Vaccine companies around the world are racing to manufacture enough bird-flu vaccine to deal with a feared influenza pandemic. Yet it's unclear whether this vaccine will work against a more virulent strain of avian influenza. What is clear is that if it fails, it would take months to years to create a new vaccine in sufficient quantities, giving the virus time to spread throughout the world.

That fear has reinvigorated the hunt for a more universal vaccine to protect against any flu -- both a pandemic strain and the seasonal virus that spreads every winter. Although such a universal vaccine is unlikely to be ready if a pandemic strikes soon, it could provide a longer-term solution to the threat of virulent flu.

"What a major step forward [a universal vaccine] would be," says Gregory Poland, a vaccinologist at the Mayo Clinic College of Medicine in Rochester, MN. "The promise is very definitely there."

Existing seasonal flu vaccines are targeted to a part of the influenza virus that mutates rapidly, so they must be made fresh every year. Public health officials must guess what next year's strain will look like, and then manufacturers can create a vaccine based on that guess.

Avian flu, the strain that public health officials fear could lead to a pandemic, presents a similar problem. Scientists have created a bird-flu vaccine based on the current circulating strain of virus. If that virus mutates, however, it's unclear how well the current vaccine will work.

"You can't predict what changes will happen to a virus," says Andrew Pekosz, a virologist at the Washington University School of Medicine in St. Louis. "That's part of the problem with flu -- you can't make a flu vaccine unless you have [the virus] in humans."

A universal flu vaccine would eliminate that problem, thus avoiding the lag time between identifying a virulent flu strain and developing and manufacturing enough vaccine to combat it. Such a vaccine could be made year-round and stored for future pandemics. What's more, manufacturers wouldn't have to dump excess quantities that become obsolete as the virus mutates.

The key hurdle in creating a universal flu vaccine, biologists say, is finding a component of the virus that remains stable over time. While conventional flu vaccines target a large protein on the surface of viruses, the best targets for a universal vaccine may be smaller, more stable proteins. Under normal circumstances, these proteins tend to slip past the immune system unnoticed. So in a traditional vaccine, which contains an inactive form of the virus, these proteins elicit no immune response.

But researchers have developed ways to modify these small proteins so that they aggravate the immune system enough to make antibodies -- and thus act as a vaccine.