AS03 _A -Adjuvanted Influenza A (H1N1) 2009 Vaccine for Adults up to 85 Years of Age

Abstract

Background. Vaccination of high-risk groups was started shortly after the emergence of the influenza A (H1N1) 2009 pandemic virus.

Methods. Healthy adults were enrolled into 2 age strata: 18–60 years and >60 years, and received monovalent influenza vaccine containing 3.75 µg of A/California/2009 (H1N1) hemagglutinin antigen, adjuvanted with AS03_A. Hemagglutination inhibition assay-based antibody titers against H1N1 vaccine were assessed after 1 vaccine dose (primary endpoint), after which subjects were randomized 1:1 to receive no further vaccination or a second dose. Immunogenicity endpoints were European licensure criteria for influenza vaccines. Exploratory analyses assessed the effect of previous seasonal influenza vaccination on responses to the H1N1 vaccine.

Results. Licensure criteria for immunogenicity were fulfilled after 1 dose of H1N1 vaccine ( N = 240). For subjects 18–60 years of age, previous vaccination against seasonal influenza within the preceding 2 seasons resulted in significantly lower geometric mean titers (adjusted for baseline antibody titer) after 1 or 2 doses of H1N1 vaccine ( P < .001 and P = .003, respectively). Transient mild or moderate injection-site pain was reported by 87.5% and 65.0% of subjects 18–60 years of age and >60 years of age, respectively, after the first dose, and in 63% of subjects overall after the second dose.

Conclusions. A single dose of 3.75 µg hemagglutinin antigen, AS03_A-adjuvanted H1N1 2009 vaccine was immunogenic and well tolerated in adults. In exploratory analyses (of subjects 18–60 years of age), postvaccination antibody titers were lower in subjects who had previously received seasonal influenza vaccination, compared with those who had not. This phenomenon warrants further investigation.

Clinical trials registration. NCT00968526.

The swine-origin influenza A H1N1v strain was first identified in April 2009, and following the rapid worldwide spread of the virus, on 11 June 2009, the World Health Organization (WHO) declared a phase 6 pandemic alert [1, 2]. As of 3 April 2010, the WHO stated that laboratory-confirmed cases of H1N1 2009 pandemic influenza had been reported in 214 countries and were associated with >17,798 deaths [3].

Disease control and prevention agencies recommended that vaccination of high-risk groups should begin as soon as vaccine against the novel H1N1 virus became available [4, 5]. However, meeting the global demand for vaccine within the constraints of the existing vaccine-manufacturing infrastructure involves reducing the amount of antigen required per dose (antigen sparing), which can be achieved by formulating the vaccine with an adjuvant [6, 7]. Experience with H5N1 pandemic vaccines shows that formulation with oil-in-water adjuvant systems allows for antigen sparing [8–10].

Recent experience has shown that robust immune responses are obtained with single doses of H1N1 2009 vaccine formulated with or without adjuvants [11–16]. Furthermore, in recent studies with H1N1 2009 vaccines, the concomitant administration of H1N1 2009 vaccine and Northern Hemisphere trivalent seasonal influenza vaccine had no effect on the immunogenicity of either vaccine [8, 13, 16, 17]. These findings support a coadministration strategy, which is favorable for the administration of vaccines to high-risk individuals for whom both pandemic and seasonal influenza vaccines are indicated.

However, recent trial data regarding H5N1 pandemic influenza vaccines have shown that previous receipt of seasonal influenza vaccination may lower immunogenicity to subsequent pandemic influenza vaccination [18, 19]. Moreover, although giving pandemic and seasonal influenza vaccination together does not appear to influence immunogenicity, the previous receipt of Northern Hemisphere trivalent influenza vaccine for any season between 2004–2005 and 2007–2008 was recently reported to reduce geometric mean titers (GMTs) against subsequent doses of nonadjuvanted H1N1 2009 vaccine [13]. As mass vaccination programs cover populations with widely heterogeneous seasonal vaccination histories, further characterization of the influence of exposure to previous seasonal influenza vaccination on immunogenicity against H1N1 2009 vaccine is needed.

We describe hemagglutination inhibition (HI) antibody responses in healthy adults 18–85 years of age after 1 or 2 doses of 3.75 µg A/California/2009 (H1N1) hemagglutinin antigen, with an oil-in-water adjuvant (AS03_A). We conducted exploratory analyses to assess the effect of previous trivalent seasonal influenza vaccination history on immune responses to AS03_AH1N1 2009 vaccine for subjects 18–60 years of age.

METHODS

This phase III, open-label study assessed the immunogenicity, reactogenicity, and safety of 1 or 2 doses of 3.75 µg hemagglutinin antigen, A/California/2009 (H1N1)-like candidate vaccine adjuvanted with AS03_A in healthy subjects 18–60 years of age ( n = 120) and in healthy subjects 160 years of age ( n = 120). The primary objective was to demonstrate that the immune response after 1 dose of H1N1 vaccine fulfilled the immunogenicity criteria recommended by European guidelines for the evaluation of candidate influenza vaccines (European Medicines Agency, Committee for Human Medicinal Products [CHMP]) [20].

The study was conducted at the Vaccine and Infectious Disease Institute, Centre for the Evaluation of Vaccination (WHO Collaborating Centre), University of Antwerp, Belgium. The protocol was approved by an independent ethics committee, and the study was conducted in accordance with the International Conference on Harmonization guidelines of Good Clinical Practice and with the Declaration of Helsinki. All subjects provided informed written consent (ClinicalTrials.gov Identifier: NCT00968526).

Vaccines. The AS03_A-adjuvanted, monovalent, split-virion H1N1 vaccine was manufactured by GlaxoSmithKline (GSK) Biologicals. The candidate vaccine was developed from the reassortant reference virus generated from the A/California/7/2009 (H1N1)v-like strain (NYMC X-179A; New York Medical College), distributed by the US Centers for Disease Control and Prevention. The vaccine contained 3.75 µg of hemagglutinin antigen and was adjuvanted with AS03_A, which contains 11.86 mg tocopherol, 10.69 mg squalene, and 4.86 mg polysorbate 80 [21]. The vaccine was produced according to licensed manufacturing and testing processes for seasonal trivalent influenza vaccine and H5N1 pandemic vaccines [22].

Subjects and study schedule. Men and women ≥18 years of age who had not received an investigational drug or any vaccines within 30 days of the study were eligible for inclusion. Eligible participants were either healthy or had controlled chronic illness. Women of child-bearing age gave a history of reliable contraceptive practices. The Higher Health Council guidelines in Belgium recommend that individuals >50 years of age and healthcare workers be vaccinated against seasonal influenza; subjects in these risk groups were eligible for the study if they fulfilled the inclusion criteria. All subjects were systematically questioned about previous seasonal influenza vaccination, and the investigator relied on the subject's recall for the information. Subjects were recruited from the University Hospital and by using the study center database.

An internet-based algorithm was used to ensure that an equal number of subjects were enrolled in the 2 age strata (18–60 years and >60 years). In the 18–60 years of age stratum, at least 40% of subjects were 18–40 years of age, and at least 40% were 41–60 years of age. In the >60 years of age stratum, at least 40% of subjects were 60–70 years of age, and at least 20% were >70 years of age. Following reports that 1 dose of H1N1 vaccine may meet the immunogenicity endpoints, the protocol was amended, and on day 21, subjects were randomized 1:1 to receive no further vaccination or a second dose. Subjects provided a new informed written consent before entry into the second-dose phase of the study.

Subjects received H1N1 vaccine administered intramuscularly (deltoid of nondominant arm). A blood sample was taken for serologic assessment before vaccination (day 0) and 21 and 42 days after the first vaccination.

Immunogenicity. The immunogenicity outcomes were (1) the proportion of subjects with antibody titers of ≥1:40 (seroprotection rate); (2) the proportion of subjects who were seronegative at baseline with postvaccination antibody titers of ≥1:40, or the proportion who were seropositive at baseline with ≥4-fold increase in antibody titer postvaccination (seroconversion rate); and (3) the geometric mean of the within-subject ratios of pre-and postvaccination reciprocal antibody titers (geometric mean ratio). The primary immunogenicity endpoint was day-21 antibody titers in the day-21 immunogenicity cohort; the secondary endpoints were day-42 responses in subjects who received 1 or 2 doses.

In each age stratum, exploratory analyses were performed to investigate the immunogenicity endpoints according to seasonal influenza vaccination history, including the current influenza season (2009–2010) and any of the preceding 2 influenza seasons (2007–2008 and 2008–2009). The immunogenicity of the H1N1 vaccine was assessed by use of HI assays to measure antibody titers against the A/California/2009 (H1N1) hemagglutinin antigen, according to an established method [23]. Hemagglutinin antibody titers were measured in thawed frozen samples by use of a standardized and validated micromethod that used 4 hemagglutination-inhibiting units of the antigen and a 0.45% chicken erythrocyte suspension. The assays were performed at the central GSK laboratories.

Reactogenicity and safety. The safety endpoints were to describe solicited and spontaneously reported adverse events. Solicited local and general events were recorded by subjects using diary cards for 7 days after the vaccination. Severity grading was based on a standard scale: the diameters of injection-site redness and swelling were >20–50 mm (grade 1), >50–100 mm (grade 2), and >100 mm (grade 3), and the intensity of other symptoms were graded as follows: “easily tolerated” (“on touch” for injection-site pain [grade 1]), “interferes with normal activity” (“when limb is moved” for injection-site pain [grade 2]), and “prevents normal activity” (“significant pain at rest” for injection-site pain [grade 3]). All local solicited events were considered H1N1 vaccine related, and the investigator provided causality assessments for solicited general events. Assessment of spontaneously reported adverse events was planned from day 0 to day 84, and serious adverse events, and events of special interest, were to be assessed prospectively from day 0 to 1 year postvaccination. Spontaneously reported adverse events up to day 42 are provided in this report.

Statistics. On basis of recent experience with AS03_A-adjuvanted pandemic influenza vaccines, it was estimated that a sample size of 120 subjects per age stratum would provide ≥95% power to demonstrate that 1 dose of the H1N1 vaccine fulfills influenza vaccine licensure criteria for immunogenicity overall [14, 24]. The safety population included all subjects who received the H1N1 vaccine. The primary immunogenicity assessment included subjects in the safety population with evaluable serum samples at day 0 and day 21 (day-21 immunogenicity cohort). The day-42 immunogenicity assessment was performed on subjects with serum samples available at day 21 and day 42 (day-42 immunogenicity cohort).

To fulfill CHMP licensure criteria for immunogenicity, the point estimates in subjects 18–60 years of age and in subjects ≥60 years of age needed to be as follows: a seroconversion rate of ≥40% and ≥30%, respectively; a seroprotection rate of ≥70% and ≥60%, respectively; and a geometric mean ratio of ≥2.5 and ≥2.0, respectively [20]. Immunogenicity and reactogenicity results were described as point estimates with 95% confidence intervals (CIs). In the exploratory analysis, the GMT ratio and the 95% CI between the groups with or without receipt of previous seasonal influenza vaccination in the previous 2 years were obtained by use of analysis of covariance for logarithm-transformed titers with log-prevaccination results as the covariate.

RESULTS

Subjects. A total of 240 subjects were enrolled during the period from 8 September to 16 September 2009 (Figure 1). The mean age was 39.7 years (range, 19–60 years) in the younger group and 69.1 years (range, 61–85 years) in the older group. No subject had received trivalent influenza vaccine for the 2009–2010 season. In the age group of 18–60 years, 53.3% of subjects had received seasonal influenza vaccine in the preceding 2 influenza seasons; in the age group of >60 years, 87.5% of subjects had received seasonal influenza vaccine in the preceding 2 influenza seasons (Table 1).

Of the 120 subjects 18–60 years of age, 10 (8.3% [95% CI, 4.1%–14.8%]) had baseline antibody titers of ≥1:40. Of the 120 subjects >60 years of age, 11 (9.2% [95% CI, 4.7%–15.8%]) had baseline antibody titers of ≥1:40 (Table 2). There was no clear relationship between seasonal influenza vaccine history and baseline antibody titers against the H1N1 2009 vaccine.

Immunogenicity. After 1 dose (day 21), all immunogenicity outcomes fulfilled the licensure criteria for candidate influenza vaccines (Table 2). The day-21 seroprotection rate was 97.5% (95% CI, 92.9%–99.5%) in subjects 18–60 years of age and 87.4% (95% CI, 80.1%–92.8%) in subjects >60 years of age. The immune responses persisted at day 42 among subjects who received only 1 dose (Table 3). At day 42 in the 2-doses group, the seroprotection rate was 100% (95% CI, 94.6%–100%) in subjects 18–60 years of age and 98.5% (95% CI, 92.0%–100%) in subjects >60 years of age (Table 3).

At day 21 in subjects 18–60 years of age, the GMT adjusted for baseline antibody titer was significantly lower in subjects who had received seasonal influenza vaccine within the preceding 2 influenza seasons, compared with subjects who had not (Table 4). The adjusted GMT ratio at day 21 was 0.42 (95% CI, 0.28–0.64; P < .001 ). The effect was also observed at day 42 in subjects 18–60 years of age who had received a second dose, with an adjusted GMT ratio of 0.60 (95% CI, 0.41–0.85; P = .003). A similar trend was observed in the age group of >60 years, but too few subjects had not received previous seasonal influenza vaccination ( n = 15), compared with those who had ( n = 105 ), to allow a statistical comparison (Table 4). In the age group of 18–60 years, day-21 and day-42 GMTs were numerically lower in subjects with previous vaccination, relative to those with no previous vaccination, regardless of baseline antibody titer (ie, <1:10 and ≥1:10) (Table 4 and Figure 2).

Reactogenicity and safety. Solicited adverse events during the initial 7-day postvaccination period are shown in Table 5. In the total vaccinated cohort after the first dose ( N = 240), local reactions (solicited and unsolicited) were reported more frequently in subjects 18–60 years of age (87.5% [95% CI, 80.2%–92.8%]) than in subjects >60 years of age (68.3% [95% CI, 59.2%–76.5%]). Injection-site pain was the most frequent local reaction during the 7-day postvaccination period, with a mean duration of 3.1 days in subjects 18–60 years of age and 2.8 days in subjects >60 years of age. The incidence of redness and swelling was low in both age groups. There were no grade3 local reactions after the first dose. Local reactions after the second dose ( n = 138 ) were similar to those after the first dose, but with 2 reports of severe pain and 1 report of severe swelling.

In the total vaccinated cohort after 1 dose, 64.2% (95% CI, 54.9%–72.7%) of subjects 18–60 years of age and 44.2% (95% CI, 35.1%–53.5%) of subjects >60 years of age experienced general reactions (solicited and unsolicited) during the 7-day postvaccination period. After 1 dose in subjects 18–60 years of age and in subjects >60 years of age, general reactions were most frequently fatigue (35.8% and 21.7%, respectively), headache (36.7% and 18.3%, respectively), and muscle aches (24.2% and 20.8%, respectively). The incidences of solicited general reactions after the first and second doses were similar.

Up to day 42, spontaneously reported adverse events occurred in 46 subjects (45.1%) who received 1 dose and in 61 subjects (44.2%) who received 2 doses. The most common spontaneously reported adverse events in subjects who received 1 or 2 doses were nasopharyngitis (12 subjects who received 1 dose and 14 subjects who received 2 doses) and headache (5 subjects who received 1 dose and 6 subjects who received 2 doses). Grade-3 unsolicited adverse events occurred in 5 subjects (4.9%) who received 1 dose and in 12 subjects (8.7%) who received 2 doses.

With regard to serious adverse events, in subjects who received 1 dose, there was 1 case of headache, and in those who received 2 doses, there was 1 case of a trial fibrillation and 1 case of intervertebral disc disorder. The serious adverse events were not considered by the investigators to be vaccine related. There were no adverse events of special interest.

DISCUSSION

A single 3.75-µg dose of AS03_A-adjuvanted H1N1 2009 vaccine elicited immune responses that fulfilled the European licensure immunogenicity criteria for immunogenicity of influenza vaccines [20]. At day 21, HI antibody titers of ≥1:40 were achieved by 97.5% of subjects 18–60 years of age and by 87.4% of subjects >60 years of age. Similar immune responses were observed at day 42 in subjects who received a second dose of vaccine, and immune responses persisted at day 42 in subjects who did not receive a second dose.

Mild or moderate injection-site pain was the most common solicited postvaccination event, and the incidence of redness and swelling was low. Three subjects reported a grade-3 local reaction after the second dose. The most frequent general reactions were fatigue, headache, and muscle aches. These findings are consistent with previous experience with avian-origin pandemic influenza vaccines, in which formulation with AS03_A increased the incidence of postvaccination reactions, compared with nonadjuvanted vaccine [22, 25]. Prospective assessments are ongoing to establish the long-term safety pro le of the H1N1 2009 vaccine.

On the basis of numerous clinical studies, it is widely accepted that vaccination history does not negatively influence protection associated with subsequent seasonal influenza vaccination [26, 27]. Nonetheless, reduced HI antibody responses in subjects with a history of influenza vaccination, compared with those without, have been repeatedly observed in clinical studies of trivalent seasonal influenza vaccines [28–30]. Furthermore, recent experience with AS03_A-adjuvanted avian-origin H5N1 influenza vaccine in adults showed that vaccination with nonadjuvanted pandemic influenza vaccine was associated with reduced immune responses to booster doses of AS03_AH5N1 vaccine, compared with responses in subjects who received the AS03_A-H5N1 vaccine for the first time [18]. And in a recent pediatric study, postvaccination neutralizing antibody titers after 2 doses of alum-adjuvanted H5N1 vaccine were significantly higher in children who had not previously received nonadjuvanted seasonal influenza vaccination, compared with those who had [19].

To investigate this phenomenon, we conducted exploratory analyses on the immunogenicity of the AS03_A-adjuvanted H1N1 2009 vaccine in subjects with and without a history of seasonal influenza vaccination. Although all immunogenicity endpoints were met, we found that, among subjects 18–60 years of age, the adjusted GMTs at day 21 and day 42 were significantly lower among subjects who had previously received at least one seasonal influenza vaccination during the preceding two seasons, compared with those who had not. It should be noted, however, that the sample was relatively small at day 42 (38 subject with previous vaccination and 28 subjects without).

A relationship between baseline serostatus and postvaccination antibody titers has been previously described, with reduced postvaccination responses observed in subjects with higher pre-vaccination titers of antibodies derived from natural influenza infection or previous influenza vaccination [31]. Contrary to this, in our study, among subjects who had previously received trivalent influenza vaccination, antibody titers after H1N1 2009 vaccine were numerically lower in subjects who were seronegative at baseline, compared with subjects who were seropositive. This suggests that the observed effect of previous seasonal influenza vaccination on subsequent responses to H1N1 2009 vaccination is not conditioned by the existence of HI antibodies at baseline.

Various studies have shown that subjects who have previously received seasonal influenza vaccine may have reduced antibody titers following subsequent pandemic influenza vaccination, yet the mechanisms underlying this phenomenon remain unknown [13, 18, 19]. It can be speculated that previous nonadjuvanted seasonal influenza vaccination led to a “skewing” of the T cell and/or B cell repertoires toward epitopes specific for those vaccines, making it more difficult to generate or select highly H1N1 pandemic-specific T and B cells after subsequent vaccination with the H1N1 pandemic vaccine, and translating into lower HI titers. Indeed, it has been shown that seasonal influenza vaccination induces a profound selection of antigen-specific B cells, with major changes in the B cell repertoire as a result [32].

Another hypothesis, not necessarily exclusive from the former, could be that previous seasonal influenza vaccination prevented natural seasonal H1N1 infection and thereby altered the development of heterosubtypic immunity against the H1N1 2009 strain. Under the assumption that virally induced immunity induces optimal cross-reactive responses, this would eventually result in a lower responsiveness state to subsequent H1N1 2009 pandemic vaccination in subjects who were previously vaccinated against seasonal influenza, compared with subjects who were not. This hypothesis has been verified in a mouse model, in which the heterosubtypic immunity against a pandemic H5N1 strain was blunted after prevention of H3N2 infection through subunit vaccination [33]. Explanations for the effect of previous seasonal influenza vaccine on immune responses to pandemic influenza vaccine remain speculative, and further investigation is warranted.

To conclude, a single dose of AS03_A-adjuvanted H1N1 2009 vaccine (3.75 µg hemagglutinin antigen) elicited immune responses that fulfilled the European licensure criteria for immunogenicity of influenza vaccines in adults. Because immunogenicity criteria were reached after 1 dose, a second dose may not be necessary. Exploratory analyses in subjects 18–60 years of age showed that seasonal influenza vaccination within the previous 2 seasons was associated with reduced responses to the H1N1 2009 vaccine.

Acknowledgments

We are grateful to the New York Medical College for providing the vaccine virus strain. We are indebted to the participating study volunteers, study doctors, trial nurses, and laboratory technicians at the study site as well as to the sponsor's project staff for their support and contributions throughout the study. We are grateful to all teams of GSK Biologicals for their contribution to this study, Karl Walravens and the clinical and serological laboratory teams, Julie De Wever for preparation of the study protocol and related study documentation, Anaëlle Delhage for global study management, Dorothy Slavin (clinical safety representative), Carine Maggetto (clinical data coordinator), Walthere Dewé, Olivier Ghys, and Vinod Bambure for input on statistical analysis, and Edith Lépine for project management. We would like to address special thanks to Paul Gillard and Robbert Van der Most for their critical reading of the manuscript and for their helpful suggestions. Finally, we thank Annick Moon (of Moon Medical Communications, Oxford, UK), who provided medical writing services, and Isabelle Gautherot for editorial assistance and manuscript coordination.

Potential conflicts of interest. P.V.D. acts as chief and principal investigator for vaccine trials conducted on behalf of the University of Antwerp, for which the university obtains research grants from several vaccine manufacturers. F.R., E.H., and T.V. report that they are employees of GSK Biologicals. E.H. reports that he owns equity or stock options. GSK Biologicals sponsored the study and, together with the authors, was involved in all stages of the study, including analysis of data. GSK Biologicals also paid the costs associated with the development and publication of the present study.

Figures and Tables

Figure 1.

Subject enrollment in study to evaluate immunogenicity and safety of an investigational influenza vaccine (H1N1) in adults (8–16 September 2009).

Figure 2.

Hemagglutination inhibition assay-based, day-21 reverse cumulative distribution curves in subjects 18–60 years of age after 1 dose of A/California/2009 (H1N1) hemagglutinin antigen, by baseline serostatus and seasonal influenza vaccination history in the day-21 immunogenicity cohort.

Table 1.

Baseline Characteristics of Study Subjects, by Age Group (Total Vaccinated Cohort)

Table 2.

Hemagglutination Inhibition Assay-Based Immune Responses against A/California/2009 (H1N1) Hemagglutinin Antigen after 1 Dose of AS03_A-H1N1 2009 Vaccine in the Day-21 Immunogenicity Cohort

Table 3.

Hemagglutination Inhibition Assay-Based Immune Responses against A/California/2009 (H1N1) Hemagglutinin Antigen after 1 or 2 Doses of AS03_A-H1N1 2009 Vaccine in the Day-42 Immunogenicity Cohort

Table 4.

Hemagglutination Inhibition Assay-Based Immune Responses against A/California/2009 (H1N1) Hemagglutinin Antigen after 1 Dose (Day-21 Immunogenicity Cohort) or 2 Doses (Day-42 Immunogenicity Cohort) of AS03_A-H1N1 2009 Vaccine, According to Age Group, Seasonal In uenza Vaccination History, and Antibody Titers

Table 5.

Data on Solicited Local and General Adverse Events Occurring within 7 Days after Receipt of AS03_A-H1N1 2009 Vaccine, According to Age Group in the Total Vaccinated Cohort after the First Dose and in the Vaccinated Cohort after the Second Dose