The major finding of this study is that an inactivated influenza A (H7N7) subunit vaccine prepared using manufacturing methods approved for seasonal influenza vaccines was poorly immunogenic in healthy young adults despite their having been given two doses a month apart of dosages up to 90 µg of the HA as determined in SRID assays. The neut assay at the central laboratory appeared to be slightly more sensitive than the HAI assay whereas the HAI assay appeared more sensitive than the neut assay at a laboratory with considerable experience with avian influenza virus serology (data not shown). Nevertheless, the serologic findings at the central laboratory were essentially confirmed. The more sensitive HAI assay performed at the study site provided a more complete assessment of immunogenicity. That test indicated that immunogenic HA was present in the vaccine although the immunogenicity was still considerably lower than what was expected for the dosages administered. The major limitation of these findings is that the sample size was not selected to produce definitive immunogenicity data on any of the dosages tested. However, the desired guidance for any further testing was obtained.
Influenza A (H7N7) is proposed as a priority for vaccine development for potential use in humans along with H5N1 and H9N2 because of human infections that have been detected with these viruses 
. The case for H7N7 was emphasized when an outbreak with H7N7 viruses occurred in 2003 in the Netherlands among persons exposed to poultry; one infected person died and evidence for human-to-human transmission was provided 
. Also, in an outbreak of H7N3 infections in poultry in Canada in 2004, transmission to humans was detected 
Considerable effort has been expended on vaccine development with H5N1 viruses for potential use in humans 
. Influenza A/H7 vaccines have been developed and shown to be immunogenic and to induce protection in animals 
. Thus, the H7 HA appears fully immunogenic in vaccinations of animals and should also be fully immunogenic for humans. Other clinical trials in humans with H7 vaccines have been reported. Low immunogenicity was reported in a clinical trial using a split virus vaccine prepared in tissue culture with A/H7N1 plasmids derived from a chicken virus 
. Two vaccinations three weeks apart of 12 and 24 µg dosages of the HA with and without alum were each given to 15 healthy adults (total 60). The 12 and 24 µg dosages without alum induced HAI and/or neut antibody responses in 21 and 23% of subjects, respectively and 50 and 62% respectively with adjuvant. A clinical trial with an H7N3 live attenuated virus using the A/Ann Arbor (H2N2) cold-adapted Master Donor Strain given to 17 subjects showed minimal virus shedding but serum HAI and neut antibody responses in 43% and 48%, respectively 
The present report and the two summarized clinical trials, in combination with the good immunogenicity in animal immunizations, indicate that a satisfactory H7 vaccine can be made for humans. Because of the potential need for an H7N7 vaccine for humans, developing a vaccine for potential use in humans seems indicated. To aid in this goal, we have directed research toward gaining some understanding of why this H7N7 vaccine of high HA dosage was so poorly immunogenic in humans. These studies are reported in a companion report on in vitro correlates of immunogenicity of avian influenza virus vaccines in humans.