The U.S. researchers said that years of exposure to the human flu subtype H1N1 - one of two types of influenza A viruses that circulate in flu season - may provide some protection against the H5N1 virus that has ravaged poultry flocks in large parts of Asia and killed at least 166 people over the last three years.
The theory is that antibodies to H1N1 might recognize and go into battle against H5N1, because the two viruses carry a related neuraminidase protein - the N in a flu virus's name - on their surfaces.
The suggestion is tantalizing and hope-provoking, given H5N1's extraordinary virulence when it infects people. To date, 60 per cent of people known to have been infected with it have died.
But experts not related to the research suggested the paper doesn't prove getting annual flu shots now will help people weather a pandemic later, should H5N1 start spreading easily among people.
"That to me has not been established by this paper," said Dr. Frederick Hayden, a scientist with the World Health Organization's global influenza program.
Flu dogma has it that the hemagglutinin protein - the H in a flu virus's name - is the major player in influenza infection and the best target against which to direct the immune system when designing a vaccine.
However, studies conducted after the 1968 flu pandemic - when the H3N2 subtype replaced H2N2 viruses that had been circulating for 11 years - suggested that the shared N2 component of those two viruses may have blunted the blow of that pandemic. In fact, the 1968 pandemic was the mildest of the last century.
That generated the hypothesis that previous exposure to a neuraminidase subtype might not prevent infection but might at least lessen the ensuing illness. Supporters of the notion point to the age distribution of H5N1 cases - 90 per cent have been younger than age 40 - to bolster the argument, saying people who are older have had that much more cumulative exposure to H1N1 viruses.
So a team of researchers led by leading influenza virologist Richard Webby of St. Jude Children's Research Hospital in Memphis, Tenn., set out to look at the question.
He and his colleagues injected mice with a DNA vaccine containing the N1 neuraminidase of contemporary human flu viruses, which has about an 80 per cent similarity to the neuraminidase of the H5N1 avian virus. Other mice were given a dummy vaccine.
They then exposed the mice to what ought to have been a lethal dose of H5N1 virus. All the mice that got the placebo died, but half of those receiving the N1 vaccine survived.
To test it out further, they injected a fresh group of mice with blood from the survivors, then exposed them to a lethal dose of H5N1. The entire second group of mice survived.
As well, the team tested the blood of healthy human volunteers to see if it mounted an antibody reaction when confronted with H5N1 virus. The blood of between one-fifth and one-quarter of the volunteers - U.S. residents unlikely to have ever been exposed to H5N1 - did generate some response.
That suggests there could be a greater role for neuraminidase proteins in flu vaccine production. But even the lead author of the study admitted that could be a difficult nut to crack.
Neuraminidase content isn't standardized in current flu vaccines; asking manufacturers to increase this component might come at a cost.
"Perhaps by increasing yields of neuraminidase you may drop yields of hemagglutinin," Webby said from Memphis.
"Certainly there's no doubt that obviously the most important part of our seasonal vaccine is hemagglutinin. And you wouldn't want to jeopardize that for the sake of boosting what is probably reasonably weak cross-reactivity to neuraminidase."
If the neuraminidase of H1N1 does offers some protection against H5N1, actually getting sick with that strain might be a better way to build up immunity, Webby acknowledged.
But public health officials would not suggest people deliberately try to catch the flu for a gain that is not yet proven.
Hayden insisted the jury is still out on this idea, arguing that if some people are being protected by their earlier exposure to H1N1, there should be some evidence of mild infection among older people in H5N1 affected countries.
So far, the small studies that have tested blood samples from people in countries battling H5N1 actually point in the opposite direction - very few mild cases have been found.
"It is premature to conclude that immunity induced by the human influenza virus N1 NA (neuraminidase) will provide significant protection from illness associated with avian influenza H5N1 virus infection," concurred Laura Gillim-Ross and Kanta Subbarao of the U.S. National Institute of Allergy and Infectious Diseases in a commentary published by the journal.
