- Email: [email protected]
Phase I and II randomised trials of the safety and immunogenicity of a prototype adjuvanted inactivated split-virus influenza A (H5N1) vaccine in healthy adults
Vaccine 26 (2008) 4160–4167
Contents lists available at ScienceDirect
Vaccine journal homepage: www.elsevier.com/locate/vaccine
Phase I and II randomised trials of the safety and immunogenicity of a prototype adjuvanted inactivated split-virus influenza A (H5N1) vaccine in healthy adults Terry M. Nolan a,∗ , Peter C. Richmond b , Maryanne V. Skeljo c , Georgina Pearce c , d ¨ Gunter Hartel c , Neil T. Formica c , Katja Hoschler , Jillian Bennet c , David Ryan c , e c Kelly Papanaoum , Russell L. Basser , Maria C. Zambon d a
Murdoch Childrens Research Institute, and the Melbourne School of Population Health, University of Melbourne, Carlton, Victoria, Australia School of Paediatrics and Child Health, The University of Western Australia, Perth, Western Australia, Australia c CSL Limited, Parkville, Victoria, Australia d Virus Reference Laboratory, The Health Protection Agency, Colindale, UK e Infectious Diseases Unit, Royal Adelaide Hospital, Adelaide, South Australia, Australia b
a r t i c l e
i n f o
Article history: Received 29 April 2008 Received in revised form 15 May 2008 Accepted 25 May 2008 Available online 13 June 2008 Keywords: Avian Influenza Vaccine Pandemic Prototype
a b s t r a c t Objective: The primary objective was to evaluate the safety and immunogenicity of a prototype inactivated, split-virus H5N1 (avian influenza A) vaccine. A secondary objective was to assess the cross-reactivity of immune responses to two variant clade 2 H5N1 strains. Methods: In two randomised, dose comparison, parallel assignment, multicentre trials conducted in Australia, healthy adult volunteers received two doses of 7.5 g or 15 g H5 haemagglutinin (HA) vaccine ± AlPO4 adjuvant (phase I trial; N = 400) or two doses of 30 g or 45 g H5 HA with AlPO4 adjuvant (phase II trial; N = 400). Revaccination with a booster dose was offered 6 months after dose 2 (phase I trial only). Main outcome measures were the change in immunogenicity at each follow-up visit from baseline, measured using HA inhibition (HI) and virus microneutralisation (MN) assays, and the frequency and nature of adverse events (AEs). Computer generated tables were used to randomly allocate treatments; participants and investigators were blinded to treatment allocation. Findings: All formulations were well-tolerated; no unexpected serious adverse events were reported. Two doses of 30 g or 45 g H5 HA adjuvanted formulations elicited the highest immune responses, with considerable MN antibody (≥1:20) persistence up to 6 months post-vaccination. The 7.5 and 15 g formulations (±adjuvant) were less immunogenic than the higher dose formulations; HI and MN antibody titres decreased to near pre-vaccination levels at 6 months but were restored to post-dose 2 levels after the booster dose. Immune responses in the phase I trial demonstrated modest levels of cross-protective MN antibodies against two currently circulating, distinct clade 2 H5N1 strains. Interpretation: Two doses of prototype 30 g or 45 g aluminium-adjuvanted, clade 1 H5N1 vaccines were immunogenic and well-tolerated with considerable 6-month antibody persistence. The prototype H5N1 vaccine also elicited modest levels of cross-protective MN antibodies against variant clade 2 H5N1 strains [ClinicalTrials.gov identifiers: NCT00136331, NCT00320346; Funding: CSL Limited, Australia]. © 2008 Elsevier Ltd. All rights reserved.
1. Introduction Pathogenic avian influenza A (H5N1) viruses are leading candidates for the next influenza pandemic. These viruses have spread
∗ Corresponding author at: Melbourne School of Population Health, University of Melbourne, Level 5, 207 Bouverie Street, Carlton, Victoria 3053, Australia. Tel.: +61 3 8344 9350; fax: +61 3 9347 6929. E-mail address: [email protected] (T.M. Nolan). 0264-410X/$ – see front matter © 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.vaccine.2008.05.077
to an unprecedented number of countries causing severe disease in poultry and associated zoonotic infections in humans with a high fatality rate [1]. Although H5N1 viruses have not yet demonstrated sustained person-to-person transmission, there is a concerted global effort to develop prototype pandemic vaccines based on currently circulating H5N1 strains. To provide a rapid response to an emergent H5 pandemic and to facilitate the speed of clinical development and regulatory approval, it is desirable that prototype pandemic vaccine development be based closely on current licensed manufacturing
T.M. Nolan et al. / Vaccine 26 (2008) 4160–4167
4161
processes for influenza vaccines [2] and on the use of adjuvants with an extensive safety record. Several prototype vaccines based on clade 1 H5N1 virus strains (A/Vietnam/1194/2004 and A/Vietnam/1203/2004) have been evaluated in adults [3–6]. Findings from these studies confirm that at least two doses of vaccine are required to elicit an immune response in na¨ıve populations. Given the diversity of circulating H5 strains and the rapid evolution of influenza A viruses, prototype pandemic vaccines that confer protection against a number of circulating H5N1 viruses will maximise the protection achieved in a vaccinated population. Although existing prototype vaccines are based on the clade 1 strains [3–6] genetically distinct clade 2 strains are responsible for the most recent human avian influenza infections [7]. The objective of this study was to evaluate prototype inactivated split-virus A/Vietnam/1194/2004 vaccine formulations in healthy adult populations. In addition, the ability of a 6-month booster vaccine to enhance immunogenicity and the cross-reactivity of immune responses to two diverse clade 2 H5N1 viruses was assessed.
June to September, 2006 at: (i) MCRI, (ii) CMAX and (iii) the Princess Margaret Hospital for Children (Western Australia). A randomised, dose comparison, parallel assignment study design was used with equal randomisation without stratification for both trials (Fig. 1). The randomisation code was prepared before each trial commenced using a SAS database (Version 9.1.3, SAS Institute, Inc., Cary, NC). Participants and investigators were blinded to treatment allocation in both trials. In the phase I trial there was a visible difference between vaccines with and without adjuvant. Therefore, personnel who prepared and administered the study vaccine had no further involvement in the study. All investigators obtained appropriate institutional ethics committee approval before the trials commenced. Both trials were conducted in accordance with the ethical principles of the Declaration of Helsinki and the Australian regulatory requirements for Good Clinical Practice.
2. Materials and methods
All participants gave voluntary, written informed consent before enrolment. The main inclusion criteria were: men or women aged 18–45 years (phase I trial) or 18–64 years (phase II trial), and for women of childbearing potential, a negative urine pregnancy test and willingness to use adequate methods of contraception. The main exclusion criteria were: hypersensitivity to vaccine components (including eggs, chicken protein, neomycin, polymixin, aluminium or thiomersal); immunosuppression; administration of immunoglobulins and/or any blood products (within 90 days); participation in a clinical study or use of an investigational compound (within 90 days); recent vaccination with a registered vaccine (within 30 days); known history of Guillain–Barre´ Syndrome; active neurological disease or current treatment with an anticoagulant. Participants in the phase I trial were ineligible for participation in the phase II trial.
2.1. Study design We report results from two prospective, observational, multicentre trials that were conducted within Australia to assess the safety, tolerability and immunogenicity of: (i) 7.5 and 15 g H5 haemagglutinin (HA) with or without AlPO4 adjuvant (phase I trial, ClinicalTrials.gov identifier: NCT00136331); and (ii) 30 g or 45 g H5 HA with AlPO4 adjuvant (phase II trial, ClinicalTrials.gov identifier: NCT00320346). The phase I trial was conducted from October 2005 to July 2006 at: (i) the Murdoch Childrens Research Institute, and the School of Population Health, at the University of Melbourne (MCRI; Victoria); and (ii) CMAX, a division of the Institute of Drug Technology (South Australia). The phase II trial was conducted from
2.2. Participants
Fig. 1. Flow of participants. Participants were randomised to receive 7.5 g or 15 g H5 haemagglutinin (HA) antigen, with or without AlPO4 adjuvant (phase I trial, NCT00136331) or 30 g or 45 g H5 haemagglutinin antigen, with AlPO4 adjuvant (phase II trial, NCT00320346). Month 6 means 6 months after receipt of the second dose of vaccine. *One participant was discontinued by the investigator after the first dose, but attended the Day 42 follow-up visit.
4162
T.M. Nolan et al. / Vaccine 26 (2008) 4160–4167
2.3. Vaccines and viruses The vaccine formulations were based on a monovalent, inactivated, split-virus H5N1 vaccine prepared by CSL Limited (Parkville, VIC, Australia). The seed virus (NIBRG-14), a reverse geneticsderived 2:6 reassortment between A/Vietnam/1194/2004 (H5N1) and a laboratory strain (A/PR/8/34, H1N1) [8] was supplied by the National Institute of Biological Standards and Controls (NIBSC, Potters Bar, UK). NIBRG-14 contains a modified HA and neuraminidase (NA) from A/Vietnam/1194/2004 and other proteins from A/PR/8/34, and is avirulent and antigenically indistinguishable to A/Vietnam/1194/2004. Study vaccine was prepared in embryonated hens’ eggs using standard techniques. The target antigen dose (0.5 mL) was: 7.5 g, 15 g, 30 g or 45 g H5 HA with or without 0.5 mg aluminium ion (AlPO4 adjuvant) and 50 g thiomersal. 2.4. Study procedures Participants received two single doses of study vaccine by intramuscular or deep subcutaneous injection into the deltoid muscle, 21 days apart (Fig. 1). At Month 6 (that is, 6 months after receiving the second dose of vaccine), participants who received the first two doses of vaccine in the phase I trial were offered a booster dose of the same formulation they received previously. Blood samples (10 mL) were collected immediately before each vaccine dose, on Day 0, Day 21, Day 42, Month 6 and, for the phase I trial, 21 days after the 6-month booster. Participants were observed for 30 min post-vaccination and were given diary cards to record solicited local and systemic adverse events (AEs) on the evening of vaccination and 6 days thereafter; 21-day diary cards were issued for any unsolicited AEs. Local reaction measurement cards were used to measure redness, swelling and ecchymosis at the injection site and a digital thermometer was provided for measuring oral temperature. 2.5. Adverse events
and the final titre was the geometric mean (GMT) of individual titres. For cross-reactivity analysis, all post-primary vaccination samples from the phase I trial were analysed for HI antibodies to the vaccine virus, NIBRG-14 (clade 1), and two variant strains: NIBRG-23 (clade 2.2) and INDO5/RG2 (clade 2.1). All post-primary vaccination samples were analysed for MN antibodies to NIBRG-14 and NIBRG-23, and a smaller subset was analysed for MN antibodies to IND05/RG2. This smaller subset contained 45 samples with the highest antibody titres to NIBRG-14 (≥1:100, ≤1:2560; GMT = 181) and 29 intermediate samples (≥1:40, <1:100, GMT = 64). 2.7. Data analyses Safety and tolerability analyses included all randomised participants and comprised descriptive statistics. The primary measures of safety and tolerability were the frequency of local and systemic solicited AEs for 6 days after each vaccination and the frequency of unsolicited AEs, for 20 days after each dose. Data from the evaluable population (participants who received at least one dose of study vaccine and whose HI or MN results at Day 0 and the appropriate post-vaccination time point could be analysed) were included in immunogenicity analyses (Fig. 1). The primary measures of immunogenicity were the proportion of participants who were seroresponsive using the HI assay or MN assay and the geometric fold-increase in antibody titres at each follow-up visit, compared to the pre-vaccination baseline. The Committee for Medicinal Products (CPMP) seroprotection immunogenicity criterion for seasonal influenza vaccines is >70% of participants 18 to 60 years of age with HI titres ≥1:40 [12]. However, as the relationship between antibody titres measured using the horse HI assay and MN is not well understood, a titre target of ≥1:32, which is at least four times the limit of quantitation (LOQ) of the horse HI assay (8 units), was used to measure seroresponsiveness. For the MN assay, seroresponsiveness (titre target ≥1:20) was set to be greater than the LOQ of the MN assay (20 units). 2.8. Role of the funding source
Vaccination site AEs (solicited local AEs) were graded from 0 to 3, where 0 represented “absent/none” and 3 represented “prevents normal activity” for pain at the injection site or represented greater than 50 mm diameter for redness, swelling, induration and ecchymosis. Systemic solicited AEs were graded from 0 to 4, where 0 was “absent/none” and 4 was “disabling” or had “life threatening consequences”. The intensity of unsolicited AEs was graded as “mild”, “moderate” or “severe” by participants. Investigators assigned AEs as “related” or “not related” to the study vaccine. Serious adverse events (SAEs) were to be reported within 12 months of the last dose of study vaccine in the phase I trial and within 6 months of the last dose in the phase II trial.
The study sponsor, CSL Limited (Parkville, VIC, Australia), was involved in the study design, laboratory assays, data collection, management and analyses, and writing of the manuscript. In compliance with the Uniform Requirements for Manuscripts, established by the International Committee of Medical Journal Editors, CSL Ltd. did not impose any impediment, directly or indirectly, on the publication of the studies’ results. All authors had full access to all study data. The corresponding author had final responsibility for the decision to submit this manuscript for publication.
2.6. Laboratory assays
3. Results
The following reverse genetic vaccine candidate reference H5N1 strains were used in laboratory assays: A/Vietnam/1194/ 2004/NIBRG14, A/turkey/Turkey/1/2005/NIBRG23 (NIBSC, Potters Bar, UK) and A/Indonesia/5/2005/PR8-IBCDC-RG2 (Centers for Disease Control and Prevention, Atlanta, USA). Haemagglutinin inhibition (HI) with horse erythrocytes [9,10] and virus microneutralisation (MN) [11] assays were conducted at the Health Protection Agency (Colindale, UK) using previously published methods. HI antibody titres that were undetectable at a dilution of 1:8 were assigned a value of 4. MN antibody titres that were undetectable at a dilution of 1:20 were assigned a value of 10. For both assays, samples were titrated twice in separate assays
3.1. Participant distribution and demographic data A total of 400 participants were enrolled in each trial and randomised to each group within each trial (Fig. 1, Table 1). All enrolled participants were included in the safety populations. Over 98% (787/800) of participants completed the primary vaccination course, including the Day 42 follow-up visit. Of the 13 participants (from both trials) who did not complete the primary vaccination course, only one withdrew because of an AE (mild muscle ache and fatigue). Of the 271 participants who received the booster dose in the phase I trial, 99% (n = 268) completed the post-booster visit and none withdrew because of an AE.
T.M. Nolan et al. / Vaccine 26 (2008) 4160–4167
4163
Table 1 Demographic characteristics of participants randomised to each vaccine formulation Characteristic
Vaccine formulation (haemagglutinin content ± adjuvant) Phase I trial
Phase II trial
7.5 g
7.5 g + Al PO4
15 g
15 g + Al PO4
30 g + Al PO4
45 g + Al PO4
No. of participants
100
100
100
100
201
199
Age (years) Mean (S.D.) Min, Max Median Men n (%) Women n (%)
32.0 (8.9) 18, 45 33 46 (46) 54 (54)
33.4 (8.7) 18, 46a 35 43 (43) 57 (57)
33.4 (8.5) 18, 45 36 40 (40) 60 (60)
33.9 (8.2) 18, 45 37 47 (47) 53 (53)
41.3 (12.2) 18, 64 42 98 (49) 103 (51)
42.0 (11.8) 18, 63 43 99 (50) 100 (50)
a
One participant did not meet the inclusion criteria for age by the time of vaccination; a protocol waiver was granted and the participant was included in the analyses.
3.2. Safety and tolerability All vaccine formulations were well-tolerated; no unexpected SAEs were reported. Most solicited vaccination site and systemic AEs were graded as mild in the phase I (80%, 1959/2435) and phase II (78%, 1585/2029) trials. There was a slight increase in the frequency of events with higher antigen content in the phase I trial, which was not evident in the phase II trial (Table 2). The frequency of events did not increase with successive doses of vaccine (data not shown). For both trials, the most frequently reported injection site reactions were pain and redness, and the most frequently reported systemic events were headache and fatigue (Table 2). Participants who received unadjuvanted formulations had fewer injection site reactions, but a similar frequency of systemic AEs, than those who received adjuvanted formulations in either trial (Table 2). Generally, the solicited injection site reactions and systemic symptoms after the booster dose for the phase I trial were comparable with those observed for the primary vaccination course (data not shown). Of all unsolicited AEs reported, 25.9% (301/1160) were considered to be related to the study vaccine; 54.3% (631/1160) were rated as mild and 8.7% (101/1160) were rated as severe. According to the regulatory definition of a SAE, there were 23 SAEs reported in the phase I trial and 13 SAEs in the phase II trial as of August 2007 (data not shown). Only one SAE was considered
possibly related to the vaccine by the investigator. This participant, who had a history of infertility and was using adequate contraception, had a miscarriage, 11 weeks after her second dose of vaccine in the phase I trial. 3.3. Immunogenicity One subject in the phase II trial had HI and MN antibody titres ≥1:40 before vaccination but had no clinical history, suggesting the potential for previous exposure to H5N1 viruses. 3.3.1. Primary vaccination course (first two doses of study vaccine) After completion of the primary vaccination course, all vaccine formulations (phase I and II trials) met the CPMP criterion for a greater than 2.5-fold increase in HI antibody titres after dose 2, but not the criterion for greater than 70% of participants achieving seroprotection (analysed as seroresponsive at HI titres ≥1:32) (Table 3). The proportion of participants with HI titres ≥1:32 increased from between 14% and 31% after the first dose to between 37% and 59% after the second dose (Table 3). The rates of seroconversion were not assessed as almost all participants had undetectable immune responses pre-vaccination (data not shown). Greater than 2.5-fold increases in MN titres were observed after the second dose with all formulations, except for the 7.5 g unadjuvanted formulation (Table 4). The proportion of participants with
Table 2 Proportion (%, 95% confidence interval) of participants reporting solicited adverse events within 7 days post-vaccination (dose 1 and 2 combined)a Adverse Event
Vaccine formulation (haemagglutinin content ± adjuvant) Phase I trial
Phase II trial
7.5 g n = 100
7.5 g + Al PO4 n = 100
15 g n = 100
15 g + Al PO4 n = 100
30 g + Al PO4 n = 201
45 g + Al PO4 n = 199
Local reactions Pain Redness Swelling Induration Ecchymosis
53.0 (43.3–62.5) 24.0 (16.7–33.2) 11.0 (6.3–18.6) NC NC
79.0 (70.0–85.8) 31.0 (22.8–40.6) 18.0 (11.7–26.7) NC NC
56.0 (46.2–65.3) 30.0 (21.9–39.6) 14.0 (8.5–22.1) NC NC
83.0 (74.5–89.1) 40.0 (30.9–49.8) 21.0 (14.2–30.0) NC NC
79.6 (73.5–84.6) 35.3 (29.0–42.1) 24.9 (19.4–31.3) 24.4 (19.0–30.8) 15.9 (11.5–21.6)
76.9 (70.6–82.2) 41.7 (35.1–48.7) 15.1 (10.8–20.7) 17.6 (12.9–23.5) 11.1 (7.4–16.2)
Systemic reactions Headache Fatigue Myalgiab Myalgia—proximal Myalgia—distal Fever Chills Sweating Nausea Vomiting
47.0 (37.5–56.7) 40.0 (30.9–49.8) NC 33.0 (24.6–42.7) 18.0 (11.7–26.7) 3.0 (1.0–8.5) 10.0 (5.5–17.4) 9.0 (4.8–16.2) 10.0 (5.5–17.4) 3.0 (1.0–8.5)
46.0 (36.6–55.7) 42.0 (32.8–51.8) NC 64.0 (54.2–72.7) 21.0 (14.2–30.0) 0.0 4.0 (1.6–9.8) 7.0 (3.4–13.7) 8.0 (4.1–15.0) 1.0 (0.2–5.4)
45.0 (35.6–54.8) 35.0 (26.4–44.7) NC 23.0 (15.8–32.2) 17.0 (10.9–25.5) 1.0 (0.2–5.4) 6.0 (2.8–12.5) 14.0 (8.5–22.1) 13.0 (7.8–21.0) 1.0 (0.2–5.4)
45.0 (35.6–54.8) 39.0 (30.0–48.8) NC 57.0 (47.2–66.3) 18.0 (11.7–26.7) 2.0 (0.6–7.0) 8.0 (4.1–15.0) 13.0 (7.8–21.0) 10.0 (5.5–17.4) 3.0 (1.0–8.5)
46.3 (39.5–53.2) 37.3 (30.9–44.2) 41.3 (34.7–48.2) NC NC 0.0 7.0 (4.2–11.4) 7.0 (4.2–11.4) 14.9 (10.7–20.5) 2.0 (0.8–5.0)
45.7 (39.0–52.7) 44.2 (37.5–51.2) 47.2 (40.4–54.2) NC NC 0.0 9.0 (5.8–13.8) 6.0 (3.5–10.2) 15.6 (11.2–21.3) 0.5 (0.1–2.8)
NC = not collected. a Adverse events included all adverse events, regardless of relationship to treatment. b Myalgia was defined as proximal or distal in the phase I trial and results are not comparable with the phase II trial.
4164
T.M. Nolan et al. / Vaccine 26 (2008) 4160–4167
Table 3 Haemagglutinin inhibition with horse erythrocytes after the first two doses of study vaccine and 6-month booster Post-dose 1 (Day 21)
CPMPa criteria
Vaccine formulation (haemagglutinin content ± adjuvant) Phase I trial
Phase II trial
7.5 g n = 99
7.5 g + Al PO4 n = 98
15 g n = 100
15 g + Al PO4 n = 100
30 g + Al PO4 n = 200
45 g + Al PO4 n = 199
9.6 (7.1–13.0) 2.4 (1.8–3.2) 21% (14–30)
7.0 (5.5–8.9) 1.7 (1.4–2.2) 14% (9–23)
13.5 (9.5–19.3) 3.3 (2.3–4.7)e 28% (20–37)
7.6 (5.9–9.9) 1.8 (1.4–2.3) 17% (11–26)
12.2 (9.9–15.0) 2.8 (2.3–3.5)e 31% (25–37)
11.0 (9.0–13.5) 2.6 (2.1–3.2)e 30% (24–37)
Post-dose 2 (Day 42)
n = 97
n = 98
n = 100
n = 100
n = 197
n = 195
GMT GMT ratio Seroresponse (titre ≥1:32)
15.9 (11.3–22.4) 3.9 (2.8–5.5)e 37% (28–47)
19.5 (13.9–27.3) 4.9 (3.5–6.8)e 40% (31–50)
18.5 (13.1–26.1) 4.5 (3.2–6.3)e 37% (28–47)
18.8 (13.5–26.3) 4.5 (3.2–6.2)e 42% (33–52)
29.4 (23.7–36.5) 6.8 (5.5–8.4)e 59% (52–66)
31.0 (24.8–38.7) 7.3 (5.8–9.1)e 58% (51–65)
n = 95
n = 95
n = 95
n = 94
5.4 (4.7–6.3) 1.3 (1.2–1.6) 5% (2–12)
5.5 (4.7–6.5) 1.4 (1.2–1.6) 5% (2–12)
6.1 (5.1–7.3) 1.5 (1.3–1.8) 11% (6–18)
5.5 (4.8–6.2) 1.3 (1.1–1.5) 4% (2–10)
NT NT NT
NT NT NT
Day 21 post-booster
n = 60
n = 71
n = 72
n = 65
GMT GMT ratio Seroresponse (titre ≥1:32)
10.0 (7.3–13.7) 2.4 (1.8–3.3) 25% (16–37)
17.1 (12.8–23.0) 4.3 (3.2–5.7)e 42% (31–54)
11.9 (9.1–15.6) 2.9 (2.2–3.7)e 28% (19–39)
13.6 (10.3–17.9) 3.4 (2.6–4.5)e 31% (21–43)
NC NC NC
NC NC NC
GMTb GMT ratioc Seroresponse (titre ≥1:32)d
>2.5 >70%
>2.5 >70%
Pre-booster (Month 6) GMT GMT ratio Seroresponse (titre ≥1:32)
>2.5 >70%
>2.5 >70%
NC = not collected, NT = not tested. a European Union Committee for Proprietary Medicinal Products (CPMP) immunogenicity criteria for licensing of seasonal influenza vaccines [12]. b GMT, geometric mean titre (95% CI); values lower than 8.0 were below the limit of detection and were assigned a value of 4.0. c GMT ratio, ratio (95% CI) of 21 days post-vaccination GMT to Day 0 pre-vaccination GMT. d Seroresponse, % (95% CI) of participants with HI titres ≥1:32. HI titres ≥1:32 represent a 4-fold increase from the limit of quantitation of the HI assay with horse erythrocytes. The CPMP benchmark for seroprotection (≥1:40) is based on a 4-fold increase from the limit of quantitation of the HI assay with turkey erythrocytes. Data for % of participants with a greater than 4-fold increase in GMT are not presented as most of the study population was seronegative before vaccination. e Formulations that meet the CPMP criteria.
Table 4 Microneutralising antibody response after the first two doses of study vaccine and 6-month booster Post-dose 1 (Day 21)
Vaccine formulation (haemagglutinin content ± adjuvant) Phase I trial
Phase II trial
7.5 g
7.5 g + Al PO4
15 g
15 g + Al PO4
30 g + Al PO4
45 g + Al PO4
n = 99
n = 98
n = 100
n = 100
n = 200
n = 199
GMTa GMT ratiob Seroresponsec (titre ≥1:20) Seroresponse (titre ≥1:40)
14.4 (12.0–17.4) 1.4 (1.2–1.7) 16% (10–25) 8% (4–15)
13.1 (11.3–15.1) 1.3 (1.1–1.5) 11% (6–19) 10% (6–18)
18.3 (14.6–22.9) 1.8 (1.5–2.3) 25% (18–34) 17% (11–26)
13.6 (11.8–15.7) 1.4 (1.2–1.6) 13% (8–21) 9% (5–16)
17.8 (15.4–20.4) 1.8 (1.5–2.0) 28% (22–35) 19% (14–24)
17.2 (14.9–19.9) 1.7 (1.5–2.0) 25% (20–32) 19% (14–25)
Post-dose 2 (Day 42)
n = 97
n = 98
n = 100
n = 100
n = 197
n = 195
GMT GMT ratio Seroresponse (titre ≥1:20) Seroresponse (titre ≥1:40)
20.9 (16.9–25.9) 2.0 (1.7–2.5) 37% (28–47) 19% (12–27)
26.8 (22.0–32.8) 2.7 (2.2–3.3) 51% (41–61) 34% (25–43)
26.3 (21.5–32.1) 2.6 (2.1–3.2) 51% (41–61) 30% (22–40)
27.8 (22.9–33.8) 2.8 (2.3–3.4) 54% (44–63) 41% (32–51)
43.3 (37.3–50.3) 4.3 (3.7–5.0) 73% (66–78) 54% (47–61)
45.4 (38.7–53.2) 4.5 (3.8–5.2) 73% (66–79) 55% (48–62)
Pre-booster (Month 6)
n = 95
n = 95
n = 95
n = 94
n = 193
n = 197
GMT GMT ratio Seroresponse (titre ≥1:20) Seroresponse (titre ≥1:40)
12.0 (10.4–13.9) 1.2 (1.0–1.3) 8% (4–16) 3% (1–9)
12.2 (10.6–14.0) 1.2 (1.0–1.4) 8% (4–16) 7% (4–14)
12.1 (10.7–13.7) 1.2 (1.1–1.4) 9% (5–17) 5% (2–12)
11.3 (10.2–12.5) 1.1 (1.0–1.2) 6% (3–13) 4% (2–10)
34.9 (29.3–41.5) 3.4 (2.9–4.1) 60% (53–67) 41% (34–48)
34.9 (29.6–41.1) 3.4 (2.9–4.1) 62% (5568). 42% (35–49)
Day 21 post-booster
n = 60
n = 71
n = 72
n = 65
GMT GMT ratio Seroresponse (titre ≥1:20) Seroresponse (titre ≥1:40)
31.3 (21.2–46.2) 3.0 (2.1–4.4) 45% (33–58) 35% (24–48)
62.2 (44.0–87.8) 6.2 (4.4–8.7) 70% (59–80) 56% (45–67)
39.3 (29.7–52.0) 3.9 (3.0–5.2) 65% (54–75) 46% (35–57)
51.3 (36.9–71.3) 5.1 (3.7–7.1) 71% (59–80) 57% (45–68)
NC NC NC NC
NC NC NC NC
NC = not collected. a GMT, geometric mean titre (95% CI); values lower than 20.0 were below the limit of detection and assigned values of 10.0. b GMT ratio, ratio (95% CI) of 21 days post-vaccination GMT to Day 0 pre-vaccination GMT. c Seroresponse, % (95% CI) of participants with MN titres ≥1:20. The percentage of participants with titres ≥1:40 are presented for comparison to the CPMP benchmark for seroprotection using the HI assay. Data for % of participants with a greater than 4-fold increase in GMT are not presented as most of the study population was seronegative before vaccination.
T.M. Nolan et al. / Vaccine 26 (2008) 4160–4167
4165
Fig. 2. Reverse cumulative distribution of haemagglutinin inhibition titres (top) and virus microneutralisation titres (bottom) at Day 42 (left), Month 6 (middle) and Day 21 post-booster vaccination (right) in the phase I trial.
MN titres ≥1:20 increased from between 11% and 28% after the first dose to between 37% and 73% after the second dose (Table 4). Participants in the phase I trial were slightly more likely to achieve MN titres ≥1:20 with adjuvant than without adjuvant (OR = 1.40, 95% CI = 0.94, 2.09, P = 0.094); a similar and statistically significant effect of adjuvant was evident with MN titres ≥1:40 (OR = 1.87, 95% CI = 1.21, 2.90, P = 0.005). The highest antibody responses were observed after the second dose of the 30 g or 45 g adjuvanted formulations (Fig. 2). Between 58% and 59% of participants had HI titres ≥1:32 and 73% had MN titres ≥1:20 (Tables 3 and 4). With the adjuvanted formulations, participants were significantly more likely to achieve MN titres ≥1:20 with 30 g HA than with 15 g HA (OR = 2.26, 95% CI = 1.37, 3.74, P = 0.002) but there was no difference between the 30 g and 45 g HA formulations (OR = 1.01, 95% CI = 0.65, 1.58, P = 0.959). 3.3.2. Six-month antibody persistence and booster vaccination At Month 6, the immune responses to the phase I trial formulations had declined with less than 12% of participants retaining HI titres ≥1:32 (Table 3) and less than 10% with MN titres ≥1:20 (Table 4). In contrast, immune responses in the phase II trial remained similar to post-dose 2 responses (Table 4). At Month 6, similar percentages (60–62%) of participants had MN titres ≥1:20 compared with those (73%) after the second dose in the primary vaccination course (Table 4). Six-month antibody persistence to phase II trial formulations was not measured using the HI assay. After the booster vaccination (Day 21 post-booster) in the phase I trial, there was a marked response to all formulations. Compared to antibody responses observed after the second dose of the 7.5 g or 15 g formulations, the post-booster HI titres were slightly lower and fewer participants were seroresponsive (Table 3). In contrast, post-booster MN titres were higher than those achieved after the second dose of 7.5 g or 15 g formulations and more participants were seroresponsive (Table 4). Moreover, when the MN assay was used to assess immunogenicity, the immune responses to the booster vaccination with the 7.5 g or 15 g adjuvanted formulations were comparable to the responses achieved after the second
dose of the 30 g or 45 g adjuvanted formulations in the phase II trial. Participants were significantly more likely to achieve MN titres ≥1:20 to the booster vaccination with adjuvanted formulations than with unadjuvanted formulations (OR = 1.95, 95% CI = 1.18–3.27, P = 0.01). Booster vaccination was not conducted in the phase II trial. 3.3.3. Cross-reactivity analysis Antibody responses to the variant clade 2 strains (NIBRG-23 and INDO5/RG2) were lower than the responses to the study vaccine virus, NIBRG-14, and demonstrated the expected antigenic discrimination between the two virus subclades. In general, there were higher levels of cross-reactive MN antibodies to the NIBRG-23 variant strain than with the INDO5/RG2 strain (Fig. 3). After the primary vaccination course (Day 42) in the phase I trial, 48% (189/395) of participants had detectable HI titres (≥1:8) and 48% (190/395) of participants had detectable MN titres (≥1:20) to the study vaccine virus, NIBRG-14. Cross-reactivity to NIBRG-23 was demonstrated in 21% (81/395) of participants using the HI assay (titre ≥1:8) and in 26% (101/395) of participants using the MN assay
Fig. 3. Cross-reactivity of immune responses against the vaccine strain (NIBRG-14), and two clade 2 variant strains (NIBRG-23 and INDO5/RG2). Reverse cumulative distribution of virus microneutralisation titres collected on Day 42 in the phase I trial.
4166
T.M. Nolan et al. / Vaccine 26 (2008) 4160–4167
(titre ≥1:20). Cross-reactivity to INDO5/RG2 was demonstrated in 19% (76/395) of participants using the HI assay and in 26% (19/74) of participants with intermediate to high antibody titres to NIBRG-14, using the MN assay. 4. Discussion This study helps define the potential dosage regimen for a prototype H5N1 pandemic vaccine in healthy adult populations and provides new data demonstrating that inactivated, split-virus, clade 1, H5N1 vaccines, with or without adjuvant, elicit modest levels of cross-reactive MN antibodies to currently circulating variant H5N1 strains. Immune responses in the phase I trial with the 7.5 and 15 g HA formulations (with or without adjuvant) were comparable to those previously observed with similar vaccines [5] and demonstrated modest improvement after two doses with the AlPO4 adjuvant, which led to the decision to test only adjuvanted formulations in the phase II trial. Administration of two doses of either 30 g or 45 g HA adjuvanted vaccine in the phase II trial elicited the highest immune responses, which were considerably maintained for up to 6 months. The immune responses to the 7.5 and 15 g formulations decreased to near pre-vaccination levels 6 months after the primary vaccination course, but were restored to post-dose 2 levels following a booster dose. These findings, based on the combined data from two large, randomised, multicentre, clinical trials with very similar populations and clinical trial conditions, demonstrate that two doses of a 30 g aluminium-adjuvanted split-virus vaccine is suitable for consideration as a prototype H5N1 pandemic vaccine dosage regimen. All vaccine formulations demonstrated acceptable safety profiles, with minimal dose level effects. The local and systemic AEs reported were similar to those observed with other prototype pandemic vaccines [3–6,13,14] and are expected reactions to influenza vaccination. Moreover, as found previously [3], the presence of the aluminium adjuvant did not substantially alter the safety profiles of the vaccine formulations. At the beginning of a pandemic, prototype vaccines that confer the highest protective effect whilst using antigen sparing strategies (i.e., the smallest amount of H5N1 antigen) will allow populations to be vaccinated quickly, before a definitive vaccine is manufactured. Our study suggests that for inactivated, split-virus H5N1 vaccines, a two-dose vaccination course with 30 g HA in an aluminium-adjuvanted formulation may be of greater individual benefit than vaccines with lower amounts of HA. Overall, the responses in our study were comparable to the findings from Bresson et al. [3] where 67% of participants achieved HI titres ≥1:32 with a 30 g HA aluminium-adjuvanted vaccine and lower dose formulations (7.5 g or 15 g HA) were less immunogenic, irrespective of the presence of adjuvant. Further, findings from our study suggest that immune responses to higher amounts of HA (30 g or 45 g) can persist for up to 6 months after a primary vaccination course, whereas responses to lower amounts of HA decline to near pre-vaccination levels after 6 months. Antibody persistence is important for reducing the impact of potential follow-on waves of pandemic activity that may occur. The prototype H5N1 vaccines in our study and those of Bresson et al. [3] were based on current licensed manufacturing processes for influenza vaccines using aluminium adjuvants with an extensive safety record. Similar to Bresson et al. [3] we found that aluminium adjuvant did not greatly improve immune responses at low HA concentrations (7.5 and 15 g). Therefore, further major improvements in immune responses to prototype avian influenza vaccines may require significant advances in adjuvant technology or the application of alternative manufacturing processes. Novel adju-
vants will require substantial use and monitoring before we have the same level of confidence provided by the aluminium adjuvant with respect to safety, given alum’s long and widespread use throughout the world. The ability to confer broad cross-protection through vaccination before a closely matched pandemic strain vaccine can be manufactured is a highly desirable property. Our preliminary findings confirm that two doses of prototype H5N1 vaccine can elicit cross-reactive MN antibodies against antigenically distinct H5N1 strains [6]. These findings are comparable to those of Stephenson et al. [15] who demonstrated cross-reactive antibodies against pathogenic avian influenza (H5N1) in individuals after two doses of non-pathogenic avian influenza (H5N3) vaccine, and higher levels of cross-reactive antibodies after a third booster dose, 16 months later. Vaccination strategies that include a primary vaccination course with a prototype pandemic vaccine, followed by a booster dose against a heterologous pandemic-specific strain, may be a useful option at the beginning of a pandemic, when the demand for vaccine is likely to surpass supply. In the absence of any established criteria for the licensing of pandemic influenza vaccines, we evaluated our vaccine formulations using the CPMP immunogenicity endpoints for licensing of seasonal influenza vaccines [12]. However, these endpoints may not be appropriate surrogates for assessing pandemic influenza vaccines as they were not designed for populations that are immunologically na¨ıve to influenza antigen, and serological correlation of protection from avian influenza has not been defined in humans or experimental systems. Recently, the United States Food and Drug Administration (FDA) released further guidelines for assessing pandemic influenza vaccines in immunologically na¨ıve populations [16]. However, both the CPMP and FDA criteria are based on the HI antibody assay for assessing protection from seasonal influenza virus infection. A serum HI titre of ≥1:40 is considered an approximate correlate of protection from seasonal influenza in humans [17], however, there are no clinical data to confirm that this seasonal HI target reflects protection from avian influenza. Moreover, as the HI assay may underestimate human immune responses to avian influenza viruses [9,11,13], other serological assays, such as the modified HI assay with horse erythrocytes and the MN assay, as used in this study, are increasingly used to evaluate immune responses to prototype H5N1 vaccines [3,5,10] However, as yet, there is no information describing the correlation between these immunogenicity end points and clinical protection. In conclusion, two doses of prototype 30 g or 45 g H5 HA, aluminium-adjuvanted vaccines were immunogenic and welltolerated, with antibody persistence for 6 months. Moreover, prototype H5N1 vaccines based on clade 1 strains of avian influenza elicited modest levels of cross-reactive MN antibodies against currently circulating, distinct clade 2, H5N1 strains. Acknowledgments The data reported here were presented, in part, at the VIII International Symposium on Respiratory Viral Infections, Hawaii, March 16–19, 2006; the 3rd WHO Meeting on Evaluation of Pandemic Influenza Vaccines in Clinical Trials, WHO/HQ, Geneva, February 15–16, 2007 and the Options for the Control of Influenza VI Conference, Toronto Canada, June 17–23, 2007. The authors would like to extend their thanks to the participants, investigators and personnel at each study site. Specifically, the authors would like to thank staff from the Vaccine and Immunisation Research Group (Murdoch Childrens Research Institute, University of Melbourne) including L. Thorn, K. Alexander, J. McVernon, M. Kefford, K. O’Grady, J. Ryrie and J. Sonego, staff from the Vaccine Trials Group (Institute for Child Health Research, Uni-
T.M. Nolan et al. / Vaccine 26 (2008) 4160–4167
versity of Western Australia) including S. Nadall, S. Bilic, J. Adams and K. Prosser and D. Lakhman, M. Rico Garcia and K. McCurrie from the Virus Reference Laboratory, The Health Protection Agency; Colindale, UK for their assistance in this study. This study was sponsored by CSL Limited (Parkville, VIC, Australia). The authors acknowledge the independent medical writing assistance provided by ProScribe Medical Communications (www.proscribe.com.au), funded from an unrestricted financial grant from CSL Limited. ProScribe’s services complied with international guidelines for Good Publication Practice. Contributors: All authors participated in the interpretation of results, writing of the manuscript and approval of the final version. T.M. Nolan had full access to all the data in the study and had final responsibility for the decision to submit for publication. T.M. Nolan, P.C. Richmond, M.C. Zambon, R.L. Basser, M.V. Skeljo, G. Pearce and J. Bennet designed the study with T.M. Nolan and P. Richmond as principal investigators. The study was led by T.M. Nolan and coordinated by T.M. Nolan, P.C. Richmond, K. Papanaoum, M.V. Skeljo, N.T. Formica and G. Pearce. D. Ryan was responsible for vaccine ¨ production. Immunological assays were conducted by K. Hoschler and M.C. Zambon and statistical and data analyses were conducted by T.M. Nolan, P.C. Richmond, M.C. Zambon, N. Formica and G. Hartel. Conflict of interest: M.V. Skeljo, G. Pearce, G. Hartel, N.T. Formica, J. Bennet, D. Ryan and R.L. Basser are employees of CSL Limited; Parkville, Victoria, Australia. T.M. Nolan and P.C. Richmond have received honorarium payments and travel support from CSL Ltd. for scientific advisory meetings on unrelated vaccine research. K. ¨ Hoschler and M.C. Zambon are employees of the Health Protection Agency, Colindale, UK, which received funding from CSL Ltd. to conduct the laboratory assays. All other authors have no conflict of interest. References [1] World Health Organisation. Cumulative number of confirmed human cases of avian influenza A (H5N1) reported to WHO; February 12, 2008. http://www. who.int/csr/disease/avian influenza/country/en/(accessed February 14, 2008). [2] Committee for Proprietary Medicinal Products (CPMP). Guideline on dossier structure and content for pandemic influenza vaccine marketing authorisation application (CPMP/VEG/4717/03). London: European Agency for the Evaluation of Medicinal Products (EMEA); April 5, 2004.
4167
[3] Bresson JL, Perronne C, Launay O, Gerdil C, Saville M, Wood J, et al. Safety and immunogenicity of an inactivated split-virion influenza A/Vietnam/1194/2004 (H5N1) vaccine: phase I randomised trial. Lancet 2006;367:1657–64. [4] Lin J, Zhang J, Dong X, Fang H, Chen J, Su N, et al. Safety and immunogenicity of an inactivated adjuvanted whole-virion influenza A (H5N1) vaccine: a phase I randomised controlled trial. Lancet 2006;368:991–7. [5] Treanor JJ, Campbell JD, Zangwill KM, Rowe T, Wolff M. Safety and immunogenicity of an inactivated subvirion influenza A (H5N1) vaccine. N Engl J Med 2006;354:1343–51. [6] Leroux-Roels I, Borkowski A, Vanwolleghem T, Drame M, Clement F, Hons E, et al. Antigen sparing and cross-reactive immunity with an adjuvanted rH5N1 prototype pandemic influenza vaccine: a randomised controlled trial. Lancet 2007;370:580–9. [7] World Health Organisation. Antigenic and genetic characteristics of H5N1 viruses and candidate H5N1 vaccine viruses developed for potential use as pre-pandemic vaccines; March, 2007. http://www.who.int/csr/disease/ avian influenza/guidelines/summaryH520070403.pdf (accessed October 18, 2007). [8] Nicolson C, Major D, Wood JM, Robertson JS. Generation of influenza vaccine viruses on Vero cells by reverse genetics: an H5N1 candidate vaccine strain produced under a quality system. Vaccine 2005;23:2943–52. [9] Stephenson I, Wood JM, Nicholson KG, Zambon MC. Sialic acid receptor specificity on erythrocytes affects detection of antibody to avian influenza haemagglutinin. J Med Virol 2003;70:391–8. [10] Stephenson I, Wood JM, Nicholson KG, Charlett A, Zambon MC. Detection of anti-H5 responses in human sera by HI using horse erythrocytes following MF59-adjuvanted influenza A/Duck/Singapore/97 vaccine. Virus Res 2004;103: 91–5. [11] Rowe T, Abernathy RA, Hu-Primmer J, Thompson WW, Lu X, Lim W, et al. Detection of antibody to avian influenza A (H5N1) virus in human serum by using a combination of serologic assays. J Clin Microbiol 1999;37:937–43. [12] Committee for Proprietary Medicinal Products (CPMP). Note for guidance on harmonisation of requirements for influenza vaccines (CPMP/BWP/214/96). London: European Agency for the Evaluation of Medicinal Products (EMEA); March 12, 1997. [13] Nicholson KG, Colegate AE, Podda A, Stephenson I, Wood J, Ypma E, et al. Safety and antigenicity of non-adjuvanted and MF59-adjuvanted influenza A/Duck/Singapore/97 (H5N3) vaccine: a randomised trial of two potential vaccines against H5N1 influenza. Lancet 2001;357:1937–43. [14] Treanor JJ, Wilkinson BE, Masseoud F, Hu-Primmer J, Battaglia R, O’Brien D, et al. Safety and immunogenicity of a recombinant hemagglutinin vaccine for H5 influenza in humans. Vaccine 2001;19:1732–7. [15] Stephenson I, Bugarini R, Nicholson KG, Podda A, Wood JM, Zambon MC, et al. Cross-reactivity to highly pathogenic avian influenza H5N1 viruses after vaccination with nonadjuvanted and MF59-adjuvanted influenza A/Duck/Singapore/97 (H5N3) vaccine: a potential priming strategy. J Infect Dis 2005;191:1210–5. [16] Food and Drug Administration (FDA). Guidance for Industry. Clinical data needed to support the licensure of pandemic influenza vaccines. Rockville: US Department of Health and Human Services; May, 2007. [17] Potter CW, Oxford JS. Determinants of immunity to influenza infection in man. Br Med Bull 1979;35:69–75.
Contents lists available at ScienceDirect
Vaccine journal homepage: www.elsevier.com/locate/vaccine
Phase I and II randomised trials of the safety and immunogenicity of a prototype adjuvanted inactivated split-virus influenza A (H5N1) vaccine in healthy adults Terry M. Nolan a,∗ , Peter C. Richmond b , Maryanne V. Skeljo c , Georgina Pearce c , d ¨ Gunter Hartel c , Neil T. Formica c , Katja Hoschler , Jillian Bennet c , David Ryan c , e c Kelly Papanaoum , Russell L. Basser , Maria C. Zambon d a
Murdoch Childrens Research Institute, and the Melbourne School of Population Health, University of Melbourne, Carlton, Victoria, Australia School of Paediatrics and Child Health, The University of Western Australia, Perth, Western Australia, Australia c CSL Limited, Parkville, Victoria, Australia d Virus Reference Laboratory, The Health Protection Agency, Colindale, UK e Infectious Diseases Unit, Royal Adelaide Hospital, Adelaide, South Australia, Australia b
a r t i c l e
i n f o
Article history: Received 29 April 2008 Received in revised form 15 May 2008 Accepted 25 May 2008 Available online 13 June 2008 Keywords: Avian Influenza Vaccine Pandemic Prototype
a b s t r a c t Objective: The primary objective was to evaluate the safety and immunogenicity of a prototype inactivated, split-virus H5N1 (avian influenza A) vaccine. A secondary objective was to assess the cross-reactivity of immune responses to two variant clade 2 H5N1 strains. Methods: In two randomised, dose comparison, parallel assignment, multicentre trials conducted in Australia, healthy adult volunteers received two doses of 7.5 g or 15 g H5 haemagglutinin (HA) vaccine ± AlPO4 adjuvant (phase I trial; N = 400) or two doses of 30 g or 45 g H5 HA with AlPO4 adjuvant (phase II trial; N = 400). Revaccination with a booster dose was offered 6 months after dose 2 (phase I trial only). Main outcome measures were the change in immunogenicity at each follow-up visit from baseline, measured using HA inhibition (HI) and virus microneutralisation (MN) assays, and the frequency and nature of adverse events (AEs). Computer generated tables were used to randomly allocate treatments; participants and investigators were blinded to treatment allocation. Findings: All formulations were well-tolerated; no unexpected serious adverse events were reported. Two doses of 30 g or 45 g H5 HA adjuvanted formulations elicited the highest immune responses, with considerable MN antibody (≥1:20) persistence up to 6 months post-vaccination. The 7.5 and 15 g formulations (±adjuvant) were less immunogenic than the higher dose formulations; HI and MN antibody titres decreased to near pre-vaccination levels at 6 months but were restored to post-dose 2 levels after the booster dose. Immune responses in the phase I trial demonstrated modest levels of cross-protective MN antibodies against two currently circulating, distinct clade 2 H5N1 strains. Interpretation: Two doses of prototype 30 g or 45 g aluminium-adjuvanted, clade 1 H5N1 vaccines were immunogenic and well-tolerated with considerable 6-month antibody persistence. The prototype H5N1 vaccine also elicited modest levels of cross-protective MN antibodies against variant clade 2 H5N1 strains [ClinicalTrials.gov identifiers: NCT00136331, NCT00320346; Funding: CSL Limited, Australia]. © 2008 Elsevier Ltd. All rights reserved.
1. Introduction Pathogenic avian influenza A (H5N1) viruses are leading candidates for the next influenza pandemic. These viruses have spread
∗ Corresponding author at: Melbourne School of Population Health, University of Melbourne, Level 5, 207 Bouverie Street, Carlton, Victoria 3053, Australia. Tel.: +61 3 8344 9350; fax: +61 3 9347 6929. E-mail address: [email protected] (T.M. Nolan). 0264-410X/$ – see front matter © 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.vaccine.2008.05.077
to an unprecedented number of countries causing severe disease in poultry and associated zoonotic infections in humans with a high fatality rate [1]. Although H5N1 viruses have not yet demonstrated sustained person-to-person transmission, there is a concerted global effort to develop prototype pandemic vaccines based on currently circulating H5N1 strains. To provide a rapid response to an emergent H5 pandemic and to facilitate the speed of clinical development and regulatory approval, it is desirable that prototype pandemic vaccine development be based closely on current licensed manufacturing
T.M. Nolan et al. / Vaccine 26 (2008) 4160–4167
4161
processes for influenza vaccines [2] and on the use of adjuvants with an extensive safety record. Several prototype vaccines based on clade 1 H5N1 virus strains (A/Vietnam/1194/2004 and A/Vietnam/1203/2004) have been evaluated in adults [3–6]. Findings from these studies confirm that at least two doses of vaccine are required to elicit an immune response in na¨ıve populations. Given the diversity of circulating H5 strains and the rapid evolution of influenza A viruses, prototype pandemic vaccines that confer protection against a number of circulating H5N1 viruses will maximise the protection achieved in a vaccinated population. Although existing prototype vaccines are based on the clade 1 strains [3–6] genetically distinct clade 2 strains are responsible for the most recent human avian influenza infections [7]. The objective of this study was to evaluate prototype inactivated split-virus A/Vietnam/1194/2004 vaccine formulations in healthy adult populations. In addition, the ability of a 6-month booster vaccine to enhance immunogenicity and the cross-reactivity of immune responses to two diverse clade 2 H5N1 viruses was assessed.
June to September, 2006 at: (i) MCRI, (ii) CMAX and (iii) the Princess Margaret Hospital for Children (Western Australia). A randomised, dose comparison, parallel assignment study design was used with equal randomisation without stratification for both trials (Fig. 1). The randomisation code was prepared before each trial commenced using a SAS database (Version 9.1.3, SAS Institute, Inc., Cary, NC). Participants and investigators were blinded to treatment allocation in both trials. In the phase I trial there was a visible difference between vaccines with and without adjuvant. Therefore, personnel who prepared and administered the study vaccine had no further involvement in the study. All investigators obtained appropriate institutional ethics committee approval before the trials commenced. Both trials were conducted in accordance with the ethical principles of the Declaration of Helsinki and the Australian regulatory requirements for Good Clinical Practice.
2. Materials and methods
All participants gave voluntary, written informed consent before enrolment. The main inclusion criteria were: men or women aged 18–45 years (phase I trial) or 18–64 years (phase II trial), and for women of childbearing potential, a negative urine pregnancy test and willingness to use adequate methods of contraception. The main exclusion criteria were: hypersensitivity to vaccine components (including eggs, chicken protein, neomycin, polymixin, aluminium or thiomersal); immunosuppression; administration of immunoglobulins and/or any blood products (within 90 days); participation in a clinical study or use of an investigational compound (within 90 days); recent vaccination with a registered vaccine (within 30 days); known history of Guillain–Barre´ Syndrome; active neurological disease or current treatment with an anticoagulant. Participants in the phase I trial were ineligible for participation in the phase II trial.
2.1. Study design We report results from two prospective, observational, multicentre trials that were conducted within Australia to assess the safety, tolerability and immunogenicity of: (i) 7.5 and 15 g H5 haemagglutinin (HA) with or without AlPO4 adjuvant (phase I trial, ClinicalTrials.gov identifier: NCT00136331); and (ii) 30 g or 45 g H5 HA with AlPO4 adjuvant (phase II trial, ClinicalTrials.gov identifier: NCT00320346). The phase I trial was conducted from October 2005 to July 2006 at: (i) the Murdoch Childrens Research Institute, and the School of Population Health, at the University of Melbourne (MCRI; Victoria); and (ii) CMAX, a division of the Institute of Drug Technology (South Australia). The phase II trial was conducted from
2.2. Participants
Fig. 1. Flow of participants. Participants were randomised to receive 7.5 g or 15 g H5 haemagglutinin (HA) antigen, with or without AlPO4 adjuvant (phase I trial, NCT00136331) or 30 g or 45 g H5 haemagglutinin antigen, with AlPO4 adjuvant (phase II trial, NCT00320346). Month 6 means 6 months after receipt of the second dose of vaccine. *One participant was discontinued by the investigator after the first dose, but attended the Day 42 follow-up visit.
4162
T.M. Nolan et al. / Vaccine 26 (2008) 4160–4167
2.3. Vaccines and viruses The vaccine formulations were based on a monovalent, inactivated, split-virus H5N1 vaccine prepared by CSL Limited (Parkville, VIC, Australia). The seed virus (NIBRG-14), a reverse geneticsderived 2:6 reassortment between A/Vietnam/1194/2004 (H5N1) and a laboratory strain (A/PR/8/34, H1N1) [8] was supplied by the National Institute of Biological Standards and Controls (NIBSC, Potters Bar, UK). NIBRG-14 contains a modified HA and neuraminidase (NA) from A/Vietnam/1194/2004 and other proteins from A/PR/8/34, and is avirulent and antigenically indistinguishable to A/Vietnam/1194/2004. Study vaccine was prepared in embryonated hens’ eggs using standard techniques. The target antigen dose (0.5 mL) was: 7.5 g, 15 g, 30 g or 45 g H5 HA with or without 0.5 mg aluminium ion (AlPO4 adjuvant) and 50 g thiomersal. 2.4. Study procedures Participants received two single doses of study vaccine by intramuscular or deep subcutaneous injection into the deltoid muscle, 21 days apart (Fig. 1). At Month 6 (that is, 6 months after receiving the second dose of vaccine), participants who received the first two doses of vaccine in the phase I trial were offered a booster dose of the same formulation they received previously. Blood samples (10 mL) were collected immediately before each vaccine dose, on Day 0, Day 21, Day 42, Month 6 and, for the phase I trial, 21 days after the 6-month booster. Participants were observed for 30 min post-vaccination and were given diary cards to record solicited local and systemic adverse events (AEs) on the evening of vaccination and 6 days thereafter; 21-day diary cards were issued for any unsolicited AEs. Local reaction measurement cards were used to measure redness, swelling and ecchymosis at the injection site and a digital thermometer was provided for measuring oral temperature. 2.5. Adverse events
and the final titre was the geometric mean (GMT) of individual titres. For cross-reactivity analysis, all post-primary vaccination samples from the phase I trial were analysed for HI antibodies to the vaccine virus, NIBRG-14 (clade 1), and two variant strains: NIBRG-23 (clade 2.2) and INDO5/RG2 (clade 2.1). All post-primary vaccination samples were analysed for MN antibodies to NIBRG-14 and NIBRG-23, and a smaller subset was analysed for MN antibodies to IND05/RG2. This smaller subset contained 45 samples with the highest antibody titres to NIBRG-14 (≥1:100, ≤1:2560; GMT = 181) and 29 intermediate samples (≥1:40, <1:100, GMT = 64). 2.7. Data analyses Safety and tolerability analyses included all randomised participants and comprised descriptive statistics. The primary measures of safety and tolerability were the frequency of local and systemic solicited AEs for 6 days after each vaccination and the frequency of unsolicited AEs, for 20 days after each dose. Data from the evaluable population (participants who received at least one dose of study vaccine and whose HI or MN results at Day 0 and the appropriate post-vaccination time point could be analysed) were included in immunogenicity analyses (Fig. 1). The primary measures of immunogenicity were the proportion of participants who were seroresponsive using the HI assay or MN assay and the geometric fold-increase in antibody titres at each follow-up visit, compared to the pre-vaccination baseline. The Committee for Medicinal Products (CPMP) seroprotection immunogenicity criterion for seasonal influenza vaccines is >70% of participants 18 to 60 years of age with HI titres ≥1:40 [12]. However, as the relationship between antibody titres measured using the horse HI assay and MN is not well understood, a titre target of ≥1:32, which is at least four times the limit of quantitation (LOQ) of the horse HI assay (8 units), was used to measure seroresponsiveness. For the MN assay, seroresponsiveness (titre target ≥1:20) was set to be greater than the LOQ of the MN assay (20 units). 2.8. Role of the funding source
Vaccination site AEs (solicited local AEs) were graded from 0 to 3, where 0 represented “absent/none” and 3 represented “prevents normal activity” for pain at the injection site or represented greater than 50 mm diameter for redness, swelling, induration and ecchymosis. Systemic solicited AEs were graded from 0 to 4, where 0 was “absent/none” and 4 was “disabling” or had “life threatening consequences”. The intensity of unsolicited AEs was graded as “mild”, “moderate” or “severe” by participants. Investigators assigned AEs as “related” or “not related” to the study vaccine. Serious adverse events (SAEs) were to be reported within 12 months of the last dose of study vaccine in the phase I trial and within 6 months of the last dose in the phase II trial.
The study sponsor, CSL Limited (Parkville, VIC, Australia), was involved in the study design, laboratory assays, data collection, management and analyses, and writing of the manuscript. In compliance with the Uniform Requirements for Manuscripts, established by the International Committee of Medical Journal Editors, CSL Ltd. did not impose any impediment, directly or indirectly, on the publication of the studies’ results. All authors had full access to all study data. The corresponding author had final responsibility for the decision to submit this manuscript for publication.
2.6. Laboratory assays
3. Results
The following reverse genetic vaccine candidate reference H5N1 strains were used in laboratory assays: A/Vietnam/1194/ 2004/NIBRG14, A/turkey/Turkey/1/2005/NIBRG23 (NIBSC, Potters Bar, UK) and A/Indonesia/5/2005/PR8-IBCDC-RG2 (Centers for Disease Control and Prevention, Atlanta, USA). Haemagglutinin inhibition (HI) with horse erythrocytes [9,10] and virus microneutralisation (MN) [11] assays were conducted at the Health Protection Agency (Colindale, UK) using previously published methods. HI antibody titres that were undetectable at a dilution of 1:8 were assigned a value of 4. MN antibody titres that were undetectable at a dilution of 1:20 were assigned a value of 10. For both assays, samples were titrated twice in separate assays
3.1. Participant distribution and demographic data A total of 400 participants were enrolled in each trial and randomised to each group within each trial (Fig. 1, Table 1). All enrolled participants were included in the safety populations. Over 98% (787/800) of participants completed the primary vaccination course, including the Day 42 follow-up visit. Of the 13 participants (from both trials) who did not complete the primary vaccination course, only one withdrew because of an AE (mild muscle ache and fatigue). Of the 271 participants who received the booster dose in the phase I trial, 99% (n = 268) completed the post-booster visit and none withdrew because of an AE.
T.M. Nolan et al. / Vaccine 26 (2008) 4160–4167
4163
Table 1 Demographic characteristics of participants randomised to each vaccine formulation Characteristic
Vaccine formulation (haemagglutinin content ± adjuvant) Phase I trial
Phase II trial
7.5 g
7.5 g + Al PO4
15 g
15 g + Al PO4
30 g + Al PO4
45 g + Al PO4
No. of participants
100
100
100
100
201
199
Age (years) Mean (S.D.) Min, Max Median Men n (%) Women n (%)
32.0 (8.9) 18, 45 33 46 (46) 54 (54)
33.4 (8.7) 18, 46a 35 43 (43) 57 (57)
33.4 (8.5) 18, 45 36 40 (40) 60 (60)
33.9 (8.2) 18, 45 37 47 (47) 53 (53)
41.3 (12.2) 18, 64 42 98 (49) 103 (51)
42.0 (11.8) 18, 63 43 99 (50) 100 (50)
a
One participant did not meet the inclusion criteria for age by the time of vaccination; a protocol waiver was granted and the participant was included in the analyses.
3.2. Safety and tolerability All vaccine formulations were well-tolerated; no unexpected SAEs were reported. Most solicited vaccination site and systemic AEs were graded as mild in the phase I (80%, 1959/2435) and phase II (78%, 1585/2029) trials. There was a slight increase in the frequency of events with higher antigen content in the phase I trial, which was not evident in the phase II trial (Table 2). The frequency of events did not increase with successive doses of vaccine (data not shown). For both trials, the most frequently reported injection site reactions were pain and redness, and the most frequently reported systemic events were headache and fatigue (Table 2). Participants who received unadjuvanted formulations had fewer injection site reactions, but a similar frequency of systemic AEs, than those who received adjuvanted formulations in either trial (Table 2). Generally, the solicited injection site reactions and systemic symptoms after the booster dose for the phase I trial were comparable with those observed for the primary vaccination course (data not shown). Of all unsolicited AEs reported, 25.9% (301/1160) were considered to be related to the study vaccine; 54.3% (631/1160) were rated as mild and 8.7% (101/1160) were rated as severe. According to the regulatory definition of a SAE, there were 23 SAEs reported in the phase I trial and 13 SAEs in the phase II trial as of August 2007 (data not shown). Only one SAE was considered
possibly related to the vaccine by the investigator. This participant, who had a history of infertility and was using adequate contraception, had a miscarriage, 11 weeks after her second dose of vaccine in the phase I trial. 3.3. Immunogenicity One subject in the phase II trial had HI and MN antibody titres ≥1:40 before vaccination but had no clinical history, suggesting the potential for previous exposure to H5N1 viruses. 3.3.1. Primary vaccination course (first two doses of study vaccine) After completion of the primary vaccination course, all vaccine formulations (phase I and II trials) met the CPMP criterion for a greater than 2.5-fold increase in HI antibody titres after dose 2, but not the criterion for greater than 70% of participants achieving seroprotection (analysed as seroresponsive at HI titres ≥1:32) (Table 3). The proportion of participants with HI titres ≥1:32 increased from between 14% and 31% after the first dose to between 37% and 59% after the second dose (Table 3). The rates of seroconversion were not assessed as almost all participants had undetectable immune responses pre-vaccination (data not shown). Greater than 2.5-fold increases in MN titres were observed after the second dose with all formulations, except for the 7.5 g unadjuvanted formulation (Table 4). The proportion of participants with
Table 2 Proportion (%, 95% confidence interval) of participants reporting solicited adverse events within 7 days post-vaccination (dose 1 and 2 combined)a Adverse Event
Vaccine formulation (haemagglutinin content ± adjuvant) Phase I trial
Phase II trial
7.5 g n = 100
7.5 g + Al PO4 n = 100
15 g n = 100
15 g + Al PO4 n = 100
30 g + Al PO4 n = 201
45 g + Al PO4 n = 199
Local reactions Pain Redness Swelling Induration Ecchymosis
53.0 (43.3–62.5) 24.0 (16.7–33.2) 11.0 (6.3–18.6) NC NC
79.0 (70.0–85.8) 31.0 (22.8–40.6) 18.0 (11.7–26.7) NC NC
56.0 (46.2–65.3) 30.0 (21.9–39.6) 14.0 (8.5–22.1) NC NC
83.0 (74.5–89.1) 40.0 (30.9–49.8) 21.0 (14.2–30.0) NC NC
79.6 (73.5–84.6) 35.3 (29.0–42.1) 24.9 (19.4–31.3) 24.4 (19.0–30.8) 15.9 (11.5–21.6)
76.9 (70.6–82.2) 41.7 (35.1–48.7) 15.1 (10.8–20.7) 17.6 (12.9–23.5) 11.1 (7.4–16.2)
Systemic reactions Headache Fatigue Myalgiab Myalgia—proximal Myalgia—distal Fever Chills Sweating Nausea Vomiting
47.0 (37.5–56.7) 40.0 (30.9–49.8) NC 33.0 (24.6–42.7) 18.0 (11.7–26.7) 3.0 (1.0–8.5) 10.0 (5.5–17.4) 9.0 (4.8–16.2) 10.0 (5.5–17.4) 3.0 (1.0–8.5)
46.0 (36.6–55.7) 42.0 (32.8–51.8) NC 64.0 (54.2–72.7) 21.0 (14.2–30.0) 0.0 4.0 (1.6–9.8) 7.0 (3.4–13.7) 8.0 (4.1–15.0) 1.0 (0.2–5.4)
45.0 (35.6–54.8) 35.0 (26.4–44.7) NC 23.0 (15.8–32.2) 17.0 (10.9–25.5) 1.0 (0.2–5.4) 6.0 (2.8–12.5) 14.0 (8.5–22.1) 13.0 (7.8–21.0) 1.0 (0.2–5.4)
45.0 (35.6–54.8) 39.0 (30.0–48.8) NC 57.0 (47.2–66.3) 18.0 (11.7–26.7) 2.0 (0.6–7.0) 8.0 (4.1–15.0) 13.0 (7.8–21.0) 10.0 (5.5–17.4) 3.0 (1.0–8.5)
46.3 (39.5–53.2) 37.3 (30.9–44.2) 41.3 (34.7–48.2) NC NC 0.0 7.0 (4.2–11.4) 7.0 (4.2–11.4) 14.9 (10.7–20.5) 2.0 (0.8–5.0)
45.7 (39.0–52.7) 44.2 (37.5–51.2) 47.2 (40.4–54.2) NC NC 0.0 9.0 (5.8–13.8) 6.0 (3.5–10.2) 15.6 (11.2–21.3) 0.5 (0.1–2.8)
NC = not collected. a Adverse events included all adverse events, regardless of relationship to treatment. b Myalgia was defined as proximal or distal in the phase I trial and results are not comparable with the phase II trial.
4164
T.M. Nolan et al. / Vaccine 26 (2008) 4160–4167
Table 3 Haemagglutinin inhibition with horse erythrocytes after the first two doses of study vaccine and 6-month booster Post-dose 1 (Day 21)
CPMPa criteria
Vaccine formulation (haemagglutinin content ± adjuvant) Phase I trial
Phase II trial
7.5 g n = 99
7.5 g + Al PO4 n = 98
15 g n = 100
15 g + Al PO4 n = 100
30 g + Al PO4 n = 200
45 g + Al PO4 n = 199
9.6 (7.1–13.0) 2.4 (1.8–3.2) 21% (14–30)
7.0 (5.5–8.9) 1.7 (1.4–2.2) 14% (9–23)
13.5 (9.5–19.3) 3.3 (2.3–4.7)e 28% (20–37)
7.6 (5.9–9.9) 1.8 (1.4–2.3) 17% (11–26)
12.2 (9.9–15.0) 2.8 (2.3–3.5)e 31% (25–37)
11.0 (9.0–13.5) 2.6 (2.1–3.2)e 30% (24–37)
Post-dose 2 (Day 42)
n = 97
n = 98
n = 100
n = 100
n = 197
n = 195
GMT GMT ratio Seroresponse (titre ≥1:32)
15.9 (11.3–22.4) 3.9 (2.8–5.5)e 37% (28–47)
19.5 (13.9–27.3) 4.9 (3.5–6.8)e 40% (31–50)
18.5 (13.1–26.1) 4.5 (3.2–6.3)e 37% (28–47)
18.8 (13.5–26.3) 4.5 (3.2–6.2)e 42% (33–52)
29.4 (23.7–36.5) 6.8 (5.5–8.4)e 59% (52–66)
31.0 (24.8–38.7) 7.3 (5.8–9.1)e 58% (51–65)
n = 95
n = 95
n = 95
n = 94
5.4 (4.7–6.3) 1.3 (1.2–1.6) 5% (2–12)
5.5 (4.7–6.5) 1.4 (1.2–1.6) 5% (2–12)
6.1 (5.1–7.3) 1.5 (1.3–1.8) 11% (6–18)
5.5 (4.8–6.2) 1.3 (1.1–1.5) 4% (2–10)
NT NT NT
NT NT NT
Day 21 post-booster
n = 60
n = 71
n = 72
n = 65
GMT GMT ratio Seroresponse (titre ≥1:32)
10.0 (7.3–13.7) 2.4 (1.8–3.3) 25% (16–37)
17.1 (12.8–23.0) 4.3 (3.2–5.7)e 42% (31–54)
11.9 (9.1–15.6) 2.9 (2.2–3.7)e 28% (19–39)
13.6 (10.3–17.9) 3.4 (2.6–4.5)e 31% (21–43)
NC NC NC
NC NC NC
GMTb GMT ratioc Seroresponse (titre ≥1:32)d
>2.5 >70%
>2.5 >70%
Pre-booster (Month 6) GMT GMT ratio Seroresponse (titre ≥1:32)
>2.5 >70%
>2.5 >70%
NC = not collected, NT = not tested. a European Union Committee for Proprietary Medicinal Products (CPMP) immunogenicity criteria for licensing of seasonal influenza vaccines [12]. b GMT, geometric mean titre (95% CI); values lower than 8.0 were below the limit of detection and were assigned a value of 4.0. c GMT ratio, ratio (95% CI) of 21 days post-vaccination GMT to Day 0 pre-vaccination GMT. d Seroresponse, % (95% CI) of participants with HI titres ≥1:32. HI titres ≥1:32 represent a 4-fold increase from the limit of quantitation of the HI assay with horse erythrocytes. The CPMP benchmark for seroprotection (≥1:40) is based on a 4-fold increase from the limit of quantitation of the HI assay with turkey erythrocytes. Data for % of participants with a greater than 4-fold increase in GMT are not presented as most of the study population was seronegative before vaccination. e Formulations that meet the CPMP criteria.
Table 4 Microneutralising antibody response after the first two doses of study vaccine and 6-month booster Post-dose 1 (Day 21)
Vaccine formulation (haemagglutinin content ± adjuvant) Phase I trial
Phase II trial
7.5 g
7.5 g + Al PO4
15 g
15 g + Al PO4
30 g + Al PO4
45 g + Al PO4
n = 99
n = 98
n = 100
n = 100
n = 200
n = 199
GMTa GMT ratiob Seroresponsec (titre ≥1:20) Seroresponse (titre ≥1:40)
14.4 (12.0–17.4) 1.4 (1.2–1.7) 16% (10–25) 8% (4–15)
13.1 (11.3–15.1) 1.3 (1.1–1.5) 11% (6–19) 10% (6–18)
18.3 (14.6–22.9) 1.8 (1.5–2.3) 25% (18–34) 17% (11–26)
13.6 (11.8–15.7) 1.4 (1.2–1.6) 13% (8–21) 9% (5–16)
17.8 (15.4–20.4) 1.8 (1.5–2.0) 28% (22–35) 19% (14–24)
17.2 (14.9–19.9) 1.7 (1.5–2.0) 25% (20–32) 19% (14–25)
Post-dose 2 (Day 42)
n = 97
n = 98
n = 100
n = 100
n = 197
n = 195
GMT GMT ratio Seroresponse (titre ≥1:20) Seroresponse (titre ≥1:40)
20.9 (16.9–25.9) 2.0 (1.7–2.5) 37% (28–47) 19% (12–27)
26.8 (22.0–32.8) 2.7 (2.2–3.3) 51% (41–61) 34% (25–43)
26.3 (21.5–32.1) 2.6 (2.1–3.2) 51% (41–61) 30% (22–40)
27.8 (22.9–33.8) 2.8 (2.3–3.4) 54% (44–63) 41% (32–51)
43.3 (37.3–50.3) 4.3 (3.7–5.0) 73% (66–78) 54% (47–61)
45.4 (38.7–53.2) 4.5 (3.8–5.2) 73% (66–79) 55% (48–62)
Pre-booster (Month 6)
n = 95
n = 95
n = 95
n = 94
n = 193
n = 197
GMT GMT ratio Seroresponse (titre ≥1:20) Seroresponse (titre ≥1:40)
12.0 (10.4–13.9) 1.2 (1.0–1.3) 8% (4–16) 3% (1–9)
12.2 (10.6–14.0) 1.2 (1.0–1.4) 8% (4–16) 7% (4–14)
12.1 (10.7–13.7) 1.2 (1.1–1.4) 9% (5–17) 5% (2–12)
11.3 (10.2–12.5) 1.1 (1.0–1.2) 6% (3–13) 4% (2–10)
34.9 (29.3–41.5) 3.4 (2.9–4.1) 60% (53–67) 41% (34–48)
34.9 (29.6–41.1) 3.4 (2.9–4.1) 62% (5568). 42% (35–49)
Day 21 post-booster
n = 60
n = 71
n = 72
n = 65
GMT GMT ratio Seroresponse (titre ≥1:20) Seroresponse (titre ≥1:40)
31.3 (21.2–46.2) 3.0 (2.1–4.4) 45% (33–58) 35% (24–48)
62.2 (44.0–87.8) 6.2 (4.4–8.7) 70% (59–80) 56% (45–67)
39.3 (29.7–52.0) 3.9 (3.0–5.2) 65% (54–75) 46% (35–57)
51.3 (36.9–71.3) 5.1 (3.7–7.1) 71% (59–80) 57% (45–68)
NC NC NC NC
NC NC NC NC
NC = not collected. a GMT, geometric mean titre (95% CI); values lower than 20.0 were below the limit of detection and assigned values of 10.0. b GMT ratio, ratio (95% CI) of 21 days post-vaccination GMT to Day 0 pre-vaccination GMT. c Seroresponse, % (95% CI) of participants with MN titres ≥1:20. The percentage of participants with titres ≥1:40 are presented for comparison to the CPMP benchmark for seroprotection using the HI assay. Data for % of participants with a greater than 4-fold increase in GMT are not presented as most of the study population was seronegative before vaccination.
T.M. Nolan et al. / Vaccine 26 (2008) 4160–4167
4165
Fig. 2. Reverse cumulative distribution of haemagglutinin inhibition titres (top) and virus microneutralisation titres (bottom) at Day 42 (left), Month 6 (middle) and Day 21 post-booster vaccination (right) in the phase I trial.
MN titres ≥1:20 increased from between 11% and 28% after the first dose to between 37% and 73% after the second dose (Table 4). Participants in the phase I trial were slightly more likely to achieve MN titres ≥1:20 with adjuvant than without adjuvant (OR = 1.40, 95% CI = 0.94, 2.09, P = 0.094); a similar and statistically significant effect of adjuvant was evident with MN titres ≥1:40 (OR = 1.87, 95% CI = 1.21, 2.90, P = 0.005). The highest antibody responses were observed after the second dose of the 30 g or 45 g adjuvanted formulations (Fig. 2). Between 58% and 59% of participants had HI titres ≥1:32 and 73% had MN titres ≥1:20 (Tables 3 and 4). With the adjuvanted formulations, participants were significantly more likely to achieve MN titres ≥1:20 with 30 g HA than with 15 g HA (OR = 2.26, 95% CI = 1.37, 3.74, P = 0.002) but there was no difference between the 30 g and 45 g HA formulations (OR = 1.01, 95% CI = 0.65, 1.58, P = 0.959). 3.3.2. Six-month antibody persistence and booster vaccination At Month 6, the immune responses to the phase I trial formulations had declined with less than 12% of participants retaining HI titres ≥1:32 (Table 3) and less than 10% with MN titres ≥1:20 (Table 4). In contrast, immune responses in the phase II trial remained similar to post-dose 2 responses (Table 4). At Month 6, similar percentages (60–62%) of participants had MN titres ≥1:20 compared with those (73%) after the second dose in the primary vaccination course (Table 4). Six-month antibody persistence to phase II trial formulations was not measured using the HI assay. After the booster vaccination (Day 21 post-booster) in the phase I trial, there was a marked response to all formulations. Compared to antibody responses observed after the second dose of the 7.5 g or 15 g formulations, the post-booster HI titres were slightly lower and fewer participants were seroresponsive (Table 3). In contrast, post-booster MN titres were higher than those achieved after the second dose of 7.5 g or 15 g formulations and more participants were seroresponsive (Table 4). Moreover, when the MN assay was used to assess immunogenicity, the immune responses to the booster vaccination with the 7.5 g or 15 g adjuvanted formulations were comparable to the responses achieved after the second
dose of the 30 g or 45 g adjuvanted formulations in the phase II trial. Participants were significantly more likely to achieve MN titres ≥1:20 to the booster vaccination with adjuvanted formulations than with unadjuvanted formulations (OR = 1.95, 95% CI = 1.18–3.27, P = 0.01). Booster vaccination was not conducted in the phase II trial. 3.3.3. Cross-reactivity analysis Antibody responses to the variant clade 2 strains (NIBRG-23 and INDO5/RG2) were lower than the responses to the study vaccine virus, NIBRG-14, and demonstrated the expected antigenic discrimination between the two virus subclades. In general, there were higher levels of cross-reactive MN antibodies to the NIBRG-23 variant strain than with the INDO5/RG2 strain (Fig. 3). After the primary vaccination course (Day 42) in the phase I trial, 48% (189/395) of participants had detectable HI titres (≥1:8) and 48% (190/395) of participants had detectable MN titres (≥1:20) to the study vaccine virus, NIBRG-14. Cross-reactivity to NIBRG-23 was demonstrated in 21% (81/395) of participants using the HI assay (titre ≥1:8) and in 26% (101/395) of participants using the MN assay
Fig. 3. Cross-reactivity of immune responses against the vaccine strain (NIBRG-14), and two clade 2 variant strains (NIBRG-23 and INDO5/RG2). Reverse cumulative distribution of virus microneutralisation titres collected on Day 42 in the phase I trial.
4166
T.M. Nolan et al. / Vaccine 26 (2008) 4160–4167
(titre ≥1:20). Cross-reactivity to INDO5/RG2 was demonstrated in 19% (76/395) of participants using the HI assay and in 26% (19/74) of participants with intermediate to high antibody titres to NIBRG-14, using the MN assay. 4. Discussion This study helps define the potential dosage regimen for a prototype H5N1 pandemic vaccine in healthy adult populations and provides new data demonstrating that inactivated, split-virus, clade 1, H5N1 vaccines, with or without adjuvant, elicit modest levels of cross-reactive MN antibodies to currently circulating variant H5N1 strains. Immune responses in the phase I trial with the 7.5 and 15 g HA formulations (with or without adjuvant) were comparable to those previously observed with similar vaccines [5] and demonstrated modest improvement after two doses with the AlPO4 adjuvant, which led to the decision to test only adjuvanted formulations in the phase II trial. Administration of two doses of either 30 g or 45 g HA adjuvanted vaccine in the phase II trial elicited the highest immune responses, which were considerably maintained for up to 6 months. The immune responses to the 7.5 and 15 g formulations decreased to near pre-vaccination levels 6 months after the primary vaccination course, but were restored to post-dose 2 levels following a booster dose. These findings, based on the combined data from two large, randomised, multicentre, clinical trials with very similar populations and clinical trial conditions, demonstrate that two doses of a 30 g aluminium-adjuvanted split-virus vaccine is suitable for consideration as a prototype H5N1 pandemic vaccine dosage regimen. All vaccine formulations demonstrated acceptable safety profiles, with minimal dose level effects. The local and systemic AEs reported were similar to those observed with other prototype pandemic vaccines [3–6,13,14] and are expected reactions to influenza vaccination. Moreover, as found previously [3], the presence of the aluminium adjuvant did not substantially alter the safety profiles of the vaccine formulations. At the beginning of a pandemic, prototype vaccines that confer the highest protective effect whilst using antigen sparing strategies (i.e., the smallest amount of H5N1 antigen) will allow populations to be vaccinated quickly, before a definitive vaccine is manufactured. Our study suggests that for inactivated, split-virus H5N1 vaccines, a two-dose vaccination course with 30 g HA in an aluminium-adjuvanted formulation may be of greater individual benefit than vaccines with lower amounts of HA. Overall, the responses in our study were comparable to the findings from Bresson et al. [3] where 67% of participants achieved HI titres ≥1:32 with a 30 g HA aluminium-adjuvanted vaccine and lower dose formulations (7.5 g or 15 g HA) were less immunogenic, irrespective of the presence of adjuvant. Further, findings from our study suggest that immune responses to higher amounts of HA (30 g or 45 g) can persist for up to 6 months after a primary vaccination course, whereas responses to lower amounts of HA decline to near pre-vaccination levels after 6 months. Antibody persistence is important for reducing the impact of potential follow-on waves of pandemic activity that may occur. The prototype H5N1 vaccines in our study and those of Bresson et al. [3] were based on current licensed manufacturing processes for influenza vaccines using aluminium adjuvants with an extensive safety record. Similar to Bresson et al. [3] we found that aluminium adjuvant did not greatly improve immune responses at low HA concentrations (7.5 and 15 g). Therefore, further major improvements in immune responses to prototype avian influenza vaccines may require significant advances in adjuvant technology or the application of alternative manufacturing processes. Novel adju-
vants will require substantial use and monitoring before we have the same level of confidence provided by the aluminium adjuvant with respect to safety, given alum’s long and widespread use throughout the world. The ability to confer broad cross-protection through vaccination before a closely matched pandemic strain vaccine can be manufactured is a highly desirable property. Our preliminary findings confirm that two doses of prototype H5N1 vaccine can elicit cross-reactive MN antibodies against antigenically distinct H5N1 strains [6]. These findings are comparable to those of Stephenson et al. [15] who demonstrated cross-reactive antibodies against pathogenic avian influenza (H5N1) in individuals after two doses of non-pathogenic avian influenza (H5N3) vaccine, and higher levels of cross-reactive antibodies after a third booster dose, 16 months later. Vaccination strategies that include a primary vaccination course with a prototype pandemic vaccine, followed by a booster dose against a heterologous pandemic-specific strain, may be a useful option at the beginning of a pandemic, when the demand for vaccine is likely to surpass supply. In the absence of any established criteria for the licensing of pandemic influenza vaccines, we evaluated our vaccine formulations using the CPMP immunogenicity endpoints for licensing of seasonal influenza vaccines [12]. However, these endpoints may not be appropriate surrogates for assessing pandemic influenza vaccines as they were not designed for populations that are immunologically na¨ıve to influenza antigen, and serological correlation of protection from avian influenza has not been defined in humans or experimental systems. Recently, the United States Food and Drug Administration (FDA) released further guidelines for assessing pandemic influenza vaccines in immunologically na¨ıve populations [16]. However, both the CPMP and FDA criteria are based on the HI antibody assay for assessing protection from seasonal influenza virus infection. A serum HI titre of ≥1:40 is considered an approximate correlate of protection from seasonal influenza in humans [17], however, there are no clinical data to confirm that this seasonal HI target reflects protection from avian influenza. Moreover, as the HI assay may underestimate human immune responses to avian influenza viruses [9,11,13], other serological assays, such as the modified HI assay with horse erythrocytes and the MN assay, as used in this study, are increasingly used to evaluate immune responses to prototype H5N1 vaccines [3,5,10] However, as yet, there is no information describing the correlation between these immunogenicity end points and clinical protection. In conclusion, two doses of prototype 30 g or 45 g H5 HA, aluminium-adjuvanted vaccines were immunogenic and welltolerated, with antibody persistence for 6 months. Moreover, prototype H5N1 vaccines based on clade 1 strains of avian influenza elicited modest levels of cross-reactive MN antibodies against currently circulating, distinct clade 2, H5N1 strains. Acknowledgments The data reported here were presented, in part, at the VIII International Symposium on Respiratory Viral Infections, Hawaii, March 16–19, 2006; the 3rd WHO Meeting on Evaluation of Pandemic Influenza Vaccines in Clinical Trials, WHO/HQ, Geneva, February 15–16, 2007 and the Options for the Control of Influenza VI Conference, Toronto Canada, June 17–23, 2007. The authors would like to extend their thanks to the participants, investigators and personnel at each study site. Specifically, the authors would like to thank staff from the Vaccine and Immunisation Research Group (Murdoch Childrens Research Institute, University of Melbourne) including L. Thorn, K. Alexander, J. McVernon, M. Kefford, K. O’Grady, J. Ryrie and J. Sonego, staff from the Vaccine Trials Group (Institute for Child Health Research, Uni-
T.M. Nolan et al. / Vaccine 26 (2008) 4160–4167
versity of Western Australia) including S. Nadall, S. Bilic, J. Adams and K. Prosser and D. Lakhman, M. Rico Garcia and K. McCurrie from the Virus Reference Laboratory, The Health Protection Agency; Colindale, UK for their assistance in this study. This study was sponsored by CSL Limited (Parkville, VIC, Australia). The authors acknowledge the independent medical writing assistance provided by ProScribe Medical Communications (www.proscribe.com.au), funded from an unrestricted financial grant from CSL Limited. ProScribe’s services complied with international guidelines for Good Publication Practice. Contributors: All authors participated in the interpretation of results, writing of the manuscript and approval of the final version. T.M. Nolan had full access to all the data in the study and had final responsibility for the decision to submit for publication. T.M. Nolan, P.C. Richmond, M.C. Zambon, R.L. Basser, M.V. Skeljo, G. Pearce and J. Bennet designed the study with T.M. Nolan and P. Richmond as principal investigators. The study was led by T.M. Nolan and coordinated by T.M. Nolan, P.C. Richmond, K. Papanaoum, M.V. Skeljo, N.T. Formica and G. Pearce. D. Ryan was responsible for vaccine ¨ production. Immunological assays were conducted by K. Hoschler and M.C. Zambon and statistical and data analyses were conducted by T.M. Nolan, P.C. Richmond, M.C. Zambon, N. Formica and G. Hartel. Conflict of interest: M.V. Skeljo, G. Pearce, G. Hartel, N.T. Formica, J. Bennet, D. Ryan and R.L. Basser are employees of CSL Limited; Parkville, Victoria, Australia. T.M. Nolan and P.C. Richmond have received honorarium payments and travel support from CSL Ltd. for scientific advisory meetings on unrelated vaccine research. K. ¨ Hoschler and M.C. Zambon are employees of the Health Protection Agency, Colindale, UK, which received funding from CSL Ltd. to conduct the laboratory assays. All other authors have no conflict of interest. References [1] World Health Organisation. Cumulative number of confirmed human cases of avian influenza A (H5N1) reported to WHO; February 12, 2008. http://www. who.int/csr/disease/avian influenza/country/en/(accessed February 14, 2008). [2] Committee for Proprietary Medicinal Products (CPMP). Guideline on dossier structure and content for pandemic influenza vaccine marketing authorisation application (CPMP/VEG/4717/03). London: European Agency for the Evaluation of Medicinal Products (EMEA); April 5, 2004.
4167
[3] Bresson JL, Perronne C, Launay O, Gerdil C, Saville M, Wood J, et al. Safety and immunogenicity of an inactivated split-virion influenza A/Vietnam/1194/2004 (H5N1) vaccine: phase I randomised trial. Lancet 2006;367:1657–64. [4] Lin J, Zhang J, Dong X, Fang H, Chen J, Su N, et al. Safety and immunogenicity of an inactivated adjuvanted whole-virion influenza A (H5N1) vaccine: a phase I randomised controlled trial. Lancet 2006;368:991–7. [5] Treanor JJ, Campbell JD, Zangwill KM, Rowe T, Wolff M. Safety and immunogenicity of an inactivated subvirion influenza A (H5N1) vaccine. N Engl J Med 2006;354:1343–51. [6] Leroux-Roels I, Borkowski A, Vanwolleghem T, Drame M, Clement F, Hons E, et al. Antigen sparing and cross-reactive immunity with an adjuvanted rH5N1 prototype pandemic influenza vaccine: a randomised controlled trial. Lancet 2007;370:580–9. [7] World Health Organisation. Antigenic and genetic characteristics of H5N1 viruses and candidate H5N1 vaccine viruses developed for potential use as pre-pandemic vaccines; March, 2007. http://www.who.int/csr/disease/ avian influenza/guidelines/summaryH520070403.pdf (accessed October 18, 2007). [8] Nicolson C, Major D, Wood JM, Robertson JS. Generation of influenza vaccine viruses on Vero cells by reverse genetics: an H5N1 candidate vaccine strain produced under a quality system. Vaccine 2005;23:2943–52. [9] Stephenson I, Wood JM, Nicholson KG, Zambon MC. Sialic acid receptor specificity on erythrocytes affects detection of antibody to avian influenza haemagglutinin. J Med Virol 2003;70:391–8. [10] Stephenson I, Wood JM, Nicholson KG, Charlett A, Zambon MC. Detection of anti-H5 responses in human sera by HI using horse erythrocytes following MF59-adjuvanted influenza A/Duck/Singapore/97 vaccine. Virus Res 2004;103: 91–5. [11] Rowe T, Abernathy RA, Hu-Primmer J, Thompson WW, Lu X, Lim W, et al. Detection of antibody to avian influenza A (H5N1) virus in human serum by using a combination of serologic assays. J Clin Microbiol 1999;37:937–43. [12] Committee for Proprietary Medicinal Products (CPMP). Note for guidance on harmonisation of requirements for influenza vaccines (CPMP/BWP/214/96). London: European Agency for the Evaluation of Medicinal Products (EMEA); March 12, 1997. [13] Nicholson KG, Colegate AE, Podda A, Stephenson I, Wood J, Ypma E, et al. Safety and antigenicity of non-adjuvanted and MF59-adjuvanted influenza A/Duck/Singapore/97 (H5N3) vaccine: a randomised trial of two potential vaccines against H5N1 influenza. Lancet 2001;357:1937–43. [14] Treanor JJ, Wilkinson BE, Masseoud F, Hu-Primmer J, Battaglia R, O’Brien D, et al. Safety and immunogenicity of a recombinant hemagglutinin vaccine for H5 influenza in humans. Vaccine 2001;19:1732–7. [15] Stephenson I, Bugarini R, Nicholson KG, Podda A, Wood JM, Zambon MC, et al. Cross-reactivity to highly pathogenic avian influenza H5N1 viruses after vaccination with nonadjuvanted and MF59-adjuvanted influenza A/Duck/Singapore/97 (H5N3) vaccine: a potential priming strategy. J Infect Dis 2005;191:1210–5. [16] Food and Drug Administration (FDA). Guidance for Industry. Clinical data needed to support the licensure of pandemic influenza vaccines. Rockville: US Department of Health and Human Services; May, 2007. [17] Potter CW, Oxford JS. Determinants of immunity to influenza infection in man. Br Med Bull 1979;35:69–75.