James Campbell, the principal investigator for the study and assistant professor of pediatrics, said he and his collaborators at the University of Rochester and the University of California in Los Angeles determined that the vaccine should be administered in two doses.

“A single dose of the vaccine did not give people protective levels of antibodies in their blood,” he said.

Four different dose amounts were tested: 7.5, 15, 45 and 90 micrograms, Campbell said. The scientists found that the two largest amounts produced the most significant immune response.

The vaccine appears to be safe, Campbell said, adding that the 154 Maryland volunteers were constantly monitored for red or sore arms or other adverse reactions to the intravenous shots.

“Still, whenever you test a new vaccine, you have to start with the assumption that there might be something that you weren’t expecting but could happen,” he said.

On Oct. 31, the scientists will draw new blood samples from the Maryland volunteers and take a last measure of antibodies in their blood. Those results will be compared to blood drawn before the first vaccine was administered and at two other stages during the study.

Antibodies are what the immune system makes to fight a foreign invader or disease, Campbell said. More antibodies detected in the blood means that the body is developing a better immune response to fight the avian flu.

Campbell said he didn’t know when the vaccine would make it on the market. “At this point, it’s hard to predict, but if the need arose, doses will be made,” he said.

According to Oct. 10 estimates from the World Health Organization, 60 people in Asia have died from the H5N1 strain since December 2003, and 117 cases of it have been confirmed.

The Centers for Disease Control says the current risk to Americans is low if an outbreak occurs in Asia but remains mostly confined. There was a brief scare in 2004 when another less harmful strain of the virus was discovered on two Delaware farms. Since 1997, millions of chickens and pigs worldwide have been killed to keep the virus at bay.

If the virus were to mutate and gain the ability to transmit from human to human, which it currently cannot do, containment becomes much more difficult, said Derek Cummings, a research associate at the Johns Hopkins Bloomberg School of Public Health.

Scientists worry that in Asia, where human and poultry contact isn’t very regulated, chances are high that someone will become infected with both a human and an avian flu strain, Cummings said. In that event, the two virus strains could easily exchange genes, and the avian strain would acquire the ability to transmit between humans. If that were to occur, CDC experts estimate that a “medium-level” pandemic could kill 89,000 to 207,000 people and render anywhere from 20 million to 47 million people sick.