TORONTO - Flu viruses that circulated decades ago may be protecting older adults from the new swine flu virus, scientists at the U.S. Centers for Disease Control suggested Thursday.

Blood samples from U.S. children don't produce any antibodies when exposed to the new H1N1 virus, but antibodies were detected in about one third of the samples from older adults, the study shows. But whether those antibodies would block infection with the new virus can't be revealed by this kind of testing.

"We don't yet understand what that means in terms of protection, if anything at all," said Dr. Jacqueline Katz, first author and chief of the immunology and pathogenesis branch of the CDC's influenza division.

Others remarked, though, that the laboratory finding is intriguing because it mirrors what is being seen in the real world as the virus spreads.

Children and young adults make up the bulk of confirmed swine flu cases. People over 60, who are normally among the most vulnerable to flu and its complications, account for a small portion of infections.

More work is underway to try to tease out what might account for this response in older adults. If it is felt the immune systems of older adults have been primed to respond to this virus by exposure to similar viruses in the past, that information could help set vaccination policies for older adults when H1N1 vaccine is ready.

"It's conceivable that potentially some age groups may have essentially been primed through exposure to H1N1 in the past and may not need two doses," Katz said in an interview.

The study appears in Friday's issue of the CDC journal Morbidity and Mortality Weekly Report.

The CDC scientists tested stored blood samples from people who had taken part in previous vaccine trials in the U.S. and Europe, though the blood samples from older adults came exclusively from the United States, said Dr. Anne Schuchat, the agency's director of immunizations and respiratory diseases.

The scientists were looking for what are called cross-reactive antibodies -- not antibodies specific to that particular virus, but ones created by exposure to earlier viruses that are nonetheless able to recognize the new invader.

Schuchat said the CDC is wondering if human H1N1 viruses that circulated in the 1930s, '40s or '50s might have been closer to this new virus than contemporary human H1N1s.

The notion is "plausible" given the findings, said Dr. Donald Kennedy, an infectious disease specialist with St. Louis University's Center for Vaccine Development.

"If these viruses that were around in people that are over 60 -- i.e. back in the '30s and '40s -- are closer to this present swine flu virus, there would be some ... immunologic memory for that particular virus and that particular strain," Kennedy said.

The CDC researchers were looking at blood samples drawn before and after vaccination with seasonal flu shots, which also allowed them to test whether vaccine made to protect against human H1N1 viruses might offer some protection against the new swine viruses.

Though the viruses carry the same name -- H1N1 -- they are distant relatives. So distant, in fact that in children the flu shot generated no antibody rise against the swine virus. Adults showed a two-fold increase in cross-reactive antibodies, but that is a meagre effect, suggesting seasonal flu shots don't protect against these viruses.

Interestingly, though, where a third of blood samples from adults over 60 showed significant levels of antibodies before the seasonal flu shot, that rose to 43 per cent after the vaccination. While that looks a bit like a booster effect, Schuchat cautioned that it's too soon to draw that conclusion.

Influenza expert Dr. Arnold Monto suggested that while the findings are interesting, he'd like to see a similar study done in another country.

That's because the United States vaccinated more than 40 million people in 1976 against a swine flu virus that was feared to be poised to start a pandemic. Until a similar study is done in a locale that didn't vaccinate against swine flu in 1976, one cannot rule out the possibility that antibodies generated by that vaccine are influencing the results of this study, Monto said.

"That (1976) swine virus is more closely related to the current (swine) virus than a lot of what's been transmitting more recently in humans," he said from Ann Arbor, Mich., where he teaches at the University of Michigan.

Dr. Allison McGeer, an influenza expert at Toronto's Mount Sinai Hospital, said Canadian scientists are embarking on just such a study here. And she offered another possible explanation for the findings.

McGeer suggested the antibodies seen in the blood of older adults may not have been generated by long ago exposure to viruses similar to the new swine H1N1. Rather, she said, it might be that the human immune system amasses protection against H1N1 viruses by years of exposure to influenza viruses and flu shots.

"It's beginning to look to me like it's more sensible to think that there's something about H1N1s that means that you accumulate antibody in a way that you just don't do for other influenza viruses," she said.

Katz said another possibility is that because older adults would have had their first influenza A infection with H1N1 viruses -- H1N1s were the only influenza A viruses circulating before 1957 -- their immune systems might be more responsive to flu viruses of this subtype.

"That may also be playing a role in making a very robust response," she said.