Vaccine Failure Is Setback in AIDS Fight

In this fiscal year, the NIH's budget for AIDS vaccine research is $497 million. The STEP and Phambili trials were each expected to cost about $32 million. Pharmaceutical giant Merck & Co. has spent an undisclosed amount developing the vaccine and helping to manage the studies.

"We can't afford to have any more trials like this," said Mark Harrington, head of the activist Treatment Action Group and a longtime observer of AIDS research. "We have to stop and reassess and recommit to basic science, or people will begin to lose faith."

At the moment, only two things are certain.

The first is that the vaccine, developed by Merck, could not have caused HIV infection because it contains only three proteins from HIV, not the entire virus. The second is that there are no obvious villains.

"I do not think that what happened in this trial is an example of scientists blindly rushing into dangerous things," said John P. Moore, an AIDS virologist at Weill Cornell Medical College, who has criticized vaccine trials he considered futile. "In the general HIV-research community, I didn't know anyone who said this was going to happen."

Both trials recruited people who were at high risk of HIV infection through sexual activity. The STEP subjects included many male homosexuals; the Phambili volunteers were male and female heterosexuals. Half the people in each trial were randomly assigned to get three shots of vaccine, and half to get three shots of a harmless liquid containing no adenovirus or HIV proteins.

Each trial was to have 3,000 participants. STEP had finished enrolling subjects in North and South America, the Caribbean and Australia. Phambili (which means "moving forward" in the Xhosa language of South Africa) had signed up 801 by the time it was shut down.

While scientists hoped the Merck vaccine might prevent some infections, its chief purpose was to stimulate "cell-mediated" immunity to produce a less severe illness. Specifically, the vaccine was expected to lower the "viral load" of HIV in the bloodstream, which in turn would both prolong survival and lessen the chance the person would infect others.

Many experts are questioning the wisdom of that strategy, even if it had worked perfectly. Urging millions of people to take an AIDS vaccine that probably would not protect them from the virus, they say, would be a hard and confusing task, even in places where the epidemic still rages.

For the moment, that is an academic question. The vaccine failed to achieve any of its goals.

In both studies, people who got vaccine were more likely -- not less -- to become infected, with trends suggesting roughly a twofold risk. In the STEP study, which has many more cases to evaluate, nearly all that added risk was in people who had high levels of antibodies to adenovirus type 5 before they got their first shot -- evidence they had been previously infected with that strain. Uncircumcised men in that group had the highest risk.

So how could this have happened?