in antibody in the groups given MF59 adjuvant and higher antigen content.5
The adjuvant effect of MF59 is provided by the emulsion but not by its single components.12,13 However, the emulsion provides an effect only when directly mixed with the antigen; administration of the adjuvant at a separate site from the antigen does not result in enhanced immunity. Local inflammatory responses at the administration site consist of neutrophils, monocytes, and macrophages that release chemokines which enhance immunogenicity. As a result, the activated dendritic cells uptake antigen and migrate to local lymph nodes, where helper T cells induce CD4 cell responses and ultimately increased antibody production.13
Boosting previously vaccinated participants with low-dose Anhui antigen with adjuvant was not associated with a significant immunogenicity difference compared with boosting without adjuvant. However, a larger study might show an advantage to adjuvanted vaccine. In this study, previously vaccinated participants experienced more malaise after receiving the adjuvanted Anhui vaccine compared with previously unvaccinated participants; however, all other systemic symptoms occurred with similar frequency between previously vaccinated and unvaccinated participants. Previous studies of MF59 adjuvanted seasonal influenza vaccines or with influenza A(H5N1) vaccines have noted increases in both acute local reactions and acute systemic reactions (mostly mild reactions13) compared with unadjuvanted seasonal influenza vaccine. In this regard, MF59 is particularly attractive as an adjuvant for novel influenza antigen because preexisting immunity would be infrequent during a prepandemic vaccine campaign.
The importance of time in allowing B cells to mature is reiterated in this study. These observations offer support for new approaches for prepandemic vaccine planning. For example, preexposure vaccination of at-risk populations such as influenza A(H5N1) investigators or persons likely to be exposed to novel influenza viruses in the field could allow rapid development of immune response after repeat vaccination with novel antigens as they become available (boosting). Whether priming has value in protecting against death is not known, but it is possible that primed individuals would have better outcomes in the face of influenza A(H5N1) infections. However, as with the 1976 experience with a swine influenza vaccine campaign, the possible occurrence of unexpected serious adverse events (such as Guillain-Barré syndrome) must be considered when balancing risks vs benefits of prepandemic vaccination.14 Certain laboratory scientists, field epidemiologists, and health care workers in areas with emerging avian influenza might elect to receive a prepandemic vaccine; long-term memory after single-dose priming is the anticipated outcome for vaccinated persons, as demonstrated in the present study.
This study has several limitations. First, although this study provides data to help guide prepandemic H5 vaccine dosing with or without MF59 adjuvant in healthy adults, it is not an efficacy study. It is not known what antibody titer would be associated with protection from infection, protection from disease, or protection from death, and each of these end points might have different correlates of protection. For example, the increased secondary response seen after earlier vaccination with a heterologous H5 antigen might correlate with reduced death but not reduced infection. Second, this study is limited to healthy adults; older adults, adults with underlying chronic conditions, pregnant women, and children need to be evaluated for safety and immunogenicity because the dose and safety profiles may be different in these groups. Third, the continued drift of H5 viruses means that the strain of H5 used for the boosting vaccine in this study will need to be updated to keep pace with the emergence of newer H5 viruses. Furthermore, the emergence of H7 avian influenza viruses as human pathogens will require a different priming vaccine than the H5 Vietnam vaccine used in the present study. Fourth, although this study suggests that low-dose H5 antigen plus MF59 adjuvant may have reduced reactogenicity compared with high doses of unadjuvanted H5 vaccine, these observations are limited to common events and do not assess the possibility of uncommon events that might be associated with H5 vaccination with or without adjuvant. That adjuvant may reduce adverse events via dose sparing of antigen needs to be evaluated further in larger studies.
Previous receipt of a single dose of influenza A/Vietnam/1203/2004(H5N1) vaccine was associated with sufficient immunologic priming to increase (boost) response to a different H5N1 antigen using low-dose influenza A/Anhui/01/2005 vaccine. In participants who had not previously received H5 vaccine, low-dose Anhui vaccine plus adjuvant was more immunogenic compared with higher doses of unadjuvanted vaccine.
Corresponding Author: Robert B. Belshe, MD, Division of Infectious Diseases, Allergy, and Immunology, Saint Louis University School of Medicine, 1100 S Grand Blvd, DRC-8, St Louis, MO 63104 (email@example.com).
Correction: This article was corrected on May 2, 2016, for a data error.
Author Contributions: Drs Belshe and Bellamy had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Mulligan, Winokur, Chen, Edwards, Creech.
Acquisition, analysis, or interpretation of data: All authors.
Drafting of the manuscript: Anderson, Mulligan, Chen, Edwards, Bellamy.
Critical revision of the manuscript for important intellectual content: Belshe, Frey, Graham, Jackson, Spearman, Edupuganti, Mulligan, Rouphael, Winokur, Dolor, Woods, Walter, Chen, Turley, Edwards, Creech, Hill.
Statistical analysis: Chen, Hill, Bellamy.
Obtained funding: Mulligan, Winokur, Walter, Edwards.
Administrative, technical, or material support: Belshe, Frey, Jackson, Spearman, Edupuganti, Mulligan, Woods, Walter, Chen, Turley, Creech.
Study supervision: Belshe, Spearman, Edupuganti, Mulligan, Rouphael, Walter, Chen, Turley, Creech.
Conflict of Interest Disclosures: Dr Belshe reports consulting/speaker fees from Medimmune and Merck. Dr Walter reports receipt of grants for unrelated vaccine studies from Novartis, Pfizer, GlaxoSmithKline, and Merck; participation in an advisory board meeting with receipt of fees from Novartis; and consultant fees from Merck. Dr Turley reports stock ownership in Pfizer, Procter and Gamble, CVS Caremark, Abbott Labs, Abbvie, and Hospira and receipt of grants for vaccine trials from Pfizer, Merck, sanofi pasteur, and the National Institute of Allergy and Infectious Diseases (NIAID). Dr Edwards reports receipt of a grant for a vaccine study from Novartis. No other disclosures were reported.
Funding/Support: This research was funded by the National Institutes of Health under NIAID-Funded Vaccine and Treatment Evaluation Unit contracts HHSN272200800003C (Saint Louis University), HHSN2722008800004C (Group Health Research Institute), HHSN272200800005C (Emory University), HHSN272200800008C (University of Iowa), HHSN272200800001C (University of Maryland), N01AI800002 (Baylor University), HHSN272200800007C (Vanderbilt University), and HHSN272200800013C (EMMES Corporation).
Role of the Funders/Sponsors: The NIAID provided advice and guidance during the design and conduct of the study; supervised collection, management, analysis, and interpretation of the data; provided advice on preparation, review, and approval of the manuscript; and encouraged the investigators in the decision to submit the manuscript for publication.