Skip Navigation

Might H5N1 Vaccine Cross Protection Allow for Pre -Pandemic Vaccine Stockpiling?

By Eric Toner, M.D. and Luciana Borio, M.D, October 4, 2006

There have been doubts as to whether our government’s strategy to stockpile vaccine in bulk in advance of a pandemic will be beneficial. This is because influenza A viruses are constantly mutating and drifting antigenically; therefore, any vaccine stockpiled in advance would not be well matched to the eventual pandemic strain.

For example, the first major H5N1 vaccine trial conducted in the U.S. used a vaccine derived from a clade 1 H5N1 virus (the strain found in Vietnam, Thailand, Cambodia and Laos); however, in vitro testing indicates that this vaccine does not appear to induce protective antibodies against clade 2 H5N1 viruses (the strain found everywhere else) [1]. Because it takes approximately 6 months or more from the isolation of a new pandemic influenza virus until vaccine specific to that strain becomes available using current egg based technology, it is now widely assumed that an effective pandemic vaccine will not be available for many months after a pandemic begins. As a consequence, pandemic planning has mostly focused on public health measures for disease containment and medical care for the ill.

However, recent experiments in ferrets conducted by Dr. Robert Webster’s group at St. Jude Children's Research Hospital have yielded promising results that suggest that even a vaccine that is not perfectly matched to a circulating strain may still provide some degree of cross protection. If there is significant cross protection among H5 strains in humans, then it is sensible to stockpile a generic H5 vaccine before the onset of a pandemic.

Evidence of cross protection in the ferret model: In the July 15 issue of the Journal of Infectious Disease [3], Govorkova and colleagues report on experiments conducted on ferrets which demonstrated two important findings, (1) vaccination against one H5N1 strain resulted in significant cross protection against antigenically distinct H5N1 strains and (2) this cross protection was not always detectable by the usual in vitro testing for the presence of antibodies. Ferrets were immunized with one or two doses of inactivated whole-virus vaccine produced by reverse genetics with the hemagglutinin and neuraminidase genes of an H5N1 virus isolated in Hong Kong in 2003 (A/HK/213/03 virus). The ferrets were subsequently challenged with lethal doses of a virus isolated in Vietnam in 2004. All the vaccinated animals lived—even those who had received only one dose of vaccine. The ferrets that received only one dose of vaccine had no detectable HI or neutralizing antibodies against the Vietnam virus, but were nevertheless protected from death. As the authors point out, the primary purpose of a pandemic vaccine is to reduce deaths, not necessarily to prevent infections.

Possible benefits of stockpiling: Even a vaccine that is only partially protective could have significant impact on the consequences of a pandemic. In a mathematical model by Longini and colleagues, early use of a poorly matched vaccine that still partially reduces transmission resulted in significant mitigation of the outbreak [2]. In addition, a reduction in the severity of disease, even without an accompanying reduction in the actual number of people infected, would have a significant impact on the consequences of a pandemic by reducing the burden on the healthcare system.

Priming with H5 antigen: A related strategy to stockpiling a “generic” H5N1 to be used at the start of a pandemic is that of “H5 priming” vaccination. In this strategy, a generic H5 antigen is included in the annual flu shots before the actual pandemic in the hope that recipients would either (1) develop some partial cross-immunity to an eventual H5 pandemic strain or (2) require only one dose of a pandemic vaccine rather than the two-dose course shown by clinical trials to be necessary now. This strategy could offer significant logistical advantages, but uncertainties remain, including whether the next pandemic will be caused by an H5 strain.  


Clearly, much more research is needed before this preliminary animal study can be used to guide policy.  For example, it is not known whether these results from ferrets can be extrapolated to humans and whether the benefit shown with whole-virus vaccine would extend to split-virus vaccines. Nonetheless, the work reported by the St. Jude group is encouraging and offers some hope that H5N1 vaccines currently in development could significantly mitigate the potentially catastrophic consequences of a severe H5N1 pandemic.


  1. Poland GA. Vaccines against Avian Influenza — A Race against Time. N Engl J Med. 2006;354:1411-3. Available at  Accessed October 4, 2006

  2. Germann TC, Kadau K, Longini IM, Macken,CA. Mitigation strategies for pandemic influenza in the United States. PNAS 2006;103: 5935-5940. Available at  Accessed October 4, 2006

  3. Govorkova EA, Webby RJ, Humberd J, Seiler JP, Webster RG. Immunization with reverse-genetics–produced H5N1 influenza vaccine protects ferrets against homologous and heterologous challenge. J Infect Dis 2006;194:159-67. Available at Accessed October 4, 2006