In a puzzling and potentially troubling development, an AIDS vaccine tested in a closely watched trial might have increased the risk among vaccine recipients of becoming infected with HIV, researchers reported yesterday at a scientific meeting in Seattle.
But the researchers said not enough data existed to determine the meaning of the findings about the vaccine, which is made by Merck.
The increased risk was principally among a group of people who had pre-existing levels of immunity to a common cold virus known as adenovirus type 5, which was modified to become a critical part of the vaccine. Researchers emphasized that the vaccine itself could not cause AIDS, but one theory is that the cold virus may have activated the immune system in some way to make certain recipients more susceptible to becoming HIV-infected when they were exposed to the AIDS virus.
But participants at the meeting also emphasized that the findings could be a statistical fluke and that nonbiological factors might have accounted for the difference. Examples of such factors are rates of circumcision and sexual practices among trial participants.
In late September, Merck unexpectedly halted the trial of its experimental HIV vaccine because it failed in its two main objectives, to prevent infection and to lower the amount of HIV in the blood among those who became infected.
The vaccine was being tested among 3,000 volunteers at high risk of developing AIDS in nine countries, including those at immunization centers organized by the National Institutes of Health in the United States. Merck's was seen as one of the most promising experimental AIDS vaccines to have been tested on people. Many scientists and advocates of AIDS research have called the failure of the experimental vaccine a major setback.
"The new analyses are both disappointing and puzzling" because they offer no explanation for the vaccine's failure, said Dr. Anthony S. Fauci, the director of the National Institute of Allergy and Infectious Diseases, a partner in the vaccine trial.
In September, a preliminary analysis of about half those in the trial suggested that those vaccinated were becoming HIV-infected at roughly the same rate as those receiving a placebo. There were 24 HIV infections among those vaccinated compared with 21 who received a placebo.
But the new analysis looked at all the trial participants and found a wider difference -- 49 in the vaccinated group compared with 33 in the placebo group. Further analysis showed that the imbalance was much more apparent among those who had the highest level of pre-existing immunity to the cold virus used in the vaccine.
Among 778 male volunteers who had a high level of pre-existing adenovirus immunity, 21 of those receiving the vaccine developed HIV infections compared with 9 in the placebo group. Researchers told reporters by telephone that when such an analysis is performed after a trial is stopped, defining what is statistically significant is difficult.
The difference between the groups with low levels of adenovirus immunity was smaller -- 28 among the vaccine recipients compared with 24 who received a placebo. That difference was not statistically significant.
Verification of the hypothesis requires extensive laboratory tests.
Findings could determine the underlying biological mechanism responsible for the vaccine's failure. It will take months to years for such tests to be completed, Dr. Keith Gottesdiener, a Merck vice president, said in an interview.
Although 35 percent to 40 percent of the trial participants were women, only one woman developed HIV. The reason for the gender difference is unknown. Her case was removed from the statistical analysis.
The new reports create even more scientific confusion about how to develop a vaccine to stop the global HIV pandemic, which has infected an estimated 39 million people and killed 25 million more.
The findings raise questions about whether adenovirus can ever be used as a crucial ingredient in an AIDS vaccine and whether new tacks will be needed. Use of a modified virus as a vector to deliver HIV genes is a new and evolving way to make an AIDS vaccine. The Merck vaccine included three synthetic HIV genes.
Scientists and groups that advocate AIDS research said the vaccine failure should not lessen the commitment to develop a vaccine.
The new data "raise more questions than answers for the field of AIDS vaccine," the AIDS Vaccine Advocacy Coalition, a community and consumer-based group, said in a statement.
The coalition urged AIDS scientists not to begin trials of other new vaccines until more definitive conclusions could be reached from further analysis of the Merck vaccine.
Meeting participants will continue discussions today about whether the trial leaders should continue to observe the participants without telling them whether they received the vaccine or a placebo and the results of their exposure to the cold virus before the study began. A recommendation will be made in about 10 days, Dr. Gottesdiener said.
"We did a beautiful experiment, but it definitely was a disappointment," Dr. Larry Corey of the University of Washington, who led the investigators, said in an interview. "One lesson is that scientists will have to look at vector-based immunity more thoroughly than we have in the past."