Although the first clinical trial of an AIDS vaccine proved to be an overall failure, researchers were sharply divided Monday over data from a subset of participants that seemed to offer hope for treating blacks and Asians.

After a five-year trial, officials of Brisbane, Calif.-based VaxGen Inc. were forced to concede that their prototype vaccine, called Aidsvax, provided no protection to the vast majority of the people who received it.

"Overall, what we see is that the trial failed," said VaxGen President Dr. Donald Francis.

But opinions varied sharply about the meaning of results obtained with a group of blacks and Asians representing less than 10 percent of the more than 5,400 people who took part in the study. To some experts, they represented the most important--or at least the most intriguing–part of the findings.

Among that small group of Asians, blacks and other non-Latino minorities, the vaccine reduced the infection rate by 67 percent. Among blacks alone, the results were even better–a 78 percent reduction in infections.

"These results are promising. The trial provides clear evidence that a vaccine can work," said Dr. Peter Piot, executive director of the United Nations AIDS agency.

Others, however, were sharply critical.

Dr. John P. Moore of the New York Weill Cornell Medical Center argued that the racial data are "a statistical artifact" based on the small number of individuals involved. There is, he added, "no biological reason based on how the vaccine works that race would make a difference."

He charged that VaxGen officials "played the race card ... because (the company) had to show some results to its public investors."

Martin Delaney of the San Francisco-based AIDS group, Project Inform, agreed, noting, "This is at best premature and irresponsible data reporting. ... It would do a great deal of harm to stir up hopes for selected groups over a vaccine that has proven ineffective overall."

But Phill Wilson, executive director of the Black AIDS Institute, cautioned that the data might not have been interpreted correctly. The majority of the blacks in the study were women, he noted, while all of the Caucasians were males. "This might be a gender effect rather than a racial one," he said.

However results with the subgroup are interpreted, people should not be overly discouraged by the findings, said Dr. Seth Berkley of the International AIDS Vaccine Initiative. Although most observers did not expect significant protection from Aidsvax, he said, "scientists remain confident that an AIDS vaccine is possible. Alternative AIDS vaccines, employing different design strategies, are now in development and some have already entered clinical trials."

Some of those look much more promising, he said.

Aidsvax is a vaccine in the classic mold, designed to stimulate the production of so-called neutralizing antibodies that bind to the virus’ surface and prevent it from attacking host cells. To achieve this, the company’s scientists genetically engineered a surface protein, called gp120, found in two strains of HIV. This protein is then injected into the recipient in an effort to stimulate antibody production.

For the trial, researchers enrolled 5,108 gay men and 309 women at high risk at 59 centers in the United States, Canada, Puerto Rico and the Netherlands. All were HIV-negative at the beginning of the trial.

Two of every three participants were randomly selected to receive at least three injections with the vaccine, while the rest received an equal number of sham injections. All were given counseling about risk avoidance. All were monitored for three years.

Even company officials did not expect the vaccine to be highly protective. Francis had noted on many occasions that he would be overjoyed if it proved to be 30 percent to 50 percent effective. As a measure of how desperate the world AIDS situation is, the Food and Drug Administration had said it would approve the vaccine even if it were only 30 percent effective. Most approved vaccines, in contrast, are 80 percent effective or higher.

But even the 30 percent figure proved to be wildly optimistic. Company officials said Monday that the reduction of infection produced by the vaccine was 3.8 percent for the entire group. The good news, they added, was that there were no adverse effects other than local tenderness at the site of the injection.

There was no evidence of increased susceptibility to infection among those who received the vaccine, and no evidence of an increased rate of progression among those who received the vaccine and subsequently became infected.

Company officials said the increased efficacy among non-Latino minorities was most likely real–there is only a one in 100 chance that it is coincidental. Phillip Berman, the company’s vice president of research and development, also added that the level of protective antibodies was higher in those groups than in whites.

But critics noted that among the 498 non-Latino minority participants, only 29 people became HIV-positive--17 of whom received the vaccine and 12 of whom did not. Among the 314 blacks, 13 became HIV-positive–four who received the vaccine and nine who received placebos.

"The numbers of participants in this group are too small to draw any conclusions about the vaccine’s effects in this subpopulation," said Chris Collins, executive director of the AIDS Vaccine Advocacy Coalition, who criticized the company for releasing the data in this form.

"Such premature conclusions run the risk of raising false hopes in a world desperate for an AIDS vaccine," he said.

Researchers are likely to argue about these results for months, especially because there is no other data available. The next major results that are expected are from a slightly different formulation of Aidsvax that is being tested among about 2,500 drug abusers in Thailand. Those results are expected in June.