Vaccine 21 (2003) 1687–1693 Boosting immunity to influenza H5N1 with MF59-adjuvanted H5N3 A/Duck/Singapore/97 vaccine in a primed human population Iain Stephenson a,d,∗ , Karl G. Nicholson a , Anthony Colegate b , Audino Podda b , John Wood c , Ellen Ypma b , Maria Zambon d a Infectious Diseases Unit, Leicester Royal Infirmary, Leicester, LE1 5WW, UK b Chiron Vaccines, Siena, Italy and Amsterdam, Amsterdam, The Netherlands c National Institute for Biological Standards and Control, Potters Bar, Hertfordshire, UK d Public Health Laboratory Service, Colindale, London, UK Received 14 May 2002; received in revised form 7 October 2002 Abstract In 1997, influenza A/Hong Kong/97 (H5N1) emerged as a potential human threat. In 1999, a randomised study comparing two doses of MF59-adjuvanted and non-adjuvanted influenza A/Duck/Singapore/97 (H5N3) surface-antigen vaccine found non-adjuvanted vac- cine was poorly immunogenic. Addition of MF59 significantly boosted antibody to H5N1 to levels associated with protection. At 16 months, we undertook a follow-up study to assess the effect of H5N3 revaccination. Geometric mean titres (GMTs) of antibody by haemagglutination-inhibition (HI), microneutralisation (MN) and single radial haemolysis (SRH) indicated that protective antibody titres did not exist at 16 months after two-dose priming. Twenty-one days after revaccination, there was significant boosting of antibody compared to GMTs achieved 21 days after two-dose priming in the original study (P< 0.001). MF59 significantly increased GMTs of antibody when compared to non-adjuvanted vaccine (P< 0.001). © 2002 Published by Elsevier Science Ltd. Keywords: Avian influenza; Vaccination; Adjuvant 1. Introduction Influenza A viruses may undergo antigenic shift and cause pandemics with high mortality and morbidity. Human-avian reassortants were responsible for pandemics in at least 1957 and 1968 [1]. Typically, newly shifted strains have emerged in south-east Asia where agricultural practices al- lowing close proximity among humans, ducks, poultry and domesticated pigs facilitate reassortment. Until recently, avian influenza was considered unable to be transmitted to humans as avian strains preferentially bind to sialic acid receptors with -2,3-galactose linkages that are lacking in human respiratory epithelial cells [2]. However, in 1997, six deaths among 18 hospital admissions in Hong Kong caused by influenza A/Hong Kong/97 (H5N1) were pre- ceded by mass circulation of a highly pathogenic H5N1 virus in poultry [3]. Although the H5N1 virus was ineffi- ciently transmitted among humans, observations indicated that a human pandemic caused by avian influenza was a ∗ Corresponding author. Tel.: +44-116-258-6952; fax: +44-116-258-5067. E-mail address: iain.stephenson@uhl-tr.nhs.uk (I. Stephenson). possibility [4]. The ability to combat the mortality, morbid- ity and disruption caused by pandemic influenza depends primarily on the availability of an effective vaccine. In June 1999, we undertook a randomised dose-escalating study of non-adjuvanted and MF59-adjuvanted influenza A/Duck/Singapore/97 (H5N3) surface-antigen vaccine in humans [5]. Sixty-five healthy adults were recruited to re- ceive two doses, 3 weeks apart, of either non-adjuvanted or MF59-adjuvanted H5N3 vaccine containing 7.5, 15 or 30 g H5 haemagglutinin content. Both vaccines were safe and well-tolerated. The antibody response was measured by haemagglutination-inhibition (HI), microneutralisation (MN), H5N3 single radial haemolysis (SRH) and H5N1 SRH. HI was found to significantly underestimate responses to avian H5 compared to MN and SRH. Non-adjuvanted vaccine was poorly immunogenic with doses up to 30 g unlikely to confer protection against H5N1. Addition of MF59 was found to significantly boost post-vaccination SRH antibody to levels fulfilling European Committee of Proprietary Medicinal Products (CPMPs) criteria for licens- ing of interpandemic vaccines [6]. Sixteen months later, a follow-up study was performed to assess revaccination on a primed immune system. Pre- and post-vaccination antibody 0264-410X/02/$ – see front matter © 2002 Published by Elsevier Science Ltd. PII:S0264-410X(02)00632-1